From fb794a708a71d7c6af55f04cc4ed2d5823fb8b33 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 25 Sep 2019 11:16:55 +1000 Subject: [PATCH 001/160] PCI: Protect pci_reassign_bridge_resources() against concurrent addition/removal pci_reassign_bridge_resources() can be called by pci_resize_resource() at runtime, it walks the PCI tree up and down, and it isn't currently protected against any changes or hotplug operation. Hold the pci_bus_sem to protect it. Link: https://lore.kernel.org/r/7339fd73ccaf58552737ab10008333fd9f7723f2.camel@kernel.crashing.org Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Bjorn Helgaas --- drivers/pci/setup-bus.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c index e7dbe21705ba..f1c51943bdfc 100644 --- a/drivers/pci/setup-bus.c +++ b/drivers/pci/setup-bus.c @@ -2066,6 +2066,8 @@ int pci_reassign_bridge_resources(struct pci_dev *bridge, unsigned long type) unsigned int i; int ret; + down_read(&pci_bus_sem); + /* Walk to the root hub, releasing bridge BARs when possible */ next = bridge; do { @@ -2100,8 +2102,10 @@ int pci_reassign_bridge_resources(struct pci_dev *bridge, unsigned long type) next = bridge->bus ? bridge->bus->self : NULL; } while (next); - if (list_empty(&saved)) + if (list_empty(&saved)) { + up_read(&pci_bus_sem); return -ENOENT; + } __pci_bus_size_bridges(bridge->subordinate, &added); __pci_bridge_assign_resources(bridge, &added, &failed); @@ -2122,6 +2126,7 @@ int pci_reassign_bridge_resources(struct pci_dev *bridge, unsigned long type) } free_list(&saved); + up_read(&pci_bus_sem); return 0; cleanup: @@ -2150,6 +2155,7 @@ cleanup: pci_setup_bridge(bridge->subordinate); } free_list(&saved); + up_read(&pci_bus_sem); return ret; } From 6acdf7e19b37cb3a9258603d0eab315079c19c5e Mon Sep 17 00:00:00 2001 From: Logan Gunthorpe Date: Tue, 10 Sep 2019 13:58:33 -0600 Subject: [PATCH 002/160] PCI/switchtec: Read all 64 bits of part_event_bitmap The part_event_bitmap register is 64 bits wide, so read it with ioread64() instead of the 32-bit ioread32(). Fixes: 52eabba5bcdb ("switchtec: Add IOCTLs to the Switchtec driver") Link: https://lore.kernel.org/r/20190910195833.3891-1-logang@deltatee.com Reported-by: Doug Meyer Signed-off-by: Logan Gunthorpe Signed-off-by: Bjorn Helgaas Cc: stable@vger.kernel.org # v4.12+ Cc: Kelvin Cao --- drivers/pci/switch/switchtec.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/switch/switchtec.c b/drivers/pci/switch/switchtec.c index 8c94cd3fd1f2..465d6afd826e 100644 --- a/drivers/pci/switch/switchtec.c +++ b/drivers/pci/switch/switchtec.c @@ -675,7 +675,7 @@ static int ioctl_event_summary(struct switchtec_dev *stdev, return -ENOMEM; s->global = ioread32(&stdev->mmio_sw_event->global_summary); - s->part_bitmap = ioread32(&stdev->mmio_sw_event->part_event_bitmap); + s->part_bitmap = ioread64(&stdev->mmio_sw_event->part_event_bitmap); s->local_part = ioread32(&stdev->mmio_part_cfg->part_event_summary); for (i = 0; i < stdev->partition_count; i++) { From 157c1062fcd86ade3c674503705033051fd3d401 Mon Sep 17 00:00:00 2001 From: Lukas Wunner Date: Fri, 9 Aug 2019 12:28:43 +0200 Subject: [PATCH 003/160] PCI: pciehp: Avoid returning prematurely from sysfs requests A sysfs request to enable or disable a PCIe hotplug slot should not return before it has been carried out. That is sought to be achieved by waiting until the controller's "pending_events" have been cleared. However the IRQ thread pciehp_ist() clears the "pending_events" before it acts on them. If pciehp_sysfs_enable_slot() / _disable_slot() happen to check the "pending_events" after they have been cleared but while pciehp_ist() is still running, the functions may return prematurely with an incorrect return value. Fix by introducing an "ist_running" flag which must be false before a sysfs request is allowed to return. Fixes: 32a8cef274fe ("PCI: pciehp: Enable/disable exclusively from IRQ thread") Link: https://lore.kernel.org/linux-pci/1562226638-54134-1-git-send-email-wangxiongfeng2@huawei.com Link: https://lore.kernel.org/r/4174210466e27eb7e2243dd1d801d5f75baaffd8.1565345211.git.lukas@wunner.de Reported-and-tested-by: Xiongfeng Wang Signed-off-by: Lukas Wunner Signed-off-by: Bjorn Helgaas Cc: stable@vger.kernel.org # v4.19+ --- drivers/pci/hotplug/pciehp.h | 2 ++ drivers/pci/hotplug/pciehp_ctrl.c | 6 ++++-- drivers/pci/hotplug/pciehp_hpc.c | 2 ++ 3 files changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index 654c972b8ea0..882ce82c4699 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -72,6 +72,7 @@ extern int pciehp_poll_time; * @reset_lock: prevents access to the Data Link Layer Link Active bit in the * Link Status register and to the Presence Detect State bit in the Slot * Status register during a slot reset which may cause them to flap + * @ist_running: flag to keep user request waiting while IRQ thread is running * @request_result: result of last user request submitted to the IRQ thread * @requester: wait queue to wake up on completion of user request, * used for synchronous slot enable/disable request via sysfs @@ -101,6 +102,7 @@ struct controller { struct hotplug_slot hotplug_slot; /* hotplug core interface */ struct rw_semaphore reset_lock; + unsigned int ist_running; int request_result; wait_queue_head_t requester; }; diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index 21af7b16d7a4..dd8e4a5fb282 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -375,7 +375,8 @@ int pciehp_sysfs_enable_slot(struct hotplug_slot *hotplug_slot) ctrl->request_result = -ENODEV; pciehp_request(ctrl, PCI_EXP_SLTSTA_PDC); wait_event(ctrl->requester, - !atomic_read(&ctrl->pending_events)); + !atomic_read(&ctrl->pending_events) && + !ctrl->ist_running); return ctrl->request_result; case POWERON_STATE: ctrl_info(ctrl, "Slot(%s): Already in powering on state\n", @@ -408,7 +409,8 @@ int pciehp_sysfs_disable_slot(struct hotplug_slot *hotplug_slot) mutex_unlock(&ctrl->state_lock); pciehp_request(ctrl, DISABLE_SLOT); wait_event(ctrl->requester, - !atomic_read(&ctrl->pending_events)); + !atomic_read(&ctrl->pending_events) && + !ctrl->ist_running); return ctrl->request_result; case POWEROFF_STATE: ctrl_info(ctrl, "Slot(%s): Already in powering off state\n", diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 1a522c1c4177..86d97f3112f0 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -583,6 +583,7 @@ static irqreturn_t pciehp_ist(int irq, void *dev_id) irqreturn_t ret; u32 events; + ctrl->ist_running = true; pci_config_pm_runtime_get(pdev); /* rerun pciehp_isr() if the port was inaccessible on interrupt */ @@ -629,6 +630,7 @@ static irqreturn_t pciehp_ist(int irq, void *dev_id) up_read(&ctrl->reset_lock); pci_config_pm_runtime_put(pdev); + ctrl->ist_running = false; wake_up(&ctrl->requester); return IRQ_HANDLED; } From 7608158df3ed87a5c938c4a0b91f5b11101a9be1 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 7 Oct 2019 20:23:25 -0500 Subject: [PATCH 004/160] PCI: Fix missing bridge dma_ranges resource list cleanup Commit e80a91ad302b ("PCI: Add dma_ranges window list") added a dma_ranges resource list, but failed to correctly free the list when devm_pci_alloc_host_bridge() is used. Only the iproc host bridge driver is using the dma_ranges list. Fixes: e80a91ad302b ("PCI: Add dma_ranges window list") Link: https://lore.kernel.org/r/20191008012325.25700-1-robh@kernel.org Signed-off-by: Rob Herring Signed-off-by: Bjorn Helgaas Cc: Srinath Mannam --- drivers/pci/probe.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 3d5271a7a849..bdbc8490f962 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -572,6 +572,7 @@ static void devm_pci_release_host_bridge_dev(struct device *dev) bridge->release_fn(bridge); pci_free_resource_list(&bridge->windows); + pci_free_resource_list(&bridge->dma_ranges); } static void pci_release_host_bridge_dev(struct device *dev) From c9c13ba428ef90a9b408a6cdf874e14ab5754516 Mon Sep 17 00:00:00 2001 From: Denis Efremov Date: Sat, 28 Sep 2019 02:43:08 +0300 Subject: [PATCH 005/160] PCI: Add PCI_STD_NUM_BARS for the number of standard BARs Code that iterates over all standard PCI BARs typically uses PCI_STD_RESOURCE_END. However, that requires the unusual test "i <= PCI_STD_RESOURCE_END" rather than something the typical "i < PCI_STD_NUM_BARS". Add a definition for PCI_STD_NUM_BARS and change loops to use the more idiomatic C style to help avoid fencepost errors. Link: https://lore.kernel.org/r/20190927234026.23342-1-efremov@linux.com Link: https://lore.kernel.org/r/20190927234308.23935-1-efremov@linux.com Link: https://lore.kernel.org/r/20190916204158.6889-3-efremov@linux.com Signed-off-by: Denis Efremov Signed-off-by: Bjorn Helgaas Acked-by: Sebastian Ott # arch/s390/ Acked-by: Bartlomiej Zolnierkiewicz # video/fbdev/ Acked-by: Gustavo Pimentel # pci/controller/dwc/ Acked-by: Jack Wang # scsi/pm8001/ Acked-by: Martin K. Petersen # scsi/pm8001/ Acked-by: Ulf Hansson # memstick/ --- arch/alpha/kernel/pci-sysfs.c | 8 ++--- arch/s390/include/asm/pci.h | 5 +-- arch/s390/include/asm/pci_clp.h | 6 ++-- arch/s390/pci/pci.c | 16 +++++----- arch/s390/pci/pci_clp.c | 6 ++-- arch/x86/pci/common.c | 2 +- arch/x86/pci/intel_mid_pci.c | 2 +- drivers/ata/pata_atp867x.c | 2 +- drivers/ata/sata_nv.c | 2 +- drivers/memstick/host/jmb38x_ms.c | 2 +- drivers/misc/pci_endpoint_test.c | 8 ++--- drivers/net/ethernet/intel/e1000/e1000.h | 1 - drivers/net/ethernet/intel/e1000/e1000_main.c | 2 +- drivers/net/ethernet/intel/ixgb/ixgb.h | 1 - drivers/net/ethernet/intel/ixgb/ixgb_main.c | 2 +- .../net/ethernet/stmicro/stmmac/stmmac_pci.c | 4 +-- .../net/ethernet/synopsys/dwc-xlgmac-pci.c | 2 +- drivers/pci/controller/dwc/pci-dra7xx.c | 2 +- .../pci/controller/dwc/pci-layerscape-ep.c | 2 +- drivers/pci/controller/dwc/pcie-artpec6.c | 2 +- .../pci/controller/dwc/pcie-designware-plat.c | 2 +- drivers/pci/controller/dwc/pcie-designware.h | 2 +- drivers/pci/controller/pci-hyperv.c | 10 +++--- drivers/pci/endpoint/functions/pci-epf-test.c | 10 +++--- drivers/pci/pci-sysfs.c | 4 +-- drivers/pci/pci.c | 13 ++++---- drivers/pci/proc.c | 4 +-- drivers/pci/quirks.c | 4 +-- drivers/rapidio/devices/tsi721.c | 2 +- drivers/scsi/pm8001/pm8001_hwi.c | 2 +- drivers/scsi/pm8001/pm8001_init.c | 2 +- drivers/staging/gasket/gasket_constants.h | 3 -- drivers/staging/gasket/gasket_core.c | 12 +++---- drivers/staging/gasket/gasket_core.h | 4 +-- drivers/tty/serial/8250/8250_pci.c | 8 ++--- drivers/usb/core/hcd-pci.c | 2 +- drivers/usb/host/pci-quirks.c | 2 +- drivers/vfio/pci/vfio_pci.c | 11 ++++--- drivers/vfio/pci/vfio_pci_config.c | 32 ++++++++++--------- drivers/vfio/pci/vfio_pci_private.h | 4 +-- drivers/video/fbdev/core/fbmem.c | 4 +-- drivers/video/fbdev/efifb.c | 2 +- include/linux/pci-epc.h | 2 +- include/linux/pci.h | 2 +- include/uapi/linux/pci_regs.h | 1 + lib/devres.c | 2 +- 46 files changed, 110 insertions(+), 113 deletions(-) diff --git a/arch/alpha/kernel/pci-sysfs.c b/arch/alpha/kernel/pci-sysfs.c index f94c732fedeb..0021580d79ad 100644 --- a/arch/alpha/kernel/pci-sysfs.c +++ b/arch/alpha/kernel/pci-sysfs.c @@ -71,10 +71,10 @@ static int pci_mmap_resource(struct kobject *kobj, struct pci_bus_region bar; int i; - for (i = 0; i < PCI_ROM_RESOURCE; i++) + for (i = 0; i < PCI_STD_NUM_BARS; i++) if (res == &pdev->resource[i]) break; - if (i >= PCI_ROM_RESOURCE) + if (i >= PCI_STD_NUM_BARS) return -ENODEV; if (res->flags & IORESOURCE_MEM && iomem_is_exclusive(res->start)) @@ -115,7 +115,7 @@ void pci_remove_resource_files(struct pci_dev *pdev) { int i; - for (i = 0; i < PCI_ROM_RESOURCE; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { struct bin_attribute *res_attr; res_attr = pdev->res_attr[i]; @@ -232,7 +232,7 @@ int pci_create_resource_files(struct pci_dev *pdev) int retval; /* Expose the PCI resources from this device as files */ - for (i = 0; i < PCI_ROM_RESOURCE; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { /* skip empty resources */ if (!pci_resource_len(pdev, i)) diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h index a2399eff84ca..3a06c264ea53 100644 --- a/arch/s390/include/asm/pci.h +++ b/arch/s390/include/asm/pci.h @@ -2,9 +2,6 @@ #ifndef __ASM_S390_PCI_H #define __ASM_S390_PCI_H -/* must be set before including pci_clp.h */ -#define PCI_BAR_COUNT 6 - #include #include #include @@ -138,7 +135,7 @@ struct zpci_dev { char res_name[16]; bool mio_capable; - struct zpci_bar_struct bars[PCI_BAR_COUNT]; + struct zpci_bar_struct bars[PCI_STD_NUM_BARS]; u64 start_dma; /* Start of available DMA addresses */ u64 end_dma; /* End of available DMA addresses */ diff --git a/arch/s390/include/asm/pci_clp.h b/arch/s390/include/asm/pci_clp.h index 50359172cc48..bd2cb4ea7d93 100644 --- a/arch/s390/include/asm/pci_clp.h +++ b/arch/s390/include/asm/pci_clp.h @@ -77,7 +77,7 @@ struct mio_info { struct { u64 wb; u64 wt; - } addr[PCI_BAR_COUNT]; + } addr[PCI_STD_NUM_BARS]; u32 reserved[6]; } __packed; @@ -98,9 +98,9 @@ struct clp_rsp_query_pci { u16 util_str_avail : 1; /* utility string available? */ u16 pfgid : 8; /* pci function group id */ u32 fid; /* pci function id */ - u8 bar_size[PCI_BAR_COUNT]; + u8 bar_size[PCI_STD_NUM_BARS]; u16 pchid; - __le32 bar[PCI_BAR_COUNT]; + __le32 bar[PCI_STD_NUM_BARS]; u8 pfip[CLP_PFIP_NR_SEGMENTS]; /* pci function internal path */ u32 : 16; u8 fmb_len; diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c index c7fea9bea8cb..7b4c2acf05a8 100644 --- a/arch/s390/pci/pci.c +++ b/arch/s390/pci/pci.c @@ -43,7 +43,7 @@ static DECLARE_BITMAP(zpci_domain, ZPCI_NR_DEVICES); static DEFINE_SPINLOCK(zpci_domain_lock); #define ZPCI_IOMAP_ENTRIES \ - min(((unsigned long) ZPCI_NR_DEVICES * PCI_BAR_COUNT / 2), \ + min(((unsigned long) ZPCI_NR_DEVICES * PCI_STD_NUM_BARS / 2), \ ZPCI_IOMAP_MAX_ENTRIES) static DEFINE_SPINLOCK(zpci_iomap_lock); @@ -294,7 +294,7 @@ static void __iomem *pci_iomap_range_mio(struct pci_dev *pdev, int bar, void __iomem *pci_iomap_range(struct pci_dev *pdev, int bar, unsigned long offset, unsigned long max) { - if (!pci_resource_len(pdev, bar) || bar >= PCI_BAR_COUNT) + if (bar >= PCI_STD_NUM_BARS || !pci_resource_len(pdev, bar)) return NULL; if (static_branch_likely(&have_mio)) @@ -324,7 +324,7 @@ static void __iomem *pci_iomap_wc_range_mio(struct pci_dev *pdev, int bar, void __iomem *pci_iomap_wc_range(struct pci_dev *pdev, int bar, unsigned long offset, unsigned long max) { - if (!pci_resource_len(pdev, bar) || bar >= PCI_BAR_COUNT) + if (bar >= PCI_STD_NUM_BARS || !pci_resource_len(pdev, bar)) return NULL; if (static_branch_likely(&have_mio)) @@ -416,7 +416,7 @@ static void zpci_map_resources(struct pci_dev *pdev) resource_size_t len; int i; - for (i = 0; i < PCI_BAR_COUNT; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { len = pci_resource_len(pdev, i); if (!len) continue; @@ -451,7 +451,7 @@ static void zpci_unmap_resources(struct pci_dev *pdev) if (zpci_use_mio(zdev)) return; - for (i = 0; i < PCI_BAR_COUNT; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { len = pci_resource_len(pdev, i); if (!len) continue; @@ -514,7 +514,7 @@ static int zpci_setup_bus_resources(struct zpci_dev *zdev, snprintf(zdev->res_name, sizeof(zdev->res_name), "PCI Bus %04x:%02x", zdev->domain, ZPCI_BUS_NR); - for (i = 0; i < PCI_BAR_COUNT; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (!zdev->bars[i].size) continue; entry = zpci_alloc_iomap(zdev); @@ -551,7 +551,7 @@ static void zpci_cleanup_bus_resources(struct zpci_dev *zdev) { int i; - for (i = 0; i < PCI_BAR_COUNT; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (!zdev->bars[i].size || !zdev->bars[i].res) continue; @@ -573,7 +573,7 @@ int pcibios_add_device(struct pci_dev *pdev) pdev->dev.dma_ops = &s390_pci_dma_ops; zpci_map_resources(pdev); - for (i = 0; i < PCI_BAR_COUNT; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { res = &pdev->resource[i]; if (res->parent || !res->flags) continue; diff --git a/arch/s390/pci/pci_clp.c b/arch/s390/pci/pci_clp.c index 9bdff4defef1..8b729b5f2972 100644 --- a/arch/s390/pci/pci_clp.c +++ b/arch/s390/pci/pci_clp.c @@ -145,7 +145,7 @@ static int clp_store_query_pci_fn(struct zpci_dev *zdev, { int i; - for (i = 0; i < PCI_BAR_COUNT; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { zdev->bars[i].val = le32_to_cpu(response->bar[i]); zdev->bars[i].size = response->bar_size[i]; } @@ -164,8 +164,8 @@ static int clp_store_query_pci_fn(struct zpci_dev *zdev, sizeof(zdev->util_str)); } zdev->mio_capable = response->mio_addr_avail; - for (i = 0; i < PCI_BAR_COUNT; i++) { - if (!(response->mio.valid & (1 << (PCI_BAR_COUNT - i - 1)))) + for (i = 0; i < PCI_STD_NUM_BARS; i++) { + if (!(response->mio.valid & (1 << (PCI_STD_NUM_BARS - i - 1)))) continue; zdev->bars[i].mio_wb = (void __iomem *) response->mio.addr[i].wb; diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c index 9acab6ac28f5..1e59df041456 100644 --- a/arch/x86/pci/common.c +++ b/arch/x86/pci/common.c @@ -135,7 +135,7 @@ static void pcibios_fixup_device_resources(struct pci_dev *dev) * resource so the kernel doesn't attempt to assign * it later on in pci_assign_unassigned_resources */ - for (bar = 0; bar <= PCI_STD_RESOURCE_END; bar++) { + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) { bar_r = &dev->resource[bar]; if (bar_r->start == 0 && bar_r->end != 0) { bar_r->flags = 0; diff --git a/arch/x86/pci/intel_mid_pci.c b/arch/x86/pci/intel_mid_pci.c index 43867bc85368..00c62115f39c 100644 --- a/arch/x86/pci/intel_mid_pci.c +++ b/arch/x86/pci/intel_mid_pci.c @@ -382,7 +382,7 @@ static void pci_fixed_bar_fixup(struct pci_dev *dev) PCI_DEVFN(2, 2) == dev->devfn) return; - for (i = 0; i < PCI_ROM_RESOURCE; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { pci_read_config_dword(dev, offset + 8 + (i * 4), &size); dev->resource[i].end = dev->resource[i].start + size - 1; dev->resource[i].flags |= IORESOURCE_PCI_FIXED; diff --git a/drivers/ata/pata_atp867x.c b/drivers/ata/pata_atp867x.c index cfd0cf2cbca6..e01a3a6e4d46 100644 --- a/drivers/ata/pata_atp867x.c +++ b/drivers/ata/pata_atp867x.c @@ -422,7 +422,7 @@ static int atp867x_ata_pci_sff_init_host(struct ata_host *host) #ifdef ATP867X_DEBUG atp867x_check_res(pdev); - for (i = 0; i < PCI_ROM_RESOURCE; i++) + for (i = 0; i < PCI_STD_NUM_BARS; i++) printk(KERN_DEBUG "ATP867X: iomap[%d]=0x%llx\n", i, (unsigned long long)(host->iomap[i])); #endif diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c index 56946012d113..6f261fbae8c2 100644 --- a/drivers/ata/sata_nv.c +++ b/drivers/ata/sata_nv.c @@ -2325,7 +2325,7 @@ static int nv_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) // Make sure this is a SATA controller by counting the number of bars // (NVIDIA SATA controllers will always have six bars). Otherwise, // it's an IDE controller and we ignore it. - for (bar = 0; bar < 6; bar++) + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) if (pci_resource_start(pdev, bar) == 0) return -ENODEV; diff --git a/drivers/memstick/host/jmb38x_ms.c b/drivers/memstick/host/jmb38x_ms.c index 32747425297d..fd281c1d39b1 100644 --- a/drivers/memstick/host/jmb38x_ms.c +++ b/drivers/memstick/host/jmb38x_ms.c @@ -848,7 +848,7 @@ static int jmb38x_ms_count_slots(struct pci_dev *pdev) { int cnt, rc = 0; - for (cnt = 0; cnt < PCI_ROM_RESOURCE; ++cnt) { + for (cnt = 0; cnt < PCI_STD_NUM_BARS; ++cnt) { if (!(IORESOURCE_MEM & pci_resource_flags(pdev, cnt))) break; diff --git a/drivers/misc/pci_endpoint_test.c b/drivers/misc/pci_endpoint_test.c index 6e208a060a58..a5e317073d95 100644 --- a/drivers/misc/pci_endpoint_test.c +++ b/drivers/misc/pci_endpoint_test.c @@ -94,7 +94,7 @@ enum pci_barno { struct pci_endpoint_test { struct pci_dev *pdev; void __iomem *base; - void __iomem *bar[6]; + void __iomem *bar[PCI_STD_NUM_BARS]; struct completion irq_raised; int last_irq; int num_irqs; @@ -687,7 +687,7 @@ static int pci_endpoint_test_probe(struct pci_dev *pdev, if (!pci_endpoint_test_request_irq(test)) goto err_disable_irq; - for (bar = BAR_0; bar <= BAR_5; bar++) { + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) { if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) { base = pci_ioremap_bar(pdev, bar); if (!base) { @@ -740,7 +740,7 @@ err_ida_remove: ida_simple_remove(&pci_endpoint_test_ida, id); err_iounmap: - for (bar = BAR_0; bar <= BAR_5; bar++) { + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) { if (test->bar[bar]) pci_iounmap(pdev, test->bar[bar]); } @@ -771,7 +771,7 @@ static void pci_endpoint_test_remove(struct pci_dev *pdev) misc_deregister(&test->miscdev); kfree(misc_device->name); ida_simple_remove(&pci_endpoint_test_ida, id); - for (bar = BAR_0; bar <= BAR_5; bar++) { + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) { if (test->bar[bar]) pci_iounmap(pdev, test->bar[bar]); } diff --git a/drivers/net/ethernet/intel/e1000/e1000.h b/drivers/net/ethernet/intel/e1000/e1000.h index c40729b2c184..7fad2f24dcad 100644 --- a/drivers/net/ethernet/intel/e1000/e1000.h +++ b/drivers/net/ethernet/intel/e1000/e1000.h @@ -45,7 +45,6 @@ #define BAR_0 0 #define BAR_1 1 -#define BAR_5 5 #define INTEL_E1000_ETHERNET_DEVICE(device_id) {\ PCI_DEVICE(PCI_VENDOR_ID_INTEL, device_id)} diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c index 86493fea56e4..78db33694494 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_main.c +++ b/drivers/net/ethernet/intel/e1000/e1000_main.c @@ -977,7 +977,7 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_ioremap; if (adapter->need_ioport) { - for (i = BAR_1; i <= BAR_5; i++) { + for (i = BAR_1; i < PCI_STD_NUM_BARS; i++) { if (pci_resource_len(pdev, i) == 0) continue; if (pci_resource_flags(pdev, i) & IORESOURCE_IO) { diff --git a/drivers/net/ethernet/intel/ixgb/ixgb.h b/drivers/net/ethernet/intel/ixgb/ixgb.h index e85271b68410..681d44cc9784 100644 --- a/drivers/net/ethernet/intel/ixgb/ixgb.h +++ b/drivers/net/ethernet/intel/ixgb/ixgb.h @@ -42,7 +42,6 @@ #define BAR_0 0 #define BAR_1 1 -#define BAR_5 5 struct ixgb_adapter; #include "ixgb_hw.h" diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_main.c b/drivers/net/ethernet/intel/ixgb/ixgb_main.c index 0940a0da16f2..3d8c051dd327 100644 --- a/drivers/net/ethernet/intel/ixgb/ixgb_main.c +++ b/drivers/net/ethernet/intel/ixgb/ixgb_main.c @@ -412,7 +412,7 @@ ixgb_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_ioremap; } - for (i = BAR_1; i <= BAR_5; i++) { + for (i = BAR_1; i < PCI_STD_NUM_BARS; i++) { if (pci_resource_len(pdev, i) == 0) continue; if (pci_resource_flags(pdev, i) & IORESOURCE_IO) { diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 292045f4581f..8237dbc3e991 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -489,7 +489,7 @@ static int stmmac_pci_probe(struct pci_dev *pdev, } /* Get the base address of device */ - for (i = 0; i <= PCI_STD_RESOURCE_END; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (pci_resource_len(pdev, i) == 0) continue; ret = pcim_iomap_regions(pdev, BIT(i), pci_name(pdev)); @@ -532,7 +532,7 @@ static void stmmac_pci_remove(struct pci_dev *pdev) if (priv->plat->stmmac_clk) clk_unregister_fixed_rate(priv->plat->stmmac_clk); - for (i = 0; i <= PCI_STD_RESOURCE_END; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (pci_resource_len(pdev, i) == 0) continue; pcim_iounmap_regions(pdev, BIT(i)); diff --git a/drivers/net/ethernet/synopsys/dwc-xlgmac-pci.c b/drivers/net/ethernet/synopsys/dwc-xlgmac-pci.c index 386bafe74c3f..fa8604d7b797 100644 --- a/drivers/net/ethernet/synopsys/dwc-xlgmac-pci.c +++ b/drivers/net/ethernet/synopsys/dwc-xlgmac-pci.c @@ -34,7 +34,7 @@ static int xlgmac_probe(struct pci_dev *pcidev, const struct pci_device_id *id) return ret; } - for (i = 0; i <= PCI_STD_RESOURCE_END; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (pci_resource_len(pcidev, i) == 0) continue; ret = pcim_iomap_regions(pcidev, BIT(i), XLGMAC_DRV_NAME); diff --git a/drivers/pci/controller/dwc/pci-dra7xx.c b/drivers/pci/controller/dwc/pci-dra7xx.c index 4234ddb4722f..b20651cea09f 100644 --- a/drivers/pci/controller/dwc/pci-dra7xx.c +++ b/drivers/pci/controller/dwc/pci-dra7xx.c @@ -353,7 +353,7 @@ static void dra7xx_pcie_ep_init(struct dw_pcie_ep *ep) struct dra7xx_pcie *dra7xx = to_dra7xx_pcie(pci); enum pci_barno bar; - for (bar = BAR_0; bar <= BAR_5; bar++) + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) dw_pcie_ep_reset_bar(pci, bar); dra7xx_pcie_enable_wrapper_interrupts(dra7xx); diff --git a/drivers/pci/controller/dwc/pci-layerscape-ep.c b/drivers/pci/controller/dwc/pci-layerscape-ep.c index ca9aa4501e7e..0d151cead1b7 100644 --- a/drivers/pci/controller/dwc/pci-layerscape-ep.c +++ b/drivers/pci/controller/dwc/pci-layerscape-ep.c @@ -58,7 +58,7 @@ static void ls_pcie_ep_init(struct dw_pcie_ep *ep) struct dw_pcie *pci = to_dw_pcie_from_ep(ep); enum pci_barno bar; - for (bar = BAR_0; bar <= BAR_5; bar++) + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) dw_pcie_ep_reset_bar(pci, bar); } diff --git a/drivers/pci/controller/dwc/pcie-artpec6.c b/drivers/pci/controller/dwc/pcie-artpec6.c index d00252bd8fae..9e2482bd7b6d 100644 --- a/drivers/pci/controller/dwc/pcie-artpec6.c +++ b/drivers/pci/controller/dwc/pcie-artpec6.c @@ -422,7 +422,7 @@ static void artpec6_pcie_ep_init(struct dw_pcie_ep *ep) artpec6_pcie_wait_for_phy(artpec6_pcie); artpec6_pcie_set_nfts(artpec6_pcie); - for (bar = BAR_0; bar <= BAR_5; bar++) + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) dw_pcie_ep_reset_bar(pci, bar); } diff --git a/drivers/pci/controller/dwc/pcie-designware-plat.c b/drivers/pci/controller/dwc/pcie-designware-plat.c index b58fdcbc664b..73646b677aff 100644 --- a/drivers/pci/controller/dwc/pcie-designware-plat.c +++ b/drivers/pci/controller/dwc/pcie-designware-plat.c @@ -70,7 +70,7 @@ static void dw_plat_pcie_ep_init(struct dw_pcie_ep *ep) struct dw_pcie *pci = to_dw_pcie_from_ep(ep); enum pci_barno bar; - for (bar = BAR_0; bar <= BAR_5; bar++) + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) dw_pcie_ep_reset_bar(pci, bar); } diff --git a/drivers/pci/controller/dwc/pcie-designware.h b/drivers/pci/controller/dwc/pcie-designware.h index 5a18e94e52c8..5accdd6bc388 100644 --- a/drivers/pci/controller/dwc/pcie-designware.h +++ b/drivers/pci/controller/dwc/pcie-designware.h @@ -214,7 +214,7 @@ struct dw_pcie_ep { phys_addr_t phys_base; size_t addr_size; size_t page_size; - u8 bar_to_atu[6]; + u8 bar_to_atu[PCI_STD_NUM_BARS]; phys_addr_t *outbound_addr; unsigned long *ib_window_map; unsigned long *ob_window_map; diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c index f1f300218fab..ad40e848d564 100644 --- a/drivers/pci/controller/pci-hyperv.c +++ b/drivers/pci/controller/pci-hyperv.c @@ -307,7 +307,7 @@ struct pci_bus_relations { struct pci_q_res_req_response { struct vmpacket_descriptor hdr; s32 status; /* negative values are failures */ - u32 probed_bar[6]; + u32 probed_bar[PCI_STD_NUM_BARS]; } __packed; struct pci_set_power { @@ -539,7 +539,7 @@ struct hv_pci_dev { * What would be observed if one wrote 0xFFFFFFFF to a BAR and then * read it back, for each of the BAR offsets within config space. */ - u32 probed_bar[6]; + u32 probed_bar[PCI_STD_NUM_BARS]; }; struct hv_pci_compl { @@ -1610,7 +1610,7 @@ static void survey_child_resources(struct hv_pcibus_device *hbus) * so it's sufficient to just add them up without tracking alignment. */ list_for_each_entry(hpdev, &hbus->children, list_entry) { - for (i = 0; i < 6; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (hpdev->probed_bar[i] & PCI_BASE_ADDRESS_SPACE_IO) dev_err(&hbus->hdev->device, "There's an I/O BAR in this list!\n"); @@ -1684,7 +1684,7 @@ static void prepopulate_bars(struct hv_pcibus_device *hbus) /* Pick addresses for the BARs. */ do { list_for_each_entry(hpdev, &hbus->children, list_entry) { - for (i = 0; i < 6; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { bar_val = hpdev->probed_bar[i]; if (bar_val == 0) continue; @@ -1841,7 +1841,7 @@ static void q_resource_requirements(void *context, struct pci_response *resp, "query resource requirements failed: %x\n", resp->status); } else { - for (i = 0; i < 6; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { completion->hpdev->probed_bar[i] = q_res_req->probed_bar[i]; } diff --git a/drivers/pci/endpoint/functions/pci-epf-test.c b/drivers/pci/endpoint/functions/pci-epf-test.c index 1cfe3687a211..5d74f81ddfe4 100644 --- a/drivers/pci/endpoint/functions/pci-epf-test.c +++ b/drivers/pci/endpoint/functions/pci-epf-test.c @@ -44,7 +44,7 @@ static struct workqueue_struct *kpcitest_workqueue; struct pci_epf_test { - void *reg[6]; + void *reg[PCI_STD_NUM_BARS]; struct pci_epf *epf; enum pci_barno test_reg_bar; struct delayed_work cmd_handler; @@ -377,7 +377,7 @@ static void pci_epf_test_unbind(struct pci_epf *epf) cancel_delayed_work(&epf_test->cmd_handler); pci_epc_stop(epc); - for (bar = BAR_0; bar <= BAR_5; bar++) { + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) { epf_bar = &epf->bar[bar]; if (epf_test->reg[bar]) { @@ -400,7 +400,7 @@ static int pci_epf_test_set_bar(struct pci_epf *epf) epc_features = epf_test->epc_features; - for (bar = BAR_0; bar <= BAR_5; bar += add) { + for (bar = 0; bar < PCI_STD_NUM_BARS; bar += add) { epf_bar = &epf->bar[bar]; /* * pci_epc_set_bar() sets PCI_BASE_ADDRESS_MEM_TYPE_64 @@ -450,7 +450,7 @@ static int pci_epf_test_alloc_space(struct pci_epf *epf) } epf_test->reg[test_reg_bar] = base; - for (bar = BAR_0; bar <= BAR_5; bar += add) { + for (bar = 0; bar < PCI_STD_NUM_BARS; bar += add) { epf_bar = &epf->bar[bar]; add = (epf_bar->flags & PCI_BASE_ADDRESS_MEM_TYPE_64) ? 2 : 1; @@ -478,7 +478,7 @@ static void pci_epf_configure_bar(struct pci_epf *epf, bool bar_fixed_64bit; int i; - for (i = BAR_0; i <= BAR_5; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { epf_bar = &epf->bar[i]; bar_fixed_64bit = !!(epc_features->bar_fixed_64bit & (1 << i)); if (bar_fixed_64bit) diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 793412954529..07bd84c50238 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -1122,7 +1122,7 @@ static void pci_remove_resource_files(struct pci_dev *pdev) { int i; - for (i = 0; i < PCI_ROM_RESOURCE; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { struct bin_attribute *res_attr; res_attr = pdev->res_attr[i]; @@ -1193,7 +1193,7 @@ static int pci_create_resource_files(struct pci_dev *pdev) int retval; /* Expose the PCI resources from this device as files */ - for (i = 0; i < PCI_ROM_RESOURCE; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { /* skip empty resources */ if (!pci_resource_len(pdev, i)) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index e7982af9a5d8..184765f12848 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -674,7 +674,7 @@ struct resource *pci_find_resource(struct pci_dev *dev, struct resource *res) { int i; - for (i = 0; i < PCI_ROM_RESOURCE; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { struct resource *r = &dev->resource[i]; if (r->start && resource_contains(r, res)) @@ -3768,7 +3768,7 @@ void pci_release_selected_regions(struct pci_dev *pdev, int bars) { int i; - for (i = 0; i < 6; i++) + for (i = 0; i < PCI_STD_NUM_BARS; i++) if (bars & (1 << i)) pci_release_region(pdev, i); } @@ -3779,7 +3779,7 @@ static int __pci_request_selected_regions(struct pci_dev *pdev, int bars, { int i; - for (i = 0; i < 6; i++) + for (i = 0; i < PCI_STD_NUM_BARS; i++) if (bars & (1 << i)) if (__pci_request_region(pdev, i, res_name, excl)) goto err_out; @@ -3827,7 +3827,7 @@ EXPORT_SYMBOL(pci_request_selected_regions_exclusive); void pci_release_regions(struct pci_dev *pdev) { - pci_release_selected_regions(pdev, (1 << 6) - 1); + pci_release_selected_regions(pdev, (1 << PCI_STD_NUM_BARS) - 1); } EXPORT_SYMBOL(pci_release_regions); @@ -3846,7 +3846,8 @@ EXPORT_SYMBOL(pci_release_regions); */ int pci_request_regions(struct pci_dev *pdev, const char *res_name) { - return pci_request_selected_regions(pdev, ((1 << 6) - 1), res_name); + return pci_request_selected_regions(pdev, + ((1 << PCI_STD_NUM_BARS) - 1), res_name); } EXPORT_SYMBOL(pci_request_regions); @@ -3868,7 +3869,7 @@ EXPORT_SYMBOL(pci_request_regions); int pci_request_regions_exclusive(struct pci_dev *pdev, const char *res_name) { return pci_request_selected_regions_exclusive(pdev, - ((1 << 6) - 1), res_name); + ((1 << PCI_STD_NUM_BARS) - 1), res_name); } EXPORT_SYMBOL(pci_request_regions_exclusive); diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c index 5495537c60c2..6ef74bf5013f 100644 --- a/drivers/pci/proc.c +++ b/drivers/pci/proc.c @@ -258,13 +258,13 @@ static int proc_bus_pci_mmap(struct file *file, struct vm_area_struct *vma) } /* Make sure the caller is mapping a real resource for this device */ - for (i = 0; i < PCI_ROM_RESOURCE; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (dev->resource[i].flags & res_bit && pci_mmap_fits(dev, i, vma, PCI_MMAP_PROCFS)) break; } - if (i >= PCI_ROM_RESOURCE) + if (i >= PCI_STD_NUM_BARS) return -ENODEV; if (fpriv->mmap_state == pci_mmap_mem && diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 320255e5e8f8..5d5e6f2f7837 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -474,7 +474,7 @@ static void quirk_extend_bar_to_page(struct pci_dev *dev) { int i; - for (i = 0; i <= PCI_STD_RESOURCE_END; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { struct resource *r = &dev->resource[i]; if (r->flags & IORESOURCE_MEM && resource_size(r) < PAGE_SIZE) { @@ -1809,7 +1809,7 @@ static void quirk_alder_ioapic(struct pci_dev *pdev) * The next five BARs all seem to be rubbish, so just clean * them out. */ - for (i = 1; i < 6; i++) + for (i = 1; i < PCI_STD_NUM_BARS; i++) memset(&pdev->resource[i], 0, sizeof(pdev->resource[i])); } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_EESSC, quirk_alder_ioapic); diff --git a/drivers/rapidio/devices/tsi721.c b/drivers/rapidio/devices/tsi721.c index 125a173bed45..4dd31dd9feea 100644 --- a/drivers/rapidio/devices/tsi721.c +++ b/drivers/rapidio/devices/tsi721.c @@ -2755,7 +2755,7 @@ static int tsi721_probe(struct pci_dev *pdev, { int i; - for (i = 0; i <= PCI_STD_RESOURCE_END; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { tsi_debug(INIT, &pdev->dev, "res%d %pR", i, &pdev->resource[i]); } diff --git a/drivers/scsi/pm8001/pm8001_hwi.c b/drivers/scsi/pm8001/pm8001_hwi.c index 68a8217032d0..1a3661d6be06 100644 --- a/drivers/scsi/pm8001/pm8001_hwi.c +++ b/drivers/scsi/pm8001/pm8001_hwi.c @@ -1186,7 +1186,7 @@ static void pm8001_hw_chip_rst(struct pm8001_hba_info *pm8001_ha) void pm8001_chip_iounmap(struct pm8001_hba_info *pm8001_ha) { s8 bar, logical = 0; - for (bar = 0; bar < 6; bar++) { + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) { /* ** logical BARs for SPC: ** bar 0 and 1 - logical BAR0 diff --git a/drivers/scsi/pm8001/pm8001_init.c b/drivers/scsi/pm8001/pm8001_init.c index 3374f553c617..aca913490eb5 100644 --- a/drivers/scsi/pm8001/pm8001_init.c +++ b/drivers/scsi/pm8001/pm8001_init.c @@ -401,7 +401,7 @@ static int pm8001_ioremap(struct pm8001_hba_info *pm8001_ha) pdev = pm8001_ha->pdev; /* map pci mem (PMC pci base 0-3)*/ - for (bar = 0; bar < 6; bar++) { + for (bar = 0; bar < PCI_STD_NUM_BARS; bar++) { /* ** logical BARs for SPC: ** bar 0 and 1 - logical BAR0 diff --git a/drivers/staging/gasket/gasket_constants.h b/drivers/staging/gasket/gasket_constants.h index 50d87c7b178c..9ea9c8833f27 100644 --- a/drivers/staging/gasket/gasket_constants.h +++ b/drivers/staging/gasket/gasket_constants.h @@ -13,9 +13,6 @@ /* The maximum devices per each type. */ #define GASKET_DEV_MAX 256 -/* The number of supported (and possible) PCI BARs. */ -#define GASKET_NUM_BARS 6 - /* The number of supported Gasket page tables per device. */ #define GASKET_MAX_NUM_PAGE_TABLES 1 diff --git a/drivers/staging/gasket/gasket_core.c b/drivers/staging/gasket/gasket_core.c index 13179f063a61..cd8be80d2076 100644 --- a/drivers/staging/gasket/gasket_core.c +++ b/drivers/staging/gasket/gasket_core.c @@ -371,7 +371,7 @@ static int gasket_setup_pci(struct pci_dev *pci_dev, { int i, mapped_bars, ret; - for (i = 0; i < GASKET_NUM_BARS; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { ret = gasket_map_pci_bar(gasket_dev, i); if (ret) { mapped_bars = i; @@ -393,7 +393,7 @@ static void gasket_cleanup_pci(struct gasket_dev *gasket_dev) { int i; - for (i = 0; i < GASKET_NUM_BARS; i++) + for (i = 0; i < PCI_STD_NUM_BARS; i++) gasket_unmap_pci_bar(gasket_dev, i); } @@ -493,7 +493,7 @@ static ssize_t gasket_sysfs_data_show(struct device *device, (enum gasket_sysfs_attribute_type)gasket_attr->data.attr_type; switch (sysfs_type) { case ATTR_BAR_OFFSETS: - for (i = 0; i < GASKET_NUM_BARS; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { bar_desc = &driver_desc->bar_descriptions[i]; if (bar_desc->size == 0) continue; @@ -505,7 +505,7 @@ static ssize_t gasket_sysfs_data_show(struct device *device, } break; case ATTR_BAR_SIZES: - for (i = 0; i < GASKET_NUM_BARS; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { bar_desc = &driver_desc->bar_descriptions[i]; if (bar_desc->size == 0) continue; @@ -556,7 +556,7 @@ static ssize_t gasket_sysfs_data_show(struct device *device, ret = snprintf(buf, PAGE_SIZE, "%d\n", gasket_dev->reset_count); break; case ATTR_USER_MEM_RANGES: - for (i = 0; i < GASKET_NUM_BARS; ++i) { + for (i = 0; i < PCI_STD_NUM_BARS; ++i) { current_written = gasket_write_mappable_regions(buf, driver_desc, i); @@ -736,7 +736,7 @@ static int gasket_get_bar_index(const struct gasket_dev *gasket_dev, const struct gasket_driver_desc *driver_desc; driver_desc = gasket_dev->internal_desc->driver_desc; - for (i = 0; i < GASKET_NUM_BARS; ++i) { + for (i = 0; i < PCI_STD_NUM_BARS; ++i) { struct gasket_bar_desc bar_desc = driver_desc->bar_descriptions[i]; diff --git a/drivers/staging/gasket/gasket_core.h b/drivers/staging/gasket/gasket_core.h index be44ac1e3118..c417acadb0d5 100644 --- a/drivers/staging/gasket/gasket_core.h +++ b/drivers/staging/gasket/gasket_core.h @@ -268,7 +268,7 @@ struct gasket_dev { char kobj_name[GASKET_NAME_MAX]; /* Virtual address of mapped BAR memory range. */ - struct gasket_bar_data bar_data[GASKET_NUM_BARS]; + struct gasket_bar_data bar_data[PCI_STD_NUM_BARS]; /* Coherent buffer. */ struct gasket_coherent_buffer coherent_buffer; @@ -369,7 +369,7 @@ struct gasket_driver_desc { /* Set of 6 bar descriptions that describe all PCIe bars. * Note that BUS/AXI devices (i.e. non PCI devices) use those. */ - struct gasket_bar_desc bar_descriptions[GASKET_NUM_BARS]; + struct gasket_bar_desc bar_descriptions[PCI_STD_NUM_BARS]; /* * Coherent buffer description. diff --git a/drivers/tty/serial/8250/8250_pci.c b/drivers/tty/serial/8250/8250_pci.c index 6adbadd6a56a..0b8784cd6d71 100644 --- a/drivers/tty/serial/8250/8250_pci.c +++ b/drivers/tty/serial/8250/8250_pci.c @@ -48,8 +48,6 @@ struct f815xxa_data { int idx; }; -#define PCI_NUM_BAR_RESOURCES 6 - struct serial_private { struct pci_dev *dev; unsigned int nr; @@ -89,7 +87,7 @@ setup_port(struct serial_private *priv, struct uart_8250_port *port, { struct pci_dev *dev = priv->dev; - if (bar >= PCI_NUM_BAR_RESOURCES) + if (bar >= PCI_STD_NUM_BARS) return -EINVAL; if (pci_resource_flags(dev, bar) & IORESOURCE_MEM) { @@ -4060,7 +4058,7 @@ serial_pci_guess_board(struct pci_dev *dev, struct pciserial_board *board) return -ENODEV; num_iomem = num_port = 0; - for (i = 0; i < PCI_NUM_BAR_RESOURCES; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (pci_resource_flags(dev, i) & IORESOURCE_IO) { num_port++; if (first_port == -1) @@ -4088,7 +4086,7 @@ serial_pci_guess_board(struct pci_dev *dev, struct pciserial_board *board) */ first_port = -1; num_port = 0; - for (i = 0; i < PCI_NUM_BAR_RESOURCES; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (pci_resource_flags(dev, i) & IORESOURCE_IO && pci_resource_len(dev, i) == 8 && (first_port == -1 || (first_port + num_port) == i)) { diff --git a/drivers/usb/core/hcd-pci.c b/drivers/usb/core/hcd-pci.c index 9e26b0143a59..9ae2a7a93df2 100644 --- a/drivers/usb/core/hcd-pci.c +++ b/drivers/usb/core/hcd-pci.c @@ -234,7 +234,7 @@ int usb_hcd_pci_probe(struct pci_dev *dev, const struct pci_device_id *id) /* UHCI */ int region; - for (region = 0; region < PCI_ROM_RESOURCE; region++) { + for (region = 0; region < PCI_STD_NUM_BARS; region++) { if (!(pci_resource_flags(dev, region) & IORESOURCE_IO)) continue; diff --git a/drivers/usb/host/pci-quirks.c b/drivers/usb/host/pci-quirks.c index f6d04491df60..6c7f0a876b96 100644 --- a/drivers/usb/host/pci-quirks.c +++ b/drivers/usb/host/pci-quirks.c @@ -728,7 +728,7 @@ static void quirk_usb_handoff_uhci(struct pci_dev *pdev) if (!pio_enabled(pdev)) return; - for (i = 0; i < PCI_ROM_RESOURCE; i++) + for (i = 0; i < PCI_STD_NUM_BARS; i++) if ((pci_resource_flags(pdev, i) & IORESOURCE_IO)) { base = pci_resource_start(pdev, i); break; diff --git a/drivers/vfio/pci/vfio_pci.c b/drivers/vfio/pci/vfio_pci.c index 02206162eaa9..379a02c36e37 100644 --- a/drivers/vfio/pci/vfio_pci.c +++ b/drivers/vfio/pci/vfio_pci.c @@ -110,13 +110,15 @@ static inline bool vfio_pci_is_vga(struct pci_dev *pdev) static void vfio_pci_probe_mmaps(struct vfio_pci_device *vdev) { struct resource *res; - int bar; + int i; struct vfio_pci_dummy_resource *dummy_res; INIT_LIST_HEAD(&vdev->dummy_resources_list); - for (bar = PCI_STD_RESOURCES; bar <= PCI_STD_RESOURCE_END; bar++) { - res = vdev->pdev->resource + bar; + for (i = 0; i < PCI_STD_NUM_BARS; i++) { + int bar = i + PCI_STD_RESOURCES; + + res = &vdev->pdev->resource[bar]; if (!IS_ENABLED(CONFIG_VFIO_PCI_MMAP)) goto no_mmap; @@ -399,7 +401,8 @@ static void vfio_pci_disable(struct vfio_pci_device *vdev) vfio_config_free(vdev); - for (bar = PCI_STD_RESOURCES; bar <= PCI_STD_RESOURCE_END; bar++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { + bar = i + PCI_STD_RESOURCES; if (!vdev->barmap[bar]) continue; pci_iounmap(pdev, vdev->barmap[bar]); diff --git a/drivers/vfio/pci/vfio_pci_config.c b/drivers/vfio/pci/vfio_pci_config.c index f0891bd8444c..90c0b80f8acf 100644 --- a/drivers/vfio/pci/vfio_pci_config.c +++ b/drivers/vfio/pci/vfio_pci_config.c @@ -450,30 +450,32 @@ static void vfio_bar_fixup(struct vfio_pci_device *vdev) { struct pci_dev *pdev = vdev->pdev; int i; - __le32 *bar; + __le32 *vbar; u64 mask; - bar = (__le32 *)&vdev->vconfig[PCI_BASE_ADDRESS_0]; + vbar = (__le32 *)&vdev->vconfig[PCI_BASE_ADDRESS_0]; - for (i = PCI_STD_RESOURCES; i <= PCI_STD_RESOURCE_END; i++, bar++) { - if (!pci_resource_start(pdev, i)) { - *bar = 0; /* Unmapped by host = unimplemented to user */ + for (i = 0; i < PCI_STD_NUM_BARS; i++, vbar++) { + int bar = i + PCI_STD_RESOURCES; + + if (!pci_resource_start(pdev, bar)) { + *vbar = 0; /* Unmapped by host = unimplemented to user */ continue; } - mask = ~(pci_resource_len(pdev, i) - 1); + mask = ~(pci_resource_len(pdev, bar) - 1); - *bar &= cpu_to_le32((u32)mask); - *bar |= vfio_generate_bar_flags(pdev, i); + *vbar &= cpu_to_le32((u32)mask); + *vbar |= vfio_generate_bar_flags(pdev, bar); - if (*bar & cpu_to_le32(PCI_BASE_ADDRESS_MEM_TYPE_64)) { - bar++; - *bar &= cpu_to_le32((u32)(mask >> 32)); + if (*vbar & cpu_to_le32(PCI_BASE_ADDRESS_MEM_TYPE_64)) { + vbar++; + *vbar &= cpu_to_le32((u32)(mask >> 32)); i++; } } - bar = (__le32 *)&vdev->vconfig[PCI_ROM_ADDRESS]; + vbar = (__le32 *)&vdev->vconfig[PCI_ROM_ADDRESS]; /* * NB. REGION_INFO will have reported zero size if we weren't able @@ -483,14 +485,14 @@ static void vfio_bar_fixup(struct vfio_pci_device *vdev) if (pci_resource_start(pdev, PCI_ROM_RESOURCE)) { mask = ~(pci_resource_len(pdev, PCI_ROM_RESOURCE) - 1); mask |= PCI_ROM_ADDRESS_ENABLE; - *bar &= cpu_to_le32((u32)mask); + *vbar &= cpu_to_le32((u32)mask); } else if (pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW) { mask = ~(0x20000 - 1); mask |= PCI_ROM_ADDRESS_ENABLE; - *bar &= cpu_to_le32((u32)mask); + *vbar &= cpu_to_le32((u32)mask); } else - *bar = 0; + *vbar = 0; vdev->bardirty = false; } diff --git a/drivers/vfio/pci/vfio_pci_private.h b/drivers/vfio/pci/vfio_pci_private.h index ee6ee91718a4..8a2c7607d513 100644 --- a/drivers/vfio/pci/vfio_pci_private.h +++ b/drivers/vfio/pci/vfio_pci_private.h @@ -86,8 +86,8 @@ struct vfio_pci_reflck { struct vfio_pci_device { struct pci_dev *pdev; - void __iomem *barmap[PCI_STD_RESOURCE_END + 1]; - bool bar_mmap_supported[PCI_STD_RESOURCE_END + 1]; + void __iomem *barmap[PCI_STD_NUM_BARS]; + bool bar_mmap_supported[PCI_STD_NUM_BARS]; u8 *pci_config_map; u8 *vconfig; struct perm_bits *msi_perm; diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c index e6a1c805064f..19af5626a317 100644 --- a/drivers/video/fbdev/core/fbmem.c +++ b/drivers/video/fbdev/core/fbmem.c @@ -1774,7 +1774,7 @@ int remove_conflicting_pci_framebuffers(struct pci_dev *pdev, int res_id, const int err, idx, bar; bool res_id_found = false; - for (idx = 0, bar = 0; bar < PCI_ROM_RESOURCE; bar++) { + for (idx = 0, bar = 0; bar < PCI_STD_NUM_BARS; bar++) { if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM)) continue; idx++; @@ -1784,7 +1784,7 @@ int remove_conflicting_pci_framebuffers(struct pci_dev *pdev, int res_id, const if (!ap) return -ENOMEM; - for (idx = 0, bar = 0; bar < PCI_ROM_RESOURCE; bar++) { + for (idx = 0, bar = 0; bar < PCI_STD_NUM_BARS; bar++) { if (!(pci_resource_flags(pdev, bar) & IORESOURCE_MEM)) continue; ap->ranges[idx].base = pci_resource_start(pdev, bar); diff --git a/drivers/video/fbdev/efifb.c b/drivers/video/fbdev/efifb.c index 51d97ec4f58f..1caa3726cb45 100644 --- a/drivers/video/fbdev/efifb.c +++ b/drivers/video/fbdev/efifb.c @@ -653,7 +653,7 @@ static void efifb_fixup_resources(struct pci_dev *dev) if (!base) return; - for (i = 0; i <= PCI_STD_RESOURCE_END; i++) { + for (i = 0; i < PCI_STD_NUM_BARS; i++) { struct resource *res = &dev->resource[i]; if (!(res->flags & IORESOURCE_MEM)) diff --git a/include/linux/pci-epc.h b/include/linux/pci-epc.h index f641badc2c61..56f1846b9d39 100644 --- a/include/linux/pci-epc.h +++ b/include/linux/pci-epc.h @@ -117,7 +117,7 @@ struct pci_epc_features { unsigned int msix_capable : 1; u8 reserved_bar; u8 bar_fixed_64bit; - u64 bar_fixed_size[BAR_5 + 1]; + u64 bar_fixed_size[PCI_STD_NUM_BARS]; size_t align; }; diff --git a/include/linux/pci.h b/include/linux/pci.h index f9088c89a534..4cc739616148 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -82,7 +82,7 @@ enum pci_mmap_state { enum { /* #0-5: standard PCI resources */ PCI_STD_RESOURCES, - PCI_STD_RESOURCE_END = 5, + PCI_STD_RESOURCE_END = PCI_STD_RESOURCES + PCI_STD_NUM_BARS - 1, /* #6: expansion ROM resource */ PCI_ROM_RESOURCE, diff --git a/include/uapi/linux/pci_regs.h b/include/uapi/linux/pci_regs.h index 29d6e93fd15e..43cf74eba29d 100644 --- a/include/uapi/linux/pci_regs.h +++ b/include/uapi/linux/pci_regs.h @@ -34,6 +34,7 @@ * of which the first 64 bytes are standardized as follows: */ #define PCI_STD_HEADER_SIZEOF 64 +#define PCI_STD_NUM_BARS 6 /* Number of standard BARs */ #define PCI_VENDOR_ID 0x00 /* 16 bits */ #define PCI_DEVICE_ID 0x02 /* 16 bits */ #define PCI_COMMAND 0x04 /* 16 bits */ diff --git a/lib/devres.c b/lib/devres.c index 6a0e9bd6524a..ab75d73122b8 100644 --- a/lib/devres.c +++ b/lib/devres.c @@ -262,7 +262,7 @@ EXPORT_SYMBOL(devm_ioport_unmap); /* * PCI iomap devres */ -#define PCIM_IOMAP_MAX PCI_ROM_RESOURCE +#define PCIM_IOMAP_MAX PCI_STD_NUM_BARS struct pcim_iomap_devres { void __iomem *table[PCIM_IOMAP_MAX]; From 364b3f1ff8f096d45f042a9c85daf7a1fc78413e Mon Sep 17 00:00:00 2001 From: Remi Pommarel Date: Wed, 22 May 2019 23:33:51 +0200 Subject: [PATCH 006/160] PCI: aardvark: Use LTSSM state to build link training flag Aardvark's PCI_EXP_LNKSTA_LT flag in its link status register is not implemented and does not reflect the actual link training state (the flag is always set to 0). In order to support link re-training feature this flag has to be emulated. The Link Training and Status State Machine (LTSSM) flag in Aardvark LMI config register could be used as a link training indicator. Indeed if the LTSSM is in L0 or upper state then link training has completed (see [1]). Unfortunately because after asking a link retraining it takes a while for the LTSSM state to become less than 0x10 (due to L0s to recovery state transition delays), LTSSM can still be in L0 while link training has not finished yet. So this waits for link to be in recovery or lesser state before returning after asking for a link retrain. [1] "PCI Express Base Specification", REV. 4.0 PCI Express, February 19 2014, Table 4-14 Fixes: 8a3ebd8de328 ("PCI: aardvark: Implement emulated root PCI bridge config space") Tested-by: Marc Zyngier Signed-off-by: Remi Pommarel Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Acked-by: Thomas Petazzoni --- drivers/pci/controller/pci-aardvark.c | 29 ++++++++++++++++++++++++++- 1 file changed, 28 insertions(+), 1 deletion(-) diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c index fc0fe4d4de49..fe471861f801 100644 --- a/drivers/pci/controller/pci-aardvark.c +++ b/drivers/pci/controller/pci-aardvark.c @@ -180,6 +180,8 @@ #define LINK_WAIT_MAX_RETRIES 10 #define LINK_WAIT_USLEEP_MIN 90000 #define LINK_WAIT_USLEEP_MAX 100000 +#define RETRAIN_WAIT_MAX_RETRIES 10 +#define RETRAIN_WAIT_USLEEP_US 2000 #define MSI_IRQ_NUM 32 @@ -239,6 +241,17 @@ static int advk_pcie_wait_for_link(struct advk_pcie *pcie) return -ETIMEDOUT; } +static void advk_pcie_wait_for_retrain(struct advk_pcie *pcie) +{ + size_t retries; + + for (retries = 0; retries < RETRAIN_WAIT_MAX_RETRIES; ++retries) { + if (!advk_pcie_link_up(pcie)) + break; + udelay(RETRAIN_WAIT_USLEEP_US); + } +} + static void advk_pcie_setup_hw(struct advk_pcie *pcie) { u32 reg; @@ -426,11 +439,20 @@ advk_pci_bridge_emul_pcie_conf_read(struct pci_bridge_emul *bridge, return PCI_BRIDGE_EMUL_HANDLED; } + case PCI_EXP_LNKCTL: { + /* u32 contains both PCI_EXP_LNKCTL and PCI_EXP_LNKSTA */ + u32 val = advk_readl(pcie, PCIE_CORE_PCIEXP_CAP + reg) & + ~(PCI_EXP_LNKSTA_LT << 16); + if (!advk_pcie_link_up(pcie)) + val |= (PCI_EXP_LNKSTA_LT << 16); + *value = val; + return PCI_BRIDGE_EMUL_HANDLED; + } + case PCI_CAP_LIST_ID: case PCI_EXP_DEVCAP: case PCI_EXP_DEVCTL: case PCI_EXP_LNKCAP: - case PCI_EXP_LNKCTL: *value = advk_readl(pcie, PCIE_CORE_PCIEXP_CAP + reg); return PCI_BRIDGE_EMUL_HANDLED; default: @@ -447,8 +469,13 @@ advk_pci_bridge_emul_pcie_conf_write(struct pci_bridge_emul *bridge, switch (reg) { case PCI_EXP_DEVCTL: + advk_writel(pcie, new, PCIE_CORE_PCIEXP_CAP + reg); + break; + case PCI_EXP_LNKCTL: advk_writel(pcie, new, PCIE_CORE_PCIEXP_CAP + reg); + if (new & PCI_EXP_LNKCTL_RL) + advk_pcie_wait_for_retrain(pcie); break; case PCI_EXP_RTCTL: From f4c7d053d7f77cd5c1a1ba7c7ce085ddba13d1d7 Mon Sep 17 00:00:00 2001 From: Remi Pommarel Date: Wed, 22 May 2019 23:33:50 +0200 Subject: [PATCH 007/160] PCI: aardvark: Wait for endpoint to be ready before training link When configuring pcie reset pin from gpio (e.g. initially set by u-boot) to pcie function this pin goes low for a brief moment asserting the PERST# signal. Thus connected device enters fundamental reset process and link configuration can only begin after a minimal 100ms delay (see [1]). Because the pin configuration comes from the "default" pinctrl it is implicitly configured before the probe callback is called: driver_probe_device() really_probe() ... pinctrl_bind_pins() /* Here pin goes from gpio to PCIE reset function and PERST# is asserted */ ... drv->probe() [1] "PCI Express Base Specification", REV. 4.0 PCI Express, February 19 2014, 6.6.1 Conventional Reset Signed-off-by: Remi Pommarel Signed-off-by: Lorenzo Pieralisi Acked-by: Thomas Petazzoni --- drivers/pci/controller/pci-aardvark.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c index fe471861f801..78c39320def4 100644 --- a/drivers/pci/controller/pci-aardvark.c +++ b/drivers/pci/controller/pci-aardvark.c @@ -337,6 +337,14 @@ static void advk_pcie_setup_hw(struct advk_pcie *pcie) reg |= PIO_CTRL_ADDR_WIN_DISABLE; advk_writel(pcie, reg, PIO_CTRL); + /* + * PERST# signal could have been asserted by pinctrl subsystem before + * probe() callback has been called, making the endpoint going into + * fundamental reset. As required by PCI Express spec a delay for at + * least 100ms after such a reset before link training is needed. + */ + msleep(PCI_PM_D3COLD_WAIT); + /* Start link training */ reg = advk_readl(pcie, PCIE_CORE_LINK_CTRL_STAT_REG); reg |= PCIE_CORE_LINK_TRAINING; From c0f05a6ab52535c1bf5f43272eede3e11c5701a5 Mon Sep 17 00:00:00 2001 From: Remi Pommarel Date: Fri, 14 Jun 2019 12:10:59 +0200 Subject: [PATCH 008/160] PCI: aardvark: Fix PCI_EXP_RTCTL register configuration PCI_EXP_RTCTL is used to activate PME interrupt only, so writing into it should not modify other interrupts' mask. The ISR mask polarity was also inverted, when PCI_EXP_RTCTL_PMEIE is set PCIE_MSG_PM_PME_MASK mask bit should actually be cleared. Fixes: 8a3ebd8de328 ("PCI: aardvark: Implement emulated root PCI bridge config space") Signed-off-by: Remi Pommarel Signed-off-by: Lorenzo Pieralisi Acked-by: Thomas Petazzoni --- drivers/pci/controller/pci-aardvark.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c index 78c39320def4..cb4327657136 100644 --- a/drivers/pci/controller/pci-aardvark.c +++ b/drivers/pci/controller/pci-aardvark.c @@ -436,7 +436,7 @@ advk_pci_bridge_emul_pcie_conf_read(struct pci_bridge_emul *bridge, case PCI_EXP_RTCTL: { u32 val = advk_readl(pcie, PCIE_ISR0_MASK_REG); - *value = (val & PCIE_MSG_PM_PME_MASK) ? PCI_EXP_RTCTL_PMEIE : 0; + *value = (val & PCIE_MSG_PM_PME_MASK) ? 0 : PCI_EXP_RTCTL_PMEIE; return PCI_BRIDGE_EMUL_HANDLED; } @@ -486,10 +486,15 @@ advk_pci_bridge_emul_pcie_conf_write(struct pci_bridge_emul *bridge, advk_pcie_wait_for_retrain(pcie); break; - case PCI_EXP_RTCTL: - new = (new & PCI_EXP_RTCTL_PMEIE) << 3; - advk_writel(pcie, new, PCIE_ISR0_MASK_REG); + case PCI_EXP_RTCTL: { + /* Only mask/unmask PME interrupt */ + u32 val = advk_readl(pcie, PCIE_ISR0_MASK_REG) & + ~PCIE_MSG_PM_PME_MASK; + if ((new & PCI_EXP_RTCTL_PMEIE) == 0) + val |= PCIE_MSG_PM_PME_MASK; + advk_writel(pcie, val, PCIE_ISR0_MASK_REG); break; + } case PCI_EXP_RTSTA: new = (new & PCI_EXP_RTSTA_PME) >> 9; From 7fbcb5da811be7d47468417c7795405058abb3da Mon Sep 17 00:00:00 2001 From: Remi Pommarel Date: Fri, 27 Sep 2019 10:55:02 +0200 Subject: [PATCH 009/160] PCI: aardvark: Don't rely on jiffies while holding spinlock advk_pcie_wait_pio() can be called while holding a spinlock (from pci_bus_read_config_dword()), then depends on jiffies in order to timeout while polling on PIO state registers. In the case the PIO transaction failed, the timeout will never happen and will also cause the cpu to stall. This decrements a variable and wait instead of using jiffies. Signed-off-by: Remi Pommarel Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Acked-by: Thomas Petazzoni --- drivers/pci/controller/pci-aardvark.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c index cb4327657136..79be0afc9b1e 100644 --- a/drivers/pci/controller/pci-aardvark.c +++ b/drivers/pci/controller/pci-aardvark.c @@ -175,7 +175,8 @@ (PCIE_CONF_BUS(bus) | PCIE_CONF_DEV(PCI_SLOT(devfn)) | \ PCIE_CONF_FUNC(PCI_FUNC(devfn)) | PCIE_CONF_REG(where)) -#define PIO_TIMEOUT_MS 1 +#define PIO_RETRY_CNT 500 +#define PIO_RETRY_DELAY 2 /* 2 us*/ #define LINK_WAIT_MAX_RETRIES 10 #define LINK_WAIT_USLEEP_MIN 90000 @@ -404,17 +405,16 @@ static void advk_pcie_check_pio_status(struct advk_pcie *pcie) static int advk_pcie_wait_pio(struct advk_pcie *pcie) { struct device *dev = &pcie->pdev->dev; - unsigned long timeout; + int i; - timeout = jiffies + msecs_to_jiffies(PIO_TIMEOUT_MS); - - while (time_before(jiffies, timeout)) { + for (i = 0; i < PIO_RETRY_CNT; i++) { u32 start, isr; start = advk_readl(pcie, PIO_START); isr = advk_readl(pcie, PIO_ISR); if (!start && isr) return 0; + udelay(PIO_RETRY_DELAY); } dev_err(dev, "config read/write timed out\n"); From 4d3186a525b3f6af7d5f468e94b79b9e92cf3853 Mon Sep 17 00:00:00 2001 From: Remi Pommarel Date: Sun, 1 Sep 2019 15:39:15 +0200 Subject: [PATCH 010/160] PCI: amlogic: Fix reset assertion via gpio descriptor Normally asserting reset signal on gpio would be achieved with: gpiod_set_value_cansleep(reset_gpio, 1); Meson PCI driver set reset value to '0' instead of '1' as it takes into account the PERST# signal polarity. The polarity should be taken care in the device tree instead. This fixes the reset assertion meaning and moves out the polarity configuration in DT (please note that there is no DT currently using this driver). Signed-off-by: Remi Pommarel Signed-off-by: Lorenzo Pieralisi Reviewed-by: Martin Blumenstingl Reviewed-by: Neil Armstrong --- drivers/pci/controller/dwc/pci-meson.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pci/controller/dwc/pci-meson.c b/drivers/pci/controller/dwc/pci-meson.c index e35e9eaa50ee..541f37a6f6a5 100644 --- a/drivers/pci/controller/dwc/pci-meson.c +++ b/drivers/pci/controller/dwc/pci-meson.c @@ -287,9 +287,9 @@ static inline void meson_cfg_writel(struct meson_pcie *mp, u32 val, u32 reg) static void meson_pcie_assert_reset(struct meson_pcie *mp) { - gpiod_set_value_cansleep(mp->reset_gpio, 0); - udelay(500); gpiod_set_value_cansleep(mp->reset_gpio, 1); + udelay(500); + gpiod_set_value_cansleep(mp->reset_gpio, 0); } static void meson_pcie_init_dw(struct meson_pcie *mp) From 0978e95253c64bda17e1c2da5c30744980ca73d3 Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Mon, 16 Sep 2019 14:50:17 +0200 Subject: [PATCH 011/160] dt-bindings: pci: amlogic, meson-pcie: Add G12A bindings Add PCIE bindings for the Amlogic G12A SoC, the support is the same but the PHY is shared with USB3 to control the differential lines. Thus this adds a phy phandle to control the PHY, and only requires the MIPI clock for the Amlogic AXG SoC Family. Signed-off-by: Neil Armstrong Signed-off-by: Lorenzo Pieralisi Reviewed-by: Rob Herring Reviewed-by: Andrew Murray --- .../devicetree/bindings/pci/amlogic,meson-pcie.txt | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/Documentation/devicetree/bindings/pci/amlogic,meson-pcie.txt b/Documentation/devicetree/bindings/pci/amlogic,meson-pcie.txt index efa2c8b9b85a..84fdc422792e 100644 --- a/Documentation/devicetree/bindings/pci/amlogic,meson-pcie.txt +++ b/Documentation/devicetree/bindings/pci/amlogic,meson-pcie.txt @@ -9,13 +9,16 @@ Additional properties are described here: Required properties: - compatible: - should contain "amlogic,axg-pcie" to identify the core. + should contain : + - "amlogic,axg-pcie" for AXG SoC Family + - "amlogic,g12a-pcie" for G12A SoC Family + to identify the core. - reg: should contain the configuration address space. - reg-names: Must be - "elbi" External local bus interface registers - "cfg" Meson specific registers - - "phy" Meson PCIE PHY registers + - "phy" Meson PCIE PHY registers for AXG SoC Family - "config" PCIe configuration space - reset-gpios: The GPIO to generate PCIe PERST# assert and deassert signal. - clocks: Must contain an entry for each entry in clock-names. @@ -23,12 +26,13 @@ Required properties: - "pclk" PCIe GEN 100M PLL clock - "port" PCIe_x(A or B) RC clock gate - "general" PCIe Phy clock - - "mipi" PCIe_x(A or B) 100M ref clock gate + - "mipi" PCIe_x(A or B) 100M ref clock gate for AXG SoC Family - resets: phandle to the reset lines. - reset-names: must contain "phy" "port" and "apb" - - "phy" Share PHY reset + - "phy" Share PHY reset for AXG SoC Family - "port" Port A or B reset - "apb" Share APB reset +- phys: should contain a phandle to the shared phy for G12A SoC Family - device_type: should be "pci". As specified in designware-pcie.txt From eacaf7dcf08eb062a1059c6c115fa3fced3374ae Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Mon, 16 Sep 2019 14:50:18 +0200 Subject: [PATCH 012/160] PCI: amlogic: Fix probed clock names Fix the clock names used in the probe function according to the bindings. Fixes: 9c0ef6d34fdb ("PCI: amlogic: Add the Amlogic Meson PCIe controller driver") Signed-off-by: Neil Armstrong Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray --- drivers/pci/controller/dwc/pci-meson.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/pci/controller/dwc/pci-meson.c b/drivers/pci/controller/dwc/pci-meson.c index 541f37a6f6a5..ab79990798f8 100644 --- a/drivers/pci/controller/dwc/pci-meson.c +++ b/drivers/pci/controller/dwc/pci-meson.c @@ -250,15 +250,15 @@ static int meson_pcie_probe_clocks(struct meson_pcie *mp) if (IS_ERR(res->port_clk)) return PTR_ERR(res->port_clk); - res->mipi_gate = meson_pcie_probe_clock(dev, "pcie_mipi_en", 0); + res->mipi_gate = meson_pcie_probe_clock(dev, "mipi", 0); if (IS_ERR(res->mipi_gate)) return PTR_ERR(res->mipi_gate); - res->general_clk = meson_pcie_probe_clock(dev, "pcie_general", 0); + res->general_clk = meson_pcie_probe_clock(dev, "general", 0); if (IS_ERR(res->general_clk)) return PTR_ERR(res->general_clk); - res->clk = meson_pcie_probe_clock(dev, "pcie", 0); + res->clk = meson_pcie_probe_clock(dev, "pclk", 0); if (IS_ERR(res->clk)) return PTR_ERR(res->clk); From 4ff9f68f8378baa8027426bb8a8853ae3f99ad6c Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Mon, 16 Sep 2019 14:50:19 +0200 Subject: [PATCH 013/160] PCI: amlogic: meson: Add support for G12A Add support for the Amlogic G12A SoC using a separate shared PHY. This adds support for fetching a PHY phandle and call the PHY init, reset and power on/off calls instead of writing in the PHY register or toggling the PHY reset line. The MIPI clock and the PHY memory resource are only required for the Amlogic AXG SoC PCIe PHY setup, thus these elements are ignored for the Amlogic G12A having a separate shared PHY. Signed-off-by: Neil Armstrong Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray --- drivers/pci/controller/dwc/pci-meson.c | 126 ++++++++++++++++++++----- 1 file changed, 104 insertions(+), 22 deletions(-) diff --git a/drivers/pci/controller/dwc/pci-meson.c b/drivers/pci/controller/dwc/pci-meson.c index ab79990798f8..3772b02a5c55 100644 --- a/drivers/pci/controller/dwc/pci-meson.c +++ b/drivers/pci/controller/dwc/pci-meson.c @@ -16,6 +16,7 @@ #include #include #include +#include #include "pcie-designware.h" @@ -96,12 +97,18 @@ struct meson_pcie_rc_reset { struct reset_control *apb; }; +struct meson_pcie_param { + bool has_shared_phy; +}; + struct meson_pcie { struct dw_pcie pci; struct meson_pcie_mem_res mem_res; struct meson_pcie_clk_res clk_res; struct meson_pcie_rc_reset mrst; struct gpio_desc *reset_gpio; + struct phy *phy; + const struct meson_pcie_param *param; }; static struct reset_control *meson_pcie_get_reset(struct meson_pcie *mp, @@ -123,10 +130,12 @@ static int meson_pcie_get_resets(struct meson_pcie *mp) { struct meson_pcie_rc_reset *mrst = &mp->mrst; - mrst->phy = meson_pcie_get_reset(mp, "phy", PCIE_SHARED_RESET); - if (IS_ERR(mrst->phy)) - return PTR_ERR(mrst->phy); - reset_control_deassert(mrst->phy); + if (!mp->param->has_shared_phy) { + mrst->phy = meson_pcie_get_reset(mp, "phy", PCIE_SHARED_RESET); + if (IS_ERR(mrst->phy)) + return PTR_ERR(mrst->phy); + reset_control_deassert(mrst->phy); + } mrst->port = meson_pcie_get_reset(mp, "port", PCIE_NORMAL_RESET); if (IS_ERR(mrst->port)) @@ -180,27 +189,52 @@ static int meson_pcie_get_mems(struct platform_device *pdev, if (IS_ERR(mp->mem_res.cfg_base)) return PTR_ERR(mp->mem_res.cfg_base); - /* Meson SoC has two PCI controllers use same phy register*/ - mp->mem_res.phy_base = meson_pcie_get_mem_shared(pdev, mp, "phy"); - if (IS_ERR(mp->mem_res.phy_base)) - return PTR_ERR(mp->mem_res.phy_base); + /* Meson AXG SoC has two PCI controllers use same phy register */ + if (!mp->param->has_shared_phy) { + mp->mem_res.phy_base = + meson_pcie_get_mem_shared(pdev, mp, "phy"); + if (IS_ERR(mp->mem_res.phy_base)) + return PTR_ERR(mp->mem_res.phy_base); + } return 0; } -static void meson_pcie_power_on(struct meson_pcie *mp) +static int meson_pcie_power_on(struct meson_pcie *mp) { - writel(MESON_PCIE_PHY_POWERUP, mp->mem_res.phy_base); + int ret = 0; + + if (mp->param->has_shared_phy) { + ret = phy_init(mp->phy); + if (ret) + return ret; + + ret = phy_power_on(mp->phy); + if (ret) { + phy_exit(mp->phy); + return ret; + } + } else + writel(MESON_PCIE_PHY_POWERUP, mp->mem_res.phy_base); + + return 0; } -static void meson_pcie_reset(struct meson_pcie *mp) +static int meson_pcie_reset(struct meson_pcie *mp) { struct meson_pcie_rc_reset *mrst = &mp->mrst; + int ret = 0; - reset_control_assert(mrst->phy); - udelay(PCIE_RESET_DELAY); - reset_control_deassert(mrst->phy); - udelay(PCIE_RESET_DELAY); + if (mp->param->has_shared_phy) { + ret = phy_reset(mp->phy); + if (ret) + return ret; + } else { + reset_control_assert(mrst->phy); + udelay(PCIE_RESET_DELAY); + reset_control_deassert(mrst->phy); + udelay(PCIE_RESET_DELAY); + } reset_control_assert(mrst->port); reset_control_assert(mrst->apb); @@ -208,6 +242,8 @@ static void meson_pcie_reset(struct meson_pcie *mp) reset_control_deassert(mrst->port); reset_control_deassert(mrst->apb); udelay(PCIE_RESET_DELAY); + + return 0; } static inline struct clk *meson_pcie_probe_clock(struct device *dev, @@ -250,9 +286,11 @@ static int meson_pcie_probe_clocks(struct meson_pcie *mp) if (IS_ERR(res->port_clk)) return PTR_ERR(res->port_clk); - res->mipi_gate = meson_pcie_probe_clock(dev, "mipi", 0); - if (IS_ERR(res->mipi_gate)) - return PTR_ERR(res->mipi_gate); + if (!mp->param->has_shared_phy) { + res->mipi_gate = meson_pcie_probe_clock(dev, "mipi", 0); + if (IS_ERR(res->mipi_gate)) + return PTR_ERR(res->mipi_gate); + } res->general_clk = meson_pcie_probe_clock(dev, "general", 0); if (IS_ERR(res->general_clk)) @@ -524,6 +562,7 @@ static const struct dw_pcie_ops dw_pcie_ops = { static int meson_pcie_probe(struct platform_device *pdev) { + const struct meson_pcie_param *match_data; struct device *dev = &pdev->dev; struct dw_pcie *pci; struct meson_pcie *mp; @@ -537,6 +576,19 @@ static int meson_pcie_probe(struct platform_device *pdev) pci->dev = dev; pci->ops = &dw_pcie_ops; + match_data = of_device_get_match_data(dev); + if (!match_data) { + dev_err(dev, "failed to get match data\n"); + return -ENODEV; + } + mp->param = match_data; + + if (mp->param->has_shared_phy) { + mp->phy = devm_phy_get(dev, "pcie"); + if (IS_ERR(mp->phy)) + return PTR_ERR(mp->phy); + } + mp->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW); if (IS_ERR(mp->reset_gpio)) { dev_err(dev, "get reset gpio failed\n"); @@ -555,13 +607,22 @@ static int meson_pcie_probe(struct platform_device *pdev) return ret; } - meson_pcie_power_on(mp); - meson_pcie_reset(mp); + ret = meson_pcie_power_on(mp); + if (ret) { + dev_err(dev, "phy power on failed, %d\n", ret); + return ret; + } + + ret = meson_pcie_reset(mp); + if (ret) { + dev_err(dev, "reset failed, %d\n", ret); + goto err_phy; + } ret = meson_pcie_probe_clocks(mp); if (ret) { dev_err(dev, "init clock resources failed, %d\n", ret); - return ret; + goto err_phy; } platform_set_drvdata(pdev, mp); @@ -569,15 +630,36 @@ static int meson_pcie_probe(struct platform_device *pdev) ret = meson_add_pcie_port(mp, pdev); if (ret < 0) { dev_err(dev, "Add PCIe port failed, %d\n", ret); - return ret; + goto err_phy; } return 0; + +err_phy: + if (mp->param->has_shared_phy) { + phy_power_off(mp->phy); + phy_exit(mp->phy); + } + + return ret; } +static struct meson_pcie_param meson_pcie_axg_param = { + .has_shared_phy = false, +}; + +static struct meson_pcie_param meson_pcie_g12a_param = { + .has_shared_phy = true, +}; + static const struct of_device_id meson_pcie_of_match[] = { { .compatible = "amlogic,axg-pcie", + .data = &meson_pcie_axg_param, + }, + { + .compatible = "amlogic,g12a-pcie", + .data = &meson_pcie_g12a_param, }, {}, }; From 631627253de29eef733cadd9385064a9fbc39823 Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Mon, 16 Sep 2019 14:50:20 +0200 Subject: [PATCH 014/160] phy: meson-g12a-usb3-pcie: Add support for PCIe mode This adds extended PCIe PHY functions for the Amlogic G12A USB3+PCIE Combo PHY to support reset, power_on and power_off for PCIe exclusively. With these callbacks, we can handle all the needed operations of the Amlogic PCIe controller driver. Signed-off-by: Neil Armstrong Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray --- .../phy/amlogic/phy-meson-g12a-usb3-pcie.c | 70 ++++++++++++++++--- 1 file changed, 61 insertions(+), 9 deletions(-) diff --git a/drivers/phy/amlogic/phy-meson-g12a-usb3-pcie.c b/drivers/phy/amlogic/phy-meson-g12a-usb3-pcie.c index ac322d643c7a..08e322789e59 100644 --- a/drivers/phy/amlogic/phy-meson-g12a-usb3-pcie.c +++ b/drivers/phy/amlogic/phy-meson-g12a-usb3-pcie.c @@ -50,6 +50,8 @@ #define PHY_R5_PHY_CR_ACK BIT(16) #define PHY_R5_PHY_BS_OUT BIT(17) +#define PCIE_RESET_DELAY 500 + struct phy_g12a_usb3_pcie_priv { struct regmap *regmap; struct regmap *regmap_cr; @@ -196,6 +198,10 @@ static int phy_g12a_usb3_init(struct phy *phy) struct phy_g12a_usb3_pcie_priv *priv = phy_get_drvdata(phy); int data, ret; + ret = reset_control_reset(priv->reset); + if (ret) + return ret; + /* Switch PHY to USB3 */ /* TODO figure out how to handle when PCIe was set in the bootloader */ regmap_update_bits(priv->regmap, PHY_R0, @@ -272,24 +278,64 @@ static int phy_g12a_usb3_init(struct phy *phy) return 0; } -static int phy_g12a_usb3_pcie_init(struct phy *phy) +static int phy_g12a_usb3_pcie_power_on(struct phy *phy) +{ + struct phy_g12a_usb3_pcie_priv *priv = phy_get_drvdata(phy); + + if (priv->mode == PHY_TYPE_USB3) + return 0; + + regmap_update_bits(priv->regmap, PHY_R0, + PHY_R0_PCIE_POWER_STATE, + FIELD_PREP(PHY_R0_PCIE_POWER_STATE, 0x1c)); + + return 0; +} + +static int phy_g12a_usb3_pcie_power_off(struct phy *phy) +{ + struct phy_g12a_usb3_pcie_priv *priv = phy_get_drvdata(phy); + + if (priv->mode == PHY_TYPE_USB3) + return 0; + + regmap_update_bits(priv->regmap, PHY_R0, + PHY_R0_PCIE_POWER_STATE, + FIELD_PREP(PHY_R0_PCIE_POWER_STATE, 0x1d)); + + return 0; +} + +static int phy_g12a_usb3_pcie_reset(struct phy *phy) { struct phy_g12a_usb3_pcie_priv *priv = phy_get_drvdata(phy); int ret; - ret = reset_control_reset(priv->reset); + if (priv->mode == PHY_TYPE_USB3) + return 0; + + ret = reset_control_assert(priv->reset); if (ret) return ret; + udelay(PCIE_RESET_DELAY); + + ret = reset_control_deassert(priv->reset); + if (ret) + return ret; + + udelay(PCIE_RESET_DELAY); + + return 0; +} + +static int phy_g12a_usb3_pcie_init(struct phy *phy) +{ + struct phy_g12a_usb3_pcie_priv *priv = phy_get_drvdata(phy); + if (priv->mode == PHY_TYPE_USB3) return phy_g12a_usb3_init(phy); - /* Power UP PCIE */ - /* TODO figure out when the bootloader has set USB3 mode before */ - regmap_update_bits(priv->regmap, PHY_R0, - PHY_R0_PCIE_POWER_STATE, - FIELD_PREP(PHY_R0_PCIE_POWER_STATE, 0x1c)); - return 0; } @@ -297,7 +343,10 @@ static int phy_g12a_usb3_pcie_exit(struct phy *phy) { struct phy_g12a_usb3_pcie_priv *priv = phy_get_drvdata(phy); - return reset_control_reset(priv->reset); + if (priv->mode == PHY_TYPE_USB3) + return reset_control_reset(priv->reset); + + return 0; } static struct phy *phy_g12a_usb3_pcie_xlate(struct device *dev, @@ -326,6 +375,9 @@ static struct phy *phy_g12a_usb3_pcie_xlate(struct device *dev, static const struct phy_ops phy_g12a_usb3_pcie_ops = { .init = phy_g12a_usb3_pcie_init, .exit = phy_g12a_usb3_pcie_exit, + .power_on = phy_g12a_usb3_pcie_power_on, + .power_off = phy_g12a_usb3_pcie_power_off, + .reset = phy_g12a_usb3_pcie_reset, .owner = THIS_MODULE, }; From 934de3415e5e63561530d1c571dc47e116a9c63a Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Mon, 16 Sep 2019 14:50:21 +0200 Subject: [PATCH 015/160] arm64: dts: meson-g12a: Add PCIe node This adds the Amlogic G12A PCI Express controller node, also using the USB3+PCIe Combo PHY. The PHY mode selection is static, thus the USB3+PCIe Combo PHY phandle would need to be removed from the USB control node if the shared differential lines are used for PCIe instead of USB3. Signed-off-by: Neil Armstrong Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray --- .../boot/dts/amlogic/meson-g12-common.dtsi | 33 +++++++++++++++++++ arch/arm64/boot/dts/amlogic/meson-sm1.dtsi | 4 +++ 2 files changed, 37 insertions(+) diff --git a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi index 3f39e020f74e..7ab71172cd3c 100644 --- a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi @@ -95,6 +95,39 @@ #size-cells = <2>; ranges; + pcie: pcie@fc000000 { + compatible = "amlogic,g12a-pcie", "snps,dw-pcie"; + reg = <0x0 0xfc000000 0x0 0x400000 + 0x0 0xff648000 0x0 0x2000 + 0x0 0xfc400000 0x0 0x200000>; + reg-names = "elbi", "cfg", "config"; + interrupts = ; + #interrupt-cells = <1>; + interrupt-map-mask = <0 0 0 0>; + interrupt-map = <0 0 0 0 &gic GIC_SPI 223 IRQ_TYPE_LEVEL_HIGH>; + bus-range = <0x0 0xff>; + #address-cells = <3>; + #size-cells = <2>; + device_type = "pci"; + ranges = <0x81000000 0 0 0x0 0xfc600000 0 0x00100000 + 0x82000000 0 0xfc700000 0x0 0xfc700000 0 0x1900000>; + + clocks = <&clkc CLKID_PCIE_PHY + &clkc CLKID_PCIE_COMB + &clkc CLKID_PCIE_PLL>; + clock-names = "general", + "pclk", + "port"; + resets = <&reset RESET_PCIE_CTRL_A>, + <&reset RESET_PCIE_APB>; + reset-names = "port", + "apb"; + num-lanes = <1>; + phys = <&usb3_pcie_phy PHY_TYPE_PCIE>; + phy-names = "pcie"; + status = "disabled"; + }; + ethmac: ethernet@ff3f0000 { compatible = "amlogic,meson-axg-dwmac", "snps,dwmac-3.70a", diff --git a/arch/arm64/boot/dts/amlogic/meson-sm1.dtsi b/arch/arm64/boot/dts/amlogic/meson-sm1.dtsi index 521573f3a5ba..256ea0349ffc 100644 --- a/arch/arm64/boot/dts/amlogic/meson-sm1.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-sm1.dtsi @@ -134,6 +134,10 @@ power-domains = <&pwrc PWRC_SM1_ETH_ID>; }; +&pcie { + power-domains = <&pwrc PWRC_SM1_PCIE_ID>; +}; + &pwrc { compatible = "amlogic,meson-sm1-pwrc"; }; From ba1f8af7f772139607d46da3fc151a403f966f46 Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Mon, 16 Sep 2019 14:50:22 +0200 Subject: [PATCH 016/160] arm64: dts: khadas-vim3: add commented support for PCIe The VIM3 on-board MCU can mux the PCIe/USB3.0 shared differential lines using a FUSB340TMX USB 3.1 SuperSpeed Data Switch between an USB3.0 Type A connector and a M.2 Key M slot. The PHY driving these differential lines is shared between the USB3.0 controller and the PCIe Controller, thus only a single controller can use it. The needed DT configuration when the MCU is configured to mux the PCIe/USB3.0 differential lines to the M.2 Key M slot is added commented and may be uncommented to disable USB3.0 from the USB Complex and enable the PCIe controller. The End User is not expected to uncomment the following except for testing purposes, but instead rely on the firmware/bootloader to update these nodes accordingly if PCIe mode is selected by the MCU. Signed-off-by: Neil Armstrong Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray --- .../amlogic/meson-g12b-a311d-khadas-vim3.dts | 25 +++++++++++++++++++ .../amlogic/meson-g12b-s922x-khadas-vim3.dts | 25 +++++++++++++++++++ .../boot/dts/amlogic/meson-khadas-vim3.dtsi | 4 +++ .../dts/amlogic/meson-sm1-khadas-vim3l.dts | 25 +++++++++++++++++++ 4 files changed, 79 insertions(+) diff --git a/arch/arm64/boot/dts/amlogic/meson-g12b-a311d-khadas-vim3.dts b/arch/arm64/boot/dts/amlogic/meson-g12b-a311d-khadas-vim3.dts index 3a6a1e0c1e32..124a80901084 100644 --- a/arch/arm64/boot/dts/amlogic/meson-g12b-a311d-khadas-vim3.dts +++ b/arch/arm64/boot/dts/amlogic/meson-g12b-a311d-khadas-vim3.dts @@ -14,3 +14,28 @@ / { compatible = "khadas,vim3", "amlogic,a311d", "amlogic,g12b"; }; + +/* + * The VIM3 on-board MCU can mux the PCIe/USB3.0 shared differential + * lines using a FUSB340TMX USB 3.1 SuperSpeed Data Switch between + * an USB3.0 Type A connector and a M.2 Key M slot. + * The PHY driving these differential lines is shared between + * the USB3.0 controller and the PCIe Controller, thus only + * a single controller can use it. + * If the MCU is configured to mux the PCIe/USB3.0 differential lines + * to the M.2 Key M slot, uncomment the following block to disable + * USB3.0 from the USB Complex and enable the PCIe controller. + * The End User is not expected to uncomment the following except for + * testing purposes, but instead rely on the firmware/bootloader to + * update these nodes accordingly if PCIe mode is selected by the MCU. + */ +/* +&pcie { + status = "okay"; +}; + +&usb { + phys = <&usb2_phy0>, <&usb2_phy1>; + phy-names = "usb2-phy0", "usb2-phy1"; +}; + */ diff --git a/arch/arm64/boot/dts/amlogic/meson-g12b-s922x-khadas-vim3.dts b/arch/arm64/boot/dts/amlogic/meson-g12b-s922x-khadas-vim3.dts index b73deb282120..bba98f982ad6 100644 --- a/arch/arm64/boot/dts/amlogic/meson-g12b-s922x-khadas-vim3.dts +++ b/arch/arm64/boot/dts/amlogic/meson-g12b-s922x-khadas-vim3.dts @@ -14,3 +14,28 @@ / { compatible = "khadas,vim3", "amlogic,s922x", "amlogic,g12b"; }; + +/* + * The VIM3 on-board MCU can mux the PCIe/USB3.0 shared differential + * lines using a FUSB340TMX USB 3.1 SuperSpeed Data Switch between + * an USB3.0 Type A connector and a M.2 Key M slot. + * The PHY driving these differential lines is shared between + * the USB3.0 controller and the PCIe Controller, thus only + * a single controller can use it. + * If the MCU is configured to mux the PCIe/USB3.0 differential lines + * to the M.2 Key M slot, uncomment the following block to disable + * USB3.0 from the USB Complex and enable the PCIe controller. + * The End User is not expected to uncomment the following except for + * testing purposes, but instead rely on the firmware/bootloader to + * update these nodes accordingly if PCIe mode is selected by the MCU. + */ +/* +&pcie { + status = "okay"; +}; + +&usb { + phys = <&usb2_phy0>, <&usb2_phy1>; + phy-names = "usb2-phy0", "usb2-phy1"; +}; + */ diff --git a/arch/arm64/boot/dts/amlogic/meson-khadas-vim3.dtsi b/arch/arm64/boot/dts/amlogic/meson-khadas-vim3.dtsi index 8647da7d6609..eac5720dc15f 100644 --- a/arch/arm64/boot/dts/amlogic/meson-khadas-vim3.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-khadas-vim3.dtsi @@ -246,6 +246,10 @@ linux,rc-map-name = "rc-khadas"; }; +&pcie { + reset-gpios = <&gpio GPIOA_8 GPIO_ACTIVE_LOW>; +}; + &pwm_ef { status = "okay"; pinctrl-0 = <&pwm_e_pins>; diff --git a/arch/arm64/boot/dts/amlogic/meson-sm1-khadas-vim3l.dts b/arch/arm64/boot/dts/amlogic/meson-sm1-khadas-vim3l.dts index 5233bd7cacfb..dbbf29a0dbf6 100644 --- a/arch/arm64/boot/dts/amlogic/meson-sm1-khadas-vim3l.dts +++ b/arch/arm64/boot/dts/amlogic/meson-sm1-khadas-vim3l.dts @@ -68,3 +68,28 @@ clock-names = "clkin1"; status = "okay"; }; + +/* + * The VIM3 on-board MCU can mux the PCIe/USB3.0 shared differential + * lines using a FUSB340TMX USB 3.1 SuperSpeed Data Switch between + * an USB3.0 Type A connector and a M.2 Key M slot. + * The PHY driving these differential lines is shared between + * the USB3.0 controller and the PCIe Controller, thus only + * a single controller can use it. + * If the MCU is configured to mux the PCIe/USB3.0 differential lines + * to the M.2 Key M slot, uncomment the following block to disable + * USB3.0 from the USB Complex and enable the PCIe controller. + * The End User is not expected to uncomment the following except for + * testing purposes, but instead rely on the firmware/bootloader to + * update these nodes accordingly if PCIe mode is selected by the MCU. + */ +/* +&pcie { + status = "okay"; +}; + +&usb { + phys = <&usb2_phy0>, <&usb2_phy1>; + phy-names = "usb2-phy0", "usb2-phy1"; +}; + */ From daee4f4e42c792997f4fee47dcdfa65dd720ec02 Mon Sep 17 00:00:00 2001 From: Alan Mikhak Date: Wed, 9 Oct 2019 10:06:56 -0700 Subject: [PATCH 017/160] PCI: endpoint: Cast the page number to phys_addr_t Modify pci_epc_mem_alloc_addr() to cast the variable 'pageno' from type 'int' to 'phys_addr_t' before shifting left. This cast is needed to avoid treating bit 31 of 'pageno' as the sign bit which would otherwise get sign-extended to produce a negative value. When added to the base address of PCI memory space, the negative value would produce an invalid physical address which falls before the start of the PCI memory space. Signed-off-by: Alan Mikhak Signed-off-by: Lorenzo Pieralisi Acked-by: Kishon Vijay Abraham I --- drivers/pci/endpoint/pci-epc-mem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/endpoint/pci-epc-mem.c b/drivers/pci/endpoint/pci-epc-mem.c index 2bf8bd1f0563..d2b174ce15de 100644 --- a/drivers/pci/endpoint/pci-epc-mem.c +++ b/drivers/pci/endpoint/pci-epc-mem.c @@ -134,7 +134,7 @@ void __iomem *pci_epc_mem_alloc_addr(struct pci_epc *epc, if (pageno < 0) return NULL; - *phys_addr = mem->phys_base + (pageno << page_shift); + *phys_addr = mem->phys_base + ((phys_addr_t)pageno << page_shift); virt_addr = ioremap(*phys_addr, size); if (!virt_addr) bitmap_release_region(mem->bitmap, pageno, order); From 4906c05b87d44c19b225935e24d62e4480ca556d Mon Sep 17 00:00:00 2001 From: Kefeng Wang Date: Fri, 4 Oct 2019 12:19:25 +0800 Subject: [PATCH 018/160] PCI: mobiveil: Fix csr_read()/write() build issue RISCV has csr_read()/write() macros in arch/riscv/include/asm/csr.h. The same function naming is used in the PCI mobiveil driver thus causing build error. Rename csr_[read,write][l,] to mobiveil_csr_read()/write() to fix it. drivers/pci/controller/pcie-mobiveil.c:238:69: error: macro "csr_read" passed 3 arguments, but takes just 1 static u32 csr_read(struct mobiveil_pcie *pcie, u32 off, size_t size) drivers/pci/controller/pcie-mobiveil.c:253:80: error: macro "csr_write" passed 4 arguments, but takes just 2 static void csr_write(struct mobiveil_pcie *pcie, u32 val, u32 off, size_t size) Fixes: bcbe0d9a8d93 ("PCI: mobiveil: Unify register accessors") Signed-off-by: Kefeng Wang Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Cc: Hou Zhiqiang Cc: Lorenzo Pieralisi Cc: Minghuan Lian Cc: Subrahmanya Lingappa Cc: Andrew Murray --- drivers/pci/controller/pcie-mobiveil.c | 119 +++++++++++++------------ 1 file changed, 62 insertions(+), 57 deletions(-) diff --git a/drivers/pci/controller/pcie-mobiveil.c b/drivers/pci/controller/pcie-mobiveil.c index a45a6447b01d..32f37d08d5bc 100644 --- a/drivers/pci/controller/pcie-mobiveil.c +++ b/drivers/pci/controller/pcie-mobiveil.c @@ -235,7 +235,7 @@ static int mobiveil_pcie_write(void __iomem *addr, int size, u32 val) return PCIBIOS_SUCCESSFUL; } -static u32 csr_read(struct mobiveil_pcie *pcie, u32 off, size_t size) +static u32 mobiveil_csr_read(struct mobiveil_pcie *pcie, u32 off, size_t size) { void *addr; u32 val; @@ -250,7 +250,8 @@ static u32 csr_read(struct mobiveil_pcie *pcie, u32 off, size_t size) return val; } -static void csr_write(struct mobiveil_pcie *pcie, u32 val, u32 off, size_t size) +static void mobiveil_csr_write(struct mobiveil_pcie *pcie, u32 val, u32 off, + size_t size) { void *addr; int ret; @@ -262,19 +263,19 @@ static void csr_write(struct mobiveil_pcie *pcie, u32 val, u32 off, size_t size) dev_err(&pcie->pdev->dev, "write CSR address failed\n"); } -static u32 csr_readl(struct mobiveil_pcie *pcie, u32 off) +static u32 mobiveil_csr_readl(struct mobiveil_pcie *pcie, u32 off) { - return csr_read(pcie, off, 0x4); + return mobiveil_csr_read(pcie, off, 0x4); } -static void csr_writel(struct mobiveil_pcie *pcie, u32 val, u32 off) +static void mobiveil_csr_writel(struct mobiveil_pcie *pcie, u32 val, u32 off) { - csr_write(pcie, val, off, 0x4); + mobiveil_csr_write(pcie, val, off, 0x4); } static bool mobiveil_pcie_link_up(struct mobiveil_pcie *pcie) { - return (csr_readl(pcie, LTSSM_STATUS) & + return (mobiveil_csr_readl(pcie, LTSSM_STATUS) & LTSSM_STATUS_L0_MASK) == LTSSM_STATUS_L0; } @@ -323,7 +324,7 @@ static void __iomem *mobiveil_pcie_map_bus(struct pci_bus *bus, PCI_SLOT(devfn) << PAB_DEVICE_SHIFT | PCI_FUNC(devfn) << PAB_FUNCTION_SHIFT; - csr_writel(pcie, value, PAB_AXI_AMAP_PEX_WIN_L(WIN_NUM_0)); + mobiveil_csr_writel(pcie, value, PAB_AXI_AMAP_PEX_WIN_L(WIN_NUM_0)); return pcie->config_axi_slave_base + where; } @@ -353,13 +354,14 @@ static void mobiveil_pcie_isr(struct irq_desc *desc) chained_irq_enter(chip, desc); /* read INTx status */ - val = csr_readl(pcie, PAB_INTP_AMBA_MISC_STAT); - mask = csr_readl(pcie, PAB_INTP_AMBA_MISC_ENB); + val = mobiveil_csr_readl(pcie, PAB_INTP_AMBA_MISC_STAT); + mask = mobiveil_csr_readl(pcie, PAB_INTP_AMBA_MISC_ENB); intr_status = val & mask; /* Handle INTx */ if (intr_status & PAB_INTP_INTX_MASK) { - shifted_status = csr_readl(pcie, PAB_INTP_AMBA_MISC_STAT); + shifted_status = mobiveil_csr_readl(pcie, + PAB_INTP_AMBA_MISC_STAT); shifted_status &= PAB_INTP_INTX_MASK; shifted_status >>= PAB_INTX_START; do { @@ -373,12 +375,13 @@ static void mobiveil_pcie_isr(struct irq_desc *desc) bit); /* clear interrupt handled */ - csr_writel(pcie, 1 << (PAB_INTX_START + bit), - PAB_INTP_AMBA_MISC_STAT); + mobiveil_csr_writel(pcie, + 1 << (PAB_INTX_START + bit), + PAB_INTP_AMBA_MISC_STAT); } - shifted_status = csr_readl(pcie, - PAB_INTP_AMBA_MISC_STAT); + shifted_status = mobiveil_csr_readl(pcie, + PAB_INTP_AMBA_MISC_STAT); shifted_status &= PAB_INTP_INTX_MASK; shifted_status >>= PAB_INTX_START; } while (shifted_status != 0); @@ -413,7 +416,7 @@ static void mobiveil_pcie_isr(struct irq_desc *desc) } /* Clear the interrupt status */ - csr_writel(pcie, intr_status, PAB_INTP_AMBA_MISC_STAT); + mobiveil_csr_writel(pcie, intr_status, PAB_INTP_AMBA_MISC_STAT); chained_irq_exit(chip, desc); } @@ -474,24 +477,24 @@ static void program_ib_windows(struct mobiveil_pcie *pcie, int win_num, return; } - value = csr_readl(pcie, PAB_PEX_AMAP_CTRL(win_num)); + value = mobiveil_csr_readl(pcie, PAB_PEX_AMAP_CTRL(win_num)); value &= ~(AMAP_CTRL_TYPE_MASK << AMAP_CTRL_TYPE_SHIFT | WIN_SIZE_MASK); value |= type << AMAP_CTRL_TYPE_SHIFT | 1 << AMAP_CTRL_EN_SHIFT | (lower_32_bits(size64) & WIN_SIZE_MASK); - csr_writel(pcie, value, PAB_PEX_AMAP_CTRL(win_num)); + mobiveil_csr_writel(pcie, value, PAB_PEX_AMAP_CTRL(win_num)); - csr_writel(pcie, upper_32_bits(size64), - PAB_EXT_PEX_AMAP_SIZEN(win_num)); + mobiveil_csr_writel(pcie, upper_32_bits(size64), + PAB_EXT_PEX_AMAP_SIZEN(win_num)); - csr_writel(pcie, lower_32_bits(cpu_addr), - PAB_PEX_AMAP_AXI_WIN(win_num)); - csr_writel(pcie, upper_32_bits(cpu_addr), - PAB_EXT_PEX_AMAP_AXI_WIN(win_num)); + mobiveil_csr_writel(pcie, lower_32_bits(cpu_addr), + PAB_PEX_AMAP_AXI_WIN(win_num)); + mobiveil_csr_writel(pcie, upper_32_bits(cpu_addr), + PAB_EXT_PEX_AMAP_AXI_WIN(win_num)); - csr_writel(pcie, lower_32_bits(pci_addr), - PAB_PEX_AMAP_PEX_WIN_L(win_num)); - csr_writel(pcie, upper_32_bits(pci_addr), - PAB_PEX_AMAP_PEX_WIN_H(win_num)); + mobiveil_csr_writel(pcie, lower_32_bits(pci_addr), + PAB_PEX_AMAP_PEX_WIN_L(win_num)); + mobiveil_csr_writel(pcie, upper_32_bits(pci_addr), + PAB_PEX_AMAP_PEX_WIN_H(win_num)); pcie->ib_wins_configured++; } @@ -515,27 +518,29 @@ static void program_ob_windows(struct mobiveil_pcie *pcie, int win_num, * program Enable Bit to 1, Type Bit to (00) base 2, AXI Window Size Bit * to 4 KB in PAB_AXI_AMAP_CTRL register */ - value = csr_readl(pcie, PAB_AXI_AMAP_CTRL(win_num)); + value = mobiveil_csr_readl(pcie, PAB_AXI_AMAP_CTRL(win_num)); value &= ~(WIN_TYPE_MASK << WIN_TYPE_SHIFT | WIN_SIZE_MASK); value |= 1 << WIN_ENABLE_SHIFT | type << WIN_TYPE_SHIFT | (lower_32_bits(size64) & WIN_SIZE_MASK); - csr_writel(pcie, value, PAB_AXI_AMAP_CTRL(win_num)); + mobiveil_csr_writel(pcie, value, PAB_AXI_AMAP_CTRL(win_num)); - csr_writel(pcie, upper_32_bits(size64), PAB_EXT_AXI_AMAP_SIZE(win_num)); + mobiveil_csr_writel(pcie, upper_32_bits(size64), + PAB_EXT_AXI_AMAP_SIZE(win_num)); /* * program AXI window base with appropriate value in * PAB_AXI_AMAP_AXI_WIN0 register */ - csr_writel(pcie, lower_32_bits(cpu_addr) & (~AXI_WINDOW_ALIGN_MASK), - PAB_AXI_AMAP_AXI_WIN(win_num)); - csr_writel(pcie, upper_32_bits(cpu_addr), - PAB_EXT_AXI_AMAP_AXI_WIN(win_num)); + mobiveil_csr_writel(pcie, + lower_32_bits(cpu_addr) & (~AXI_WINDOW_ALIGN_MASK), + PAB_AXI_AMAP_AXI_WIN(win_num)); + mobiveil_csr_writel(pcie, upper_32_bits(cpu_addr), + PAB_EXT_AXI_AMAP_AXI_WIN(win_num)); - csr_writel(pcie, lower_32_bits(pci_addr), - PAB_AXI_AMAP_PEX_WIN_L(win_num)); - csr_writel(pcie, upper_32_bits(pci_addr), - PAB_AXI_AMAP_PEX_WIN_H(win_num)); + mobiveil_csr_writel(pcie, lower_32_bits(pci_addr), + PAB_AXI_AMAP_PEX_WIN_L(win_num)); + mobiveil_csr_writel(pcie, upper_32_bits(pci_addr), + PAB_AXI_AMAP_PEX_WIN_H(win_num)); pcie->ob_wins_configured++; } @@ -579,42 +584,42 @@ static int mobiveil_host_init(struct mobiveil_pcie *pcie) struct resource_entry *win; /* setup bus numbers */ - value = csr_readl(pcie, PCI_PRIMARY_BUS); + value = mobiveil_csr_readl(pcie, PCI_PRIMARY_BUS); value &= 0xff000000; value |= 0x00ff0100; - csr_writel(pcie, value, PCI_PRIMARY_BUS); + mobiveil_csr_writel(pcie, value, PCI_PRIMARY_BUS); /* * program Bus Master Enable Bit in Command Register in PAB Config * Space */ - value = csr_readl(pcie, PCI_COMMAND); + value = mobiveil_csr_readl(pcie, PCI_COMMAND); value |= PCI_COMMAND_IO | PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER; - csr_writel(pcie, value, PCI_COMMAND); + mobiveil_csr_writel(pcie, value, PCI_COMMAND); /* * program PIO Enable Bit to 1 (and PEX PIO Enable to 1) in PAB_CTRL * register */ - pab_ctrl = csr_readl(pcie, PAB_CTRL); + pab_ctrl = mobiveil_csr_readl(pcie, PAB_CTRL); pab_ctrl |= (1 << AMBA_PIO_ENABLE_SHIFT) | (1 << PEX_PIO_ENABLE_SHIFT); - csr_writel(pcie, pab_ctrl, PAB_CTRL); + mobiveil_csr_writel(pcie, pab_ctrl, PAB_CTRL); - csr_writel(pcie, (PAB_INTP_INTX_MASK | PAB_INTP_MSI_MASK), - PAB_INTP_AMBA_MISC_ENB); + mobiveil_csr_writel(pcie, (PAB_INTP_INTX_MASK | PAB_INTP_MSI_MASK), + PAB_INTP_AMBA_MISC_ENB); /* * program PIO Enable Bit to 1 and Config Window Enable Bit to 1 in * PAB_AXI_PIO_CTRL Register */ - value = csr_readl(pcie, PAB_AXI_PIO_CTRL); + value = mobiveil_csr_readl(pcie, PAB_AXI_PIO_CTRL); value |= APIO_EN_MASK; - csr_writel(pcie, value, PAB_AXI_PIO_CTRL); + mobiveil_csr_writel(pcie, value, PAB_AXI_PIO_CTRL); /* Enable PCIe PIO master */ - value = csr_readl(pcie, PAB_PEX_PIO_CTRL); + value = mobiveil_csr_readl(pcie, PAB_PEX_PIO_CTRL); value |= 1 << PIO_ENABLE_SHIFT; - csr_writel(pcie, value, PAB_PEX_PIO_CTRL); + mobiveil_csr_writel(pcie, value, PAB_PEX_PIO_CTRL); /* * we'll program one outbound window for config reads and @@ -647,10 +652,10 @@ static int mobiveil_host_init(struct mobiveil_pcie *pcie) } /* fixup for PCIe class register */ - value = csr_readl(pcie, PAB_INTP_AXI_PIO_CLASS); + value = mobiveil_csr_readl(pcie, PAB_INTP_AXI_PIO_CLASS); value &= 0xff; value |= (PCI_CLASS_BRIDGE_PCI << 16); - csr_writel(pcie, value, PAB_INTP_AXI_PIO_CLASS); + mobiveil_csr_writel(pcie, value, PAB_INTP_AXI_PIO_CLASS); /* setup MSI hardware registers */ mobiveil_pcie_enable_msi(pcie); @@ -668,9 +673,9 @@ static void mobiveil_mask_intx_irq(struct irq_data *data) pcie = irq_desc_get_chip_data(desc); mask = 1 << ((data->hwirq + PAB_INTX_START) - 1); raw_spin_lock_irqsave(&pcie->intx_mask_lock, flags); - shifted_val = csr_readl(pcie, PAB_INTP_AMBA_MISC_ENB); + shifted_val = mobiveil_csr_readl(pcie, PAB_INTP_AMBA_MISC_ENB); shifted_val &= ~mask; - csr_writel(pcie, shifted_val, PAB_INTP_AMBA_MISC_ENB); + mobiveil_csr_writel(pcie, shifted_val, PAB_INTP_AMBA_MISC_ENB); raw_spin_unlock_irqrestore(&pcie->intx_mask_lock, flags); } @@ -684,9 +689,9 @@ static void mobiveil_unmask_intx_irq(struct irq_data *data) pcie = irq_desc_get_chip_data(desc); mask = 1 << ((data->hwirq + PAB_INTX_START) - 1); raw_spin_lock_irqsave(&pcie->intx_mask_lock, flags); - shifted_val = csr_readl(pcie, PAB_INTP_AMBA_MISC_ENB); + shifted_val = mobiveil_csr_readl(pcie, PAB_INTP_AMBA_MISC_ENB); shifted_val |= mask; - csr_writel(pcie, shifted_val, PAB_INTP_AMBA_MISC_ENB); + mobiveil_csr_writel(pcie, shifted_val, PAB_INTP_AMBA_MISC_ENB); raw_spin_unlock_irqrestore(&pcie->intx_mask_lock, flags); } From 1137e61dcb99f7f8b54e77ed83f68b5b485a3e34 Mon Sep 17 00:00:00 2001 From: Niklas Cassel Date: Wed, 4 Sep 2019 18:03:38 +0200 Subject: [PATCH 019/160] PCI: dwc: Fix find_next_bit() usage find_next_bit() takes a parameter of size long, and performs arithmetic that assumes that the argument is of size long. Therefore we cannot pass a u32, since this will cause find_next_bit() to read outside the stack buffer and will produce the following print: BUG: KASAN: stack-out-of-bounds in find_next_bit+0x38/0xb0 Fixes: 1b497e6493c4 ("PCI: dwc: Fix uninitialized variable in dw_handle_msi_irq()") Tested-by: Bjorn Andersson Signed-off-by: Niklas Cassel Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Acked-by: Gustavo Pimentel --- drivers/pci/controller/dwc/pcie-designware-host.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c index 0f36a926059a..8615f1548882 100644 --- a/drivers/pci/controller/dwc/pcie-designware-host.c +++ b/drivers/pci/controller/dwc/pcie-designware-host.c @@ -78,7 +78,8 @@ static struct msi_domain_info dw_pcie_msi_domain_info = { irqreturn_t dw_handle_msi_irq(struct pcie_port *pp) { int i, pos, irq; - u32 val, num_ctrls; + unsigned long val; + u32 status, num_ctrls; irqreturn_t ret = IRQ_NONE; num_ctrls = pp->num_vectors / MAX_MSI_IRQS_PER_CTRL; @@ -86,14 +87,14 @@ irqreturn_t dw_handle_msi_irq(struct pcie_port *pp) for (i = 0; i < num_ctrls; i++) { dw_pcie_rd_own_conf(pp, PCIE_MSI_INTR0_STATUS + (i * MSI_REG_CTRL_BLOCK_SIZE), - 4, &val); - if (!val) + 4, &status); + if (!status) continue; ret = IRQ_HANDLED; + val = status; pos = 0; - while ((pos = find_next_bit((unsigned long *) &val, - MAX_MSI_IRQS_PER_CTRL, + while ((pos = find_next_bit(&val, MAX_MSI_IRQS_PER_CTRL, pos)) != MAX_MSI_IRQS_PER_CTRL) { irq = irq_find_mapping(pp->irq_domain, (i * MAX_MSI_IRQS_PER_CTRL) + From 5e0c21c75e8c08375a69710527e4a921b897cb7e Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 9 Oct 2019 08:00:12 -0500 Subject: [PATCH 020/160] PCI/ASPM: Remove pcie_aspm_enabled() unnecessary locking The lifetime of the link_state structure (bridge->link_state) is not the same as the lifetime of "bridge" itself. The link_state is allocated by pcie_aspm_init_link_state() after children of the bridge have been enumerated, and it is deallocated by pcie_aspm_exit_link_state() after all children of the bridge (but not the bridge itself) have been removed. Previously pcie_aspm_enabled() acquired aspm_lock to ensure that link_state was not deallocated while we're looking at it. But the fact that the caller of pcie_aspm_enabled() holds a reference to @pdev means there's always at least one child of the bridge, which means link_state can't be deallocated. Remove the unnecessary locking in pcie_aspm_enabled(). Signed-off-by: Bjorn Helgaas Acked-by: Rafael J. Wysocki --- drivers/pci/pcie/aspm.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index 652ef23bba35..f5c7138a34aa 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -1172,20 +1172,20 @@ module_param_call(policy, pcie_aspm_set_policy, pcie_aspm_get_policy, /** * pcie_aspm_enabled - Check if PCIe ASPM has been enabled for a device. * @pdev: Target device. + * + * Relies on the upstream bridge's link_state being valid. The link_state + * is deallocated only when the last child of the bridge (i.e., @pdev or a + * sibling) is removed, and the caller should be holding a reference to + * @pdev, so this should be safe. */ bool pcie_aspm_enabled(struct pci_dev *pdev) { struct pci_dev *bridge = pci_upstream_bridge(pdev); - bool ret; if (!bridge) return false; - mutex_lock(&aspm_lock); - ret = bridge->link_state ? !!bridge->link_state->aspm_enabled : false; - mutex_unlock(&aspm_lock); - - return ret; + return bridge->link_state ? !!bridge->link_state->aspm_enabled : false; } EXPORT_SYMBOL_GPL(pcie_aspm_enabled); From aff5d0552da4055da3faa27ee4252e48bb1f5821 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 5 Oct 2019 14:04:36 +0200 Subject: [PATCH 021/160] PCI/ASPM: Add L1 PM substate support to pci_disable_link_state() Add support for disabling states L1.1 and L1.2 to pci_disable_link_state(). Allow separate control of ASPM and PCI PM L1 substates. Link: https://lore.kernel.org/r/d81f8036-c236-6463-48e7-ebcdcda85bba@gmail.com Signed-off-by: Heiner Kallweit Signed-off-by: Bjorn Helgaas --- drivers/pci/pcie/aspm.c | 11 ++++++++++- include/linux/pci.h | 10 +++++++--- 2 files changed, 17 insertions(+), 4 deletions(-) diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index f5c7138a34aa..f8572a523a15 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -1094,7 +1094,16 @@ static int __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem) if (state & PCIE_LINK_STATE_L0S) link->aspm_disable |= ASPM_STATE_L0S; if (state & PCIE_LINK_STATE_L1) - link->aspm_disable |= ASPM_STATE_L1; + /* L1 PM substates require L1 */ + link->aspm_disable |= ASPM_STATE_L1 | ASPM_STATE_L1SS; + if (state & PCIE_LINK_STATE_L1_1) + link->aspm_disable |= ASPM_STATE_L1_1; + if (state & PCIE_LINK_STATE_L1_2) + link->aspm_disable |= ASPM_STATE_L1_2; + if (state & PCIE_LINK_STATE_L1_1_PCIPM) + link->aspm_disable |= ASPM_STATE_L1_1_PCIPM; + if (state & PCIE_LINK_STATE_L1_2_PCIPM) + link->aspm_disable |= ASPM_STATE_L1_2_PCIPM; pcie_config_aspm_link(link, policy_to_aspm_state(link)); if (state & PCIE_LINK_STATE_CLKPM) { diff --git a/include/linux/pci.h b/include/linux/pci.h index f9088c89a534..9dc5bee14ae9 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1544,9 +1544,13 @@ extern bool pcie_ports_native; #define pcie_ports_native false #endif -#define PCIE_LINK_STATE_L0S 1 -#define PCIE_LINK_STATE_L1 2 -#define PCIE_LINK_STATE_CLKPM 4 +#define PCIE_LINK_STATE_L0S BIT(0) +#define PCIE_LINK_STATE_L1 BIT(1) +#define PCIE_LINK_STATE_CLKPM BIT(2) +#define PCIE_LINK_STATE_L1_1 BIT(3) +#define PCIE_LINK_STATE_L1_2 BIT(4) +#define PCIE_LINK_STATE_L1_1_PCIPM BIT(5) +#define PCIE_LINK_STATE_L1_2_PCIPM BIT(6) #ifdef CONFIG_PCIEASPM int pci_disable_link_state(struct pci_dev *pdev, int state); From 35efea32b26f9aacc99bf07e0d2cdfba2028b099 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 5 Oct 2019 14:03:57 +0200 Subject: [PATCH 022/160] PCI/ASPM: Allow re-enabling Clock PM Previously Clock PM could not be re-enabled after being disabled by pci_disable_link_state() because clkpm_capable was reset. Change this by adding a clkpm_disable field similar to aspm_disable. Link: https://lore.kernel.org/r/4e8a66db-7d53-4a66-c26c-f0037ffaa705@gmail.com Signed-off-by: Heiner Kallweit Signed-off-by: Bjorn Helgaas --- drivers/pci/pcie/aspm.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index f8572a523a15..23e1d6b88655 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -64,6 +64,7 @@ struct pcie_link_state { u32 clkpm_capable:1; /* Clock PM capable? */ u32 clkpm_enabled:1; /* Current Clock PM state */ u32 clkpm_default:1; /* Default Clock PM state by BIOS */ + u32 clkpm_disable:1; /* Clock PM disabled */ /* Exit latencies */ struct aspm_latency latency_up; /* Upstream direction exit latency */ @@ -161,8 +162,11 @@ static void pcie_set_clkpm_nocheck(struct pcie_link_state *link, int enable) static void pcie_set_clkpm(struct pcie_link_state *link, int enable) { - /* Don't enable Clock PM if the link is not Clock PM capable */ - if (!link->clkpm_capable) + /* + * Don't enable Clock PM if the link is not Clock PM capable + * or Clock PM is disabled + */ + if (!link->clkpm_capable || link->clkpm_disable) enable = 0; /* Need nothing if the specified equals to current state */ if (link->clkpm_enabled == enable) @@ -192,7 +196,8 @@ static void pcie_clkpm_cap_init(struct pcie_link_state *link, int blacklist) } link->clkpm_enabled = enabled; link->clkpm_default = enabled; - link->clkpm_capable = (blacklist) ? 0 : capable; + link->clkpm_capable = capable; + link->clkpm_disable = blacklist ? 1 : 0; } static bool pcie_retrain_link(struct pcie_link_state *link) @@ -1106,10 +1111,9 @@ static int __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem) link->aspm_disable |= ASPM_STATE_L1_2_PCIPM; pcie_config_aspm_link(link, policy_to_aspm_state(link)); - if (state & PCIE_LINK_STATE_CLKPM) { - link->clkpm_capable = 0; - pcie_set_clkpm(link, 0); - } + if (state & PCIE_LINK_STATE_CLKPM) + link->clkpm_disable = 1; + pcie_set_clkpm(link, policy_to_clkpm_state(link)); mutex_unlock(&aspm_lock); if (sem) up_read(&pci_bus_sem); From 687aaf386aeb551130f31705ce40d1341047a936 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 5 Oct 2019 14:07:18 +0200 Subject: [PATCH 023/160] PCI/ASPM: Add pcie_aspm_get_link() Factor out getting the link associated with a pci_dev and use this helper where appropriate. In addition this helper will be used in a subsequent patch of this series. Link: https://lore.kernel.org/r/19d33770-29de-a9af-4d85-f2b30269d383@gmail.com Signed-off-by: Heiner Kallweit Signed-off-by: Bjorn Helgaas --- drivers/pci/pcie/aspm.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index 23e1d6b88655..2d5ecede8fa3 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -1066,19 +1066,26 @@ void pcie_aspm_powersave_config_link(struct pci_dev *pdev) up_read(&pci_bus_sem); } -static int __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem) +static struct pcie_link_state *pcie_aspm_get_link(struct pci_dev *pdev) { - struct pci_dev *parent = pdev->bus->self; - struct pcie_link_state *link; + struct pci_dev *bridge; if (!pci_is_pcie(pdev)) - return 0; + return NULL; - if (pcie_downstream_port(pdev)) - parent = pdev; - if (!parent || !parent->link_state) + bridge = pci_upstream_bridge(pdev); + if (!bridge || !pci_is_pcie(bridge)) + return NULL; + + return bridge->link_state; +} + +static int __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem) +{ + struct pcie_link_state *link = pcie_aspm_get_link(pdev); + + if (!link) return -EINVAL; - /* * A driver requested that ASPM be disabled on this device, but * if we don't have permission to manage ASPM (e.g., on ACPI @@ -1095,7 +1102,6 @@ static int __pci_disable_link_state(struct pci_dev *pdev, int state, bool sem) if (sem) down_read(&pci_bus_sem); mutex_lock(&aspm_lock); - link = parent->link_state; if (state & PCIE_LINK_STATE_L0S) link->aspm_disable |= ASPM_STATE_L0S; if (state & PCIE_LINK_STATE_L1) @@ -1193,12 +1199,12 @@ module_param_call(policy, pcie_aspm_set_policy, pcie_aspm_get_policy, */ bool pcie_aspm_enabled(struct pci_dev *pdev) { - struct pci_dev *bridge = pci_upstream_bridge(pdev); + struct pcie_link_state *link = pcie_aspm_get_link(pdev); - if (!bridge) + if (!link) return false; - return bridge->link_state ? !!bridge->link_state->aspm_enabled : false; + return link->aspm_enabled; } EXPORT_SYMBOL_GPL(pcie_aspm_enabled); From fd872843ecd52820654dcacf85e680d8101a9fde Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 9 Oct 2019 15:14:05 -0500 Subject: [PATCH 024/160] iommu/vt-d: Select PCI_PRI for INTEL_IOMMU_SVM Previously intel-iommu.c depended on CONFIG_AMD_IOMMU in an undesirable way. When CONFIG_INTEL_IOMMU_SVM=y, iommu_enable_dev_iotlb() calls PRI interfaces (pci_reset_pri() and pci_enable_pri()), but those are only implemented when CONFIG_PCI_PRI is enabled. The INTEL_IOMMU_SVM Kconfig did nothing with PCI_PRI, but AMD_IOMMU selects PCI_PRI. So if AMD_IOMMU was enabled, intel-iommu.c got the full PRI interfaces, but if AMD_IOMMU was not enabled, it got the PRI stubs. Make the iommu_enable_dev_iotlb() behavior independent of AMD_IOMMU by having INTEL_IOMMU_SVM select PCI_PRI so iommu_enable_dev_iotlb() always uses the full implementations of PRI interfaces. Signed-off-by: Bjorn Helgaas Reviewed-by: Kuppuswamy Sathyanarayanan Reviewed-by: Jerry Snitselaar Reviewed-by: Joerg Roedel Acked-by: Joerg Roedel --- drivers/iommu/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig index e3842eabcfdd..b183c9f916b0 100644 --- a/drivers/iommu/Kconfig +++ b/drivers/iommu/Kconfig @@ -207,6 +207,7 @@ config INTEL_IOMMU_SVM bool "Support for Shared Virtual Memory with Intel IOMMU" depends on INTEL_IOMMU && X86 select PCI_PASID + select PCI_PRI select MMU_NOTIFIER help Shared Virtual Memory (SVM) provides a facility for devices From 8cbb8a9374a271099bacdc890fb16d374261332b Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 9 Oct 2019 14:54:01 -0500 Subject: [PATCH 025/160] PCI/ATS: Move pci_prg_resp_pasid_required() to CONFIG_PCI_PRI pci_prg_resp_pasid_required() returns the value of the "PRG Response PASID Required" bit from the PRI capability, but the interface was previously defined under #ifdef CONFIG_PCI_PASID. Move it from CONFIG_PCI_PASID to CONFIG_PCI_PRI so it's with the other PRI-related things. Signed-off-by: Bjorn Helgaas Reviewed-by: Kuppuswamy Sathyanarayanan Reviewed-by: Joerg Roedel --- drivers/pci/ats.c | 55 +++++++++++++++++++---------------------- include/linux/pci-ats.h | 11 ++++----- 2 files changed, 30 insertions(+), 36 deletions(-) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index e18499243f84..0d06177252c7 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -280,6 +280,31 @@ int pci_reset_pri(struct pci_dev *pdev) return 0; } EXPORT_SYMBOL_GPL(pci_reset_pri); + +/** + * pci_prg_resp_pasid_required - Return PRG Response PASID Required bit + * status. + * @pdev: PCI device structure + * + * Returns 1 if PASID is required in PRG Response Message, 0 otherwise. + */ +int pci_prg_resp_pasid_required(struct pci_dev *pdev) +{ + u16 status; + int pos; + + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); + if (!pos) + return 0; + + pci_read_config_word(pdev, pos + PCI_PRI_STATUS, &status); + + if (status & PCI_PRI_STATUS_PASID) + return 1; + + return 0; +} +EXPORT_SYMBOL_GPL(pci_prg_resp_pasid_required); #endif /* CONFIG_PCI_PRI */ #ifdef CONFIG_PCI_PASID @@ -395,36 +420,6 @@ int pci_pasid_features(struct pci_dev *pdev) } EXPORT_SYMBOL_GPL(pci_pasid_features); -/** - * pci_prg_resp_pasid_required - Return PRG Response PASID Required bit - * status. - * @pdev: PCI device structure - * - * Returns 1 if PASID is required in PRG Response Message, 0 otherwise. - * - * Even though the PRG response PASID status is read from PRI Status - * Register, since this API will mainly be used by PASID users, this - * function is defined within #ifdef CONFIG_PCI_PASID instead of - * CONFIG_PCI_PRI. - */ -int pci_prg_resp_pasid_required(struct pci_dev *pdev) -{ - u16 status; - int pos; - - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); - if (!pos) - return 0; - - pci_read_config_word(pdev, pos + PCI_PRI_STATUS, &status); - - if (status & PCI_PRI_STATUS_PASID) - return 1; - - return 0; -} -EXPORT_SYMBOL_GPL(pci_prg_resp_pasid_required); - #define PASID_NUMBER_SHIFT 8 #define PASID_NUMBER_MASK (0x1f << PASID_NUMBER_SHIFT) /** diff --git a/include/linux/pci-ats.h b/include/linux/pci-ats.h index 1ebb88e7c184..a7a2b3d94fcc 100644 --- a/include/linux/pci-ats.h +++ b/include/linux/pci-ats.h @@ -10,6 +10,7 @@ int pci_enable_pri(struct pci_dev *pdev, u32 reqs); void pci_disable_pri(struct pci_dev *pdev); void pci_restore_pri_state(struct pci_dev *pdev); int pci_reset_pri(struct pci_dev *pdev); +int pci_prg_resp_pasid_required(struct pci_dev *pdev); #else /* CONFIG_PCI_PRI */ @@ -31,6 +32,10 @@ static inline int pci_reset_pri(struct pci_dev *pdev) return -ENODEV; } +static inline int pci_prg_resp_pasid_required(struct pci_dev *pdev) +{ + return 0; +} #endif /* CONFIG_PCI_PRI */ #ifdef CONFIG_PCI_PASID @@ -40,7 +45,6 @@ void pci_disable_pasid(struct pci_dev *pdev); void pci_restore_pasid_state(struct pci_dev *pdev); int pci_pasid_features(struct pci_dev *pdev); int pci_max_pasids(struct pci_dev *pdev); -int pci_prg_resp_pasid_required(struct pci_dev *pdev); #else /* CONFIG_PCI_PASID */ @@ -66,11 +70,6 @@ static inline int pci_max_pasids(struct pci_dev *pdev) { return -EINVAL; } - -static inline int pci_prg_resp_pasid_required(struct pci_dev *pdev) -{ - return 0; -} #endif /* CONFIG_PCI_PASID */ From 9bf49e36d7183a170a9906d19acc5254818fc574 Mon Sep 17 00:00:00 2001 From: Kuppuswamy Sathyanarayanan Date: Thu, 5 Sep 2019 14:31:42 -0500 Subject: [PATCH 026/160] PCI/ATS: Handle sharing of PF PRI Capability with all VFs Per PCIe r5.0, sec 9.3.7.11, VFs must not implement the PRI Capability. If the PF implements PRI, it is shared by the VFs. Since VFs don't have a PRI Capability, pci_enable_pri() always failed, which caused IOMMU setup to fail. Update the PRI interfaces so for VFs they reflect the state of the PF PRI. [bhelgaas: rebase without pri_cap caching, commit log] Suggested-by: Ashok Raj Link: https://lore.kernel.org/r/b971e31f8695980da8e4a7f93e3b6a3edba3edaa.1567029860.git.sathyanarayanan.kuppuswamy@linux.intel.com Link: https://lore.kernel.org/r/20190905193146.90250-2-helgaas@kernel.org Signed-off-by: Kuppuswamy Sathyanarayanan Signed-off-by: Bjorn Helgaas Cc: Ashok Raj Cc: Keith Busch --- drivers/pci/ats.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index 0d06177252c7..b0f68c0ea91e 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -182,6 +182,17 @@ int pci_enable_pri(struct pci_dev *pdev, u32 reqs) u32 max_requests; int pos; + /* + * VFs must not implement the PRI Capability. If their PF + * implements PRI, it is shared by the VFs, so if the PF PRI is + * enabled, it is also enabled for the VF. + */ + if (pdev->is_virtfn) { + if (pci_physfn(pdev)->pri_enabled) + return 0; + return -EINVAL; + } + if (WARN_ON(pdev->pri_enabled)) return -EBUSY; @@ -218,6 +229,10 @@ void pci_disable_pri(struct pci_dev *pdev) u16 control; int pos; + /* VFs share the PF PRI */ + if (pdev->is_virtfn) + return; + if (WARN_ON(!pdev->pri_enabled)) return; @@ -243,6 +258,9 @@ void pci_restore_pri_state(struct pci_dev *pdev) u32 reqs = pdev->pri_reqs_alloc; int pos; + if (pdev->is_virtfn) + return; + if (!pdev->pri_enabled) return; @@ -267,6 +285,9 @@ int pci_reset_pri(struct pci_dev *pdev) u16 control; int pos; + if (pdev->is_virtfn) + return 0; + if (WARN_ON(pdev->pri_enabled)) return -EBUSY; @@ -293,6 +314,9 @@ int pci_prg_resp_pasid_required(struct pci_dev *pdev) u16 status; int pos; + if (pdev->is_virtfn) + pdev = pci_physfn(pdev); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); if (!pos) return 0; From 2b0ae7cc3bfc3fae124c25870f41291c670b4549 Mon Sep 17 00:00:00 2001 From: Kuppuswamy Sathyanarayanan Date: Thu, 5 Sep 2019 14:31:43 -0500 Subject: [PATCH 027/160] PCI/ATS: Handle sharing of PF PASID Capability with all VFs Per PCIe r5.0, sec 9.3.7.14, if a PF implements the PASID Capability, the PF PASID configuration is shared by its VFs. VFs must not implement their own PASID Capability. Since VFs don't have a PASID Capability, pci_enable_pasid() always failed, which caused IOMMU setup to fail. Update the PASID interfaces so for VFs they reflect the state of the PF PASID. [bhelgaas: rebase without pasid_cap caching, commit log] Suggested-by: Ashok Raj Link: https://lore.kernel.org/r/8ba1ac192e4ac737508b6ac15002158e176bab91.1567029860.git.sathyanarayanan.kuppuswamy@linux.intel.com Link: https://lore.kernel.org/r/20190905193146.90250-3-helgaas@kernel.org Signed-off-by: Kuppuswamy Sathyanarayanan Signed-off-by: Bjorn Helgaas Cc: Ashok Raj Cc: Keith Busch --- drivers/pci/ats.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index b0f68c0ea91e..fb5cfd27dd3c 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -346,6 +346,16 @@ int pci_enable_pasid(struct pci_dev *pdev, int features) u16 control, supported; int pos; + /* + * VFs must not implement the PASID Capability, but if a PF + * supports PASID, its VFs share the PF PASID configuration. + */ + if (pdev->is_virtfn) { + if (pci_physfn(pdev)->pasid_enabled) + return 0; + return -EINVAL; + } + if (WARN_ON(pdev->pasid_enabled)) return -EBUSY; @@ -383,6 +393,10 @@ void pci_disable_pasid(struct pci_dev *pdev) u16 control = 0; int pos; + /* VFs share the PF PASID configuration */ + if (pdev->is_virtfn) + return; + if (WARN_ON(!pdev->pasid_enabled)) return; @@ -405,6 +419,9 @@ void pci_restore_pasid_state(struct pci_dev *pdev) u16 control; int pos; + if (pdev->is_virtfn) + return; + if (!pdev->pasid_enabled) return; @@ -432,6 +449,9 @@ int pci_pasid_features(struct pci_dev *pdev) u16 supported; int pos; + if (pdev->is_virtfn) + pdev = pci_physfn(pdev); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); if (!pos) return -EINVAL; @@ -458,6 +478,9 @@ int pci_max_pasids(struct pci_dev *pdev) u16 supported; int pos; + if (pdev->is_virtfn) + pdev = pci_physfn(pdev); + pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); if (!pos) return -EINVAL; From 3ad62192097443e8c3a8e244475bacaecb894d4e Mon Sep 17 00:00:00 2001 From: Kuppuswamy Sathyanarayanan Date: Thu, 5 Sep 2019 14:31:44 -0500 Subject: [PATCH 028/160] PCI/ATS: Disable PF/VF ATS service independently Previously we didn't disable the PF ATS until all associated VFs had disabled it. But per PCIe spec r5.0, sec 9.3.7.8, the ATS Capability in VFs and associated PFs may be enabled independently. Leaving ATS enabled in the PF unnecessarily may have power and performance impacts. Remove this dependency logic in the ATS enable/disable code. [bhelgaas: commit log] Suggested-by: Ashok Raj Link: https://lore.kernel.org/r/8163ab8fa66afd2cba514ae95d29ab12104781aa.1567029860.git.sathyanarayanan.kuppuswamy@linux.intel.com Link: https://lore.kernel.org/r/20190905193146.90250-4-helgaas@kernel.org Signed-off-by: Kuppuswamy Sathyanarayanan Signed-off-by: Bjorn Helgaas Cc: Ashok Raj Cc: Keith Busch --- drivers/pci/ats.c | 11 ----------- include/linux/pci.h | 1 - 2 files changed, 12 deletions(-) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index fb5cfd27dd3c..a708ed4146ca 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -60,8 +60,6 @@ int pci_enable_ats(struct pci_dev *dev, int ps) pdev = pci_physfn(dev); if (pdev->ats_stu != ps) return -EINVAL; - - atomic_inc(&pdev->ats_ref_cnt); /* count enabled VFs */ } else { dev->ats_stu = ps; ctrl |= PCI_ATS_CTRL_STU(dev->ats_stu - PCI_ATS_MIN_STU); @@ -79,20 +77,11 @@ EXPORT_SYMBOL_GPL(pci_enable_ats); */ void pci_disable_ats(struct pci_dev *dev) { - struct pci_dev *pdev; u16 ctrl; if (WARN_ON(!dev->ats_enabled)) return; - if (atomic_read(&dev->ats_ref_cnt)) - return; /* VFs still enabled */ - - if (dev->is_virtfn) { - pdev = pci_physfn(dev); - atomic_dec(&pdev->ats_ref_cnt); - } - pci_read_config_word(dev, dev->ats_cap + PCI_ATS_CTRL, &ctrl); ctrl &= ~PCI_ATS_CTRL_ENABLE; pci_write_config_word(dev, dev->ats_cap + PCI_ATS_CTRL, ctrl); diff --git a/include/linux/pci.h b/include/linux/pci.h index f9088c89a534..c028883c8460 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -452,7 +452,6 @@ struct pci_dev { }; u16 ats_cap; /* ATS Capability offset */ u8 ats_stu; /* ATS Smallest Translation Unit */ - atomic_t ats_ref_cnt; /* Number of VFs with ATS enabled */ #endif #ifdef CONFIG_PCI_PRI u32 pri_reqs_alloc; /* Number of PRI requests allocated */ From c065190bbcd4fb54ce9c5fd34fcad71acf2a0ea4 Mon Sep 17 00:00:00 2001 From: Kuppuswamy Sathyanarayanan Date: Thu, 5 Sep 2019 14:31:45 -0500 Subject: [PATCH 029/160] PCI/ATS: Cache PRI Capability offset Previously each PRI interface searched for the PRI Capability. Cache the capability offset the first time we use it instead of searching each time. [bhelgaas: commit log, reorder patch to later, call pci_pri_init() from pci_init_capabilities()] Link: https://lore.kernel.org/r/0c5495d376faf6dbb8eb2165204c474438aaae65.156 7029860.git.sathyanarayanan.kuppuswamy@linux.intel.com Link: https://lore.kernel.org/r/20190905193146.90250-5-helgaas@kernel.org Signed-off-by: Kuppuswamy Sathyanarayanan Signed-off-by: Bjorn Helgaas --- drivers/pci/ats.c | 51 +++++++++++++++++++++++---------------------- drivers/pci/pci.h | 6 ++++++ drivers/pci/probe.c | 3 +++ include/linux/pci.h | 1 + 4 files changed, 36 insertions(+), 25 deletions(-) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index a708ed4146ca..c97e862b538a 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -159,6 +159,11 @@ int pci_ats_page_aligned(struct pci_dev *pdev) EXPORT_SYMBOL_GPL(pci_ats_page_aligned); #ifdef CONFIG_PCI_PRI +void pci_pri_init(struct pci_dev *pdev) +{ + pdev->pri_cap = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); +} + /** * pci_enable_pri - Enable PRI capability * @ pdev: PCI device structure @@ -169,7 +174,7 @@ int pci_enable_pri(struct pci_dev *pdev, u32 reqs) { u16 control, status; u32 max_requests; - int pos; + int pri = pdev->pri_cap; /* * VFs must not implement the PRI Capability. If their PF @@ -185,21 +190,20 @@ int pci_enable_pri(struct pci_dev *pdev, u32 reqs) if (WARN_ON(pdev->pri_enabled)) return -EBUSY; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); - if (!pos) + if (!pri) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PRI_STATUS, &status); + pci_read_config_word(pdev, pri + PCI_PRI_STATUS, &status); if (!(status & PCI_PRI_STATUS_STOPPED)) return -EBUSY; - pci_read_config_dword(pdev, pos + PCI_PRI_MAX_REQ, &max_requests); + pci_read_config_dword(pdev, pri + PCI_PRI_MAX_REQ, &max_requests); reqs = min(max_requests, reqs); pdev->pri_reqs_alloc = reqs; - pci_write_config_dword(pdev, pos + PCI_PRI_ALLOC_REQ, reqs); + pci_write_config_dword(pdev, pri + PCI_PRI_ALLOC_REQ, reqs); control = PCI_PRI_CTRL_ENABLE; - pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control); + pci_write_config_word(pdev, pri + PCI_PRI_CTRL, control); pdev->pri_enabled = 1; @@ -216,7 +220,7 @@ EXPORT_SYMBOL_GPL(pci_enable_pri); void pci_disable_pri(struct pci_dev *pdev) { u16 control; - int pos; + int pri = pdev->pri_cap; /* VFs share the PF PRI */ if (pdev->is_virtfn) @@ -225,13 +229,12 @@ void pci_disable_pri(struct pci_dev *pdev) if (WARN_ON(!pdev->pri_enabled)) return; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); - if (!pos) + if (!pri) return; - pci_read_config_word(pdev, pos + PCI_PRI_CTRL, &control); + pci_read_config_word(pdev, pri + PCI_PRI_CTRL, &control); control &= ~PCI_PRI_CTRL_ENABLE; - pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control); + pci_write_config_word(pdev, pri + PCI_PRI_CTRL, control); pdev->pri_enabled = 0; } @@ -245,7 +248,7 @@ void pci_restore_pri_state(struct pci_dev *pdev) { u16 control = PCI_PRI_CTRL_ENABLE; u32 reqs = pdev->pri_reqs_alloc; - int pos; + int pri = pdev->pri_cap; if (pdev->is_virtfn) return; @@ -253,12 +256,11 @@ void pci_restore_pri_state(struct pci_dev *pdev) if (!pdev->pri_enabled) return; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); - if (!pos) + if (!pri) return; - pci_write_config_dword(pdev, pos + PCI_PRI_ALLOC_REQ, reqs); - pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control); + pci_write_config_dword(pdev, pri + PCI_PRI_ALLOC_REQ, reqs); + pci_write_config_word(pdev, pri + PCI_PRI_CTRL, control); } EXPORT_SYMBOL_GPL(pci_restore_pri_state); @@ -272,7 +274,7 @@ EXPORT_SYMBOL_GPL(pci_restore_pri_state); int pci_reset_pri(struct pci_dev *pdev) { u16 control; - int pos; + int pri = pdev->pri_cap; if (pdev->is_virtfn) return 0; @@ -280,12 +282,11 @@ int pci_reset_pri(struct pci_dev *pdev) if (WARN_ON(pdev->pri_enabled)) return -EBUSY; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); - if (!pos) + if (!pri) return -EINVAL; control = PCI_PRI_CTRL_RESET; - pci_write_config_word(pdev, pos + PCI_PRI_CTRL, control); + pci_write_config_word(pdev, pri + PCI_PRI_CTRL, control); return 0; } @@ -301,16 +302,16 @@ EXPORT_SYMBOL_GPL(pci_reset_pri); int pci_prg_resp_pasid_required(struct pci_dev *pdev) { u16 status; - int pos; + int pri; if (pdev->is_virtfn) pdev = pci_physfn(pdev); - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); - if (!pos) + pri = pdev->pri_cap; + if (!pri) return 0; - pci_read_config_word(pdev, pos + PCI_PRI_STATUS, &status); + pci_read_config_word(pdev, pri + PCI_PRI_STATUS, &status); if (status & PCI_PRI_STATUS_PASID) return 1; diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 3f6947ee3324..aa08cd35bf87 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -456,6 +456,12 @@ static inline void pci_ats_init(struct pci_dev *d) { } static inline void pci_restore_ats_state(struct pci_dev *dev) { } #endif /* CONFIG_PCI_ATS */ +#ifdef CONFIG_PCI_PRI +void pci_pri_init(struct pci_dev *dev); +#else +static inline void pci_pri_init(struct pci_dev *dev) { } +#endif + #ifdef CONFIG_PCI_IOV int pci_iov_init(struct pci_dev *dev); void pci_iov_release(struct pci_dev *dev); diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 3d5271a7a849..d145165c799c 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -2324,6 +2324,9 @@ static void pci_init_capabilities(struct pci_dev *dev) /* Address Translation Services */ pci_ats_init(dev); + /* Page Request Interface */ + pci_pri_init(dev); + /* Enable ACS P2P upstream forwarding */ pci_enable_acs(dev); diff --git a/include/linux/pci.h b/include/linux/pci.h index c028883c8460..e7770d990c46 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -454,6 +454,7 @@ struct pci_dev { u8 ats_stu; /* ATS Smallest Translation Unit */ #endif #ifdef CONFIG_PCI_PRI + u16 pri_cap; /* PRI Capability offset */ u32 pri_reqs_alloc; /* Number of PRI requests allocated */ #endif #ifdef CONFIG_PCI_PASID From 751035b8dc061ae434c3311bac9cd6d0e5e00f94 Mon Sep 17 00:00:00 2001 From: Kuppuswamy Sathyanarayanan Date: Thu, 5 Sep 2019 14:31:46 -0500 Subject: [PATCH 030/160] PCI/ATS: Cache PASID Capability offset Previously each PASID interface searched for the PASID Capability. Cache the capability offset the first time we use it instead of searching each time. [bhelgaas: commit log, reorder patch to later, call pci_pasid_init() from pci_init_capabilities()] Link: https://lore.kernel.org/r/4957778959fa34eab3e8b3065d1951989c61cb0f.1567029860.git.sathyanarayanan.kuppuswamy@linux.intel.com Link: https://lore.kernel.org/r/20190905193146.90250-6-helgaas@kernel.org Signed-off-by: Kuppuswamy Sathyanarayanan Signed-off-by: Bjorn Helgaas --- drivers/pci/ats.c | 42 +++++++++++++++++++++--------------------- drivers/pci/pci.h | 6 ++++++ drivers/pci/probe.c | 3 +++ include/linux/pci.h | 1 + 4 files changed, 31 insertions(+), 21 deletions(-) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index c97e862b538a..d5ac808cae21 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -322,6 +322,11 @@ EXPORT_SYMBOL_GPL(pci_prg_resp_pasid_required); #endif /* CONFIG_PCI_PRI */ #ifdef CONFIG_PCI_PASID +void pci_pasid_init(struct pci_dev *pdev) +{ + pdev->pasid_cap = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); +} + /** * pci_enable_pasid - Enable the PASID capability * @pdev: PCI device structure @@ -334,7 +339,7 @@ EXPORT_SYMBOL_GPL(pci_prg_resp_pasid_required); int pci_enable_pasid(struct pci_dev *pdev, int features) { u16 control, supported; - int pos; + int pasid = pdev->pasid_cap; /* * VFs must not implement the PASID Capability, but if a PF @@ -352,11 +357,10 @@ int pci_enable_pasid(struct pci_dev *pdev, int features) if (!pdev->eetlp_prefix_path) return -EINVAL; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); - if (!pos) + if (!pasid) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported); + pci_read_config_word(pdev, pasid + PCI_PASID_CAP, &supported); supported &= PCI_PASID_CAP_EXEC | PCI_PASID_CAP_PRIV; /* User wants to enable anything unsupported? */ @@ -366,7 +370,7 @@ int pci_enable_pasid(struct pci_dev *pdev, int features) control = PCI_PASID_CTRL_ENABLE | features; pdev->pasid_features = features; - pci_write_config_word(pdev, pos + PCI_PASID_CTRL, control); + pci_write_config_word(pdev, pasid + PCI_PASID_CTRL, control); pdev->pasid_enabled = 1; @@ -381,7 +385,7 @@ EXPORT_SYMBOL_GPL(pci_enable_pasid); void pci_disable_pasid(struct pci_dev *pdev) { u16 control = 0; - int pos; + int pasid = pdev->pasid_cap; /* VFs share the PF PASID configuration */ if (pdev->is_virtfn) @@ -390,11 +394,10 @@ void pci_disable_pasid(struct pci_dev *pdev) if (WARN_ON(!pdev->pasid_enabled)) return; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); - if (!pos) + if (!pasid) return; - pci_write_config_word(pdev, pos + PCI_PASID_CTRL, control); + pci_write_config_word(pdev, pasid + PCI_PASID_CTRL, control); pdev->pasid_enabled = 0; } @@ -407,7 +410,7 @@ EXPORT_SYMBOL_GPL(pci_disable_pasid); void pci_restore_pasid_state(struct pci_dev *pdev) { u16 control; - int pos; + int pasid = pdev->pasid_cap; if (pdev->is_virtfn) return; @@ -415,12 +418,11 @@ void pci_restore_pasid_state(struct pci_dev *pdev) if (!pdev->pasid_enabled) return; - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); - if (!pos) + if (!pasid) return; control = PCI_PASID_CTRL_ENABLE | pdev->pasid_features; - pci_write_config_word(pdev, pos + PCI_PASID_CTRL, control); + pci_write_config_word(pdev, pasid + PCI_PASID_CTRL, control); } EXPORT_SYMBOL_GPL(pci_restore_pasid_state); @@ -437,16 +439,15 @@ EXPORT_SYMBOL_GPL(pci_restore_pasid_state); int pci_pasid_features(struct pci_dev *pdev) { u16 supported; - int pos; + int pasid = pdev->pasid_cap; if (pdev->is_virtfn) pdev = pci_physfn(pdev); - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); - if (!pos) + if (!pasid) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported); + pci_read_config_word(pdev, pasid + PCI_PASID_CAP, &supported); supported &= PCI_PASID_CAP_EXEC | PCI_PASID_CAP_PRIV; @@ -466,16 +467,15 @@ EXPORT_SYMBOL_GPL(pci_pasid_features); int pci_max_pasids(struct pci_dev *pdev) { u16 supported; - int pos; + int pasid = pdev->pasid_cap; if (pdev->is_virtfn) pdev = pci_physfn(pdev); - pos = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PASID); - if (!pos) + if (!pasid) return -EINVAL; - pci_read_config_word(pdev, pos + PCI_PASID_CAP, &supported); + pci_read_config_word(pdev, pasid + PCI_PASID_CAP, &supported); supported = (supported & PASID_NUMBER_MASK) >> PASID_NUMBER_SHIFT; diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index aa08cd35bf87..ae84d28ba03a 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -462,6 +462,12 @@ void pci_pri_init(struct pci_dev *dev); static inline void pci_pri_init(struct pci_dev *dev) { } #endif +#ifdef CONFIG_PCI_PASID +void pci_pasid_init(struct pci_dev *dev); +#else +static inline void pci_pasid_init(struct pci_dev *dev) { } +#endif + #ifdef CONFIG_PCI_IOV int pci_iov_init(struct pci_dev *dev); void pci_iov_release(struct pci_dev *dev); diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index d145165c799c..df2b77866f3b 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -2327,6 +2327,9 @@ static void pci_init_capabilities(struct pci_dev *dev) /* Page Request Interface */ pci_pri_init(dev); + /* Process Address Space ID */ + pci_pasid_init(dev); + /* Enable ACS P2P upstream forwarding */ pci_enable_acs(dev); diff --git a/include/linux/pci.h b/include/linux/pci.h index e7770d990c46..6542100bd2dd 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -458,6 +458,7 @@ struct pci_dev { u32 pri_reqs_alloc; /* Number of PRI requests allocated */ #endif #ifdef CONFIG_PCI_PASID + u16 pasid_cap; /* PASID Capability offset */ u16 pasid_features; #endif #ifdef CONFIG_PCI_P2PDMA From e5adf79a1d8086aefa56f48eeb08f8fe4e054a3d Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 9 Oct 2019 16:07:51 -0500 Subject: [PATCH 031/160] PCI/ATS: Cache PRI PRG Response PASID Required bit The PRG Response PASID Required bit in the PRI Capability is read-only. Read it once when we enumerate the device and cache the value so we don't need to read it again. Based-on-patch-by: Kuppuswamy Sathyanarayanan Signed-off-by: Bjorn Helgaas --- drivers/pci/ats.c | 23 ++++++++++------------- include/linux/pci.h | 1 + 2 files changed, 11 insertions(+), 13 deletions(-) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index d5ac808cae21..76ae518d55f4 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -161,7 +161,16 @@ EXPORT_SYMBOL_GPL(pci_ats_page_aligned); #ifdef CONFIG_PCI_PRI void pci_pri_init(struct pci_dev *pdev) { + u16 status; + pdev->pri_cap = pci_find_ext_capability(pdev, PCI_EXT_CAP_ID_PRI); + + if (!pdev->pri_cap) + return; + + pci_read_config_word(pdev, pdev->pri_cap + PCI_PRI_STATUS, &status); + if (status & PCI_PRI_STATUS_PASID) + pdev->pasid_required = 1; } /** @@ -301,22 +310,10 @@ EXPORT_SYMBOL_GPL(pci_reset_pri); */ int pci_prg_resp_pasid_required(struct pci_dev *pdev) { - u16 status; - int pri; - if (pdev->is_virtfn) pdev = pci_physfn(pdev); - pri = pdev->pri_cap; - if (!pri) - return 0; - - pci_read_config_word(pdev, pri + PCI_PRI_STATUS, &status); - - if (status & PCI_PRI_STATUS_PASID) - return 1; - - return 0; + return pdev->pasid_required; } EXPORT_SYMBOL_GPL(pci_prg_resp_pasid_required); #endif /* CONFIG_PCI_PRI */ diff --git a/include/linux/pci.h b/include/linux/pci.h index 6542100bd2dd..64d35e730fab 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -456,6 +456,7 @@ struct pci_dev { #ifdef CONFIG_PCI_PRI u16 pri_cap; /* PRI Capability offset */ u32 pri_reqs_alloc; /* Number of PRI requests allocated */ + unsigned int pasid_required:1; /* PRG Response PASID Required */ #endif #ifdef CONFIG_PCI_PASID u16 pasid_cap; /* PASID Capability offset */ From b24d5c2098596a41cf187af41287777a2e0dd753 Mon Sep 17 00:00:00 2001 From: Krzysztof Wilczynski Date: Sat, 14 Sep 2019 23:30:32 +0200 Subject: [PATCH 032/160] PCI/ATS: Consolidate ATS declarations in linux/pci-ats.h Move ATS function prototypes from include/linux/pci.h to include/linux/pci-ats.h as the ATS, PRI, and PASID interfaces are related and are used only by the IOMMU drivers. This effectively reverts ff9bee895c4d ("PCI: Move ATS declarations to linux/pci.h so they're all together"). Also, remove surplus forward declaration of struct pci_ats from include/linux/pci.h, as it is no longer needed, since struct pci_ats was embedded directly into struct pci_dev by d544d75ac96a ("PCI: Embed ATS info directly into struct pci_dev"). No functional changes intended. Link: https://lore.kernel.org/r/20190914213032.22314-1-kw@linux.com Signed-off-by: Krzysztof Wilczynski Signed-off-by: Bjorn Helgaas --- include/linux/pci-ats.h | 75 +++++++++++++++-------------------------- include/linux/pci.h | 14 -------- 2 files changed, 28 insertions(+), 61 deletions(-) diff --git a/include/linux/pci-ats.h b/include/linux/pci-ats.h index a7a2b3d94fcc..67de3a9499bb 100644 --- a/include/linux/pci-ats.h +++ b/include/linux/pci-ats.h @@ -4,73 +4,54 @@ #include -#ifdef CONFIG_PCI_PRI +#ifdef CONFIG_PCI_ATS +/* Address Translation Service */ +int pci_enable_ats(struct pci_dev *dev, int ps); +void pci_disable_ats(struct pci_dev *dev); +int pci_ats_queue_depth(struct pci_dev *dev); +int pci_ats_page_aligned(struct pci_dev *dev); +#else /* CONFIG_PCI_ATS */ +static inline int pci_enable_ats(struct pci_dev *d, int ps) +{ return -ENODEV; } +static inline void pci_disable_ats(struct pci_dev *d) { } +static inline int pci_ats_queue_depth(struct pci_dev *d) +{ return -ENODEV; } +static inline int pci_ats_page_aligned(struct pci_dev *dev) +{ return 0; } +#endif /* CONFIG_PCI_ATS */ +#ifdef CONFIG_PCI_PRI int pci_enable_pri(struct pci_dev *pdev, u32 reqs); void pci_disable_pri(struct pci_dev *pdev); void pci_restore_pri_state(struct pci_dev *pdev); int pci_reset_pri(struct pci_dev *pdev); int pci_prg_resp_pasid_required(struct pci_dev *pdev); - #else /* CONFIG_PCI_PRI */ - static inline int pci_enable_pri(struct pci_dev *pdev, u32 reqs) -{ - return -ENODEV; -} - -static inline void pci_disable_pri(struct pci_dev *pdev) -{ -} - -static inline void pci_restore_pri_state(struct pci_dev *pdev) -{ -} - +{ return -ENODEV; } +static inline void pci_disable_pri(struct pci_dev *pdev) { } +static inline void pci_restore_pri_state(struct pci_dev *pdev) { } static inline int pci_reset_pri(struct pci_dev *pdev) -{ - return -ENODEV; -} - +{ return -ENODEV; } static inline int pci_prg_resp_pasid_required(struct pci_dev *pdev) -{ - return 0; -} +{ return 0; } #endif /* CONFIG_PCI_PRI */ #ifdef CONFIG_PCI_PASID - int pci_enable_pasid(struct pci_dev *pdev, int features); void pci_disable_pasid(struct pci_dev *pdev); void pci_restore_pasid_state(struct pci_dev *pdev); int pci_pasid_features(struct pci_dev *pdev); int pci_max_pasids(struct pci_dev *pdev); - -#else /* CONFIG_PCI_PASID */ - +#else /* CONFIG_PCI_PASID */ static inline int pci_enable_pasid(struct pci_dev *pdev, int features) -{ - return -EINVAL; -} - -static inline void pci_disable_pasid(struct pci_dev *pdev) -{ -} - -static inline void pci_restore_pasid_state(struct pci_dev *pdev) -{ -} - +{ return -EINVAL; } +static inline void pci_disable_pasid(struct pci_dev *pdev) { } +static inline void pci_restore_pasid_state(struct pci_dev *pdev) { } static inline int pci_pasid_features(struct pci_dev *pdev) -{ - return -EINVAL; -} - +{ return -EINVAL; } static inline int pci_max_pasids(struct pci_dev *pdev) -{ - return -EINVAL; -} +{ return -EINVAL; } #endif /* CONFIG_PCI_PASID */ - -#endif /* LINUX_PCI_ATS_H*/ +#endif /* LINUX_PCI_ATS_H */ diff --git a/include/linux/pci.h b/include/linux/pci.h index 64d35e730fab..9fc22f48055e 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -284,7 +284,6 @@ struct irq_affinity; struct pcie_link_state; struct pci_vpd; struct pci_sriov; -struct pci_ats; struct pci_p2pdma; /* The pci_dev structure describes PCI devices */ @@ -1778,19 +1777,6 @@ pci_alloc_irq_vectors(struct pci_dev *dev, unsigned int min_vecs, NULL); } -#ifdef CONFIG_PCI_ATS -/* Address Translation Service */ -int pci_enable_ats(struct pci_dev *dev, int ps); -void pci_disable_ats(struct pci_dev *dev); -int pci_ats_queue_depth(struct pci_dev *dev); -int pci_ats_page_aligned(struct pci_dev *dev); -#else -static inline int pci_enable_ats(struct pci_dev *d, int ps) { return -ENODEV; } -static inline void pci_disable_ats(struct pci_dev *d) { } -static inline int pci_ats_queue_depth(struct pci_dev *d) { return -ENODEV; } -static inline int pci_ats_page_aligned(struct pci_dev *dev) { return 0; } -#endif - /* Include architecture-dependent settings and functions */ #include From c6e9aefbf9db818d60818aa5540d78c1da289aae Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 9 Oct 2019 16:27:30 -0500 Subject: [PATCH 033/160] PCI/ATS: Remove unused PRI and PASID stubs The following functions are only used by amd_iommu.c and intel-iommu.c (when CONFIG_INTEL_IOMMU_SVM is enabled). CONFIG_PCI_PRI and CONFIG_PCI_PASID are always defined in those cases, so there's no need for the stubs. pci_enable_pri() pci_disable_pri() pci_reset_pri() pci_prg_resp_pasid_required() pci_enable_pasid() pci_disable_pasid() Remove the unused stubs. Signed-off-by: Bjorn Helgaas Reviewed-by: Kuppuswamy Sathyanarayanan Reviewed-by: Joerg Roedel --- include/linux/pci-ats.h | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/include/linux/pci-ats.h b/include/linux/pci-ats.h index 67de3a9499bb..963c11f7c56b 100644 --- a/include/linux/pci-ats.h +++ b/include/linux/pci-ats.h @@ -27,14 +27,7 @@ void pci_restore_pri_state(struct pci_dev *pdev); int pci_reset_pri(struct pci_dev *pdev); int pci_prg_resp_pasid_required(struct pci_dev *pdev); #else /* CONFIG_PCI_PRI */ -static inline int pci_enable_pri(struct pci_dev *pdev, u32 reqs) -{ return -ENODEV; } -static inline void pci_disable_pri(struct pci_dev *pdev) { } static inline void pci_restore_pri_state(struct pci_dev *pdev) { } -static inline int pci_reset_pri(struct pci_dev *pdev) -{ return -ENODEV; } -static inline int pci_prg_resp_pasid_required(struct pci_dev *pdev) -{ return 0; } #endif /* CONFIG_PCI_PRI */ #ifdef CONFIG_PCI_PASID @@ -44,9 +37,6 @@ void pci_restore_pasid_state(struct pci_dev *pdev); int pci_pasid_features(struct pci_dev *pdev); int pci_max_pasids(struct pci_dev *pdev); #else /* CONFIG_PCI_PASID */ -static inline int pci_enable_pasid(struct pci_dev *pdev, int features) -{ return -EINVAL; } -static inline void pci_disable_pasid(struct pci_dev *pdev) { } static inline void pci_restore_pasid_state(struct pci_dev *pdev) { } static inline int pci_pasid_features(struct pci_dev *pdev) { return -EINVAL; } From d355bb2097834a977a6f47cec003b7d7748adbd6 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 9 Oct 2019 16:41:04 -0500 Subject: [PATCH 034/160] PCI/ATS: Remove unnecessary EXPORT_SYMBOL_GPL() The following functions are only used by the PCI core or by IOMMU drivers that cannot be modular, so there's no need to export them at all: pci_enable_ats() pci_disable_ats() pci_restore_ats_state() pci_ats_queue_depth() pci_ats_page_aligned() pci_enable_pri() pci_restore_pri_state() pci_reset_pri() pci_prg_resp_pasid_required() pci_enable_pasid() pci_disable_pasid() pci_restore_pasid_state() pci_pasid_features() pci_max_pasids() Remove the unnecessary EXPORT_SYMBOL_GPL()s. Signed-off-by: Bjorn Helgaas Reviewed-by: Joerg Roedel --- drivers/pci/ats.c | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/drivers/pci/ats.c b/drivers/pci/ats.c index 76ae518d55f4..982b46f0a54d 100644 --- a/drivers/pci/ats.c +++ b/drivers/pci/ats.c @@ -69,7 +69,6 @@ int pci_enable_ats(struct pci_dev *dev, int ps) dev->ats_enabled = 1; return 0; } -EXPORT_SYMBOL_GPL(pci_enable_ats); /** * pci_disable_ats - disable the ATS capability @@ -88,7 +87,6 @@ void pci_disable_ats(struct pci_dev *dev) dev->ats_enabled = 0; } -EXPORT_SYMBOL_GPL(pci_disable_ats); void pci_restore_ats_state(struct pci_dev *dev) { @@ -102,7 +100,6 @@ void pci_restore_ats_state(struct pci_dev *dev) ctrl |= PCI_ATS_CTRL_STU(dev->ats_stu - PCI_ATS_MIN_STU); pci_write_config_word(dev, dev->ats_cap + PCI_ATS_CTRL, ctrl); } -EXPORT_SYMBOL_GPL(pci_restore_ats_state); /** * pci_ats_queue_depth - query the ATS Invalidate Queue Depth @@ -129,7 +126,6 @@ int pci_ats_queue_depth(struct pci_dev *dev) pci_read_config_word(dev, dev->ats_cap + PCI_ATS_CAP, &cap); return PCI_ATS_CAP_QDEP(cap) ? PCI_ATS_CAP_QDEP(cap) : PCI_ATS_MAX_QDEP; } -EXPORT_SYMBOL_GPL(pci_ats_queue_depth); /** * pci_ats_page_aligned - Return Page Aligned Request bit status. @@ -156,7 +152,6 @@ int pci_ats_page_aligned(struct pci_dev *pdev) return 0; } -EXPORT_SYMBOL_GPL(pci_ats_page_aligned); #ifdef CONFIG_PCI_PRI void pci_pri_init(struct pci_dev *pdev) @@ -218,7 +213,6 @@ int pci_enable_pri(struct pci_dev *pdev, u32 reqs) return 0; } -EXPORT_SYMBOL_GPL(pci_enable_pri); /** * pci_disable_pri - Disable PRI capability @@ -271,7 +265,6 @@ void pci_restore_pri_state(struct pci_dev *pdev) pci_write_config_dword(pdev, pri + PCI_PRI_ALLOC_REQ, reqs); pci_write_config_word(pdev, pri + PCI_PRI_CTRL, control); } -EXPORT_SYMBOL_GPL(pci_restore_pri_state); /** * pci_reset_pri - Resets device's PRI state @@ -299,7 +292,6 @@ int pci_reset_pri(struct pci_dev *pdev) return 0; } -EXPORT_SYMBOL_GPL(pci_reset_pri); /** * pci_prg_resp_pasid_required - Return PRG Response PASID Required bit @@ -315,7 +307,6 @@ int pci_prg_resp_pasid_required(struct pci_dev *pdev) return pdev->pasid_required; } -EXPORT_SYMBOL_GPL(pci_prg_resp_pasid_required); #endif /* CONFIG_PCI_PRI */ #ifdef CONFIG_PCI_PASID @@ -373,7 +364,6 @@ int pci_enable_pasid(struct pci_dev *pdev, int features) return 0; } -EXPORT_SYMBOL_GPL(pci_enable_pasid); /** * pci_disable_pasid - Disable the PASID capability @@ -398,7 +388,6 @@ void pci_disable_pasid(struct pci_dev *pdev) pdev->pasid_enabled = 0; } -EXPORT_SYMBOL_GPL(pci_disable_pasid); /** * pci_restore_pasid_state - Restore PASID capabilities @@ -421,7 +410,6 @@ void pci_restore_pasid_state(struct pci_dev *pdev) control = PCI_PASID_CTRL_ENABLE | pdev->pasid_features; pci_write_config_word(pdev, pasid + PCI_PASID_CTRL, control); } -EXPORT_SYMBOL_GPL(pci_restore_pasid_state); /** * pci_pasid_features - Check which PASID features are supported @@ -450,7 +438,6 @@ int pci_pasid_features(struct pci_dev *pdev) return supported; } -EXPORT_SYMBOL_GPL(pci_pasid_features); #define PASID_NUMBER_SHIFT 8 #define PASID_NUMBER_MASK (0x1f << PASID_NUMBER_SHIFT) @@ -478,5 +465,4 @@ int pci_max_pasids(struct pci_dev *pdev) return (1 << supported); } -EXPORT_SYMBOL_GPL(pci_max_pasids); #endif /* CONFIG_PCI_PASID */ From fef2dd8b3966517172514ea5a89104ba7745678b Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 9 Oct 2019 16:47:15 -0500 Subject: [PATCH 035/160] PCI/ATS: Make pci_restore_pri_state(), pci_restore_pasid_state() private These interfaces: void pci_restore_pri_state(struct pci_dev *pdev); void pci_restore_pasid_state(struct pci_dev *pdev); are only used in drivers/pci and do not need to be seen by the rest of the kernel. Most them to drivers/pci/pci.h so they're private to the PCI subsystem. Signed-off-by: Bjorn Helgaas Reviewed-by: Joerg Roedel --- drivers/pci/pci.h | 4 ++++ include/linux/pci-ats.h | 5 ----- 2 files changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index ae84d28ba03a..e6b46d2b9846 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -458,14 +458,18 @@ static inline void pci_restore_ats_state(struct pci_dev *dev) { } #ifdef CONFIG_PCI_PRI void pci_pri_init(struct pci_dev *dev); +void pci_restore_pri_state(struct pci_dev *pdev); #else static inline void pci_pri_init(struct pci_dev *dev) { } +static inline void pci_restore_pri_state(struct pci_dev *pdev) { } #endif #ifdef CONFIG_PCI_PASID void pci_pasid_init(struct pci_dev *dev); +void pci_restore_pasid_state(struct pci_dev *pdev); #else static inline void pci_pasid_init(struct pci_dev *dev) { } +static inline void pci_restore_pasid_state(struct pci_dev *pdev) { } #endif #ifdef CONFIG_PCI_IOV diff --git a/include/linux/pci-ats.h b/include/linux/pci-ats.h index 963c11f7c56b..5d62e78946a3 100644 --- a/include/linux/pci-ats.h +++ b/include/linux/pci-ats.h @@ -23,21 +23,16 @@ static inline int pci_ats_page_aligned(struct pci_dev *dev) #ifdef CONFIG_PCI_PRI int pci_enable_pri(struct pci_dev *pdev, u32 reqs); void pci_disable_pri(struct pci_dev *pdev); -void pci_restore_pri_state(struct pci_dev *pdev); int pci_reset_pri(struct pci_dev *pdev); int pci_prg_resp_pasid_required(struct pci_dev *pdev); -#else /* CONFIG_PCI_PRI */ -static inline void pci_restore_pri_state(struct pci_dev *pdev) { } #endif /* CONFIG_PCI_PRI */ #ifdef CONFIG_PCI_PASID int pci_enable_pasid(struct pci_dev *pdev, int features); void pci_disable_pasid(struct pci_dev *pdev); -void pci_restore_pasid_state(struct pci_dev *pdev); int pci_pasid_features(struct pci_dev *pdev); int pci_max_pasids(struct pci_dev *pdev); #else /* CONFIG_PCI_PASID */ -static inline void pci_restore_pasid_state(struct pci_dev *pdev) { } static inline int pci_pasid_features(struct pci_dev *pdev) { return -EINVAL; } static inline int pci_max_pasids(struct pci_dev *pdev) From d8558ac8c93d429d65d7490b512a3a67e559d0d4 Mon Sep 17 00:00:00 2001 From: Steffen Liebergeld Date: Wed, 18 Sep 2019 15:16:52 +0200 Subject: [PATCH 036/160] PCI: Fix Intel ACS quirk UPDCR register address According to documentation [0] the correct offset for the Upstream Peer Decode Configuration Register (UPDCR) is 0x1014. It was previously defined as 0x1114. d99321b63b1f ("PCI: Enable quirks for PCIe ACS on Intel PCH root ports") intended to enforce isolation between PCI devices allowing them to be put into separate IOMMU groups. Due to the wrong register offset the intended isolation was not fully enforced. This is fixed with this patch. Please note that I did not test this patch because I have no hardware that implements this register. [0] https://www.intel.com/content/dam/www/public/us/en/documents/datasheets/4th-gen-core-family-mobile-i-o-datasheet.pdf (page 325) Fixes: d99321b63b1f ("PCI: Enable quirks for PCIe ACS on Intel PCH root ports") Link: https://lore.kernel.org/r/7a3505df-79ba-8a28-464c-88b83eefffa6@kernkonzept.com Signed-off-by: Steffen Liebergeld Signed-off-by: Bjorn Helgaas Reviewed-by: Andrew Murray Acked-by: Ashok Raj Cc: stable@vger.kernel.org # v3.15+ --- drivers/pci/quirks.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 320255e5e8f8..cd3e84ae742e 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -4706,7 +4706,7 @@ int pci_dev_specific_acs_enabled(struct pci_dev *dev, u16 acs_flags) #define INTEL_BSPR_REG_BPPD (1 << 9) /* Upstream Peer Decode Configuration Register */ -#define INTEL_UPDCR_REG 0x1114 +#define INTEL_UPDCR_REG 0x1014 /* 5:0 Peer Decode Enable bits */ #define INTEL_UPDCR_REG_MASK 0x3f From 56b4cd4b7da9ee95778eb5c8abea49f641ebfd91 Mon Sep 17 00:00:00 2001 From: Slawomir Pawlowski Date: Tue, 17 Sep 2019 09:20:48 +0000 Subject: [PATCH 037/160] PCI: Add DMA alias quirk for Intel VCA NTB Intel Visual Compute Accelerator (VCA) is a family of PCIe add-in devices exposing computational units via Non Transparent Bridges (NTB, PEX 87xx). Similarly to MIC x200, we need to add DMA aliases to allow buffer access when IOMMU is enabled. Add aliases to allow computational unit access to host memory. These aliases mark the whole VCA device as one IOMMU group. All possible slot numbers (0x20) are used, since we are unable to tell what slot is used on other side. This quirk is intended for both host and computational unit sides. The VCA devices have up to five functions: four for DMA channels and one additional. Link: https://lore.kernel.org/r/5683A335CC8BE1438C3C30C49DCC38DF637CED8E@IRSMSX102.ger.corp.intel.com Signed-off-by: Slawomir Pawlowski Signed-off-by: Przemek Kitszel Signed-off-by: Bjorn Helgaas --- drivers/pci/quirks.c | 34 ++++++++++++++++++++++++++++++++++ 1 file changed, 34 insertions(+) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index cd3e84ae742e..d5d57cd91a5e 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -4080,6 +4080,40 @@ static void quirk_mic_x200_dma_alias(struct pci_dev *pdev) DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2260, quirk_mic_x200_dma_alias); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2264, quirk_mic_x200_dma_alias); +/* + * Intel Visual Compute Accelerator (VCA) is a family of PCIe add-in devices + * exposing computational units via Non Transparent Bridges (NTB, PEX 87xx). + * + * Similarly to MIC x200, we need to add DMA aliases to allow buffer access + * when IOMMU is enabled. These aliases allow computational unit access to + * host memory. These aliases mark the whole VCA device as one IOMMU + * group. + * + * All possible slot numbers (0x20) are used, since we are unable to tell + * what slot is used on other side. This quirk is intended for both host + * and computational unit sides. The VCA devices have up to five functions + * (four for DMA channels and one additional). + */ +static void quirk_pex_vca_alias(struct pci_dev *pdev) +{ + const unsigned int num_pci_slots = 0x20; + unsigned int slot; + + for (slot = 0; slot < num_pci_slots; slot++) { + pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x0)); + pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x1)); + pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x2)); + pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x3)); + pci_add_dma_alias(pdev, PCI_DEVFN(slot, 0x4)); + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2954, quirk_pex_vca_alias); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2955, quirk_pex_vca_alias); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2956, quirk_pex_vca_alias); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2958, quirk_pex_vca_alias); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2959, quirk_pex_vca_alias); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x295A, quirk_pex_vca_alias); + /* * The IOMMU and interrupt controller on Broadcom Vulcan/Cavium ThunderX2 are * associated not at the root bus, but at a bridge below. This quirk avoids From 35ff867b76576e32f34c698ccd11343f7d616204 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Pierre=20Cr=C3=A9gut?= Date: Wed, 11 Sep 2019 09:27:36 +0200 Subject: [PATCH 038/160] PCI/IOV: Serialize sysfs sriov_numvfs reads vs writes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When sriov_numvfs is being updated, we call the driver->sriov_configure() function, which may enable VFs and call probe functions, which may make new devices visible. This all happens before before sriov_numvfs_store() updates sriov->num_VFs, so previously, concurrent sysfs reads of sriov_numvfs returned stale values. Serialize the sysfs read vs the write so the read returns the correct num_VFs value. [bhelgaas: hold device_lock instead of checking mutex_is_locked()] Link: https://bugzilla.kernel.org/show_bug.cgi?id=202991 Link: https://lore.kernel.org/r/20190911072736.32091-1-pierre.cregut@orange.com Signed-off-by: Pierre Crégut Signed-off-by: Bjorn Helgaas --- drivers/pci/iov.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c index b3f972e8cfed..1d3de1ea081d 100644 --- a/drivers/pci/iov.c +++ b/drivers/pci/iov.c @@ -254,8 +254,14 @@ static ssize_t sriov_numvfs_show(struct device *dev, char *buf) { struct pci_dev *pdev = to_pci_dev(dev); + u16 num_vfs; - return sprintf(buf, "%u\n", pdev->sriov->num_VFs); + /* Serialize vs sriov_numvfs_store() so readers see valid num_VFs */ + device_lock(&pdev->dev); + num_vfs = pdev->sriov->num_VFs; + device_unlock(&pdev->dev); + + return sprintf(buf, "%u\n", num_vfs); } /* From a243bf39d8be133f9d6d8c8b2565830c6d3d247b Mon Sep 17 00:00:00 2001 From: "Ben Dooks (Codethink)" Date: Tue, 15 Oct 2019 17:07:02 +0100 Subject: [PATCH 039/160] PCI: iproc-msi: Fix __iomem annotation in decode_msi_hwirq() Fix __iomem attribute on msg variable passed to readl() in the decode_msi_hwirq() function. Fixes the following sparse warning: drivers/pci/controller/pcie-iproc-msi.c:301:17: warning: incorrect type in argument 1 (different address spaces) drivers/pci/controller/pcie-iproc-msi.c:301:17: expected void const volatile [noderef] *addr drivers/pci/controller/pcie-iproc-msi.c:301:17: got unsigned int [usertype] *[assigned] msg Signed-off-by: Ben Dooks Signed-off-by: Lorenzo Pieralisi Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas Cc: Ray Jui Cc: Scott Branden Cc: bcm-kernel-feedback-list@broadcom.com Cc: linux-pci@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org Cc: linux-kernel@vger.kernel.org --- drivers/pci/controller/pcie-iproc-msi.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/pci/controller/pcie-iproc-msi.c b/drivers/pci/controller/pcie-iproc-msi.c index 0a3f61be5625..3176ad3ab0e5 100644 --- a/drivers/pci/controller/pcie-iproc-msi.c +++ b/drivers/pci/controller/pcie-iproc-msi.c @@ -293,11 +293,12 @@ static const struct irq_domain_ops msi_domain_ops = { static inline u32 decode_msi_hwirq(struct iproc_msi *msi, u32 eq, u32 head) { - u32 *msg, hwirq; + u32 __iomem *msg; + u32 hwirq; unsigned int offs; offs = iproc_msi_eq_offset(msi, eq) + head * sizeof(u32); - msg = (u32 *)(msi->eq_cpu + offs); + msg = (u32 __iomem *)(msi->eq_cpu + offs); hwirq = readl(msg); hwirq = (hwirq >> 5) + (hwirq & 0x1f); From 8990e381d1888f82ca65a31469f2327cf1f56d52 Mon Sep 17 00:00:00 2001 From: "Ben Dooks (Codethink)" Date: Tue, 15 Oct 2019 17:09:30 +0100 Subject: [PATCH 040/160] PCI: mvebu: Make mvebu_pci_bridge_emul_ops static The mvebu_pci_bridge_emul_ops is not exported outside of the driver, so make it static to avoid the following sparse warning: drivers/pci/controller/pci-mvebu.c:557:28: warning: symbol 'mvebu_pci_bridge_emul_ops' was not declared. Should it be static? Signed-off-by: Ben Dooks Signed-off-by: Lorenzo Pieralisi Cc: Thomas Petazzoni Cc: Jason Cooper Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas Cc: linux-pci@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org Cc: linux-kernel@vger.kernel.org --- drivers/pci/controller/pci-mvebu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/controller/pci-mvebu.c b/drivers/pci/controller/pci-mvebu.c index d3a0419e42f2..ed032e9a3156 100644 --- a/drivers/pci/controller/pci-mvebu.c +++ b/drivers/pci/controller/pci-mvebu.c @@ -554,7 +554,7 @@ mvebu_pci_bridge_emul_pcie_conf_write(struct pci_bridge_emul *bridge, } } -struct pci_bridge_emul_ops mvebu_pci_bridge_emul_ops = { +static struct pci_bridge_emul_ops mvebu_pci_bridge_emul_ops = { .write_base = mvebu_pci_bridge_emul_base_conf_write, .read_pcie = mvebu_pci_bridge_emul_pcie_conf_read, .write_pcie = mvebu_pci_bridge_emul_pcie_conf_write, From 80aed7dc6d36748a8b125c12ca4abba1f2b17664 Mon Sep 17 00:00:00 2001 From: "Ben Dooks (Codethink)" Date: Tue, 15 Oct 2019 17:11:48 +0100 Subject: [PATCH 041/160] PCI: mvebu: mvebu_pcie_map_registers __iomem fix Fix the return type of mvebu_pcie_map_registers in the error path to have __iomem on it. Fixes the following sparse warning: drivers/pci/controller/pci-mvebu.c:716:31: warning: incorrect type in return expression (different address spaces) drivers/pci/controller/pci-mvebu.c:716:31: expected void [noderef] * drivers/pci/controller/pci-mvebu.c:716:31: got void * Signed-off-by: Ben Dooks Signed-off-by: Lorenzo Pieralisi Cc: Thomas Petazzoni Cc: Jason Cooper Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas Cc: linux-pci@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org Cc: linux-kernel@vger.kernel.org --- drivers/pci/controller/pci-mvebu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/controller/pci-mvebu.c b/drivers/pci/controller/pci-mvebu.c index ed032e9a3156..153a64676bc9 100644 --- a/drivers/pci/controller/pci-mvebu.c +++ b/drivers/pci/controller/pci-mvebu.c @@ -713,7 +713,7 @@ static void __iomem *mvebu_pcie_map_registers(struct platform_device *pdev, ret = of_address_to_resource(np, 0, ®s); if (ret) - return ERR_PTR(ret); + return (void __iomem *)ERR_PTR(ret); return devm_ioremap_resource(&pdev->dev, ®s); } From e078723f9cccd509482fd7f30a4afb1125ca7a2a Mon Sep 17 00:00:00 2001 From: Grzegorz Jaszczyk Date: Tue, 16 Jul 2019 14:12:07 +0200 Subject: [PATCH 042/160] PCI: aardvark: Fix big endian support Initialise every multiple-byte field of emulated PCI bridge config space with proper cpu_to_le* macro. This is required since the structure describing config space of emulated bridge assumes little-endian convention. Signed-off-by: Grzegorz Jaszczyk Signed-off-by: Lorenzo Pieralisi --- drivers/pci/controller/pci-aardvark.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c index 79be0afc9b1e..9f56f3bee39d 100644 --- a/drivers/pci/controller/pci-aardvark.c +++ b/drivers/pci/controller/pci-aardvark.c @@ -519,18 +519,20 @@ static void advk_sw_pci_bridge_init(struct advk_pcie *pcie) { struct pci_bridge_emul *bridge = &pcie->bridge; - bridge->conf.vendor = advk_readl(pcie, PCIE_CORE_DEV_ID_REG) & 0xffff; - bridge->conf.device = advk_readl(pcie, PCIE_CORE_DEV_ID_REG) >> 16; + bridge->conf.vendor = + cpu_to_le16(advk_readl(pcie, PCIE_CORE_DEV_ID_REG) & 0xffff); + bridge->conf.device = + cpu_to_le16(advk_readl(pcie, PCIE_CORE_DEV_ID_REG) >> 16); bridge->conf.class_revision = - advk_readl(pcie, PCIE_CORE_DEV_REV_REG) & 0xff; + cpu_to_le32(advk_readl(pcie, PCIE_CORE_DEV_REV_REG) & 0xff); /* Support 32 bits I/O addressing */ bridge->conf.iobase = PCI_IO_RANGE_TYPE_32; bridge->conf.iolimit = PCI_IO_RANGE_TYPE_32; /* Support 64 bits memory pref */ - bridge->conf.pref_mem_base = PCI_PREF_RANGE_TYPE_64; - bridge->conf.pref_mem_limit = PCI_PREF_RANGE_TYPE_64; + bridge->conf.pref_mem_base = cpu_to_le16(PCI_PREF_RANGE_TYPE_64); + bridge->conf.pref_mem_limit = cpu_to_le16(PCI_PREF_RANGE_TYPE_64); /* Support interrupt A for MSI feature */ bridge->conf.intpin = PCIE_CORE_INT_A_ASSERT_ENABLE; From e0d9d30b73548fbfe5c024ed630169bdc9a08aee Mon Sep 17 00:00:00 2001 From: Grzegorz Jaszczyk Date: Tue, 16 Jul 2019 14:13:46 +0200 Subject: [PATCH 043/160] PCI: pci-bridge-emul: Fix big-endian support Perform conversion to little-endian before every write to configuration space and convert it back to CPU endianness on reads. Additionally, initialise every multiple byte field of config space with the cpu_to_le* macro, which is required since the structure describing config space of emulated bridge assumes little-endian convention. Signed-off-by: Grzegorz Jaszczyk Signed-off-by: Lorenzo Pieralisi --- drivers/pci/pci-bridge-emul.c | 25 +++++------ drivers/pci/pci-bridge-emul.h | 78 +++++++++++++++++------------------ 2 files changed, 52 insertions(+), 51 deletions(-) diff --git a/drivers/pci/pci-bridge-emul.c b/drivers/pci/pci-bridge-emul.c index 5fd90105510d..fffa77093c08 100644 --- a/drivers/pci/pci-bridge-emul.c +++ b/drivers/pci/pci-bridge-emul.c @@ -270,10 +270,10 @@ static const struct pci_bridge_reg_behavior pcie_cap_regs_behavior[] = { int pci_bridge_emul_init(struct pci_bridge_emul *bridge, unsigned int flags) { - bridge->conf.class_revision |= PCI_CLASS_BRIDGE_PCI << 16; + bridge->conf.class_revision |= cpu_to_le32(PCI_CLASS_BRIDGE_PCI << 16); bridge->conf.header_type = PCI_HEADER_TYPE_BRIDGE; bridge->conf.cache_line_size = 0x10; - bridge->conf.status = PCI_STATUS_CAP_LIST; + bridge->conf.status = cpu_to_le16(PCI_STATUS_CAP_LIST); bridge->pci_regs_behavior = kmemdup(pci_regs_behavior, sizeof(pci_regs_behavior), GFP_KERNEL); @@ -284,8 +284,9 @@ int pci_bridge_emul_init(struct pci_bridge_emul *bridge, bridge->conf.capabilities_pointer = PCI_CAP_PCIE_START; bridge->pcie_conf.cap_id = PCI_CAP_ID_EXP; /* Set PCIe v2, root port, slot support */ - bridge->pcie_conf.cap = PCI_EXP_TYPE_ROOT_PORT << 4 | 2 | - PCI_EXP_FLAGS_SLOT; + bridge->pcie_conf.cap = + cpu_to_le16(PCI_EXP_TYPE_ROOT_PORT << 4 | 2 | + PCI_EXP_FLAGS_SLOT); bridge->pcie_cap_regs_behavior = kmemdup(pcie_cap_regs_behavior, sizeof(pcie_cap_regs_behavior), @@ -327,7 +328,7 @@ int pci_bridge_emul_conf_read(struct pci_bridge_emul *bridge, int where, int reg = where & ~3; pci_bridge_emul_read_status_t (*read_op)(struct pci_bridge_emul *bridge, int reg, u32 *value); - u32 *cfgspace; + __le32 *cfgspace; const struct pci_bridge_reg_behavior *behavior; if (bridge->has_pcie && reg >= PCI_CAP_PCIE_END) { @@ -343,11 +344,11 @@ int pci_bridge_emul_conf_read(struct pci_bridge_emul *bridge, int where, if (bridge->has_pcie && reg >= PCI_CAP_PCIE_START) { reg -= PCI_CAP_PCIE_START; read_op = bridge->ops->read_pcie; - cfgspace = (u32 *) &bridge->pcie_conf; + cfgspace = (__le32 *) &bridge->pcie_conf; behavior = bridge->pcie_cap_regs_behavior; } else { read_op = bridge->ops->read_base; - cfgspace = (u32 *) &bridge->conf; + cfgspace = (__le32 *) &bridge->conf; behavior = bridge->pci_regs_behavior; } @@ -357,7 +358,7 @@ int pci_bridge_emul_conf_read(struct pci_bridge_emul *bridge, int where, ret = PCI_BRIDGE_EMUL_NOT_HANDLED; if (ret == PCI_BRIDGE_EMUL_NOT_HANDLED) - *value = cfgspace[reg / 4]; + *value = le32_to_cpu(cfgspace[reg / 4]); /* * Make sure we never return any reserved bit with a value @@ -387,7 +388,7 @@ int pci_bridge_emul_conf_write(struct pci_bridge_emul *bridge, int where, int mask, ret, old, new, shift; void (*write_op)(struct pci_bridge_emul *bridge, int reg, u32 old, u32 new, u32 mask); - u32 *cfgspace; + __le32 *cfgspace; const struct pci_bridge_reg_behavior *behavior; if (bridge->has_pcie && reg >= PCI_CAP_PCIE_END) @@ -414,11 +415,11 @@ int pci_bridge_emul_conf_write(struct pci_bridge_emul *bridge, int where, if (bridge->has_pcie && reg >= PCI_CAP_PCIE_START) { reg -= PCI_CAP_PCIE_START; write_op = bridge->ops->write_pcie; - cfgspace = (u32 *) &bridge->pcie_conf; + cfgspace = (__le32 *) &bridge->pcie_conf; behavior = bridge->pcie_cap_regs_behavior; } else { write_op = bridge->ops->write_base; - cfgspace = (u32 *) &bridge->conf; + cfgspace = (__le32 *) &bridge->conf; behavior = bridge->pci_regs_behavior; } @@ -431,7 +432,7 @@ int pci_bridge_emul_conf_write(struct pci_bridge_emul *bridge, int where, /* Clear the W1C bits */ new &= ~((value << shift) & (behavior[reg / 4].w1c & mask)); - cfgspace[reg / 4] = new; + cfgspace[reg / 4] = cpu_to_le32(new); if (write_op) write_op(bridge, reg, old, new, mask); diff --git a/drivers/pci/pci-bridge-emul.h b/drivers/pci/pci-bridge-emul.h index e65b1b79899d..b31883022a8e 100644 --- a/drivers/pci/pci-bridge-emul.h +++ b/drivers/pci/pci-bridge-emul.h @@ -6,65 +6,65 @@ /* PCI configuration space of a PCI-to-PCI bridge. */ struct pci_bridge_emul_conf { - u16 vendor; - u16 device; - u16 command; - u16 status; - u32 class_revision; + __le16 vendor; + __le16 device; + __le16 command; + __le16 status; + __le32 class_revision; u8 cache_line_size; u8 latency_timer; u8 header_type; u8 bist; - u32 bar[2]; + __le32 bar[2]; u8 primary_bus; u8 secondary_bus; u8 subordinate_bus; u8 secondary_latency_timer; u8 iobase; u8 iolimit; - u16 secondary_status; - u16 membase; - u16 memlimit; - u16 pref_mem_base; - u16 pref_mem_limit; - u32 prefbaseupper; - u32 preflimitupper; - u16 iobaseupper; - u16 iolimitupper; + __le16 secondary_status; + __le16 membase; + __le16 memlimit; + __le16 pref_mem_base; + __le16 pref_mem_limit; + __le32 prefbaseupper; + __le32 preflimitupper; + __le16 iobaseupper; + __le16 iolimitupper; u8 capabilities_pointer; u8 reserve[3]; - u32 romaddr; + __le32 romaddr; u8 intline; u8 intpin; - u16 bridgectrl; + __le16 bridgectrl; }; /* PCI configuration space of the PCIe capabilities */ struct pci_bridge_emul_pcie_conf { u8 cap_id; u8 next; - u16 cap; - u32 devcap; - u16 devctl; - u16 devsta; - u32 lnkcap; - u16 lnkctl; - u16 lnksta; - u32 slotcap; - u16 slotctl; - u16 slotsta; - u16 rootctl; - u16 rsvd; - u32 rootsta; - u32 devcap2; - u16 devctl2; - u16 devsta2; - u32 lnkcap2; - u16 lnkctl2; - u16 lnksta2; - u32 slotcap2; - u16 slotctl2; - u16 slotsta2; + __le16 cap; + __le32 devcap; + __le16 devctl; + __le16 devsta; + __le32 lnkcap; + __le16 lnkctl; + __le16 lnksta; + __le32 slotcap; + __le16 slotctl; + __le16 slotsta; + __le16 rootctl; + __le16 rsvd; + __le32 rootsta; + __le32 devcap2; + __le16 devctl2; + __le16 devsta2; + __le32 lnkcap2; + __le16 lnkctl2; + __le16 lnksta2; + __le32 slotcap2; + __le16 slotctl2; + __le16 slotsta2; }; struct pci_bridge_emul; From 35a82a3795105eb2ccdb206659343d82fefbf51f Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Wed, 16 Oct 2019 14:02:49 +0200 Subject: [PATCH 044/160] MAINTAINERS: Add Marek and Shimoda-san as R-Car PCIE co-maintainers At the end of the v5.3 upstream development cycle I stepped down from my role at Renesas. Pass maintainership of the R-Car PCIE to Marek and Shimoda-san. Signed-off-by: Simon Horman Signed-off-by: Lorenzo Pieralisi Acked-by: Yoshihiro Shimoda Acked-by: Marek Vasut --- MAINTAINERS | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 296de2b51c83..53a4b003895f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -12485,7 +12485,8 @@ F: Documentation/devicetree/bindings/pci/nvidia,tegra20-pcie.txt F: drivers/pci/controller/pci-tegra.c PCI DRIVER FOR RENESAS R-CAR -M: Simon Horman +M: Marek Vasut +M: Yoshihiro Shimoda L: linux-pci@vger.kernel.org L: linux-renesas-soc@vger.kernel.org S: Maintained From eceb86028d23f19d88202153de13201b92300b27 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 1 Oct 2019 17:51:59 -0500 Subject: [PATCH 045/160] PCI: Remove unnecessary includes Remove unnecessary includes. Signed-off-by: Bjorn Helgaas --- drivers/pci/iov.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c index b3f972e8cfed..8165f257ec1b 100644 --- a/drivers/pci/iov.c +++ b/drivers/pci/iov.c @@ -9,7 +9,6 @@ #include #include -#include #include #include #include From 9d8b738bb9f8008848b62f58671de3537a6d0734 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 3 Oct 2019 16:28:26 -0500 Subject: [PATCH 046/160] PCI: Remove useless comments and tidy others Remove useless comments and tidy others for better readability. Whitespace changes only. Signed-off-by: Bjorn Helgaas --- drivers/pci/probe.c | 37 +++++++++---------------------------- 1 file changed, 9 insertions(+), 28 deletions(-) diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 3d5271a7a849..4afe327df7ae 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -2300,8 +2300,7 @@ void pcie_report_downtraining(struct pci_dev *dev) static void pci_init_capabilities(struct pci_dev *dev) { - /* Enhanced Allocation */ - pci_ea_init(dev); + pci_ea_init(dev); /* Enhanced Allocation */ /* Setup MSI caps & disable MSI/MSI-X interrupts */ pci_msi_setup_pci_dev(dev); @@ -2309,29 +2308,14 @@ static void pci_init_capabilities(struct pci_dev *dev) /* Buffers for saving PCIe and PCI-X capabilities */ pci_allocate_cap_save_buffers(dev); - /* Power Management */ - pci_pm_init(dev); - - /* Vital Product Data */ - pci_vpd_init(dev); - - /* Alternative Routing-ID Forwarding */ - pci_configure_ari(dev); - - /* Single Root I/O Virtualization */ - pci_iov_init(dev); - - /* Address Translation Services */ - pci_ats_init(dev); - - /* Enable ACS P2P upstream forwarding */ - pci_enable_acs(dev); - - /* Precision Time Measurement */ - pci_ptm_init(dev); - - /* Advanced Error Reporting */ - pci_aer_init(dev); + pci_pm_init(dev); /* Power Management */ + pci_vpd_init(dev); /* Vital Product Data */ + pci_configure_ari(dev); /* Alternative Routing-ID Forwarding */ + pci_iov_init(dev); /* Single Root I/O Virtualization */ + pci_ats_init(dev); /* Address Translation Services */ + pci_enable_acs(dev); /* Enable ACS P2P upstream forwarding */ + pci_ptm_init(dev); /* Precision Time Measurement */ + pci_aer_init(dev); /* Advanced Error Reporting */ pcie_report_downtraining(dev); @@ -2403,13 +2387,10 @@ void pci_device_add(struct pci_dev *dev, struct pci_bus *bus) /* Fix up broken headers */ pci_fixup_device(pci_fixup_header, dev); - /* Moved out from quirk header fixup code */ pci_reassigndev_resource_alignment(dev); - /* Clear the state_saved flag */ dev->state_saved = false; - /* Initialize various capabilities */ pci_init_capabilities(dev); /* From ff5c2bb9c6f5ee461293e337754360836b05b7f4 Mon Sep 17 00:00:00 2001 From: Vidya Sagar Date: Sat, 5 Oct 2019 22:12:11 +0530 Subject: [PATCH 047/160] PCI: tegra: Fix CLKREQ dependency programming Corrects the programming to provide REFCLK to the downstream device when there is no CLKREQ sideband signal routing present from root port to the endpont. Signed-off-by: Vidya Sagar Signed-off-by: Lorenzo Pieralisi Acked-by: Thierry Reding --- drivers/pci/controller/dwc/pcie-tegra194.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/pci/controller/dwc/pcie-tegra194.c b/drivers/pci/controller/dwc/pcie-tegra194.c index f89f5acee72d..cbe95f0ea0ca 100644 --- a/drivers/pci/controller/dwc/pcie-tegra194.c +++ b/drivers/pci/controller/dwc/pcie-tegra194.c @@ -40,8 +40,6 @@ #define APPL_PINMUX_CLKREQ_OVERRIDE BIT(3) #define APPL_PINMUX_CLK_OUTPUT_IN_OVERRIDE_EN BIT(4) #define APPL_PINMUX_CLK_OUTPUT_IN_OVERRIDE BIT(5) -#define APPL_PINMUX_CLKREQ_OUT_OVRD_EN BIT(9) -#define APPL_PINMUX_CLKREQ_OUT_OVRD BIT(10) #define APPL_CTRL 0x4 #define APPL_CTRL_SYS_PRE_DET_STATE BIT(6) @@ -1193,8 +1191,8 @@ static int tegra_pcie_config_controller(struct tegra_pcie_dw *pcie, if (!pcie->supports_clkreq) { val = appl_readl(pcie, APPL_PINMUX); - val |= APPL_PINMUX_CLKREQ_OUT_OVRD_EN; - val |= APPL_PINMUX_CLKREQ_OUT_OVRD; + val |= APPL_PINMUX_CLKREQ_OVERRIDE_EN; + val &= ~APPL_PINMUX_CLKREQ_OVERRIDE; appl_writel(pcie, val, APPL_PINMUX); } From af65d1ad416bc6e069ccb9e649faeda224248f96 Mon Sep 17 00:00:00 2001 From: "Patel, Mayurkumar" Date: Fri, 18 Oct 2019 16:52:21 +0000 Subject: [PATCH 048/160] PCI/AER: Save AER Capability for suspend/resume Previously we did not save and restore the AER configuration on suspend/resume, so the configuration may be lost after resume. Save the AER configuration during suspend and restore it during resume. [bhelgaas: commit log] Link: https://lore.kernel.org/r/92EBB4272BF81E4089A7126EC1E7B28492C3B007@IRSMSX101.ger.corp.intel.com Signed-off-by: Mayurkumar Patel Signed-off-by: Kuppuswamy Sathyanarayanan Signed-off-by: Bjorn Helgaas Reviewed-by: Andy Shevchenko --- drivers/pci/access.c | 2 +- drivers/pci/pci.c | 2 ++ drivers/pci/pci.h | 1 + drivers/pci/pcie/aer.c | 64 ++++++++++++++++++++++++++++++++++++++++-- include/linux/aer.h | 4 +++ 5 files changed, 69 insertions(+), 4 deletions(-) diff --git a/drivers/pci/access.c b/drivers/pci/access.c index 2fccb5762c76..79c4a2ef269a 100644 --- a/drivers/pci/access.c +++ b/drivers/pci/access.c @@ -355,7 +355,7 @@ static inline bool pcie_cap_has_sltctl(const struct pci_dev *dev) pcie_caps_reg(dev) & PCI_EXP_FLAGS_SLOT; } -static inline bool pcie_cap_has_rtctl(const struct pci_dev *dev) +bool pcie_cap_has_rtctl(const struct pci_dev *dev) { int type = pci_pcie_type(dev); diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index e7982af9a5d8..d02636540ca1 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1361,6 +1361,7 @@ int pci_save_state(struct pci_dev *dev) pci_save_ltr_state(dev); pci_save_dpc_state(dev); + pci_save_aer_state(dev); return pci_save_vc_state(dev); } EXPORT_SYMBOL(pci_save_state); @@ -1474,6 +1475,7 @@ void pci_restore_state(struct pci_dev *dev) pci_restore_dpc_state(dev); pci_cleanup_aer_error_status_regs(dev); + pci_restore_aer_state(dev); pci_restore_config_space(dev); diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 3f6947ee3324..b96988f90315 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -12,6 +12,7 @@ extern const unsigned char pcie_link_speed[]; extern bool pci_early_dump; bool pcie_cap_has_lnkctl(const struct pci_dev *dev); +bool pcie_cap_has_rtctl(const struct pci_dev *dev); /* Functions internal to the PCI core code */ diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c index b45bc47d04fe..443512882be4 100644 --- a/drivers/pci/pcie/aer.c +++ b/drivers/pci/pcie/aer.c @@ -448,12 +448,70 @@ int pci_cleanup_aer_error_status_regs(struct pci_dev *dev) return 0; } +void pci_save_aer_state(struct pci_dev *dev) +{ + struct pci_cap_saved_state *save_state; + u32 *cap; + int pos; + + pos = dev->aer_cap; + if (!pos) + return; + + save_state = pci_find_saved_ext_cap(dev, PCI_EXT_CAP_ID_ERR); + if (!save_state) + return; + + cap = &save_state->cap.data[0]; + pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_MASK, cap++); + pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_SEVER, cap++); + pci_read_config_dword(dev, pos + PCI_ERR_COR_MASK, cap++); + pci_read_config_dword(dev, pos + PCI_ERR_CAP, cap++); + if (pcie_cap_has_rtctl(dev)) + pci_read_config_dword(dev, pos + PCI_ERR_ROOT_COMMAND, cap++); +} + +void pci_restore_aer_state(struct pci_dev *dev) +{ + struct pci_cap_saved_state *save_state; + u32 *cap; + int pos; + + pos = dev->aer_cap; + if (!pos) + return; + + save_state = pci_find_saved_ext_cap(dev, PCI_EXT_CAP_ID_ERR); + if (!save_state) + return; + + cap = &save_state->cap.data[0]; + pci_write_config_dword(dev, pos + PCI_ERR_UNCOR_MASK, *cap++); + pci_write_config_dword(dev, pos + PCI_ERR_UNCOR_SEVER, *cap++); + pci_write_config_dword(dev, pos + PCI_ERR_COR_MASK, *cap++); + pci_write_config_dword(dev, pos + PCI_ERR_CAP, *cap++); + if (pcie_cap_has_rtctl(dev)) + pci_write_config_dword(dev, pos + PCI_ERR_ROOT_COMMAND, *cap++); +} + void pci_aer_init(struct pci_dev *dev) { - dev->aer_cap = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR); + int n; - if (dev->aer_cap) - dev->aer_stats = kzalloc(sizeof(struct aer_stats), GFP_KERNEL); + dev->aer_cap = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR); + if (!dev->aer_cap) + return; + + dev->aer_stats = kzalloc(sizeof(struct aer_stats), GFP_KERNEL); + + /* + * We save/restore PCI_ERR_UNCOR_MASK, PCI_ERR_UNCOR_SEVER, + * PCI_ERR_COR_MASK, and PCI_ERR_CAP. Root and Root Complex Event + * Collectors also implement PCI_ERR_ROOT_COMMAND (PCIe r5.0, sec + * 7.8.4). + */ + n = pcie_cap_has_rtctl(dev) ? 5 : 4; + pci_add_ext_cap_save_buffer(dev, PCI_EXT_CAP_ID_ERR, sizeof(u32) * n); pci_cleanup_aer_error_status_regs(dev); } diff --git a/include/linux/aer.h b/include/linux/aer.h index 514bffa11dbb..fa19e01f418a 100644 --- a/include/linux/aer.h +++ b/include/linux/aer.h @@ -46,6 +46,8 @@ int pci_enable_pcie_error_reporting(struct pci_dev *dev); int pci_disable_pcie_error_reporting(struct pci_dev *dev); int pci_cleanup_aer_uncorrect_error_status(struct pci_dev *dev); int pci_cleanup_aer_error_status_regs(struct pci_dev *dev); +void pci_save_aer_state(struct pci_dev *dev); +void pci_restore_aer_state(struct pci_dev *dev); #else static inline int pci_enable_pcie_error_reporting(struct pci_dev *dev) { @@ -63,6 +65,8 @@ static inline int pci_cleanup_aer_error_status_regs(struct pci_dev *dev) { return -EINVAL; } +static inline void pci_save_aer_state(struct pci_dev *dev) {} +static inline void pci_restore_aer_state(struct pci_dev *dev) {} #endif void cper_print_aer(struct pci_dev *dev, int aer_severity, From 6458b438ebc12bec732290bf80c53c4eeeaed1c0 Mon Sep 17 00:00:00 2001 From: Rajat Jain Date: Tue, 27 Aug 2019 15:21:44 -0700 Subject: [PATCH 049/160] PCI/AER: Add PoisonTLPBlocked to Uncorrectable error counters The elements in the aer_uncorrectable_error_string[] refer to the bit names in Uncorrectable Error Status Register. Add PoisonTLPBlocked, which was added in PCIe r3.1, sec 7.10.2. Link: https://lore.kernel.org/r/20190827222145.32642-1-rajatja@google.com Signed-off-by: Rajat Jain Signed-off-by: Bjorn Helgaas --- drivers/pci/pcie/aer.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c index 443512882be4..0742e3a10ed8 100644 --- a/drivers/pci/pcie/aer.c +++ b/drivers/pci/pcie/aer.c @@ -36,7 +36,7 @@ #define AER_ERROR_SOURCES_MAX 128 #define AER_MAX_TYPEOF_COR_ERRS 16 /* as per PCI_ERR_COR_STATUS */ -#define AER_MAX_TYPEOF_UNCOR_ERRS 26 /* as per PCI_ERR_UNCOR_STATUS*/ +#define AER_MAX_TYPEOF_UNCOR_ERRS 27 /* as per PCI_ERR_UNCOR_STATUS*/ struct aer_err_source { unsigned int status; @@ -618,6 +618,7 @@ static const char *aer_uncorrectable_error_string[AER_MAX_TYPEOF_UNCOR_ERRS] = { "BlockedTLP", /* Bit Position 23 */ "AtomicOpBlocked", /* Bit Position 24 */ "TLPBlockedErr", /* Bit Position 25 */ + "PoisonTLPBlocked", /* Bit Position 26 */ }; static const char *aer_agent_string[] = { From 6a8c97345a15f9c60ff6c6ac1629a3e9ec140320 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 27 Aug 2019 18:18:22 +0300 Subject: [PATCH 050/160] PCI/AER: Use for_each_set_bit() to simplify code Simplify error counting code by using for_each_set_bit() library function. Link: https://lore.kernel.org/r/20190827151823.75312-1-andriy.shevchenko@linux.intel.com Signed-off-by: Andy Shevchenko Signed-off-by: Bjorn Helgaas Reviewed-by: Kuppuswamy Sathyanarayanan --- drivers/pci/pcie/aer.c | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c index 0742e3a10ed8..bf3c77957a87 100644 --- a/drivers/pci/pcie/aer.c +++ b/drivers/pci/pcie/aer.c @@ -15,6 +15,7 @@ #define pr_fmt(fmt) "AER: " fmt #define dev_fmt pr_fmt +#include #include #include #include @@ -716,7 +717,8 @@ const struct attribute_group aer_stats_attr_group = { static void pci_dev_aer_stats_incr(struct pci_dev *pdev, struct aer_err_info *info) { - int status, i, max = -1; + unsigned long status = info->status & ~info->mask; + int i, max = -1; u64 *counter = NULL; struct aer_stats *aer_stats = pdev->aer_stats; @@ -741,10 +743,8 @@ static void pci_dev_aer_stats_incr(struct pci_dev *pdev, break; } - status = (info->status & ~info->mask); - for (i = 0; i < max; i++) - if (status & (1 << i)) - counter[i]++; + for_each_set_bit(i, &status, max) + counter[i]++; } static void pci_rootport_aer_stats_incr(struct pci_dev *pdev, @@ -776,14 +776,11 @@ static void __print_tlp_header(struct pci_dev *dev, static void __aer_print_error(struct pci_dev *dev, struct aer_err_info *info) { - int i, status; + unsigned long status = info->status & ~info->mask; const char *errmsg = NULL; - status = (info->status & ~info->mask); - - for (i = 0; i < 32; i++) { - if (!(status & (1 << i))) - continue; + int i; + for_each_set_bit(i, &status, 32) { if (info->severity == AER_CORRECTABLE) errmsg = i < ARRAY_SIZE(aer_correctable_error_string) ? aer_correctable_error_string[i] : NULL; From 161eea1b25268a1f7fa8d220a3f0c350b4cc491c Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 27 Aug 2019 18:18:23 +0300 Subject: [PATCH 051/160] PCI/AER: Fix kernel-doc warnings Kernel-doc validator complains: aer.c:207: warning: Function parameter or member 'str' not described in 'pcie_ecrc_get_policy' aer.c:1209: warning: Function parameter or member 'irq' not described in 'aer_isr' aer.c:1209: warning: Function parameter or member 'context' not described in 'aer_isr' aer.c:1209: warning: Excess function parameter 'work' description in 'aer_isr' Fix the above accordingly. Link: https://lore.kernel.org/r/20190827151823.75312-2-andriy.shevchenko@linux.intel.com Signed-off-by: Andy Shevchenko Signed-off-by: Bjorn Helgaas Reviewed-by: Kuppuswamy Sathyanarayanan --- drivers/pci/pcie/aer.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/pci/pcie/aer.c b/drivers/pci/pcie/aer.c index bf3c77957a87..1ca86f2e0166 100644 --- a/drivers/pci/pcie/aer.c +++ b/drivers/pci/pcie/aer.c @@ -202,6 +202,7 @@ void pcie_set_ecrc_checking(struct pci_dev *dev) /** * pcie_ecrc_get_policy - parse kernel command-line ecrc option + * @str: ECRC policy from kernel command line to use */ void pcie_ecrc_get_policy(char *str) { @@ -1260,7 +1261,8 @@ static void aer_isr_one_error(struct aer_rpc *rpc, /** * aer_isr - consume errors detected by root port - * @work: definition of this work item + * @irq: IRQ assigned to Root Port + * @context: pointer to Root Port data structure * * Invoked, as DPC, when root port records new detected error */ From d7b8a217521ca21e2c6391da88d4928c6ce1f539 Mon Sep 17 00:00:00 2001 From: Nicholas Johnson Date: Wed, 23 Oct 2019 12:12:29 +0000 Subject: [PATCH 052/160] PCI: Add "pci=hpmmiosize" and "pci=hpmmioprefsize" parameters The existing "pci=hpmemsize=nn[KMG]" kernel parameter overrides the default size of both the non-prefetchable and the prefetchable MMIO windows for hotplug bridges. Add "pci=hpmmiosize=nn[KMG]" to override the default size of only the non-prefetchable MMIO window. Add "pci=hpmmioprefsize=nn[KMG]" to override the default size of only the prefetchable MMIO window. Link: https://lore.kernel.org/r/SL2P216MB0187E4D0055791957B7E2660806B0@SL2P216MB0187.KORP216.PROD.OUTLOOK.COM Signed-off-by: Nicholas Johnson Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg --- .../admin-guide/kernel-parameters.txt | 9 ++++++- drivers/pci/pci.c | 20 ++++++++++++---- drivers/pci/pci.h | 3 ++- drivers/pci/setup-bus.c | 24 ++++++++++--------- 4 files changed, 39 insertions(+), 17 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index c7ac2f3ac99f..62fbf7c0ff1b 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -3492,8 +3492,15 @@ hpiosize=nn[KMG] The fixed amount of bus space which is reserved for hotplug bridge's IO window. Default size is 256 bytes. + hpmmiosize=nn[KMG] The fixed amount of bus space which is + reserved for hotplug bridge's MMIO window. + Default size is 2 megabytes. + hpmmioprefsize=nn[KMG] The fixed amount of bus space which is + reserved for hotplug bridge's MMIO_PREF window. + Default size is 2 megabytes. hpmemsize=nn[KMG] The fixed amount of bus space which is - reserved for hotplug bridge's memory window. + reserved for hotplug bridge's MMIO and + MMIO_PREF window. Default size is 2 megabytes. hpbussize=nn The minimum amount of additional bus numbers reserved for buses below a hotplug bridge. diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 184765f12848..66ff1ca5b688 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -85,10 +85,17 @@ unsigned long pci_cardbus_io_size = DEFAULT_CARDBUS_IO_SIZE; unsigned long pci_cardbus_mem_size = DEFAULT_CARDBUS_MEM_SIZE; #define DEFAULT_HOTPLUG_IO_SIZE (256) -#define DEFAULT_HOTPLUG_MEM_SIZE (2*1024*1024) -/* pci=hpmemsize=nnM,hpiosize=nn can override this */ +#define DEFAULT_HOTPLUG_MMIO_SIZE (2*1024*1024) +#define DEFAULT_HOTPLUG_MMIO_PREF_SIZE (2*1024*1024) +/* hpiosize=nn can override this */ unsigned long pci_hotplug_io_size = DEFAULT_HOTPLUG_IO_SIZE; -unsigned long pci_hotplug_mem_size = DEFAULT_HOTPLUG_MEM_SIZE; +/* + * pci=hpmmiosize=nnM overrides non-prefetchable MMIO size, + * pci=hpmmioprefsize=nnM overrides prefetchable MMIO size; + * pci=hpmemsize=nnM overrides both + */ +unsigned long pci_hotplug_mmio_size = DEFAULT_HOTPLUG_MMIO_SIZE; +unsigned long pci_hotplug_mmio_pref_size = DEFAULT_HOTPLUG_MMIO_PREF_SIZE; #define DEFAULT_HOTPLUG_BUS_SIZE 1 unsigned long pci_hotplug_bus_size = DEFAULT_HOTPLUG_BUS_SIZE; @@ -6289,8 +6296,13 @@ static int __init pci_setup(char *str) pcie_ecrc_get_policy(str + 5); } else if (!strncmp(str, "hpiosize=", 9)) { pci_hotplug_io_size = memparse(str + 9, &str); + } else if (!strncmp(str, "hpmmiosize=", 11)) { + pci_hotplug_mmio_size = memparse(str + 11, &str); + } else if (!strncmp(str, "hpmmioprefsize=", 15)) { + pci_hotplug_mmio_pref_size = memparse(str + 15, &str); } else if (!strncmp(str, "hpmemsize=", 10)) { - pci_hotplug_mem_size = memparse(str + 10, &str); + pci_hotplug_mmio_size = memparse(str + 10, &str); + pci_hotplug_mmio_pref_size = pci_hotplug_mmio_size; } else if (!strncmp(str, "hpbussize=", 10)) { pci_hotplug_bus_size = simple_strtoul(str + 10, &str, 0); diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 3f6947ee3324..9faa55a151cb 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -218,7 +218,8 @@ extern const struct device_type pci_dev_type; extern const struct attribute_group *pci_bus_groups[]; extern unsigned long pci_hotplug_io_size; -extern unsigned long pci_hotplug_mem_size; +extern unsigned long pci_hotplug_mmio_size; +extern unsigned long pci_hotplug_mmio_pref_size; extern unsigned long pci_hotplug_bus_size; /** diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c index f1c51943bdfc..b5f9f57f436a 100644 --- a/drivers/pci/setup-bus.c +++ b/drivers/pci/setup-bus.c @@ -1178,7 +1178,8 @@ void __pci_bus_size_bridges(struct pci_bus *bus, struct list_head *realloc_head) { struct pci_dev *dev; unsigned long mask, prefmask, type2 = 0, type3 = 0; - resource_size_t additional_mem_size = 0, additional_io_size = 0; + resource_size_t additional_io_size = 0, additional_mmio_size = 0, + additional_mmio_pref_size = 0; struct resource *b_res; int ret; @@ -1212,7 +1213,8 @@ void __pci_bus_size_bridges(struct pci_bus *bus, struct list_head *realloc_head) pci_bridge_check_ranges(bus); if (bus->self->is_hotplug_bridge) { additional_io_size = pci_hotplug_io_size; - additional_mem_size = pci_hotplug_mem_size; + additional_mmio_size = pci_hotplug_mmio_size; + additional_mmio_pref_size = pci_hotplug_mmio_pref_size; } /* Fall through */ default: @@ -1230,9 +1232,9 @@ void __pci_bus_size_bridges(struct pci_bus *bus, struct list_head *realloc_head) if (b_res[2].flags & IORESOURCE_MEM_64) { prefmask |= IORESOURCE_MEM_64; ret = pbus_size_mem(bus, prefmask, prefmask, - prefmask, prefmask, - realloc_head ? 0 : additional_mem_size, - additional_mem_size, realloc_head); + prefmask, prefmask, + realloc_head ? 0 : additional_mmio_pref_size, + additional_mmio_pref_size, realloc_head); /* * If successful, all non-prefetchable resources @@ -1254,9 +1256,9 @@ void __pci_bus_size_bridges(struct pci_bus *bus, struct list_head *realloc_head) if (!type2) { prefmask &= ~IORESOURCE_MEM_64; ret = pbus_size_mem(bus, prefmask, prefmask, - prefmask, prefmask, - realloc_head ? 0 : additional_mem_size, - additional_mem_size, realloc_head); + prefmask, prefmask, + realloc_head ? 0 : additional_mmio_pref_size, + additional_mmio_pref_size, realloc_head); /* * If successful, only non-prefetchable resources @@ -1265,7 +1267,7 @@ void __pci_bus_size_bridges(struct pci_bus *bus, struct list_head *realloc_head) if (ret == 0) mask = prefmask; else - additional_mem_size += additional_mem_size; + additional_mmio_size += additional_mmio_pref_size; type2 = type3 = IORESOURCE_MEM; } @@ -1285,8 +1287,8 @@ void __pci_bus_size_bridges(struct pci_bus *bus, struct list_head *realloc_head) * prefetchable resource in a 64-bit prefetchable window. */ pbus_size_mem(bus, mask, IORESOURCE_MEM, type2, type3, - realloc_head ? 0 : additional_mem_size, - additional_mem_size, realloc_head); + realloc_head ? 0 : additional_mmio_size, + additional_mmio_size, realloc_head); break; } } From ad5086108b9f0361929aa9a79cf959ab5681d249 Mon Sep 17 00:00:00 2001 From: Yunsheng Lin Date: Sat, 19 Oct 2019 14:45:43 +0800 Subject: [PATCH 053/160] PCI: Warn if no host bridge NUMA node info In pci_call_probe(), we try to run driver probe functions on the node where the device is attached. If we don't know which node the device is attached to, the driver will likely run on the wrong node. This will still work, but performance will not be as good as it could be. On NUMA systems, warn if we don't know which node a PCI host bridge is attached to. This is likely an indication that ACPI didn't supply a _PXM method or the DT didn't supply a "numa-node-id" property. [bhelgaas: commit log, check bus node] Link: https://lore.kernel.org/r/1571467543-26125-1-git-send-email-linyunsheng@huawei.com Signed-off-by: Yunsheng Lin Signed-off-by: Bjorn Helgaas --- drivers/pci/probe.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 3d5271a7a849..40259c38d66a 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -897,6 +897,9 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge) else pr_info("PCI host bridge to bus %s\n", name); + if (nr_node_ids > 1 && pcibus_to_node(bus) == NUMA_NO_NODE) + dev_warn(&bus->dev, "Unknown NUMA node; performance will be reduced\n"); + /* Add initial resources to the bus */ resource_list_for_each_entry_safe(window, n, &resources) { list_move_tail(&window->node, &bridge->windows); From 9415743e4c8a029bfa46654e3a0a681a8db473f8 Mon Sep 17 00:00:00 2001 From: Abhishek Shah Date: Fri, 6 Sep 2019 09:28:13 +0530 Subject: [PATCH 054/160] PCI: iproc: Invalidate PAXB address mapping before programming it Invalidate PAXB inbound/outbound address mapping on probe before programming it. Kernel relies on outbound/inbound windows VALID bit in OARR registers to detect if a window was programmed and if it is set it does not overwrite it. This causes issues on soft reboot (eg kexec) since the host controller does not go through a HW reset on softboot so the kernel detects valid outbound/inbound windows configuration and is not able to reprogramme it as expected. Therefore, in order to make sure outbound/inbound windows are reprogrammed on soft reboot (eg kexec), invalidate memory windows on each probe to fix the issue. Signed-off-by: Abhishek Shah Signed-off-by: Lorenzo Pieralisi Reviewed-by: Ray Jui Reviewed-by: Andrew Murray --- drivers/pci/controller/pcie-iproc.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/drivers/pci/controller/pcie-iproc.c b/drivers/pci/controller/pcie-iproc.c index 2d457bfdaf66..9a5c35225944 100644 --- a/drivers/pci/controller/pcie-iproc.c +++ b/drivers/pci/controller/pcie-iproc.c @@ -1245,6 +1245,32 @@ out: return ret; } +static void iproc_pcie_invalidate_mapping(struct iproc_pcie *pcie) +{ + struct iproc_pcie_ib *ib = &pcie->ib; + struct iproc_pcie_ob *ob = &pcie->ob; + int idx; + + if (pcie->ep_is_internal) + return; + + if (pcie->need_ob_cfg) { + /* iterate through all OARR mapping regions */ + for (idx = ob->nr_windows - 1; idx >= 0; idx--) { + iproc_pcie_write_reg(pcie, + MAP_REG(IPROC_PCIE_OARR0, idx), 0); + } + } + + if (pcie->need_ib_cfg) { + /* iterate through all IARR mapping regions */ + for (idx = 0; idx < ib->nr_regions; idx++) { + iproc_pcie_write_reg(pcie, + MAP_REG(IPROC_PCIE_IARR0, idx), 0); + } + } +} + static int iproce_pcie_get_msi(struct iproc_pcie *pcie, struct device_node *msi_node, u64 *msi_addr) @@ -1517,6 +1543,8 @@ int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res) iproc_pcie_perst_ctrl(pcie, true); iproc_pcie_perst_ctrl(pcie, false); + iproc_pcie_invalidate_mapping(pcie); + if (pcie->need_ob_cfg) { ret = iproc_pcie_map_ranges(pcie, res); if (ret) { From 35a0b2378c199d4f26e458b2ca38ea56aaf2d9b8 Mon Sep 17 00:00:00 2001 From: Olof Johansson Date: Wed, 23 Oct 2019 12:22:05 -0700 Subject: [PATCH 055/160] PCI/DPC: Add "pcie_ports=dpc-native" to allow DPC without AER control Prior to eed85ff4c0da7 ("PCI/DPC: Enable DPC only if AER is available"), Linux handled DPC events regardless of whether firmware had granted it ownership of AER or DPC, e.g., via _OSC. PCIe r5.0, sec 6.2.10, recommends that the OS link control of DPC to control of AER, so after eed85ff4c0da7, Linux handles DPC events only if it has control of AER. On platforms that do not grant OS control of AER via _OSC, Linux DPC handling worked before eed85ff4c0da7 but not after. To make Linux DPC handling work on those platforms the same way they did before, add a "pcie_ports=dpc-native" kernel parameter that makes Linux handle DPC events regardless of whether it has control of AER. [bhelgaas: commit log, move pcie_ports_dpc_native to drivers/pci/] Link: https://lore.kernel.org/r/20191023192205.97024-1-olof@lixom.net Signed-off-by: Olof Johansson Signed-off-by: Bjorn Helgaas --- Documentation/admin-guide/kernel-parameters.txt | 2 ++ drivers/pci/pcie/dpc.c | 2 +- drivers/pci/pcie/portdrv.h | 2 ++ drivers/pci/pcie/portdrv_core.c | 7 ++++++- drivers/pci/pcie/portdrv_pci.c | 8 ++++++++ 5 files changed, 19 insertions(+), 2 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index c7ac2f3ac99f..806c89f79be8 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -3540,6 +3540,8 @@ even if the platform doesn't give the OS permission to use them. This may cause conflicts if the platform also tries to use these services. + dpc-native Use native PCIe service for DPC only. May + cause conflicts if firmware uses AER or DPC. compat Disable native PCIe services (PME, AER, DPC, PCIe hotplug). diff --git a/drivers/pci/pcie/dpc.c b/drivers/pci/pcie/dpc.c index a32ec3487a8d..e06f42f58d3d 100644 --- a/drivers/pci/pcie/dpc.c +++ b/drivers/pci/pcie/dpc.c @@ -291,7 +291,7 @@ static int dpc_probe(struct pcie_device *dev) int status; u16 ctl, cap; - if (pcie_aer_get_firmware_first(pdev)) + if (pcie_aer_get_firmware_first(pdev) && !pcie_ports_dpc_native) return -ENOTSUPP; dpc = devm_kzalloc(device, sizeof(*dpc), GFP_KERNEL); diff --git a/drivers/pci/pcie/portdrv.h b/drivers/pci/pcie/portdrv.h index 944827a8c7d3..1e673619b101 100644 --- a/drivers/pci/pcie/portdrv.h +++ b/drivers/pci/pcie/portdrv.h @@ -25,6 +25,8 @@ #define PCIE_PORT_DEVICE_MAXSERVICES 5 +extern bool pcie_ports_dpc_native; + #ifdef CONFIG_PCIEAER int pcie_aer_init(void); #else diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c index 1b330129089f..5075cb9e850c 100644 --- a/drivers/pci/pcie/portdrv_core.c +++ b/drivers/pci/pcie/portdrv_core.c @@ -250,8 +250,13 @@ static int get_port_device_capability(struct pci_dev *dev) pcie_pme_interrupt_enable(dev, false); } + /* + * With dpc-native, allow Linux to use DPC even if it doesn't have + * permission to use AER. + */ if (pci_find_ext_capability(dev, PCI_EXT_CAP_ID_DPC) && - pci_aer_available() && services & PCIE_PORT_SERVICE_AER) + pci_aer_available() && + (pcie_ports_dpc_native || (services & PCIE_PORT_SERVICE_AER))) services |= PCIE_PORT_SERVICE_DPC; if (pci_pcie_type(dev) == PCI_EXP_TYPE_DOWNSTREAM || diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c index 0a87091a0800..160d67c59310 100644 --- a/drivers/pci/pcie/portdrv_pci.c +++ b/drivers/pci/pcie/portdrv_pci.c @@ -29,12 +29,20 @@ bool pcie_ports_disabled; */ bool pcie_ports_native; +/* + * If the user specified "pcie_ports=dpc-native", use the Linux DPC PCIe + * service even if the platform hasn't given us permission. + */ +bool pcie_ports_dpc_native; + static int __init pcie_port_setup(char *str) { if (!strncmp(str, "compat", 6)) pcie_ports_disabled = true; else if (!strncmp(str, "native", 6)) pcie_ports_native = true; + else if (!strncmp(str, "dpc-native", 10)) + pcie_ports_dpc_native = true; return 1; } From 494f8b10d832456a96be4ee7317425f6936cabc8 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:32 -0500 Subject: [PATCH 056/160] resource: Add a resource_list_first_type helper A common pattern is looping over a resource_list just to get a matching entry with a specific type. Add resource_list_first_type() helper which implements this. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi --- include/linux/resource_ext.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/include/linux/resource_ext.h b/include/linux/resource_ext.h index 06da59b23b79..ff0339df56af 100644 --- a/include/linux/resource_ext.h +++ b/include/linux/resource_ext.h @@ -66,4 +66,16 @@ resource_list_destroy_entry(struct resource_entry *entry) #define resource_list_for_each_entry_safe(entry, tmp, list) \ list_for_each_entry_safe((entry), (tmp), (list), node) +static inline struct resource_entry * +resource_list_first_type(struct list_head *list, unsigned long type) +{ + struct resource_entry *entry; + + resource_list_for_each_entry(entry, list) { + if (resource_type(entry->res) == type) + return entry; + } + return NULL; +} + #endif /* _LINUX_RESOURCE_EXT_H */ From 65991f43769941bea14158e0e731f9362051b618 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:33 -0500 Subject: [PATCH 057/160] PCI: Export pci_parse_request_of_pci_ranges() pci_parse_request_of_pci_ranges() is missing a module export, so add it. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Cc: Bjorn Helgaas --- drivers/pci/of.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/pci/of.c b/drivers/pci/of.c index 36891e7deee3..f3da49a31db4 100644 --- a/drivers/pci/of.c +++ b/drivers/pci/of.c @@ -530,6 +530,7 @@ int pci_parse_request_of_pci_ranges(struct device *dev, pci_free_resource_list(resources); return err; } +EXPORT_SYMBOL_GPL(pci_parse_request_of_pci_ranges); #endif /* CONFIG_PCI */ From 4e5be6f81be72c980580676e31c2b542c0267757 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:34 -0500 Subject: [PATCH 058/160] PCI: aardvark: Use pci_parse_request_of_pci_ranges() Convert aardvark to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list first. Just use bridge->windows directly and remove all the temporary list handling. Tested-by: Thomas Petazzoni Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/controller/pci-aardvark.c | 60 ++------------------------- 1 file changed, 4 insertions(+), 56 deletions(-) diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c index fc0fe4d4de49..9cbeba507f0c 100644 --- a/drivers/pci/controller/pci-aardvark.c +++ b/drivers/pci/controller/pci-aardvark.c @@ -186,7 +186,6 @@ struct advk_pcie { struct platform_device *pdev; void __iomem *base; - struct list_head resources; struct irq_domain *irq_domain; struct irq_chip irq_chip; struct irq_domain *msi_domain; @@ -910,63 +909,11 @@ static irqreturn_t advk_pcie_irq_handler(int irq, void *arg) return IRQ_HANDLED; } -static int advk_pcie_parse_request_of_pci_ranges(struct advk_pcie *pcie) -{ - int err, res_valid = 0; - struct device *dev = &pcie->pdev->dev; - struct resource_entry *win, *tmp; - resource_size_t iobase; - - INIT_LIST_HEAD(&pcie->resources); - - err = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, - &pcie->resources, &iobase); - if (err) - return err; - - err = devm_request_pci_bus_resources(dev, &pcie->resources); - if (err) - goto out_release_res; - - resource_list_for_each_entry_safe(win, tmp, &pcie->resources) { - struct resource *res = win->res; - - switch (resource_type(res)) { - case IORESOURCE_IO: - err = devm_pci_remap_iospace(dev, res, iobase); - if (err) { - dev_warn(dev, "error %d: failed to map resource %pR\n", - err, res); - resource_list_destroy_entry(win); - } - break; - case IORESOURCE_MEM: - res_valid |= !(res->flags & IORESOURCE_PREFETCH); - break; - case IORESOURCE_BUS: - pcie->root_bus_nr = res->start; - break; - } - } - - if (!res_valid) { - dev_err(dev, "non-prefetchable memory resource required\n"); - err = -EINVAL; - goto out_release_res; - } - - return 0; - -out_release_res: - pci_free_resource_list(&pcie->resources); - return err; -} - static int advk_pcie_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct advk_pcie *pcie; - struct resource *res; + struct resource *res, *bus; struct pci_host_bridge *bridge; int ret, irq; @@ -991,11 +938,13 @@ static int advk_pcie_probe(struct platform_device *pdev) return ret; } - ret = advk_pcie_parse_request_of_pci_ranges(pcie); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + &bus); if (ret) { dev_err(dev, "Failed to parse resources\n"); return ret; } + pcie->root_bus_nr = bus->start; advk_pcie_setup_hw(pcie); @@ -1014,7 +963,6 @@ static int advk_pcie_probe(struct platform_device *pdev) return ret; } - list_splice_init(&pcie->resources, &bridge->windows); bridge->dev.parent = dev; bridge->sysdata = pcie; bridge->busnr = 0; From e634e3e0b790789ee16f4df503ccab13f1169134 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:35 -0500 Subject: [PATCH 059/160] PCI: altera: Use pci_parse_request_of_pci_ranges() Convert altera host bridge to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list first. Just use bridge->windows directly and remove all the temporary list handling. If an I/O range is present, then it will now be mapped. It's expected that h/w which doesn't support I/O range will not define one. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Cc: Ley Foon Tan Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas Cc: rfi@lists.rocketboards.org --- drivers/pci/controller/pcie-altera.c | 41 ++-------------------------- 1 file changed, 2 insertions(+), 39 deletions(-) diff --git a/drivers/pci/controller/pcie-altera.c b/drivers/pci/controller/pcie-altera.c index d2497ca43828..ba025efeae28 100644 --- a/drivers/pci/controller/pcie-altera.c +++ b/drivers/pci/controller/pcie-altera.c @@ -92,7 +92,6 @@ struct altera_pcie { u8 root_bus_nr; struct irq_domain *irq_domain; struct resource bus_range; - struct list_head resources; const struct altera_pcie_data *pcie_data; }; @@ -670,39 +669,6 @@ static void altera_pcie_isr(struct irq_desc *desc) chained_irq_exit(chip, desc); } -static int altera_pcie_parse_request_of_pci_ranges(struct altera_pcie *pcie) -{ - int err, res_valid = 0; - struct device *dev = &pcie->pdev->dev; - struct resource_entry *win; - - err = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, - &pcie->resources, NULL); - if (err) - return err; - - err = devm_request_pci_bus_resources(dev, &pcie->resources); - if (err) - goto out_release_res; - - resource_list_for_each_entry(win, &pcie->resources) { - struct resource *res = win->res; - - if (resource_type(res) == IORESOURCE_MEM) - res_valid |= !(res->flags & IORESOURCE_PREFETCH); - } - - if (res_valid) - return 0; - - dev_err(dev, "non-prefetchable memory resource required\n"); - err = -EINVAL; - -out_release_res: - pci_free_resource_list(&pcie->resources); - return err; -} - static int altera_pcie_init_irq_domain(struct altera_pcie *pcie) { struct device *dev = &pcie->pdev->dev; @@ -833,9 +799,8 @@ static int altera_pcie_probe(struct platform_device *pdev) return ret; } - INIT_LIST_HEAD(&pcie->resources); - - ret = altera_pcie_parse_request_of_pci_ranges(pcie); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + NULL); if (ret) { dev_err(dev, "Failed add resources\n"); return ret; @@ -853,7 +818,6 @@ static int altera_pcie_probe(struct platform_device *pdev) cra_writel(pcie, P2A_INT_ENA_ALL, P2A_INT_ENABLE); altera_pcie_host_init(pcie); - list_splice_init(&pcie->resources, &bridge->windows); bridge->dev.parent = dev; bridge->sysdata = pcie; bridge->busnr = pcie->root_bus_nr; @@ -884,7 +848,6 @@ static int altera_pcie_remove(struct platform_device *pdev) pci_stop_root_bus(bridge->bus); pci_remove_root_bus(bridge->bus); - pci_free_resource_list(&pcie->resources); altera_pcie_irq_teardown(pcie); return 0; From 7fe71aa84b438d7830f812692f2e0fa55ecdf413 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:36 -0500 Subject: [PATCH 060/160] PCI: dwc: Use pci_parse_request_of_pci_ranges() Convert the Designware host bridge to use the common pci_parse_request_of_pci_ranges(). Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Cc: Jingoo Han Cc: Gustavo Pimentel Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas --- .../pci/controller/dwc/pcie-designware-host.c | 28 ++++++------------- 1 file changed, 8 insertions(+), 20 deletions(-) diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c index 0f36a926059a..aeec8b65eb97 100644 --- a/drivers/pci/controller/dwc/pcie-designware-host.c +++ b/drivers/pci/controller/dwc/pcie-designware-host.c @@ -319,7 +319,7 @@ int dw_pcie_host_init(struct pcie_port *pp) struct device *dev = pci->dev; struct device_node *np = dev->of_node; struct platform_device *pdev = to_platform_device(dev); - struct resource_entry *win, *tmp; + struct resource_entry *win; struct pci_bus *child; struct pci_host_bridge *bridge; struct resource *cfg_res; @@ -342,31 +342,19 @@ int dw_pcie_host_init(struct pcie_port *pp) if (!bridge) return -ENOMEM; - ret = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, - &bridge->windows, &pp->io_base); - if (ret) - return ret; - - ret = devm_request_pci_bus_resources(dev, &bridge->windows); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); if (ret) return ret; /* Get the I/O and memory ranges from DT */ - resource_list_for_each_entry_safe(win, tmp, &bridge->windows) { + resource_list_for_each_entry(win, &bridge->windows) { switch (resource_type(win->res)) { case IORESOURCE_IO: - ret = devm_pci_remap_iospace(dev, win->res, - pp->io_base); - if (ret) { - dev_warn(dev, "Error %d: failed to map resource %pR\n", - ret, win->res); - resource_list_destroy_entry(win); - } else { - pp->io = win->res; - pp->io->name = "I/O"; - pp->io_size = resource_size(pp->io); - pp->io_bus_addr = pp->io->start - win->offset; - } + pp->io = win->res; + pp->io->name = "I/O"; + pp->io_size = resource_size(pp->io); + pp->io_bus_addr = pp->io->start - win->offset; + pp->io_base = pci_pio_to_address(pp->io->start); break; case IORESOURCE_MEM: pp->mem = win->res; From 783a862563f71e5a3253efa0653eb0ebf9188cf8 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:37 -0500 Subject: [PATCH 061/160] PCI: faraday: Use pci_parse_request_of_pci_ranges() Convert the Faraday host bridge to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list first. Just use bridge->windows directly and remove all the temporary list handling. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/controller/pci-ftpci100.c | 51 ++++++--------------------- 1 file changed, 11 insertions(+), 40 deletions(-) diff --git a/drivers/pci/controller/pci-ftpci100.c b/drivers/pci/controller/pci-ftpci100.c index bf5ece5d9291..75603348b88a 100644 --- a/drivers/pci/controller/pci-ftpci100.c +++ b/drivers/pci/controller/pci-ftpci100.c @@ -430,10 +430,8 @@ static int faraday_pci_probe(struct platform_device *pdev) const struct faraday_pci_variant *variant = of_device_get_match_data(dev); struct resource *regs; - resource_size_t io_base; struct resource_entry *win; struct faraday_pci *p; - struct resource *mem; struct resource *io; struct pci_host_bridge *host; struct clk *clk; @@ -441,7 +439,6 @@ static int faraday_pci_probe(struct platform_device *pdev) unsigned char cur_bus_speed = PCI_SPEED_33MHz; int ret; u32 val; - LIST_HEAD(res); host = devm_pci_alloc_host_bridge(dev, sizeof(*p)); if (!host) @@ -480,44 +477,20 @@ static int faraday_pci_probe(struct platform_device *pdev) if (IS_ERR(p->base)) return PTR_ERR(p->base); - ret = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, - &res, &io_base); + ret = pci_parse_request_of_pci_ranges(dev, &host->windows, NULL); if (ret) return ret; - ret = devm_request_pci_bus_resources(dev, &res); - if (ret) - return ret; - - /* Get the I/O and memory ranges from DT */ - resource_list_for_each_entry(win, &res) { - switch (resource_type(win->res)) { - case IORESOURCE_IO: - io = win->res; - io->name = "Gemini PCI I/O"; - if (!faraday_res_to_memcfg(io->start - win->offset, - resource_size(io), &val)) { - /* setup I/O space size */ - writel(val, p->base + PCI_IOSIZE); - } else { - dev_err(dev, "illegal IO mem size\n"); - return -EINVAL; - } - ret = devm_pci_remap_iospace(dev, io, io_base); - if (ret) { - dev_warn(dev, "error %d: failed to map resource %pR\n", - ret, io); - continue; - } - break; - case IORESOURCE_MEM: - mem = win->res; - mem->name = "Gemini PCI MEM"; - break; - case IORESOURCE_BUS: - break; - default: - break; + win = resource_list_first_type(&host->windows, IORESOURCE_IO); + if (win) { + io = win->res; + if (!faraday_res_to_memcfg(io->start - win->offset, + resource_size(io), &val)) { + /* setup I/O space size */ + writel(val, p->base + PCI_IOSIZE); + } else { + dev_err(dev, "illegal IO mem size\n"); + return -EINVAL; } } @@ -569,7 +542,6 @@ static int faraday_pci_probe(struct platform_device *pdev) if (ret) return ret; - list_splice_init(&res, &host->windows); ret = pci_scan_root_bus_bridge(host); if (ret) { dev_err(dev, "failed to scan host: %d\n", ret); @@ -581,7 +553,6 @@ static int faraday_pci_probe(struct platform_device *pdev) pci_bus_assign_resources(p->bus); pci_bus_add_devices(p->bus); - pci_free_resource_list(&res); return 0; } From 7ef1c871da16125ae58db13716fe82795dcbe3e8 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:38 -0500 Subject: [PATCH 062/160] PCI: iproc: Use pci_parse_request_of_pci_ranges() Convert the iProc host bridge to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list, so just use bridge->windows directly. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas Cc: Ray Jui Cc: Scott Branden Cc: bcm-kernel-feedback-list@broadcom.com --- drivers/pci/controller/pcie-iproc-platform.c | 8 ++------ drivers/pci/controller/pcie-iproc.c | 5 ----- 2 files changed, 2 insertions(+), 11 deletions(-) diff --git a/drivers/pci/controller/pcie-iproc-platform.c b/drivers/pci/controller/pcie-iproc-platform.c index 9ee6200a66f4..375d815f7301 100644 --- a/drivers/pci/controller/pcie-iproc-platform.c +++ b/drivers/pci/controller/pcie-iproc-platform.c @@ -43,8 +43,6 @@ static int iproc_pcie_pltfm_probe(struct platform_device *pdev) struct iproc_pcie *pcie; struct device_node *np = dev->of_node; struct resource reg; - resource_size_t iobase = 0; - LIST_HEAD(resources); struct pci_host_bridge *bridge; int ret; @@ -97,8 +95,7 @@ static int iproc_pcie_pltfm_probe(struct platform_device *pdev) if (IS_ERR(pcie->phy)) return PTR_ERR(pcie->phy); - ret = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, &resources, - &iobase); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); if (ret) { dev_err(dev, "unable to get PCI host bridge resources\n"); return ret; @@ -113,10 +110,9 @@ static int iproc_pcie_pltfm_probe(struct platform_device *pdev) pcie->map_irq = of_irq_parse_and_map_pci; } - ret = iproc_pcie_setup(pcie, &resources); + ret = iproc_pcie_setup(pcie, &bridge->windows); if (ret) { dev_err(dev, "PCIe controller setup failed\n"); - pci_free_resource_list(&resources); return ret; } diff --git a/drivers/pci/controller/pcie-iproc.c b/drivers/pci/controller/pcie-iproc.c index 2d457bfdaf66..223335ee791a 100644 --- a/drivers/pci/controller/pcie-iproc.c +++ b/drivers/pci/controller/pcie-iproc.c @@ -1498,10 +1498,6 @@ int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res) return ret; } - ret = devm_request_pci_bus_resources(dev, res); - if (ret) - return ret; - ret = phy_init(pcie->phy); if (ret) { dev_err(dev, "unable to initialize PCIe PHY\n"); @@ -1543,7 +1539,6 @@ int iproc_pcie_setup(struct iproc_pcie *pcie, struct list_head *res) if (iproc_pcie_msi_enable(pcie)) dev_info(dev, "not using iProc MSI\n"); - list_splice_init(res, &host->windows); host->busnr = 0; host->dev.parent = dev; host->ops = &iproc_pcie_ops; From 8a26f861b815c997dd85cc2f6210020e7a700a62 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:39 -0500 Subject: [PATCH 063/160] PCI: mediatek: Use pci_parse_request_of_pci_ranges() Convert Mediatek host bridge to use the common pci_parse_request_of_pci_ranges(). Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Cc: Ryder Lee Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas Cc: Matthias Brugger Cc: linux-mediatek@lists.infradead.org --- drivers/pci/controller/pcie-mediatek.c | 43 ++++++++------------------ 1 file changed, 13 insertions(+), 30 deletions(-) diff --git a/drivers/pci/controller/pcie-mediatek.c b/drivers/pci/controller/pcie-mediatek.c index 626a7c352dfd..d9206a3cd56b 100644 --- a/drivers/pci/controller/pcie-mediatek.c +++ b/drivers/pci/controller/pcie-mediatek.c @@ -216,7 +216,6 @@ struct mtk_pcie { void __iomem *base; struct clk *free_ck; - struct resource mem; struct list_head ports; const struct mtk_pcie_soc *soc; unsigned int busnr; @@ -661,11 +660,19 @@ static int mtk_pcie_setup_irq(struct mtk_pcie_port *port, static int mtk_pcie_startup_port_v2(struct mtk_pcie_port *port) { struct mtk_pcie *pcie = port->pcie; - struct resource *mem = &pcie->mem; + struct pci_host_bridge *host = pci_host_bridge_from_priv(pcie); + struct resource *mem = NULL; + struct resource_entry *entry; const struct mtk_pcie_soc *soc = port->pcie->soc; u32 val; int err; + entry = resource_list_first_type(&host->windows, IORESOURCE_MEM); + if (entry) + mem = entry->res; + if (!mem) + return -EINVAL; + /* MT7622 platforms need to enable LTSSM and ASPM from PCIe subsys */ if (pcie->base) { val = readl(pcie->base + PCIE_SYS_CFG_V2); @@ -1023,39 +1030,15 @@ static int mtk_pcie_setup(struct mtk_pcie *pcie) struct mtk_pcie_port *port, *tmp; struct pci_host_bridge *host = pci_host_bridge_from_priv(pcie); struct list_head *windows = &host->windows; - struct resource_entry *win, *tmp_win; - resource_size_t io_base; + struct resource *bus; int err; - err = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, - windows, &io_base); + err = pci_parse_request_of_pci_ranges(dev, windows, + &bus); if (err) return err; - err = devm_request_pci_bus_resources(dev, windows); - if (err < 0) - return err; - - /* Get the I/O and memory ranges from DT */ - resource_list_for_each_entry_safe(win, tmp_win, windows) { - switch (resource_type(win->res)) { - case IORESOURCE_IO: - err = devm_pci_remap_iospace(dev, win->res, io_base); - if (err) { - dev_warn(dev, "error %d: failed to map resource %pR\n", - err, win->res); - resource_list_destroy_entry(win); - } - break; - case IORESOURCE_MEM: - memcpy(&pcie->mem, win->res, sizeof(*win->res)); - pcie->mem.name = "non-prefetchable"; - break; - case IORESOURCE_BUS: - pcie->busnr = win->res->start; - break; - } - } + pcie->busnr = bus->start; for_each_available_child_of_node(node, child) { int slot; From 6c6a0dff064176a5a0c6b1a9ee32f868ecd5e0f1 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:40 -0500 Subject: [PATCH 064/160] PCI: mobiveil: Use pci_parse_request_of_pci_ranges() Convert the Mobiveil host bridge to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list first. Just use bridge->windows directly and remove all the temporary list handling. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Reviewed-by: Hou Zhiqiang Cc: Karthikeyan Mitran Cc: Hou Zhiqiang Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/controller/pcie-mobiveil.c | 26 +++++++------------------- 1 file changed, 7 insertions(+), 19 deletions(-) diff --git a/drivers/pci/controller/pcie-mobiveil.c b/drivers/pci/controller/pcie-mobiveil.c index a45a6447b01d..4eab8624ce4d 100644 --- a/drivers/pci/controller/pcie-mobiveil.c +++ b/drivers/pci/controller/pcie-mobiveil.c @@ -140,7 +140,6 @@ struct mobiveil_msi { /* MSI information */ struct mobiveil_pcie { struct platform_device *pdev; - struct list_head resources; void __iomem *config_axi_slave_base; /* endpoint config base */ void __iomem *csr_axi_slave_base; /* root port config base */ void __iomem *apb_csr_base; /* MSI register base */ @@ -575,6 +574,7 @@ static void mobiveil_pcie_enable_msi(struct mobiveil_pcie *pcie) static int mobiveil_host_init(struct mobiveil_pcie *pcie) { + struct pci_host_bridge *bridge = pci_host_bridge_from_priv(pcie); u32 value, pab_ctrl, type; struct resource_entry *win; @@ -631,7 +631,7 @@ static int mobiveil_host_init(struct mobiveil_pcie *pcie) program_ib_windows(pcie, WIN_NUM_0, 0, 0, MEM_WINDOW_TYPE, IB_WIN_SIZE); /* Get the I/O and memory ranges from DT */ - resource_list_for_each_entry(win, &pcie->resources) { + resource_list_for_each_entry(win, &bridge->windows) { if (resource_type(win->res) == IORESOURCE_MEM) type = MEM_WINDOW_TYPE; else if (resource_type(win->res) == IORESOURCE_IO) @@ -857,7 +857,6 @@ static int mobiveil_pcie_probe(struct platform_device *pdev) struct pci_bus *child; struct pci_host_bridge *bridge; struct device *dev = &pdev->dev; - resource_size_t iobase; int ret; /* allocate the PCIe port */ @@ -875,11 +874,8 @@ static int mobiveil_pcie_probe(struct platform_device *pdev) return ret; } - INIT_LIST_HEAD(&pcie->resources); - /* parse the host bridge base addresses from the device tree file */ - ret = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, - &pcie->resources, &iobase); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); if (ret) { dev_err(dev, "Getting bridge resources failed\n"); return ret; @@ -892,24 +888,19 @@ static int mobiveil_pcie_probe(struct platform_device *pdev) ret = mobiveil_host_init(pcie); if (ret) { dev_err(dev, "Failed to initialize host\n"); - goto error; + return ret; } /* initialize the IRQ domains */ ret = mobiveil_pcie_init_irq_domain(pcie); if (ret) { dev_err(dev, "Failed creating IRQ Domain\n"); - goto error; + return ret; } irq_set_chained_handler_and_data(pcie->irq, mobiveil_pcie_isr, pcie); - ret = devm_request_pci_bus_resources(dev, &pcie->resources); - if (ret) - goto error; - /* Initialize bridge */ - list_splice_init(&pcie->resources, &bridge->windows); bridge->dev.parent = dev; bridge->sysdata = pcie; bridge->busnr = pcie->root_bus_nr; @@ -920,13 +911,13 @@ static int mobiveil_pcie_probe(struct platform_device *pdev) ret = mobiveil_bringup_link(pcie); if (ret) { dev_info(dev, "link bring-up failed\n"); - goto error; + return ret; } /* setup the kernel resources for the newly added PCIe root bus */ ret = pci_scan_root_bus_bridge(bridge); if (ret) - goto error; + return ret; bus = bridge->bus; @@ -936,9 +927,6 @@ static int mobiveil_pcie_probe(struct platform_device *pdev) pci_bus_add_devices(bus); return 0; -error: - pci_free_resource_list(&pcie->resources); - return ret; } static const struct of_device_id mobiveil_pcie_of_match[] = { From 5c1306a0fde67e5a39bef79932a0cb5cec5fd629 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:41 -0500 Subject: [PATCH 065/160] PCI: rockchip: Use pci_parse_request_of_pci_ranges() Convert the Rockchip host bridge to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list first. Just use bridge->windows directly and remove all the temporary list handling. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Cc: Shawn Lin Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas Cc: Heiko Stuebner Cc: linux-rockchip@lists.infradead.org --- drivers/pci/controller/pcie-rockchip-host.c | 36 ++++----------------- 1 file changed, 7 insertions(+), 29 deletions(-) diff --git a/drivers/pci/controller/pcie-rockchip-host.c b/drivers/pci/controller/pcie-rockchip-host.c index ef8e677ce9d1..8d2e6f2e141e 100644 --- a/drivers/pci/controller/pcie-rockchip-host.c +++ b/drivers/pci/controller/pcie-rockchip-host.c @@ -950,14 +950,10 @@ static int rockchip_pcie_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; struct pci_bus *bus, *child; struct pci_host_bridge *bridge; + struct resource *bus_res; struct resource_entry *win; - resource_size_t io_base; - struct resource *mem; - struct resource *io; int err; - LIST_HEAD(res); - if (!dev->of_node) return -ENODEV; @@ -995,29 +991,20 @@ static int rockchip_pcie_probe(struct platform_device *pdev) if (err < 0) goto err_deinit_port; - err = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, - &res, &io_base); + err = pci_parse_request_of_pci_ranges(dev, &bridge->windows, &bus_res); if (err) goto err_remove_irq_domain; - err = devm_request_pci_bus_resources(dev, &res); - if (err) - goto err_free_res; + rockchip->root_bus_nr = bus_res->start; /* Get the I/O and memory ranges from DT */ - resource_list_for_each_entry(win, &res) { + resource_list_for_each_entry(win, &bridge->windows) { switch (resource_type(win->res)) { case IORESOURCE_IO: io = win->res; io->name = "I/O"; rockchip->io_size = resource_size(io); rockchip->io_bus_addr = io->start - win->offset; - err = pci_remap_iospace(io, io_base); - if (err) { - dev_warn(dev, "error %d: failed to map resource %pR\n", - err, io); - continue; - } rockchip->io = io; break; case IORESOURCE_MEM: @@ -1026,9 +1013,6 @@ static int rockchip_pcie_probe(struct platform_device *pdev) rockchip->mem_size = resource_size(mem); rockchip->mem_bus_addr = mem->start - win->offset; break; - case IORESOURCE_BUS: - rockchip->root_bus_nr = win->res->start; - break; default: continue; } @@ -1036,15 +1020,14 @@ static int rockchip_pcie_probe(struct platform_device *pdev) err = rockchip_pcie_cfg_atu(rockchip); if (err) - goto err_unmap_iospace; + goto err_remove_irq_domain; rockchip->msg_region = devm_ioremap(dev, rockchip->msg_bus_addr, SZ_1M); if (!rockchip->msg_region) { err = -ENOMEM; - goto err_unmap_iospace; + goto err_remove_irq_domain; } - list_splice_init(&res, &bridge->windows); bridge->dev.parent = dev; bridge->sysdata = rockchip; bridge->busnr = 0; @@ -1054,7 +1037,7 @@ static int rockchip_pcie_probe(struct platform_device *pdev) err = pci_scan_root_bus_bridge(bridge); if (err < 0) - goto err_unmap_iospace; + goto err_remove_irq_domain; bus = bridge->bus; @@ -1068,10 +1051,6 @@ static int rockchip_pcie_probe(struct platform_device *pdev) pci_bus_add_devices(bus); return 0; -err_unmap_iospace: - pci_unmap_iospace(rockchip->io); -err_free_res: - pci_free_resource_list(&res); err_remove_irq_domain: irq_domain_remove(rockchip->irq_domain); err_deinit_port: @@ -1097,7 +1076,6 @@ static int rockchip_pcie_remove(struct platform_device *pdev) pci_stop_root_bus(rockchip->root_bus); pci_remove_root_bus(rockchip->root_bus); - pci_unmap_iospace(rockchip->io); irq_domain_remove(rockchip->irq_domain); rockchip_pcie_deinit_phys(rockchip); From 62240a88004b0205beb0c1faca1c875c392b53f0 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:42 -0500 Subject: [PATCH 066/160] PCI: rockchip: Drop storing driver private outbound resource data The Rockchip host bridge driver doesn't need to store outboard resources in its private struct as they are already stored in struct pci_host_bridge. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Cc: Shawn Lin Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas Cc: Heiko Stuebner Cc: linux-rockchip@lists.infradead.org --- drivers/pci/controller/pcie-rockchip-host.c | 54 +++++++++------------ drivers/pci/controller/pcie-rockchip.h | 5 -- 2 files changed, 23 insertions(+), 36 deletions(-) diff --git a/drivers/pci/controller/pcie-rockchip-host.c b/drivers/pci/controller/pcie-rockchip-host.c index 8d2e6f2e141e..f375e55ea02e 100644 --- a/drivers/pci/controller/pcie-rockchip-host.c +++ b/drivers/pci/controller/pcie-rockchip-host.c @@ -806,19 +806,28 @@ static int rockchip_pcie_prog_ib_atu(struct rockchip_pcie *rockchip, static int rockchip_pcie_cfg_atu(struct rockchip_pcie *rockchip) { struct device *dev = rockchip->dev; + struct pci_host_bridge *bridge = pci_host_bridge_from_priv(rockchip); + struct resource_entry *entry; + u64 pci_addr, size; int offset; int err; int reg_no; rockchip_pcie_cfg_configuration_accesses(rockchip, AXI_WRAPPER_TYPE0_CFG); + entry = resource_list_first_type(&bridge->windows, IORESOURCE_MEM); + if (!entry) + return -ENODEV; - for (reg_no = 0; reg_no < (rockchip->mem_size >> 20); reg_no++) { + size = resource_size(entry->res); + pci_addr = entry->res->start - entry->offset; + rockchip->msg_bus_addr = pci_addr; + + for (reg_no = 0; reg_no < (size >> 20); reg_no++) { err = rockchip_pcie_prog_ob_atu(rockchip, reg_no + 1, AXI_WRAPPER_MEM_WRITE, 20 - 1, - rockchip->mem_bus_addr + - (reg_no << 20), + pci_addr + (reg_no << 20), 0); if (err) { dev_err(dev, "program RC mem outbound ATU failed\n"); @@ -832,14 +841,20 @@ static int rockchip_pcie_cfg_atu(struct rockchip_pcie *rockchip) return err; } - offset = rockchip->mem_size >> 20; - for (reg_no = 0; reg_no < (rockchip->io_size >> 20); reg_no++) { + entry = resource_list_first_type(&bridge->windows, IORESOURCE_IO); + if (!entry) + return -ENODEV; + + size = resource_size(entry->res); + pci_addr = entry->res->start - entry->offset; + + offset = size >> 20; + for (reg_no = 0; reg_no < (size >> 20); reg_no++) { err = rockchip_pcie_prog_ob_atu(rockchip, reg_no + 1 + offset, AXI_WRAPPER_IO_WRITE, 20 - 1, - rockchip->io_bus_addr + - (reg_no << 20), + pci_addr + (reg_no << 20), 0); if (err) { dev_err(dev, "program RC io outbound ATU failed\n"); @@ -852,8 +867,7 @@ static int rockchip_pcie_cfg_atu(struct rockchip_pcie *rockchip) AXI_WRAPPER_NOR_MSG, 20 - 1, 0, 0); - rockchip->msg_bus_addr = rockchip->mem_bus_addr + - ((reg_no + offset) << 20); + rockchip->msg_bus_addr += ((reg_no + offset) << 20); return err; } @@ -951,7 +965,6 @@ static int rockchip_pcie_probe(struct platform_device *pdev) struct pci_bus *bus, *child; struct pci_host_bridge *bridge; struct resource *bus_res; - struct resource_entry *win; int err; if (!dev->of_node) @@ -997,27 +1010,6 @@ static int rockchip_pcie_probe(struct platform_device *pdev) rockchip->root_bus_nr = bus_res->start; - /* Get the I/O and memory ranges from DT */ - resource_list_for_each_entry(win, &bridge->windows) { - switch (resource_type(win->res)) { - case IORESOURCE_IO: - io = win->res; - io->name = "I/O"; - rockchip->io_size = resource_size(io); - rockchip->io_bus_addr = io->start - win->offset; - rockchip->io = io; - break; - case IORESOURCE_MEM: - mem = win->res; - mem->name = "MEM"; - rockchip->mem_size = resource_size(mem); - rockchip->mem_bus_addr = mem->start - win->offset; - break; - default: - continue; - } - } - err = rockchip_pcie_cfg_atu(rockchip); if (err) goto err_remove_irq_domain; diff --git a/drivers/pci/controller/pcie-rockchip.h b/drivers/pci/controller/pcie-rockchip.h index 8e87a059ce73..bef42a803b56 100644 --- a/drivers/pci/controller/pcie-rockchip.h +++ b/drivers/pci/controller/pcie-rockchip.h @@ -304,13 +304,8 @@ struct rockchip_pcie { struct irq_domain *irq_domain; int offset; struct pci_bus *root_bus; - struct resource *io; - phys_addr_t io_bus_addr; - u32 io_size; void __iomem *msg_region; - u32 mem_size; phys_addr_t msg_bus_addr; - phys_addr_t mem_bus_addr; bool is_rc; struct resource *mem_res; }; From e0aebfe84a2fb02ca1f195054595af9c3c01f12e Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:43 -0500 Subject: [PATCH 067/160] PCI: v3-semi: Use pci_parse_request_of_pci_ranges() Convert V3 host bridge to use the common pci_parse_request_of_pci_ranges(). Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Acked-by: Linus Walleij Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas --- drivers/pci/controller/pci-v3-semi.c | 35 +++++----------------------- 1 file changed, 6 insertions(+), 29 deletions(-) diff --git a/drivers/pci/controller/pci-v3-semi.c b/drivers/pci/controller/pci-v3-semi.c index d219404bad92..96677520f6c1 100644 --- a/drivers/pci/controller/pci-v3-semi.c +++ b/drivers/pci/controller/pci-v3-semi.c @@ -241,10 +241,8 @@ struct v3_pci { void __iomem *config_base; struct pci_bus *bus; u32 config_mem; - u32 io_mem; u32 non_pre_mem; u32 pre_mem; - phys_addr_t io_bus_addr; phys_addr_t non_pre_bus_addr; phys_addr_t pre_bus_addr; struct regmap *map; @@ -520,35 +518,22 @@ static int v3_integrator_init(struct v3_pci *v3) } static int v3_pci_setup_resource(struct v3_pci *v3, - resource_size_t io_base, struct pci_host_bridge *host, struct resource_entry *win) { struct device *dev = v3->dev; struct resource *mem; struct resource *io; - int ret; switch (resource_type(win->res)) { case IORESOURCE_IO: io = win->res; - io->name = "V3 PCI I/O"; - v3->io_mem = io_base; - v3->io_bus_addr = io->start - win->offset; - dev_dbg(dev, "I/O window %pR, bus addr %pap\n", - io, &v3->io_bus_addr); - ret = devm_pci_remap_iospace(dev, io, io_base); - if (ret) { - dev_warn(dev, - "error %d: failed to map resource %pR\n", - ret, io); - return ret; - } + /* Setup window 2 - PCI I/O */ - writel(v3_addr_to_lb_base2(v3->io_mem) | + writel(v3_addr_to_lb_base2(pci_pio_to_address(io->start)) | V3_LB_BASE2_ENABLE, v3->base + V3_LB_BASE2); - writew(v3_addr_to_lb_map2(v3->io_bus_addr), + writew(v3_addr_to_lb_map2(io->start - win->offset), v3->base + V3_LB_MAP2); break; case IORESOURCE_MEM: @@ -732,7 +717,6 @@ static int v3_pci_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct device_node *np = dev->of_node; - resource_size_t io_base; struct resource *regs; struct resource_entry *win; struct v3_pci *v3; @@ -741,7 +725,6 @@ static int v3_pci_probe(struct platform_device *pdev) u16 val; int irq; int ret; - LIST_HEAD(res); host = pci_alloc_host_bridge(sizeof(*v3)); if (!host) @@ -793,12 +776,7 @@ static int v3_pci_probe(struct platform_device *pdev) if (IS_ERR(v3->config_base)) return PTR_ERR(v3->config_base); - ret = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, &res, - &io_base); - if (ret) - return ret; - - ret = devm_request_pci_bus_resources(dev, &res); + ret = pci_parse_request_of_pci_ranges(dev, &host->windows, NULL); if (ret) return ret; @@ -852,8 +830,8 @@ static int v3_pci_probe(struct platform_device *pdev) writew(val, v3->base + V3_PCI_CMD); /* Get the I/O and memory ranges from DT */ - resource_list_for_each_entry(win, &res) { - ret = v3_pci_setup_resource(v3, io_base, host, win); + resource_list_for_each_entry(win, &host->windows) { + ret = v3_pci_setup_resource(v3, host, win); if (ret) { dev_err(dev, "error setting up resources\n"); return ret; @@ -931,7 +909,6 @@ static int v3_pci_probe(struct platform_device *pdev) val |= V3_SYSTEM_M_LOCK; writew(val, v3->base + V3_SYSTEM); - list_splice_init(&res, &host->windows); ret = pci_scan_root_bus_bridge(host); if (ret) { dev_err(dev, "failed to register host: %d\n", ret); From 83083e241d48aba4d0b92c3f5e274351c1560c97 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:44 -0500 Subject: [PATCH 068/160] PCI: xgene: Use pci_parse_request_of_pci_ranges() Convert the xgene host bridge to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list first. Just use bridge->windows directly and remove all the temporary list handling. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Cc: Toan Le Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas --- drivers/pci/controller/pci-xgene.c | 39 +++++++++--------------------- 1 file changed, 11 insertions(+), 28 deletions(-) diff --git a/drivers/pci/controller/pci-xgene.c b/drivers/pci/controller/pci-xgene.c index ffda3e8b4742..7d0f0395a479 100644 --- a/drivers/pci/controller/pci-xgene.c +++ b/drivers/pci/controller/pci-xgene.c @@ -405,15 +405,13 @@ static void xgene_pcie_setup_cfg_reg(struct xgene_pcie_port *port) xgene_pcie_writel(port, CFGCTL, EN_REG); } -static int xgene_pcie_map_ranges(struct xgene_pcie_port *port, - struct list_head *res, - resource_size_t io_base) +static int xgene_pcie_map_ranges(struct xgene_pcie_port *port) { + struct pci_host_bridge *bridge = pci_host_bridge_from_priv(port); struct resource_entry *window; struct device *dev = port->dev; - int ret; - resource_list_for_each_entry(window, res) { + resource_list_for_each_entry(window, &bridge->windows) { struct resource *res = window->res; u64 restype = resource_type(res); @@ -421,11 +419,9 @@ static int xgene_pcie_map_ranges(struct xgene_pcie_port *port, switch (restype) { case IORESOURCE_IO: - xgene_pcie_setup_ob_reg(port, res, OMR3BARL, io_base, + xgene_pcie_setup_ob_reg(port, res, OMR3BARL, + pci_pio_to_address(res->start), res->start - window->offset); - ret = devm_pci_remap_iospace(dev, res, io_base); - if (ret < 0) - return ret; break; case IORESOURCE_MEM: if (res->flags & IORESOURCE_PREFETCH) @@ -567,8 +563,7 @@ static void xgene_pcie_clear_config(struct xgene_pcie_port *port) xgene_pcie_writel(port, i, 0); } -static int xgene_pcie_setup(struct xgene_pcie_port *port, struct list_head *res, - resource_size_t io_base) +static int xgene_pcie_setup(struct xgene_pcie_port *port) { struct device *dev = port->dev; u32 val, lanes = 0, speed = 0; @@ -580,7 +575,7 @@ static int xgene_pcie_setup(struct xgene_pcie_port *port, struct list_head *res, val = (XGENE_PCIE_DEVICEID << 16) | XGENE_PCIE_VENDORID; xgene_pcie_writel(port, BRIDGE_CFG_0, val); - ret = xgene_pcie_map_ranges(port, res, io_base); + ret = xgene_pcie_map_ranges(port); if (ret) return ret; @@ -607,11 +602,9 @@ static int xgene_pcie_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; struct device_node *dn = dev->of_node; struct xgene_pcie_port *port; - resource_size_t iobase = 0; struct pci_bus *bus, *child; struct pci_host_bridge *bridge; int ret; - LIST_HEAD(res); bridge = devm_pci_alloc_host_bridge(dev, sizeof(*port)); if (!bridge) @@ -634,20 +627,14 @@ static int xgene_pcie_probe(struct platform_device *pdev) if (ret) return ret; - ret = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, &res, - &iobase); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); if (ret) return ret; - ret = devm_request_pci_bus_resources(dev, &res); + ret = xgene_pcie_setup(port); if (ret) - goto error; + return ret; - ret = xgene_pcie_setup(port, &res, iobase); - if (ret) - goto error; - - list_splice_init(&res, &bridge->windows); bridge->dev.parent = dev; bridge->sysdata = port; bridge->busnr = 0; @@ -657,7 +644,7 @@ static int xgene_pcie_probe(struct platform_device *pdev) ret = pci_scan_root_bus_bridge(bridge); if (ret < 0) - goto error; + return ret; bus = bridge->bus; @@ -666,10 +653,6 @@ static int xgene_pcie_probe(struct platform_device *pdev) pcie_bus_configure_settings(child); pci_bus_add_devices(bus); return 0; - -error: - pci_free_resource_list(&res); - return ret; } static const struct of_device_id xgene_pcie_match_table[] = { From ee352c272e41c67aac887cbfd8c6aa71e5721997 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:45 -0500 Subject: [PATCH 069/160] PCI: xilinx: Use pci_parse_request_of_pci_ranges() Convert the Xilinx host bridge to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list first. Just use bridge->windows directly and remove all the temporary list handling. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas Cc: Michal Simek --- drivers/pci/controller/pcie-xilinx.c | 17 ++--------------- 1 file changed, 2 insertions(+), 15 deletions(-) diff --git a/drivers/pci/controller/pcie-xilinx.c b/drivers/pci/controller/pcie-xilinx.c index 5bf3af3b28e6..257702288787 100644 --- a/drivers/pci/controller/pcie-xilinx.c +++ b/drivers/pci/controller/pcie-xilinx.c @@ -619,8 +619,6 @@ static int xilinx_pcie_probe(struct platform_device *pdev) struct pci_bus *bus, *child; struct pci_host_bridge *bridge; int err; - resource_size_t iobase = 0; - LIST_HEAD(res); if (!dev->of_node) return -ENODEV; @@ -647,19 +645,12 @@ static int xilinx_pcie_probe(struct platform_device *pdev) return err; } - err = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, &res, - &iobase); + err = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); if (err) { dev_err(dev, "Getting bridge resources failed\n"); return err; } - err = devm_request_pci_bus_resources(dev, &res); - if (err) - goto error; - - - list_splice_init(&res, &bridge->windows); bridge->dev.parent = dev; bridge->sysdata = port; bridge->busnr = 0; @@ -673,7 +664,7 @@ static int xilinx_pcie_probe(struct platform_device *pdev) #endif err = pci_scan_root_bus_bridge(bridge); if (err < 0) - goto error; + return err; bus = bridge->bus; @@ -682,10 +673,6 @@ static int xilinx_pcie_probe(struct platform_device *pdev) pcie_bus_configure_settings(child); pci_bus_add_devices(bus); return 0; - -error: - pci_free_resource_list(&res); - return err; } static const struct of_device_id xilinx_pcie_of_match[] = { From 3c65ebff8faedfc3386e6e1ad91adf2bdb8eeaa7 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:46 -0500 Subject: [PATCH 070/160] PCI: xilinx-nwl: Use pci_parse_request_of_pci_ranges() Convert the xilinx-nwl host bridge to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list first. Just use bridge->windows directly and remove all the temporary list handling. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas Cc: Michal Simek --- drivers/pci/controller/pcie-xilinx-nwl.c | 20 ++++---------------- 1 file changed, 4 insertions(+), 16 deletions(-) diff --git a/drivers/pci/controller/pcie-xilinx-nwl.c b/drivers/pci/controller/pcie-xilinx-nwl.c index 45c0f344ccd1..e135a4b60489 100644 --- a/drivers/pci/controller/pcie-xilinx-nwl.c +++ b/drivers/pci/controller/pcie-xilinx-nwl.c @@ -821,8 +821,6 @@ static int nwl_pcie_probe(struct platform_device *pdev) struct pci_bus *child; struct pci_host_bridge *bridge; int err; - resource_size_t iobase = 0; - LIST_HEAD(res); bridge = devm_pci_alloc_host_bridge(dev, sizeof(*pcie)); if (!bridge) @@ -845,24 +843,18 @@ static int nwl_pcie_probe(struct platform_device *pdev) return err; } - err = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, &res, - &iobase); + err = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); if (err) { dev_err(dev, "Getting bridge resources failed\n"); return err; } - err = devm_request_pci_bus_resources(dev, &res); - if (err) - goto error; - err = nwl_pcie_init_irq_domain(pcie); if (err) { dev_err(dev, "Failed creating IRQ Domain\n"); - goto error; + return err; } - list_splice_init(&res, &bridge->windows); bridge->dev.parent = dev; bridge->sysdata = pcie; bridge->busnr = pcie->root_busno; @@ -874,13 +866,13 @@ static int nwl_pcie_probe(struct platform_device *pdev) err = nwl_pcie_enable_msi(pcie); if (err < 0) { dev_err(dev, "failed to enable MSI support: %d\n", err); - goto error; + return err; } } err = pci_scan_root_bus_bridge(bridge); if (err) - goto error; + return err; bus = bridge->bus; @@ -889,10 +881,6 @@ static int nwl_pcie_probe(struct platform_device *pdev) pcie_bus_configure_settings(child); pci_bus_add_devices(bus); return 0; - -error: - pci_free_resource_list(&res); - return err; } static struct platform_driver nwl_pcie_driver = { From f9f4fdaa3509b804410b4742bcad9469151f4ee0 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:47 -0500 Subject: [PATCH 071/160] PCI: versatile: Use pci_parse_request_of_pci_ranges() Convert ARM Versatile host bridge to use the common pci_parse_request_of_pci_ranges(). There's no need to assign the resources to a temporary list first. Just use bridge->windows directly and remove all the temporary list handling. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Acked-by: Linus Walleij Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/controller/pci-versatile.c | 64 +++++--------------------- 1 file changed, 11 insertions(+), 53 deletions(-) diff --git a/drivers/pci/controller/pci-versatile.c b/drivers/pci/controller/pci-versatile.c index f59ad2728c0b..18697f2ea345 100644 --- a/drivers/pci/controller/pci-versatile.c +++ b/drivers/pci/controller/pci-versatile.c @@ -62,65 +62,16 @@ static struct pci_ops pci_versatile_ops = { .write = pci_generic_config_write, }; -static int versatile_pci_parse_request_of_pci_ranges(struct device *dev, - struct list_head *res) -{ - int err, mem = 1, res_valid = 0; - resource_size_t iobase; - struct resource_entry *win, *tmp; - - err = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, res, &iobase); - if (err) - return err; - - err = devm_request_pci_bus_resources(dev, res); - if (err) - goto out_release_res; - - resource_list_for_each_entry_safe(win, tmp, res) { - struct resource *res = win->res; - - switch (resource_type(res)) { - case IORESOURCE_IO: - err = devm_pci_remap_iospace(dev, res, iobase); - if (err) { - dev_warn(dev, "error %d: failed to map resource %pR\n", - err, res); - resource_list_destroy_entry(win); - } - break; - case IORESOURCE_MEM: - res_valid |= !(res->flags & IORESOURCE_PREFETCH); - - writel(res->start >> 28, PCI_IMAP(mem)); - writel(PHYS_OFFSET >> 28, PCI_SMAP(mem)); - mem++; - - break; - } - } - - if (res_valid) - return 0; - - dev_err(dev, "non-prefetchable memory resource required\n"); - err = -EINVAL; - -out_release_res: - pci_free_resource_list(res); - return err; -} - static int versatile_pci_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct resource *res; - int ret, i, myslot = -1; + struct resource_entry *entry; + int ret, i, myslot = -1, mem = 1; u32 val; void __iomem *local_pci_cfg_base; struct pci_bus *bus, *child; struct pci_host_bridge *bridge; - LIST_HEAD(pci_res); bridge = devm_pci_alloc_host_bridge(dev, 0); if (!bridge) @@ -141,10 +92,18 @@ static int versatile_pci_probe(struct platform_device *pdev) if (IS_ERR(versatile_cfg_base[1])) return PTR_ERR(versatile_cfg_base[1]); - ret = versatile_pci_parse_request_of_pci_ranges(dev, &pci_res); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); if (ret) return ret; + resource_list_for_each_entry(entry, &bridge->windows) { + if (resource_type(entry->res) == IORESOURCE_MEM) { + writel(entry->res->start >> 28, PCI_IMAP(mem)); + writel(PHYS_OFFSET >> 28, PCI_SMAP(mem)); + mem++; + } + } + /* * We need to discover the PCI core first to configure itself * before the main PCI probing is performed @@ -197,7 +156,6 @@ static int versatile_pci_probe(struct platform_device *pdev) pci_add_flags(PCI_ENABLE_PROC_DOMAINS); pci_add_flags(PCI_REASSIGN_ALL_BUS); - list_splice_init(&pci_res, &bridge->windows); bridge->dev.parent = dev; bridge->sysdata = NULL; bridge->busnr = 0; From 2999dea8e94a8e32dadfe17970aa29eba46985b9 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:48 -0500 Subject: [PATCH 072/160] PCI: versatile: Remove usage of PHYS_OFFSET PHYS_OFFSET is not universally defined on all arches and using it prevents enabling COMPILE_TEST. PAGE_OFFSET and __pa() are always available, so use them to get the physical start of memory address. This should have probably used 'dma-ranges' to get the address, but we don't want to force a DT update to do that. At least in QEMU, the SMAP registers have no effect (or perhaps the only value that is handled is 0). Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Acked-by: Linus Walleij Cc: Lorenzo Pieralisi Cc: Andrew Murray Cc: Bjorn Helgaas --- drivers/pci/controller/pci-versatile.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/pci/controller/pci-versatile.c b/drivers/pci/controller/pci-versatile.c index 18697f2ea345..eae1b859990b 100644 --- a/drivers/pci/controller/pci-versatile.c +++ b/drivers/pci/controller/pci-versatile.c @@ -99,7 +99,7 @@ static int versatile_pci_probe(struct platform_device *pdev) resource_list_for_each_entry(entry, &bridge->windows) { if (resource_type(entry->res) == IORESOURCE_MEM) { writel(entry->res->start >> 28, PCI_IMAP(mem)); - writel(PHYS_OFFSET >> 28, PCI_SMAP(mem)); + writel(__pa(PAGE_OFFSET) >> 28, PCI_SMAP(mem)); mem++; } } @@ -136,9 +136,9 @@ static int versatile_pci_probe(struct platform_device *pdev) /* * Configure the PCI inbound memory windows to be 1:1 mapped to SDRAM */ - writel(PHYS_OFFSET, local_pci_cfg_base + PCI_BASE_ADDRESS_0); - writel(PHYS_OFFSET, local_pci_cfg_base + PCI_BASE_ADDRESS_1); - writel(PHYS_OFFSET, local_pci_cfg_base + PCI_BASE_ADDRESS_2); + writel(__pa(PAGE_OFFSET), local_pci_cfg_base + PCI_BASE_ADDRESS_0); + writel(__pa(PAGE_OFFSET), local_pci_cfg_base + PCI_BASE_ADDRESS_1); + writel(__pa(PAGE_OFFSET), local_pci_cfg_base + PCI_BASE_ADDRESS_2); /* * For many years the kernel and QEMU were symbiotically buggy From ecf8fd6d917dca1065e1021ff7fc6b6c282373f9 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:49 -0500 Subject: [PATCH 073/160] PCI: versatile: Enable COMPILE_TEST Since commit a574795bc383 ("PCI: generic,versatile: Remove unused pci_sys_data structures") the build dependency on ARM is gone, so let's enable COMPILE_TEST for versatile. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Acked-by: Linus Walleij Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/controller/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/controller/Kconfig b/drivers/pci/controller/Kconfig index 70e078238899..f5de9119e8d3 100644 --- a/drivers/pci/controller/Kconfig +++ b/drivers/pci/controller/Kconfig @@ -135,7 +135,7 @@ config PCI_V3_SEMI config PCI_VERSATILE bool "ARM Versatile PB PCI controller" - depends on ARCH_VERSATILE + depends on ARCH_VERSATILE || COMPILE_TEST config PCIE_IPROC tristate From 3c379a59b4795d7279d38c623e74b9790345a32b Mon Sep 17 00:00:00 2001 From: Hewenliang Date: Fri, 25 Oct 2019 21:35:55 -0400 Subject: [PATCH 074/160] tools: PCI: Fix fd leakage We should close fd before the return of run_test. Fixes: 3f2ed8134834 ("tools: PCI: Add a userspace tool to test PCI endpoint") Signed-off-by: Hewenliang Signed-off-by: Lorenzo Pieralisi Acked-by: Kishon Vijay Abraham I --- tools/pci/pcitest.c | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/pci/pcitest.c b/tools/pci/pcitest.c index cb1e51fcc84e..32b7c6f9043d 100644 --- a/tools/pci/pcitest.c +++ b/tools/pci/pcitest.c @@ -129,6 +129,7 @@ static int run_test(struct pci_test *test) } fflush(stdout); + close(fd); return (ret < 0) ? ret : 1 - ret; /* return 0 if test succeeded */ } From d8725e38dd9fe88bc7dab6e53acdf2e257c241db Mon Sep 17 00:00:00 2001 From: Xiaowei Bao Date: Mon, 2 Sep 2019 11:43:17 +0800 Subject: [PATCH 075/160] dt-bindings: pci: layerscape-pci: add compatible strings "fsl, ls1028a-pcie" Add the PCIe compatible string for LS1028A. Signed-off-by: Xiaowei Bao Signed-off-by: Hou Zhiqiang Signed-off-by: Lorenzo Pieralisi Reviewed-by: Rob Herring Reviewed-by: Andrew Murray --- Documentation/devicetree/bindings/pci/layerscape-pci.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/pci/layerscape-pci.txt b/Documentation/devicetree/bindings/pci/layerscape-pci.txt index e20ceaab9b38..99a386ea691c 100644 --- a/Documentation/devicetree/bindings/pci/layerscape-pci.txt +++ b/Documentation/devicetree/bindings/pci/layerscape-pci.txt @@ -21,6 +21,7 @@ Required properties: "fsl,ls1046a-pcie" "fsl,ls1043a-pcie" "fsl,ls1012a-pcie" + "fsl,ls1028a-pcie" EP mode: "fsl,ls1046a-pcie-ep", "fsl,ls-pcie-ep" - reg: base addresses and lengths of the PCIe controller register blocks. From 7973eb13aecfeebe2881c4285b0bedbfe89ff0fd Mon Sep 17 00:00:00 2001 From: Xiaowei Bao Date: Mon, 2 Sep 2019 11:43:19 +0800 Subject: [PATCH 076/160] PCI: layerscape: Add LS1028a support Add support for the LS1028a PCIe controller. Signed-off-by: Xiaowei Bao Signed-off-by: Hou Zhiqiang Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Acked-by: Minghuan Lian --- drivers/pci/controller/dwc/pci-layerscape.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/pci/controller/dwc/pci-layerscape.c b/drivers/pci/controller/dwc/pci-layerscape.c index 3a5fa26d5e56..f24f79a70d9a 100644 --- a/drivers/pci/controller/dwc/pci-layerscape.c +++ b/drivers/pci/controller/dwc/pci-layerscape.c @@ -263,6 +263,7 @@ static const struct ls_pcie_drvdata ls2088_drvdata = { static const struct of_device_id ls_pcie_of_match[] = { { .compatible = "fsl,ls1012a-pcie", .data = &ls1046_drvdata }, { .compatible = "fsl,ls1021a-pcie", .data = &ls1021_drvdata }, + { .compatible = "fsl,ls1028a-pcie", .data = &ls2088_drvdata }, { .compatible = "fsl,ls1043a-pcie", .data = &ls1043_drvdata }, { .compatible = "fsl,ls1046a-pcie", .data = &ls1046_drvdata }, { .compatible = "fsl,ls2080a-pcie", .data = &ls2080_drvdata }, From af072edb83555e768d2b30c6a31629d3eb3e0527 Mon Sep 17 00:00:00 2001 From: Andrew Murray Date: Thu, 5 Sep 2019 16:05:28 +0100 Subject: [PATCH 077/160] PCI: rcar: Remove unnecessary header include (../pci.h) Remove unnecessary header include (../pci.h) since it doesn't provide any needed symbols. Signed-off-by: Andrew Murray Signed-off-by: Lorenzo Pieralisi Reviewed-by: Simon Horman Reviewed-by: Kieran Bingham Reviewed-by: Geert Uytterhoeven --- drivers/pci/controller/pcie-rcar.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/pci/controller/pcie-rcar.c b/drivers/pci/controller/pcie-rcar.c index f6a669a9af41..ee1c38c2fac9 100644 --- a/drivers/pci/controller/pcie-rcar.c +++ b/drivers/pci/controller/pcie-rcar.c @@ -30,8 +30,6 @@ #include #include -#include "../pci.h" - #define PCIECAR 0x000010 #define PCIECCTLR 0x000018 #define CONFIG_SEND_ENABLE BIT(31) From 85bff4c3d320bffceab0c96ee2be4d5f55a3a4e7 Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Sat, 26 Oct 2019 20:26:58 +0200 Subject: [PATCH 078/160] PCI: rcar: Move the inbound index check Since the 'idx' variable value is stored across multiple calls to rcar_pcie_inbound_ranges() function, and the 'idx' value is used to index registers which are written, subsequent calls might cause the 'idx' value to be high enough to trigger writes into nonexistent registers. Fix this by moving the 'idx' value check to the beginning of the loop. Signed-off-by: Marek Vasut Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Reviewed-by: Yoshihiro Shimoda Cc: Geert Uytterhoeven Cc: Lorenzo Pieralisi Cc: Wolfram Sang Cc: linux-renesas-soc@vger.kernel.org --- drivers/pci/controller/pcie-rcar.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/pci/controller/pcie-rcar.c b/drivers/pci/controller/pcie-rcar.c index ee1c38c2fac9..04ff6c4baa70 100644 --- a/drivers/pci/controller/pcie-rcar.c +++ b/drivers/pci/controller/pcie-rcar.c @@ -1046,6 +1046,10 @@ static int rcar_pcie_inbound_ranges(struct rcar_pcie *pcie, mask &= ~0xf; while (cpu_addr < cpu_end) { + if (idx >= MAX_NR_INBOUND_MAPS - 1) { + dev_err(pcie->dev, "Failed to map inbound regions!\n"); + return -EINVAL; + } /* * Set up 64-bit inbound regions as the range parser doesn't * distinguish between 32 and 64-bit types. @@ -1065,11 +1069,6 @@ static int rcar_pcie_inbound_ranges(struct rcar_pcie *pcie, pci_addr += size; cpu_addr += size; idx += 2; - - if (idx > MAX_NR_INBOUND_MAPS) { - dev_err(pcie->dev, "Failed to map inbound regions!\n"); - return -EINVAL; - } } *index = idx; From 767c7846419cc562c9dd4f14cc617c2b9b1b96cd Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Sat, 26 Oct 2019 20:26:59 +0200 Subject: [PATCH 079/160] PCI: rcar: Recalculate inbound range alignment for each controller entry Due to hardware constraints, the size of each inbound range entry populated into the controller cannot be larger than the alignment of the entry's start address. Currently, the alignment for each "dma-ranges" inbound range is calculated only once for each range and the increment for programming the controller is also derived from it only once. Thus, a "dma-ranges" entry describing a memory at 0x48000000 and size 0x38000000 would lead to multiple controller entries, each 0x08000000 long. This is inefficient, especially considering that by adding the size to the start address, the alignment increases. This patch moves the alignment calculation into the loop populating the controller entries, thus updating the alignment for each controller entry. Tested-by: Yoshihiro Shimoda Signed-off-by: Marek Vasut Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Reviewed-by: Simon Horman Reviewed-by: Yoshihiro Shimoda Cc: Geert Uytterhoeven Cc: Lorenzo Pieralisi Cc: Wolfram Sang Cc: linux-renesas-soc@vger.kernel.org --- drivers/pci/controller/pcie-rcar.c | 37 +++++++++++++++--------------- 1 file changed, 19 insertions(+), 18 deletions(-) diff --git a/drivers/pci/controller/pcie-rcar.c b/drivers/pci/controller/pcie-rcar.c index 04ff6c4baa70..40d8c54a17d1 100644 --- a/drivers/pci/controller/pcie-rcar.c +++ b/drivers/pci/controller/pcie-rcar.c @@ -1027,29 +1027,30 @@ static int rcar_pcie_inbound_ranges(struct rcar_pcie *pcie, if (restype & IORESOURCE_PREFETCH) flags |= LAM_PREFETCH; - /* - * If the size of the range is larger than the alignment of the start - * address, we have to use multiple entries to perform the mapping. - */ - if (cpu_addr > 0) { - unsigned long nr_zeros = __ffs64(cpu_addr); - u64 alignment = 1ULL << nr_zeros; - - size = min(range->size, alignment); - } else { - size = range->size; - } - /* Hardware supports max 4GiB inbound region */ - size = min(size, 1ULL << 32); - - mask = roundup_pow_of_two(size) - 1; - mask &= ~0xf; - while (cpu_addr < cpu_end) { if (idx >= MAX_NR_INBOUND_MAPS - 1) { dev_err(pcie->dev, "Failed to map inbound regions!\n"); return -EINVAL; } + /* + * If the size of the range is larger than the alignment of + * the start address, we have to use multiple entries to + * perform the mapping. + */ + if (cpu_addr > 0) { + unsigned long nr_zeros = __ffs64(cpu_addr); + u64 alignment = 1ULL << nr_zeros; + + size = min(range->size, alignment); + } else { + size = range->size; + } + /* Hardware supports max 4GiB inbound region */ + size = min(size, 1ULL << 32); + + mask = roundup_pow_of_two(size) - 1; + mask &= ~0xf; + /* * Set up 64-bit inbound regions as the range parser doesn't * distinguish between 32 and 64-bit types. From bd22885aa188f135fd98382febfec650601ec998 Mon Sep 17 00:00:00 2001 From: Tom Joseph Date: Mon, 11 Nov 2019 12:30:43 +0000 Subject: [PATCH 080/160] PCI: cadence: Refactor driver to use as a core library Cadence PCIe host and endpoint IP may be embedded into a variety of SoCs/platforms. Let's extract the platform related APIs/Structures in the current driver to a separate file (pcie-cadence-plat.c), such that the common functionality can be used by future platforms. Signed-off-by: Tom Joseph Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray --- drivers/pci/controller/Kconfig | 31 +++- drivers/pci/controller/Makefile | 1 + drivers/pci/controller/pcie-cadence-ep.c | 96 +----------- drivers/pci/controller/pcie-cadence-host.c | 95 +---------- drivers/pci/controller/pcie-cadence-plat.c | 174 +++++++++++++++++++++ drivers/pci/controller/pcie-cadence.h | 77 +++++++++ 6 files changed, 287 insertions(+), 187 deletions(-) create mode 100644 drivers/pci/controller/pcie-cadence-plat.c diff --git a/drivers/pci/controller/Kconfig b/drivers/pci/controller/Kconfig index 70e078238899..b593616e9f52 100644 --- a/drivers/pci/controller/Kconfig +++ b/drivers/pci/controller/Kconfig @@ -28,23 +28,38 @@ config PCIE_CADENCE bool config PCIE_CADENCE_HOST - bool "Cadence PCIe host controller" + bool depends on OF - depends on PCI select IRQ_DOMAIN select PCIE_CADENCE - help - Say Y here if you want to support the Cadence PCIe controller in host - mode. This PCIe controller may be embedded into many different vendors - SoCs. config PCIE_CADENCE_EP - bool "Cadence PCIe endpoint controller" + bool depends on OF depends on PCI_ENDPOINT select PCIE_CADENCE + +config PCIE_CADENCE_PLAT + bool + +config PCIE_CADENCE_PLAT_HOST + bool "Cadence PCIe platform host controller" + depends on OF + select PCIE_CADENCE_HOST + select PCIE_CADENCE_PLAT help - Say Y here if you want to support the Cadence PCIe controller in + Say Y here if you want to support the Cadence PCIe platform controller in + host mode. This PCIe controller may be embedded into many different + vendors SoCs. + +config PCIE_CADENCE_PLAT_EP + bool "Cadence PCIe platform endpoint controller" + depends on OF + depends on PCI_ENDPOINT + select PCIE_CADENCE_EP + select PCIE_CADENCE_PLAT + help + Say Y here if you want to support the Cadence PCIe platform controller in endpoint mode. This PCIe controller may be embedded into many different vendors SoCs. diff --git a/drivers/pci/controller/Makefile b/drivers/pci/controller/Makefile index a2a22c9d91af..42a6363d38f1 100644 --- a/drivers/pci/controller/Makefile +++ b/drivers/pci/controller/Makefile @@ -2,6 +2,7 @@ obj-$(CONFIG_PCIE_CADENCE) += pcie-cadence.o obj-$(CONFIG_PCIE_CADENCE_HOST) += pcie-cadence-host.o obj-$(CONFIG_PCIE_CADENCE_EP) += pcie-cadence-ep.o +obj-$(CONFIG_PCIE_CADENCE_PLAT) += pcie-cadence-plat.o obj-$(CONFIG_PCI_FTPCI100) += pci-ftpci100.o obj-$(CONFIG_PCI_HYPERV) += pci-hyperv.o obj-$(CONFIG_PCI_HYPERV_INTERFACE) += pci-hyperv-intf.o diff --git a/drivers/pci/controller/pcie-cadence-ep.c b/drivers/pci/controller/pcie-cadence-ep.c index def7820cb824..1c173dad67d1 100644 --- a/drivers/pci/controller/pcie-cadence-ep.c +++ b/drivers/pci/controller/pcie-cadence-ep.c @@ -17,35 +17,6 @@ #define CDNS_PCIE_EP_IRQ_PCI_ADDR_NONE 0x1 #define CDNS_PCIE_EP_IRQ_PCI_ADDR_LEGACY 0x3 -/** - * struct cdns_pcie_ep - private data for this PCIe endpoint controller driver - * @pcie: Cadence PCIe controller - * @max_regions: maximum number of regions supported by hardware - * @ob_region_map: bitmask of mapped outbound regions - * @ob_addr: base addresses in the AXI bus where the outbound regions start - * @irq_phys_addr: base address on the AXI bus where the MSI/legacy IRQ - * dedicated outbound regions is mapped. - * @irq_cpu_addr: base address in the CPU space where a write access triggers - * the sending of a memory write (MSI) / normal message (legacy - * IRQ) TLP through the PCIe bus. - * @irq_pci_addr: used to save the current mapping of the MSI/legacy IRQ - * dedicated outbound region. - * @irq_pci_fn: the latest PCI function that has updated the mapping of - * the MSI/legacy IRQ dedicated outbound region. - * @irq_pending: bitmask of asserted legacy IRQs. - */ -struct cdns_pcie_ep { - struct cdns_pcie pcie; - u32 max_regions; - unsigned long ob_region_map; - phys_addr_t *ob_addr; - phys_addr_t irq_phys_addr; - void __iomem *irq_cpu_addr; - u64 irq_pci_addr; - u8 irq_pci_fn; - u8 irq_pending; -}; - static int cdns_pcie_ep_write_header(struct pci_epc *epc, u8 fn, struct pci_epf_header *hdr) { @@ -424,28 +395,17 @@ static const struct pci_epc_ops cdns_pcie_epc_ops = { .get_features = cdns_pcie_ep_get_features, }; -static const struct of_device_id cdns_pcie_ep_of_match[] = { - { .compatible = "cdns,cdns-pcie-ep" }, - { }, -}; - -static int cdns_pcie_ep_probe(struct platform_device *pdev) +int cdns_pcie_ep_setup(struct cdns_pcie_ep *ep) { - struct device *dev = &pdev->dev; + struct device *dev = ep->pcie.dev; + struct platform_device *pdev = to_platform_device(dev); struct device_node *np = dev->of_node; - struct cdns_pcie_ep *ep; - struct cdns_pcie *pcie; - struct pci_epc *epc; + struct cdns_pcie *pcie = &ep->pcie; struct resource *res; + struct pci_epc *epc; int ret; - int phy_count; - ep = devm_kzalloc(dev, sizeof(*ep), GFP_KERNEL); - if (!ep) - return -ENOMEM; - - pcie = &ep->pcie; pcie->is_rc = false; res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "reg"); @@ -474,19 +434,6 @@ static int cdns_pcie_ep_probe(struct platform_device *pdev) if (!ep->ob_addr) return -ENOMEM; - ret = cdns_pcie_init_phy(dev, pcie); - if (ret) { - dev_err(dev, "failed to init phy\n"); - return ret; - } - platform_set_drvdata(pdev, pcie); - pm_runtime_enable(dev); - ret = pm_runtime_get_sync(dev); - if (ret < 0) { - dev_err(dev, "pm_runtime_get_sync() failed\n"); - goto err_get_sync; - } - /* Disable all but function 0 (anyway BIT(0) is hardwired to 1). */ cdns_pcie_writel(pcie, CDNS_PCIE_LM_EP_FUNC_CFG, BIT(0)); @@ -528,38 +475,5 @@ static int cdns_pcie_ep_probe(struct platform_device *pdev) err_init: pm_runtime_put_sync(dev); - err_get_sync: - pm_runtime_disable(dev); - cdns_pcie_disable_phy(pcie); - phy_count = pcie->phy_count; - while (phy_count--) - device_link_del(pcie->link[phy_count]); - return ret; } - -static void cdns_pcie_ep_shutdown(struct platform_device *pdev) -{ - struct device *dev = &pdev->dev; - struct cdns_pcie *pcie = dev_get_drvdata(dev); - int ret; - - ret = pm_runtime_put_sync(dev); - if (ret < 0) - dev_dbg(dev, "pm_runtime_put_sync failed\n"); - - pm_runtime_disable(dev); - - cdns_pcie_disable_phy(pcie); -} - -static struct platform_driver cdns_pcie_ep_driver = { - .driver = { - .name = "cdns-pcie-ep", - .of_match_table = cdns_pcie_ep_of_match, - .pm = &cdns_pcie_pm_ops, - }, - .probe = cdns_pcie_ep_probe, - .shutdown = cdns_pcie_ep_shutdown, -}; -builtin_platform_driver(cdns_pcie_ep_driver); diff --git a/drivers/pci/controller/pcie-cadence-host.c b/drivers/pci/controller/pcie-cadence-host.c index 97e251090b4f..8a42afd1f3fa 100644 --- a/drivers/pci/controller/pcie-cadence-host.c +++ b/drivers/pci/controller/pcie-cadence-host.c @@ -11,33 +11,6 @@ #include "pcie-cadence.h" -/** - * struct cdns_pcie_rc - private data for this PCIe Root Complex driver - * @pcie: Cadence PCIe controller - * @dev: pointer to PCIe device - * @cfg_res: start/end offsets in the physical system memory to map PCI - * configuration space accesses - * @bus_range: first/last buses behind the PCIe host controller - * @cfg_base: IO mapped window to access the PCI configuration space of a - * single function at a time - * @max_regions: maximum number of regions supported by the hardware - * @no_bar_nbits: Number of bits to keep for inbound (PCIe -> CPU) address - * translation (nbits sets into the "no BAR match" register) - * @vendor_id: PCI vendor ID - * @device_id: PCI device ID - */ -struct cdns_pcie_rc { - struct cdns_pcie pcie; - struct device *dev; - struct resource *cfg_res; - struct resource *bus_range; - void __iomem *cfg_base; - u32 max_regions; - u32 no_bar_nbits; - u16 vendor_id; - u16 device_id; -}; - static void __iomem *cdns_pci_map_bus(struct pci_bus *bus, unsigned int devfn, int where) { @@ -92,11 +65,6 @@ static struct pci_ops cdns_pcie_host_ops = { .write = pci_generic_config_write, }; -static const struct of_device_id cdns_pcie_host_of_match[] = { - { .compatible = "cdns,cdns-pcie-host" }, - - { }, -}; static int cdns_pcie_host_init_root_port(struct cdns_pcie_rc *rc) { @@ -136,10 +104,10 @@ static int cdns_pcie_host_init_root_port(struct cdns_pcie_rc *rc) static int cdns_pcie_host_init_address_translation(struct cdns_pcie_rc *rc) { struct cdns_pcie *pcie = &rc->pcie; - struct resource *cfg_res = rc->cfg_res; struct resource *mem_res = pcie->mem_res; struct resource *bus_range = rc->bus_range; - struct device *dev = rc->dev; + struct resource *cfg_res = rc->cfg_res; + struct device *dev = pcie->dev; struct device_node *np = dev->of_node; struct of_pci_range_parser parser; struct of_pci_range range; @@ -233,25 +201,21 @@ static int cdns_pcie_host_init(struct device *dev, return err; } -static int cdns_pcie_host_probe(struct platform_device *pdev) +int cdns_pcie_host_setup(struct cdns_pcie_rc *rc) { - struct device *dev = &pdev->dev; + struct device *dev = rc->pcie.dev; + struct platform_device *pdev = to_platform_device(dev); struct device_node *np = dev->of_node; struct pci_host_bridge *bridge; struct list_head resources; - struct cdns_pcie_rc *rc; struct cdns_pcie *pcie; struct resource *res; int ret; - int phy_count; - bridge = devm_pci_alloc_host_bridge(dev, sizeof(*rc)); + bridge = pci_host_bridge_from_priv(rc); if (!bridge) return -ENOMEM; - rc = pci_host_bridge_priv(bridge); - rc->dev = dev; - pcie = &rc->pcie; pcie->is_rc = true; @@ -287,22 +251,9 @@ static int cdns_pcie_host_probe(struct platform_device *pdev) dev_err(dev, "missing \"mem\"\n"); return -EINVAL; } + pcie->mem_res = res; - ret = cdns_pcie_init_phy(dev, pcie); - if (ret) { - dev_err(dev, "failed to init phy\n"); - return ret; - } - platform_set_drvdata(pdev, pcie); - - pm_runtime_enable(dev); - ret = pm_runtime_get_sync(dev); - if (ret < 0) { - dev_err(dev, "pm_runtime_get_sync() failed\n"); - goto err_get_sync; - } - ret = cdns_pcie_host_init(dev, &resources, rc); if (ret) goto err_init; @@ -326,37 +277,5 @@ static int cdns_pcie_host_probe(struct platform_device *pdev) err_init: pm_runtime_put_sync(dev); - err_get_sync: - pm_runtime_disable(dev); - cdns_pcie_disable_phy(pcie); - phy_count = pcie->phy_count; - while (phy_count--) - device_link_del(pcie->link[phy_count]); - return ret; } - -static void cdns_pcie_shutdown(struct platform_device *pdev) -{ - struct device *dev = &pdev->dev; - struct cdns_pcie *pcie = dev_get_drvdata(dev); - int ret; - - ret = pm_runtime_put_sync(dev); - if (ret < 0) - dev_dbg(dev, "pm_runtime_put_sync failed\n"); - - pm_runtime_disable(dev); - cdns_pcie_disable_phy(pcie); -} - -static struct platform_driver cdns_pcie_host_driver = { - .driver = { - .name = "cdns-pcie-host", - .of_match_table = cdns_pcie_host_of_match, - .pm = &cdns_pcie_pm_ops, - }, - .probe = cdns_pcie_host_probe, - .shutdown = cdns_pcie_shutdown, -}; -builtin_platform_driver(cdns_pcie_host_driver); diff --git a/drivers/pci/controller/pcie-cadence-plat.c b/drivers/pci/controller/pcie-cadence-plat.c new file mode 100644 index 000000000000..f5c6bf6dfcb8 --- /dev/null +++ b/drivers/pci/controller/pcie-cadence-plat.c @@ -0,0 +1,174 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Cadence PCIe platform driver. + * + * Copyright (c) 2019, Cadence Design Systems + * Author: Tom Joseph + */ +#include +#include +#include +#include +#include +#include +#include "pcie-cadence.h" + +/** + * struct cdns_plat_pcie - private data for this PCIe platform driver + * @pcie: Cadence PCIe controller + * @is_rc: Set to 1 indicates the PCIe controller mode is Root Complex, + * if 0 it is in Endpoint mode. + */ +struct cdns_plat_pcie { + struct cdns_pcie *pcie; + bool is_rc; +}; + +struct cdns_plat_pcie_of_data { + bool is_rc; +}; + +static const struct of_device_id cdns_plat_pcie_of_match[]; + +static int cdns_plat_pcie_probe(struct platform_device *pdev) +{ + const struct cdns_plat_pcie_of_data *data; + struct cdns_plat_pcie *cdns_plat_pcie; + const struct of_device_id *match; + struct device *dev = &pdev->dev; + struct pci_host_bridge *bridge; + struct cdns_pcie_ep *ep; + struct cdns_pcie_rc *rc; + int phy_count; + bool is_rc; + int ret; + + match = of_match_device(cdns_plat_pcie_of_match, dev); + if (!match) + return -EINVAL; + + data = (struct cdns_plat_pcie_of_data *)match->data; + is_rc = data->is_rc; + + pr_debug(" Started %s with is_rc: %d\n", __func__, is_rc); + cdns_plat_pcie = devm_kzalloc(dev, sizeof(*cdns_plat_pcie), GFP_KERNEL); + if (!cdns_plat_pcie) + return -ENOMEM; + + platform_set_drvdata(pdev, cdns_plat_pcie); + if (is_rc) { + if (!IS_ENABLED(CONFIG_PCIE_CADENCE_PLAT_HOST)) + return -ENODEV; + + bridge = devm_pci_alloc_host_bridge(dev, sizeof(*rc)); + if (!bridge) + return -ENOMEM; + + rc = pci_host_bridge_priv(bridge); + rc->pcie.dev = dev; + cdns_plat_pcie->pcie = &rc->pcie; + cdns_plat_pcie->is_rc = is_rc; + + ret = cdns_pcie_init_phy(dev, cdns_plat_pcie->pcie); + if (ret) { + dev_err(dev, "failed to init phy\n"); + return ret; + } + pm_runtime_enable(dev); + ret = pm_runtime_get_sync(dev); + if (ret < 0) { + dev_err(dev, "pm_runtime_get_sync() failed\n"); + goto err_get_sync; + } + + ret = cdns_pcie_host_setup(rc); + if (ret) + goto err_init; + } else { + if (!IS_ENABLED(CONFIG_PCIE_CADENCE_PLAT_EP)) + return -ENODEV; + + ep = devm_kzalloc(dev, sizeof(*ep), GFP_KERNEL); + if (!ep) + return -ENOMEM; + + ep->pcie.dev = dev; + cdns_plat_pcie->pcie = &ep->pcie; + cdns_plat_pcie->is_rc = is_rc; + + ret = cdns_pcie_init_phy(dev, cdns_plat_pcie->pcie); + if (ret) { + dev_err(dev, "failed to init phy\n"); + return ret; + } + + pm_runtime_enable(dev); + ret = pm_runtime_get_sync(dev); + if (ret < 0) { + dev_err(dev, "pm_runtime_get_sync() failed\n"); + goto err_get_sync; + } + + ret = cdns_pcie_ep_setup(ep); + if (ret) + goto err_init; + } + + err_init: + pm_runtime_put_sync(dev); + + err_get_sync: + pm_runtime_disable(dev); + cdns_pcie_disable_phy(cdns_plat_pcie->pcie); + phy_count = cdns_plat_pcie->pcie->phy_count; + while (phy_count--) + device_link_del(cdns_plat_pcie->pcie->link[phy_count]); + + return 0; +} + +static void cdns_plat_pcie_shutdown(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct cdns_pcie *pcie = dev_get_drvdata(dev); + int ret; + + ret = pm_runtime_put_sync(dev); + if (ret < 0) + dev_dbg(dev, "pm_runtime_put_sync failed\n"); + + pm_runtime_disable(dev); + + cdns_pcie_disable_phy(pcie); +} + +static const struct cdns_plat_pcie_of_data cdns_plat_pcie_host_of_data = { + .is_rc = true, +}; + +static const struct cdns_plat_pcie_of_data cdns_plat_pcie_ep_of_data = { + .is_rc = false, +}; + +static const struct of_device_id cdns_plat_pcie_of_match[] = { + { + .compatible = "cdns,cdns-pcie-host", + .data = &cdns_plat_pcie_host_of_data, + }, + { + .compatible = "cdns,cdns-pcie-ep", + .data = &cdns_plat_pcie_ep_of_data, + }, + {}, +}; + +static struct platform_driver cdns_plat_pcie_driver = { + .driver = { + .name = "cdns-pcie", + .of_match_table = cdns_plat_pcie_of_match, + .pm = &cdns_pcie_pm_ops, + }, + .probe = cdns_plat_pcie_probe, + .shutdown = cdns_plat_pcie_shutdown, +}; +builtin_platform_driver(cdns_plat_pcie_driver); diff --git a/drivers/pci/controller/pcie-cadence.h b/drivers/pci/controller/pcie-cadence.h index ae6bf2a2b3d3..c98e85825776 100644 --- a/drivers/pci/controller/pcie-cadence.h +++ b/drivers/pci/controller/pcie-cadence.h @@ -190,6 +190,8 @@ enum cdns_pcie_rp_bar { (((code) << 8) & CDNS_PCIE_NORMAL_MSG_CODE_MASK) #define CDNS_PCIE_MSG_NO_DATA BIT(16) +struct cdns_pcie; + enum cdns_pcie_msg_code { MSG_CODE_ASSERT_INTA = 0x20, MSG_CODE_ASSERT_INTB = 0x21, @@ -231,13 +233,71 @@ enum cdns_pcie_msg_routing { struct cdns_pcie { void __iomem *reg_base; struct resource *mem_res; + struct device *dev; bool is_rc; u8 bus; int phy_count; struct phy **phy; struct device_link **link; + const struct cdns_pcie_common_ops *ops; }; +/** + * struct cdns_pcie_rc - private data for this PCIe Root Complex driver + * @pcie: Cadence PCIe controller + * @dev: pointer to PCIe device + * @cfg_res: start/end offsets in the physical system memory to map PCI + * configuration space accesses + * @bus_range: first/last buses behind the PCIe host controller + * @cfg_base: IO mapped window to access the PCI configuration space of a + * single function at a time + * @max_regions: maximum number of regions supported by the hardware + * @no_bar_nbits: Number of bits to keep for inbound (PCIe -> CPU) address + * translation (nbits sets into the "no BAR match" register) + * @vendor_id: PCI vendor ID + * @device_id: PCI device ID + */ +struct cdns_pcie_rc { + struct cdns_pcie pcie; + struct resource *cfg_res; + struct resource *bus_range; + void __iomem *cfg_base; + u32 max_regions; + u32 no_bar_nbits; + u16 vendor_id; + u16 device_id; +}; + +/** + * struct cdns_pcie_ep - private data for this PCIe endpoint controller driver + * @pcie: Cadence PCIe controller + * @max_regions: maximum number of regions supported by hardware + * @ob_region_map: bitmask of mapped outbound regions + * @ob_addr: base addresses in the AXI bus where the outbound regions start + * @irq_phys_addr: base address on the AXI bus where the MSI/legacy IRQ + * dedicated outbound regions is mapped. + * @irq_cpu_addr: base address in the CPU space where a write access triggers + * the sending of a memory write (MSI) / normal message (legacy + * IRQ) TLP through the PCIe bus. + * @irq_pci_addr: used to save the current mapping of the MSI/legacy IRQ + * dedicated outbound region. + * @irq_pci_fn: the latest PCI function that has updated the mapping of + * the MSI/legacy IRQ dedicated outbound region. + * @irq_pending: bitmask of asserted legacy IRQs. + */ +struct cdns_pcie_ep { + struct cdns_pcie pcie; + u32 max_regions; + unsigned long ob_region_map; + phys_addr_t *ob_addr; + phys_addr_t irq_phys_addr; + void __iomem *irq_cpu_addr; + u64 irq_pci_addr; + u8 irq_pci_fn; + u8 irq_pending; +}; + + /* Register access */ static inline void cdns_pcie_writeb(struct cdns_pcie *pcie, u32 reg, u8 value) { @@ -306,6 +366,23 @@ static inline u32 cdns_pcie_ep_fn_readl(struct cdns_pcie *pcie, u8 fn, u32 reg) return readl(pcie->reg_base + CDNS_PCIE_EP_FUNC_BASE(fn) + reg); } +#ifdef CONFIG_PCIE_CADENCE_HOST +int cdns_pcie_host_setup(struct cdns_pcie_rc *rc); +#else +static inline int cdns_pcie_host_setup(struct cdns_pcie_rc *rc) +{ + return 0; +} +#endif + +#ifdef CONFIG_PCIE_CADENCE_EP +int cdns_pcie_ep_setup(struct cdns_pcie_ep *ep); +#else +static inline int cdns_pcie_ep_setup(struct cdns_pcie_ep *ep) +{ + return 0; +} +#endif void cdns_pcie_set_outbound_region(struct cdns_pcie *pcie, u8 fn, u32 r, bool is_io, u64 cpu_addr, u64 pci_addr, size_t size); From de80f95ccb9c4de5a0fae0334de5ab438acf3453 Mon Sep 17 00:00:00 2001 From: Tom Joseph Date: Mon, 11 Nov 2019 12:30:44 +0000 Subject: [PATCH 081/160] PCI: cadence: Move all files to per-device cadence directory Cadence core library files may be used by various platform drivers. Add a new directory "cadence" to group all the Cadence core library files and the platforms using Cadence core library. Signed-off-by: Tom Joseph Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray --- drivers/pci/controller/Kconfig | 44 +----------------- drivers/pci/controller/Makefile | 5 +-- drivers/pci/controller/cadence/Kconfig | 45 +++++++++++++++++++ drivers/pci/controller/cadence/Makefile | 5 +++ .../{ => cadence}/pcie-cadence-ep.c | 0 .../{ => cadence}/pcie-cadence-host.c | 0 .../{ => cadence}/pcie-cadence-plat.c | 0 .../controller/{ => cadence}/pcie-cadence.c | 0 .../controller/{ => cadence}/pcie-cadence.h | 0 9 files changed, 52 insertions(+), 47 deletions(-) create mode 100644 drivers/pci/controller/cadence/Kconfig create mode 100644 drivers/pci/controller/cadence/Makefile rename drivers/pci/controller/{ => cadence}/pcie-cadence-ep.c (100%) rename drivers/pci/controller/{ => cadence}/pcie-cadence-host.c (100%) rename drivers/pci/controller/{ => cadence}/pcie-cadence-plat.c (100%) rename drivers/pci/controller/{ => cadence}/pcie-cadence.c (100%) rename drivers/pci/controller/{ => cadence}/pcie-cadence.h (100%) diff --git a/drivers/pci/controller/Kconfig b/drivers/pci/controller/Kconfig index b593616e9f52..5da00343bce7 100644 --- a/drivers/pci/controller/Kconfig +++ b/drivers/pci/controller/Kconfig @@ -22,49 +22,6 @@ config PCI_AARDVARK controller is part of the South Bridge of the Marvel Armada 3700 SoC. -menu "Cadence PCIe controllers support" - -config PCIE_CADENCE - bool - -config PCIE_CADENCE_HOST - bool - depends on OF - select IRQ_DOMAIN - select PCIE_CADENCE - -config PCIE_CADENCE_EP - bool - depends on OF - depends on PCI_ENDPOINT - select PCIE_CADENCE - -config PCIE_CADENCE_PLAT - bool - -config PCIE_CADENCE_PLAT_HOST - bool "Cadence PCIe platform host controller" - depends on OF - select PCIE_CADENCE_HOST - select PCIE_CADENCE_PLAT - help - Say Y here if you want to support the Cadence PCIe platform controller in - host mode. This PCIe controller may be embedded into many different - vendors SoCs. - -config PCIE_CADENCE_PLAT_EP - bool "Cadence PCIe platform endpoint controller" - depends on OF - depends on PCI_ENDPOINT - select PCIE_CADENCE_EP - select PCIE_CADENCE_PLAT - help - Say Y here if you want to support the Cadence PCIe platform controller in - endpoint mode. This PCIe controller may be embedded into many - different vendors SoCs. - -endmenu - config PCIE_XILINX_NWL bool "NWL PCIe Core" depends on ARCH_ZYNQMP || COMPILE_TEST @@ -304,4 +261,5 @@ config PCI_HYPERV_INTERFACE have a common interface with the Hyper-V PCI frontend driver. source "drivers/pci/controller/dwc/Kconfig" +source "drivers/pci/controller/cadence/Kconfig" endmenu diff --git a/drivers/pci/controller/Makefile b/drivers/pci/controller/Makefile index 42a6363d38f1..3d4f597f15ce 100644 --- a/drivers/pci/controller/Makefile +++ b/drivers/pci/controller/Makefile @@ -1,8 +1,5 @@ # SPDX-License-Identifier: GPL-2.0 -obj-$(CONFIG_PCIE_CADENCE) += pcie-cadence.o -obj-$(CONFIG_PCIE_CADENCE_HOST) += pcie-cadence-host.o -obj-$(CONFIG_PCIE_CADENCE_EP) += pcie-cadence-ep.o -obj-$(CONFIG_PCIE_CADENCE_PLAT) += pcie-cadence-plat.o +obj-$(CONFIG_PCIE_CADENCE) += cadence/ obj-$(CONFIG_PCI_FTPCI100) += pci-ftpci100.o obj-$(CONFIG_PCI_HYPERV) += pci-hyperv.o obj-$(CONFIG_PCI_HYPERV_INTERFACE) += pci-hyperv-intf.o diff --git a/drivers/pci/controller/cadence/Kconfig b/drivers/pci/controller/cadence/Kconfig new file mode 100644 index 000000000000..b76b3cf55ce5 --- /dev/null +++ b/drivers/pci/controller/cadence/Kconfig @@ -0,0 +1,45 @@ +# SPDX-License-Identifier: GPL-2.0 + +menu "Cadence PCIe controllers support" + depends on PCI + +config PCIE_CADENCE + bool + +config PCIE_CADENCE_HOST + bool + depends on OF + select IRQ_DOMAIN + select PCIE_CADENCE + +config PCIE_CADENCE_EP + bool + depends on OF + depends on PCI_ENDPOINT + select PCIE_CADENCE + +config PCIE_CADENCE_PLAT + bool + +config PCIE_CADENCE_PLAT_HOST + bool "Cadence PCIe platform host controller" + depends on OF + select PCIE_CADENCE_HOST + select PCIE_CADENCE_PLAT + help + Say Y here if you want to support the Cadence PCIe platform controller in + host mode. This PCIe controller may be embedded into many different + vendors SoCs. + +config PCIE_CADENCE_PLAT_EP + bool "Cadence PCIe platform endpoint controller" + depends on OF + depends on PCI_ENDPOINT + select PCIE_CADENCE_EP + select PCIE_CADENCE_PLAT + help + Say Y here if you want to support the Cadence PCIe platform controller in + endpoint mode. This PCIe controller may be embedded into many + different vendors SoCs. + +endmenu diff --git a/drivers/pci/controller/cadence/Makefile b/drivers/pci/controller/cadence/Makefile new file mode 100644 index 000000000000..232a3f20876a --- /dev/null +++ b/drivers/pci/controller/cadence/Makefile @@ -0,0 +1,5 @@ +# SPDX-License-Identifier: GPL-2.0 +obj-$(CONFIG_PCIE_CADENCE) += pcie-cadence.o +obj-$(CONFIG_PCIE_CADENCE_HOST) += pcie-cadence-host.o +obj-$(CONFIG_PCIE_CADENCE_EP) += pcie-cadence-ep.o +obj-$(CONFIG_PCIE_CADENCE_PLAT) += pcie-cadence-plat.o diff --git a/drivers/pci/controller/pcie-cadence-ep.c b/drivers/pci/controller/cadence/pcie-cadence-ep.c similarity index 100% rename from drivers/pci/controller/pcie-cadence-ep.c rename to drivers/pci/controller/cadence/pcie-cadence-ep.c diff --git a/drivers/pci/controller/pcie-cadence-host.c b/drivers/pci/controller/cadence/pcie-cadence-host.c similarity index 100% rename from drivers/pci/controller/pcie-cadence-host.c rename to drivers/pci/controller/cadence/pcie-cadence-host.c diff --git a/drivers/pci/controller/pcie-cadence-plat.c b/drivers/pci/controller/cadence/pcie-cadence-plat.c similarity index 100% rename from drivers/pci/controller/pcie-cadence-plat.c rename to drivers/pci/controller/cadence/pcie-cadence-plat.c diff --git a/drivers/pci/controller/pcie-cadence.c b/drivers/pci/controller/cadence/pcie-cadence.c similarity index 100% rename from drivers/pci/controller/pcie-cadence.c rename to drivers/pci/controller/cadence/pcie-cadence.c diff --git a/drivers/pci/controller/pcie-cadence.h b/drivers/pci/controller/cadence/pcie-cadence.h similarity index 100% rename from drivers/pci/controller/pcie-cadence.h rename to drivers/pci/controller/cadence/pcie-cadence.h From f338bb9f0179cb959977b74e8331b312264d720b Mon Sep 17 00:00:00 2001 From: George Cherian Date: Mon, 11 Nov 2019 02:43:03 +0000 Subject: [PATCH 082/160] PCI: Apply Cavium ACS quirk to ThunderX2 and ThunderX3 Enhance the ACS quirk for Cavium Processors. Add the root port vendor IDs for ThunderX2 and ThunderX3 series of processors. [bhelgaas: add Fixes: and stable tag] Fixes: f2ddaf8dfd4a ("PCI: Apply Cavium ThunderX ACS quirk to more Root Ports") Link: https://lore.kernel.org/r/20191111024243.GA11408@dc5-eodlnx05.marvell.com Signed-off-by: George Cherian Signed-off-by: Bjorn Helgaas Reviewed-by: Robert Richter Cc: stable@vger.kernel.org # v4.12+ --- drivers/pci/quirks.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index d5d57cd91a5e..2544e210b984 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -4347,15 +4347,21 @@ static int pci_quirk_amd_sb_acs(struct pci_dev *dev, u16 acs_flags) static bool pci_quirk_cavium_acs_match(struct pci_dev *dev) { + if (!pci_is_pcie(dev) || pci_pcie_type(dev) != PCI_EXP_TYPE_ROOT_PORT) + return false; + + switch (dev->device) { /* - * Effectively selects all downstream ports for whole ThunderX 1 - * family by 0xf800 mask (which represents 8 SoCs), while the lower - * bits of device ID are used to indicate which subdevice is used - * within the SoC. + * Effectively selects all downstream ports for whole ThunderX1 + * (which represents 8 SoCs). */ - return (pci_is_pcie(dev) && - (pci_pcie_type(dev) == PCI_EXP_TYPE_ROOT_PORT) && - ((dev->device & 0xf800) == 0xa000)); + case 0xa000 ... 0xa7ff: /* ThunderX1 */ + case 0xaf84: /* ThunderX2 */ + case 0xb884: /* ThunderX3 */ + return true; + default: + return false; + } } static int pci_quirk_cavium_acs(struct pci_dev *dev, u16 acs_flags) From b94ec12dfaee41bdd6a8b6d75e5715d9ba2c92cc Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 8 Nov 2019 13:18:55 +0200 Subject: [PATCH 083/160] PCI: pciehp: Refactor infinite loop in pcie_poll_cmd() Infinite timeout loops are hard to read. Refactor it to plausible 'do {} while ()'. Note, the supplied timeout can't be negative for current use, though if it's not dividable to 10, we may go below 0, that's why type of the parameter is int. And thus, we may move the check to the loop condition. No functional change intended. Link: https://lore.kernel.org/r/20191108111855.85866-1-andriy.shevchenko@linux.intel.com Signed-off-by: Andy Shevchenko Signed-off-by: Bjorn Helgaas Reviewed-by: Andrew Murray --- drivers/pci/hotplug/pciehp_hpc.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 86d97f3112f0..764384153c7d 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -68,7 +68,7 @@ static int pcie_poll_cmd(struct controller *ctrl, int timeout) struct pci_dev *pdev = ctrl_dev(ctrl); u16 slot_status; - while (true) { + do { pcie_capability_read_word(pdev, PCI_EXP_SLTSTA, &slot_status); if (slot_status == (u16) ~0) { ctrl_info(ctrl, "%s: no response from device\n", @@ -81,11 +81,9 @@ static int pcie_poll_cmd(struct controller *ctrl, int timeout) PCI_EXP_SLTSTA_CC); return 1; } - if (timeout < 0) - break; msleep(10); timeout -= 10; - } + } while (timeout >= 0); return 0; /* timeout */ } From 7c7e53e1c93df14690bd12c1f84730fef927a6f1 Mon Sep 17 00:00:00 2001 From: Yoshihiro Shimoda Date: Tue, 5 Nov 2019 19:51:29 +0900 Subject: [PATCH 084/160] PCI: rcar: Fix missing MACCTLR register setting in initialization sequence The R-Car Gen2/3 manual - available at: https://www.renesas.com/eu/en/products/microcontrollers-microprocessors/rz/rzg/rzg1m.html#documents "RZ/G Series User's Manual: Hardware" section strictly enforces the MACCTLR inizialization value - 39.3.1 - "Initial Setting of PCI Express": "Be sure to write the initial value (= H'80FF 0000) to MACCTLR before enabling PCIETCTLR.CFINIT". To avoid unexpected behavior and to match the SW initialization sequence guidelines, this patch programs the MACCTLR with the correct value. Note that the MACCTLR.SPCHG bit in the MACCTLR register description reports that "Only writing 1 is valid and writing 0 is invalid" but this "invalid" has to be interpreted as a write-ignore aka "ignored", not "prohibited". Reported-by: Eugeniu Rosca Fixes: c25da4778803 ("PCI: rcar: Add Renesas R-Car PCIe driver") Fixes: be20bbcb0a8c ("PCI: rcar: Add the initialization of PCIe link in resume_noirq()") Signed-off-by: Yoshihiro Shimoda Signed-off-by: Lorenzo Pieralisi Reviewed-by: Geert Uytterhoeven Cc: # v5.2+ --- drivers/pci/controller/pcie-rcar.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/pci/controller/pcie-rcar.c b/drivers/pci/controller/pcie-rcar.c index 40d8c54a17d1..94ba4fe21923 100644 --- a/drivers/pci/controller/pcie-rcar.c +++ b/drivers/pci/controller/pcie-rcar.c @@ -91,8 +91,11 @@ #define LINK_SPEED_2_5GTS (1 << 16) #define LINK_SPEED_5_0GTS (2 << 16) #define MACCTLR 0x011058 +#define MACCTLR_NFTS_MASK GENMASK(23, 16) /* The name is from SH7786 */ #define SPEED_CHANGE BIT(24) #define SCRAMBLE_DISABLE BIT(27) +#define LTSMDIS BIT(31) +#define MACCTLR_INIT_VAL (LTSMDIS | MACCTLR_NFTS_MASK) #define PMSR 0x01105c #define MACS2R 0x011078 #define MACCGSPSETR 0x011084 @@ -613,6 +616,8 @@ static int rcar_pcie_hw_init(struct rcar_pcie *pcie) if (IS_ENABLED(CONFIG_PCI_MSI)) rcar_pci_write_reg(pcie, 0x801f0000, PCIEMSITXR); + rcar_pci_write_reg(pcie, MACCTLR_INIT_VAL, MACCTLR); + /* Finish initialization - establish a PCI Express link */ rcar_pci_write_reg(pcie, CFINIT, PCIETCTLR); @@ -1235,6 +1240,7 @@ static int rcar_pcie_resume_noirq(struct device *dev) return 0; /* Re-establish the PCIe link */ + rcar_pci_write_reg(pcie, MACCTLR_INIT_VAL, MACCTLR); rcar_pci_write_reg(pcie, CFINIT, PCIETCTLR); return rcar_pcie_wait_for_dl(pcie); } From 75fcc0ce72e5cea2e357cdde858216c5bad40442 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Tue, 29 Oct 2019 20:00:21 +0300 Subject: [PATCH 085/160] PCI: pciehp: Do not disable interrupt twice on suspend We try to keep PCIe hotplug ports runtime suspended when entering system suspend. Because the PCIe portdrv sets the DPM_FLAG_NEVER_SKIP flag, the PM core always calls system suspend/resume hooks even if the device is left runtime suspended. Since PCIe hotplug driver re-used the same function for both runtime suspend and system suspend, it ended up disabling hotplug interrupt twice and the second time following was printed: pciehp 0000:03:01.0:pcie204: pcie_do_write_cmd: no response from device Prevent this from happening by checking whether the device is already runtime suspended when the system suspend hook is called. Fixes: 9c62f0bfb832 ("PCI: pciehp: Implement runtime PM callbacks") Link: https://lore.kernel.org/r/20191029170022.57528-1-mika.westerberg@linux.intel.com Reported-by: Kai-Heng Feng Tested-by: Kai-Heng Feng Signed-off-by: Mika Westerberg Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/hotplug/pciehp_core.c | 25 +++++++++++++++++++++++-- 1 file changed, 23 insertions(+), 2 deletions(-) diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index b3122c151b80..56daad828c9e 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -253,7 +253,7 @@ static bool pme_is_native(struct pcie_device *dev) return pcie_ports_native || host->native_pme; } -static int pciehp_suspend(struct pcie_device *dev) +static void pciehp_disable_interrupt(struct pcie_device *dev) { /* * Disable hotplug interrupt so that it does not trigger @@ -261,7 +261,19 @@ static int pciehp_suspend(struct pcie_device *dev) */ if (pme_is_native(dev)) pcie_disable_interrupt(get_service_data(dev)); +} +#ifdef CONFIG_PM_SLEEP +static int pciehp_suspend(struct pcie_device *dev) +{ + /* + * If the port is already runtime suspended we can keep it that + * way. + */ + if (dev_pm_smart_suspend_and_suspended(&dev->port->dev)) + return 0; + + pciehp_disable_interrupt(dev); return 0; } @@ -279,6 +291,7 @@ static int pciehp_resume_noirq(struct pcie_device *dev) return 0; } +#endif static int pciehp_resume(struct pcie_device *dev) { @@ -292,6 +305,12 @@ static int pciehp_resume(struct pcie_device *dev) return 0; } +static int pciehp_runtime_suspend(struct pcie_device *dev) +{ + pciehp_disable_interrupt(dev); + return 0; +} + static int pciehp_runtime_resume(struct pcie_device *dev) { struct controller *ctrl = get_service_data(dev); @@ -318,10 +337,12 @@ static struct pcie_port_service_driver hpdriver_portdrv = { .remove = pciehp_remove, #ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP .suspend = pciehp_suspend, .resume_noirq = pciehp_resume_noirq, .resume = pciehp_resume, - .runtime_suspend = pciehp_suspend, +#endif + .runtime_suspend = pciehp_runtime_suspend, .runtime_resume = pciehp_runtime_resume, #endif /* PM */ }; From 87d0f2a5536fdf5053a6d341880f96135549a644 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Tue, 29 Oct 2019 20:00:22 +0300 Subject: [PATCH 086/160] PCI: pciehp: Prevent deadlock on disconnect This addresses deadlocks in these common cases in hierarchies containing two switches: - All involved ports are runtime suspended and they are unplugged. This can happen easily if the drivers involved automatically enable runtime PM (xHCI for example does that). - System is suspended (e.g., closing the lid on a laptop) with a dock + something else connected, and the dock is unplugged while suspended. These cases lead to the following deadlock: INFO: task irq/126-pciehp:198 blocked for more than 120 seconds. irq/126-pciehp D 0 198 2 0x80000000 Call Trace: schedule+0x2c/0x80 schedule_timeout+0x246/0x350 wait_for_completion+0xb7/0x140 kthread_stop+0x49/0x110 free_irq+0x32/0x70 pcie_shutdown_notification+0x2f/0x50 pciehp_remove+0x27/0x50 pcie_port_remove_service+0x36/0x50 device_release_driver+0x12/0x20 bus_remove_device+0xec/0x160 device_del+0x13b/0x350 device_unregister+0x1a/0x60 remove_iter+0x1e/0x30 device_for_each_child+0x56/0x90 pcie_port_device_remove+0x22/0x40 pcie_portdrv_remove+0x20/0x60 pci_device_remove+0x3e/0xc0 device_release_driver_internal+0x18c/0x250 device_release_driver+0x12/0x20 pci_stop_bus_device+0x6f/0x90 pci_stop_bus_device+0x31/0x90 pci_stop_and_remove_bus_device+0x12/0x20 pciehp_unconfigure_device+0x88/0x140 pciehp_disable_slot+0x6a/0x110 pciehp_handle_presence_or_link_change+0x263/0x400 pciehp_ist+0x1c9/0x1d0 irq_thread_fn+0x24/0x60 irq_thread+0xeb/0x190 kthread+0x120/0x140 INFO: task irq/190-pciehp:2288 blocked for more than 120 seconds. irq/190-pciehp D 0 2288 2 0x80000000 Call Trace: __schedule+0x2a2/0x880 schedule+0x2c/0x80 schedule_preempt_disabled+0xe/0x10 mutex_lock+0x2c/0x30 pci_lock_rescan_remove+0x15/0x20 pciehp_unconfigure_device+0x4d/0x140 pciehp_disable_slot+0x6a/0x110 pciehp_handle_presence_or_link_change+0x263/0x400 pciehp_ist+0x1c9/0x1d0 irq_thread_fn+0x24/0x60 irq_thread+0xeb/0x190 kthread+0x120/0x140 What happens here is that the whole hierarchy is runtime resumed and the parent PCIe downstream port, which got the hot-remove event, starts removing devices below it, taking pci_lock_rescan_remove() lock. When the child PCIe port is runtime resumed it calls pciehp_check_presence() which ends up calling pciehp_card_present() and pciehp_check_link_active(). Both of these use pcie_capability_read_word(), which notices that the underlying device is already gone and returns PCIBIOS_DEVICE_NOT_FOUND with the capability value set to 0. When pciehp gets this value it thinks that its child device is also hot-removed and schedules its IRQ thread to handle the event. The deadlock happens when the child's IRQ thread runs and tries to acquire pci_lock_rescan_remove() which is already taken by the parent and the parent waits for the child's IRQ thread to finish. Prevent this from happening by checking the return value of pcie_capability_read_word() and if it is PCIBIOS_DEVICE_NOT_FOUND stop performing any hot-removal activities. [bhelgaas: add common scenarios to commit log] Link: https://lore.kernel.org/r/20191029170022.57528-2-mika.westerberg@linux.intel.com Tested-by: Kai-Heng Feng Signed-off-by: Mika Westerberg Signed-off-by: Bjorn Helgaas --- drivers/pci/hotplug/pciehp.h | 6 ++-- drivers/pci/hotplug/pciehp_core.c | 11 ++++-- drivers/pci/hotplug/pciehp_ctrl.c | 4 +-- drivers/pci/hotplug/pciehp_hpc.c | 59 +++++++++++++++++++++++++------ 4 files changed, 61 insertions(+), 19 deletions(-) diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index 882ce82c4699..aa61d4c219d7 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -174,10 +174,10 @@ void pciehp_set_indicators(struct controller *ctrl, int pwr, int attn); void pciehp_get_latch_status(struct controller *ctrl, u8 *status); int pciehp_query_power_fault(struct controller *ctrl); -bool pciehp_card_present(struct controller *ctrl); -bool pciehp_card_present_or_link_active(struct controller *ctrl); +int pciehp_card_present(struct controller *ctrl); +int pciehp_card_present_or_link_active(struct controller *ctrl); int pciehp_check_link_status(struct controller *ctrl); -bool pciehp_check_link_active(struct controller *ctrl); +int pciehp_check_link_active(struct controller *ctrl); void pciehp_release_ctrl(struct controller *ctrl); int pciehp_sysfs_enable_slot(struct hotplug_slot *hotplug_slot); diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index 56daad828c9e..312cc45c44c7 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -139,10 +139,15 @@ static int get_adapter_status(struct hotplug_slot *hotplug_slot, u8 *value) { struct controller *ctrl = to_ctrl(hotplug_slot); struct pci_dev *pdev = ctrl->pcie->port; + int ret; pci_config_pm_runtime_get(pdev); - *value = pciehp_card_present_or_link_active(ctrl); + ret = pciehp_card_present_or_link_active(ctrl); pci_config_pm_runtime_put(pdev); + if (ret < 0) + return ret; + + *value = ret; return 0; } @@ -158,13 +163,13 @@ static int get_adapter_status(struct hotplug_slot *hotplug_slot, u8 *value) */ static void pciehp_check_presence(struct controller *ctrl) { - bool occupied; + int occupied; down_read(&ctrl->reset_lock); mutex_lock(&ctrl->state_lock); occupied = pciehp_card_present_or_link_active(ctrl); - if ((occupied && (ctrl->state == OFF_STATE || + if ((occupied > 0 && (ctrl->state == OFF_STATE || ctrl->state == BLINKINGON_STATE)) || (!occupied && (ctrl->state == ON_STATE || ctrl->state == BLINKINGOFF_STATE))) diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index dd8e4a5fb282..6503d15effbb 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -226,7 +226,7 @@ void pciehp_handle_disable_request(struct controller *ctrl) void pciehp_handle_presence_or_link_change(struct controller *ctrl, u32 events) { - bool present, link_active; + int present, link_active; /* * If the slot is on and presence or link has changed, turn it off. @@ -257,7 +257,7 @@ void pciehp_handle_presence_or_link_change(struct controller *ctrl, u32 events) mutex_lock(&ctrl->state_lock); present = pciehp_card_present(ctrl); link_active = pciehp_check_link_active(ctrl); - if (!present && !link_active) { + if (present <= 0 && link_active <= 0) { mutex_unlock(&ctrl->state_lock); return; } diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 764384153c7d..8a2cb1764386 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -199,17 +199,29 @@ static void pcie_write_cmd_nowait(struct controller *ctrl, u16 cmd, u16 mask) pcie_do_write_cmd(ctrl, cmd, mask, false); } -bool pciehp_check_link_active(struct controller *ctrl) +/** + * pciehp_check_link_active() - Is the link active + * @ctrl: PCIe hotplug controller + * + * Check whether the downstream link is currently active. Note it is + * possible that the card is removed immediately after this so the + * caller may need to take it into account. + * + * If the hotplug controller itself is not available anymore returns + * %-ENODEV. + */ +int pciehp_check_link_active(struct controller *ctrl) { struct pci_dev *pdev = ctrl_dev(ctrl); u16 lnk_status; - bool ret; + int ret; + + ret = pcie_capability_read_word(pdev, PCI_EXP_LNKSTA, &lnk_status); + if (ret == PCIBIOS_DEVICE_NOT_FOUND || lnk_status == (u16)~0) + return -ENODEV; - pcie_capability_read_word(pdev, PCI_EXP_LNKSTA, &lnk_status); ret = !!(lnk_status & PCI_EXP_LNKSTA_DLLLA); - - if (ret) - ctrl_dbg(ctrl, "%s: lnk_status = %x\n", __func__, lnk_status); + ctrl_dbg(ctrl, "%s: lnk_status = %x\n", __func__, lnk_status); return ret; } @@ -371,13 +383,29 @@ void pciehp_get_latch_status(struct controller *ctrl, u8 *status) *status = !!(slot_status & PCI_EXP_SLTSTA_MRLSS); } -bool pciehp_card_present(struct controller *ctrl) +/** + * pciehp_card_present() - Is the card present + * @ctrl: PCIe hotplug controller + * + * Function checks whether the card is currently present in the slot and + * in that case returns true. Note it is possible that the card is + * removed immediately after the check so the caller may need to take + * this into account. + * + * It the hotplug controller itself is not available anymore returns + * %-ENODEV. + */ +int pciehp_card_present(struct controller *ctrl) { struct pci_dev *pdev = ctrl_dev(ctrl); u16 slot_status; + int ret; - pcie_capability_read_word(pdev, PCI_EXP_SLTSTA, &slot_status); - return slot_status & PCI_EXP_SLTSTA_PDS; + ret = pcie_capability_read_word(pdev, PCI_EXP_SLTSTA, &slot_status); + if (ret == PCIBIOS_DEVICE_NOT_FOUND || slot_status == (u16)~0) + return -ENODEV; + + return !!(slot_status & PCI_EXP_SLTSTA_PDS); } /** @@ -388,10 +416,19 @@ bool pciehp_card_present(struct controller *ctrl) * Presence Detect State bit, this helper also returns true if the Link Active * bit is set. This is a concession to broken hotplug ports which hardwire * Presence Detect State to zero, such as Wilocity's [1ae9:0200]. + * + * Returns: %1 if the slot is occupied and %0 if it is not. If the hotplug + * port is not present anymore returns %-ENODEV. */ -bool pciehp_card_present_or_link_active(struct controller *ctrl) +int pciehp_card_present_or_link_active(struct controller *ctrl) { - return pciehp_card_present(ctrl) || pciehp_check_link_active(ctrl); + int ret; + + ret = pciehp_card_present(ctrl); + if (ret) + return ret; + + return pciehp_check_link_active(ctrl); } int pciehp_query_power_fault(struct controller *ctrl) From 77adf9355304f8dcf09054280af5e23fc451ab3d Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Wed, 30 Oct 2019 18:05:45 +0300 Subject: [PATCH 087/160] ACPI / hotplug / PCI: Allocate resources directly under the non-hotplug bridge Valerio and others reported that commit 84c8b58ed3ad ("ACPI / hotplug / PCI: Don't scan bridges managed by native hotplug") prevents some recent LG and HP laptops from booting with endless loop of: ACPI Error: No handler or method for GPE 08, disabling event (20190215/evgpe-835) ACPI Error: No handler or method for GPE 09, disabling event (20190215/evgpe-835) ACPI Error: No handler or method for GPE 0A, disabling event (20190215/evgpe-835) ... What seems to happen is that during boot, after the initial PCI enumeration when EC is enabled the platform triggers ACPI Notify() to one of the root ports. The root port itself looks like this: pci 0000:00:1b.0: PCI bridge to [bus 02-3a] pci 0000:00:1b.0: bridge window [mem 0xc4000000-0xda0fffff] pci 0000:00:1b.0: bridge window [mem 0x80000000-0xa1ffffff 64bit pref] The BIOS has configured the root port so that it does not have I/O bridge window. Now when the ACPI Notify() is triggered ACPI hotplug handler calls acpiphp_native_scan_bridge() for each non-hotplug bridge (as this system is using native PCIe hotplug) and pci_assign_unassigned_bridge_resources() to allocate resources. The device connected to the root port is a PCIe switch (Thunderbolt controller) with two hotplug downstream ports. Because of the hotplug ports __pci_bus_size_bridges() tries to add "additional I/O" of 256 bytes to each (DEFAULT_HOTPLUG_IO_SIZE). This gets further aligned to 4k as that's the minimum I/O window size so each hotplug port gets 4k I/O window and the same happens for the root port (which is also hotplug port). This means 3 * 4k = 12k I/O window. Because of this pci_assign_unassigned_bridge_resources() ends up opening a I/O bridge window for the root port at first available I/O address which seems to be in range 0x1000 - 0x3fff. Normally this range is used for ACPI stuff such as GPE bits (below is part of /proc/ioports): 1800-1803 : ACPI PM1a_EVT_BLK 1804-1805 : ACPI PM1a_CNT_BLK 1808-180b : ACPI PM_TMR 1810-1815 : ACPI CPU throttle 1850-1850 : ACPI PM2_CNT_BLK 1854-1857 : pnp 00:05 1860-187f : ACPI GPE0_BLK However, when the ACPI Notify() happened this range was not yet reserved for ACPI/PNP (that happens later) so PCI gets it. It then starts writing to this range and accidentally stomps over GPE bits among other things causing the endless stream of messages about missing GPE handler. This problem does not happen if "pci=hpiosize=0" is passed in the kernel command line. The reason is that then the kernel does not try to allocate the additional 256 bytes for each hotplug port. Fix this by allocating resources directly below the non-hotplug bridges where a new device may appear as a result of ACPI Notify(). This avoids the hotplug bridges and prevents opening the additional I/O window. Fixes: 84c8b58ed3ad ("ACPI / hotplug / PCI: Don't scan bridges managed by native hotplug") Link: https://bugzilla.kernel.org/show_bug.cgi?id=203617 Link: https://lore.kernel.org/r/20191030150545.19885-1-mika.westerberg@linux.intel.com Reported-by: Valerio Passini Signed-off-by: Mika Westerberg Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki Cc: stable@vger.kernel.org --- drivers/pci/hotplug/acpiphp_glue.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index e4c46637f32f..b3869951c0eb 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -449,8 +449,15 @@ static void acpiphp_native_scan_bridge(struct pci_dev *bridge) /* Scan non-hotplug bridges that need to be reconfigured */ for_each_pci_bridge(dev, bus) { - if (!hotplug_is_native(dev)) - max = pci_scan_bridge(bus, dev, max, 1); + if (hotplug_is_native(dev)) + continue; + + max = pci_scan_bridge(bus, dev, max, 1); + if (dev->subordinate) { + pcibios_resource_survey_bus(dev->subordinate); + pci_bus_size_bridges(dev->subordinate); + pci_bus_assign_resources(dev->subordinate); + } } } @@ -480,7 +487,6 @@ static void enable_slot(struct acpiphp_slot *slot, bool bridge) if (PCI_SLOT(dev->devfn) == slot->device) acpiphp_native_scan_bridge(dev); } - pci_assign_unassigned_bridge_resources(bus->self); } else { LIST_HEAD(add_list); int max, pass; From c13704f5685deb7d6eb21e293233e0901ed77377 Mon Sep 17 00:00:00 2001 From: Nicholas Johnson Date: Wed, 13 Nov 2019 15:25:28 +0000 Subject: [PATCH 088/160] PCI: Avoid double hpmemsize MMIO window assignment Previously, the kernel sometimes assigned more MMIO or MMIO_PREF space than desired. For example, if the user requested 128M of space with "pci=realloc,hpmemsize=128M", we sometimes assigned 256M: pci 0000:06:01.0: BAR 14: assigned [mem 0x90100000-0xa00fffff] = 256M pci 0000:06:04.0: BAR 14: assigned [mem 0xa0200000-0xb01fffff] = 256M With this patch applied: pci 0000:06:01.0: BAR 14: assigned [mem 0x90100000-0x980fffff] = 128M pci 0000:06:04.0: BAR 14: assigned [mem 0x98200000-0xa01fffff] = 128M This happened when in the first pass, the MMIO_PREF succeeded but the MMIO failed. In the next pass, because MMIO_PREF was already assigned, the attempt to assign MMIO_PREF returned an error code instead of success (nothing more to do, already allocated). Hence, the size which was actually allocated, but thought to have failed, was placed in the MMIO window. The bug resulted in the MMIO_PREF being added to the MMIO window, which meant doubling if MMIO_PREF size = MMIO size. With a large MMIO_PREF, the MMIO window would likely fail to be assigned altogether due to lack of 32-bit address space. Change find_free_bus_resource() to do the following: - Return first unassigned resource of the correct type. - If there is none, return first assigned resource of the correct type. - If none of the above, return NULL. Returning an assigned resource of the correct type allows the caller to distinguish between already assigned and no resource of the correct type. Add checks in pbus_size_io() and pbus_size_mem() to return success if resource returned from find_free_bus_resource() is already allocated. This avoids pbus_size_io() and pbus_size_mem() returning error code to __pci_bus_size_bridges() when a resource has been successfully assigned in a previous pass. This fixes the existing behaviour where space for a resource could be reserved multiple times in different parent bridge windows. Link: https://lore.kernel.org/lkml/20190531171216.20532-2-logang@deltatee.com/T/#u Link: https://bugzilla.kernel.org/show_bug.cgi?id=203243 Link: https://lore.kernel.org/r/PS2P216MB075563AA6AD242AA666EDC6A80760@PS2P216MB0755.KORP216.PROD.OUTLOOK.COM Reported-by: Kit Chow Reported-by: Nicholas Johnson Signed-off-by: Nicholas Johnson Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg Reviewed-by: Logan Gunthorpe --- drivers/pci/setup-bus.c | 38 +++++++++++++++++++++++++++----------- 1 file changed, 27 insertions(+), 11 deletions(-) diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c index b5f9f57f436a..f279826204eb 100644 --- a/drivers/pci/setup-bus.c +++ b/drivers/pci/setup-bus.c @@ -752,24 +752,32 @@ static void pci_bridge_check_ranges(struct pci_bus *bus) } /* - * Helper function for sizing routines: find first available bus resource - * of a given type. Note: we intentionally skip the bus resources which - * have already been assigned (that is, have non-NULL parent resource). + * Helper function for sizing routines. Assigned resources have non-NULL + * parent resource. + * + * Return first unassigned resource of the correct type. If there is none, + * return first assigned resource of the correct type. If none of the + * above, return NULL. + * + * Returning an assigned resource of the correct type allows the caller to + * distinguish between already assigned and no resource of the correct type. */ -static struct resource *find_free_bus_resource(struct pci_bus *bus, - unsigned long type_mask, - unsigned long type) +static struct resource *find_bus_resource_of_type(struct pci_bus *bus, + unsigned long type_mask, + unsigned long type) { + struct resource *r, *r_assigned = NULL; int i; - struct resource *r; pci_bus_for_each_resource(bus, r, i) { if (r == &ioport_resource || r == &iomem_resource) continue; if (r && (r->flags & type_mask) == type && !r->parent) return r; + if (r && (r->flags & type_mask) == type && !r_assigned) + r_assigned = r; } - return NULL; + return r_assigned; } static resource_size_t calculate_iosize(resource_size_t size, @@ -866,8 +874,8 @@ static void pbus_size_io(struct pci_bus *bus, resource_size_t min_size, struct list_head *realloc_head) { struct pci_dev *dev; - struct resource *b_res = find_free_bus_resource(bus, IORESOURCE_IO, - IORESOURCE_IO); + struct resource *b_res = find_bus_resource_of_type(bus, IORESOURCE_IO, + IORESOURCE_IO); resource_size_t size = 0, size0 = 0, size1 = 0; resource_size_t children_add_size = 0; resource_size_t min_align, align; @@ -875,6 +883,10 @@ static void pbus_size_io(struct pci_bus *bus, resource_size_t min_size, if (!b_res) return; + /* If resource is already assigned, nothing more to do */ + if (b_res->parent) + return; + min_align = window_alignment(bus, IORESOURCE_IO); list_for_each_entry(dev, &bus->devices, bus_list) { int i; @@ -978,7 +990,7 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, resource_size_t min_align, align, size, size0, size1; resource_size_t aligns[18]; /* Alignments from 1MB to 128GB */ int order, max_order; - struct resource *b_res = find_free_bus_resource(bus, + struct resource *b_res = find_bus_resource_of_type(bus, mask | IORESOURCE_PREFETCH, type); resource_size_t children_add_size = 0; resource_size_t children_add_align = 0; @@ -987,6 +999,10 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, if (!b_res) return -ENOSPC; + /* If resource is already assigned, nothing more to do */ + if (b_res->parent) + return 0; + memset(aligns, 0, sizeof(aligns)); max_order = 0; size = 0; From 73884a7082f466ce6686bb8dd7e6571dd42313b4 Mon Sep 17 00:00:00 2001 From: Subbaraya Sundeep Date: Mon, 4 Nov 2019 12:27:44 +0530 Subject: [PATCH 089/160] PCI: Do not use bus number zero from EA capability As per PCIe r5.0, sec 7.8.5.2, fixed bus numbers of a bridge must be zero when no function that uses EA is located behind it. Hence, if EA supplies bus numbers of zero, assign bus numbers normally. A secondary bus can never have a bus number of zero, so setting a bridge's Secondary Bus Number to zero makes downstream devices unreachable. [bhelgaas: retain bool return value so "zero is invalid" logic is local] Fixes: 2dbce5901179 ("PCI: Assign bus numbers present in EA capability for bridges") Link: https://lore.kernel.org/r/1572850664-9861-1-git-send-email-sundeep.lkml@gmail.com Signed-off-by: Subbaraya Sundeep Signed-off-by: Bjorn Helgaas Cc: stable@vger.kernel.org # v5.2+ --- drivers/pci/probe.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index bdbc8490f962..d3033873395d 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -1090,14 +1090,15 @@ static unsigned int pci_scan_child_bus_extend(struct pci_bus *bus, * @sec: updated with secondary bus number from EA * @sub: updated with subordinate bus number from EA * - * If @dev is a bridge with EA capability, update @sec and @sub with - * fixed bus numbers from the capability and return true. Otherwise, - * return false. + * If @dev is a bridge with EA capability that specifies valid secondary + * and subordinate bus numbers, return true with the bus numbers in @sec + * and @sub. Otherwise return false. */ static bool pci_ea_fixed_busnrs(struct pci_dev *dev, u8 *sec, u8 *sub) { int ea, offset; u32 dw; + u8 ea_sec, ea_sub; if (dev->hdr_type != PCI_HEADER_TYPE_BRIDGE) return false; @@ -1109,8 +1110,13 @@ static bool pci_ea_fixed_busnrs(struct pci_dev *dev, u8 *sec, u8 *sub) offset = ea + PCI_EA_FIRST_ENT; pci_read_config_dword(dev, offset, &dw); - *sec = dw & PCI_EA_SEC_BUS_MASK; - *sub = (dw & PCI_EA_SUB_BUS_MASK) >> PCI_EA_SUB_BUS_SHIFT; + ea_sec = dw & PCI_EA_SEC_BUS_MASK; + ea_sub = (dw & PCI_EA_SUB_BUS_MASK) >> PCI_EA_SUB_BUS_SHIFT; + if (ea_sec == 0 || ea_sub < ea_sec) + return false; + + *sec = ea_sec; + *sub = ea_sub; return true; } From c8de8ed2dcaac82e5d76d467dc0b02e0ee79809b Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 5 Sep 2019 17:54:42 -0500 Subject: [PATCH 090/160] PCI: Make ACS quirk implementations more uniform The ACS quirks differ in needless ways, which makes them look more different than they really are. Reorder the ACS flags in order of definitions in the spec: PCI_ACS_SV Source Validation PCI_ACS_TB Translation Blocking PCI_ACS_RR P2P Request Redirect PCI_ACS_CR P2P Completion Redirect PCI_ACS_UF Upstream Forwarding PCI_ACS_EC P2P Egress Control PCI_ACS_DT Direct Translated P2P (PCIe r5.0, sec 7.7.8.2) and use similar code structure in all. No functional change intended. Signed-off-by: Bjorn Helgaas Reviewed-by: Logan Gunthorpe Reviewed-by: Alex Williamson --- drivers/pci/quirks.c | 41 +++++++++++++++++++---------------------- 1 file changed, 19 insertions(+), 22 deletions(-) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 2544e210b984..59f73d084e1d 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -4366,18 +4366,18 @@ static bool pci_quirk_cavium_acs_match(struct pci_dev *dev) static int pci_quirk_cavium_acs(struct pci_dev *dev, u16 acs_flags) { + if (!pci_quirk_cavium_acs_match(dev)) + return -ENOTTY; + /* - * Cavium root ports don't advertise an ACS capability. However, + * Cavium Root Ports don't advertise an ACS capability. However, * the RTL internally implements similar protection as if ACS had - * Request Redirection, Completion Redirection, Source Validation, + * Source Validation, Request Redirection, Completion Redirection, * and Upstream Forwarding features enabled. Assert that the * hardware implements and enables equivalent ACS functionality for * these flags. */ - acs_flags &= ~(PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_SV | PCI_ACS_UF); - - if (!pci_quirk_cavium_acs_match(dev)) - return -ENOTTY; + acs_flags &= ~(PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); return acs_flags ? 0 : 1; } @@ -4395,7 +4395,7 @@ static int pci_quirk_xgene_acs(struct pci_dev *dev, u16 acs_flags) } /* - * Many Intel PCH root ports do provide ACS-like features to disable peer + * Many Intel PCH Root Ports do provide ACS-like features to disable peer * transactions and validate bus numbers in requests, but do not provide an * actual PCIe ACS capability. This is the list of device IDs known to fall * into that category as provided by Intel in Red Hat bugzilla 1037684. @@ -4443,37 +4443,34 @@ static bool pci_quirk_intel_pch_acs_match(struct pci_dev *dev) return false; } -#define INTEL_PCH_ACS_FLAGS (PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_SV) +#define INTEL_PCH_ACS_FLAGS (PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF) static int pci_quirk_intel_pch_acs(struct pci_dev *dev, u16 acs_flags) { - u16 flags = dev->dev_flags & PCI_DEV_FLAGS_ACS_ENABLED_QUIRK ? - INTEL_PCH_ACS_FLAGS : 0; - if (!pci_quirk_intel_pch_acs_match(dev)) return -ENOTTY; - return acs_flags & ~flags ? 0 : 1; + if (dev->dev_flags & PCI_DEV_FLAGS_ACS_ENABLED_QUIRK) + acs_flags &= ~(INTEL_PCH_ACS_FLAGS); + + return acs_flags ? 0 : 1; } /* - * These QCOM root ports do provide ACS-like features to disable peer + * These QCOM Root Ports do provide ACS-like features to disable peer * transactions and validate bus numbers in requests, but do not provide an * actual PCIe ACS capability. Hardware supports source validation but it * will report the issue as Completer Abort instead of ACS Violation. - * Hardware doesn't support peer-to-peer and each root port is a root - * complex with unique segment numbers. It is not possible for one root - * port to pass traffic to another root port. All PCIe transactions are - * terminated inside the root port. + * Hardware doesn't support peer-to-peer and each Root Port is a Root + * Complex with unique segment numbers. It is not possible for one Root + * Port to pass traffic to another Root Port. All PCIe transactions are + * terminated inside the Root Port. */ static int pci_quirk_qcom_rp_acs(struct pci_dev *dev, u16 acs_flags) { - u16 flags = (PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_SV); - int ret = acs_flags & ~flags ? 0 : 1; + acs_flags &= ~(PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); - pci_info(dev, "Using QCOM ACS Quirk (%d)\n", ret); - - return ret; + return acs_flags ? 0 : 1; } static int pci_quirk_al_acs(struct pci_dev *dev, u16 acs_flags) From 7cf2cba43f15c74bac46dc5f0326805d25ef514d Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Fri, 6 Sep 2019 18:36:06 -0500 Subject: [PATCH 091/160] PCI: Unify ACS quirk desired vs provided checking Most of the ACS quirks have a similar pattern of: acs_flags &= ~( ); return acs_flags ? 0 : 1; Pull this out into a helper function to simplify the quirks slightly. The helper function is also a convenient place for comments about what the list of ACS controls means. No functional change intended. Signed-off-by: Bjorn Helgaas Reviewed-by: Logan Gunthorpe Reviewed-by: Alex Williamson --- drivers/pci/quirks.c | 67 +++++++++++++++++++++++++++++--------------- 1 file changed, 45 insertions(+), 22 deletions(-) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 59f73d084e1d..9a1051071a81 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -4296,6 +4296,24 @@ static void quirk_chelsio_T5_disable_root_port_attributes(struct pci_dev *pdev) DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_CHELSIO, PCI_ANY_ID, quirk_chelsio_T5_disable_root_port_attributes); +/* + * pci_acs_ctrl_enabled - compare desired ACS controls with those provided + * by a device + * @acs_ctrl_req: Bitmask of desired ACS controls + * @acs_ctrl_ena: Bitmask of ACS controls enabled or provided implicitly by + * the hardware design + * + * Return 1 if all ACS controls in the @acs_ctrl_req bitmask are included + * in @acs_ctrl_ena, i.e., the device provides all the access controls the + * caller desires. Return 0 otherwise. + */ +static int pci_acs_ctrl_enabled(u16 acs_ctrl_req, u16 acs_ctrl_ena) +{ + if ((acs_ctrl_req & acs_ctrl_ena) == acs_ctrl_req) + return 1; + return 0; +} + /* * AMD has indicated that the devices below do not support peer-to-peer * in any system where they are found in the southbridge with an AMD @@ -4339,7 +4357,7 @@ static int pci_quirk_amd_sb_acs(struct pci_dev *dev, u16 acs_flags) /* Filter out flags not applicable to multifunction */ acs_flags &= (PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_EC | PCI_ACS_DT); - return acs_flags & ~(PCI_ACS_RR | PCI_ACS_CR) ? 0 : 1; + return pci_acs_ctrl_enabled(acs_flags, PCI_ACS_RR | PCI_ACS_CR); #else return -ENODEV; #endif @@ -4377,9 +4395,8 @@ static int pci_quirk_cavium_acs(struct pci_dev *dev, u16 acs_flags) * hardware implements and enables equivalent ACS functionality for * these flags. */ - acs_flags &= ~(PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); - - return acs_flags ? 0 : 1; + return pci_acs_ctrl_enabled(acs_flags, + PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); } static int pci_quirk_xgene_acs(struct pci_dev *dev, u16 acs_flags) @@ -4389,9 +4406,8 @@ static int pci_quirk_xgene_acs(struct pci_dev *dev, u16 acs_flags) * transactions with others, allowing masking out these bits as if they * were unimplemented in the ACS capability. */ - acs_flags &= ~(PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); - - return acs_flags ? 0 : 1; + return pci_acs_ctrl_enabled(acs_flags, + PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); } /* @@ -4443,17 +4459,16 @@ static bool pci_quirk_intel_pch_acs_match(struct pci_dev *dev) return false; } -#define INTEL_PCH_ACS_FLAGS (PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF) - static int pci_quirk_intel_pch_acs(struct pci_dev *dev, u16 acs_flags) { if (!pci_quirk_intel_pch_acs_match(dev)) return -ENOTTY; if (dev->dev_flags & PCI_DEV_FLAGS_ACS_ENABLED_QUIRK) - acs_flags &= ~(INTEL_PCH_ACS_FLAGS); + return pci_acs_ctrl_enabled(acs_flags, + PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); - return acs_flags ? 0 : 1; + return pci_acs_ctrl_enabled(acs_flags, 0); } /* @@ -4468,9 +4483,8 @@ static int pci_quirk_intel_pch_acs(struct pci_dev *dev, u16 acs_flags) */ static int pci_quirk_qcom_rp_acs(struct pci_dev *dev, u16 acs_flags) { - acs_flags &= ~(PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); - - return acs_flags ? 0 : 1; + return pci_acs_ctrl_enabled(acs_flags, + PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); } static int pci_quirk_al_acs(struct pci_dev *dev, u16 acs_flags) @@ -4571,7 +4585,7 @@ static int pci_quirk_intel_spt_pch_acs(struct pci_dev *dev, u16 acs_flags) pci_read_config_dword(dev, pos + INTEL_SPT_ACS_CTRL, &ctrl); - return acs_flags & ~ctrl ? 0 : 1; + return pci_acs_ctrl_enabled(acs_flags, ctrl); } static int pci_quirk_mf_endpoint_acs(struct pci_dev *dev, u16 acs_flags) @@ -4585,10 +4599,9 @@ static int pci_quirk_mf_endpoint_acs(struct pci_dev *dev, u16 acs_flags) * perform peer-to-peer with other functions, allowing us to mask out * these bits as if they were unimplemented in the ACS capability. */ - acs_flags &= ~(PCI_ACS_SV | PCI_ACS_TB | PCI_ACS_RR | - PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_DT); - - return acs_flags ? 0 : 1; + return pci_acs_ctrl_enabled(acs_flags, + PCI_ACS_SV | PCI_ACS_TB | PCI_ACS_RR | + PCI_ACS_CR | PCI_ACS_UF | PCI_ACS_DT); } static int pci_quirk_brcm_acs(struct pci_dev *dev, u16 acs_flags) @@ -4599,9 +4612,8 @@ static int pci_quirk_brcm_acs(struct pci_dev *dev, u16 acs_flags) * Allow each Root Port to be in a separate IOMMU group by masking * SV/RR/CR/UF bits. */ - acs_flags &= ~(PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); - - return acs_flags ? 0 : 1; + return pci_acs_ctrl_enabled(acs_flags, + PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF); } static const struct pci_dev_acs_enabled { @@ -4703,6 +4715,17 @@ static const struct pci_dev_acs_enabled { { 0 } }; +/* + * pci_dev_specific_acs_enabled - check whether device provides ACS controls + * @dev: PCI device + * @acs_flags: Bitmask of desired ACS controls + * + * Returns: + * -ENOTTY: No quirk applies to this device; we can't tell whether the + * device provides the desired controls + * 0: Device does not provide all the desired controls + * >0: Device provides all the controls in @acs_flags + */ int pci_dev_specific_acs_enabled(struct pci_dev *dev, u16 acs_flags) { const struct pci_dev_acs_enabled *i; From 08bcdd22ecdb01dd60d9284a55f8220a8a40150e Mon Sep 17 00:00:00 2001 From: Jon Derrick Date: Tue, 12 Nov 2019 05:47:52 -0700 Subject: [PATCH 092/160] PCI: vmd: Add bus 224-255 restriction decode VMD bus restrictions are required when IO fabric is multiplexed such that VMD cannot use the entire bus range. This patch adds another bus restriction decode bit that can be set by firmware to restrict the VMD bus range to between 224-255. Signed-off-by: Jon Derrick Signed-off-by: Lorenzo Pieralisi --- drivers/pci/controller/vmd.c | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-) diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c index a35d3f3996d7..15302a17ee3f 100644 --- a/drivers/pci/controller/vmd.c +++ b/drivers/pci/controller/vmd.c @@ -602,16 +602,30 @@ static int vmd_enable_domain(struct vmd_dev *vmd, unsigned long features) /* * Certain VMD devices may have a root port configuration option which - * limits the bus range to between 0-127 or 128-255 + * limits the bus range to between 0-127, 128-255, or 224-255 */ if (features & VMD_FEAT_HAS_BUS_RESTRICTIONS) { - u32 vmcap, vmconfig; + u16 reg16; - pci_read_config_dword(vmd->dev, PCI_REG_VMCAP, &vmcap); - pci_read_config_dword(vmd->dev, PCI_REG_VMCONFIG, &vmconfig); - if (BUS_RESTRICT_CAP(vmcap) && - (BUS_RESTRICT_CFG(vmconfig) == 0x1)) - vmd->busn_start = 128; + pci_read_config_word(vmd->dev, PCI_REG_VMCAP, ®16); + if (BUS_RESTRICT_CAP(reg16)) { + pci_read_config_word(vmd->dev, PCI_REG_VMCONFIG, + ®16); + + switch (BUS_RESTRICT_CFG(reg16)) { + case 1: + vmd->busn_start = 128; + break; + case 2: + vmd->busn_start = 224; + break; + case 3: + pci_err(vmd->dev, "Unknown Bus Offset Setting\n"); + return -ENODEV; + default: + break; + } + } } res = &vmd->dev->resource[VMD_CFGBAR]; From ec11e5c213cc20cac5e8310728b06793448b9f6d Mon Sep 17 00:00:00 2001 From: Jon Derrick Date: Tue, 12 Nov 2019 05:47:53 -0700 Subject: [PATCH 093/160] PCI: vmd: Add device id for VMD device 8086:9A0B This patch adds support for this VMD device which supports the bus restriction mode. Signed-off-by: Jon Derrick Signed-off-by: Lorenzo Pieralisi --- drivers/pci/controller/vmd.c | 2 ++ include/linux/pci_ids.h | 1 + 2 files changed, 3 insertions(+) diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c index 15302a17ee3f..6bff95115d28 100644 --- a/drivers/pci/controller/vmd.c +++ b/drivers/pci/controller/vmd.c @@ -868,6 +868,8 @@ static const struct pci_device_id vmd_ids[] = { {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_VMD_28C0), .driver_data = VMD_FEAT_HAS_MEMBAR_SHADOW | VMD_FEAT_HAS_BUS_RESTRICTIONS,}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_VMD_9A0B), + .driver_data = VMD_FEAT_HAS_BUS_RESTRICTIONS,}, {0,} }; MODULE_DEVICE_TABLE(pci, vmd_ids); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 21a572469a4e..2302d133af6f 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -3006,6 +3006,7 @@ #define PCI_DEVICE_ID_INTEL_84460GX 0x84ea #define PCI_DEVICE_ID_INTEL_IXP4XX 0x8500 #define PCI_DEVICE_ID_INTEL_IXP2800 0x9004 +#define PCI_DEVICE_ID_INTEL_VMD_9A0B 0x9a0b #define PCI_DEVICE_ID_INTEL_S21152BB 0xb152 #define PCI_VENDOR_ID_SCALEMP 0x8686 From 331f63457165a30c708280de2c77f1742c6351dc Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Wed, 30 Oct 2019 17:30:57 -0500 Subject: [PATCH 094/160] PCI: of: Add inbound resource parsing to helpers Extend devm_of_pci_get_host_bridge_resources() and pci_parse_request_of_pci_ranges() helpers to also parse the inbound addresses from DT 'dma-ranges' and populate a resource list with the translated addresses. This will help ensure 'dma-ranges' is always parsed in a consistent way. Tested-by: Srinath Mannam Tested-by: Thomas Petazzoni # for AArdvark Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Srinath Mannam Reviewed-by: Andrew Murray Acked-by: Gustavo Pimentel Cc: Jingoo Han Cc: Gustavo Pimentel Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas Cc: Thomas Petazzoni Cc: Will Deacon Cc: Linus Walleij Cc: Toan Le Cc: Ley Foon Tan Cc: Tom Joseph Cc: Ray Jui Cc: Scott Branden Cc: bcm-kernel-feedback-list@broadcom.com Cc: Ryder Lee Cc: Karthikeyan Mitran Cc: Hou Zhiqiang Cc: Simon Horman Cc: Shawn Lin Cc: Heiko Stuebner Cc: Michal Simek Cc: rfi@lists.rocketboards.org Cc: linux-mediatek@lists.infradead.org Cc: linux-renesas-soc@vger.kernel.org Cc: linux-rockchip@lists.infradead.org --- .../pci/controller/dwc/pcie-designware-host.c | 3 +- drivers/pci/controller/pci-aardvark.c | 2 +- drivers/pci/controller/pci-ftpci100.c | 3 +- drivers/pci/controller/pci-host-common.c | 2 +- drivers/pci/controller/pci-v3-semi.c | 3 +- drivers/pci/controller/pci-versatile.c | 3 +- drivers/pci/controller/pci-xgene.c | 3 +- drivers/pci/controller/pcie-altera.c | 2 +- drivers/pci/controller/pcie-cadence-host.c | 2 +- drivers/pci/controller/pcie-iproc-platform.c | 3 +- drivers/pci/controller/pcie-mediatek.c | 2 +- drivers/pci/controller/pcie-mobiveil.c | 3 +- drivers/pci/controller/pcie-rcar.c | 3 +- drivers/pci/controller/pcie-rockchip-host.c | 3 +- drivers/pci/controller/pcie-xilinx-nwl.c | 3 +- drivers/pci/controller/pcie-xilinx.c | 3 +- drivers/pci/of.c | 61 ++++++++++++++++--- drivers/pci/pci.h | 8 ++- include/linux/pci.h | 9 ++- 19 files changed, 93 insertions(+), 28 deletions(-) diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c index aeec8b65eb97..f7b1d80c4a0a 100644 --- a/drivers/pci/controller/dwc/pcie-designware-host.c +++ b/drivers/pci/controller/dwc/pcie-designware-host.c @@ -342,7 +342,8 @@ int dw_pcie_host_init(struct pcie_port *pp) if (!bridge) return -ENOMEM; - ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + &bridge->dma_ranges, NULL); if (ret) return ret; diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c index 9cbeba507f0c..b34eaa2cd762 100644 --- a/drivers/pci/controller/pci-aardvark.c +++ b/drivers/pci/controller/pci-aardvark.c @@ -939,7 +939,7 @@ static int advk_pcie_probe(struct platform_device *pdev) } ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, - &bus); + &bridge->dma_ranges, &bus); if (ret) { dev_err(dev, "Failed to parse resources\n"); return ret; diff --git a/drivers/pci/controller/pci-ftpci100.c b/drivers/pci/controller/pci-ftpci100.c index 75603348b88a..66288b94e92d 100644 --- a/drivers/pci/controller/pci-ftpci100.c +++ b/drivers/pci/controller/pci-ftpci100.c @@ -477,7 +477,8 @@ static int faraday_pci_probe(struct platform_device *pdev) if (IS_ERR(p->base)) return PTR_ERR(p->base); - ret = pci_parse_request_of_pci_ranges(dev, &host->windows, NULL); + ret = pci_parse_request_of_pci_ranges(dev, &host->windows, + &host->dma_ranges, NULL); if (ret) return ret; diff --git a/drivers/pci/controller/pci-host-common.c b/drivers/pci/controller/pci-host-common.c index c8cb9c5188a4..250a3fc80ec6 100644 --- a/drivers/pci/controller/pci-host-common.c +++ b/drivers/pci/controller/pci-host-common.c @@ -27,7 +27,7 @@ static struct pci_config_window *gen_pci_init(struct device *dev, struct pci_config_window *cfg; /* Parse our PCI ranges and request their resources */ - err = pci_parse_request_of_pci_ranges(dev, resources, &bus_range); + err = pci_parse_request_of_pci_ranges(dev, resources, NULL, &bus_range); if (err) return ERR_PTR(err); diff --git a/drivers/pci/controller/pci-v3-semi.c b/drivers/pci/controller/pci-v3-semi.c index 96677520f6c1..2209c7671115 100644 --- a/drivers/pci/controller/pci-v3-semi.c +++ b/drivers/pci/controller/pci-v3-semi.c @@ -776,7 +776,8 @@ static int v3_pci_probe(struct platform_device *pdev) if (IS_ERR(v3->config_base)) return PTR_ERR(v3->config_base); - ret = pci_parse_request_of_pci_ranges(dev, &host->windows, NULL); + ret = pci_parse_request_of_pci_ranges(dev, &host->windows, + &host->dma_ranges, NULL); if (ret) return ret; diff --git a/drivers/pci/controller/pci-versatile.c b/drivers/pci/controller/pci-versatile.c index eae1b859990b..b911359b6d81 100644 --- a/drivers/pci/controller/pci-versatile.c +++ b/drivers/pci/controller/pci-versatile.c @@ -92,7 +92,8 @@ static int versatile_pci_probe(struct platform_device *pdev) if (IS_ERR(versatile_cfg_base[1])) return PTR_ERR(versatile_cfg_base[1]); - ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + NULL, NULL); if (ret) return ret; diff --git a/drivers/pci/controller/pci-xgene.c b/drivers/pci/controller/pci-xgene.c index 7d0f0395a479..9408269d943d 100644 --- a/drivers/pci/controller/pci-xgene.c +++ b/drivers/pci/controller/pci-xgene.c @@ -627,7 +627,8 @@ static int xgene_pcie_probe(struct platform_device *pdev) if (ret) return ret; - ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + &bridge->dma_ranges, NULL); if (ret) return ret; diff --git a/drivers/pci/controller/pcie-altera.c b/drivers/pci/controller/pcie-altera.c index ba025efeae28..b447c3e4abad 100644 --- a/drivers/pci/controller/pcie-altera.c +++ b/drivers/pci/controller/pcie-altera.c @@ -800,7 +800,7 @@ static int altera_pcie_probe(struct platform_device *pdev) } ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, - NULL); + &bridge->dma_ranges, NULL); if (ret) { dev_err(dev, "Failed add resources\n"); return ret; diff --git a/drivers/pci/controller/pcie-cadence-host.c b/drivers/pci/controller/pcie-cadence-host.c index 97e251090b4f..a8f7a6284c3e 100644 --- a/drivers/pci/controller/pcie-cadence-host.c +++ b/drivers/pci/controller/pcie-cadence-host.c @@ -211,7 +211,7 @@ static int cdns_pcie_host_init(struct device *dev, int err; /* Parse our PCI ranges and request their resources */ - err = pci_parse_request_of_pci_ranges(dev, resources, &bus_range); + err = pci_parse_request_of_pci_ranges(dev, resources, NULL, &bus_range); if (err) return err; diff --git a/drivers/pci/controller/pcie-iproc-platform.c b/drivers/pci/controller/pcie-iproc-platform.c index 375d815f7301..ff0a81a632a1 100644 --- a/drivers/pci/controller/pcie-iproc-platform.c +++ b/drivers/pci/controller/pcie-iproc-platform.c @@ -95,7 +95,8 @@ static int iproc_pcie_pltfm_probe(struct platform_device *pdev) if (IS_ERR(pcie->phy)) return PTR_ERR(pcie->phy); - ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + &bridge->dma_ranges, NULL); if (ret) { dev_err(dev, "unable to get PCI host bridge resources\n"); return ret; diff --git a/drivers/pci/controller/pcie-mediatek.c b/drivers/pci/controller/pcie-mediatek.c index d9206a3cd56b..cb982891b22b 100644 --- a/drivers/pci/controller/pcie-mediatek.c +++ b/drivers/pci/controller/pcie-mediatek.c @@ -1034,7 +1034,7 @@ static int mtk_pcie_setup(struct mtk_pcie *pcie) int err; err = pci_parse_request_of_pci_ranges(dev, windows, - &bus); + &host->dma_ranges, &bus); if (err) return err; diff --git a/drivers/pci/controller/pcie-mobiveil.c b/drivers/pci/controller/pcie-mobiveil.c index 4eab8624ce4d..257ba49c177c 100644 --- a/drivers/pci/controller/pcie-mobiveil.c +++ b/drivers/pci/controller/pcie-mobiveil.c @@ -875,7 +875,8 @@ static int mobiveil_pcie_probe(struct platform_device *pdev) } /* parse the host bridge base addresses from the device tree file */ - ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); + ret = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + &bridge->dma_ranges, NULL); if (ret) { dev_err(dev, "Getting bridge resources failed\n"); return ret; diff --git a/drivers/pci/controller/pcie-rcar.c b/drivers/pci/controller/pcie-rcar.c index f6a669a9af41..b8d6e86a5539 100644 --- a/drivers/pci/controller/pcie-rcar.c +++ b/drivers/pci/controller/pcie-rcar.c @@ -1138,7 +1138,8 @@ static int rcar_pcie_probe(struct platform_device *pdev) pcie->dev = dev; platform_set_drvdata(pdev, pcie); - err = pci_parse_request_of_pci_ranges(dev, &pcie->resources, NULL); + err = pci_parse_request_of_pci_ranges(dev, &pcie->resources, + &bridge->dma_ranges, NULL); if (err) goto err_free_bridge; diff --git a/drivers/pci/controller/pcie-rockchip-host.c b/drivers/pci/controller/pcie-rockchip-host.c index f375e55ea02e..ee83f8494ee9 100644 --- a/drivers/pci/controller/pcie-rockchip-host.c +++ b/drivers/pci/controller/pcie-rockchip-host.c @@ -1004,7 +1004,8 @@ static int rockchip_pcie_probe(struct platform_device *pdev) if (err < 0) goto err_deinit_port; - err = pci_parse_request_of_pci_ranges(dev, &bridge->windows, &bus_res); + err = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + &bridge->dma_ranges, &bus_res); if (err) goto err_remove_irq_domain; diff --git a/drivers/pci/controller/pcie-xilinx-nwl.c b/drivers/pci/controller/pcie-xilinx-nwl.c index e135a4b60489..9bd1427f2fd6 100644 --- a/drivers/pci/controller/pcie-xilinx-nwl.c +++ b/drivers/pci/controller/pcie-xilinx-nwl.c @@ -843,7 +843,8 @@ static int nwl_pcie_probe(struct platform_device *pdev) return err; } - err = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); + err = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + &bridge->dma_ranges, NULL); if (err) { dev_err(dev, "Getting bridge resources failed\n"); return err; diff --git a/drivers/pci/controller/pcie-xilinx.c b/drivers/pci/controller/pcie-xilinx.c index 257702288787..98e55297815b 100644 --- a/drivers/pci/controller/pcie-xilinx.c +++ b/drivers/pci/controller/pcie-xilinx.c @@ -645,7 +645,8 @@ static int xilinx_pcie_probe(struct platform_device *pdev) return err; } - err = pci_parse_request_of_pci_ranges(dev, &bridge->windows, NULL); + err = pci_parse_request_of_pci_ranges(dev, &bridge->windows, + &bridge->dma_ranges, NULL); if (err) { dev_err(dev, "Getting bridge resources failed\n"); return err; diff --git a/drivers/pci/of.c b/drivers/pci/of.c index f3da49a31db4..1dfde6f9e82d 100644 --- a/drivers/pci/of.c +++ b/drivers/pci/of.c @@ -257,14 +257,16 @@ EXPORT_SYMBOL_GPL(of_pci_check_probe_only); */ int devm_of_pci_get_host_bridge_resources(struct device *dev, unsigned char busno, unsigned char bus_max, - struct list_head *resources, resource_size_t *io_base) + struct list_head *resources, + struct list_head *ib_resources, + resource_size_t *io_base) { struct device_node *dev_node = dev->of_node; struct resource *res, tmp_res; struct resource *bus_range; struct of_pci_range range; struct of_pci_range_parser parser; - char range_type[4]; + const char *range_type; int err; if (io_base) @@ -298,12 +300,12 @@ int devm_of_pci_get_host_bridge_resources(struct device *dev, for_each_of_pci_range(&parser, &range) { /* Read next ranges element */ if ((range.flags & IORESOURCE_TYPE_BITS) == IORESOURCE_IO) - snprintf(range_type, 4, " IO"); + range_type = "IO"; else if ((range.flags & IORESOURCE_TYPE_BITS) == IORESOURCE_MEM) - snprintf(range_type, 4, "MEM"); + range_type = "MEM"; else - snprintf(range_type, 4, "err"); - dev_info(dev, " %s %#010llx..%#010llx -> %#010llx\n", + range_type = "err"; + dev_info(dev, " %6s %#012llx..%#012llx -> %#012llx\n", range_type, range.cpu_addr, range.cpu_addr + range.size - 1, range.pci_addr); @@ -340,6 +342,48 @@ int devm_of_pci_get_host_bridge_resources(struct device *dev, pci_add_resource_offset(resources, res, res->start - range.pci_addr); } + /* Check for dma-ranges property */ + if (!ib_resources) + return 0; + err = of_pci_dma_range_parser_init(&parser, dev_node); + if (err) + return 0; + + dev_dbg(dev, "Parsing dma-ranges property...\n"); + for_each_of_pci_range(&parser, &range) { + struct resource_entry *entry; + /* + * If we failed translation or got a zero-sized region + * then skip this range + */ + if (((range.flags & IORESOURCE_TYPE_BITS) != IORESOURCE_MEM) || + range.cpu_addr == OF_BAD_ADDR || range.size == 0) + continue; + + dev_info(dev, " %6s %#012llx..%#012llx -> %#012llx\n", + "IB MEM", range.cpu_addr, + range.cpu_addr + range.size - 1, range.pci_addr); + + + err = of_pci_range_to_resource(&range, dev_node, &tmp_res); + if (err) + continue; + + res = devm_kmemdup(dev, &tmp_res, sizeof(tmp_res), GFP_KERNEL); + if (!res) { + err = -ENOMEM; + goto failed; + } + + /* Keep the resource list sorted */ + resource_list_for_each_entry(entry, ib_resources) + if (entry->res->start > res->start) + break; + + pci_add_resource_offset(&entry->node, res, + res->start - range.pci_addr); + } + return 0; failed: @@ -482,6 +526,7 @@ EXPORT_SYMBOL_GPL(of_irq_parse_and_map_pci); int pci_parse_request_of_pci_ranges(struct device *dev, struct list_head *resources, + struct list_head *ib_resources, struct resource **bus_range) { int err, res_valid = 0; @@ -489,8 +534,10 @@ int pci_parse_request_of_pci_ranges(struct device *dev, struct resource_entry *win, *tmp; INIT_LIST_HEAD(resources); + if (ib_resources) + INIT_LIST_HEAD(ib_resources); err = devm_of_pci_get_host_bridge_resources(dev, 0, 0xff, resources, - &iobase); + ib_resources, &iobase); if (err) return err; diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 3f6947ee3324..6692c4fe4290 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -633,11 +633,15 @@ static inline void pci_release_bus_of_node(struct pci_bus *bus) { } #if defined(CONFIG_OF_ADDRESS) int devm_of_pci_get_host_bridge_resources(struct device *dev, unsigned char busno, unsigned char bus_max, - struct list_head *resources, resource_size_t *io_base); + struct list_head *resources, + struct list_head *ib_resources, + resource_size_t *io_base); #else static inline int devm_of_pci_get_host_bridge_resources(struct device *dev, unsigned char busno, unsigned char bus_max, - struct list_head *resources, resource_size_t *io_base) + struct list_head *resources, + struct list_head *ib_resources, + resource_size_t *io_base) { return -EINVAL; } diff --git a/include/linux/pci.h b/include/linux/pci.h index f9088c89a534..5cb94916eaa1 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -2278,6 +2278,7 @@ struct irq_domain; struct irq_domain *pci_host_bridge_of_msi_domain(struct pci_bus *bus); int pci_parse_request_of_pci_ranges(struct device *dev, struct list_head *resources, + struct list_head *ib_resources, struct resource **bus_range); /* Arch may override this (weak) */ @@ -2286,9 +2287,11 @@ struct device_node *pcibios_get_phb_of_node(struct pci_bus *bus); #else /* CONFIG_OF */ static inline struct irq_domain * pci_host_bridge_of_msi_domain(struct pci_bus *bus) { return NULL; } -static inline int pci_parse_request_of_pci_ranges(struct device *dev, - struct list_head *resources, - struct resource **bus_range) +static inline int +pci_parse_request_of_pci_ranges(struct device *dev, + struct list_head *resources, + struct list_head *ib_resources, + struct resource **bus_range) { return -EINVAL; } From ea4f718e8455e8d94ec5fcf270b9d37988fc5bbd Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:51 -0500 Subject: [PATCH 095/160] PCI: ftpci100: Use inbound resources for setup Now that the helpers provide the inbound resources in the host bridge 'dma_ranges' resource list, convert Faraday ftpci100 host bridge to use the resource list to setup the inbound addresses. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Acked-by: Linus Walleij Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/controller/pci-ftpci100.c | 27 ++++++++++----------------- 1 file changed, 10 insertions(+), 17 deletions(-) diff --git a/drivers/pci/controller/pci-ftpci100.c b/drivers/pci/controller/pci-ftpci100.c index 66288b94e92d..1b67564de7af 100644 --- a/drivers/pci/controller/pci-ftpci100.c +++ b/drivers/pci/controller/pci-ftpci100.c @@ -375,12 +375,11 @@ static int faraday_pci_setup_cascaded_irq(struct faraday_pci *p) return 0; } -static int faraday_pci_parse_map_dma_ranges(struct faraday_pci *p, - struct device_node *np) +static int faraday_pci_parse_map_dma_ranges(struct faraday_pci *p) { - struct of_pci_range range; - struct of_pci_range_parser parser; struct device *dev = p->dev; + struct pci_host_bridge *bridge = pci_host_bridge_from_priv(p); + struct resource_entry *entry; u32 confreg[3] = { FARADAY_PCI_MEM1_BASE_SIZE, FARADAY_PCI_MEM2_BASE_SIZE, @@ -389,19 +388,13 @@ static int faraday_pci_parse_map_dma_ranges(struct faraday_pci *p, int i = 0; u32 val; - if (of_pci_dma_range_parser_init(&parser, np)) { - dev_err(dev, "missing dma-ranges property\n"); - return -EINVAL; - } - - /* - * Get the dma-ranges from the device tree - */ - for_each_of_pci_range(&parser, &range) { - u64 end = range.pci_addr + range.size - 1; + resource_list_for_each_entry(entry, &bridge->dma_ranges) { + u64 pci_addr = entry->res->start - entry->offset; + u64 end = entry->res->end - entry->offset; int ret; - ret = faraday_res_to_memcfg(range.pci_addr, range.size, &val); + ret = faraday_res_to_memcfg(pci_addr, + resource_size(entry->res), &val); if (ret) { dev_err(dev, "DMA range %d: illegal MEM resource size\n", i); @@ -409,7 +402,7 @@ static int faraday_pci_parse_map_dma_ranges(struct faraday_pci *p, } dev_info(dev, "DMA MEM%d BASE: 0x%016llx -> 0x%016llx config %08x\n", - i + 1, range.pci_addr, end, val); + i + 1, pci_addr, end, val); if (i <= 2) { faraday_raw_pci_write_config(p, 0, 0, confreg[i], 4, val); @@ -539,7 +532,7 @@ static int faraday_pci_probe(struct platform_device *pdev) cur_bus_speed = PCI_SPEED_66MHz; } - ret = faraday_pci_parse_map_dma_ranges(p, dev->of_node); + ret = faraday_pci_parse_map_dma_ranges(p); if (ret) return ret; From 070d7d70291c3a6b61fb10f0e9cf5874df91e214 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:52 -0500 Subject: [PATCH 096/160] PCI: v3-semi: Use inbound resources for setup Now that the helpers provide the inbound resources in the host bridge 'dma_ranges' resource list, convert the v3-semi host bridge to use the resource list to setup the inbound addresses. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Reviewed-by: Linus Walleij Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/controller/pci-v3-semi.c | 38 ++++++++++++---------------- 1 file changed, 16 insertions(+), 22 deletions(-) diff --git a/drivers/pci/controller/pci-v3-semi.c b/drivers/pci/controller/pci-v3-semi.c index 2209c7671115..bd05221f5a22 100644 --- a/drivers/pci/controller/pci-v3-semi.c +++ b/drivers/pci/controller/pci-v3-semi.c @@ -598,28 +598,30 @@ static int v3_pci_setup_resource(struct v3_pci *v3, } static int v3_get_dma_range_config(struct v3_pci *v3, - struct of_pci_range *range, + struct resource_entry *entry, u32 *pci_base, u32 *pci_map) { struct device *dev = v3->dev; - u64 cpu_end = range->cpu_addr + range->size - 1; - u64 pci_end = range->pci_addr + range->size - 1; + u64 cpu_addr = entry->res->start; + u64 cpu_end = entry->res->end; + u64 pci_end = cpu_end - entry->offset; + u64 pci_addr = entry->res->start - entry->offset; u32 val; - if (range->pci_addr & ~V3_PCI_BASE_M_ADR_BASE) { + if (pci_addr & ~V3_PCI_BASE_M_ADR_BASE) { dev_err(dev, "illegal range, only PCI bits 31..20 allowed\n"); return -EINVAL; } - val = ((u32)range->pci_addr) & V3_PCI_BASE_M_ADR_BASE; + val = ((u32)pci_addr) & V3_PCI_BASE_M_ADR_BASE; *pci_base = val; - if (range->cpu_addr & ~V3_PCI_MAP_M_MAP_ADR) { + if (cpu_addr & ~V3_PCI_MAP_M_MAP_ADR) { dev_err(dev, "illegal range, only CPU bits 31..20 allowed\n"); return -EINVAL; } - val = ((u32)range->cpu_addr) & V3_PCI_MAP_M_MAP_ADR; + val = ((u32)cpu_addr) & V3_PCI_MAP_M_MAP_ADR; - switch (range->size) { + switch (resource_size(entry->res)) { case SZ_1M: val |= V3_LB_BASE_ADR_SIZE_1MB; break; @@ -667,8 +669,8 @@ static int v3_get_dma_range_config(struct v3_pci *v3, dev_dbg(dev, "DMA MEM CPU: 0x%016llx -> 0x%016llx => " "PCI: 0x%016llx -> 0x%016llx base %08x map %08x\n", - range->cpu_addr, cpu_end, - range->pci_addr, pci_end, + cpu_addr, cpu_end, + pci_addr, pci_end, *pci_base, *pci_map); return 0; @@ -677,24 +679,16 @@ static int v3_get_dma_range_config(struct v3_pci *v3, static int v3_pci_parse_map_dma_ranges(struct v3_pci *v3, struct device_node *np) { - struct of_pci_range range; - struct of_pci_range_parser parser; + struct pci_host_bridge *bridge = pci_host_bridge_from_priv(v3); struct device *dev = v3->dev; + struct resource_entry *entry; int i = 0; - if (of_pci_dma_range_parser_init(&parser, np)) { - dev_err(dev, "missing dma-ranges property\n"); - return -EINVAL; - } - - /* - * Get the dma-ranges from the device tree - */ - for_each_of_pci_range(&parser, &range) { + resource_list_for_each_entry(entry, &bridge->dma_ranges) { int ret; u32 pci_base, pci_map; - ret = v3_get_dma_range_config(v3, &range, &pci_base, &pci_map); + ret = v3_get_dma_range_config(v3, entry, &pci_base, &pci_map); if (ret) return ret; From 6dce5aa59e0bf2430733d7a8b11c205ec10f408e Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:53 -0500 Subject: [PATCH 097/160] PCI: xgene: Use inbound resources for setup Now that the helpers provide the inbound resources in the host bridge 'dma_ranges' resource list, convert the Xgene host bridge to use the resource list to setup the inbound addresses. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Cc: Toan Le Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/controller/pci-xgene.c | 33 ++++++++++-------------------- 1 file changed, 11 insertions(+), 22 deletions(-) diff --git a/drivers/pci/controller/pci-xgene.c b/drivers/pci/controller/pci-xgene.c index 9408269d943d..de195fd430dc 100644 --- a/drivers/pci/controller/pci-xgene.c +++ b/drivers/pci/controller/pci-xgene.c @@ -481,27 +481,28 @@ static int xgene_pcie_select_ib_reg(u8 *ib_reg_mask, u64 size) } static void xgene_pcie_setup_ib_reg(struct xgene_pcie_port *port, - struct of_pci_range *range, u8 *ib_reg_mask) + struct resource_entry *entry, + u8 *ib_reg_mask) { void __iomem *cfg_base = port->cfg_base; struct device *dev = port->dev; void *bar_addr; u32 pim_reg; - u64 cpu_addr = range->cpu_addr; - u64 pci_addr = range->pci_addr; - u64 size = range->size; + u64 cpu_addr = entry->res->start; + u64 pci_addr = cpu_addr - entry->offset; + u64 size = resource_size(entry->res); u64 mask = ~(size - 1) | EN_REG; u32 flags = PCI_BASE_ADDRESS_MEM_TYPE_64; u32 bar_low; int region; - region = xgene_pcie_select_ib_reg(ib_reg_mask, range->size); + region = xgene_pcie_select_ib_reg(ib_reg_mask, size); if (region < 0) { dev_warn(dev, "invalid pcie dma-range config\n"); return; } - if (range->flags & IORESOURCE_PREFETCH) + if (entry->res->flags & IORESOURCE_PREFETCH) flags |= PCI_BASE_ADDRESS_MEM_PREFETCH; bar_low = pcie_bar_low_val((u32)cpu_addr, flags); @@ -532,25 +533,13 @@ static void xgene_pcie_setup_ib_reg(struct xgene_pcie_port *port, static int xgene_pcie_parse_map_dma_ranges(struct xgene_pcie_port *port) { - struct device_node *np = port->node; - struct of_pci_range range; - struct of_pci_range_parser parser; - struct device *dev = port->dev; + struct pci_host_bridge *bridge = pci_host_bridge_from_priv(port); + struct resource_entry *entry; u8 ib_reg_mask = 0; - if (of_pci_dma_range_parser_init(&parser, np)) { - dev_err(dev, "missing dma-ranges property\n"); - return -EINVAL; - } + resource_list_for_each_entry(entry, &bridge->dma_ranges) + xgene_pcie_setup_ib_reg(port, entry, &ib_reg_mask); - /* Get the dma-ranges from DT */ - for_each_of_pci_range(&parser, &range) { - u64 end = range.cpu_addr + range.size - 1; - - dev_dbg(dev, "0x%08x 0x%016llx..0x%016llx -> 0x%016llx\n", - range.flags, range.cpu_addr, end, range.pci_addr); - xgene_pcie_setup_ib_reg(port, &range, &ib_reg_mask); - } return 0; } From b9ae59b30bcf62691aec4170ce4bafc20e968490 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:54 -0500 Subject: [PATCH 098/160] PCI: iproc: Use inbound resources for setup Now that the helpers provide the inbound resources in the host bridge 'dma_ranges' resource list, convert Broadcom iProc host bridge to use the resource list to setup the inbound addresses. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas Cc: Ray Jui Cc: Scott Branden Cc: bcm-kernel-feedback-list@broadcom.com --- drivers/pci/controller/pcie-iproc.c | 77 +++++++---------------------- 1 file changed, 17 insertions(+), 60 deletions(-) diff --git a/drivers/pci/controller/pcie-iproc.c b/drivers/pci/controller/pcie-iproc.c index 223335ee791a..f4d78e66846e 100644 --- a/drivers/pci/controller/pcie-iproc.c +++ b/drivers/pci/controller/pcie-iproc.c @@ -1122,15 +1122,16 @@ static int iproc_pcie_ib_write(struct iproc_pcie *pcie, int region_idx, } static int iproc_pcie_setup_ib(struct iproc_pcie *pcie, - struct of_pci_range *range, + struct resource_entry *entry, enum iproc_pcie_ib_map_type type) { struct device *dev = pcie->dev; struct iproc_pcie_ib *ib = &pcie->ib; int ret; unsigned int region_idx, size_idx; - u64 axi_addr = range->cpu_addr, pci_addr = range->pci_addr; - resource_size_t size = range->size; + u64 axi_addr = entry->res->start; + u64 pci_addr = entry->res->start - entry->offset; + resource_size_t size = resource_size(entry->res); /* iterate through all IARR mapping regions */ for (region_idx = 0; region_idx < ib->nr_regions; region_idx++) { @@ -1182,66 +1183,19 @@ err_ib: return ret; } -static int iproc_pcie_add_dma_range(struct device *dev, - struct list_head *resources, - struct of_pci_range *range) -{ - struct resource *res; - struct resource_entry *entry, *tmp; - struct list_head *head = resources; - - res = devm_kzalloc(dev, sizeof(struct resource), GFP_KERNEL); - if (!res) - return -ENOMEM; - - resource_list_for_each_entry(tmp, resources) { - if (tmp->res->start < range->cpu_addr) - head = &tmp->node; - } - - res->start = range->cpu_addr; - res->end = res->start + range->size - 1; - - entry = resource_list_create_entry(res, 0); - if (!entry) - return -ENOMEM; - - entry->offset = res->start - range->cpu_addr; - resource_list_add(entry, head); - - return 0; -} - static int iproc_pcie_map_dma_ranges(struct iproc_pcie *pcie) { struct pci_host_bridge *host = pci_host_bridge_from_priv(pcie); - struct of_pci_range range; - struct of_pci_range_parser parser; - int ret; - LIST_HEAD(resources); + struct resource_entry *entry; + int ret = 0; - /* Get the dma-ranges from DT */ - ret = of_pci_dma_range_parser_init(&parser, pcie->dev->of_node); - if (ret) - return ret; - - for_each_of_pci_range(&parser, &range) { - ret = iproc_pcie_add_dma_range(pcie->dev, - &resources, - &range); - if (ret) - goto out; + resource_list_for_each_entry(entry, &host->dma_ranges) { /* Each range entry corresponds to an inbound mapping region */ - ret = iproc_pcie_setup_ib(pcie, &range, IPROC_PCIE_IB_MAP_MEM); + ret = iproc_pcie_setup_ib(pcie, entry, IPROC_PCIE_IB_MAP_MEM); if (ret) - goto out; + break; } - list_splice_init(&resources, &host->dma_ranges); - - return 0; -out: - pci_free_resource_list(&resources); return ret; } @@ -1276,13 +1230,16 @@ static int iproce_pcie_get_msi(struct iproc_pcie *pcie, static int iproc_pcie_paxb_v2_msi_steer(struct iproc_pcie *pcie, u64 msi_addr) { int ret; - struct of_pci_range range; + struct resource_entry entry; - memset(&range, 0, sizeof(range)); - range.size = SZ_32K; - range.pci_addr = range.cpu_addr = msi_addr & ~(range.size - 1); + memset(&entry, 0, sizeof(entry)); + entry.res = &entry.__res; - ret = iproc_pcie_setup_ib(pcie, &range, IPROC_PCIE_IB_MAP_IO); + msi_addr &= ~(SZ_32K - 1); + entry.res->start = msi_addr; + entry.res->end = msi_addr + SZ_32K - 1; + + ret = iproc_pcie_setup_ib(pcie, &entry, IPROC_PCIE_IB_MAP_IO); return ret; } From 085f793984adcbf3966176bac088c32e0b66235a Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:55 -0500 Subject: [PATCH 099/160] PCI: rcar: Use inbound resources for setup Now that the helpers provide the inbound resources in the host bridge 'dma_ranges' resource list, convert Renesas R-Car PCIe host bridge to use the resource list to setup the inbound addresses. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Cc: Simon Horman Cc: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/controller/pcie-rcar.c | 45 +++++++++++------------------- 1 file changed, 16 insertions(+), 29 deletions(-) diff --git a/drivers/pci/controller/pcie-rcar.c b/drivers/pci/controller/pcie-rcar.c index b8d6e86a5539..453c931aaf77 100644 --- a/drivers/pci/controller/pcie-rcar.c +++ b/drivers/pci/controller/pcie-rcar.c @@ -1014,16 +1014,16 @@ err_irq1: } static int rcar_pcie_inbound_ranges(struct rcar_pcie *pcie, - struct of_pci_range *range, + struct resource_entry *entry, int *index) { - u64 restype = range->flags; - u64 cpu_addr = range->cpu_addr; - u64 cpu_end = range->cpu_addr + range->size; - u64 pci_addr = range->pci_addr; + u64 restype = entry->res->flags; + u64 cpu_addr = entry->res->start; + u64 cpu_end = entry->res->end; + u64 pci_addr = entry->res->start - entry->offset; u32 flags = LAM_64BIT | LAR_ENABLE; u64 mask; - u64 size; + u64 size = resource_size(entry->res); int idx = *index; if (restype & IORESOURCE_PREFETCH) @@ -1037,9 +1037,7 @@ static int rcar_pcie_inbound_ranges(struct rcar_pcie *pcie, unsigned long nr_zeros = __ffs64(cpu_addr); u64 alignment = 1ULL << nr_zeros; - size = min(range->size, alignment); - } else { - size = range->size; + size = min(size, alignment); } /* Hardware supports max 4GiB inbound region */ size = min(size, 1ULL << 32); @@ -1078,30 +1076,19 @@ static int rcar_pcie_inbound_ranges(struct rcar_pcie *pcie, return 0; } -static int rcar_pcie_parse_map_dma_ranges(struct rcar_pcie *pcie, - struct device_node *np) +static int rcar_pcie_parse_map_dma_ranges(struct rcar_pcie *pcie) { - struct of_pci_range range; - struct of_pci_range_parser parser; - int index = 0; - int err; + struct pci_host_bridge *bridge = pci_host_bridge_from_priv(pcie); + struct resource_entry *entry; + int index = 0, err = 0; - if (of_pci_dma_range_parser_init(&parser, np)) - return -EINVAL; - - /* Get the dma-ranges from DT */ - for_each_of_pci_range(&parser, &range) { - u64 end = range.cpu_addr + range.size - 1; - - dev_dbg(pcie->dev, "0x%08x 0x%016llx..0x%016llx -> 0x%016llx\n", - range.flags, range.cpu_addr, end, range.pci_addr); - - err = rcar_pcie_inbound_ranges(pcie, &range, &index); + resource_list_for_each_entry(entry, &bridge->dma_ranges) { + err = rcar_pcie_inbound_ranges(pcie, entry, &index); if (err) - return err; + break; } - return 0; + return err; } static const struct of_device_id rcar_pcie_of_match[] = { @@ -1162,7 +1149,7 @@ static int rcar_pcie_probe(struct platform_device *pdev) goto err_unmap_msi_irqs; } - err = rcar_pcie_parse_map_dma_ranges(pcie, dev->of_node); + err = rcar_pcie_parse_map_dma_ranges(pcie); if (err) goto err_clk_disable; From 3b55809cf91f54fa5add4cc4cfed934565568ed7 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 28 Oct 2019 11:32:56 -0500 Subject: [PATCH 100/160] PCI: Make devm_of_pci_get_host_bridge_resources() static Now that all the PCI host drivers are using pci_parse_request_of_pci_ranges(), make devm_of_pci_get_host_bridge_resources() static. Signed-off-by: Rob Herring Signed-off-by: Lorenzo Pieralisi Cc: Bjorn Helgaas --- drivers/pci/of.c | 5 +---- drivers/pci/pci.h | 17 ----------------- 2 files changed, 1 insertion(+), 21 deletions(-) diff --git a/drivers/pci/of.c b/drivers/pci/of.c index 1dfde6f9e82d..81ceeaa6f1d5 100644 --- a/drivers/pci/of.c +++ b/drivers/pci/of.c @@ -236,7 +236,6 @@ void of_pci_check_probe_only(void) } EXPORT_SYMBOL_GPL(of_pci_check_probe_only); -#if defined(CONFIG_OF_ADDRESS) /** * devm_of_pci_get_host_bridge_resources() - Resource-managed parsing of PCI * host bridge resources from DT @@ -255,7 +254,7 @@ EXPORT_SYMBOL_GPL(of_pci_check_probe_only); * It returns zero if the range parsing has been successful or a standard error * value if it failed. */ -int devm_of_pci_get_host_bridge_resources(struct device *dev, +static int devm_of_pci_get_host_bridge_resources(struct device *dev, unsigned char busno, unsigned char bus_max, struct list_head *resources, struct list_head *ib_resources, @@ -390,8 +389,6 @@ failed: pci_free_resource_list(resources); return err; } -EXPORT_SYMBOL_GPL(devm_of_pci_get_host_bridge_resources); -#endif /* CONFIG_OF_ADDRESS */ #if IS_ENABLED(CONFIG_OF_IRQ) /** diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 6692c4fe4290..118a4974537b 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -630,23 +630,6 @@ static inline void pci_set_bus_of_node(struct pci_bus *bus) { } static inline void pci_release_bus_of_node(struct pci_bus *bus) { } #endif /* CONFIG_OF */ -#if defined(CONFIG_OF_ADDRESS) -int devm_of_pci_get_host_bridge_resources(struct device *dev, - unsigned char busno, unsigned char bus_max, - struct list_head *resources, - struct list_head *ib_resources, - resource_size_t *io_base); -#else -static inline int devm_of_pci_get_host_bridge_resources(struct device *dev, - unsigned char busno, unsigned char bus_max, - struct list_head *resources, - struct list_head *ib_resources, - resource_size_t *io_base) -{ - return -EINVAL; -} -#endif - #ifdef CONFIG_PCIEAER void pci_no_aer(void); void pci_aer_init(struct pci_dev *dev); From 1e4d401860268bbed124eb93a0010237bd8ed26c Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Sat, 16 Nov 2019 12:54:19 +0000 Subject: [PATCH 101/160] PCI: rockchip: Make some regulators non-optional The 0V9 and 1V8 supplies power the PCIe block in the SoC itself, and are thus fundamental to PCIe being usable at all. As such, it makes sense to treat them as non-optional and rely on dummy regulators if not explicitly described. Signed-off-by: Robin Murphy Signed-off-by: Lorenzo Pieralisi Reviewed-by: Mark Brown Reviewed-by: Andrew Murray --- drivers/pci/controller/pcie-rockchip-host.c | 69 ++++++++------------- 1 file changed, 25 insertions(+), 44 deletions(-) diff --git a/drivers/pci/controller/pcie-rockchip-host.c b/drivers/pci/controller/pcie-rockchip-host.c index ef8e677ce9d1..68525f8ac4d9 100644 --- a/drivers/pci/controller/pcie-rockchip-host.c +++ b/drivers/pci/controller/pcie-rockchip-host.c @@ -620,19 +620,13 @@ static int rockchip_pcie_parse_host_dt(struct rockchip_pcie *rockchip) dev_info(dev, "no vpcie3v3 regulator found\n"); } - rockchip->vpcie1v8 = devm_regulator_get_optional(dev, "vpcie1v8"); - if (IS_ERR(rockchip->vpcie1v8)) { - if (PTR_ERR(rockchip->vpcie1v8) != -ENODEV) - return PTR_ERR(rockchip->vpcie1v8); - dev_info(dev, "no vpcie1v8 regulator found\n"); - } + rockchip->vpcie1v8 = devm_regulator_get(dev, "vpcie1v8"); + if (IS_ERR(rockchip->vpcie1v8)) + return PTR_ERR(rockchip->vpcie1v8); - rockchip->vpcie0v9 = devm_regulator_get_optional(dev, "vpcie0v9"); - if (IS_ERR(rockchip->vpcie0v9)) { - if (PTR_ERR(rockchip->vpcie0v9) != -ENODEV) - return PTR_ERR(rockchip->vpcie0v9); - dev_info(dev, "no vpcie0v9 regulator found\n"); - } + rockchip->vpcie0v9 = devm_regulator_get(dev, "vpcie0v9"); + if (IS_ERR(rockchip->vpcie0v9)) + return PTR_ERR(rockchip->vpcie0v9); return 0; } @@ -658,27 +652,22 @@ static int rockchip_pcie_set_vpcie(struct rockchip_pcie *rockchip) } } - if (!IS_ERR(rockchip->vpcie1v8)) { - err = regulator_enable(rockchip->vpcie1v8); - if (err) { - dev_err(dev, "fail to enable vpcie1v8 regulator\n"); - goto err_disable_3v3; - } + err = regulator_enable(rockchip->vpcie1v8); + if (err) { + dev_err(dev, "fail to enable vpcie1v8 regulator\n"); + goto err_disable_3v3; } - if (!IS_ERR(rockchip->vpcie0v9)) { - err = regulator_enable(rockchip->vpcie0v9); - if (err) { - dev_err(dev, "fail to enable vpcie0v9 regulator\n"); - goto err_disable_1v8; - } + err = regulator_enable(rockchip->vpcie0v9); + if (err) { + dev_err(dev, "fail to enable vpcie0v9 regulator\n"); + goto err_disable_1v8; } return 0; err_disable_1v8: - if (!IS_ERR(rockchip->vpcie1v8)) - regulator_disable(rockchip->vpcie1v8); + regulator_disable(rockchip->vpcie1v8); err_disable_3v3: if (!IS_ERR(rockchip->vpcie3v3)) regulator_disable(rockchip->vpcie3v3); @@ -897,8 +886,7 @@ static int __maybe_unused rockchip_pcie_suspend_noirq(struct device *dev) rockchip_pcie_disable_clocks(rockchip); - if (!IS_ERR(rockchip->vpcie0v9)) - regulator_disable(rockchip->vpcie0v9); + regulator_disable(rockchip->vpcie0v9); return ret; } @@ -908,12 +896,10 @@ static int __maybe_unused rockchip_pcie_resume_noirq(struct device *dev) struct rockchip_pcie *rockchip = dev_get_drvdata(dev); int err; - if (!IS_ERR(rockchip->vpcie0v9)) { - err = regulator_enable(rockchip->vpcie0v9); - if (err) { - dev_err(dev, "fail to enable vpcie0v9 regulator\n"); - return err; - } + err = regulator_enable(rockchip->vpcie0v9); + if (err) { + dev_err(dev, "fail to enable vpcie0v9 regulator\n"); + return err; } err = rockchip_pcie_enable_clocks(rockchip); @@ -939,8 +925,7 @@ err_err_deinit_port: err_pcie_resume: rockchip_pcie_disable_clocks(rockchip); err_disable_0v9: - if (!IS_ERR(rockchip->vpcie0v9)) - regulator_disable(rockchip->vpcie0v9); + regulator_disable(rockchip->vpcie0v9); return err; } @@ -1081,10 +1066,8 @@ err_vpcie: regulator_disable(rockchip->vpcie12v); if (!IS_ERR(rockchip->vpcie3v3)) regulator_disable(rockchip->vpcie3v3); - if (!IS_ERR(rockchip->vpcie1v8)) - regulator_disable(rockchip->vpcie1v8); - if (!IS_ERR(rockchip->vpcie0v9)) - regulator_disable(rockchip->vpcie0v9); + regulator_disable(rockchip->vpcie1v8); + regulator_disable(rockchip->vpcie0v9); err_set_vpcie: rockchip_pcie_disable_clocks(rockchip); return err; @@ -1108,10 +1091,8 @@ static int rockchip_pcie_remove(struct platform_device *pdev) regulator_disable(rockchip->vpcie12v); if (!IS_ERR(rockchip->vpcie3v3)) regulator_disable(rockchip->vpcie3v3); - if (!IS_ERR(rockchip->vpcie1v8)) - regulator_disable(rockchip->vpcie1v8); - if (!IS_ERR(rockchip->vpcie0v9)) - regulator_disable(rockchip->vpcie0v9); + regulator_disable(rockchip->vpcie1v8); + regulator_disable(rockchip->vpcie0v9); return 0; } From 9d09e5a95c54d2b0cbb4ae00929edbcadc847c7f Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 15 Oct 2019 17:11:22 -0500 Subject: [PATCH 102/160] PCI: Fix typos Fix typos in drivers/pci. Comment and whitespace changes only. Signed-off-by: Bjorn Helgaas --- drivers/pci/hotplug/rpaphp_core.c | 4 ++-- drivers/pci/quirks.c | 1 - 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/pci/hotplug/rpaphp_core.c b/drivers/pci/hotplug/rpaphp_core.c index 18627bb21e9e..0c71b4a8f807 100644 --- a/drivers/pci/hotplug/rpaphp_core.c +++ b/drivers/pci/hotplug/rpaphp_core.c @@ -185,8 +185,8 @@ static int get_children_props(struct device_node *dn, const int **drc_indexes, /* Verify the existence of 'drc_name' and/or 'drc_type' within the - * current node. First obtain it's my-drc-index property. Next, - * obtain the DRC info from it's parent. Use the my-drc-index for + * current node. First obtain its my-drc-index property. Next, + * obtain the DRC info from its parent. Use the my-drc-index for * correlation, and obtain/validate the requested properties. */ diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 320255e5e8f8..cf14423e4fa0 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -4033,7 +4033,6 @@ static void quirk_fixed_dma_alias(struct pci_dev *dev) if (id) pci_add_dma_alias(dev, id->driver_data); } - DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ADAPTEC2, 0x0285, quirk_fixed_dma_alias); /* From f2c33ccacb2d4bbeae2a255a7ca0cbfd03017b7c Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Wed, 14 Aug 2019 01:06:55 +0000 Subject: [PATCH 103/160] PCI/PM: Always return devices to D0 when thawing pci_pm_thaw_noirq() is supposed to return the device to D0 and restore its configuration registers, but previously it only did that for devices whose drivers implemented the new power management ops. Hibernation, e.g., via "echo disk > /sys/power/state", involves freezing devices, creating a hibernation image, thawing devices, writing the image, and powering off. The fact that thawing did not return devices with legacy power management to D0 caused errors, e.g., in this path: pci_pm_thaw_noirq if (pci_has_legacy_pm_support(pci_dev)) # true for Mellanox VF driver return pci_legacy_resume_early(dev) # ... legacy PM skips the rest pci_set_power_state(pci_dev, PCI_D0) pci_restore_state(pci_dev) pci_pm_thaw if (pci_has_legacy_pm_support(pci_dev)) pci_legacy_resume drv->resume mlx4_resume ... pci_enable_msix_range ... if (dev->current_state != PCI_D0) # <--- return -EINVAL; which caused these warnings: mlx4_core a6d1:00:02.0: INTx is not supported in multi-function mode, aborting PM: dpm_run_callback(): pci_pm_thaw+0x0/0xd7 returns -95 PM: Device a6d1:00:02.0 failed to thaw: error -95 Return devices to D0 and restore config registers for all devices, not just those whose drivers support new power management. [bhelgaas: also call pci_restore_state() before pci_legacy_resume_early(), update comment, add stable tag, commit log] Link: https://lore.kernel.org/r/KU1P153MB016637CAEAD346F0AA8E3801BFAD0@KU1P153MB0166.APCP153.PROD.OUTLOOK.COM Signed-off-by: Dexuan Cui Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki Cc: stable@vger.kernel.org # v4.13+ --- drivers/pci/pci-driver.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index a8124e47bf6e..d4ac8ce8c1f9 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -1076,17 +1076,22 @@ static int pci_pm_thaw_noirq(struct device *dev) return error; } - if (pci_has_legacy_pm_support(pci_dev)) - return pci_legacy_resume_early(dev); - /* - * pci_restore_state() requires the device to be in D0 (because of MSI - * restoration among other things), so force it into D0 in case the - * driver's "freeze" callbacks put it into a low-power state directly. + * Both the legacy ->resume_early() and the new pm->thaw_noirq() + * callbacks assume the device has been returned to D0 and its + * config state has been restored. + * + * In addition, pci_restore_state() restores MSI-X state in MMIO + * space, which requires the device to be in D0, so return it to D0 + * in case the driver's "freeze" callbacks put it into a low-power + * state. */ pci_set_power_state(pci_dev, PCI_D0); pci_restore_state(pci_dev); + if (pci_has_legacy_pm_support(pci_dev)) + return pci_legacy_resume_early(dev); + if (drv && drv->pm && drv->pm->thaw_noirq) error = drv->pm->thaw_noirq(dev); From dc68b406783edf33af0e997fb98e9e048b4d46e8 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Mon, 14 Oct 2019 14:14:06 -0500 Subject: [PATCH 104/160] PCI/PM: Correct pci_pm_thaw_noirq() documentation According to the documentation, pci_pm_thaw_noirq() did not put the device into the full-power state and restore its standard configuration registers. This is incorrect, so update the documentation to match the code. Link: https://lore.kernel.org/r/20191014230016.240912-3-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- Documentation/power/pci.rst | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index 0e2ef7429304..1525c594d631 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -600,17 +600,17 @@ using the following PCI bus type's callbacks:: respectively. -The first of them, pci_pm_thaw_noirq(), is analogous to pci_pm_resume_noirq(), -but it doesn't put the device into the full power state and doesn't attempt to -restore its standard configuration registers. It also executes the device -driver's pm->thaw_noirq() callback, if defined, instead of pm->resume_noirq(). +The first of them, pci_pm_thaw_noirq(), is analogous to pci_pm_resume_noirq(). +It puts the device into the full power state and restores its standard +configuration registers. It also executes the device driver's pm->thaw_noirq() +callback, if defined, instead of pm->resume_noirq(). The pci_pm_thaw() routine is similar to pci_pm_resume(), but it runs the device driver's pm->thaw() callback instead of pm->resume(). It is executed asynchronously for different PCI devices that don't depend on each other in a known way. -The complete phase it the same as for system resume. +The complete phase is the same as for system resume. After saving the image, devices need to be powered down before the system can enter the target sleep state (ACPI S4 for ACPI-based systems). This is done in From ec6a75ef8e33fe33f963b916fd902c52a0be33ff Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 10 Oct 2019 16:54:36 -0500 Subject: [PATCH 105/160] PCI/PM: Clear PCIe PME Status even for legacy power management Previously, pci_pm_resume_noirq() cleared the PME Status bit in the Root Status register only if the device had no driver or the driver did not implement legacy power management. It should clear PME Status regardless of what sort of power management the driver supports, so do this before checking for legacy power management. This affects Root Ports and Root Complex Event Collectors, for which the usual driver is the PCIe portdrv, which implements new power management, so this change is just on principle, not to fix any actual defects. Fixes: a39bd851dccf ("PCI/PM: Clear PCIe PME Status bit in core, not PCIe port driver") Link: https://lore.kernel.org/r/20191014230016.240912-4-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/pci-driver.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index d4ac8ce8c1f9..0c3086793e4e 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -941,12 +941,11 @@ static int pci_pm_resume_noirq(struct device *dev) pci_pm_default_resume_early(pci_dev); pci_fixup_device(pci_fixup_resume_early, pci_dev); + pcie_pme_root_status_cleanup(pci_dev); if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_resume_early(dev); - pcie_pme_root_status_cleanup(pci_dev); - if (drv && drv->pm && drv->pm->resume_noirq) error = drv->pm->resume_noirq(dev); From f7b32a86e455b035dd45a334c203abf6fe118cf3 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Sat, 12 Oct 2019 17:15:57 -0500 Subject: [PATCH 106/160] PCI/PM: Run resume fixups before disabling wakeup events pci_pm_resume() and pci_pm_restore() call pci_pm_default_resume(), which runs resume fixups before disabling wakeup events: static void pci_pm_default_resume(struct pci_dev *pci_dev) { pci_fixup_device(pci_fixup_resume, pci_dev); pci_enable_wake(pci_dev, PCI_D0, false); } pci_pm_runtime_resume() does both of these, but in the opposite order: pci_enable_wake(pci_dev, PCI_D0, false); pci_fixup_device(pci_fixup_resume, pci_dev); We should always use the same ordering unless there's a reason to do otherwise. Change pci_pm_runtime_resume() to call pci_pm_default_resume() instead of open-coding this, so the fixups are always done before disabling wakeup events. pci_pm_default_resume() is called from pci_pm_runtime_resume(), which is under #ifdef CONFIG_PM. If SUSPEND and HIBERNATION are disabled, PM_SLEEP is disabled also, so move pci_pm_default_resume() from #ifdef CONFIG_PM_SLEEP to #ifdef CONFIG_PM. Link: https://lore.kernel.org/r/20191014230016.240912-5-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/pci-driver.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 0c3086793e4e..abee2a790a10 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -517,6 +517,12 @@ static int pci_restore_standard_config(struct pci_dev *pci_dev) return 0; } +static void pci_pm_default_resume(struct pci_dev *pci_dev) +{ + pci_fixup_device(pci_fixup_resume, pci_dev); + pci_enable_wake(pci_dev, PCI_D0, false); +} + #endif #ifdef CONFIG_PM_SLEEP @@ -645,12 +651,6 @@ static int pci_legacy_resume(struct device *dev) /* Auxiliary functions used by the new power management framework */ -static void pci_pm_default_resume(struct pci_dev *pci_dev) -{ - pci_fixup_device(pci_fixup_resume, pci_dev); - pci_enable_wake(pci_dev, PCI_D0, false); -} - static void pci_pm_default_suspend(struct pci_dev *pci_dev) { /* Disable non-bridge devices without PM support */ @@ -992,7 +992,6 @@ static int pci_pm_resume(struct device *dev) #ifdef CONFIG_HIBERNATE_CALLBACKS - /* * pcibios_pm_ops - provide arch-specific hooks when a PCI device is doing * a hibernate transition @@ -1345,8 +1344,7 @@ static int pci_pm_runtime_resume(struct device *dev) return 0; pci_fixup_device(pci_fixup_resume_early, pci_dev); - pci_enable_wake(pci_dev, PCI_D0, false); - pci_fixup_device(pci_fixup_resume, pci_dev); + pci_pm_default_resume(pci_dev); if (pm && pm->runtime_resume) rc = pm->runtime_resume(dev); From 6da2f2ccfd2deb81a63fc23a505ccd72de005c39 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Mon, 14 Oct 2019 13:46:50 -0500 Subject: [PATCH 107/160] PCI/PM: Make power management op coding style consistent Some of the power management ops use this style: struct device_driver *drv = dev->driver; if (drv && drv->pm && drv->pm->prepare(dev)) drv->pm->prepare(dev); while others use this: const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; if (pm && pm->runtime_resume) pm->runtime_resume(dev); Convert the first style to the second so they're all consistent. Remove local "error" variables when unnecessary. No functional change intended. Link: https://lore.kernel.org/r/20191014230016.240912-6-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/pci-driver.c | 76 +++++++++++++++++++--------------------- 1 file changed, 36 insertions(+), 40 deletions(-) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index abee2a790a10..dd70ab2519c9 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -679,11 +679,11 @@ static bool pci_has_legacy_pm_support(struct pci_dev *pci_dev) static int pci_pm_prepare(struct device *dev) { - struct device_driver *drv = dev->driver; struct pci_dev *pci_dev = to_pci_dev(dev); + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - if (drv && drv->pm && drv->pm->prepare) { - int error = drv->pm->prepare(dev); + if (pm && pm->prepare) { + int error = pm->prepare(dev); if (error < 0) return error; @@ -917,8 +917,7 @@ Fixup: static int pci_pm_resume_noirq(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct device_driver *drv = dev->driver; - int error = 0; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; if (dev_pm_may_skip_resume(dev)) return 0; @@ -946,17 +945,16 @@ static int pci_pm_resume_noirq(struct device *dev) if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_resume_early(dev); - if (drv && drv->pm && drv->pm->resume_noirq) - error = drv->pm->resume_noirq(dev); + if (pm && pm->resume_noirq) + return pm->resume_noirq(dev); - return error; + return 0; } static int pci_pm_resume(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - int error = 0; /* * This is necessary for the suspend error path in which resume is @@ -972,12 +970,12 @@ static int pci_pm_resume(struct device *dev) if (pm) { if (pm->resume) - error = pm->resume(dev); + return pm->resume(dev); } else { pci_pm_reenable_device(pci_dev); } - return error; + return 0; } #else /* !CONFIG_SUSPEND */ @@ -1037,16 +1035,16 @@ static int pci_pm_freeze(struct device *dev) static int pci_pm_freeze_noirq(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct device_driver *drv = dev->driver; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend_late(dev, PMSG_FREEZE); - if (drv && drv->pm && drv->pm->freeze_noirq) { + if (pm && pm->freeze_noirq) { int error; - error = drv->pm->freeze_noirq(dev); - suspend_report_result(drv->pm->freeze_noirq, error); + error = pm->freeze_noirq(dev); + suspend_report_result(pm->freeze_noirq, error); if (error) return error; } @@ -1065,8 +1063,8 @@ static int pci_pm_freeze_noirq(struct device *dev) static int pci_pm_thaw_noirq(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct device_driver *drv = dev->driver; - int error = 0; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + int error; if (pcibios_pm_ops.thaw_noirq) { error = pcibios_pm_ops.thaw_noirq(dev); @@ -1090,10 +1088,10 @@ static int pci_pm_thaw_noirq(struct device *dev) if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_resume_early(dev); - if (drv && drv->pm && drv->pm->thaw_noirq) - error = drv->pm->thaw_noirq(dev); + if (pm && pm->thaw_noirq) + return pm->thaw_noirq(dev); - return error; + return 0; } static int pci_pm_thaw(struct device *dev) @@ -1164,24 +1162,24 @@ static int pci_pm_poweroff_late(struct device *dev) static int pci_pm_poweroff_noirq(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct device_driver *drv = dev->driver; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; if (dev_pm_smart_suspend_and_suspended(dev)) return 0; - if (pci_has_legacy_pm_support(to_pci_dev(dev))) + if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend_late(dev, PMSG_HIBERNATE); - if (!drv || !drv->pm) { + if (!pm) { pci_fixup_device(pci_fixup_suspend_late, pci_dev); return 0; } - if (drv->pm->poweroff_noirq) { + if (pm->poweroff_noirq) { int error; - error = drv->pm->poweroff_noirq(dev); - suspend_report_result(drv->pm->poweroff_noirq, error); + error = pm->poweroff_noirq(dev); + suspend_report_result(pm->poweroff_noirq, error); if (error) return error; } @@ -1207,8 +1205,8 @@ static int pci_pm_poweroff_noirq(struct device *dev) static int pci_pm_restore_noirq(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct device_driver *drv = dev->driver; - int error = 0; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + int error; if (pcibios_pm_ops.restore_noirq) { error = pcibios_pm_ops.restore_noirq(dev); @@ -1222,17 +1220,16 @@ static int pci_pm_restore_noirq(struct device *dev) if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_resume_early(dev); - if (drv && drv->pm && drv->pm->restore_noirq) - error = drv->pm->restore_noirq(dev); + if (pm && pm->restore_noirq) + return pm->restore_noirq(dev); - return error; + return 0; } static int pci_pm_restore(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - int error = 0; /* * This is necessary for the hibernation error path in which restore is @@ -1248,12 +1245,12 @@ static int pci_pm_restore(struct device *dev) if (pm) { if (pm->restore) - error = pm->restore(dev); + return pm->restore(dev); } else { pci_pm_reenable_device(pci_dev); } - return error; + return 0; } #else /* !CONFIG_HIBERNATE_CALLBACKS */ @@ -1329,9 +1326,9 @@ static int pci_pm_runtime_suspend(struct device *dev) static int pci_pm_runtime_resume(struct device *dev) { - int rc = 0; struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + int error = 0; /* * Restoring config space is necessary even if the device is not bound @@ -1347,18 +1344,17 @@ static int pci_pm_runtime_resume(struct device *dev) pci_pm_default_resume(pci_dev); if (pm && pm->runtime_resume) - rc = pm->runtime_resume(dev); + error = pm->runtime_resume(dev); pci_dev->runtime_d3cold = false; - return rc; + return error; } static int pci_pm_runtime_idle(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; - int ret = 0; /* * If pci_dev->driver is not set (unbound), the device should @@ -1371,9 +1367,9 @@ static int pci_pm_runtime_idle(struct device *dev) return -ENOSYS; if (pm->runtime_idle) - ret = pm->runtime_idle(dev); + return pm->runtime_idle(dev); - return ret; + return 0; } static const struct dev_pm_ops pci_dev_pm_ops = { From 85a9b0507db2fbbfe7912dc3b33d322f200e2ca7 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 8 Oct 2019 15:28:00 -0500 Subject: [PATCH 108/160] PCI/PM: Note that PME can be generated from D0 Per PCIe r5.0 sec 7.5.2.1, PME may be generated from D0, so update Documentation/power/pci.rst to reflect that. Link: https://lore.kernel.org/r/20191016194450.68959-1-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- Documentation/power/pci.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index 1525c594d631..8a4abb7c7363 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -130,8 +130,8 @@ a full power-on reset sequence and the power-on defaults are restored to the device by hardware just as at initial power up. PCI devices supporting the PCI PM Spec can be programmed to generate PMEs -while in a low-power state (D1-D3), but they are not required to be capable -of generating PMEs from all supported low-power states. In particular, the +while in any power state (D0-D3), but they are not required to be capable +of generating PMEs from all supported power states. In particular, the capability of generating PMEs from D3cold is optional and depends on the presence of additional voltage (3.3Vaux) allowing the device to remain sufficiently active to generate a wakeup signal. From b64cf7a1711d7796fdac19a23bc63d378a6e7ed1 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 8 Oct 2019 15:25:23 -0500 Subject: [PATCH 109/160] PCI/PM: Wrap long lines in documentation Documentation/power/pci.rst is wrapped to fit in 80 columns, but directory structure changes made a few lines longer. Wrap them so they all fit in 80 columns again. Link: https://lore.kernel.org/r/20191014230016.240912-7-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- Documentation/power/pci.rst | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index 8a4abb7c7363..a90e82c70a3b 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -426,12 +426,12 @@ pm->runtime_idle() callback. 2.4. System-Wide Power Transitions ---------------------------------- There are a few different types of system-wide power transitions, described in -Documentation/driver-api/pm/devices.rst. Each of them requires devices to be handled -in a specific way and the PM core executes subsystem-level power management -callbacks for this purpose. They are executed in phases such that each phase -involves executing the same subsystem-level callback for every device belonging -to the given subsystem before the next phase begins. These phases always run -after tasks have been frozen. +Documentation/driver-api/pm/devices.rst. Each of them requires devices to be +handled in a specific way and the PM core executes subsystem-level power +management callbacks for this purpose. They are executed in phases such that +each phase involves executing the same subsystem-level callback for every device +belonging to the given subsystem before the next phase begins. These phases +always run after tasks have been frozen. 2.4.1. System Suspend ^^^^^^^^^^^^^^^^^^^^^ @@ -636,12 +636,12 @@ System restore requires a hibernation image to be loaded into memory and the pre-hibernation memory contents to be restored before the pre-hibernation system activity can be resumed. -As described in Documentation/driver-api/pm/devices.rst, the hibernation image is loaded -into memory by a fresh instance of the kernel, called the boot kernel, which in -turn is loaded and run by a boot loader in the usual way. After the boot kernel -has loaded the image, it needs to replace its own code and data with the code -and data of the "hibernated" kernel stored within the image, called the image -kernel. For this purpose all devices are frozen just like before creating +As described in Documentation/driver-api/pm/devices.rst, the hibernation image +is loaded into memory by a fresh instance of the kernel, called the boot kernel, +which in turn is loaded and run by a boot loader in the usual way. After the +boot kernel has loaded the image, it needs to replace its own code and data with +the code and data of the "hibernated" kernel stored within the image, called the +image kernel. For this purpose all devices are frozen just like before creating the image during hibernation, in the prepare, freeze, freeze_noirq @@ -691,8 +691,8 @@ controlling the runtime power management of their devices. At the time of this writing there are two ways to define power management callbacks for a PCI device driver, the recommended one, based on using a -dev_pm_ops structure described in Documentation/driver-api/pm/devices.rst, and the -"legacy" one, in which the .suspend(), .suspend_late(), .resume_early(), and +dev_pm_ops structure described in Documentation/driver-api/pm/devices.rst, and +the "legacy" one, in which the .suspend(), .suspend_late(), .resume_early(), and .resume() callbacks from struct pci_driver are used. The legacy approach, however, doesn't allow one to define runtime power management callbacks and is not really suitable for any new drivers. Therefore it is not covered by this From 6941a0c2bdedfb729fb166091e12d06e4fce177f Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Mon, 7 Oct 2019 07:55:18 -0500 Subject: [PATCH 110/160] PCI/PM: Use PCI dev_printk() wrappers for consistency Use the PCI dev_printk() wrappers for consistency with the rest of the PCI core. No functional change intended. Link: https://lore.kernel.org/r/20191017212851.54237-2-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/pci-driver.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index dd70ab2519c9..407b1df5ea7c 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -315,7 +315,8 @@ static long local_pci_probe(void *_ddi) * Probe function should return < 0 for failure, 0 for success * Treat values > 0 as success, but warn. */ - dev_warn(dev, "Driver probe function unexpectedly returned %d\n", rc); + pci_warn(pci_dev, "Driver probe function unexpectedly returned %d\n", + rc); return 0; } @@ -865,7 +866,7 @@ static int pci_pm_suspend_noirq(struct device *dev) pci_prepare_to_sleep(pci_dev); } - dev_dbg(dev, "PCI PM: Suspend power state: %s\n", + pci_dbg(pci_dev, "PCI PM: Suspend power state: %s\n", pci_power_name(pci_dev->current_state)); if (pci_dev->current_state == PCI_D0) { @@ -880,7 +881,7 @@ static int pci_pm_suspend_noirq(struct device *dev) } if (pci_dev->skip_bus_pm && pm_suspend_no_platform()) { - dev_dbg(dev, "PCI PM: Skipped\n"); + pci_dbg(pci_dev, "PCI PM: Skipped\n"); goto Fixup; } @@ -1295,11 +1296,11 @@ static int pci_pm_runtime_suspend(struct device *dev) * log level. */ if (error == -EBUSY || error == -EAGAIN) { - dev_dbg(dev, "can't suspend now (%ps returned %d)\n", + pci_dbg(pci_dev, "can't suspend now (%ps returned %d)\n", pm->runtime_suspend, error); return error; } else if (error) { - dev_err(dev, "can't suspend (%ps returned %d)\n", + pci_err(pci_dev, "can't suspend (%ps returned %d)\n", pm->runtime_suspend, error); return error; } From 12bcae44bf48595c71898330076576075590e15b Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Mon, 7 Oct 2019 07:52:28 -0500 Subject: [PATCH 111/160] PCI/PM: Use pci_WARN() to include device information Add and use pci_WARN() wrappers so warnings include device information. Link: https://lore.kernel.org/r/20191017212851.54237-3-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/pci-driver.c | 34 +++++++++++++++++----------------- include/linux/pci.h | 8 ++++++++ 2 files changed, 25 insertions(+), 17 deletions(-) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 407b1df5ea7c..5337cbbd69de 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -585,9 +585,9 @@ static int pci_legacy_suspend(struct device *dev, pm_message_t state) if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0 && pci_dev->current_state != PCI_UNKNOWN) { - WARN_ONCE(pci_dev->current_state != prev, - "PCI PM: Device state not saved by %pS\n", - drv->suspend); + pci_WARN_ONCE(pci_dev, pci_dev->current_state != prev, + "PCI PM: Device state not saved by %pS\n", + drv->suspend); } } @@ -612,9 +612,9 @@ static int pci_legacy_suspend_late(struct device *dev, pm_message_t state) if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0 && pci_dev->current_state != PCI_UNKNOWN) { - WARN_ONCE(pci_dev->current_state != prev, - "PCI PM: Device state not saved by %pS\n", - drv->suspend_late); + pci_WARN_ONCE(pci_dev, pci_dev->current_state != prev, + "PCI PM: Device state not saved by %pS\n", + drv->suspend_late); goto Fixup; } } @@ -670,8 +670,8 @@ static bool pci_has_legacy_pm_support(struct pci_dev *pci_dev) * supported as well. Drivers are supposed to support either the * former, or the latter, but not both at the same time. */ - WARN(ret && drv->driver.pm, "driver %s device %04x:%04x\n", - drv->name, pci_dev->vendor, pci_dev->device); + pci_WARN(pci_dev, ret && drv->driver.pm, "device %04x:%04x\n", + pci_dev->vendor, pci_dev->device); return ret; } @@ -794,9 +794,9 @@ static int pci_pm_suspend(struct device *dev) if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0 && pci_dev->current_state != PCI_UNKNOWN) { - WARN_ONCE(pci_dev->current_state != prev, - "PCI PM: State of device not saved by %pS\n", - pm->suspend); + pci_WARN_ONCE(pci_dev, pci_dev->current_state != prev, + "PCI PM: State of device not saved by %pS\n", + pm->suspend); } } @@ -842,9 +842,9 @@ static int pci_pm_suspend_noirq(struct device *dev) if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0 && pci_dev->current_state != PCI_UNKNOWN) { - WARN_ONCE(pci_dev->current_state != prev, - "PCI PM: State of device not saved by %pS\n", - pm->suspend_noirq); + pci_WARN_ONCE(pci_dev, pci_dev->current_state != prev, + "PCI PM: State of device not saved by %pS\n", + pm->suspend_noirq); goto Fixup; } } @@ -1311,9 +1311,9 @@ static int pci_pm_runtime_suspend(struct device *dev) if (pm && pm->runtime_suspend && !pci_dev->state_saved && pci_dev->current_state != PCI_D0 && pci_dev->current_state != PCI_UNKNOWN) { - WARN_ONCE(pci_dev->current_state != prev, - "PCI PM: State of device not saved by %pS\n", - pm->runtime_suspend); + pci_WARN_ONCE(pci_dev, pci_dev->current_state != prev, + "PCI PM: State of device not saved by %pS\n", + pm->runtime_suspend); return 0; } diff --git a/include/linux/pci.h b/include/linux/pci.h index f9088c89a534..4846306d521c 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -2400,4 +2400,12 @@ void pci_uevent_ers(struct pci_dev *pdev, enum pci_ers_result err_type); #define pci_info_ratelimited(pdev, fmt, arg...) \ dev_info_ratelimited(&(pdev)->dev, fmt, ##arg) +#define pci_WARN(pdev, condition, fmt, arg...) \ + WARN(condition, "%s %s: " fmt, \ + dev_driver_string(&(pdev)->dev), pci_name(pdev), ##arg) + +#define pci_WARN_ONCE(pdev, condition, fmt, arg...) \ + WARN_ONCE(condition, "%s %s: " fmt, \ + dev_driver_string(&(pdev)->dev), pci_name(pdev), ##arg) + #endif /* LINUX_PCI_H */ From 7e24bc347e57992d532bc2ed700209b0fc0a4bf5 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 23 Oct 2019 17:40:52 -0500 Subject: [PATCH 112/160] PCI/PM: Apply D2 delay as milliseconds, not microseconds PCI_PM_D2_DELAY is defined as 200, which is milliseconds, but previously we used udelay(), which only waited for 200 microseconds. Use msleep() instead so we wait the correct amount of time. See PCIe r5.0, sec 5.9. Link: https://lore.kernel.org/r/20191101204558.210235-2-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index a97e2571a527..4dfbde4f944e 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -886,7 +886,7 @@ static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state) if (state == PCI_D3hot || dev->current_state == PCI_D3hot) pci_dev_d3_sleep(dev); else if (state == PCI_D2 || dev->current_state == PCI_D2) - udelay(PCI_PM_D2_DELAY); + msleep(PCI_PM_D2_DELAY); pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr); dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK); From 993cc6d1bd3af734693a74a3ccc9445dd5b31f9c Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Mon, 28 Oct 2019 08:27:00 -0500 Subject: [PATCH 113/160] PCI/PM: Expand PM reset messages to mention D3hot (not just D3) pci_pm_reset() resets a device by putting it in D3hot and bringing it back to D0. Clarify related messages to mention "D3hot" explicitly instead of just "D3". Link: https://lore.kernel.org/r/20191101204558.210235-3-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 4dfbde4f944e..9378fe5fe475 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -4603,7 +4603,7 @@ static int pci_pm_reset(struct pci_dev *dev, int probe) pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr); pci_dev_d3_sleep(dev); - return pci_dev_wait(dev, "PM D3->D0", PCIE_RESET_READY_POLL_MS); + return pci_dev_wait(dev, "PM D3hot->D0", PCIE_RESET_READY_POLL_MS); } /** * pcie_wait_for_link - Wait until link is active or inactive From baef7f8e5e91f85ce7625c11370479f9f0778fae Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 16 Oct 2019 15:23:20 -0500 Subject: [PATCH 114/160] PCI/PM: Simplify pci_set_power_state() Check for the PCI_DEV_FLAGS_NO_D3 quirk early, before calling __pci_start_power_transition(). This way all the cases where we don't need to do anything at all are checked up front. This doesn't fix anything because if the caller requested D3hot or D3cold, __pci_start_power_transition() is a no-op. But calling it is pointless and makes the code harder to analyze. Link: https://lore.kernel.org/r/20191101204558.210235-4-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/pci.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 9378fe5fe475..ea98c77a6512 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1117,8 +1117,6 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state) if (dev->current_state == state) return 0; - __pci_start_power_transition(dev, state); - /* * This device is quirked not to be put into D3, so don't put it in * D3 @@ -1126,6 +1124,8 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state) if (state >= PCI_D3hot && (dev->dev_flags & PCI_DEV_FLAGS_NO_D3)) return 0; + __pci_start_power_transition(dev, state); + /* * To put device in D3cold, we put device into D3hot in native * way, then put device into D3cold with platform ops From 77b84bb306fd0d5d1d3a4bf1f2acf73dc971e372 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Fri, 1 Nov 2019 08:20:40 -0500 Subject: [PATCH 115/160] xen-platform: Convert to generic power management Convert xen-platform from the legacy PCI power management callbacks to the generic operations. This is one step towards removing support for the legacy PCI callbacks. The generic .resume_noirq() operation is called by pci_pm_resume_noirq() at the same point the legacy PCI .resume_early() callback was, so this patch should not change the xen-platform behavior. Link: https://lore.kernel.org/r/20191101204558.210235-5-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Acked-by: Rafael J. Wysocki Cc: Stefano Stabellini Cc: KarimAllah Ahmed --- drivers/xen/platform-pci.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/xen/platform-pci.c b/drivers/xen/platform-pci.c index 5e30602fdbad..59e85e408c23 100644 --- a/drivers/xen/platform-pci.c +++ b/drivers/xen/platform-pci.c @@ -74,7 +74,7 @@ static int xen_allocate_irq(struct pci_dev *pdev) "xen-platform-pci", pdev); } -static int platform_pci_resume(struct pci_dev *pdev) +static int platform_pci_resume(struct device *dev) { int err; @@ -83,7 +83,7 @@ static int platform_pci_resume(struct pci_dev *pdev) err = xen_set_callback_via(callback_via); if (err) { - dev_err(&pdev->dev, "platform_pci_resume failure!\n"); + dev_err(dev, "platform_pci_resume failure!\n"); return err; } return 0; @@ -168,13 +168,17 @@ static const struct pci_device_id platform_pci_tbl[] = { {0,} }; +static struct dev_pm_ops platform_pm_ops = { + .resume_noirq = platform_pci_resume, +}; + static struct pci_driver platform_driver = { .name = DRV_NAME, .probe = platform_pci_probe, .id_table = platform_pci_tbl, -#ifdef CONFIG_PM - .resume_early = platform_pci_resume, -#endif + .driver = { + .pm = &platform_pm_ops, + }, }; builtin_pci_driver(platform_driver); From 89cdbc3546354c359558a1809133902028c57da4 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 31 Oct 2019 17:53:04 -0500 Subject: [PATCH 116/160] PCI/PM: Remove unused pci_driver.resume_early() hook The struct pci_driver.resume_early() hook is one of the legacy PCI power management callbacks, and there are no remaining users of it. Remove it. Link: https://lore.kernel.org/r/20191101204558.210235-6-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- Documentation/power/pci.rst | 2 +- drivers/pci/pci-driver.c | 23 ++++++----------------- include/linux/pci.h | 2 -- 3 files changed, 7 insertions(+), 20 deletions(-) diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index a90e82c70a3b..ff7029b94068 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -692,7 +692,7 @@ controlling the runtime power management of their devices. At the time of this writing there are two ways to define power management callbacks for a PCI device driver, the recommended one, based on using a dev_pm_ops structure described in Documentation/driver-api/pm/devices.rst, and -the "legacy" one, in which the .suspend(), .suspend_late(), .resume_early(), and +the "legacy" one, in which the .suspend(), .suspend_late(), and .resume() callbacks from struct pci_driver are used. The legacy approach, however, doesn't allow one to define runtime power management callbacks and is not really suitable for any new drivers. Therefore it is not covered by this diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 5337cbbd69de..fc372c2d529a 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -630,15 +630,6 @@ Fixup: return 0; } -static int pci_legacy_resume_early(struct device *dev) -{ - struct pci_dev *pci_dev = to_pci_dev(dev); - struct pci_driver *drv = pci_dev->driver; - - return drv && drv->resume_early ? - drv->resume_early(pci_dev) : 0; -} - static int pci_legacy_resume(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); @@ -662,8 +653,7 @@ static void pci_pm_default_suspend(struct pci_dev *pci_dev) static bool pci_has_legacy_pm_support(struct pci_dev *pci_dev) { struct pci_driver *drv = pci_dev->driver; - bool ret = drv && (drv->suspend || drv->suspend_late || drv->resume - || drv->resume_early); + bool ret = drv && (drv->suspend || drv->suspend_late || drv->resume); /* * Legacy PM support is used by default, so warn if the new framework is @@ -944,7 +934,7 @@ static int pci_pm_resume_noirq(struct device *dev) pcie_pme_root_status_cleanup(pci_dev); if (pci_has_legacy_pm_support(pci_dev)) - return pci_legacy_resume_early(dev); + return 0; if (pm && pm->resume_noirq) return pm->resume_noirq(dev); @@ -1074,9 +1064,8 @@ static int pci_pm_thaw_noirq(struct device *dev) } /* - * Both the legacy ->resume_early() and the new pm->thaw_noirq() - * callbacks assume the device has been returned to D0 and its - * config state has been restored. + * The pm->thaw_noirq() callback assumes the device has been + * returned to D0 and its config state has been restored. * * In addition, pci_restore_state() restores MSI-X state in MMIO * space, which requires the device to be in D0, so return it to D0 @@ -1087,7 +1076,7 @@ static int pci_pm_thaw_noirq(struct device *dev) pci_restore_state(pci_dev); if (pci_has_legacy_pm_support(pci_dev)) - return pci_legacy_resume_early(dev); + return 0; if (pm && pm->thaw_noirq) return pm->thaw_noirq(dev); @@ -1219,7 +1208,7 @@ static int pci_pm_restore_noirq(struct device *dev) pci_fixup_device(pci_fixup_resume_early, pci_dev); if (pci_has_legacy_pm_support(pci_dev)) - return pci_legacy_resume_early(dev); + return 0; if (pm && pm->restore_noirq) return pm->restore_noirq(dev); diff --git a/include/linux/pci.h b/include/linux/pci.h index 4846306d521c..dd4596fc1208 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -806,7 +806,6 @@ struct module; * context, so it can sleep. * @suspend: Put device into low power state. * @suspend_late: Put device into low power state. - * @resume_early: Wake device from low power state. * @resume: Wake device from low power state. * (Please see Documentation/power/pci.rst for descriptions * of PCI Power Management and the related functions.) @@ -830,7 +829,6 @@ struct pci_driver { void (*remove)(struct pci_dev *dev); /* Device removed (NULL if not a hot-plug capable driver) */ int (*suspend)(struct pci_dev *dev, pm_message_t state); /* Device suspended */ int (*suspend_late)(struct pci_dev *dev, pm_message_t state); - int (*resume_early)(struct pci_dev *dev); int (*resume)(struct pci_dev *dev); /* Device woken up */ void (*shutdown)(struct pci_dev *dev); int (*sriov_configure)(struct pci_dev *dev, int num_vfs); /* On PF */ From 1a1daf097e21e544dd3e7c0ff620d78a9795fbf2 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 31 Oct 2019 17:37:54 -0500 Subject: [PATCH 117/160] PCI/PM: Remove unused pci_driver.suspend_late() hook The struct pci_driver.suspend_late() hook is one of the legacy PCI power management callbacks, and there are no remaining users of it. Remove it. Link: https://lore.kernel.org/r/20191101204558.210235-7-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- Documentation/power/pci.rst | 10 +++++----- drivers/pci/pci-driver.c | 22 +--------------------- include/linux/pci.h | 2 -- 3 files changed, 6 insertions(+), 28 deletions(-) diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index ff7029b94068..0924d29636ad 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -692,11 +692,11 @@ controlling the runtime power management of their devices. At the time of this writing there are two ways to define power management callbacks for a PCI device driver, the recommended one, based on using a dev_pm_ops structure described in Documentation/driver-api/pm/devices.rst, and -the "legacy" one, in which the .suspend(), .suspend_late(), and -.resume() callbacks from struct pci_driver are used. The legacy approach, -however, doesn't allow one to define runtime power management callbacks and is -not really suitable for any new drivers. Therefore it is not covered by this -document (refer to the source code to learn more about it). +the "legacy" one, in which the .suspend() and .resume() callbacks from struct +pci_driver are used. The legacy approach, however, doesn't allow one to define +runtime power management callbacks and is not really suitable for any new +drivers. Therefore it is not covered by this document (refer to the source code +to learn more about it). It is recommended that all PCI device drivers define a struct dev_pm_ops object containing pointers to power management (PM) callbacks that will be executed by diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index fc372c2d529a..e89fd90eaa93 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -599,32 +599,12 @@ static int pci_legacy_suspend(struct device *dev, pm_message_t state) static int pci_legacy_suspend_late(struct device *dev, pm_message_t state) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct pci_driver *drv = pci_dev->driver; - - if (drv && drv->suspend_late) { - pci_power_t prev = pci_dev->current_state; - int error; - - error = drv->suspend_late(pci_dev, state); - suspend_report_result(drv->suspend_late, error); - if (error) - return error; - - if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0 - && pci_dev->current_state != PCI_UNKNOWN) { - pci_WARN_ONCE(pci_dev, pci_dev->current_state != prev, - "PCI PM: Device state not saved by %pS\n", - drv->suspend_late); - goto Fixup; - } - } if (!pci_dev->state_saved) pci_save_state(pci_dev); pci_pm_set_unknown_state(pci_dev); -Fixup: pci_fixup_device(pci_fixup_suspend_late, pci_dev); return 0; @@ -653,7 +633,7 @@ static void pci_pm_default_suspend(struct pci_dev *pci_dev) static bool pci_has_legacy_pm_support(struct pci_dev *pci_dev) { struct pci_driver *drv = pci_dev->driver; - bool ret = drv && (drv->suspend || drv->suspend_late || drv->resume); + bool ret = drv && (drv->suspend || drv->resume); /* * Legacy PM support is used by default, so warn if the new framework is diff --git a/include/linux/pci.h b/include/linux/pci.h index dd4596fc1208..9b0e35e09874 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -805,7 +805,6 @@ struct module; * The remove function always gets called from process * context, so it can sleep. * @suspend: Put device into low power state. - * @suspend_late: Put device into low power state. * @resume: Wake device from low power state. * (Please see Documentation/power/pci.rst for descriptions * of PCI Power Management and the related functions.) @@ -828,7 +827,6 @@ struct pci_driver { int (*probe)(struct pci_dev *dev, const struct pci_device_id *id); /* New device inserted */ void (*remove)(struct pci_dev *dev); /* Device removed (NULL if not a hot-plug capable driver) */ int (*suspend)(struct pci_dev *dev, pm_message_t state); /* Device suspended */ - int (*suspend_late)(struct pci_dev *dev, pm_message_t state); int (*resume)(struct pci_dev *dev); /* Device woken up */ void (*shutdown)(struct pci_dev *dev); int (*sriov_configure)(struct pci_dev *dev, int num_vfs); /* On PF */ From 81cfa5908fd6fe610b7c47b742fe30d6d897ba0f Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 5 Nov 2019 11:13:43 +0100 Subject: [PATCH 118/160] PCI/PM: Move power state update away from pci_power_up() Move the invocation of pci_update_current_state() from pci_power_up() to pci_pm_default_resume_early(), which is the only caller of that function. Preparatory change, no functional impact. Link: https://lore.kernel.org/r/37482337.udjOGdOKNb@kreacher Signed-off-by: Rafael J. Wysocki Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg --- drivers/pci/pci-driver.c | 1 + drivers/pci/pci.c | 1 - 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index e89fd90eaa93..08d3bdbc8c04 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -531,6 +531,7 @@ static void pci_pm_default_resume(struct pci_dev *pci_dev) static void pci_pm_default_resume_early(struct pci_dev *pci_dev) { pci_power_up(pci_dev); + pci_update_current_state(pci_dev, PCI_D0); pci_restore_state(pci_dev); pci_pme_restore(pci_dev); } diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index ea98c77a6512..11b4d9274d96 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1148,7 +1148,6 @@ void pci_power_up(struct pci_dev *dev) { __pci_start_power_transition(dev, PCI_D0); pci_raw_set_power_state(dev, PCI_D0); - pci_update_current_state(dev, PCI_D0); } /** From adfac8f6b7396b408fa9a8f40ea41112bebb980f Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 5 Nov 2019 11:27:49 +0100 Subject: [PATCH 119/160] PCI/PM: Use pci_power_up() in pci_set_power_state() Make it explicitly clear that the code to put devices into D0 in pci_set_power_state() and in pci_pm_default_resume_early() is the same by making the latter use pci_power_up() for transitions into D0. Code rearrangement, no intentional functional impact. Link: https://lore.kernel.org/r/2520019.OZ1nXS5aSj@kreacher Signed-off-by: Rafael J. Wysocki Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg --- drivers/pci/pci.c | 25 +++++++++++++------------ drivers/pci/pci.h | 2 +- 2 files changed, 14 insertions(+), 13 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 11b4d9274d96..9f0977d0ac5a 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1032,6 +1032,16 @@ static void __pci_start_power_transition(struct pci_dev *dev, pci_power_t state) } } +/** + * pci_power_up - Put the given device into D0 + * @dev: PCI device to power up + */ +int pci_power_up(struct pci_dev *dev) +{ + __pci_start_power_transition(dev, PCI_D0); + return pci_raw_set_power_state(dev, PCI_D0); +} + /** * __pci_dev_set_current_state - Set current state of a PCI device * @dev: Device to handle @@ -1117,6 +1127,9 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state) if (dev->current_state == state) return 0; + if (state == PCI_D0) + return pci_power_up(dev); + /* * This device is quirked not to be put into D3, so don't put it in * D3 @@ -1124,8 +1137,6 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state) if (state >= PCI_D3hot && (dev->dev_flags & PCI_DEV_FLAGS_NO_D3)) return 0; - __pci_start_power_transition(dev, state); - /* * To put device in D3cold, we put device into D3hot in native * way, then put device into D3cold with platform ops @@ -1140,16 +1151,6 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state) } EXPORT_SYMBOL(pci_set_power_state); -/** - * pci_power_up - Put the given device into D0 forcibly - * @dev: PCI device to power up - */ -void pci_power_up(struct pci_dev *dev) -{ - __pci_start_power_transition(dev, PCI_D0); - pci_raw_set_power_state(dev, PCI_D0); -} - /** * pci_choose_state - Choose the power state of a PCI device * @dev: PCI device to be suspended diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 3f6947ee3324..7c68312c72f4 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -85,7 +85,7 @@ struct pci_platform_pm_ops { int pci_set_platform_pm(const struct pci_platform_pm_ops *ops); void pci_update_current_state(struct pci_dev *dev, pci_power_t state); void pci_refresh_power_state(struct pci_dev *dev); -void pci_power_up(struct pci_dev *dev); +int pci_power_up(struct pci_dev *dev); void pci_disable_enabled_device(struct pci_dev *dev); int pci_finish_runtime_suspend(struct pci_dev *dev); void pcie_clear_root_pme_status(struct pci_dev *dev); From dc2256b0735d03664a92a6cb94ea4e564dfa237b Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 5 Nov 2019 11:29:16 +0100 Subject: [PATCH 120/160] PCI/PM: Fold __pci_start_power_transition() into its caller Because pci_power_up() has become the only caller of __pci_start_power_transition(), there is no need for the latter to be a separate function any more, so fold it into the former, drop a redundant check and reduce the number of lines of code somewhat. Code rearrangement, no intentional functional impact. Link: https://lore.kernel.org/r/3458080.lsoDbfkST9@kreacher Signed-off-by: Rafael J. Wysocki Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg --- drivers/pci/pci.c | 50 ++++++++++++++++++----------------------------- 1 file changed, 19 insertions(+), 31 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 9f0977d0ac5a..56bc79e33286 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1002,43 +1002,31 @@ void pci_wakeup_bus(struct pci_bus *bus) pci_walk_bus(bus, pci_wakeup, NULL); } -/** - * __pci_start_power_transition - Start power transition of a PCI device - * @dev: PCI device to handle. - * @state: State to put the device into. - */ -static void __pci_start_power_transition(struct pci_dev *dev, pci_power_t state) -{ - if (state == PCI_D0) { - pci_platform_power_transition(dev, PCI_D0); - /* - * Mandatory power management transition delays, see - * PCI Express Base Specification Revision 2.0 Section - * 6.6.1: Conventional Reset. Do not delay for - * devices powered on/off by corresponding bridge, - * because have already delayed for the bridge. - */ - if (dev->runtime_d3cold) { - if (dev->d3cold_delay && !dev->imm_ready) - msleep(dev->d3cold_delay); - /* - * When powering on a bridge from D3cold, the - * whole hierarchy may be powered on into - * D0uninitialized state, resume them to give - * them a chance to suspend again - */ - pci_wakeup_bus(dev->subordinate); - } - } -} - /** * pci_power_up - Put the given device into D0 * @dev: PCI device to power up */ int pci_power_up(struct pci_dev *dev) { - __pci_start_power_transition(dev, PCI_D0); + pci_platform_power_transition(dev, PCI_D0); + + /* + * Mandatory power management transition delays, see PCI Express Base + * Specification Revision 2.0 Section 6.6.1: Conventional Reset. Do not + * delay for devices powered on/off by corresponding bridge, because + * have already delayed for the bridge. + */ + if (dev->runtime_d3cold) { + if (dev->d3cold_delay && !dev->imm_ready) + msleep(dev->d3cold_delay); + /* + * When powering on a bridge from D3cold, the whole hierarchy + * may be powered on into D0uninitialized state, resume them to + * give them a chance to suspend again + */ + pci_wakeup_bus(dev->subordinate); + } + return pci_raw_set_power_state(dev, PCI_D0); } From d6aa37cd04fdafaf31ae89691e537535df43ca78 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 5 Nov 2019 11:30:36 +0100 Subject: [PATCH 121/160] PCI/PM: Avoid exporting __pci_complete_power_transition() Notice that radeon_set_suspend(), which is the only caller of __pci_complete_power_transition() outside of pci.c, really only cares about the pci_platform_power_transition() invoked by it, so export the latter instead of it, update the radeon driver to call pci_platform_power_transition() directly and make __pci_complete_power_transition() static. Code rearrangement, no intentional functional impact. Link: https://lore.kernel.org/r/1731661.ykamz2Tiuf@kreacher Signed-off-by: Rafael J. Wysocki Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg --- drivers/pci/pci.c | 6 +++--- drivers/video/fbdev/aty/radeon_pm.c | 2 +- include/linux/pci.h | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 56bc79e33286..2f53234f9e91 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -963,7 +963,7 @@ void pci_refresh_power_state(struct pci_dev *dev) * @dev: PCI device to handle. * @state: State to put the device into. */ -static int pci_platform_power_transition(struct pci_dev *dev, pci_power_t state) +int pci_platform_power_transition(struct pci_dev *dev, pci_power_t state) { int error; @@ -979,6 +979,7 @@ static int pci_platform_power_transition(struct pci_dev *dev, pci_power_t state) return error; } +EXPORT_SYMBOL_GPL(pci_platform_power_transition); /** * pci_wakeup - Wake up a PCI device @@ -1061,7 +1062,7 @@ void pci_bus_set_current_state(struct pci_bus *bus, pci_power_t state) * * This function should not be called directly by device drivers. */ -int __pci_complete_power_transition(struct pci_dev *dev, pci_power_t state) +static int __pci_complete_power_transition(struct pci_dev *dev, pci_power_t state) { int ret; @@ -1073,7 +1074,6 @@ int __pci_complete_power_transition(struct pci_dev *dev, pci_power_t state) pci_bus_set_current_state(dev->subordinate, PCI_D3cold); return ret; } -EXPORT_SYMBOL_GPL(__pci_complete_power_transition); /** * pci_set_power_state - Set the power state of a PCI device diff --git a/drivers/video/fbdev/aty/radeon_pm.c b/drivers/video/fbdev/aty/radeon_pm.c index 2dc5703eac51..7c4483c7f313 100644 --- a/drivers/video/fbdev/aty/radeon_pm.c +++ b/drivers/video/fbdev/aty/radeon_pm.c @@ -2593,7 +2593,7 @@ static void radeon_set_suspend(struct radeonfb_info *rinfo, int suspend) * calling pci_set_power_state() */ radeonfb_whack_power_state(rinfo, PCI_D2); - __pci_complete_power_transition(rinfo->pdev, PCI_D2); + pci_platform_power_transition(rinfo->pdev, PCI_D2); } else { printk(KERN_DEBUG "radeonfb (%s): switching to D0 state...\n", pci_name(rinfo->pdev)); diff --git a/include/linux/pci.h b/include/linux/pci.h index 9b0e35e09874..86976cccdfe3 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1228,7 +1228,7 @@ struct pci_cap_saved_state *pci_find_saved_ext_cap(struct pci_dev *dev, int pci_add_cap_save_buffer(struct pci_dev *dev, char cap, unsigned int size); int pci_add_ext_cap_save_buffer(struct pci_dev *dev, u16 cap, unsigned int size); -int __pci_complete_power_transition(struct pci_dev *dev, pci_power_t state); +int pci_platform_power_transition(struct pci_dev *dev, pci_power_t state); int pci_set_power_state(struct pci_dev *dev, pci_power_t state); pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state); bool pci_pme_capable(struct pci_dev *dev, pci_power_t state); From 9c77e63bd8dcbb3b59294e9176c00c5fcab3c9c6 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 5 Nov 2019 17:32:08 +0100 Subject: [PATCH 122/160] PCI/PM: Fold __pci_complete_power_transition() into its caller Because pci_set_power_state() has become the only caller of __pci_complete_power_transition(), there is no need for the latter to be a separate function any more, so fold it into the former, drop a redundant check and reduce the number of lines of code somewhat. Code rearrangement, no intentional functional impact. Link: https://lore.kernel.org/r/15576968.k611qn3UU0@kreacher Signed-off-by: Rafael J. Wysocki Signed-off-by: Bjorn Helgaas Reviewed-by: Mika Westerberg --- drivers/pci/pci.c | 30 +++++++----------------------- 1 file changed, 7 insertions(+), 23 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 2f53234f9e91..37fca1e51d16 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1055,26 +1055,6 @@ void pci_bus_set_current_state(struct pci_bus *bus, pci_power_t state) pci_walk_bus(bus, __pci_dev_set_current_state, &state); } -/** - * __pci_complete_power_transition - Complete power transition of a PCI device - * @dev: PCI device to handle. - * @state: State to put the device into. - * - * This function should not be called directly by device drivers. - */ -static int __pci_complete_power_transition(struct pci_dev *dev, pci_power_t state) -{ - int ret; - - if (state <= PCI_D0) - return -EINVAL; - ret = pci_platform_power_transition(dev, state); - /* Power off the bridge may power off the whole hierarchy */ - if (!ret && state == PCI_D3cold) - pci_bus_set_current_state(dev->subordinate, PCI_D3cold); - return ret; -} - /** * pci_set_power_state - Set the power state of a PCI device * @dev: PCI device to handle. @@ -1132,10 +1112,14 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state) error = pci_raw_set_power_state(dev, state > PCI_D3hot ? PCI_D3hot : state); - if (!__pci_complete_power_transition(dev, state)) - error = 0; + if (pci_platform_power_transition(dev, state)) + return error; - return error; + /* Powering off a bridge may power off the whole hierarchy */ + if (state == PCI_D3cold) + pci_bus_set_current_state(dev->subordinate, PCI_D3cold); + + return 0; } EXPORT_SYMBOL(pci_set_power_state); From e43f15ea2f6d6858675fa1baa5cb624f17269af0 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Fri, 2 Aug 2019 18:47:22 -0500 Subject: [PATCH 123/160] PCI/PM: Decode D3cold power state correctly Use pci_power_name() to print pci_power_t correctly. This changes: "state 0" or "D0" to "D0" "state 1" or "D1" to "D1" "state 2" or "D2" to "D2" "state 3" or "D3" to "D3hot" "state 4" or "D4" to "D3cold" Changes dmesg logging only, no other functional change intended. Link: https://lore.kernel.org/r/20190822200551.129039-3-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki Reviewed-by: Keith Busch Reviewed-by: Mika Westerberg --- drivers/pci/pci.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 37fca1e51d16..3376c663035d 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -834,14 +834,16 @@ static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state) return -EINVAL; /* - * Validate current state: - * Can enter D0 from any state, but if we can only go deeper - * to sleep if we're already in a low power state + * Validate transition: We can enter D0 from any state, but if + * we're already in a low-power state, we can only go deeper. E.g., + * we can go from D1 to D3, but we can't go directly from D3 to D1; + * we'd have to go from D3 to D0, then to D1. */ if (state != PCI_D0 && dev->current_state <= PCI_D3cold && dev->current_state > state) { - pci_err(dev, "invalid power transition (from state %d to %d)\n", - dev->current_state, state); + pci_err(dev, "invalid power transition (from %s to %s)\n", + pci_power_name(dev->current_state), + pci_power_name(state)); return -EINVAL; } @@ -891,8 +893,9 @@ static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state) pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr); dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK); if (dev->current_state != state) - pci_info_ratelimited(dev, "Refused to change power state, currently in D%d\n", - dev->current_state); + pci_info_ratelimited(dev, "refused to change power state from %s to %s\n", + pci_power_name(dev->current_state), + pci_power_name(state)); /* * According to section 5.4.1 of the "PCI BUS POWER MANAGEMENT From 327ccbbcc1497bff6d6f543c1f37c43a1653671f Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 1 Aug 2019 11:50:56 -0500 Subject: [PATCH 124/160] PCI/PM: Return error when changing power state from D3cold pci_raw_set_power_state() uses the Power Management capability to change a device's power state. The capability is in config space, which is accessible in D0, D1, D2, and D3hot, but not in D3cold. If we call pci_raw_set_power_state() on a device that's in D3cold, config reads fail and return ~0 data, which we erroneously interpreted as "the device is in D3hot", leading to messages like this: pcieport 0000:03:00.0: Refused to change power state, currently in D3 The PCI_PM_CTRL has several RsvdP fields, so ~0 is never a valid register value. If we get that value, print a more informative message and return an error. Changing the power state of a device from D3cold must be done by a platform power management method or some other non-config space mechanism. Link: https://lore.kernel.org/r/20190822200551.129039-4-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki Reviewed-by: Keith Busch Reviewed-by: Mika Westerberg --- drivers/pci/pci.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 3376c663035d..d6ff1a98bd8d 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -853,6 +853,12 @@ static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state) return -EIO; pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr); + if (pmcsr == (u16) ~0) { + pci_err(dev, "can't change power state from %s to %s (config space inaccessible)\n", + pci_power_name(dev->current_state), + pci_power_name(state)); + return -EIO; + } /* * If we're (effectively) in D3, force entire word to 0. From 4827d63891b6a839dac49c6ab62e61c4b011c4f2 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Tue, 12 Nov 2019 12:16:16 +0300 Subject: [PATCH 125/160] PCI/PM: Add pcie_wait_for_link_delay() Add pcie_wait_for_link_delay(). Similar to pcie_wait_for_link() but allows passing custom activation delay in milliseconds. Link: https://lore.kernel.org/r/20191112091617.70282-2-mika.westerberg@linux.intel.com Signed-off-by: Mika Westerberg Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/pci.c | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index d6ff1a98bd8d..11e4b495cfac 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -4586,14 +4586,17 @@ static int pci_pm_reset(struct pci_dev *dev, int probe) return pci_dev_wait(dev, "PM D3hot->D0", PCIE_RESET_READY_POLL_MS); } + /** - * pcie_wait_for_link - Wait until link is active or inactive + * pcie_wait_for_link_delay - Wait until link is active or inactive * @pdev: Bridge device * @active: waiting for active or inactive? + * @delay: Delay to wait after link has become active (in ms) * * Use this to wait till link becomes active or inactive. */ -bool pcie_wait_for_link(struct pci_dev *pdev, bool active) +static bool pcie_wait_for_link_delay(struct pci_dev *pdev, bool active, + int delay) { int timeout = 1000; bool ret; @@ -4630,13 +4633,25 @@ bool pcie_wait_for_link(struct pci_dev *pdev, bool active) timeout -= 10; } if (active && ret) - msleep(100); + msleep(delay); else if (ret != active) pci_info(pdev, "Data Link Layer Link Active not %s in 1000 msec\n", active ? "set" : "cleared"); return ret == active; } +/** + * pcie_wait_for_link - Wait until link is active or inactive + * @pdev: Bridge device + * @active: waiting for active or inactive? + * + * Use this to wait till link becomes active or inactive. + */ +bool pcie_wait_for_link(struct pci_dev *pdev, bool active) +{ + return pcie_wait_for_link_delay(pdev, active, 100); +} + void pci_reset_secondary_bus(struct pci_dev *dev) { u16 ctrl; From ad9001f2f41198784b0423646450ba2cb24793a3 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Tue, 12 Nov 2019 12:16:17 +0300 Subject: [PATCH 126/160] PCI/PM: Add missing link delays required by the PCIe spec Currently Linux does not follow PCIe spec regarding the required delays after reset. A concrete example is a Thunderbolt add-in-card that consists of a PCIe switch and two PCIe endpoints: +-1b.0-[01-6b]----00.0-[02-6b]--+-00.0-[03]----00.0 TBT controller +-01.0-[04-36]-- DS hotplug port +-02.0-[37]----00.0 xHCI controller \-04.0-[38-6b]-- DS hotplug port The root port (1b.0) and the PCIe switch downstream ports are all PCIe Gen3 so they support 8GT/s link speeds. We wait for the PCIe hierarchy to enter D3cold (runtime): pcieport 0000:00:1b.0: power state changed by ACPI to D3cold When it wakes up from D3cold, according to the PCIe 5.0 section 5.8 the PCIe switch is put to reset and its power is re-applied. This means that we must follow the rules in PCIe 5.0 section 6.6.1. For the PCIe Gen3 ports we are dealing with here, the following applies: With a Downstream Port that supports Link speeds greater than 5.0 GT/s, software must wait a minimum of 100 ms after Link training completes before sending a Configuration Request to the device immediately below that Port. Software can determine when Link training completes by polling the Data Link Layer Link Active bit or by setting up an associated interrupt (see Section 6.7.3.3). Translating this into the above topology we would need to do this (DLLLA stands for Data Link Layer Link Active): 0000:00:1b.0: wait for 100 ms after DLLLA is set before access to 0000:01:00.0 0000:02:00.0: wait for 100 ms after DLLLA is set before access to 0000:03:00.0 0000:02:02.0: wait for 100 ms after DLLLA is set before access to 0000:37:00.0 I've instrumented the kernel with some additional logging so we can see the actual delays performed: pcieport 0000:00:1b.0: power state changed by ACPI to D0 pcieport 0000:00:1b.0: waiting for D3cold delay of 100 ms pcieport 0000:00:1b.0: waiting for D3hot delay of 10 ms pcieport 0000:02:01.0: waiting for D3hot delay of 10 ms pcieport 0000:02:04.0: waiting for D3hot delay of 10 ms For the switch upstream port (01:00.0 reachable through 00:1b.0 root port) we wait for 100 ms but not taking into account the DLLLA requirement. We then wait 10 ms for D3hot -> D0 transition of the root port and the two downstream hotplug ports. This means that we deviate from what the spec requires. Performing the same check for system sleep (s2idle) transitions it turns out to be even worse. None of the mandatory delays are performed. If this would be S3 instead of s2idle then according to PCI FW spec 3.2 section 4.6.8. there is a specific _DSM that allows the OS to skip the delays but this platform does not provide the _DSM and does not go to S3 anyway so no firmware is involved that could already handle these delays. On this particular platform these delays are not actually needed because there is an additional delay as part of the ACPI power resource that is used to turn on power to the hierarchy but since that additional delay is not required by any of standards (PCIe, ACPI) it is not present in the Intel Ice Lake, for example where missing the mandatory delays causes pciehp to start tearing down the stack too early (links are not yet trained). Below is an example how it looks like when this happens: pcieport 0000:83:04.0: pciehp: Slot(4): Card not present pcieport 0000:87:04.0: PME# disabled pcieport 0000:83:04.0: pciehp: pciehp_unconfigure_device: domain:bus:dev = 0000:86:00 pcieport 0000:86:00.0: Refused to change power state, currently in D3 pcieport 0000:86:00.0: restoring config space at offset 0x3c (was 0xffffffff, writing 0x201ff) pcieport 0000:86:00.0: restoring config space at offset 0x38 (was 0xffffffff, writing 0x0) ... There is also one reported case (see the bugzilla link below) where the missing delay causes xHCI on a Titan Ridge controller fail to runtime resume when USB-C dock is plugged. This does not involve pciehp but instead the PCI core fails to runtime resume the xHCI device: pcieport 0000:04:02.0: restoring config space at offset 0xc (was 0x10000, writing 0x10020) pcieport 0000:04:02.0: restoring config space at offset 0x4 (was 0x100000, writing 0x100406) xhci_hcd 0000:39:00.0: Refused to change power state, currently in D3 xhci_hcd 0000:39:00.0: restoring config space at offset 0x3c (was 0xffffffff, writing 0x1ff) xhci_hcd 0000:39:00.0: restoring config space at offset 0x38 (was 0xffffffff, writing 0x0) ... Add a new function pci_bridge_wait_for_secondary_bus() that is called on PCI core resume and runtime resume paths accordingly if the bridge entered D3cold (and thus went through reset). This is second attempt to add the missing delays. The previous solution in c2bf1fc212f7 ("PCI: Add missing link delays required by the PCIe spec") was reverted because of two issues it caused: 1. One system become unresponsive after S3 resume due to PME service spinning in pcie_pme_work_fn(). The root port in question reports that the xHCI sent PME but the xHCI device itself does not have PME status set. The PME status bit is never cleared in the root port resulting the indefinite loop in pcie_pme_work_fn(). 2. Slows down resume if the root/downstream port does not support Data Link Layer Active Reporting because pcie_wait_for_link_delay() waits 1100 ms in that case. This version should avoid the above issues because we restrict the delay to happen only if the port went into D3cold. Link: https://lore.kernel.org/linux-pci/SL2P216MB01878BBCD75F21D882AEEA2880C60@SL2P216MB0187.KORP216.PROD.OUTLOOK.COM/ Link: https://bugzilla.kernel.org/show_bug.cgi?id=203885 Link: https://lore.kernel.org/r/20191112091617.70282-3-mika.westerberg@linux.intel.com Reported-by: Kai-Heng Feng Tested-by: Kai-Heng Feng Signed-off-by: Mika Westerberg Signed-off-by: Bjorn Helgaas --- drivers/pci/pci-driver.c | 11 +++- drivers/pci/pci.c | 128 +++++++++++++++++++++++++++++++++++++-- drivers/pci/pci.h | 1 + 3 files changed, 133 insertions(+), 7 deletions(-) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 08d3bdbc8c04..0454ca0e4e3f 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -890,6 +890,8 @@ static int pci_pm_resume_noirq(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + pci_power_t prev_state = pci_dev->current_state; + bool skip_bus_pm = pci_dev->skip_bus_pm; if (dev_pm_may_skip_resume(dev)) return 0; @@ -908,12 +910,15 @@ static int pci_pm_resume_noirq(struct device *dev) * configuration here and attempting to put them into D0 again is * pointless, so avoid doing that. */ - if (!(pci_dev->skip_bus_pm && pm_suspend_no_platform())) + if (!(skip_bus_pm && pm_suspend_no_platform())) pci_pm_default_resume_early(pci_dev); pci_fixup_device(pci_fixup_resume_early, pci_dev); pcie_pme_root_status_cleanup(pci_dev); + if (!skip_bus_pm && prev_state == PCI_D3cold) + pci_bridge_wait_for_secondary_bus(pci_dev); + if (pci_has_legacy_pm_support(pci_dev)) return 0; @@ -1299,6 +1304,7 @@ static int pci_pm_runtime_resume(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + pci_power_t prev_state = pci_dev->current_state; int error = 0; /* @@ -1314,6 +1320,9 @@ static int pci_pm_runtime_resume(struct device *dev) pci_fixup_device(pci_fixup_resume_early, pci_dev); pci_pm_default_resume(pci_dev); + if (prev_state == PCI_D3cold) + pci_bridge_wait_for_secondary_bus(pci_dev); + if (pm && pm->runtime_resume) error = pm->runtime_resume(dev); diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 11e4b495cfac..b9d91370856a 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1021,14 +1021,11 @@ int pci_power_up(struct pci_dev *dev) pci_platform_power_transition(dev, PCI_D0); /* - * Mandatory power management transition delays, see PCI Express Base - * Specification Revision 2.0 Section 6.6.1: Conventional Reset. Do not - * delay for devices powered on/off by corresponding bridge, because - * have already delayed for the bridge. + * Mandatory power management transition delays are handled in + * pci_pm_resume_noirq() and pci_pm_runtime_resume() of the + * corresponding bridge. */ if (dev->runtime_d3cold) { - if (dev->d3cold_delay && !dev->imm_ready) - msleep(dev->d3cold_delay); /* * When powering on a bridge from D3cold, the whole hierarchy * may be powered on into D0uninitialized state, resume them to @@ -4652,6 +4649,125 @@ bool pcie_wait_for_link(struct pci_dev *pdev, bool active) return pcie_wait_for_link_delay(pdev, active, 100); } +/* + * Find maximum D3cold delay required by all the devices on the bus. The + * spec says 100 ms, but firmware can lower it and we allow drivers to + * increase it as well. + * + * Called with @pci_bus_sem locked for reading. + */ +static int pci_bus_max_d3cold_delay(const struct pci_bus *bus) +{ + const struct pci_dev *pdev; + int min_delay = 100; + int max_delay = 0; + + list_for_each_entry(pdev, &bus->devices, bus_list) { + if (pdev->d3cold_delay < min_delay) + min_delay = pdev->d3cold_delay; + if (pdev->d3cold_delay > max_delay) + max_delay = pdev->d3cold_delay; + } + + return max(min_delay, max_delay); +} + +/** + * pci_bridge_wait_for_secondary_bus - Wait for secondary bus to be accessible + * @dev: PCI bridge + * + * Handle necessary delays before access to the devices on the secondary + * side of the bridge are permitted after D3cold to D0 transition. + * + * For PCIe this means the delays in PCIe 5.0 section 6.6.1. For + * conventional PCI it means Tpvrh + Trhfa specified in PCI 3.0 section + * 4.3.2. + */ +void pci_bridge_wait_for_secondary_bus(struct pci_dev *dev) +{ + struct pci_dev *child; + int delay; + + if (pci_dev_is_disconnected(dev)) + return; + + if (!pci_is_bridge(dev) || !dev->bridge_d3) + return; + + down_read(&pci_bus_sem); + + /* + * We only deal with devices that are present currently on the bus. + * For any hot-added devices the access delay is handled in pciehp + * board_added(). In case of ACPI hotplug the firmware is expected + * to configure the devices before OS is notified. + */ + if (!dev->subordinate || list_empty(&dev->subordinate->devices)) { + up_read(&pci_bus_sem); + return; + } + + /* Take d3cold_delay requirements into account */ + delay = pci_bus_max_d3cold_delay(dev->subordinate); + if (!delay) { + up_read(&pci_bus_sem); + return; + } + + child = list_first_entry(&dev->subordinate->devices, struct pci_dev, + bus_list); + up_read(&pci_bus_sem); + + /* + * Conventional PCI and PCI-X we need to wait Tpvrh + Trhfa before + * accessing the device after reset (that is 1000 ms + 100 ms). In + * practice this should not be needed because we don't do power + * management for them (see pci_bridge_d3_possible()). + */ + if (!pci_is_pcie(dev)) { + pci_dbg(dev, "waiting %d ms for secondary bus\n", 1000 + delay); + msleep(1000 + delay); + return; + } + + /* + * For PCIe downstream and root ports that do not support speeds + * greater than 5 GT/s need to wait minimum 100 ms. For higher + * speeds (gen3) we need to wait first for the data link layer to + * become active. + * + * However, 100 ms is the minimum and the PCIe spec says the + * software must allow at least 1s before it can determine that the + * device that did not respond is a broken device. There is + * evidence that 100 ms is not always enough, for example certain + * Titan Ridge xHCI controller does not always respond to + * configuration requests if we only wait for 100 ms (see + * https://bugzilla.kernel.org/show_bug.cgi?id=203885). + * + * Therefore we wait for 100 ms and check for the device presence. + * If it is still not present give it an additional 100 ms. + */ + if (!pcie_downstream_port(dev)) + return; + + if (pcie_get_speed_cap(dev) <= PCIE_SPEED_5_0GT) { + pci_dbg(dev, "waiting %d ms for downstream link\n", delay); + msleep(delay); + } else { + pci_dbg(dev, "waiting %d ms for downstream link, after activation\n", + delay); + if (!pcie_wait_for_link_delay(dev, true, delay)) { + /* Did not train, no need to wait any further */ + return; + } + } + + if (!pci_device_is_present(child)) { + pci_dbg(child, "waiting additional %d ms to become accessible\n", delay); + msleep(delay); + } +} + void pci_reset_secondary_bus(struct pci_dev *dev) { u16 ctrl; diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 7c68312c72f4..bf3c7e5de6fa 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -104,6 +104,7 @@ void pci_allocate_cap_save_buffers(struct pci_dev *dev); void pci_free_cap_save_buffers(struct pci_dev *dev); bool pci_bridge_d3_possible(struct pci_dev *dev); void pci_bridge_d3_update(struct pci_dev *dev); +void pci_bridge_wait_for_secondary_bus(struct pci_dev *dev); static inline void pci_wakeup_event(struct pci_dev *dev) { From bae26849372b83c65da73d19ff58e987d70e6600 Mon Sep 17 00:00:00 2001 From: Vidya Sagar Date: Wed, 20 Nov 2019 10:47:42 +0530 Subject: [PATCH 127/160] PCI/PM: Move pci_dev_wait() definition earlier Move the definition of pci_dev_wait() above pci_power_up() so that it can be called from the latter with no change in functionality. This is a pure code move with no functional change. Link: https://lore.kernel.org/r/20191120051743.23124-1-vidyas@nvidia.com Signed-off-by: Vidya Sagar Signed-off-by: Bjorn Helgaas --- drivers/pci/pci.c | 82 +++++++++++++++++++++++------------------------ 1 file changed, 41 insertions(+), 41 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index b9d91370856a..d6b44ae8baa4 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1012,6 +1012,47 @@ void pci_wakeup_bus(struct pci_bus *bus) pci_walk_bus(bus, pci_wakeup, NULL); } +static int pci_dev_wait(struct pci_dev *dev, char *reset_type, int timeout) +{ + int delay = 1; + u32 id; + + /* + * After reset, the device should not silently discard config + * requests, but it may still indicate that it needs more time by + * responding to them with CRS completions. The Root Port will + * generally synthesize ~0 data to complete the read (except when + * CRS SV is enabled and the read was for the Vendor ID; in that + * case it synthesizes 0x0001 data). + * + * Wait for the device to return a non-CRS completion. Read the + * Command register instead of Vendor ID so we don't have to + * contend with the CRS SV value. + */ + pci_read_config_dword(dev, PCI_COMMAND, &id); + while (id == ~0) { + if (delay > timeout) { + pci_warn(dev, "not ready %dms after %s; giving up\n", + delay - 1, reset_type); + return -ENOTTY; + } + + if (delay > 1000) + pci_info(dev, "not ready %dms after %s; waiting\n", + delay - 1, reset_type); + + msleep(delay); + delay *= 2; + pci_read_config_dword(dev, PCI_COMMAND, &id); + } + + if (delay > 1000) + pci_info(dev, "ready %dms after %s\n", delay - 1, + reset_type); + + return 0; +} + /** * pci_power_up - Put the given device into D0 * @dev: PCI device to power up @@ -4406,47 +4447,6 @@ int pci_wait_for_pending_transaction(struct pci_dev *dev) } EXPORT_SYMBOL(pci_wait_for_pending_transaction); -static int pci_dev_wait(struct pci_dev *dev, char *reset_type, int timeout) -{ - int delay = 1; - u32 id; - - /* - * After reset, the device should not silently discard config - * requests, but it may still indicate that it needs more time by - * responding to them with CRS completions. The Root Port will - * generally synthesize ~0 data to complete the read (except when - * CRS SV is enabled and the read was for the Vendor ID; in that - * case it synthesizes 0x0001 data). - * - * Wait for the device to return a non-CRS completion. Read the - * Command register instead of Vendor ID so we don't have to - * contend with the CRS SV value. - */ - pci_read_config_dword(dev, PCI_COMMAND, &id); - while (id == ~0) { - if (delay > timeout) { - pci_warn(dev, "not ready %dms after %s; giving up\n", - delay - 1, reset_type); - return -ENOTTY; - } - - if (delay > 1000) - pci_info(dev, "not ready %dms after %s; waiting\n", - delay - 1, reset_type); - - msleep(delay); - delay *= 2; - pci_read_config_dword(dev, PCI_COMMAND, &id); - } - - if (delay > 1000) - pci_info(dev, "ready %dms after %s\n", delay - 1, - reset_type); - - return 0; -} - /** * pcie_has_flr - check if a device supports function level resets * @dev: device to check From 4257ac5acdee880ed2251278199b3569dfa3dc49 Mon Sep 17 00:00:00 2001 From: Krzysztof Wilczynski Date: Mon, 30 Sep 2019 17:48:09 -0500 Subject: [PATCH 128/160] x86/PCI: Add NumaChip SPDX GPL-2.0 to replace COPYING boilerplate Add SPDX GPL-2.0 to numachip.c, which referred to the kernel default "COPYING" file, which specifies GPL version 2. Remove the boilerplate language referring to the GPL and "COPYING", relying on the assertion in b24413180f56 ("License cleanup: add SPDX GPL-2.0 license identifier to files with no license") that the SPDX identifier may be used instead of the full boilerplate text. [bhelgaas: split to separate patch] Link: https://lore.kernel.org/r/20190828135322.10370-1-kw@linux.com Signed-off-by: Krzysztof Wilczynski Signed-off-by: Bjorn Helgaas --- arch/x86/pci/numachip.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/arch/x86/pci/numachip.c b/arch/x86/pci/numachip.c index 2e565e65c893..01a085d9135a 100644 --- a/arch/x86/pci/numachip.c +++ b/arch/x86/pci/numachip.c @@ -1,8 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0 /* - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * * Numascale NumaConnect-specific PCI code * * Copyright (C) 2012 Numascale AS. All rights reserved. From 65e3c803e7a49bea983c63d145e15bdb274faf27 Mon Sep 17 00:00:00 2001 From: Krzysztof Wilczynski Date: Wed, 28 Aug 2019 15:53:22 +0200 Subject: [PATCH 129/160] x86/PCI: Correct SPDX comment style Change: drivers/pci/controller/pcie-cadence.h drivers/pci/controller/pcie-rockchip.h to use the correct SPDX comment style per section 2 of Documentation/process/license-rules.rst. These resolve the following checkpatch.pl warning: WARNING: Missing or malformed SPDX-License-Identifier tag in line 1 [bhelgaas: split to separate patch] Link: https://lore.kernel.org/r/20190828135322.10370-1-kw@linux.com Signed-off-by: Krzysztof Wilczynski Signed-off-by: Bjorn Helgaas --- drivers/pci/controller/pcie-cadence.h | 2 +- drivers/pci/controller/pcie-rockchip.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pci/controller/pcie-cadence.h b/drivers/pci/controller/pcie-cadence.h index ae6bf2a2b3d3..f1cba3931b99 100644 --- a/drivers/pci/controller/pcie-cadence.h +++ b/drivers/pci/controller/pcie-cadence.h @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: GPL-2.0 +/* SPDX-License-Identifier: GPL-2.0 */ // Copyright (c) 2017 Cadence // Cadence PCIe controller driver. // Author: Cyrille Pitchen diff --git a/drivers/pci/controller/pcie-rockchip.h b/drivers/pci/controller/pcie-rockchip.h index 8e87a059ce73..53e4f9e59624 100644 --- a/drivers/pci/controller/pcie-rockchip.h +++ b/drivers/pci/controller/pcie-rockchip.h @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: GPL-2.0+ +/* SPDX-License-Identifier: GPL-2.0+ */ /* * Rockchip AXI PCIe controller driver * From 0d2f4d62ff4164df5611df53ae5546421d40cfb4 Mon Sep 17 00:00:00 2001 From: Krzysztof Wilczynski Date: Mon, 19 Aug 2019 08:05:32 +0200 Subject: [PATCH 130/160] x86/PCI: Replace deprecated EXTRA_CFLAGS with ccflags-y Update arch/x86/pci/Makefile replacing the deprecated EXTRA_CFLAGS with the ccflags-y matching recommendation per section 3.7 of Documentation/kbuild/makefiles.txt. Link: https://lore.kernel.org/r/20190819060532.17093-1-kw@linux.com Signed-off-by: Krzysztof Wilczynski Signed-off-by: Bjorn Helgaas --- arch/x86/pci/Makefile | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/arch/x86/pci/Makefile b/arch/x86/pci/Makefile index c806b57d3f22..48bcada5cabe 100644 --- a/arch/x86/pci/Makefile +++ b/arch/x86/pci/Makefile @@ -24,6 +24,4 @@ obj-y += bus_numa.o obj-$(CONFIG_AMD_NB) += amd_bus.o obj-$(CONFIG_PCI_CNB20LE_QUIRK) += broadcom_bus.o -ifeq ($(CONFIG_PCI_DEBUG),y) -EXTRA_CFLAGS += -DDEBUG -endif +ccflags-$(CONFIG_PCI_DEBUG) += -DDEBUG From bbd8810d399812f2016713565e4d8ff8f1508aa6 Mon Sep 17 00:00:00 2001 From: Krzysztof Wilczynski Date: Tue, 3 Sep 2019 13:30:59 +0200 Subject: [PATCH 131/160] PCI: Remove unused includes and superfluous struct declaration Remove and from being included directly as part of the include/linux/of_pci.h, and remove superfluous declaration of struct of_phandle_args. Move users of include to include and directly rather than rely on both being included transitively through . Link: https://lore.kernel.org/r/20190903113059.2901-1-kw@linux.com Signed-off-by: Krzysztof Wilczynski Signed-off-by: Bjorn Helgaas Reviewed-by: Rob Herring --- drivers/iommu/of_iommu.c | 2 ++ drivers/irqchip/irq-gic-v2m.c | 1 + drivers/irqchip/irq-gic-v3-its-pci-msi.c | 1 + drivers/pci/controller/dwc/pcie-designware-host.c | 1 + drivers/pci/controller/pci-aardvark.c | 1 + drivers/pci/controller/pci-thunder-pem.c | 1 + drivers/pci/pci.c | 1 + drivers/pci/probe.c | 1 + include/linux/of_pci.h | 5 ++--- 9 files changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/iommu/of_iommu.c b/drivers/iommu/of_iommu.c index 614a93aa5305..026ad2b29dcd 100644 --- a/drivers/iommu/of_iommu.c +++ b/drivers/iommu/of_iommu.c @@ -8,6 +8,8 @@ #include #include #include +#include +#include #include #include #include diff --git a/drivers/irqchip/irq-gic-v2m.c b/drivers/irqchip/irq-gic-v2m.c index e88e75c22b6a..fbec07d634ad 100644 --- a/drivers/irqchip/irq-gic-v2m.c +++ b/drivers/irqchip/irq-gic-v2m.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/irqchip/irq-gic-v3-its-pci-msi.c b/drivers/irqchip/irq-gic-v3-its-pci-msi.c index 229d586c3d7a..87711e0f8014 100644 --- a/drivers/irqchip/irq-gic-v3-its-pci-msi.c +++ b/drivers/irqchip/irq-gic-v3-its-pci-msi.c @@ -5,6 +5,7 @@ */ #include +#include #include #include #include diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c index 0f36a926059a..43ce6de70374 100644 --- a/drivers/pci/controller/dwc/pcie-designware-host.c +++ b/drivers/pci/controller/dwc/pcie-designware-host.c @@ -10,6 +10,7 @@ #include #include +#include #include #include #include diff --git a/drivers/pci/controller/pci-aardvark.c b/drivers/pci/controller/pci-aardvark.c index fc0fe4d4de49..3a05f6ca95b0 100644 --- a/drivers/pci/controller/pci-aardvark.c +++ b/drivers/pci/controller/pci-aardvark.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include diff --git a/drivers/pci/controller/pci-thunder-pem.c b/drivers/pci/controller/pci-thunder-pem.c index f127ce8bd4ef..9491e266b1ea 100644 --- a/drivers/pci/controller/pci-thunder-pem.c +++ b/drivers/pci/controller/pci-thunder-pem.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index e7982af9a5d8..2d36995e4ea5 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 3d5271a7a849..7c5d68b807ef 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include diff --git a/include/linux/of_pci.h b/include/linux/of_pci.h index 21a89c4880fa..29658c0ee71f 100644 --- a/include/linux/of_pci.h +++ b/include/linux/of_pci.h @@ -2,11 +2,10 @@ #ifndef __OF_PCI_H #define __OF_PCI_H -#include -#include +#include +#include struct pci_dev; -struct of_phandle_args; struct device_node; #if IS_ENABLED(CONFIG_OF) && IS_ENABLED(CONFIG_PCI) From 7e8ce0e2b036dbc6617184317983aea4f2c52099 Mon Sep 17 00:00:00 2001 From: Kai-Heng Feng Date: Mon, 2 Sep 2019 22:52:52 +0800 Subject: [PATCH 132/160] x86/PCI: Avoid AMD FCH XHCI USB PME# from D0 defect The AMD FCH USB XHCI Controller advertises support for generating PME# while in D0. When in D0, it does signal PME# for USB 3.0 connect events, but not for USB 2.0 or USB 1.1 connect events, which means the controller doesn't wake correctly for those events. 00:10.0 USB controller [0c03]: Advanced Micro Devices, Inc. [AMD] FCH USB XHCI Controller [1022:7914] (rev 20) (prog-if 30 [XHCI]) Subsystem: Dell FCH USB XHCI Controller [1028:087e] Capabilities: [50] Power Management version 3 Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0+,D1-,D2-,D3hot+,D3cold+) Clear PCI_PM_CAP_PME_D0 in dev->pme_support to indicate the device will not assert PME# from D0 so we don't rely on it. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=203673 Link: https://lore.kernel.org/r/20190902145252.32111-1-kai.heng.feng@canonical.com Signed-off-by: Kai-Heng Feng Signed-off-by: Bjorn Helgaas Cc: stable@vger.kernel.org --- arch/x86/pci/fixup.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/arch/x86/pci/fixup.c b/arch/x86/pci/fixup.c index 527e69b12002..e723559c386a 100644 --- a/arch/x86/pci/fixup.c +++ b/arch/x86/pci/fixup.c @@ -588,6 +588,17 @@ static void pci_fixup_amd_ehci_pme(struct pci_dev *dev) } DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, 0x7808, pci_fixup_amd_ehci_pme); +/* + * Device [1022:7914] + * When in D0, PME# doesn't get asserted when plugging USB 2.0 device. + */ +static void pci_fixup_amd_fch_xhci_pme(struct pci_dev *dev) +{ + dev_info(&dev->dev, "PME# does not work under D0, disabling it\n"); + dev->pme_support &= ~(PCI_PM_CAP_PME_D0 >> PCI_PM_CAP_PME_SHIFT); +} +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, 0x7914, pci_fixup_amd_fch_xhci_pme); + /* * Apple MacBook Pro: Avoid [mem 0x7fa00000-0x7fbfffff] * From ca22d1f5474a7278d317bb86296723e9e563c44d Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Wed, 16 Oct 2019 09:03:24 +0100 Subject: [PATCH 133/160] PCI: sysfs: Remove unused attribute groups 56c1af4606f0 ("PCI: Add sysfs max_link_speed/width, current_link_speed/width, etc") added the following objects, but they are unused, so remove them: pci_bridge_group pci_bridge_groups pcie_dev_group pcie_dev_groups This fixes the following warnings from sparse: drivers/pci/pci-sysfs.c:1546:30: warning: symbol 'pci_bridge_groups' was not declared. Should it be static? drivers/pci/pci-sysfs.c:1555:30: warning: symbol 'pcie_dev_groups' was not declared. Should it be static? Link: https://lore.kernel.org/r/20191016080324.12864-1-ben.dooks@codethink.co.uk Signed-off-by: Ben Dooks Signed-off-by: Bjorn Helgaas --- drivers/pci/pci-sysfs.c | 18 ------------------ 1 file changed, 18 deletions(-) diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 793412954529..eaffb477c5bf 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -1539,24 +1539,6 @@ const struct attribute_group *pci_dev_groups[] = { NULL, }; -static const struct attribute_group pci_bridge_group = { - .attrs = pci_bridge_attrs, -}; - -const struct attribute_group *pci_bridge_groups[] = { - &pci_bridge_group, - NULL, -}; - -static const struct attribute_group pcie_dev_group = { - .attrs = pcie_dev_attrs, -}; - -const struct attribute_group *pcie_dev_groups[] = { - &pcie_dev_group, - NULL, -}; - static const struct attribute_group pci_dev_hp_attr_group = { .attrs = pci_dev_hp_attrs, .is_visible = pci_dev_hp_attrs_are_visible, From 127a7709495db52a41012deaebbb7afc231dad91 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 6 Nov 2019 15:30:48 -0600 Subject: [PATCH 134/160] PCI/PTM: Remove spurious "d" from granularity message The granularity message has an extra "d": pci 0000:02:00.0: PTM enabled, 4dns granularity Remove the "d" so the message is simply "PTM enabled, 4ns granularity". Fixes: 8b2ec318eece ("PCI: Add PTM clock granularity information") Link: https://lore.kernel.org/r/20191106222420.10216-2-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Andrew Murray Cc: Jonathan Yong --- drivers/pci/pcie/ptm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/pcie/ptm.c b/drivers/pci/pcie/ptm.c index 98cfa30f3fae..9361f3aa26ab 100644 --- a/drivers/pci/pcie/ptm.c +++ b/drivers/pci/pcie/ptm.c @@ -21,7 +21,7 @@ static void pci_ptm_info(struct pci_dev *dev) snprintf(clock_desc, sizeof(clock_desc), ">254ns"); break; default: - snprintf(clock_desc, sizeof(clock_desc), "%udns", + snprintf(clock_desc, sizeof(clock_desc), "%uns", dev->ptm_granularity); break; } From 97a0ac8a4678fb6340f6b2646e7c46e9c2872e02 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 6 Nov 2019 16:01:15 -0600 Subject: [PATCH 135/160] PCI/PTM: Remove dependency on PCIEPORTBUS The PTM support does not depend on the portdrv, so remove the Kconfig dependency. Link: https://lore.kernel.org/r/20191106222420.10216-3-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Andrew Murray Cc: Jonathan Yong --- drivers/pci/pcie/Kconfig | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/pci/pcie/Kconfig b/drivers/pci/pcie/Kconfig index 362eb8cfa53b..b0d781d72d1b 100644 --- a/drivers/pci/pcie/Kconfig +++ b/drivers/pci/pcie/Kconfig @@ -135,7 +135,6 @@ config PCIE_DPC config PCIE_PTM bool "PCI Express Precision Time Measurement support" - depends on PCIEPORTBUS help This enables PCI Express Precision Time Measurement (PTM) support. From 33ce09ef17329be407fc238a214d6f49fdc8a007 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 6 Nov 2019 16:07:36 -0600 Subject: [PATCH 136/160] PCI/ASPM: Remove dependency on PCIEPORTBUS The ASPM support does not depend on the portdrv, so remove the Kconfig dependency. Link: https://lore.kernel.org/r/20191106222420.10216-4-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Andrew Murray --- drivers/pci/pcie/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/pcie/Kconfig b/drivers/pci/pcie/Kconfig index b0d781d72d1b..b196ad816129 100644 --- a/drivers/pci/pcie/Kconfig +++ b/drivers/pci/pcie/Kconfig @@ -63,7 +63,7 @@ config PCIE_ECRC # config PCIEASPM bool "PCI Express ASPM control" if EXPERT - depends on PCI && PCIEPORTBUS + depends on PCI default y help This enables OS control over PCI Express ASPM (Active State From b6479581e6823239c3e03b1d006f160c6c491ba0 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 6 Nov 2019 16:09:40 -0600 Subject: [PATCH 137/160] PCI: Remove PCIe Kconfig dependencies on PCI drivers/pci/pcie/Kconfig is only sourced by drivers/pci/Kconfig, and only when PCI is defined, so there's no need to depend on PCI again. Remove the unnecessary dependencies. Link: https://lore.kernel.org/r/20191106222420.10216-5-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Andrew Murray --- drivers/pci/pcie/Kconfig | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/pci/pcie/Kconfig b/drivers/pci/pcie/Kconfig index b196ad816129..207dac2fd588 100644 --- a/drivers/pci/pcie/Kconfig +++ b/drivers/pci/pcie/Kconfig @@ -4,7 +4,6 @@ # config PCIEPORTBUS bool "PCI Express Port Bus support" - depends on PCI help This enables PCI Express Port Bus support. Users can then enable support for Native Hot-Plug, Advanced Error Reporting, Power @@ -63,7 +62,6 @@ config PCIE_ECRC # config PCIEASPM bool "PCI Express ASPM control" if EXPERT - depends on PCI default y help This enables OS control over PCI Express ASPM (Active State From b7da3d4df05232d637eff1c0874d20996b98769c Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 6 Nov 2019 16:13:43 -0600 Subject: [PATCH 138/160] PCI: Allow building PCIe things without PCIEPORTBUS Some things in drivers/pci/pcie (aspm.c and ptm.c) do not depend on the PCIe portdrv, so we should be able to build them even if PCIEPORTBUS is not selected. Remove the PCIEPORTBUS guard from building pcie/. Link: https://lore.kernel.org/r/20191106222420.10216-6-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Andrew Murray --- drivers/pci/Makefile | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile index 28cdd8c0213a..522d2b974e91 100644 --- a/drivers/pci/Makefile +++ b/drivers/pci/Makefile @@ -7,6 +7,8 @@ obj-$(CONFIG_PCI) += access.o bus.o probe.o host-bridge.o \ pci-sysfs.o rom.o setup-res.o irq.o vpd.o \ setup-bus.o vc.o mmap.o setup-irq.o +obj-$(CONFIG_PCI) += pcie/ + ifdef CONFIG_PCI obj-$(CONFIG_PROC_FS) += proc.o obj-$(CONFIG_SYSFS) += slot.o @@ -15,7 +17,6 @@ endif obj-$(CONFIG_OF) += of.o obj-$(CONFIG_PCI_QUIRKS) += quirks.o -obj-$(CONFIG_PCIEPORTBUS) += pcie/ obj-$(CONFIG_HOTPLUG_PCI) += hotplug/ obj-$(CONFIG_PCI_MSI) += msi.o obj-$(CONFIG_PCI_ATS) += ats.o From bbdb2f5ecdf1e66b2f09710134db3c2e5c43a958 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 12 Nov 2019 11:07:36 -0600 Subject: [PATCH 139/160] PCI: Add #defines for Enter Compliance, Transmit Margin Add definitions for the Enter Compliance and Transmit Margin fields of the PCIe Link Control 2 register. Link: https://lore.kernel.org/r/20191112173503.176611-2-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Alex Deucher --- include/uapi/linux/pci_regs.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/uapi/linux/pci_regs.h b/include/uapi/linux/pci_regs.h index 29d6e93fd15e..5869e5778a05 100644 --- a/include/uapi/linux/pci_regs.h +++ b/include/uapi/linux/pci_regs.h @@ -673,6 +673,8 @@ #define PCI_EXP_LNKCTL2_TLS_8_0GT 0x0003 /* Supported Speed 8GT/s */ #define PCI_EXP_LNKCTL2_TLS_16_0GT 0x0004 /* Supported Speed 16GT/s */ #define PCI_EXP_LNKCTL2_TLS_32_0GT 0x0005 /* Supported Speed 32GT/s */ +#define PCI_EXP_LNKCTL2_ENTER_COMP 0x0010 /* Enter Compliance */ +#define PCI_EXP_LNKCTL2_TX_MARGIN 0x0380 /* Transmit Margin */ #define PCI_EXP_LNKSTA2 50 /* Link Status 2 */ #define PCI_CAP_EXP_ENDPOINT_SIZEOF_V2 52 /* v2 endpoints with link end here */ #define PCI_EXP_SLTCAP2 52 /* Slot Capabilities 2 */ From 19d7a95a8ba66b198f759cf610cc935ce9840d5b Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 20 Nov 2019 17:52:48 -0600 Subject: [PATCH 140/160] drm/amdgpu: Correct Transmit Margin masks Previously we masked PCIe Link Control 2 register values with "7 << 9", which was apparently intended to be the Transmit Margin field, but instead was the high order bit of Transmit Margin, the Enter Modified Compliance bit, and the Compliance SOS bit. Correct the mask to "7 << 7", which is the Transmit Margin field. Link: https://lore.kernel.org/r/20191112173503.176611-3-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/cik.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/si.c | 8 ++++---- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c index b81bb414fcb3..13a5696d2a6a 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.c +++ b/drivers/gpu/drm/amd/amdgpu/cik.c @@ -1498,13 +1498,13 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) /* linkctl2 */ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 9)); - tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 9))); + tmp16 &= ~((1 << 4) | (7 << 7)); + tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 7))); pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 9)); - tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 9))); + tmp16 &= ~((1 << 4) | (7 << 7)); + tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 7))); pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); tmp = RREG32_PCIE(ixPCIE_LC_CNTL4); diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c index 493af42152f2..1e350172dc7b 100644 --- a/drivers/gpu/drm/amd/amdgpu/si.c +++ b/drivers/gpu/drm/amd/amdgpu/si.c @@ -1737,13 +1737,13 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 9)); - tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 9))); + tmp16 &= ~((1 << 4) | (7 << 7)); + tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 7))); pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 9)); - tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 9))); + tmp16 &= ~((1 << 4) | (7 << 7)); + tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 7))); pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); From 35e768e296729ac96a8c33b7810b6cb1673ae961 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 21 Nov 2019 07:23:41 -0600 Subject: [PATCH 141/160] drm/amdgpu: Replace numbers with PCI_EXP_LNKCTL2 definitions Replace hard-coded magic numbers with the descriptive PCI_EXP_LNKCTL2 definitions. No functional change intended. Link: https://lore.kernel.org/r/20191112173503.176611-4-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/cik.c | 22 ++++++++++++++-------- drivers/gpu/drm/amd/amdgpu/si.c | 22 ++++++++++++++-------- 2 files changed, 28 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c index 13a5696d2a6a..3067bb874032 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.c +++ b/drivers/gpu/drm/amd/amdgpu/cik.c @@ -1498,13 +1498,19 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) /* linkctl2 */ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 7)); - tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 7))); + tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN); + tmp16 |= (bridge_cfg2 & + (PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN)); pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 7)); - tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 7))); + tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN); + tmp16 |= (gpu_cfg2 & + (PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN)); pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); tmp = RREG32_PCIE(ixPCIE_LC_CNTL4); @@ -1521,13 +1527,13 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) WREG32_PCIE(ixPCIE_LC_SPEED_CNTL, speed_cntl); pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~0xf; + tmp16 &= ~PCI_EXP_LNKCTL2_TLS; if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) - tmp16 |= 3; /* gen3 */ + tmp16 |= PCI_EXP_LNKCTL2_TLS_8_0GT; /* gen3 */ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2) - tmp16 |= 2; /* gen2 */ + tmp16 |= PCI_EXP_LNKCTL2_TLS_5_0GT; /* gen2 */ else - tmp16 |= 1; /* gen1 */ + tmp16 |= PCI_EXP_LNKCTL2_TLS_2_5GT; /* gen1 */ pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); speed_cntl = RREG32_PCIE(ixPCIE_LC_SPEED_CNTL); diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c index 1e350172dc7b..a7dcb0d0f039 100644 --- a/drivers/gpu/drm/amd/amdgpu/si.c +++ b/drivers/gpu/drm/amd/amdgpu/si.c @@ -1737,13 +1737,19 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 7)); - tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 7))); + tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN); + tmp16 |= (bridge_cfg2 & + (PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN)); pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 7)); - tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 7))); + tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN); + tmp16 |= (gpu_cfg2 & + (PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN)); pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); @@ -1758,13 +1764,13 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl); pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~0xf; + tmp16 &= ~PCI_EXP_LNKCTL2_TLS; if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) - tmp16 |= 3; + tmp16 |= PCI_EXP_LNKCTL2_TLS_8_0GT; /* gen3 */ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2) - tmp16 |= 2; + tmp16 |= PCI_EXP_LNKCTL2_TLS_5_0GT; /* gen2 */ else - tmp16 |= 1; + tmp16 |= PCI_EXP_LNKCTL2_TLS_2_5GT; /* gen1 */ pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL); From 4b1140ade8f5c1ced640286cce79371ade2bd2d1 Mon Sep 17 00:00:00 2001 From: Kunihiko Hayashi Date: Thu, 7 Nov 2019 13:58:14 +0900 Subject: [PATCH 142/160] PCI: uniphier: Set mode register to host mode Set the mode register to host(RC) mode so that the host controller mode is set-up consistently across SoCs. Signed-off-by: Kunihiko Hayashi [lorenzo.pieralisi@arm.com: updated log] Signed-off-by: Lorenzo Pieralisi Reviewed-by: Andrew Murray --- drivers/pci/controller/dwc/pcie-uniphier.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/pci/controller/dwc/pcie-uniphier.c b/drivers/pci/controller/dwc/pcie-uniphier.c index 3f30ee4a00b3..8fd7badd59c2 100644 --- a/drivers/pci/controller/dwc/pcie-uniphier.c +++ b/drivers/pci/controller/dwc/pcie-uniphier.c @@ -33,6 +33,10 @@ #define PCL_PIPEMON 0x0044 #define PCL_PCLK_ALIVE BIT(15) +#define PCL_MODE 0x8000 +#define PCL_MODE_REGEN BIT(8) +#define PCL_MODE_REGVAL BIT(0) + #define PCL_APP_READY_CTRL 0x8008 #define PCL_APP_LTSSM_ENABLE BIT(0) @@ -85,6 +89,12 @@ static void uniphier_pcie_init_rc(struct uniphier_pcie_priv *priv) { u32 val; + /* set RC MODE */ + val = readl(priv->base + PCL_MODE); + val |= PCL_MODE_REGEN; + val &= ~PCL_MODE_REGVAL; + writel(val, priv->base + PCL_MODE); + /* use auxiliary power detection */ val = readl(priv->base + PCL_APP_PM0); val |= PCL_SYS_AUX_PWR_DET; From 88027c89ea146e32485251f1c2dddcde43c8d04e Mon Sep 17 00:00:00 2001 From: Frederick Lawler Date: Sun, 17 Nov 2019 18:35:13 -0600 Subject: [PATCH 143/160] drm/amdgpu: Prefer pcie_capability_read_word() Commit 8c0d3a02c130 ("PCI: Add accessors for PCI Express Capability") added accessors for the PCI Express Capability so that drivers didn't need to be aware of differences between v1 and v2 of the PCI Express Capability. Replace pci_read_config_word() and pci_write_config_word() calls with pcie_capability_read_word() and pcie_capability_write_word(). [bhelgaas: fix a couple remaining instances in cik.c] Link: https://lore.kernel.org/r/20191118003513.10852-1-fred@fredlawl.com Signed-off-by: Frederick Lawler Signed-off-by: Bjorn Helgaas Reviewed-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/cik.c | 71 ++++++++++++++++++++------------ drivers/gpu/drm/amd/amdgpu/si.c | 71 ++++++++++++++++++++------------ 2 files changed, 90 insertions(+), 52 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c index 3067bb874032..38b06ae6357a 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.c +++ b/drivers/gpu/drm/amd/amdgpu/cik.c @@ -1384,7 +1384,6 @@ static int cik_set_vce_clocks(struct amdgpu_device *adev, u32 evclk, u32 ecclk) static void cik_pcie_gen3_enable(struct amdgpu_device *adev) { struct pci_dev *root = adev->pdev->bus->self; - int bridge_pos, gpu_pos; u32 speed_cntl, current_data_rate; int i; u16 tmp16; @@ -1419,12 +1418,7 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) DRM_INFO("enabling PCIE gen 2 link speeds, disable with amdgpu.pcie_gen2=0\n"); } - bridge_pos = pci_pcie_cap(root); - if (!bridge_pos) - return; - - gpu_pos = pci_pcie_cap(adev->pdev); - if (!gpu_pos) + if (!pci_is_pcie(root) || !pci_is_pcie(adev->pdev)) return; if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) { @@ -1434,14 +1428,17 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) u16 bridge_cfg2, gpu_cfg2; u32 max_lw, current_lw, tmp; - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &bridge_cfg); + pcie_capability_read_word(adev->pdev, PCI_EXP_LNKCTL, + &gpu_cfg); tmp16 = bridge_cfg | PCI_EXP_LNKCTL_HAWD; - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(root, PCI_EXP_LNKCTL, tmp16); tmp16 = gpu_cfg | PCI_EXP_LNKCTL_HAWD; - pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(adev->pdev, PCI_EXP_LNKCTL, + tmp16); tmp = RREG32_PCIE(ixPCIE_LC_STATUS1); max_lw = (tmp & PCIE_LC_STATUS1__LC_DETECTED_LINK_WIDTH_MASK) >> @@ -1465,15 +1462,23 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) for (i = 0; i < 10; i++) { /* check status */ - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_DEVSTA, &tmp16); + pcie_capability_read_word(adev->pdev, + PCI_EXP_DEVSTA, + &tmp16); if (tmp16 & PCI_EXP_DEVSTA_TRPND) break; - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &bridge_cfg); + pcie_capability_read_word(adev->pdev, + PCI_EXP_LNKCTL, + &gpu_cfg); - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &bridge_cfg2); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &gpu_cfg2); + pcie_capability_read_word(root, PCI_EXP_LNKCTL2, + &bridge_cfg2); + pcie_capability_read_word(adev->pdev, + PCI_EXP_LNKCTL2, + &gpu_cfg2); tmp = RREG32_PCIE(ixPCIE_LC_CNTL4); tmp |= PCIE_LC_CNTL4__LC_SET_QUIESCE_MASK; @@ -1486,32 +1491,45 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) msleep(100); /* linkctl */ - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &tmp16); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &tmp16); tmp16 &= ~PCI_EXP_LNKCTL_HAWD; tmp16 |= (bridge_cfg & PCI_EXP_LNKCTL_HAWD); - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(root, PCI_EXP_LNKCTL, + tmp16); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, &tmp16); + pcie_capability_read_word(adev->pdev, + PCI_EXP_LNKCTL, + &tmp16); tmp16 &= ~PCI_EXP_LNKCTL_HAWD; tmp16 |= (gpu_cfg & PCI_EXP_LNKCTL_HAWD); - pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(adev->pdev, + PCI_EXP_LNKCTL, + tmp16); /* linkctl2 */ - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(root, PCI_EXP_LNKCTL2, + &tmp16); tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN); tmp16 |= (bridge_cfg2 & (PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN)); - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(root, + PCI_EXP_LNKCTL2, + tmp16); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(adev->pdev, + PCI_EXP_LNKCTL2, + &tmp16); tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN); tmp16 |= (gpu_cfg2 & (PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN)); - pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(adev->pdev, + PCI_EXP_LNKCTL2, + tmp16); tmp = RREG32_PCIE(ixPCIE_LC_CNTL4); tmp &= ~PCIE_LC_CNTL4__LC_SET_QUIESCE_MASK; @@ -1526,15 +1544,16 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) speed_cntl &= ~PCIE_LC_SPEED_CNTL__LC_FORCE_DIS_SW_SPEED_CHANGE_MASK; WREG32_PCIE(ixPCIE_LC_SPEED_CNTL, speed_cntl); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(adev->pdev, PCI_EXP_LNKCTL2, &tmp16); tmp16 &= ~PCI_EXP_LNKCTL2_TLS; + if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) tmp16 |= PCI_EXP_LNKCTL2_TLS_8_0GT; /* gen3 */ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2) tmp16 |= PCI_EXP_LNKCTL2_TLS_5_0GT; /* gen2 */ else tmp16 |= PCI_EXP_LNKCTL2_TLS_2_5GT; /* gen1 */ - pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(adev->pdev, PCI_EXP_LNKCTL2, tmp16); speed_cntl = RREG32_PCIE(ixPCIE_LC_SPEED_CNTL); speed_cntl |= PCIE_LC_SPEED_CNTL__LC_INITIATE_LINK_SPEED_CHANGE_MASK; diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c index a7dcb0d0f039..9f82be879224 100644 --- a/drivers/gpu/drm/amd/amdgpu/si.c +++ b/drivers/gpu/drm/amd/amdgpu/si.c @@ -1633,7 +1633,6 @@ static void si_init_golden_registers(struct amdgpu_device *adev) static void si_pcie_gen3_enable(struct amdgpu_device *adev) { struct pci_dev *root = adev->pdev->bus->self; - int bridge_pos, gpu_pos; u32 speed_cntl, current_data_rate; int i; u16 tmp16; @@ -1668,12 +1667,7 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) DRM_INFO("enabling PCIE gen 2 link speeds, disable with amdgpu.pcie_gen2=0\n"); } - bridge_pos = pci_pcie_cap(root); - if (!bridge_pos) - return; - - gpu_pos = pci_pcie_cap(adev->pdev); - if (!gpu_pos) + if (!pci_is_pcie(root) || !pci_is_pcie(adev->pdev)) return; if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) { @@ -1682,14 +1676,17 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) u16 bridge_cfg2, gpu_cfg2; u32 max_lw, current_lw, tmp; - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &bridge_cfg); + pcie_capability_read_word(adev->pdev, PCI_EXP_LNKCTL, + &gpu_cfg); tmp16 = bridge_cfg | PCI_EXP_LNKCTL_HAWD; - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(root, PCI_EXP_LNKCTL, tmp16); tmp16 = gpu_cfg | PCI_EXP_LNKCTL_HAWD; - pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(adev->pdev, PCI_EXP_LNKCTL, + tmp16); tmp = RREG32_PCIE(PCIE_LC_STATUS1); max_lw = (tmp & LC_DETECTED_LINK_WIDTH_MASK) >> LC_DETECTED_LINK_WIDTH_SHIFT; @@ -1706,15 +1703,23 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) } for (i = 0; i < 10; i++) { - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_DEVSTA, &tmp16); + pcie_capability_read_word(adev->pdev, + PCI_EXP_DEVSTA, + &tmp16); if (tmp16 & PCI_EXP_DEVSTA_TRPND) break; - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &bridge_cfg); + pcie_capability_read_word(adev->pdev, + PCI_EXP_LNKCTL, + &gpu_cfg); - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &bridge_cfg2); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &gpu_cfg2); + pcie_capability_read_word(root, PCI_EXP_LNKCTL2, + &bridge_cfg2); + pcie_capability_read_word(adev->pdev, + PCI_EXP_LNKCTL2, + &gpu_cfg2); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); tmp |= LC_SET_QUIESCE; @@ -1726,31 +1731,44 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) mdelay(100); - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &tmp16); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &tmp16); tmp16 &= ~PCI_EXP_LNKCTL_HAWD; tmp16 |= (bridge_cfg & PCI_EXP_LNKCTL_HAWD); - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(root, PCI_EXP_LNKCTL, + tmp16); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, &tmp16); + pcie_capability_read_word(adev->pdev, + PCI_EXP_LNKCTL, + &tmp16); tmp16 &= ~PCI_EXP_LNKCTL_HAWD; tmp16 |= (gpu_cfg & PCI_EXP_LNKCTL_HAWD); - pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(adev->pdev, + PCI_EXP_LNKCTL, + tmp16); - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(root, PCI_EXP_LNKCTL2, + &tmp16); tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN); tmp16 |= (bridge_cfg2 & (PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN)); - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(root, + PCI_EXP_LNKCTL2, + tmp16); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(adev->pdev, + PCI_EXP_LNKCTL2, + &tmp16); tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN); tmp16 |= (gpu_cfg2 & (PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN)); - pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(adev->pdev, + PCI_EXP_LNKCTL2, + tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); tmp &= ~LC_SET_QUIESCE; @@ -1763,15 +1781,16 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) speed_cntl &= ~LC_FORCE_DIS_SW_SPEED_CHANGE; WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl); - pci_read_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(adev->pdev, PCI_EXP_LNKCTL2, &tmp16); tmp16 &= ~PCI_EXP_LNKCTL2_TLS; + if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) tmp16 |= PCI_EXP_LNKCTL2_TLS_8_0GT; /* gen3 */ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2) tmp16 |= PCI_EXP_LNKCTL2_TLS_5_0GT; /* gen2 */ else tmp16 |= PCI_EXP_LNKCTL2_TLS_2_5GT; /* gen1 */ - pci_write_config_word(adev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(adev->pdev, PCI_EXP_LNKCTL2, tmp16); speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL); speed_cntl |= LC_INITIATE_LINK_SPEED_CHANGE; From 40bd4be5a652ce56068a8273b68caa38cb0d8f4b Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 20 Nov 2019 17:54:13 -0600 Subject: [PATCH 144/160] drm/radeon: Correct Transmit Margin masks Previously we masked PCIe Link Control 2 register values with "7 << 9", which was apparently intended to be the Transmit Margin field, but instead was the high order bit of Transmit Margin, the Enter Modified Compliance bit, and the Compliance SOS bit. Correct the mask to "7 << 7", which is the Transmit Margin field. Link: https://lore.kernel.org/r/20191112173503.176611-3-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Alex Deucher --- drivers/gpu/drm/radeon/cik.c | 8 ++++---- drivers/gpu/drm/radeon/si.c | 8 ++++---- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index 62eab82a64f9..14cdfdf78bde 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -9619,13 +9619,13 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) /* linkctl2 */ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 9)); - tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 9))); + tmp16 &= ~((1 << 4) | (7 << 7)); + tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 7))); pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 9)); - tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 9))); + tmp16 &= ~((1 << 4) | (7 << 7)); + tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 7))); pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 05894d198a79..9b7042d3ef1b 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -7202,13 +7202,13 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) /* linkctl2 */ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 9)); - tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 9))); + tmp16 &= ~((1 << 4) | (7 << 7)); + tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 7))); pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 9)); - tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 9))); + tmp16 &= ~((1 << 4) | (7 << 7)); + tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 7))); pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); From ca56f99c18cafdeae6961ce9d87fc978506152ca Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 21 Nov 2019 07:24:24 -0600 Subject: [PATCH 145/160] drm/radeon: Replace numbers with PCI_EXP_LNKCTL2 definitions Replace hard-coded magic numbers with the descriptive PCI_EXP_LNKCTL2 definitions. No functional change intended. Link: https://lore.kernel.org/r/20191112173503.176611-4-helgaas@kernel.org Signed-off-by: Bjorn Helgaas Reviewed-by: Alex Deucher --- drivers/gpu/drm/radeon/cik.c | 22 ++++++++++++++-------- drivers/gpu/drm/radeon/si.c | 22 ++++++++++++++-------- 2 files changed, 28 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index 14cdfdf78bde..a280442c81aa 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -9619,13 +9619,19 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) /* linkctl2 */ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 7)); - tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 7))); + tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN); + tmp16 |= (bridge_cfg2 & + (PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN)); pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 7)); - tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 7))); + tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN); + tmp16 |= (gpu_cfg2 & + (PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN)); pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); @@ -9641,13 +9647,13 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl); pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~0xf; + tmp16 &= ~PCI_EXP_LNKCTL2_TLS; if (speed_cap == PCIE_SPEED_8_0GT) - tmp16 |= 3; /* gen3 */ + tmp16 |= PCI_EXP_LNKCTL2_TLS_8_0GT; /* gen3 */ else if (speed_cap == PCIE_SPEED_5_0GT) - tmp16 |= 2; /* gen2 */ + tmp16 |= PCI_EXP_LNKCTL2_TLS_5_0GT; /* gen2 */ else - tmp16 |= 1; /* gen1 */ + tmp16 |= PCI_EXP_LNKCTL2_TLS_2_5GT; /* gen1 */ pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL); diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 9b7042d3ef1b..529e70a42019 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -7202,13 +7202,19 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) /* linkctl2 */ pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 7)); - tmp16 |= (bridge_cfg2 & ((1 << 4) | (7 << 7))); + tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN); + tmp16 |= (bridge_cfg2 & + (PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN)); pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~((1 << 4) | (7 << 7)); - tmp16 |= (gpu_cfg2 & ((1 << 4) | (7 << 7))); + tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN); + tmp16 |= (gpu_cfg2 & + (PCI_EXP_LNKCTL2_ENTER_COMP | + PCI_EXP_LNKCTL2_TX_MARGIN)); pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); @@ -7224,13 +7230,13 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl); pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); - tmp16 &= ~0xf; + tmp16 &= ~PCI_EXP_LNKCTL2_TLS; if (speed_cap == PCIE_SPEED_8_0GT) - tmp16 |= 3; /* gen3 */ + tmp16 |= PCI_EXP_LNKCTL2_TLS_8_0GT; /* gen3 */ else if (speed_cap == PCIE_SPEED_5_0GT) - tmp16 |= 2; /* gen2 */ + tmp16 |= PCI_EXP_LNKCTL2_TLS_5_0GT; /* gen2 */ else - tmp16 |= 1; /* gen1 */ + tmp16 |= PCI_EXP_LNKCTL2_TLS_2_5GT; /* gen1 */ pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL); From 3d581b11e34a92350983e5d3ecf469b5c677e295 Mon Sep 17 00:00:00 2001 From: Frederick Lawler Date: Sun, 17 Nov 2019 18:35:13 -0600 Subject: [PATCH 146/160] drm/radeon: Prefer pcie_capability_read_word() Commit 8c0d3a02c130 ("PCI: Add accessors for PCI Express Capability") added accessors for the PCI Express Capability so that drivers didn't need to be aware of differences between v1 and v2 of the PCI Express Capability. Replace pci_read_config_word() and pci_write_config_word() calls with pcie_capability_read_word() and pcie_capability_write_word(). Link: https://lore.kernel.org/r/20191118003513.10852-1-fred@fredlawl.com Signed-off-by: Frederick Lawler Signed-off-by: Bjorn Helgaas Reviewed-by: Alex Deucher --- drivers/gpu/drm/radeon/cik.c | 70 +++++++++++++++++++++------------- drivers/gpu/drm/radeon/si.c | 73 +++++++++++++++++++++++------------- 2 files changed, 90 insertions(+), 53 deletions(-) diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index a280442c81aa..09a4709e67f0 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -9504,7 +9504,6 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) { struct pci_dev *root = rdev->pdev->bus->self; enum pci_bus_speed speed_cap; - int bridge_pos, gpu_pos; u32 speed_cntl, current_data_rate; int i; u16 tmp16; @@ -9546,12 +9545,7 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) DRM_INFO("enabling PCIE gen 2 link speeds, disable with radeon.pcie_gen2=0\n"); } - bridge_pos = pci_pcie_cap(root); - if (!bridge_pos) - return; - - gpu_pos = pci_pcie_cap(rdev->pdev); - if (!gpu_pos) + if (!pci_is_pcie(root) || !pci_is_pcie(rdev->pdev)) return; if (speed_cap == PCIE_SPEED_8_0GT) { @@ -9561,14 +9555,17 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) u16 bridge_cfg2, gpu_cfg2; u32 max_lw, current_lw, tmp; - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &bridge_cfg); + pcie_capability_read_word(rdev->pdev, PCI_EXP_LNKCTL, + &gpu_cfg); tmp16 = bridge_cfg | PCI_EXP_LNKCTL_HAWD; - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(root, PCI_EXP_LNKCTL, tmp16); tmp16 = gpu_cfg | PCI_EXP_LNKCTL_HAWD; - pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(rdev->pdev, PCI_EXP_LNKCTL, + tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_STATUS1); max_lw = (tmp & LC_DETECTED_LINK_WIDTH_MASK) >> LC_DETECTED_LINK_WIDTH_SHIFT; @@ -9586,15 +9583,23 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) for (i = 0; i < 10; i++) { /* check status */ - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_DEVSTA, &tmp16); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_DEVSTA, + &tmp16); if (tmp16 & PCI_EXP_DEVSTA_TRPND) break; - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &bridge_cfg); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_LNKCTL, + &gpu_cfg); - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &bridge_cfg2); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &gpu_cfg2); + pcie_capability_read_word(root, PCI_EXP_LNKCTL2, + &bridge_cfg2); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_LNKCTL2, + &gpu_cfg2); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); tmp |= LC_SET_QUIESCE; @@ -9607,32 +9612,45 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) msleep(100); /* linkctl */ - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &tmp16); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &tmp16); tmp16 &= ~PCI_EXP_LNKCTL_HAWD; tmp16 |= (bridge_cfg & PCI_EXP_LNKCTL_HAWD); - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(root, PCI_EXP_LNKCTL, + tmp16); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, &tmp16); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_LNKCTL, + &tmp16); tmp16 &= ~PCI_EXP_LNKCTL_HAWD; tmp16 |= (gpu_cfg & PCI_EXP_LNKCTL_HAWD); - pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(rdev->pdev, + PCI_EXP_LNKCTL, + tmp16); /* linkctl2 */ - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(root, PCI_EXP_LNKCTL2, + &tmp16); tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN); tmp16 |= (bridge_cfg2 & (PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN)); - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(root, + PCI_EXP_LNKCTL2, + tmp16); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_LNKCTL2, + &tmp16); tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN); tmp16 |= (gpu_cfg2 & (PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN)); - pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(rdev->pdev, + PCI_EXP_LNKCTL2, + tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); tmp &= ~LC_SET_QUIESCE; @@ -9646,7 +9664,7 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) speed_cntl &= ~LC_FORCE_DIS_SW_SPEED_CHANGE; WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(rdev->pdev, PCI_EXP_LNKCTL2, &tmp16); tmp16 &= ~PCI_EXP_LNKCTL2_TLS; if (speed_cap == PCIE_SPEED_8_0GT) tmp16 |= PCI_EXP_LNKCTL2_TLS_8_0GT; /* gen3 */ @@ -9654,7 +9672,7 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) tmp16 |= PCI_EXP_LNKCTL2_TLS_5_0GT; /* gen2 */ else tmp16 |= PCI_EXP_LNKCTL2_TLS_2_5GT; /* gen1 */ - pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(rdev->pdev, PCI_EXP_LNKCTL2, tmp16); speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL); speed_cntl |= LC_INITIATE_LINK_SPEED_CHANGE; diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 529e70a42019..67a98b3370d1 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -3257,7 +3257,7 @@ static void si_gpu_init(struct radeon_device *rdev) /* XXX what about 12? */ rdev->config.si.tile_config |= (3 << 0); break; - } + } switch ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) { case 0: /* four banks */ rdev->config.si.tile_config |= 0 << 4; @@ -7087,7 +7087,6 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) { struct pci_dev *root = rdev->pdev->bus->self; enum pci_bus_speed speed_cap; - int bridge_pos, gpu_pos; u32 speed_cntl, current_data_rate; int i; u16 tmp16; @@ -7129,12 +7128,7 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) DRM_INFO("enabling PCIE gen 2 link speeds, disable with radeon.pcie_gen2=0\n"); } - bridge_pos = pci_pcie_cap(root); - if (!bridge_pos) - return; - - gpu_pos = pci_pcie_cap(rdev->pdev); - if (!gpu_pos) + if (!pci_is_pcie(root) || !pci_is_pcie(rdev->pdev)) return; if (speed_cap == PCIE_SPEED_8_0GT) { @@ -7144,14 +7138,17 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) u16 bridge_cfg2, gpu_cfg2; u32 max_lw, current_lw, tmp; - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &bridge_cfg); + pcie_capability_read_word(rdev->pdev, PCI_EXP_LNKCTL, + &gpu_cfg); tmp16 = bridge_cfg | PCI_EXP_LNKCTL_HAWD; - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(root, PCI_EXP_LNKCTL, tmp16); tmp16 = gpu_cfg | PCI_EXP_LNKCTL_HAWD; - pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(rdev->pdev, PCI_EXP_LNKCTL, + tmp16); tmp = RREG32_PCIE(PCIE_LC_STATUS1); max_lw = (tmp & LC_DETECTED_LINK_WIDTH_MASK) >> LC_DETECTED_LINK_WIDTH_SHIFT; @@ -7169,15 +7166,23 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) for (i = 0; i < 10; i++) { /* check status */ - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_DEVSTA, &tmp16); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_DEVSTA, + &tmp16); if (tmp16 & PCI_EXP_DEVSTA_TRPND) break; - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &bridge_cfg); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, &gpu_cfg); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &bridge_cfg); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_LNKCTL, + &gpu_cfg); - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &bridge_cfg2); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &gpu_cfg2); + pcie_capability_read_word(root, PCI_EXP_LNKCTL2, + &bridge_cfg2); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_LNKCTL2, + &gpu_cfg2); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); tmp |= LC_SET_QUIESCE; @@ -7190,32 +7195,46 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) msleep(100); /* linkctl */ - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL, &tmp16); + pcie_capability_read_word(root, PCI_EXP_LNKCTL, + &tmp16); tmp16 &= ~PCI_EXP_LNKCTL_HAWD; tmp16 |= (bridge_cfg & PCI_EXP_LNKCTL_HAWD); - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(root, + PCI_EXP_LNKCTL, + tmp16); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, &tmp16); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_LNKCTL, + &tmp16); tmp16 &= ~PCI_EXP_LNKCTL_HAWD; tmp16 |= (gpu_cfg & PCI_EXP_LNKCTL_HAWD); - pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL, tmp16); + pcie_capability_write_word(rdev->pdev, + PCI_EXP_LNKCTL, + tmp16); /* linkctl2 */ - pci_read_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(root, PCI_EXP_LNKCTL2, + &tmp16); tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN); tmp16 |= (bridge_cfg2 & (PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN)); - pci_write_config_word(root, bridge_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(root, + PCI_EXP_LNKCTL2, + tmp16); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(rdev->pdev, + PCI_EXP_LNKCTL2, + &tmp16); tmp16 &= ~(PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN); tmp16 |= (gpu_cfg2 & (PCI_EXP_LNKCTL2_ENTER_COMP | PCI_EXP_LNKCTL2_TX_MARGIN)); - pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(rdev->pdev, + PCI_EXP_LNKCTL2, + tmp16); tmp = RREG32_PCIE_PORT(PCIE_LC_CNTL4); tmp &= ~LC_SET_QUIESCE; @@ -7229,7 +7248,7 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) speed_cntl &= ~LC_FORCE_DIS_SW_SPEED_CHANGE; WREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL, speed_cntl); - pci_read_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, &tmp16); + pcie_capability_read_word(rdev->pdev, PCI_EXP_LNKCTL2, &tmp16); tmp16 &= ~PCI_EXP_LNKCTL2_TLS; if (speed_cap == PCIE_SPEED_8_0GT) tmp16 |= PCI_EXP_LNKCTL2_TLS_8_0GT; /* gen3 */ @@ -7237,7 +7256,7 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) tmp16 |= PCI_EXP_LNKCTL2_TLS_5_0GT; /* gen2 */ else tmp16 |= PCI_EXP_LNKCTL2_TLS_2_5GT; /* gen1 */ - pci_write_config_word(rdev->pdev, gpu_pos + PCI_EXP_LNKCTL2, tmp16); + pcie_capability_write_word(rdev->pdev, PCI_EXP_LNKCTL2, tmp16); speed_cntl = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL); speed_cntl |= LC_INITIATE_LINK_SPEED_CHANGE; From 36533f355b1ad14ec4352f7e254a5bfd4f4923d5 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Wed, 20 Nov 2019 21:40:36 +0800 Subject: [PATCH 147/160] PCI: Fix indentation Adjust indentation from spaces to tab (+optional two spaces) as in coding style with command like: $ sed -e 's/^ /\t/' -i */Kconfig [bhelgaas: do same in vmd.c] Link: https://lore.kernel.org/r/20191120134036.14502-1-krzk@kernel.org Signed-off-by: Krzysztof Kozlowski Signed-off-by: Bjorn Helgaas --- drivers/pci/Kconfig | 24 ++++++++++++------------ drivers/pci/controller/dwc/Kconfig | 6 +++--- drivers/pci/controller/vmd.c | 2 +- drivers/pci/hotplug/Kconfig | 2 +- 4 files changed, 17 insertions(+), 17 deletions(-) diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index a304f5ea11b9..bd50765f30cd 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -106,14 +106,14 @@ config PCI_PF_STUB When in doubt, say N. config XEN_PCIDEV_FRONTEND - tristate "Xen PCI Frontend" - depends on X86 && XEN - select PCI_XEN + tristate "Xen PCI Frontend" + depends on X86 && XEN + select PCI_XEN select XEN_XENBUS_FRONTEND - default y - help - The PCI device frontend driver allows the kernel to import arbitrary - PCI devices from a PCI backend to support PCI driver domains. + default y + help + The PCI device frontend driver allows the kernel to import arbitrary + PCI devices from a PCI backend to support PCI driver domains. config PCI_ATS bool @@ -180,12 +180,12 @@ config PCI_LABEL select NLS config PCI_HYPERV - tristate "Hyper-V PCI Frontend" - depends on X86_64 && HYPERV && PCI_MSI && PCI_MSI_IRQ_DOMAIN && SYSFS + tristate "Hyper-V PCI Frontend" + depends on X86_64 && HYPERV && PCI_MSI && PCI_MSI_IRQ_DOMAIN && SYSFS select PCI_HYPERV_INTERFACE - help - The PCI device frontend driver allows the kernel to import arbitrary - PCI devices from a PCI backend to support PCI driver domains. + help + The PCI device frontend driver allows the kernel to import arbitrary + PCI devices from a PCI backend to support PCI driver domains. source "drivers/pci/hotplug/Kconfig" source "drivers/pci/controller/Kconfig" diff --git a/drivers/pci/controller/dwc/Kconfig b/drivers/pci/controller/dwc/Kconfig index 0ba988b5b5bc..625a031b2193 100644 --- a/drivers/pci/controller/dwc/Kconfig +++ b/drivers/pci/controller/dwc/Kconfig @@ -7,9 +7,9 @@ config PCIE_DW bool config PCIE_DW_HOST - bool + bool depends on PCI_MSI_IRQ_DOMAIN - select PCIE_DW + select PCIE_DW config PCIE_DW_EP bool @@ -224,7 +224,7 @@ config PCIE_HISI_STB depends on PCI_MSI_IRQ_DOMAIN select PCIE_DW_HOST help - Say Y here if you want PCIe controller support on HiSilicon STB SoCs + Say Y here if you want PCIe controller support on HiSilicon STB SoCs config PCI_MESON bool "MESON PCIe controller" diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c index a35d3f3996d7..5d21c9c73336 100644 --- a/drivers/pci/controller/vmd.c +++ b/drivers/pci/controller/vmd.c @@ -823,7 +823,7 @@ static int vmd_suspend(struct device *dev) int i; for (i = 0; i < vmd->msix_count; i++) - devm_free_irq(dev, pci_irq_vector(pdev, i), &vmd->irqs[i]); + devm_free_irq(dev, pci_irq_vector(pdev, i), &vmd->irqs[i]); pci_save_state(pdev); return 0; diff --git a/drivers/pci/hotplug/Kconfig b/drivers/pci/hotplug/Kconfig index e7b493c22bf3..32455a79372d 100644 --- a/drivers/pci/hotplug/Kconfig +++ b/drivers/pci/hotplug/Kconfig @@ -83,7 +83,7 @@ config HOTPLUG_PCI_CPCI_ZT5550 depends on HOTPLUG_PCI_CPCI && X86 help Say Y here if you have an Performance Technologies (formerly Intel, - formerly just Ziatech) Ziatech ZT5550 CompactPCI system card. + formerly just Ziatech) Ziatech ZT5550 CompactPCI system card. To compile this driver as a module, choose M here: the module will be called cpcihp_zt5550. From 72ea91afbfb08619696ccde610ee4d0d29cf4a1d Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 5 Oct 2019 14:07:56 +0200 Subject: [PATCH 148/160] PCI/ASPM: Add sysfs attributes for controlling ASPM link states Add sysfs attributes to Endpoints and other Upstream Ports to control ASPM, Clock PM, and L1 PM Substates. The new attributes are: /sys/devices/pci*/.../link/clkpm /sys/devices/pci*/.../link/l0s_aspm /sys/devices/pci*/.../link/l1_aspm /sys/devices/pci*/.../link/l1_1_aspm /sys/devices/pci*/.../link/l1_2_aspm /sys/devices/pci*/.../link/l1_1_pcipm /sys/devices/pci*/.../link/l1_2_pcipm An attribute is only visible if both ends of the Link leading to the device support the state. Writing y/1/on to the file enables the state; n/0/off disables it. These attributes can be used to tune the power/performance tradeoff for individual devices. [bhelgaas: commit log, rename directory to "link"] Link: https://lore.kernel.org/r/b1c83f8a-9bf6-eac5-82d0-cf5b90128fbf@gmail.com Signed-off-by: Heiner Kallweit Signed-off-by: Bjorn Helgaas --- Documentation/ABI/testing/sysfs-bus-pci | 13 +++ drivers/pci/pci-sysfs.c | 3 + drivers/pci/pci.h | 4 + drivers/pci/pcie/aspm.c | 149 ++++++++++++++++++++++++ 4 files changed, 169 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-bus-pci b/Documentation/ABI/testing/sysfs-bus-pci index 8bfee557e50e..450296cc7948 100644 --- a/Documentation/ABI/testing/sysfs-bus-pci +++ b/Documentation/ABI/testing/sysfs-bus-pci @@ -347,3 +347,16 @@ Description: If the device has any Peer-to-Peer memory registered, this file contains a '1' if the memory has been published for use outside the driver that owns the device. + +What: /sys/bus/pci/devices/.../link/clkpm + /sys/bus/pci/devices/.../link/l0s_aspm + /sys/bus/pci/devices/.../link/l1_aspm + /sys/bus/pci/devices/.../link/l1_1_aspm + /sys/bus/pci/devices/.../link/l1_2_aspm + /sys/bus/pci/devices/.../link/l1_1_pcipm + /sys/bus/pci/devices/.../link/l1_2_pcipm +Date: October 2019 +Contact: Heiner Kallweit +Description: If ASPM is supported for an endpoint, these files can be + used to disable or enable the individual power management + states. Write y/1/on to enable, n/0/off to disable. diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 793412954529..0e76c02e0b7d 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -1587,6 +1587,9 @@ static const struct attribute_group *pci_dev_attr_groups[] = { &pcie_dev_attr_group, #ifdef CONFIG_PCIEAER &aer_stats_attr_group, +#endif +#ifdef CONFIG_PCIEASPM + &aspm_ctrl_attr_group, #endif NULL, }; diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 3f6947ee3324..b2cd21e8cb51 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -667,4 +667,8 @@ static inline int pci_acpi_program_hp_params(struct pci_dev *dev) } #endif +#ifdef CONFIG_PCIEASPM +extern const struct attribute_group aspm_ctrl_attr_group; +#endif + #endif /* DRIVERS_PCI_H */ diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index 2d5ecede8fa3..9cd251827a67 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -899,6 +899,14 @@ static struct pcie_link_state *alloc_pcie_link_state(struct pci_dev *pdev) return link; } +static void pcie_aspm_update_sysfs_visibility(struct pci_dev *pdev) +{ + struct pci_dev *child; + + list_for_each_entry(child, &pdev->subordinate->devices, bus_list) + sysfs_update_group(&child->dev.kobj, &aspm_ctrl_attr_group); +} + /* * pcie_aspm_init_link_state: Initiate PCI express link state. * It is called after the pcie and its children devices are scanned. @@ -960,6 +968,8 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev) pcie_set_clkpm(link, policy_to_clkpm_state(link)); } + pcie_aspm_update_sysfs_visibility(pdev); + unlock: mutex_unlock(&aspm_lock); out: @@ -1313,6 +1323,145 @@ void pcie_aspm_remove_sysfs_dev_files(struct pci_dev *pdev) } #endif +static ssize_t aspm_attr_show_common(struct device *dev, + struct device_attribute *attr, + char *buf, u8 state) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct pcie_link_state *link = pcie_aspm_get_link(pdev); + + return sprintf(buf, "%d\n", (link->aspm_enabled & state) ? 1 : 0); +} + +static ssize_t aspm_attr_store_common(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t len, u8 state) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct pcie_link_state *link = pcie_aspm_get_link(pdev); + bool state_enable; + + if (strtobool(buf, &state_enable) < 0) + return -EINVAL; + + down_read(&pci_bus_sem); + mutex_lock(&aspm_lock); + + if (state_enable) { + link->aspm_disable &= ~state; + /* need to enable L1 for substates */ + if (state & ASPM_STATE_L1SS) + link->aspm_disable &= ~ASPM_STATE_L1; + } else { + link->aspm_disable |= state; + } + + pcie_config_aspm_link(link, policy_to_aspm_state(link)); + + mutex_unlock(&aspm_lock); + up_read(&pci_bus_sem); + + return len; +} + +#define ASPM_ATTR(_f, _s) \ +static ssize_t _f##_show(struct device *dev, \ + struct device_attribute *attr, char *buf) \ +{ return aspm_attr_show_common(dev, attr, buf, ASPM_STATE_##_s); } \ + \ +static ssize_t _f##_store(struct device *dev, \ + struct device_attribute *attr, \ + const char *buf, size_t len) \ +{ return aspm_attr_store_common(dev, attr, buf, len, ASPM_STATE_##_s); } + +ASPM_ATTR(l0s_aspm, L0S) +ASPM_ATTR(l1_aspm, L1) +ASPM_ATTR(l1_1_aspm, L1_1) +ASPM_ATTR(l1_2_aspm, L1_2) +ASPM_ATTR(l1_1_pcipm, L1_1_PCIPM) +ASPM_ATTR(l1_2_pcipm, L1_2_PCIPM) + +static ssize_t clkpm_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct pcie_link_state *link = pcie_aspm_get_link(pdev); + + return sprintf(buf, "%d\n", link->clkpm_enabled); +} + +static ssize_t clkpm_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t len) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct pcie_link_state *link = pcie_aspm_get_link(pdev); + bool state_enable; + + if (strtobool(buf, &state_enable) < 0) + return -EINVAL; + + down_read(&pci_bus_sem); + mutex_lock(&aspm_lock); + + link->clkpm_disable = !state_enable; + pcie_set_clkpm(link, policy_to_clkpm_state(link)); + + mutex_unlock(&aspm_lock); + up_read(&pci_bus_sem); + + return len; +} + +static DEVICE_ATTR_RW(clkpm); +static DEVICE_ATTR_RW(l0s_aspm); +static DEVICE_ATTR_RW(l1_aspm); +static DEVICE_ATTR_RW(l1_1_aspm); +static DEVICE_ATTR_RW(l1_2_aspm); +static DEVICE_ATTR_RW(l1_1_pcipm); +static DEVICE_ATTR_RW(l1_2_pcipm); + +static struct attribute *aspm_ctrl_attrs[] = { + &dev_attr_clkpm.attr, + &dev_attr_l0s_aspm.attr, + &dev_attr_l1_aspm.attr, + &dev_attr_l1_1_aspm.attr, + &dev_attr_l1_2_aspm.attr, + &dev_attr_l1_1_pcipm.attr, + &dev_attr_l1_2_pcipm.attr, + NULL +}; + +static umode_t aspm_ctrl_attrs_are_visible(struct kobject *kobj, + struct attribute *a, int n) +{ + struct device *dev = kobj_to_dev(kobj); + struct pci_dev *pdev = to_pci_dev(dev); + struct pcie_link_state *link = pcie_aspm_get_link(pdev); + static const u8 aspm_state_map[] = { + ASPM_STATE_L0S, + ASPM_STATE_L1, + ASPM_STATE_L1_1, + ASPM_STATE_L1_2, + ASPM_STATE_L1_1_PCIPM, + ASPM_STATE_L1_2_PCIPM, + }; + + if (aspm_disabled || !link) + return 0; + + if (n == 0) + return link->clkpm_capable ? a->mode : 0; + + return link->aspm_capable & aspm_state_map[n - 1] ? a->mode : 0; +} + +const struct attribute_group aspm_ctrl_attr_group = { + .name = "link", + .attrs = aspm_ctrl_attrs, + .is_visible = aspm_ctrl_attrs_are_visible, +}; + static int __init pcie_aspm_disable(char *str) { if (!strcmp(str, "off")) { From 87e90283c94c76ee11d379ab5a0973382bbd0baf Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Sat, 5 Oct 2019 14:08:52 +0200 Subject: [PATCH 149/160] PCI/ASPM: Remove PCIEASPM_DEBUG Kconfig option and related code Previously, CONFIG_PCIEASPM_DEBUG enabled "link_state" and "clk_ctl" sysfs files that controlled ASPM. We believe these files were rarely if ever used. We recently added sysfs ASPM controls that are always present, so the debug code is no longer needed. Removing this debug code has been discussed for quite some time, see e.g. [0]. Remove PCIEASPM_DEBUG and the related code. [0] https://lore.kernel.org/lkml/20180727202619.GD173328@bhelgaas-glaptop.roam.corp.google.com/ Link: https://lore.kernel.org/r/ec935d8e-c084-3938-f1d1-748617596b25@gmail.com Signed-off-by: Heiner Kallweit Signed-off-by: Bjorn Helgaas --- drivers/pci/pci-sysfs.c | 3 -- drivers/pci/pci.h | 8 --- drivers/pci/pcie/Kconfig | 7 --- drivers/pci/pcie/aspm.c | 105 --------------------------------------- 4 files changed, 123 deletions(-) diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 0e76c02e0b7d..92bf1b1ce280 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -1330,7 +1330,6 @@ static int pci_create_capabilities_sysfs(struct pci_dev *dev) int retval; pcie_vpd_create_sysfs_dev_files(dev); - pcie_aspm_create_sysfs_dev_files(dev); if (dev->reset_fn) { retval = device_create_file(&dev->dev, &dev_attr_reset); @@ -1340,7 +1339,6 @@ static int pci_create_capabilities_sysfs(struct pci_dev *dev) return 0; error: - pcie_aspm_remove_sysfs_dev_files(dev); pcie_vpd_remove_sysfs_dev_files(dev); return retval; } @@ -1416,7 +1414,6 @@ err: static void pci_remove_capabilities_sysfs(struct pci_dev *dev) { pcie_vpd_remove_sysfs_dev_files(dev); - pcie_aspm_remove_sysfs_dev_files(dev); if (dev->reset_fn) { device_remove_file(&dev->dev, &dev_attr_reset); dev->reset_fn = 0; diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index b2cd21e8cb51..ae231e3cdd69 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -541,14 +541,6 @@ static inline void pcie_aspm_pm_state_change(struct pci_dev *pdev) { } static inline void pcie_aspm_powersave_config_link(struct pci_dev *pdev) { } #endif -#ifdef CONFIG_PCIEASPM_DEBUG -void pcie_aspm_create_sysfs_dev_files(struct pci_dev *pdev); -void pcie_aspm_remove_sysfs_dev_files(struct pci_dev *pdev); -#else -static inline void pcie_aspm_create_sysfs_dev_files(struct pci_dev *pdev) { } -static inline void pcie_aspm_remove_sysfs_dev_files(struct pci_dev *pdev) { } -#endif - #ifdef CONFIG_PCIE_ECRC void pcie_set_ecrc_checking(struct pci_dev *dev); void pcie_ecrc_get_policy(char *str); diff --git a/drivers/pci/pcie/Kconfig b/drivers/pci/pcie/Kconfig index 362eb8cfa53b..a2e862d4ec21 100644 --- a/drivers/pci/pcie/Kconfig +++ b/drivers/pci/pcie/Kconfig @@ -79,13 +79,6 @@ config PCIEASPM When in doubt, say Y. -config PCIEASPM_DEBUG - bool "Debug PCI Express ASPM" - depends on PCIEASPM - help - This enables PCI Express ASPM debug support. It will add per-device - interface to control ASPM. - choice prompt "Default ASPM policy" default PCIEASPM_DEFAULT diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index 9cd251827a67..0dcd44308228 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -1218,111 +1218,6 @@ bool pcie_aspm_enabled(struct pci_dev *pdev) } EXPORT_SYMBOL_GPL(pcie_aspm_enabled); -#ifdef CONFIG_PCIEASPM_DEBUG -static ssize_t link_state_show(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct pci_dev *pci_device = to_pci_dev(dev); - struct pcie_link_state *link_state = pci_device->link_state; - - return sprintf(buf, "%d\n", link_state->aspm_enabled); -} - -static ssize_t link_state_store(struct device *dev, - struct device_attribute *attr, - const char *buf, - size_t n) -{ - struct pci_dev *pdev = to_pci_dev(dev); - struct pcie_link_state *link, *root = pdev->link_state->root; - u32 state; - - if (aspm_disabled) - return -EPERM; - - if (kstrtouint(buf, 10, &state)) - return -EINVAL; - if ((state & ~ASPM_STATE_ALL) != 0) - return -EINVAL; - - down_read(&pci_bus_sem); - mutex_lock(&aspm_lock); - list_for_each_entry(link, &link_list, sibling) { - if (link->root != root) - continue; - pcie_config_aspm_link(link, state); - } - mutex_unlock(&aspm_lock); - up_read(&pci_bus_sem); - return n; -} - -static ssize_t clk_ctl_show(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct pci_dev *pci_device = to_pci_dev(dev); - struct pcie_link_state *link_state = pci_device->link_state; - - return sprintf(buf, "%d\n", link_state->clkpm_enabled); -} - -static ssize_t clk_ctl_store(struct device *dev, - struct device_attribute *attr, - const char *buf, - size_t n) -{ - struct pci_dev *pdev = to_pci_dev(dev); - bool state; - - if (strtobool(buf, &state)) - return -EINVAL; - - down_read(&pci_bus_sem); - mutex_lock(&aspm_lock); - pcie_set_clkpm_nocheck(pdev->link_state, state); - mutex_unlock(&aspm_lock); - up_read(&pci_bus_sem); - - return n; -} - -static DEVICE_ATTR_RW(link_state); -static DEVICE_ATTR_RW(clk_ctl); - -static char power_group[] = "power"; -void pcie_aspm_create_sysfs_dev_files(struct pci_dev *pdev) -{ - struct pcie_link_state *link_state = pdev->link_state; - - if (!link_state) - return; - - if (link_state->aspm_support) - sysfs_add_file_to_group(&pdev->dev.kobj, - &dev_attr_link_state.attr, power_group); - if (link_state->clkpm_capable) - sysfs_add_file_to_group(&pdev->dev.kobj, - &dev_attr_clk_ctl.attr, power_group); -} - -void pcie_aspm_remove_sysfs_dev_files(struct pci_dev *pdev) -{ - struct pcie_link_state *link_state = pdev->link_state; - - if (!link_state) - return; - - if (link_state->aspm_support) - sysfs_remove_file_from_group(&pdev->dev.kobj, - &dev_attr_link_state.attr, power_group); - if (link_state->clkpm_capable) - sysfs_remove_file_from_group(&pdev->dev.kobj, - &dev_attr_clk_ctl.attr, power_group); -} -#endif - static ssize_t aspm_attr_show_common(struct device *dev, struct device_attribute *attr, char *buf, u8 state) From e3d023b8952b78b82b1f9c6d7b708680a99ea831 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Sat, 23 Nov 2019 01:25:01 +0900 Subject: [PATCH 150/160] MAINTAINERS: Remove Keith from VMD maintainer I no longer work in this capacity on the VMD driver. Signed-off-by: Keith Busch Signed-off-by: Lorenzo Pieralisi Acked-by: Jon Derrick --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 296de2b51c83..25211e94b0a9 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -12442,7 +12442,6 @@ F: Documentation/devicetree/bindings/pci/fsl,imx6q-pcie.txt F: drivers/pci/controller/dwc/*imx6* PCI DRIVER FOR INTEL VOLUME MANAGEMENT DEVICE (VMD) -M: Keith Busch M: Jonathan Derrick L: linux-pci@vger.kernel.org S: Supported From a8e37506e79a7a08d0ee217f389fa1710e7a2ea4 Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Sun, 24 Nov 2019 21:33:51 -0800 Subject: [PATCH 151/160] PCI: hv: Reorganize the code in preparation of hibernation There is no functional change. This is just preparatory for a later patch which adds the hibernation support for the pci-hyperv driver. Signed-off-by: Dexuan Cui Signed-off-by: Lorenzo Pieralisi Reviewed-by: Michael Kelley --- drivers/pci/controller/pci-hyperv.c | 43 +++++++++++++++++++---------- 1 file changed, 28 insertions(+), 15 deletions(-) diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c index f1f300218fab..65f18f840ce9 100644 --- a/drivers/pci/controller/pci-hyperv.c +++ b/drivers/pci/controller/pci-hyperv.c @@ -2379,7 +2379,9 @@ static void hv_pci_onchannelcallback(void *context) * failing if the host doesn't support the necessary protocol * level. */ -static int hv_pci_protocol_negotiation(struct hv_device *hdev) +static int hv_pci_protocol_negotiation(struct hv_device *hdev, + enum pci_protocol_version_t version[], + int num_version) { struct pci_version_request *version_req; struct hv_pci_compl comp_pkt; @@ -2403,8 +2405,8 @@ static int hv_pci_protocol_negotiation(struct hv_device *hdev) version_req = (struct pci_version_request *)&pkt->message; version_req->message_type.type = PCI_QUERY_PROTOCOL_VERSION; - for (i = 0; i < ARRAY_SIZE(pci_protocol_versions); i++) { - version_req->protocol_version = pci_protocol_versions[i]; + for (i = 0; i < num_version; i++) { + version_req->protocol_version = version[i]; ret = vmbus_sendpacket(hdev->channel, version_req, sizeof(struct pci_version_request), (unsigned long)pkt, VM_PKT_DATA_INBAND, @@ -2420,7 +2422,7 @@ static int hv_pci_protocol_negotiation(struct hv_device *hdev) } if (comp_pkt.completion_status >= 0) { - pci_protocol_version = pci_protocol_versions[i]; + pci_protocol_version = version[i]; dev_info(&hdev->device, "PCI VMBus probing: Using version %#x\n", pci_protocol_version); @@ -2930,7 +2932,8 @@ static int hv_pci_probe(struct hv_device *hdev, hv_set_drvdata(hdev, hbus); - ret = hv_pci_protocol_negotiation(hdev); + ret = hv_pci_protocol_negotiation(hdev, pci_protocol_versions, + ARRAY_SIZE(pci_protocol_versions)); if (ret) goto close; @@ -3011,7 +3014,7 @@ free_bus: return ret; } -static void hv_pci_bus_exit(struct hv_device *hdev) +static int hv_pci_bus_exit(struct hv_device *hdev, bool hibernating) { struct hv_pcibus_device *hbus = hv_get_drvdata(hdev); struct { @@ -3027,16 +3030,20 @@ static void hv_pci_bus_exit(struct hv_device *hdev) * access the per-channel ringbuffer any longer. */ if (hdev->channel->rescind) - return; + return 0; - /* Delete any children which might still exist. */ - memset(&relations, 0, sizeof(relations)); - hv_pci_devices_present(hbus, &relations); + if (!hibernating) { + /* Delete any children which might still exist. */ + memset(&relations, 0, sizeof(relations)); + hv_pci_devices_present(hbus, &relations); + } ret = hv_send_resources_released(hdev); - if (ret) + if (ret) { dev_err(&hdev->device, "Couldn't send resources released packet(s)\n"); + return ret; + } memset(&pkt.teardown_packet, 0, sizeof(pkt.teardown_packet)); init_completion(&comp_pkt.host_event); @@ -3049,8 +3056,13 @@ static void hv_pci_bus_exit(struct hv_device *hdev) (unsigned long)&pkt.teardown_packet, VM_PKT_DATA_INBAND, VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED); - if (!ret) - wait_for_completion_timeout(&comp_pkt.host_event, 10 * HZ); + if (ret) + return ret; + + if (wait_for_completion_timeout(&comp_pkt.host_event, 10 * HZ) == 0) + return -ETIMEDOUT; + + return 0; } /** @@ -3062,6 +3074,7 @@ static void hv_pci_bus_exit(struct hv_device *hdev) static int hv_pci_remove(struct hv_device *hdev) { struct hv_pcibus_device *hbus; + int ret; hbus = hv_get_drvdata(hdev); if (hbus->state == hv_pcibus_installed) { @@ -3074,7 +3087,7 @@ static int hv_pci_remove(struct hv_device *hdev) hbus->state = hv_pcibus_removed; } - hv_pci_bus_exit(hdev); + ret = hv_pci_bus_exit(hdev, false); vmbus_close(hdev->channel); @@ -3091,7 +3104,7 @@ static int hv_pci_remove(struct hv_device *hdev) hv_put_dom_num(hbus->sysdata.domain); free_page((unsigned long)hbus); - return 0; + return ret; } static const struct hv_vmbus_device_id hv_pci_id_table[] = { From ac82fc83270884adea31d9dec22db09392058bf7 Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Sun, 24 Nov 2019 21:33:52 -0800 Subject: [PATCH 152/160] PCI: hv: Add hibernation support Add suspend() and resume() functions so that Hyper-V virtual PCI devices are handled properly when the VM hibernates and resumes from hibernation. Note that the suspend() function must make sure there are no pending work items before calling vmbus_close(), since it runs in a process context as a callback in dpm_suspend(). When it starts to run, the channel callback hv_pci_onchannelcallback(), which runs in a tasklet context, can be still running concurrently and scheduling new work items onto hbus->wq in hv_pci_devices_present() and hv_pci_eject_device(), and the work item handlers can access the vmbus channel, which can be being closed by hv_pci_suspend(), e.g. the work item handler pci_devices_present_work() -> new_pcichild_device() writes to the vmbus channel. To eliminate the race, hv_pci_suspend() disables the channel callback tasklet, sets hbus->state to hv_pcibus_removing, and re-enables the tasklet. This way, when hv_pci_suspend() proceeds, it knows that no new work item can be scheduled, and then it flushes hbus->wq and safely closes the vmbus channel. Signed-off-by: Dexuan Cui Signed-off-by: Lorenzo Pieralisi Reviewed-by: Michael Kelley --- drivers/pci/controller/pci-hyperv.c | 125 +++++++++++++++++++++++++++- 1 file changed, 123 insertions(+), 2 deletions(-) diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c index 65f18f840ce9..3c4996b073ca 100644 --- a/drivers/pci/controller/pci-hyperv.c +++ b/drivers/pci/controller/pci-hyperv.c @@ -455,6 +455,7 @@ enum hv_pcibus_state { hv_pcibus_init = 0, hv_pcibus_probed, hv_pcibus_installed, + hv_pcibus_removing, hv_pcibus_removed, hv_pcibus_maximum }; @@ -1681,6 +1682,23 @@ static void prepopulate_bars(struct hv_pcibus_device *hbus) spin_lock_irqsave(&hbus->device_list_lock, flags); + /* + * Clear the memory enable bit, in case it's already set. This occurs + * in the suspend path of hibernation, where the device is suspended, + * resumed and suspended again: see hibernation_snapshot() and + * hibernation_platform_enter(). + * + * If the memory enable bit is already set, Hyper-V sliently ignores + * the below BAR updates, and the related PCI device driver can not + * work, because reading from the device register(s) always returns + * 0xFFFFFFFF. + */ + list_for_each_entry(hpdev, &hbus->children, list_entry) { + _hv_pcifront_read_config(hpdev, PCI_COMMAND, 2, &command); + command &= ~PCI_COMMAND_MEMORY; + _hv_pcifront_write_config(hpdev, PCI_COMMAND, 2, command); + } + /* Pick addresses for the BARs. */ do { list_for_each_entry(hpdev, &hbus->children, list_entry) { @@ -2107,6 +2125,12 @@ static void hv_pci_devices_present(struct hv_pcibus_device *hbus, unsigned long flags; bool pending_dr; + if (hbus->state == hv_pcibus_removing) { + dev_info(&hbus->hdev->device, + "PCI VMBus BUS_RELATIONS: ignored\n"); + return; + } + dr_wrk = kzalloc(sizeof(*dr_wrk), GFP_NOWAIT); if (!dr_wrk) return; @@ -2223,11 +2247,19 @@ static void hv_eject_device_work(struct work_struct *work) */ static void hv_pci_eject_device(struct hv_pci_dev *hpdev) { + struct hv_pcibus_device *hbus = hpdev->hbus; + struct hv_device *hdev = hbus->hdev; + + if (hbus->state == hv_pcibus_removing) { + dev_info(&hdev->device, "PCI VMBus EJECT: ignored\n"); + return; + } + hpdev->state = hv_pcichild_ejecting; get_pcichild(hpdev); INIT_WORK(&hpdev->wrk, hv_eject_device_work); - get_hvpcibus(hpdev->hbus); - queue_work(hpdev->hbus->wq, &hpdev->wrk); + get_hvpcibus(hbus); + queue_work(hbus->wq, &hpdev->wrk); } /** @@ -3107,6 +3139,93 @@ static int hv_pci_remove(struct hv_device *hdev) return ret; } +static int hv_pci_suspend(struct hv_device *hdev) +{ + struct hv_pcibus_device *hbus = hv_get_drvdata(hdev); + enum hv_pcibus_state old_state; + int ret; + + /* + * hv_pci_suspend() must make sure there are no pending work items + * before calling vmbus_close(), since it runs in a process context + * as a callback in dpm_suspend(). When it starts to run, the channel + * callback hv_pci_onchannelcallback(), which runs in a tasklet + * context, can be still running concurrently and scheduling new work + * items onto hbus->wq in hv_pci_devices_present() and + * hv_pci_eject_device(), and the work item handlers can access the + * vmbus channel, which can be being closed by hv_pci_suspend(), e.g. + * the work item handler pci_devices_present_work() -> + * new_pcichild_device() writes to the vmbus channel. + * + * To eliminate the race, hv_pci_suspend() disables the channel + * callback tasklet, sets hbus->state to hv_pcibus_removing, and + * re-enables the tasklet. This way, when hv_pci_suspend() proceeds, + * it knows that no new work item can be scheduled, and then it flushes + * hbus->wq and safely closes the vmbus channel. + */ + tasklet_disable(&hdev->channel->callback_event); + + /* Change the hbus state to prevent new work items. */ + old_state = hbus->state; + if (hbus->state == hv_pcibus_installed) + hbus->state = hv_pcibus_removing; + + tasklet_enable(&hdev->channel->callback_event); + + if (old_state != hv_pcibus_installed) + return -EINVAL; + + flush_workqueue(hbus->wq); + + ret = hv_pci_bus_exit(hdev, true); + if (ret) + return ret; + + vmbus_close(hdev->channel); + + return 0; +} + +static int hv_pci_resume(struct hv_device *hdev) +{ + struct hv_pcibus_device *hbus = hv_get_drvdata(hdev); + enum pci_protocol_version_t version[1]; + int ret; + + hbus->state = hv_pcibus_init; + + ret = vmbus_open(hdev->channel, pci_ring_size, pci_ring_size, NULL, 0, + hv_pci_onchannelcallback, hbus); + if (ret) + return ret; + + /* Only use the version that was in use before hibernation. */ + version[0] = pci_protocol_version; + ret = hv_pci_protocol_negotiation(hdev, version, 1); + if (ret) + goto out; + + ret = hv_pci_query_relations(hdev); + if (ret) + goto out; + + ret = hv_pci_enter_d0(hdev); + if (ret) + goto out; + + ret = hv_send_resources_allocated(hdev); + if (ret) + goto out; + + prepopulate_bars(hbus); + + hbus->state = hv_pcibus_installed; + return 0; +out: + vmbus_close(hdev->channel); + return ret; +} + static const struct hv_vmbus_device_id hv_pci_id_table[] = { /* PCI Pass-through Class ID */ /* 44C4F61D-4444-4400-9D52-802E27EDE19F */ @@ -3121,6 +3240,8 @@ static struct hv_driver hv_pci_drv = { .id_table = hv_pci_id_table, .probe = hv_pci_probe, .remove = hv_pci_remove, + .suspend = hv_pci_suspend, + .resume = hv_pci_resume, }; static void __exit exit_hv_pci_drv(void) From 14ef39fddd2367523de25b87dccfe6422c3f3efa Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Sun, 24 Nov 2019 21:33:53 -0800 Subject: [PATCH 153/160] PCI: hv: Change pci_protocol_version to per-hbus A VM can have multiple Hyper-V hbus. It's incorrect to set the global variable 'pci_protocol_version' when *every* hbus is initialized in hv_pci_protocol_negotiation(). This is not an issue in practice since every hbus should have the same value of hbus->protocol_version, but we should make the variable per-hbus, so in case we have busses with different protocol versions, the driver can still work correctly. Signed-off-by: Dexuan Cui Signed-off-by: Lorenzo Pieralisi Reviewed-by: Michael Kelley --- drivers/pci/controller/pci-hyperv.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c index 3c4996b073ca..910fa016d095 100644 --- a/drivers/pci/controller/pci-hyperv.c +++ b/drivers/pci/controller/pci-hyperv.c @@ -76,11 +76,6 @@ static enum pci_protocol_version_t pci_protocol_versions[] = { PCI_PROTOCOL_VERSION_1_1, }; -/* - * Protocol version negotiated by hv_pci_protocol_negotiation(). - */ -static enum pci_protocol_version_t pci_protocol_version; - #define PCI_CONFIG_MMIO_LENGTH 0x2000 #define CFG_PAGE_OFFSET 0x1000 #define CFG_PAGE_SIZE (PCI_CONFIG_MMIO_LENGTH - CFG_PAGE_OFFSET) @@ -462,6 +457,8 @@ enum hv_pcibus_state { struct hv_pcibus_device { struct pci_sysdata sysdata; + /* Protocol version negotiated with the host */ + enum pci_protocol_version_t protocol_version; enum hv_pcibus_state state; refcount_t remove_lock; struct hv_device *hdev; @@ -1225,7 +1222,7 @@ static void hv_irq_unmask(struct irq_data *data) * negative effect (yet?). */ - if (pci_protocol_version >= PCI_PROTOCOL_VERSION_1_2) { + if (hbus->protocol_version >= PCI_PROTOCOL_VERSION_1_2) { /* * PCI_PROTOCOL_VERSION_1_2 supports the VP_SET version of the * HVCALL_RETARGET_INTERRUPT hypercall, which also coincides @@ -1395,7 +1392,7 @@ static void hv_compose_msi_msg(struct irq_data *data, struct msi_msg *msg) ctxt.pci_pkt.completion_func = hv_pci_compose_compl; ctxt.pci_pkt.compl_ctxt = ∁ - switch (pci_protocol_version) { + switch (hbus->protocol_version) { case PCI_PROTOCOL_VERSION_1_1: size = hv_compose_msi_req_v1(&ctxt.int_pkts.v1, dest, @@ -2415,6 +2412,7 @@ static int hv_pci_protocol_negotiation(struct hv_device *hdev, enum pci_protocol_version_t version[], int num_version) { + struct hv_pcibus_device *hbus = hv_get_drvdata(hdev); struct pci_version_request *version_req; struct hv_pci_compl comp_pkt; struct pci_packet *pkt; @@ -2454,10 +2452,10 @@ static int hv_pci_protocol_negotiation(struct hv_device *hdev, } if (comp_pkt.completion_status >= 0) { - pci_protocol_version = version[i]; + hbus->protocol_version = version[i]; dev_info(&hdev->device, "PCI VMBus probing: Using version %#x\n", - pci_protocol_version); + hbus->protocol_version); goto exit; } @@ -2741,7 +2739,7 @@ static int hv_send_resources_allocated(struct hv_device *hdev) u32 wslot; int ret; - size_res = (pci_protocol_version < PCI_PROTOCOL_VERSION_1_2) + size_res = (hbus->protocol_version < PCI_PROTOCOL_VERSION_1_2) ? sizeof(*res_assigned) : sizeof(*res_assigned2); pkt = kmalloc(sizeof(*pkt) + size_res, GFP_KERNEL); @@ -2760,7 +2758,7 @@ static int hv_send_resources_allocated(struct hv_device *hdev) pkt->completion_func = hv_pci_generic_compl; pkt->compl_ctxt = &comp_pkt; - if (pci_protocol_version < PCI_PROTOCOL_VERSION_1_2) { + if (hbus->protocol_version < PCI_PROTOCOL_VERSION_1_2) { res_assigned = (struct pci_resources_assigned *)&pkt->message; res_assigned->message_type.type = @@ -3200,7 +3198,7 @@ static int hv_pci_resume(struct hv_device *hdev) return ret; /* Only use the version that was in use before hibernation. */ - version[0] = pci_protocol_version; + version[0] = hbus->protocol_version; ret = hv_pci_protocol_negotiation(hdev, version, 1); if (ret) goto out; From 877b911a5ba0733f62239055ee869f2e117b57da Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Sun, 24 Nov 2019 21:33:54 -0800 Subject: [PATCH 154/160] PCI: hv: Avoid a kmemleak false positive caused by the hbus buffer With the recent 59bb47985c1d ("mm, sl[aou]b: guarantee natural alignment for kmalloc(power-of-two)"), kzalloc() is able to allocate a 4KB buffer that is guaranteed to be 4KB-aligned. Here the size and alignment of hbus is important because hbus's field retarget_msi_interrupt_params must not cross a 4KB page boundary. Here we prefer kzalloc to get_zeroed_page(), because a buffer allocated by the latter is not tracked and scanned by kmemleak, and hence kmemleak reports the pointer contained in the hbus buffer (i.e. the hpdev struct, which is created in new_pcichild_device() and is tracked by hbus->children) as memory leak (false positive). If the kernel doesn't have 59bb47985c1d, get_zeroed_page() *must* be used to allocate the hbus buffer and we can avoid the kmemleak false positive by using kmemleak_alloc() and kmemleak_free() to ask kmemleak to track and scan the hbus buffer. Reported-by: Lili Deng Signed-off-by: Dexuan Cui Signed-off-by: Lorenzo Pieralisi Reviewed-by: Michael Kelley --- drivers/pci/controller/pci-hyperv.c | 24 +++++++++++++++++++++--- 1 file changed, 21 insertions(+), 3 deletions(-) diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c index 910fa016d095..be99862166d0 100644 --- a/drivers/pci/controller/pci-hyperv.c +++ b/drivers/pci/controller/pci-hyperv.c @@ -2902,9 +2902,27 @@ static int hv_pci_probe(struct hv_device *hdev, * hv_pcibus_device contains the hypercall arguments for retargeting in * hv_irq_unmask(). Those must not cross a page boundary. */ - BUILD_BUG_ON(sizeof(*hbus) > PAGE_SIZE); + BUILD_BUG_ON(sizeof(*hbus) > HV_HYP_PAGE_SIZE); - hbus = (struct hv_pcibus_device *)get_zeroed_page(GFP_KERNEL); + /* + * With the recent 59bb47985c1d ("mm, sl[aou]b: guarantee natural + * alignment for kmalloc(power-of-two)"), kzalloc() is able to allocate + * a 4KB buffer that is guaranteed to be 4KB-aligned. Here the size and + * alignment of hbus is important because hbus's field + * retarget_msi_interrupt_params must not cross a 4KB page boundary. + * + * Here we prefer kzalloc to get_zeroed_page(), because a buffer + * allocated by the latter is not tracked and scanned by kmemleak, and + * hence kmemleak reports the pointer contained in the hbus buffer + * (i.e. the hpdev struct, which is created in new_pcichild_device() and + * is tracked by hbus->children) as memory leak (false positive). + * + * If the kernel doesn't have 59bb47985c1d, get_zeroed_page() *must* be + * used to allocate the hbus buffer and we can avoid the kmemleak false + * positive by using kmemleak_alloc() and kmemleak_free() to ask + * kmemleak to track and scan the hbus buffer. + */ + hbus = (struct hv_pcibus_device *)kzalloc(HV_HYP_PAGE_SIZE, GFP_KERNEL); if (!hbus) return -ENOMEM; hbus->state = hv_pcibus_init; @@ -3133,7 +3151,7 @@ static int hv_pci_remove(struct hv_device *hdev) hv_put_dom_num(hbus->sysdata.domain); - free_page((unsigned long)hbus); + kfree(hbus); return ret; } From d41b0e64d206f8948212b4d0f30c330db95c9636 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 14 Oct 2019 12:04:52 +0200 Subject: [PATCH 155/160] PCI/MSI: Remove unused pci_irq_get_node() The function pci_irq_get_node() is not used by anyone in the tree, so just delete it. Link: https://lore.kernel.org/r/20191014100452.GA6699@kroah.com Signed-off-by: Greg Kroah-Hartman Signed-off-by: Bjorn Helgaas Reviewed-by: Andrew Murray --- drivers/pci/msi.c | 16 ---------------- include/linux/pci.h | 6 ------ 2 files changed, 22 deletions(-) diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 0884bedcfc7a..f95fe23830f0 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -1315,22 +1315,6 @@ const struct cpumask *pci_irq_get_affinity(struct pci_dev *dev, int nr) } EXPORT_SYMBOL(pci_irq_get_affinity); -/** - * pci_irq_get_node - return the NUMA node of a particular MSI vector - * @pdev: PCI device to operate on - * @vec: device-relative interrupt vector index (0-based). - */ -int pci_irq_get_node(struct pci_dev *pdev, int vec) -{ - const struct cpumask *mask; - - mask = pci_irq_get_affinity(pdev, vec); - if (mask) - return local_memory_node(cpu_to_node(cpumask_first(mask))); - return dev_to_node(&pdev->dev); -} -EXPORT_SYMBOL(pci_irq_get_node); - struct pci_dev *msi_desc_to_pci_dev(struct msi_desc *desc) { return to_pci_dev(desc->dev); diff --git a/include/linux/pci.h b/include/linux/pci.h index f9088c89a534..755d8c0176b9 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1454,7 +1454,6 @@ int pci_alloc_irq_vectors_affinity(struct pci_dev *dev, unsigned int min_vecs, void pci_free_irq_vectors(struct pci_dev *dev); int pci_irq_vector(struct pci_dev *dev, unsigned int nr); const struct cpumask *pci_irq_get_affinity(struct pci_dev *pdev, int vec); -int pci_irq_get_node(struct pci_dev *pdev, int vec); #else static inline int pci_msi_vec_count(struct pci_dev *dev) { return -ENOSYS; } @@ -1497,11 +1496,6 @@ static inline const struct cpumask *pci_irq_get_affinity(struct pci_dev *pdev, { return cpu_possible_mask; } - -static inline int pci_irq_get_node(struct pci_dev *pdev, int vec) -{ - return first_online_node; -} #endif /** From 901c4ddbe277a766fd081ffbbd526d7b4bdbacdf Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Mon, 14 Oct 2019 16:17:05 -0500 Subject: [PATCH 156/160] PCI/MSI: Move power state check out of pci_msi_supported() 27e20603c54b ("PCI/MSI: Move D0 check into pci_msi_check_device()") moved the power state check into pci_msi_check_device(), which was subsequently renamed to pci_msi_supported(). This didn't change the behavior, since both callers checked the power state. However, it doesn't fit the current "pci_msi_supported()" name, which should return what the device is capable of, independent of the power state. Move the power state check back into the callers for readability. No functional change intended. Signed-off-by: Bjorn Helgaas Reviewed-by: Rafael J. Wysocki --- drivers/pci/msi.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index f95fe23830f0..fa90f08e55cf 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -861,7 +861,7 @@ static int pci_msi_supported(struct pci_dev *dev, int nvec) if (!pci_msi_enable) return 0; - if (!dev || dev->no_msi || dev->current_state != PCI_D0) + if (!dev || dev->no_msi) return 0; /* @@ -972,7 +972,7 @@ static int __pci_enable_msix(struct pci_dev *dev, struct msix_entry *entries, int nr_entries; int i, j; - if (!pci_msi_supported(dev, nvec)) + if (!pci_msi_supported(dev, nvec) || dev->current_state != PCI_D0) return -EINVAL; nr_entries = pci_msix_vec_count(dev); @@ -1058,7 +1058,7 @@ static int __pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec, int nvec; int rc; - if (!pci_msi_supported(dev, minvec)) + if (!pci_msi_supported(dev, minvec) || dev->current_state != PCI_D0) return -EINVAL; /* Check whether driver already requested MSI-X IRQs */ From e045fa29e89383c717e308609edd19d2fd29e1be Mon Sep 17 00:00:00 2001 From: Jian-Hong Pan Date: Tue, 8 Oct 2019 11:42:39 +0800 Subject: [PATCH 157/160] PCI/MSI: Fix incorrect MSI-X masking on resume When a driver enables MSI-X, msix_program_entries() reads the MSI-X Vector Control register for each vector and saves it in desc->masked. Each register is 32 bits and bit 0 is the actual Mask bit. When we restored these registers during resume, we previously set the Mask bit if *any* bit in desc->masked was set instead of when the Mask bit itself was set: pci_restore_state pci_restore_msi_state __pci_restore_msix_state for_each_pci_msi_entry msix_mask_irq(entry, entry->masked) <-- entire u32 word __pci_msix_desc_mask_irq(desc, flag) mask_bits = desc->masked & ~PCI_MSIX_ENTRY_CTRL_MASKBIT if (flag) <-- testing entire u32, not just bit 0 mask_bits |= PCI_MSIX_ENTRY_CTRL_MASKBIT writel(mask_bits, desc_addr + PCI_MSIX_ENTRY_VECTOR_CTRL) This means that after resume, MSI-X vectors were masked when they shouldn't be, which leads to timeouts like this: nvme nvme0: I/O 978 QID 3 timeout, completion polled On resume, set the Mask bit only when the saved Mask bit from suspend was set. This should remove the need for 19ea025e1d28 ("nvme: Add quirk for Kingston NVME SSD running FW E8FK11.T"). [bhelgaas: commit log, move fix to __pci_msix_desc_mask_irq()] Link: https://bugzilla.kernel.org/show_bug.cgi?id=204887 Link: https://lore.kernel.org/r/20191008034238.2503-1-jian-hong@endlessm.com Fixes: f2440d9acbe8 ("PCI MSI: Refactor interrupt masking code") Signed-off-by: Jian-Hong Pan Signed-off-by: Bjorn Helgaas Cc: stable@vger.kernel.org --- drivers/pci/msi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index fa90f08e55cf..c7709e49f0e4 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -213,12 +213,13 @@ u32 __pci_msix_desc_mask_irq(struct msi_desc *desc, u32 flag) if (pci_msi_ignore_mask) return 0; + desc_addr = pci_msix_desc_addr(desc); if (!desc_addr) return 0; mask_bits &= ~PCI_MSIX_ENTRY_CTRL_MASKBIT; - if (flag) + if (flag & PCI_MSIX_ENTRY_CTRL_MASKBIT) mask_bits |= PCI_MSIX_ENTRY_CTRL_MASKBIT; writel(mask_bits, desc_addr + PCI_MSIX_ENTRY_VECTOR_CTRL); From 655e7aee1f0398602627a485f7dca6c29cc96cae Mon Sep 17 00:00:00 2001 From: Jian-Hong Pan Date: Thu, 31 Oct 2019 17:34:09 +0800 Subject: [PATCH 158/160] Revert "nvme: Add quirk for Kingston NVME SSD running FW E8FK11.T" Since e045fa29e893 ("PCI/MSI: Fix incorrect MSI-X masking on resume") is merged, we can revert the previous quirk now. This reverts commit 19ea025e1d28c629b369c3532a85b3df478cc5c6. Buglink: https://bugzilla.kernel.org/show_bug.cgi?id=204887 Fixes: 19ea025e1d28 ("nvme: Add quirk for Kingston NVME SSD running FW E8FK11.T") Link: https://lore.kernel.org/r/20191031093408.9322-1-jian-hong@endlessm.com Signed-off-by: Jian-Hong Pan Signed-off-by: Bjorn Helgaas Acked-by: Christoph Hellwig Cc: stable@vger.kernel.org --- drivers/nvme/host/core.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index fa7ba09dca77..94bfbee1e5f7 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -2404,16 +2404,6 @@ static const struct nvme_core_quirk_entry core_quirks[] = { .vid = 0x14a4, .fr = "22301111", .quirks = NVME_QUIRK_SIMPLE_SUSPEND, - }, - { - /* - * This Kingston E8FK11.T firmware version has no interrupt - * after resume with actions related to suspend to idle - * https://bugzilla.kernel.org/show_bug.cgi?id=204887 - */ - .vid = 0x2646, - .fr = "E8FK11.T", - .quirks = NVME_QUIRK_SIMPLE_SUSPEND, } }; From a1b39bae16a62ce4aae02d958224f19316d98b24 Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Fri, 25 Oct 2019 08:10:37 +0200 Subject: [PATCH 159/160] asm-generic: Make msi.h a mandatory include/asm header msi.h is generic for all architectures except x86, which has its own version. Enabling MSI by adding msi.h to every architecture's Kbuild is just an additional step which doesn't need to be done. Make msi.h mandatory in the asm-generic/Kbuild so we don't have to do it for each architecture. Suggested-by: Christoph Hellwig Link: https://lore.kernel.org/r/c991669e29a79b1a8e28c3b4b3a125801a693de8.1571983829.git.michal.simek@xilinx.com Tested-by: Paul Walmsley # build only, rv32/rv64 Signed-off-by: Michal Simek Signed-off-by: Bjorn Helgaas Reviewed-by: Masahiro Yamada Acked-by: Waiman Long Acked-by: Paul Walmsley # arch/riscv --- arch/arc/include/asm/Kbuild | 1 - arch/arm/include/asm/Kbuild | 1 - arch/arm64/include/asm/Kbuild | 1 - arch/mips/include/asm/Kbuild | 1 - arch/powerpc/include/asm/Kbuild | 1 - arch/riscv/include/asm/Kbuild | 1 - arch/sparc/include/asm/Kbuild | 1 - include/asm-generic/Kbuild | 1 + 8 files changed, 1 insertion(+), 7 deletions(-) diff --git a/arch/arc/include/asm/Kbuild b/arch/arc/include/asm/Kbuild index 393d4f5e1450..1b505694691e 100644 --- a/arch/arc/include/asm/Kbuild +++ b/arch/arc/include/asm/Kbuild @@ -17,7 +17,6 @@ generic-y += local64.h generic-y += mcs_spinlock.h generic-y += mm-arch-hooks.h generic-y += mmiowb.h -generic-y += msi.h generic-y += parport.h generic-y += percpu.h generic-y += preempt.h diff --git a/arch/arm/include/asm/Kbuild b/arch/arm/include/asm/Kbuild index 68ca86f85eb7..fa579b23b4df 100644 --- a/arch/arm/include/asm/Kbuild +++ b/arch/arm/include/asm/Kbuild @@ -12,7 +12,6 @@ generic-y += local.h generic-y += local64.h generic-y += mm-arch-hooks.h generic-y += mmiowb.h -generic-y += msi.h generic-y += parport.h generic-y += preempt.h generic-y += seccomp.h diff --git a/arch/arm64/include/asm/Kbuild b/arch/arm64/include/asm/Kbuild index 98a5405c8558..bd23f87d6c55 100644 --- a/arch/arm64/include/asm/Kbuild +++ b/arch/arm64/include/asm/Kbuild @@ -16,7 +16,6 @@ generic-y += local64.h generic-y += mcs_spinlock.h generic-y += mm-arch-hooks.h generic-y += mmiowb.h -generic-y += msi.h generic-y += qrwlock.h generic-y += qspinlock.h generic-y += serial.h diff --git a/arch/mips/include/asm/Kbuild b/arch/mips/include/asm/Kbuild index c8b595c60910..61b0fc2026e6 100644 --- a/arch/mips/include/asm/Kbuild +++ b/arch/mips/include/asm/Kbuild @@ -13,7 +13,6 @@ generic-y += irq_work.h generic-y += local64.h generic-y += mcs_spinlock.h generic-y += mm-arch-hooks.h -generic-y += msi.h generic-y += parport.h generic-y += percpu.h generic-y += preempt.h diff --git a/arch/powerpc/include/asm/Kbuild b/arch/powerpc/include/asm/Kbuild index 64870c7be4a3..17726f2e46de 100644 --- a/arch/powerpc/include/asm/Kbuild +++ b/arch/powerpc/include/asm/Kbuild @@ -10,4 +10,3 @@ generic-y += local64.h generic-y += mcs_spinlock.h generic-y += preempt.h generic-y += vtime.h -generic-y += msi.h diff --git a/arch/riscv/include/asm/Kbuild b/arch/riscv/include/asm/Kbuild index 16970f246860..1efaeddf1e4b 100644 --- a/arch/riscv/include/asm/Kbuild +++ b/arch/riscv/include/asm/Kbuild @@ -22,7 +22,6 @@ generic-y += kvm_para.h generic-y += local.h generic-y += local64.h generic-y += mm-arch-hooks.h -generic-y += msi.h generic-y += percpu.h generic-y += preempt.h generic-y += sections.h diff --git a/arch/sparc/include/asm/Kbuild b/arch/sparc/include/asm/Kbuild index b6212164847b..62de2eb2773d 100644 --- a/arch/sparc/include/asm/Kbuild +++ b/arch/sparc/include/asm/Kbuild @@ -18,7 +18,6 @@ generic-y += mcs_spinlock.h generic-y += mm-arch-hooks.h generic-y += mmiowb.h generic-y += module.h -generic-y += msi.h generic-y += preempt.h generic-y += serial.h generic-y += trace_clock.h diff --git a/include/asm-generic/Kbuild b/include/asm-generic/Kbuild index adff14fcb8e4..ddfee1bd9dc1 100644 --- a/include/asm-generic/Kbuild +++ b/include/asm-generic/Kbuild @@ -4,4 +4,5 @@ # (This file is not included when SRCARCH=um since UML borrows several # asm headers from the host architecutre.) +mandatory-y += msi.h mandatory-y += simd.h From 191d6f91f283dfb007499bb8529d54c3ac434bd7 Mon Sep 17 00:00:00 2001 From: Palmer Dabbelt Date: Fri, 25 Oct 2019 08:10:38 +0200 Subject: [PATCH 160/160] PCI: Remove PCI_MSI_IRQ_DOMAIN architecture whitelist The only apparent reason for the PCI_MSI_IRQ_DOMAIN architecture whitelist was that it requires msi.h. Now that msi.h is mandatory in asm-generic/Kbuild, every arch should have at least the default version, so remove the whitelist. Built for all the architectures that play nice with make.cross, but not boot tested anywhere. Link: https://lore.kernel.org/r/514e7b040be8ccd69088193aba260da1b89e919c.1571983829.git.michal.simek@xilinx.com Signed-off-by: Palmer Dabbelt Signed-off-by: Michal Simek Signed-off-by: Bjorn Helgaas Acked-by: Waiman Long --- drivers/pci/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index a304f5ea11b9..77c1428cd945 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -52,7 +52,7 @@ config PCI_MSI If you don't know what to do here, say Y. config PCI_MSI_IRQ_DOMAIN - def_bool ARC || ARM || ARM64 || X86 || RISCV + def_bool y depends on PCI_MSI select GENERIC_MSI_IRQ_DOMAIN