From 55d554f5d14071f7c2c5dbd88d0a2eb695c97d16 Mon Sep 17 00:00:00 2001 From: Shuah Khan Date: Thu, 26 Sep 2019 19:13:44 -0600 Subject: [PATCH 001/145] tools: bpf: Use !building_out_of_srctree to determine srctree make TARGETS=bpf kselftest fails with: Makefile:127: tools/build/Makefile.include: No such file or directory When the bpf tool make is invoked from tools Makefile, srctree is cleared and the current logic check for srctree equals to empty string to determine srctree location from CURDIR. When the build in invoked from selftests/bpf Makefile, the srctree is set to "." and the same logic used for srctree equals to empty is needed to determine srctree. Check building_out_of_srctree undefined as the condition for both cases to fix "make TARGETS=bpf kselftest" build failure. Signed-off-by: Shuah Khan Signed-off-by: Daniel Borkmann Acked-by: Song Liu Link: https://lore.kernel.org/bpf/20190927011344.4695-1-skhan@linuxfoundation.org --- tools/bpf/Makefile | 6 +++++- tools/lib/bpf/Makefile | 6 +++++- 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/tools/bpf/Makefile b/tools/bpf/Makefile index fbf5e4a0cb9c..5d1995fd369c 100644 --- a/tools/bpf/Makefile +++ b/tools/bpf/Makefile @@ -12,7 +12,11 @@ INSTALL ?= install CFLAGS += -Wall -O2 CFLAGS += -D__EXPORTED_HEADERS__ -I$(srctree)/include/uapi -I$(srctree)/include -ifeq ($(srctree),) +# This will work when bpf is built in tools env. where srctree +# isn't set and when invoked from selftests build, where srctree +# is set to ".". building_out_of_srctree is undefined for in srctree +# builds +ifndef building_out_of_srctree srctree := $(patsubst %/,%,$(dir $(CURDIR))) srctree := $(patsubst %/,%,$(dir $(srctree))) endif diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile index c6f94cffe06e..20772663d3e1 100644 --- a/tools/lib/bpf/Makefile +++ b/tools/lib/bpf/Makefile @@ -8,7 +8,11 @@ LIBBPF_MAJOR_VERSION := $(firstword $(subst ., ,$(LIBBPF_VERSION))) MAKEFLAGS += --no-print-directory -ifeq ($(srctree),) +# This will work when bpf is built in tools env. where srctree +# isn't set and when invoked from selftests build, where srctree +# is a ".". building_out_of_srctree is undefined for in srctree +# builds +ifndef building_out_of_srctree srctree := $(patsubst %/,%,$(dir $(CURDIR))) srctree := $(patsubst %/,%,$(dir $(srctree))) srctree := $(patsubst %/,%,$(dir $(srctree))) From 1bd63524593b964934a33afd442df16b8f90e2b5 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Mon, 30 Sep 2019 14:02:03 -0700 Subject: [PATCH 002/145] libbpf: handle symbol versioning properly for libbpf.a bcc uses libbpf repo as a submodule. It brings in libbpf source code and builds everything together to produce shared libraries. With latest libbpf, I got the following errors: /bin/ld: libbcc_bpf.so.0.10.0: version node not found for symbol xsk_umem__create@LIBBPF_0.0.2 /bin/ld: failed to set dynamic section sizes: Bad value collect2: error: ld returned 1 exit status make[2]: *** [src/cc/libbcc_bpf.so.0.10.0] Error 1 In xsk.c, we have asm(".symver xsk_umem__create_v0_0_2, xsk_umem__create@LIBBPF_0.0.2"); asm(".symver xsk_umem__create_v0_0_4, xsk_umem__create@@LIBBPF_0.0.4"); The linker thinks the built is for LIBBPF but cannot find proper version LIBBPF_0.0.2/4, so emit errors. I also confirmed that using libbpf.a to produce a shared library also has issues: -bash-4.4$ cat t.c extern void *xsk_umem__create; void * test() { return xsk_umem__create; } -bash-4.4$ gcc -c -fPIC t.c -bash-4.4$ gcc -shared t.o libbpf.a -o t.so /bin/ld: t.so: version node not found for symbol xsk_umem__create@LIBBPF_0.0.2 /bin/ld: failed to set dynamic section sizes: Bad value collect2: error: ld returned 1 exit status -bash-4.4$ Symbol versioning does happens in commonly used libraries, e.g., elfutils and glibc. For static libraries, for a versioned symbol, the old definitions will be ignored, and the symbol will be an alias to the latest definition. For example, glibc sched_setaffinity is versioned. -bash-4.4$ readelf -s /usr/lib64/libc.so.6 | grep sched_setaffinity 756: 000000000013d3d0 13 FUNC GLOBAL DEFAULT 13 sched_setaffinity@GLIBC_2.3.3 757: 00000000000e2e70 455 FUNC GLOBAL DEFAULT 13 sched_setaffinity@@GLIBC_2.3.4 1800: 0000000000000000 0 FILE LOCAL DEFAULT ABS sched_setaffinity.c 4228: 00000000000e2e70 455 FUNC LOCAL DEFAULT 13 __sched_setaffinity_new 4648: 000000000013d3d0 13 FUNC LOCAL DEFAULT 13 __sched_setaffinity_old 7338: 000000000013d3d0 13 FUNC GLOBAL DEFAULT 13 sched_setaffinity@GLIBC_2 7380: 00000000000e2e70 455 FUNC GLOBAL DEFAULT 13 sched_setaffinity@@GLIBC_ -bash-4.4$ For static library, the definition of sched_setaffinity aliases to the new definition. -bash-4.4$ readelf -s /usr/lib64/libc.a | grep sched_setaffinity File: /usr/lib64/libc.a(sched_setaffinity.o) 8: 0000000000000000 455 FUNC GLOBAL DEFAULT 1 __sched_setaffinity_new 12: 0000000000000000 455 FUNC WEAK DEFAULT 1 sched_setaffinity For both elfutils and glibc, additional macros are used to control different handling of symbol versioning w.r.t static and shared libraries. For elfutils, the macro is SYMBOL_VERSIONING (https://sourceware.org/git/?p=elfutils.git;a=blob;f=lib/eu-config.h). For glibc, the macro is SHARED (https://sourceware.org/git/?p=glibc.git;a=blob;f=include/shlib-compat.h;hb=refs/heads/master) This patch used SHARED as the macro name. After this patch, the libbpf.a has -bash-4.4$ readelf -s libbpf.a | grep xsk_umem__create 372: 0000000000017145 1190 FUNC GLOBAL DEFAULT 1 xsk_umem__create_v0_0_4 405: 0000000000017145 1190 FUNC GLOBAL DEFAULT 1 xsk_umem__create 499: 00000000000175eb 103 FUNC GLOBAL DEFAULT 1 xsk_umem__create_v0_0_2 -bash-4.4$ No versioned symbols for xsk_umem__create. The libbpf.a can be used to build a shared library succesfully. -bash-4.4$ cat t.c extern void *xsk_umem__create; void * test() { return xsk_umem__create; } -bash-4.4$ gcc -c -fPIC t.c -bash-4.4$ gcc -shared t.o libbpf.a -o t.so -bash-4.4$ Fixes: 10d30e301732 ("libbpf: add flags to umem config") Cc: Kevin Laatz Cc: Arnaldo Carvalho de Melo Cc: Andrii Nakryiko Acked-by: Andrii Nakryiko Signed-off-by: Yonghong Song Signed-off-by: Alexei Starovoitov --- tools/lib/bpf/Makefile | 27 ++++++++++++++++++--------- tools/lib/bpf/libbpf_internal.h | 16 ++++++++++++++++ tools/lib/bpf/xsk.c | 4 ++-- 3 files changed, 36 insertions(+), 11 deletions(-) diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile index 20772663d3e1..56ce6292071b 100644 --- a/tools/lib/bpf/Makefile +++ b/tools/lib/bpf/Makefile @@ -114,6 +114,9 @@ override CFLAGS += $(INCLUDES) override CFLAGS += -fvisibility=hidden override CFLAGS += -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 +# flags specific for shared library +SHLIB_FLAGS := -DSHARED + ifeq ($(VERBOSE),1) Q = else @@ -130,14 +133,17 @@ all: export srctree OUTPUT CC LD CFLAGS V include $(srctree)/tools/build/Makefile.include -BPF_IN := $(OUTPUT)libbpf-in.o +SHARED_OBJDIR := $(OUTPUT)sharedobjs/ +STATIC_OBJDIR := $(OUTPUT)staticobjs/ +BPF_IN_SHARED := $(SHARED_OBJDIR)libbpf-in.o +BPF_IN_STATIC := $(STATIC_OBJDIR)libbpf-in.o VERSION_SCRIPT := libbpf.map LIB_TARGET := $(addprefix $(OUTPUT),$(LIB_TARGET)) LIB_FILE := $(addprefix $(OUTPUT),$(LIB_FILE)) PC_FILE := $(addprefix $(OUTPUT),$(PC_FILE)) -GLOBAL_SYM_COUNT = $(shell readelf -s --wide $(BPF_IN) | \ +GLOBAL_SYM_COUNT = $(shell readelf -s --wide $(BPF_IN_SHARED) | \ cut -d "@" -f1 | sed 's/_v[0-9]_[0-9]_[0-9].*//' | \ awk '/GLOBAL/ && /DEFAULT/ && !/UND/ {print $$8}' | \ sort -u | wc -l) @@ -159,7 +165,7 @@ all: fixdep all_cmd: $(CMD_TARGETS) check -$(BPF_IN): force elfdep bpfdep +$(BPF_IN_SHARED): force elfdep bpfdep @(test -f ../../include/uapi/linux/bpf.h -a -f ../../../include/uapi/linux/bpf.h && ( \ (diff -B ../../include/uapi/linux/bpf.h ../../../include/uapi/linux/bpf.h >/dev/null) || \ echo "Warning: Kernel ABI header at 'tools/include/uapi/linux/bpf.h' differs from latest version at 'include/uapi/linux/bpf.h'" >&2 )) || true @@ -175,17 +181,20 @@ $(BPF_IN): force elfdep bpfdep @(test -f ../../include/uapi/linux/if_xdp.h -a -f ../../../include/uapi/linux/if_xdp.h && ( \ (diff -B ../../include/uapi/linux/if_xdp.h ../../../include/uapi/linux/if_xdp.h >/dev/null) || \ echo "Warning: Kernel ABI header at 'tools/include/uapi/linux/if_xdp.h' differs from latest version at 'include/uapi/linux/if_xdp.h'" >&2 )) || true - $(Q)$(MAKE) $(build)=libbpf + $(Q)$(MAKE) $(build)=libbpf OUTPUT=$(SHARED_OBJDIR) CFLAGS="$(CFLAGS) $(SHLIB_FLAGS)" + +$(BPF_IN_STATIC): force elfdep bpfdep + $(Q)$(MAKE) $(build)=libbpf OUTPUT=$(STATIC_OBJDIR) $(OUTPUT)libbpf.so: $(OUTPUT)libbpf.so.$(LIBBPF_VERSION) -$(OUTPUT)libbpf.so.$(LIBBPF_VERSION): $(BPF_IN) +$(OUTPUT)libbpf.so.$(LIBBPF_VERSION): $(BPF_IN_SHARED) $(QUIET_LINK)$(CC) --shared -Wl,-soname,libbpf.so.$(LIBBPF_MAJOR_VERSION) \ -Wl,--version-script=$(VERSION_SCRIPT) $^ -lelf -o $@ @ln -sf $(@F) $(OUTPUT)libbpf.so @ln -sf $(@F) $(OUTPUT)libbpf.so.$(LIBBPF_MAJOR_VERSION) -$(OUTPUT)libbpf.a: $(BPF_IN) +$(OUTPUT)libbpf.a: $(BPF_IN_STATIC) $(QUIET_LINK)$(RM) $@; $(AR) rcs $@ $^ $(OUTPUT)test_libbpf: test_libbpf.cpp $(OUTPUT)libbpf.a @@ -201,7 +210,7 @@ check: check_abi check_abi: $(OUTPUT)libbpf.so @if [ "$(GLOBAL_SYM_COUNT)" != "$(VERSIONED_SYM_COUNT)" ]; then \ - echo "Warning: Num of global symbols in $(BPF_IN)" \ + echo "Warning: Num of global symbols in $(BPF_IN_SHARED)" \ "($(GLOBAL_SYM_COUNT)) does NOT match with num of" \ "versioned symbols in $^ ($(VERSIONED_SYM_COUNT))." \ "Please make sure all LIBBPF_API symbols are" \ @@ -259,9 +268,9 @@ config-clean: $(Q)$(MAKE) -C $(srctree)/tools/build/feature/ clean >/dev/null clean: - $(call QUIET_CLEAN, libbpf) $(RM) $(TARGETS) $(CXX_TEST_TARGET) \ + $(call QUIET_CLEAN, libbpf) $(RM) -rf $(TARGETS) $(CXX_TEST_TARGET) \ *.o *~ *.a *.so *.so.$(LIBBPF_MAJOR_VERSION) .*.d .*.cmd \ - *.pc LIBBPF-CFLAGS + *.pc LIBBPF-CFLAGS $(SHARED_OBJDIR) $(STATIC_OBJDIR) $(call QUIET_CLEAN, core-gen) $(RM) $(OUTPUT)FEATURE-DUMP.libbpf diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h index 2e83a34f8c79..98216a69c32f 100644 --- a/tools/lib/bpf/libbpf_internal.h +++ b/tools/lib/bpf/libbpf_internal.h @@ -34,6 +34,22 @@ (offsetof(TYPE, FIELD) + sizeof(((TYPE *)0)->FIELD)) #endif +/* Symbol versioning is different between static and shared library. + * Properly versioned symbols are needed for shared library, but + * only the symbol of the new version is needed for static library. + */ +#ifdef SHARED +# define COMPAT_VERSION(internal_name, api_name, version) \ + asm(".symver " #internal_name "," #api_name "@" #version); +# define DEFAULT_VERSION(internal_name, api_name, version) \ + asm(".symver " #internal_name "," #api_name "@@" #version); +#else +# define COMPAT_VERSION(internal_name, api_name, version) +# define DEFAULT_VERSION(internal_name, api_name, version) \ + extern typeof(internal_name) api_name \ + __attribute__((alias(#internal_name))); +#endif + extern void libbpf_print(enum libbpf_print_level level, const char *format, ...) __attribute__((format(printf, 2, 3))); diff --git a/tools/lib/bpf/xsk.c b/tools/lib/bpf/xsk.c index 24fa313524fb..a902838f9fcc 100644 --- a/tools/lib/bpf/xsk.c +++ b/tools/lib/bpf/xsk.c @@ -261,8 +261,8 @@ int xsk_umem__create_v0_0_2(struct xsk_umem **umem_ptr, void *umem_area, return xsk_umem__create_v0_0_4(umem_ptr, umem_area, size, fill, comp, &config); } -asm(".symver xsk_umem__create_v0_0_2, xsk_umem__create@LIBBPF_0.0.2"); -asm(".symver xsk_umem__create_v0_0_4, xsk_umem__create@@LIBBPF_0.0.4"); +COMPAT_VERSION(xsk_umem__create_v0_0_2, xsk_umem__create, LIBBPF_0.0.2) +DEFAULT_VERSION(xsk_umem__create_v0_0_4, xsk_umem__create, LIBBPF_0.0.4) static int xsk_load_xdp_prog(struct xsk_socket *xsk) { From 86c1aea84b97120a6d428ce17a2ebd55be677f56 Mon Sep 17 00:00:00 2001 From: Brian Vazquez Date: Tue, 1 Oct 2019 10:37:27 -0700 Subject: [PATCH 003/145] selftests/bpf: test_progs: Don't leak server_fd in tcp_rtt server_fd needs to be closed if pthread can't be created. Fixes: 8a03222f508b ("selftests/bpf: test_progs: fix client/server race in tcp_rtt") Signed-off-by: Brian Vazquez Signed-off-by: Daniel Borkmann Reviewed-by: Stanislav Fomichev Link: https://lore.kernel.org/bpf/20191001173728.149786-2-brianvv@google.com --- tools/testing/selftests/bpf/prog_tests/tcp_rtt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/bpf/prog_tests/tcp_rtt.c b/tools/testing/selftests/bpf/prog_tests/tcp_rtt.c index a82da555b1b0..f4cd60d6fba2 100644 --- a/tools/testing/selftests/bpf/prog_tests/tcp_rtt.c +++ b/tools/testing/selftests/bpf/prog_tests/tcp_rtt.c @@ -260,13 +260,14 @@ void test_tcp_rtt(void) if (CHECK_FAIL(pthread_create(&tid, NULL, server_thread, (void *)&server_fd))) - goto close_cgroup_fd; + goto close_server_fd; pthread_mutex_lock(&server_started_mtx); pthread_cond_wait(&server_started, &server_started_mtx); pthread_mutex_unlock(&server_started_mtx); CHECK_FAIL(run_test(cgroup_fd, server_fd)); +close_server_fd: close(server_fd); close_cgroup_fd: close(cgroup_fd); From a2d074e4c6e81ec9ab359d54f0b88273c738de37 Mon Sep 17 00:00:00 2001 From: Brian Vazquez Date: Tue, 1 Oct 2019 10:37:28 -0700 Subject: [PATCH 004/145] selftests/bpf: test_progs: Don't leak server_fd in test_sockopt_inherit server_fd needs to be closed if pthread can't be created. Fixes: e3e02e1d9c24 ("selftests/bpf: test_progs: convert test_sockopt_inherit") Signed-off-by: Brian Vazquez Signed-off-by: Daniel Borkmann Reviewed-by: Stanislav Fomichev Link: https://lore.kernel.org/bpf/20191001173728.149786-3-brianvv@google.com --- tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c b/tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c index 6cbeea7b4bf1..8547ecbdc61f 100644 --- a/tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c +++ b/tools/testing/selftests/bpf/prog_tests/sockopt_inherit.c @@ -195,7 +195,7 @@ static void run_test(int cgroup_fd) if (CHECK_FAIL(pthread_create(&tid, NULL, server_thread, (void *)&server_fd))) - goto close_bpf_object; + goto close_server_fd; pthread_mutex_lock(&server_started_mtx); pthread_cond_wait(&server_started, &server_started_mtx); From c91a9cfe9f6d136172a52ff6e01b3f83ba850c19 Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Thu, 26 Sep 2019 10:54:33 +0200 Subject: [PATCH 005/145] rt2x00: initialize last_reset Initialize last_reset variable to INITIAL_JIFFIES, otherwise it is not possible to test H/W reset for first 5 minutes of system run. Fixes: e403fa31ed71 ("rt2x00: add restart hw") Reported-and-tested-by: Jonathan Liu Signed-off-by: Stanislaw Gruszka Signed-off-by: Kalle Valo --- drivers/net/wireless/ralink/rt2x00/rt2x00debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c b/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c index 4d4e3888ef20..f2395309ec00 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00debug.c @@ -555,7 +555,7 @@ static ssize_t rt2x00debug_write_restart_hw(struct file *file, { struct rt2x00debug_intf *intf = file->private_data; struct rt2x00_dev *rt2x00dev = intf->rt2x00dev; - static unsigned long last_reset; + static unsigned long last_reset = INITIAL_JIFFIES; if (!rt2x00_has_cap_restart_hw(rt2x00dev)) return -EOPNOTSUPP; From e55190f26f92927e95aba22b069679311d5379ea Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B6rn=20T=C3=B6pel?= Date: Tue, 1 Oct 2019 13:22:49 +0200 Subject: [PATCH 006/145] samples/bpf: Fix build for task_fd_query_user.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add missing include for which was removed from perf-sys.h in commit 91854f9a077e ("perf tools: Move everything related to sys_perf_event_open() to perf-sys.h"). Fixes: 91854f9a077e ("perf tools: Move everything related to sys_perf_event_open() to perf-sys.h") Reported-by: KP Singh Reported-by: Florent Revest Signed-off-by: Björn Töpel Signed-off-by: Daniel Borkmann Tested-by: KP Singh Acked-by: Song Liu Link: https://lore.kernel.org/bpf/20191001112249.27341-1-bjorn.topel@gmail.com --- samples/bpf/task_fd_query_user.c | 1 + 1 file changed, 1 insertion(+) diff --git a/samples/bpf/task_fd_query_user.c b/samples/bpf/task_fd_query_user.c index e39938058223..4c31b305e6ef 100644 --- a/samples/bpf/task_fd_query_user.c +++ b/samples/bpf/task_fd_query_user.c @@ -13,6 +13,7 @@ #include #include #include +#include #include "libbpf.h" #include "bpf_load.h" From df551058f7a303bd3a17a4cef001349974962ce8 Mon Sep 17 00:00:00 2001 From: Magnus Karlsson Date: Wed, 2 Oct 2019 08:31:59 +0200 Subject: [PATCH 007/145] xsk: Fix crash in poll when device does not support ndo_xsk_wakeup Fixes a crash in poll() when an AF_XDP socket is opened in copy mode and the bound device does not have ndo_xsk_wakeup defined. Avoid trying to call the non-existing ndo and instead call the internal xsk sendmsg function to send packets in the same way (from the application's point of view) as calling sendmsg() in any mode or poll() in zero-copy mode would have done. The application should behave in the same way independent on if zero-copy mode or copy mode is used. Fixes: 77cd0d7b3f25 ("xsk: add support for need_wakeup flag in AF_XDP rings") Reported-by: syzbot+a5765ed8cdb1cca4d249@syzkaller.appspotmail.com Signed-off-by: Magnus Karlsson Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/1569997919-11541-1-git-send-email-magnus.karlsson@intel.com --- net/xdp/xsk.c | 42 +++++++++++++++++++++++++++--------------- 1 file changed, 27 insertions(+), 15 deletions(-) diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index fa8fbb8fa3c8..9044073fbf22 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -305,9 +305,8 @@ out: } EXPORT_SYMBOL(xsk_umem_consume_tx); -static int xsk_zc_xmit(struct sock *sk) +static int xsk_zc_xmit(struct xdp_sock *xs) { - struct xdp_sock *xs = xdp_sk(sk); struct net_device *dev = xs->dev; return dev->netdev_ops->ndo_xsk_wakeup(dev, xs->queue_id, @@ -327,11 +326,10 @@ static void xsk_destruct_skb(struct sk_buff *skb) sock_wfree(skb); } -static int xsk_generic_xmit(struct sock *sk, struct msghdr *m, - size_t total_len) +static int xsk_generic_xmit(struct sock *sk) { - u32 max_batch = TX_BATCH_SIZE; struct xdp_sock *xs = xdp_sk(sk); + u32 max_batch = TX_BATCH_SIZE; bool sent_frame = false; struct xdp_desc desc; struct sk_buff *skb; @@ -394,6 +392,18 @@ out: return err; } +static int __xsk_sendmsg(struct sock *sk) +{ + struct xdp_sock *xs = xdp_sk(sk); + + if (unlikely(!(xs->dev->flags & IFF_UP))) + return -ENETDOWN; + if (unlikely(!xs->tx)) + return -ENOBUFS; + + return xs->zc ? xsk_zc_xmit(xs) : xsk_generic_xmit(sk); +} + static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len) { bool need_wait = !(m->msg_flags & MSG_DONTWAIT); @@ -402,21 +412,18 @@ static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t total_len) if (unlikely(!xsk_is_bound(xs))) return -ENXIO; - if (unlikely(!(xs->dev->flags & IFF_UP))) - return -ENETDOWN; - if (unlikely(!xs->tx)) - return -ENOBUFS; - if (need_wait) + if (unlikely(need_wait)) return -EOPNOTSUPP; - return (xs->zc) ? xsk_zc_xmit(sk) : xsk_generic_xmit(sk, m, total_len); + return __xsk_sendmsg(sk); } static unsigned int xsk_poll(struct file *file, struct socket *sock, struct poll_table_struct *wait) { unsigned int mask = datagram_poll(file, sock, wait); - struct xdp_sock *xs = xdp_sk(sock->sk); + struct sock *sk = sock->sk; + struct xdp_sock *xs = xdp_sk(sk); struct net_device *dev; struct xdp_umem *umem; @@ -426,9 +433,14 @@ static unsigned int xsk_poll(struct file *file, struct socket *sock, dev = xs->dev; umem = xs->umem; - if (umem->need_wakeup) - dev->netdev_ops->ndo_xsk_wakeup(dev, xs->queue_id, - umem->need_wakeup); + if (umem->need_wakeup) { + if (dev->netdev_ops->ndo_xsk_wakeup) + dev->netdev_ops->ndo_xsk_wakeup(dev, xs->queue_id, + umem->need_wakeup); + else + /* Poll needs to drive Tx also in copy mode */ + __xsk_sendmsg(sk); + } if (xs->rx && !xskq_empty_desc(xs->rx)) mask |= POLLIN | POLLRDNORM; From 98beb3edeb974e906a81f305d88f7bc96b2ec83e Mon Sep 17 00:00:00 2001 From: KP Singh Date: Wed, 2 Oct 2019 21:16:52 +0200 Subject: [PATCH 008/145] samples/bpf: Add a workaround for asm_inline This was added in commit eb111869301e ("compiler-types.h: add asm_inline definition") and breaks samples/bpf as clang does not support asm __inline. Fixes: eb111869301e ("compiler-types.h: add asm_inline definition") Co-developed-by: Florent Revest Signed-off-by: Florent Revest Signed-off-by: KP Singh Signed-off-by: Daniel Borkmann Acked-by: Song Liu Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20191002191652.11432-1-kpsingh@chromium.org --- samples/bpf/asm_goto_workaround.h | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/samples/bpf/asm_goto_workaround.h b/samples/bpf/asm_goto_workaround.h index 7409722727ca..7048bb3594d6 100644 --- a/samples/bpf/asm_goto_workaround.h +++ b/samples/bpf/asm_goto_workaround.h @@ -3,7 +3,8 @@ #ifndef __ASM_GOTO_WORKAROUND_H #define __ASM_GOTO_WORKAROUND_H -/* this will bring in asm_volatile_goto macro definition +/* + * This will bring in asm_volatile_goto and asm_inline macro definitions * if enabled by compiler and config options. */ #include @@ -13,5 +14,15 @@ #define asm_volatile_goto(x...) asm volatile("invalid use of asm_volatile_goto") #endif +/* + * asm_inline is defined as asm __inline in "include/linux/compiler_types.h" + * if supported by the kernel's CC (i.e CONFIG_CC_HAS_ASM_INLINE) which is not + * supported by CLANG. + */ +#ifdef asm_inline +#undef asm_inline +#define asm_inline asm +#endif + #define volatile(x...) volatile("") #endif From 313c3fe9c2348e7147eca38bb446f295b45403a0 Mon Sep 17 00:00:00 2001 From: Michael Vassernis Date: Thu, 3 Oct 2019 07:31:38 +0000 Subject: [PATCH 009/145] mac80211_hwsim: fix incorrect dev_alloc_name failure goto If dev_alloc_name fails, hwsim_mon's memory allocated in alloc_netdev needs to be freed. Change goto command in dev_alloc_name failure to out_free_mon in order to perform free_netdev. Signed-off-by: Michael Vassernis Link: https://lore.kernel.org/r/20191003073049.3760-1-michael.vassernis@tandemg.com Signed-off-by: Johannes Berg --- drivers/net/wireless/mac80211_hwsim.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 45c73a6f09a1..14f562cd715c 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -4026,7 +4026,7 @@ static int __init init_mac80211_hwsim(void) err = dev_alloc_name(hwsim_mon, hwsim_mon->name); if (err < 0) { rtnl_unlock(); - goto out_free_radios; + goto out_free_mon; } err = register_netdevice(hwsim_mon); From 4152561f5da3fca92af7179dd538ea89e248f9d0 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 4 Oct 2019 10:51:31 +0100 Subject: [PATCH 010/145] mac80211: Reject malformed SSID elements Although this shouldn't occur in practice, it's a good idea to bounds check the length field of the SSID element prior to using it for things like allocations or memcpy operations. Cc: Cc: Kees Cook Reported-by: Nicolas Waisman Signed-off-by: Will Deacon Link: https://lore.kernel.org/r/20191004095132.15777-1-will@kernel.org Signed-off-by: Johannes Berg --- net/mac80211/mlme.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 26a2f49208b6..54dd8849d1cc 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -2633,7 +2633,8 @@ struct sk_buff *ieee80211_ap_probereq_get(struct ieee80211_hw *hw, rcu_read_lock(); ssid = ieee80211_bss_get_ie(cbss, WLAN_EID_SSID); - if (WARN_ON_ONCE(ssid == NULL)) + if (WARN_ONCE(!ssid || ssid[1] > IEEE80211_MAX_SSID_LEN, + "invalid SSID element (len=%d)", ssid ? ssid[1] : -1)) ssid_len = 0; else ssid_len = ssid[1]; @@ -5233,7 +5234,7 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata, rcu_read_lock(); ssidie = ieee80211_bss_get_ie(req->bss, WLAN_EID_SSID); - if (!ssidie) { + if (!ssidie || ssidie[1] > sizeof(assoc_data->ssid)) { rcu_read_unlock(); kfree(assoc_data); return -EINVAL; From 4ac2813cc867ae563a1ba5a9414bfb554e5796fa Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 4 Oct 2019 10:51:32 +0100 Subject: [PATCH 011/145] cfg80211: wext: avoid copying malformed SSIDs Ensure the SSID element is bounds-checked prior to invoking memcpy() with its length field, when copying to userspace. Cc: Cc: Kees Cook Reported-by: Nicolas Waisman Signed-off-by: Will Deacon Link: https://lore.kernel.org/r/20191004095132.15777-2-will@kernel.org [adjust commit log a bit] Signed-off-by: Johannes Berg --- net/wireless/wext-sme.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/wireless/wext-sme.c b/net/wireless/wext-sme.c index c67d7a82ab13..73fd0eae08ca 100644 --- a/net/wireless/wext-sme.c +++ b/net/wireless/wext-sme.c @@ -202,6 +202,7 @@ int cfg80211_mgd_wext_giwessid(struct net_device *dev, struct iw_point *data, char *ssid) { struct wireless_dev *wdev = dev->ieee80211_ptr; + int ret = 0; /* call only for station! */ if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION)) @@ -219,7 +220,10 @@ int cfg80211_mgd_wext_giwessid(struct net_device *dev, if (ie) { data->flags = 1; data->length = ie[1]; - memcpy(ssid, ie + 2, data->length); + if (data->length > IW_ESSID_MAX_SIZE) + ret = -EINVAL; + else + memcpy(ssid, ie + 2, data->length); } rcu_read_unlock(); } else if (wdev->wext.connect.ssid && wdev->wext.connect.ssid_len) { @@ -229,7 +233,7 @@ int cfg80211_mgd_wext_giwessid(struct net_device *dev, } wdev_unlock(wdev); - return 0; + return ret; } int cfg80211_mgd_wext_siwap(struct net_device *dev, From c763ac436b668d7417f0979430ec0312ede4093d Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Sat, 5 Oct 2019 15:05:18 -0700 Subject: [PATCH 012/145] net: dsa: b53: Do not clear existing mirrored port mask Clearing the existing bitmask of mirrored ports essentially prevents us from capturing more than one port at any given time. This is clearly wrong, do not clear the bitmask prior to setting up the new port. Reported-by: Hubert Feurstein Fixes: ed3af5fd08eb ("net: dsa: b53: Add support for port mirroring") Signed-off-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller --- drivers/net/dsa/b53/b53_common.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 526ba2ab66f1..cc3536315eff 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1845,7 +1845,6 @@ int b53_mirror_add(struct dsa_switch *ds, int port, loc = B53_EG_MIR_CTL; b53_read16(dev, B53_MGMT_PAGE, loc, ®); - reg &= ~MIRROR_MASK; reg |= BIT(port); b53_write16(dev, B53_MGMT_PAGE, loc, reg); From b870b0f867c77b92d7fd17ace8421904270d3c6a Mon Sep 17 00:00:00 2001 From: Jose Abreu Date: Sun, 6 Oct 2019 13:08:55 +0200 Subject: [PATCH 013/145] net: stmmac: selftests: Check if filtering is available before running We need to check if the number of available Hash Filters is enough to run the test, otherwise we will get false failures. Fixes: 091810dbded9 ("net: stmmac: Introduce selftests support") Signed-off-by: Jose Abreu Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c index cc76a42c7466..ed3926d4471d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c @@ -496,6 +496,9 @@ static int stmmac_test_hfilt(struct stmmac_priv *priv) if (ret) return ret; + if (netdev_mc_count(priv->dev) >= priv->hw->multicast_filter_bins) + return -EOPNOTSUPP; + ret = dev_mc_add(priv->dev, gd_addr); if (ret) return ret; @@ -573,6 +576,8 @@ static int stmmac_test_mcfilt(struct stmmac_priv *priv) if (stmmac_filter_check(priv)) return -EOPNOTSUPP; + if (!priv->hw->multicast_filter_bins) + return -EOPNOTSUPP; /* Remove all MC addresses */ __dev_mc_unsync(priv->dev, NULL); @@ -611,6 +616,8 @@ static int stmmac_test_ucfilt(struct stmmac_priv *priv) if (stmmac_filter_check(priv)) return -EOPNOTSUPP; + if (!priv->hw->multicast_filter_bins) + return -EOPNOTSUPP; /* Remove all UC addresses */ __dev_uc_unsync(priv->dev, NULL); From 25683bab09a70542b9f8e3e28f79b3369e56701f Mon Sep 17 00:00:00 2001 From: Jose Abreu Date: Sun, 6 Oct 2019 13:08:56 +0200 Subject: [PATCH 014/145] net: stmmac: gmac4+: Not all Unicast addresses may be available Some setups may not have all Unicast addresses filters available. Check the number of available filters before trying to setup it. Fixes: 477286b53f55 ("stmmac: add GMAC4 core support") Signed-off-by: Jose Abreu Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 2cb9c53f93b8..5a7b0aca1d31 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -448,7 +448,7 @@ static void dwmac4_set_filter(struct mac_device_info *hw, value |= GMAC_PACKET_FILTER_HPF; /* Handle multiple unicast addresses */ - if (netdev_uc_count(dev) > GMAC_MAX_PERFECT_ADDRESSES) { + if (netdev_uc_count(dev) > hw->unicast_filter_entries) { /* Switch to promiscuous mode if more than 128 addrs * are required */ From 2809fc13163f4946d7cde092807bea44bbae4478 Mon Sep 17 00:00:00 2001 From: Jose Abreu Date: Sun, 6 Oct 2019 13:08:57 +0200 Subject: [PATCH 015/145] net: stmmac: selftests: Fix L2 Hash Filter test With the current MAC addresses hard-coded in the test we can get some false positives as we use the Hash Filtering method. Let's change the MAC addresses in the tests to be unique when hashed. Fixes: 091810dbded9 ("net: stmmac: Introduce selftests support") Signed-off-by: Jose Abreu Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c index ed3926d4471d..e4ac3c401432 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_selftests.c @@ -487,8 +487,8 @@ static int stmmac_filter_check(struct stmmac_priv *priv) static int stmmac_test_hfilt(struct stmmac_priv *priv) { - unsigned char gd_addr[ETH_ALEN] = {0x01, 0x00, 0xcc, 0xcc, 0xdd, 0xdd}; - unsigned char bd_addr[ETH_ALEN] = {0x09, 0x00, 0xaa, 0xaa, 0xbb, 0xbb}; + unsigned char gd_addr[ETH_ALEN] = {0x01, 0xee, 0xdd, 0xcc, 0xbb, 0xaa}; + unsigned char bd_addr[ETH_ALEN] = {0x01, 0x01, 0x02, 0x03, 0x04, 0x05}; struct stmmac_packet_attrs attr = { }; int ret; From c48fc11b69e95007109206311b0187a3090591f3 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 7 Oct 2019 10:58:28 +0100 Subject: [PATCH 016/145] rxrpc: Fix call ref leak When sendmsg() finds a call to continue on with, if the call is in an inappropriate state, it doesn't release the ref it just got on that call before returning an error. This causes the following symptom to show up with kasan: BUG: KASAN: use-after-free in rxrpc_send_keepalive+0x8a2/0x940 net/rxrpc/output.c:635 Read of size 8 at addr ffff888064219698 by task kworker/0:3/11077 where line 635 is: whdr.epoch = htonl(peer->local->rxnet->epoch); The local endpoint (which cannot be pinned by the call) has been released, but not the peer (which is pinned by the call). Fix this by releasing the call in the error path. Fixes: 37411cad633f ("rxrpc: Fix potential NULL-pointer exception") Reported-by: syzbot+d850c266e3df14da1d31@syzkaller.appspotmail.com Signed-off-by: David Howells --- net/rxrpc/sendmsg.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index 6a1547b270fe..22f51a7e356e 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -661,6 +661,7 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) case RXRPC_CALL_SERVER_PREALLOC: case RXRPC_CALL_SERVER_SECURING: case RXRPC_CALL_SERVER_ACCEPTING: + rxrpc_put_call(call, rxrpc_call_put); ret = -EBUSY; goto error_release_sock; default: From 55f6c98e3674ce16038a1949c3f9ca5a9a99f289 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 7 Oct 2019 10:58:29 +0100 Subject: [PATCH 017/145] rxrpc: Fix trace-after-put looking at the put peer record rxrpc_put_peer() calls trace_rxrpc_peer() after it has done the decrement of the refcount - which looks at the debug_id in the peer record. But unless the refcount was reduced to zero, we no longer have the right to look in the record and, indeed, it may be deleted by some other thread. Fix this by getting the debug_id out before decrementing the refcount and then passing that into the tracepoint. This can cause the following symptoms: BUG: KASAN: use-after-free in __rxrpc_put_peer net/rxrpc/peer_object.c:411 [inline] BUG: KASAN: use-after-free in rxrpc_put_peer+0x685/0x6a0 net/rxrpc/peer_object.c:435 Read of size 8 at addr ffff888097ec0058 by task syz-executor823/24216 Fixes: 1159d4b496f5 ("rxrpc: Add a tracepoint to track rxrpc_peer refcounting") Reported-by: syzbot+b9be979c55f2bea8ed30@syzkaller.appspotmail.com Signed-off-by: David Howells --- include/trace/events/rxrpc.h | 6 +++--- net/rxrpc/peer_object.c | 11 +++++++---- 2 files changed, 10 insertions(+), 7 deletions(-) diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h index edc5c887a44c..45556fe771c3 100644 --- a/include/trace/events/rxrpc.h +++ b/include/trace/events/rxrpc.h @@ -519,10 +519,10 @@ TRACE_EVENT(rxrpc_local, ); TRACE_EVENT(rxrpc_peer, - TP_PROTO(struct rxrpc_peer *peer, enum rxrpc_peer_trace op, + TP_PROTO(unsigned int peer_debug_id, enum rxrpc_peer_trace op, int usage, const void *where), - TP_ARGS(peer, op, usage, where), + TP_ARGS(peer_debug_id, op, usage, where), TP_STRUCT__entry( __field(unsigned int, peer ) @@ -532,7 +532,7 @@ TRACE_EVENT(rxrpc_peer, ), TP_fast_assign( - __entry->peer = peer->debug_id; + __entry->peer = peer_debug_id; __entry->op = op; __entry->usage = usage; __entry->where = where; diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index 9c3ac96f71cb..b700b7ecaa3d 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -382,7 +382,7 @@ struct rxrpc_peer *rxrpc_get_peer(struct rxrpc_peer *peer) int n; n = atomic_inc_return(&peer->usage); - trace_rxrpc_peer(peer, rxrpc_peer_got, n, here); + trace_rxrpc_peer(peer->debug_id, rxrpc_peer_got, n, here); return peer; } @@ -396,7 +396,7 @@ struct rxrpc_peer *rxrpc_get_peer_maybe(struct rxrpc_peer *peer) if (peer) { int n = atomic_fetch_add_unless(&peer->usage, 1, 0); if (n > 0) - trace_rxrpc_peer(peer, rxrpc_peer_got, n + 1, here); + trace_rxrpc_peer(peer->debug_id, rxrpc_peer_got, n + 1, here); else peer = NULL; } @@ -426,11 +426,13 @@ static void __rxrpc_put_peer(struct rxrpc_peer *peer) void rxrpc_put_peer(struct rxrpc_peer *peer) { const void *here = __builtin_return_address(0); + unsigned int debug_id; int n; if (peer) { + debug_id = peer->debug_id; n = atomic_dec_return(&peer->usage); - trace_rxrpc_peer(peer, rxrpc_peer_put, n, here); + trace_rxrpc_peer(debug_id, rxrpc_peer_put, n, here); if (n == 0) __rxrpc_put_peer(peer); } @@ -443,10 +445,11 @@ void rxrpc_put_peer(struct rxrpc_peer *peer) void rxrpc_put_peer_locked(struct rxrpc_peer *peer) { const void *here = __builtin_return_address(0); + unsigned int debug_id = peer->debug_id; int n; n = atomic_dec_return(&peer->usage); - trace_rxrpc_peer(peer, rxrpc_peer_put, n, here); + trace_rxrpc_peer(debug_id, rxrpc_peer_put, n, here); if (n == 0) { hash_del_rcu(&peer->hash_link); list_del_init(&peer->keepalive_link); From 4c1295dccc0afe0905b6ca4c62ade7f2406f2cfb Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 7 Oct 2019 10:58:29 +0100 Subject: [PATCH 018/145] rxrpc: Fix trace-after-put looking at the put connection record rxrpc_put_*conn() calls trace_rxrpc_conn() after they have done the decrement of the refcount - which looks at the debug_id in the connection record. But unless the refcount was reduced to zero, we no longer have the right to look in the record and, indeed, it may be deleted by some other thread. Fix this by getting the debug_id out before decrementing the refcount and then passing that into the tracepoint. Fixes: 363deeab6d0f ("rxrpc: Add connection tracepoint and client conn state tracepoint") Signed-off-by: David Howells --- include/trace/events/rxrpc.h | 6 +++--- net/rxrpc/call_accept.c | 2 +- net/rxrpc/conn_client.c | 6 ++++-- net/rxrpc/conn_object.c | 13 +++++++------ net/rxrpc/conn_service.c | 2 +- 5 files changed, 16 insertions(+), 13 deletions(-) diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h index 45556fe771c3..38a97e890cb6 100644 --- a/include/trace/events/rxrpc.h +++ b/include/trace/events/rxrpc.h @@ -546,10 +546,10 @@ TRACE_EVENT(rxrpc_peer, ); TRACE_EVENT(rxrpc_conn, - TP_PROTO(struct rxrpc_connection *conn, enum rxrpc_conn_trace op, + TP_PROTO(unsigned int conn_debug_id, enum rxrpc_conn_trace op, int usage, const void *where), - TP_ARGS(conn, op, usage, where), + TP_ARGS(conn_debug_id, op, usage, where), TP_STRUCT__entry( __field(unsigned int, conn ) @@ -559,7 +559,7 @@ TRACE_EVENT(rxrpc_conn, ), TP_fast_assign( - __entry->conn = conn->debug_id; + __entry->conn = conn_debug_id; __entry->op = op; __entry->usage = usage; __entry->where = where; diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 00c095d74145..c1b1b7dd2924 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -84,7 +84,7 @@ static int rxrpc_service_prealloc_one(struct rxrpc_sock *rx, smp_store_release(&b->conn_backlog_head, (head + 1) & (size - 1)); - trace_rxrpc_conn(conn, rxrpc_conn_new_service, + trace_rxrpc_conn(conn->debug_id, rxrpc_conn_new_service, atomic_read(&conn->usage), here); } diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index 3f1da1b49f69..700eb77173fc 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -212,7 +212,8 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, gfp_t gfp) rxrpc_get_local(conn->params.local); key_get(conn->params.key); - trace_rxrpc_conn(conn, rxrpc_conn_new_client, atomic_read(&conn->usage), + trace_rxrpc_conn(conn->debug_id, rxrpc_conn_new_client, + atomic_read(&conn->usage), __builtin_return_address(0)); trace_rxrpc_client(conn, -1, rxrpc_client_alloc); _leave(" = %p", conn); @@ -985,11 +986,12 @@ rxrpc_put_one_client_conn(struct rxrpc_connection *conn) void rxrpc_put_client_conn(struct rxrpc_connection *conn) { const void *here = __builtin_return_address(0); + unsigned int debug_id = conn->debug_id; int n; do { n = atomic_dec_return(&conn->usage); - trace_rxrpc_conn(conn, rxrpc_conn_put_client, n, here); + trace_rxrpc_conn(debug_id, rxrpc_conn_put_client, n, here); if (n > 0) return; ASSERTCMP(n, >=, 0); diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index ed05b6922132..38d718e90dc6 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -269,7 +269,7 @@ bool rxrpc_queue_conn(struct rxrpc_connection *conn) if (n == 0) return false; if (rxrpc_queue_work(&conn->processor)) - trace_rxrpc_conn(conn, rxrpc_conn_queued, n + 1, here); + trace_rxrpc_conn(conn->debug_id, rxrpc_conn_queued, n + 1, here); else rxrpc_put_connection(conn); return true; @@ -284,7 +284,7 @@ void rxrpc_see_connection(struct rxrpc_connection *conn) if (conn) { int n = atomic_read(&conn->usage); - trace_rxrpc_conn(conn, rxrpc_conn_seen, n, here); + trace_rxrpc_conn(conn->debug_id, rxrpc_conn_seen, n, here); } } @@ -296,7 +296,7 @@ void rxrpc_get_connection(struct rxrpc_connection *conn) const void *here = __builtin_return_address(0); int n = atomic_inc_return(&conn->usage); - trace_rxrpc_conn(conn, rxrpc_conn_got, n, here); + trace_rxrpc_conn(conn->debug_id, rxrpc_conn_got, n, here); } /* @@ -310,7 +310,7 @@ rxrpc_get_connection_maybe(struct rxrpc_connection *conn) if (conn) { int n = atomic_fetch_add_unless(&conn->usage, 1, 0); if (n > 0) - trace_rxrpc_conn(conn, rxrpc_conn_got, n + 1, here); + trace_rxrpc_conn(conn->debug_id, rxrpc_conn_got, n + 1, here); else conn = NULL; } @@ -333,10 +333,11 @@ static void rxrpc_set_service_reap_timer(struct rxrpc_net *rxnet, void rxrpc_put_service_conn(struct rxrpc_connection *conn) { const void *here = __builtin_return_address(0); + unsigned int debug_id = conn->debug_id; int n; n = atomic_dec_return(&conn->usage); - trace_rxrpc_conn(conn, rxrpc_conn_put_service, n, here); + trace_rxrpc_conn(debug_id, rxrpc_conn_put_service, n, here); ASSERTCMP(n, >=, 0); if (n == 1) rxrpc_set_service_reap_timer(conn->params.local->rxnet, @@ -420,7 +421,7 @@ void rxrpc_service_connection_reaper(struct work_struct *work) */ if (atomic_cmpxchg(&conn->usage, 1, 0) != 1) continue; - trace_rxrpc_conn(conn, rxrpc_conn_reap_service, 0, NULL); + trace_rxrpc_conn(conn->debug_id, rxrpc_conn_reap_service, 0, NULL); if (rxrpc_conn_is_client(conn)) BUG(); diff --git a/net/rxrpc/conn_service.c b/net/rxrpc/conn_service.c index b30e13f6d95f..123d6ceab15c 100644 --- a/net/rxrpc/conn_service.c +++ b/net/rxrpc/conn_service.c @@ -134,7 +134,7 @@ struct rxrpc_connection *rxrpc_prealloc_service_connection(struct rxrpc_net *rxn list_add_tail(&conn->proc_link, &rxnet->conn_proc_list); write_unlock(&rxnet->conn_lock); - trace_rxrpc_conn(conn, rxrpc_conn_new_service, + trace_rxrpc_conn(conn->debug_id, rxrpc_conn_new_service, atomic_read(&conn->usage), __builtin_return_address(0)); } From 48c9e0ec7cbbb7370448f859ccc8e3b7eb69e755 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 7 Oct 2019 10:58:29 +0100 Subject: [PATCH 019/145] rxrpc: Fix trace-after-put looking at the put call record rxrpc_put_call() calls trace_rxrpc_call() after it has done the decrement of the refcount - which looks at the debug_id in the call record. But unless the refcount was reduced to zero, we no longer have the right to look in the record and, indeed, it may be deleted by some other thread. Fix this by getting the debug_id out before decrementing the refcount and then passing that into the tracepoint. Fixes: e34d4234b0b7 ("rxrpc: Trace rxrpc_call usage") Signed-off-by: David Howells --- include/trace/events/rxrpc.h | 6 +++--- net/rxrpc/call_accept.c | 2 +- net/rxrpc/call_object.c | 28 +++++++++++++++++----------- 3 files changed, 21 insertions(+), 15 deletions(-) diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h index 38a97e890cb6..191fe447f990 100644 --- a/include/trace/events/rxrpc.h +++ b/include/trace/events/rxrpc.h @@ -606,10 +606,10 @@ TRACE_EVENT(rxrpc_client, ); TRACE_EVENT(rxrpc_call, - TP_PROTO(struct rxrpc_call *call, enum rxrpc_call_trace op, + TP_PROTO(unsigned int call_debug_id, enum rxrpc_call_trace op, int usage, const void *where, const void *aux), - TP_ARGS(call, op, usage, where, aux), + TP_ARGS(call_debug_id, op, usage, where, aux), TP_STRUCT__entry( __field(unsigned int, call ) @@ -620,7 +620,7 @@ TRACE_EVENT(rxrpc_call, ), TP_fast_assign( - __entry->call = call->debug_id; + __entry->call = call_debug_id; __entry->op = op; __entry->usage = usage; __entry->where = where; diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index c1b1b7dd2924..1f778102ed8d 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -97,7 +97,7 @@ static int rxrpc_service_prealloc_one(struct rxrpc_sock *rx, call->flags |= (1 << RXRPC_CALL_IS_SERVICE); call->state = RXRPC_CALL_SERVER_PREALLOC; - trace_rxrpc_call(call, rxrpc_call_new_service, + trace_rxrpc_call(call->debug_id, rxrpc_call_new_service, atomic_read(&call->usage), here, (const void *)user_call_ID); diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 32d8dc677142..6dace078971a 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -240,7 +240,8 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, if (p->intr) __set_bit(RXRPC_CALL_IS_INTR, &call->flags); call->tx_total_len = p->tx_total_len; - trace_rxrpc_call(call, rxrpc_call_new_client, atomic_read(&call->usage), + trace_rxrpc_call(call->debug_id, rxrpc_call_new_client, + atomic_read(&call->usage), here, (const void *)p->user_call_ID); /* We need to protect a partially set up call against the user as we @@ -290,8 +291,8 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, if (ret < 0) goto error; - trace_rxrpc_call(call, rxrpc_call_connected, atomic_read(&call->usage), - here, NULL); + trace_rxrpc_call(call->debug_id, rxrpc_call_connected, + atomic_read(&call->usage), here, NULL); rxrpc_start_call_timer(call); @@ -313,8 +314,8 @@ error_dup_user_ID: error: __rxrpc_set_call_completion(call, RXRPC_CALL_LOCAL_ERROR, RX_CALL_DEAD, ret); - trace_rxrpc_call(call, rxrpc_call_error, atomic_read(&call->usage), - here, ERR_PTR(ret)); + trace_rxrpc_call(call->debug_id, rxrpc_call_error, + atomic_read(&call->usage), here, ERR_PTR(ret)); rxrpc_release_call(rx, call); mutex_unlock(&call->user_mutex); rxrpc_put_call(call, rxrpc_call_put); @@ -376,7 +377,8 @@ bool rxrpc_queue_call(struct rxrpc_call *call) if (n == 0) return false; if (rxrpc_queue_work(&call->processor)) - trace_rxrpc_call(call, rxrpc_call_queued, n + 1, here, NULL); + trace_rxrpc_call(call->debug_id, rxrpc_call_queued, n + 1, + here, NULL); else rxrpc_put_call(call, rxrpc_call_put_noqueue); return true; @@ -391,7 +393,8 @@ bool __rxrpc_queue_call(struct rxrpc_call *call) int n = atomic_read(&call->usage); ASSERTCMP(n, >=, 1); if (rxrpc_queue_work(&call->processor)) - trace_rxrpc_call(call, rxrpc_call_queued_ref, n, here, NULL); + trace_rxrpc_call(call->debug_id, rxrpc_call_queued_ref, n, + here, NULL); else rxrpc_put_call(call, rxrpc_call_put_noqueue); return true; @@ -406,7 +409,8 @@ void rxrpc_see_call(struct rxrpc_call *call) if (call) { int n = atomic_read(&call->usage); - trace_rxrpc_call(call, rxrpc_call_seen, n, here, NULL); + trace_rxrpc_call(call->debug_id, rxrpc_call_seen, n, + here, NULL); } } @@ -418,7 +422,7 @@ void rxrpc_get_call(struct rxrpc_call *call, enum rxrpc_call_trace op) const void *here = __builtin_return_address(0); int n = atomic_inc_return(&call->usage); - trace_rxrpc_call(call, op, n, here, NULL); + trace_rxrpc_call(call->debug_id, op, n, here, NULL); } /* @@ -445,7 +449,8 @@ void rxrpc_release_call(struct rxrpc_sock *rx, struct rxrpc_call *call) _enter("{%d,%d}", call->debug_id, atomic_read(&call->usage)); - trace_rxrpc_call(call, rxrpc_call_release, atomic_read(&call->usage), + trace_rxrpc_call(call->debug_id, rxrpc_call_release, + atomic_read(&call->usage), here, (const void *)call->flags); ASSERTCMP(call->state, ==, RXRPC_CALL_COMPLETE); @@ -534,12 +539,13 @@ void rxrpc_put_call(struct rxrpc_call *call, enum rxrpc_call_trace op) { struct rxrpc_net *rxnet = call->rxnet; const void *here = __builtin_return_address(0); + unsigned int debug_id = call->debug_id; int n; ASSERT(call != NULL); n = atomic_dec_return(&call->usage); - trace_rxrpc_call(call, op, n, here, NULL); + trace_rxrpc_call(debug_id, op, n, here, NULL); ASSERTCMP(n, >=, 0); if (n == 0) { _debug("call %d dead", call->debug_id); From 9ebeddef58c41bd700419cdcece24cf64ce32276 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 7 Oct 2019 10:58:29 +0100 Subject: [PATCH 020/145] rxrpc: rxrpc_peer needs to hold a ref on the rxrpc_local record The rxrpc_peer record needs to hold a reference on the rxrpc_local record it points as the peer is used as a base to access information in the rxrpc_local record. This can cause problems in __rxrpc_put_peer(), where we need the network namespace pointer, and in rxrpc_send_keepalive(), where we need to access the UDP socket, leading to symptoms like: BUG: KASAN: use-after-free in __rxrpc_put_peer net/rxrpc/peer_object.c:411 [inline] BUG: KASAN: use-after-free in rxrpc_put_peer+0x685/0x6a0 net/rxrpc/peer_object.c:435 Read of size 8 at addr ffff888097ec0058 by task syz-executor823/24216 Fix this by taking a ref on the local record for the peer record. Fixes: ace45bec6d77 ("rxrpc: Fix firewall route keepalive") Fixes: 2baec2c3f854 ("rxrpc: Support network namespacing") Reported-by: syzbot+b9be979c55f2bea8ed30@syzkaller.appspotmail.com Signed-off-by: David Howells --- net/rxrpc/peer_object.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index b700b7ecaa3d..64830d8c1fdb 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -216,7 +216,7 @@ struct rxrpc_peer *rxrpc_alloc_peer(struct rxrpc_local *local, gfp_t gfp) peer = kzalloc(sizeof(struct rxrpc_peer), gfp); if (peer) { atomic_set(&peer->usage, 1); - peer->local = local; + peer->local = rxrpc_get_local(local); INIT_HLIST_HEAD(&peer->error_targets); peer->service_conns = RB_ROOT; seqlock_init(&peer->service_conn_lock); @@ -307,7 +307,6 @@ void rxrpc_new_incoming_peer(struct rxrpc_sock *rx, struct rxrpc_local *local, unsigned long hash_key; hash_key = rxrpc_peer_hash_key(local, &peer->srx); - peer->local = local; rxrpc_init_peer(rx, peer, hash_key); spin_lock(&rxnet->peer_hash_lock); @@ -417,6 +416,7 @@ static void __rxrpc_put_peer(struct rxrpc_peer *peer) list_del_init(&peer->keepalive_link); spin_unlock_bh(&rxnet->peer_hash_lock); + rxrpc_put_local(peer->local); kfree_rcu(peer, rcu); } @@ -453,6 +453,7 @@ void rxrpc_put_peer_locked(struct rxrpc_peer *peer) if (n == 0) { hash_del_rcu(&peer->hash_link); list_del_init(&peer->keepalive_link); + rxrpc_put_local(peer->local); kfree_rcu(peer, rcu); } } From 91fcfbe8852edb929ff8702534525031a15d0aa6 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 7 Oct 2019 10:58:29 +0100 Subject: [PATCH 021/145] rxrpc: Fix call crypto state cleanup Fix the cleanup of the crypto state on a call after the call has been disconnected. As the call has been disconnected, its connection ref has been discarded and so we can't go through that to get to the security ops table. Fix this by caching the security ops pointer in the rxrpc_call struct and using that when freeing the call security state. Also use this in other places we're dealing with call-specific security. The symptoms look like: BUG: KASAN: use-after-free in rxrpc_release_call+0xb2d/0xb60 net/rxrpc/call_object.c:481 Read of size 8 at addr ffff888062ffeb50 by task syz-executor.5/4764 Fixes: 1db88c534371 ("rxrpc: Fix -Wframe-larger-than= warnings from on-stack crypto") Reported-by: syzbot+eed305768ece6682bb7f@syzkaller.appspotmail.com Signed-off-by: David Howells --- net/rxrpc/ar-internal.h | 1 + net/rxrpc/call_accept.c | 1 + net/rxrpc/call_object.c | 6 +++--- net/rxrpc/conn_client.c | 3 +++ net/rxrpc/recvmsg.c | 6 +++--- net/rxrpc/sendmsg.c | 2 +- 6 files changed, 12 insertions(+), 7 deletions(-) diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 1091bf35a199..ecc17dabec8f 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -556,6 +556,7 @@ struct rxrpc_call { struct rxrpc_peer *peer; /* Peer record for remote address */ struct rxrpc_sock __rcu *socket; /* socket responsible */ struct rxrpc_net *rxnet; /* Network namespace to which call belongs */ + const struct rxrpc_security *security; /* applied security module */ struct mutex user_mutex; /* User access mutex */ unsigned long ack_at; /* When deferred ACK needs to happen */ unsigned long ack_lost_at; /* When ACK is figured as lost */ diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 1f778102ed8d..135bf5cd8dd5 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -307,6 +307,7 @@ static struct rxrpc_call *rxrpc_alloc_incoming_call(struct rxrpc_sock *rx, rxrpc_see_call(call); call->conn = conn; + call->security = conn->security; call->peer = rxrpc_get_peer(conn->params.peer); call->cong_cwnd = call->peer->cong_cwnd; return call; diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 6dace078971a..a31c18c09894 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -493,10 +493,10 @@ void rxrpc_release_call(struct rxrpc_sock *rx, struct rxrpc_call *call) _debug("RELEASE CALL %p (%d CONN %p)", call, call->debug_id, conn); - if (conn) { + if (conn) rxrpc_disconnect_call(call); - conn->security->free_call_crypto(call); - } + if (call->security) + call->security->free_call_crypto(call); rxrpc_cleanup_ring(call); _leave(""); diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index 700eb77173fc..376370cd9285 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -353,6 +353,7 @@ static int rxrpc_get_client_conn(struct rxrpc_sock *rx, if (cp->exclusive) { call->conn = candidate; + call->security = candidate->security; call->security_ix = candidate->security_ix; call->service_id = candidate->service_id; _leave(" = 0 [exclusive %d]", candidate->debug_id); @@ -404,6 +405,7 @@ static int rxrpc_get_client_conn(struct rxrpc_sock *rx, candidate_published: set_bit(RXRPC_CONN_IN_CLIENT_CONNS, &candidate->flags); call->conn = candidate; + call->security = candidate->security; call->security_ix = candidate->security_ix; call->service_id = candidate->service_id; spin_unlock(&local->client_conns_lock); @@ -426,6 +428,7 @@ found_extant_conn: spin_lock(&conn->channel_lock); call->conn = conn; + call->security = conn->security; call->security_ix = conn->security_ix; call->service_id = conn->service_id; list_add_tail(&call->chan_wait_link, &conn->waiting_calls); diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index 3b0becb12041..a4090797c9b2 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -251,8 +251,8 @@ static int rxrpc_verify_packet(struct rxrpc_call *call, struct sk_buff *skb, seq += subpacket; } - return call->conn->security->verify_packet(call, skb, offset, len, - seq, cksum); + return call->security->verify_packet(call, skb, offset, len, + seq, cksum); } /* @@ -291,7 +291,7 @@ static int rxrpc_locate_data(struct rxrpc_call *call, struct sk_buff *skb, *_offset = offset; *_len = len; - call->conn->security->locate_data(call, skb, _offset, _len); + call->security->locate_data(call, skb, _offset, _len); return 0; } diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index 22f51a7e356e..813fd6888142 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -419,7 +419,7 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, call->tx_winsize) sp->hdr.flags |= RXRPC_MORE_PACKETS; - ret = conn->security->secure_packet( + ret = call->security->secure_packet( call, skb, skb->mark, skb->head); if (ret < 0) goto out; From 4b5149365faa8fddf7dc58e47cfed51b19db051e Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Sun, 6 Oct 2019 13:21:07 -0700 Subject: [PATCH 022/145] ionic: fix stats memory dereference When the netdev is down, the queues and their debug stats do not exist, so don't try using a pointer to them when when printing the ethtool stats. Fixes: e470355bd96a ("ionic: Add driver stats") Signed-off-by: Shannon Nelson Signed-off-by: David S. Miller --- .../net/ethernet/pensando/ionic/ionic_lif.h | 2 ++ .../net/ethernet/pensando/ionic/ionic_stats.c | 29 ++++++++++++------- 2 files changed, 21 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h index 812190e729c2..6a95b42a8d8c 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h @@ -182,6 +182,8 @@ struct ionic_lif { #define lif_to_txqcq(lif, i) ((lif)->txqcqs[i].qcq) #define lif_to_rxqcq(lif, i) ((lif)->rxqcqs[i].qcq) +#define lif_to_txstats(lif, i) ((lif)->txqcqs[i].stats->tx) +#define lif_to_rxstats(lif, i) ((lif)->rxqcqs[i].stats->rx) #define lif_to_txq(lif, i) (&lif_to_txqcq((lif), i)->q) #define lif_to_rxq(lif, i) (&lif_to_txqcq((lif), i)->q) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_stats.c b/drivers/net/ethernet/pensando/ionic/ionic_stats.c index e2907884f843..03916b6d47f2 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_stats.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_stats.c @@ -117,7 +117,8 @@ static u64 ionic_sw_stats_get_count(struct ionic_lif *lif) /* rx stats */ total += MAX_Q(lif) * IONIC_NUM_RX_STATS; - if (test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_UP, lif->state) && + test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { /* tx debug stats */ total += MAX_Q(lif) * (IONIC_NUM_DBG_CQ_STATS + IONIC_NUM_TX_Q_STATS + @@ -149,7 +150,8 @@ static void ionic_sw_stats_get_strings(struct ionic_lif *lif, u8 **buf) *buf += ETH_GSTRING_LEN; } - if (test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_UP, lif->state) && + test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { for (i = 0; i < IONIC_NUM_TX_Q_STATS; i++) { snprintf(*buf, ETH_GSTRING_LEN, "txq_%d_%s", @@ -187,7 +189,8 @@ static void ionic_sw_stats_get_strings(struct ionic_lif *lif, u8 **buf) *buf += ETH_GSTRING_LEN; } - if (test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_UP, lif->state) && + test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { for (i = 0; i < IONIC_NUM_DBG_CQ_STATS; i++) { snprintf(*buf, ETH_GSTRING_LEN, "rxq_%d_cq_%s", @@ -223,6 +226,8 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) { struct ionic_lif_sw_stats lif_stats; struct ionic_qcq *txqcq, *rxqcq; + struct ionic_tx_stats *txstats; + struct ionic_rx_stats *rxstats; int i, q_num; ionic_get_lif_stats(lif, &lif_stats); @@ -233,15 +238,17 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) } for (q_num = 0; q_num < MAX_Q(lif); q_num++) { - txqcq = lif_to_txqcq(lif, q_num); + txstats = &lif_to_txstats(lif, q_num); for (i = 0; i < IONIC_NUM_TX_STATS; i++) { - **buf = IONIC_READ_STAT64(&txqcq->stats->tx, + **buf = IONIC_READ_STAT64(txstats, &ionic_tx_stats_desc[i]); (*buf)++; } - if (test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_UP, lif->state) && + test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + txqcq = lif_to_txqcq(lif, q_num); for (i = 0; i < IONIC_NUM_TX_Q_STATS; i++) { **buf = IONIC_READ_STAT64(&txqcq->q, &ionic_txq_stats_desc[i]); @@ -258,22 +265,24 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) (*buf)++; } for (i = 0; i < IONIC_MAX_NUM_SG_CNTR; i++) { - **buf = txqcq->stats->tx.sg_cntr[i]; + **buf = txstats->sg_cntr[i]; (*buf)++; } } } for (q_num = 0; q_num < MAX_Q(lif); q_num++) { - rxqcq = lif_to_rxqcq(lif, q_num); + rxstats = &lif_to_rxstats(lif, q_num); for (i = 0; i < IONIC_NUM_RX_STATS; i++) { - **buf = IONIC_READ_STAT64(&rxqcq->stats->rx, + **buf = IONIC_READ_STAT64(rxstats, &ionic_rx_stats_desc[i]); (*buf)++; } - if (test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + if (test_bit(IONIC_LIF_UP, lif->state) && + test_bit(IONIC_LIF_SW_DEBUG_STATS, lif->state)) { + rxqcq = lif_to_rxqcq(lif, q_num); for (i = 0; i < IONIC_NUM_DBG_CQ_STATS; i++) { **buf = IONIC_READ_STAT64(&rxqcq->cq, &ionic_dbg_cq_stats_desc[i]); From 503c9addef613c872679e24fc8a78f3febeb5a08 Mon Sep 17 00:00:00 2001 From: Antonio Borneo Date: Mon, 7 Oct 2019 17:43:02 +0200 Subject: [PATCH 023/145] ptp: fix typo of "mechanism" in Kconfig help text Fix typo s/mechansim/mechanism/ Signed-off-by: Antonio Borneo Signed-off-by: David S. Miller --- drivers/ptp/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig index 960961fb0d7c..0517272a268e 100644 --- a/drivers/ptp/Kconfig +++ b/drivers/ptp/Kconfig @@ -97,8 +97,8 @@ config PTP_1588_CLOCK_PCH help This driver adds support for using the PCH EG20T as a PTP clock. The hardware supports time stamping of PTP packets - when using the end-to-end delay (E2E) mechansim. The peer - delay mechansim (P2P) is not supported. + when using the end-to-end delay (E2E) mechanism. The peer + delay mechanism (P2P) is not supported. This clock is only useful if your PTP programs are getting hardware time stamps on the PTP Ethernet packets using the From 1399c59fa92984836db90538cf92397fe7caaa57 Mon Sep 17 00:00:00 2001 From: Navid Emamdoost Date: Fri, 4 Oct 2019 14:42:19 -0500 Subject: [PATCH 024/145] nl80211: fix memory leak in nl80211_get_ftm_responder_stats In nl80211_get_ftm_responder_stats, a new skb is created via nlmsg_new named msg. If nl80211hdr_put() fails, then msg should be released. The return statement should be replace by goto to error handling code. Fixes: 81e54d08d9d8 ("cfg80211: support FTM responder configuration/statistics") Signed-off-by: Navid Emamdoost Link: https://lore.kernel.org/r/20191004194220.19412-1-navid.emamdoost@gmail.com Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 141cdb171665..4453dd375de9 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -13682,7 +13682,7 @@ static int nl80211_get_ftm_responder_stats(struct sk_buff *skb, hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0, NL80211_CMD_GET_FTM_RESPONDER_STATS); if (!hdr) - return -ENOBUFS; + goto nla_put_failure; if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex)) goto nla_put_failure; From 461c4c2b4c0731d7452bad4e77c0cdbdcea1804c Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Fri, 4 Oct 2019 15:37:06 +0300 Subject: [PATCH 025/145] cfg80211: fix a bunch of RCU issues in multi-bssid code cfg80211_update_notlisted_nontrans() leaves the RCU critical session too early, while still using nontrans_ssid which is RCU protected. In addition, it performs a bunch of RCU pointer update operations such as rcu_access_pointer and rcu_assign_pointer. The caller, cfg80211_inform_bss_frame_data(), also accesses the RCU pointer without holding the lock. Just wrap all of this with bss_lock. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho Link: https://lore.kernel.org/r/20191004123706.15768-3-luca@coelho.fi Signed-off-by: Johannes Berg --- net/wireless/scan.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) diff --git a/net/wireless/scan.c b/net/wireless/scan.c index ff1016607f0b..aef240fdf8df 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -1703,8 +1703,7 @@ cfg80211_parse_mbssid_frame_data(struct wiphy *wiphy, static void cfg80211_update_notlisted_nontrans(struct wiphy *wiphy, struct cfg80211_bss *nontrans_bss, - struct ieee80211_mgmt *mgmt, size_t len, - gfp_t gfp) + struct ieee80211_mgmt *mgmt, size_t len) { u8 *ie, *new_ie, *pos; const u8 *nontrans_ssid, *trans_ssid, *mbssid; @@ -1715,6 +1714,8 @@ cfg80211_update_notlisted_nontrans(struct wiphy *wiphy, const struct cfg80211_bss_ies *old; u8 cpy_len; + lockdep_assert_held(&wiphy_to_rdev(wiphy)->bss_lock); + ie = mgmt->u.probe_resp.variable; new_ie_len = ielen; @@ -1731,23 +1732,22 @@ cfg80211_update_notlisted_nontrans(struct wiphy *wiphy, if (!mbssid || mbssid < trans_ssid) return; new_ie_len -= mbssid[1]; - rcu_read_lock(); + nontrans_ssid = ieee80211_bss_get_ie(nontrans_bss, WLAN_EID_SSID); - if (!nontrans_ssid) { - rcu_read_unlock(); + if (!nontrans_ssid) return; - } + new_ie_len += nontrans_ssid[1]; - rcu_read_unlock(); /* generate new ie for nontrans BSS * 1. replace SSID with nontrans BSS' SSID * 2. skip MBSSID IE */ - new_ie = kzalloc(new_ie_len, gfp); + new_ie = kzalloc(new_ie_len, GFP_ATOMIC); if (!new_ie) return; - new_ies = kzalloc(sizeof(*new_ies) + new_ie_len, gfp); + + new_ies = kzalloc(sizeof(*new_ies) + new_ie_len, GFP_ATOMIC); if (!new_ies) goto out_free; @@ -1901,6 +1901,8 @@ cfg80211_inform_bss_frame_data(struct wiphy *wiphy, cfg80211_parse_mbssid_frame_data(wiphy, data, mgmt, len, &non_tx_data, gfp); + spin_lock_bh(&wiphy_to_rdev(wiphy)->bss_lock); + /* check if the res has other nontransmitting bss which is not * in MBSSID IE */ @@ -1915,8 +1917,9 @@ cfg80211_inform_bss_frame_data(struct wiphy *wiphy, ies2 = rcu_access_pointer(tmp_bss->ies); if (ies2->tsf < ies1->tsf) cfg80211_update_notlisted_nontrans(wiphy, tmp_bss, - mgmt, len, gfp); + mgmt, len); } + spin_unlock_bh(&wiphy_to_rdev(wiphy)->bss_lock); return res; } From 95697f9907bfe3eab0ef20265a766b22e27dde64 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 4 Oct 2019 15:37:05 +0300 Subject: [PATCH 026/145] mac80211: accept deauth frames in IBSS mode We can process deauth frames and all, but we drop them very early in the RX path today - this could never have worked. Fixes: 2cc59e784b54 ("mac80211: reply to AUTH with DEAUTH if sta allocation fails in IBSS") Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho Link: https://lore.kernel.org/r/20191004123706.15768-2-luca@coelho.fi Signed-off-by: Johannes Berg --- net/mac80211/rx.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 768d14c9a716..0e05ff037672 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -3467,9 +3467,18 @@ ieee80211_rx_h_mgmt(struct ieee80211_rx_data *rx) case cpu_to_le16(IEEE80211_STYPE_PROBE_RESP): /* process for all: mesh, mlme, ibss */ break; + case cpu_to_le16(IEEE80211_STYPE_DEAUTH): + if (is_multicast_ether_addr(mgmt->da) && + !is_broadcast_ether_addr(mgmt->da)) + return RX_DROP_MONITOR; + + /* process only for station/IBSS */ + if (sdata->vif.type != NL80211_IFTYPE_STATION && + sdata->vif.type != NL80211_IFTYPE_ADHOC) + return RX_DROP_MONITOR; + break; case cpu_to_le16(IEEE80211_STYPE_ASSOC_RESP): case cpu_to_le16(IEEE80211_STYPE_REASSOC_RESP): - case cpu_to_le16(IEEE80211_STYPE_DEAUTH): case cpu_to_le16(IEEE80211_STYPE_DISASSOC): if (is_multicast_ether_addr(mgmt->da) && !is_broadcast_ether_addr(mgmt->da)) From dc0c18ed229cdcca283dd78fefa38273ec37a42c Mon Sep 17 00:00:00 2001 From: Aaron Komisar Date: Wed, 2 Oct 2019 13:59:07 +0000 Subject: [PATCH 027/145] mac80211: fix scan when operating on DFS channels in ETSI domains In non-ETSI regulatory domains scan is blocked when operating channel is a DFS channel. For ETSI, however, once DFS channel is marked as available after the CAC, this channel will remain available (for some time) even after leaving this channel. Therefore a scan can be done without any impact on the availability of the DFS channel as no new CAC is required after the scan. Enable scan in mac80211 in these cases. Signed-off-by: Aaron Komisar Link: https://lore.kernel.org/r/1570024728-17284-1-git-send-email-aaron.komisar@tandemg.com Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 8 ++++++++ net/mac80211/scan.c | 30 ++++++++++++++++++++++++++++-- net/wireless/reg.c | 1 + net/wireless/reg.h | 8 -------- 4 files changed, 37 insertions(+), 10 deletions(-) diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index ff45c3e1abff..4ab2c49423dc 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -5549,6 +5549,14 @@ const struct ieee80211_reg_rule *freq_reg_info(struct wiphy *wiphy, */ const char *reg_initiator_name(enum nl80211_reg_initiator initiator); +/** + * regulatory_pre_cac_allowed - check if pre-CAC allowed in the current regdom + * @wiphy: wiphy for which pre-CAC capability is checked. + * + * Pre-CAC is allowed only in some regdomains (notable ETSI). + */ +bool regulatory_pre_cac_allowed(struct wiphy *wiphy); + /** * DOC: Internal regulatory db functions * diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index adf94ba1ed77..4d31d9688dc2 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -520,10 +520,33 @@ static int ieee80211_start_sw_scan(struct ieee80211_local *local, return 0; } +static bool __ieee80211_can_leave_ch(struct ieee80211_sub_if_data *sdata) +{ + struct ieee80211_local *local = sdata->local; + struct ieee80211_sub_if_data *sdata_iter; + + if (!ieee80211_is_radar_required(local)) + return true; + + if (!regulatory_pre_cac_allowed(local->hw.wiphy)) + return false; + + mutex_lock(&local->iflist_mtx); + list_for_each_entry(sdata_iter, &local->interfaces, list) { + if (sdata_iter->wdev.cac_started) { + mutex_unlock(&local->iflist_mtx); + return false; + } + } + mutex_unlock(&local->iflist_mtx); + + return true; +} + static bool ieee80211_can_scan(struct ieee80211_local *local, struct ieee80211_sub_if_data *sdata) { - if (ieee80211_is_radar_required(local)) + if (!__ieee80211_can_leave_ch(sdata)) return false; if (!list_empty(&local->roc_list)) @@ -630,7 +653,10 @@ static int __ieee80211_start_scan(struct ieee80211_sub_if_data *sdata, lockdep_assert_held(&local->mtx); - if (local->scan_req || ieee80211_is_radar_required(local)) + if (local->scan_req) + return -EBUSY; + + if (!__ieee80211_can_leave_ch(sdata)) return -EBUSY; if (!ieee80211_can_scan(local, sdata)) { diff --git a/net/wireless/reg.c b/net/wireless/reg.c index 420c4207ab59..446c76d44e65 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -3883,6 +3883,7 @@ bool regulatory_pre_cac_allowed(struct wiphy *wiphy) return pre_cac_allowed; } +EXPORT_SYMBOL(regulatory_pre_cac_allowed); void regulatory_propagate_dfs_state(struct wiphy *wiphy, struct cfg80211_chan_def *chandef, diff --git a/net/wireless/reg.h b/net/wireless/reg.h index 504133d76de4..dc8f689bd469 100644 --- a/net/wireless/reg.h +++ b/net/wireless/reg.h @@ -155,14 +155,6 @@ bool regulatory_indoor_allowed(void); */ #define REG_PRE_CAC_EXPIRY_GRACE_MS 2000 -/** - * regulatory_pre_cac_allowed - if pre-CAC allowed in the current dfs domain - * @wiphy: wiphy for which pre-CAC capability is checked. - - * Pre-CAC is allowed only in ETSI domain. - */ -bool regulatory_pre_cac_allowed(struct wiphy *wiphy); - /** * regulatory_propagate_dfs_state - Propagate DFS channel state to other wiphys * @wiphy - wiphy on which radar is detected and the event will be propagated From 98d22b01f9f6f85eb8870290006241b316829fd3 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 2 Oct 2019 14:50:52 -0700 Subject: [PATCH 028/145] rt2x00: remove input-polldev.h header The driver does not use input subsystem so we do not need this header, and it is being removed, so stop pulling it in. Signed-off-by: Dmitry Torokhov Signed-off-by: Kalle Valo --- drivers/net/wireless/ralink/rt2x00/rt2x00.h | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00.h b/drivers/net/wireless/ralink/rt2x00/rt2x00.h index 2b216edd0c7d..a90a518b40d3 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00.h +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00.h @@ -23,7 +23,6 @@ #include #include #include -#include #include #include #include From c6ee11c39fcc1fb55130748990a8f199e76263b4 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Sun, 6 Oct 2019 14:24:24 -0700 Subject: [PATCH 029/145] llc: fix sk_buff leak in llc_sap_state_process() syzbot reported: BUG: memory leak unreferenced object 0xffff888116270800 (size 224): comm "syz-executor641", pid 7047, jiffies 4294947360 (age 13.860s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 20 e1 2a 81 88 ff ff 00 40 3d 2a 81 88 ff ff . .*.....@=*.... backtrace: [<000000004d41b4cc>] kmemleak_alloc_recursive include/linux/kmemleak.h:55 [inline] [<000000004d41b4cc>] slab_post_alloc_hook mm/slab.h:439 [inline] [<000000004d41b4cc>] slab_alloc_node mm/slab.c:3269 [inline] [<000000004d41b4cc>] kmem_cache_alloc_node+0x153/0x2a0 mm/slab.c:3579 [<00000000506a5965>] __alloc_skb+0x6e/0x210 net/core/skbuff.c:198 [<000000001ba5a161>] alloc_skb include/linux/skbuff.h:1058 [inline] [<000000001ba5a161>] alloc_skb_with_frags+0x5f/0x250 net/core/skbuff.c:5327 [<0000000047d9c78b>] sock_alloc_send_pskb+0x269/0x2a0 net/core/sock.c:2225 [<000000003828fe54>] sock_alloc_send_skb+0x32/0x40 net/core/sock.c:2242 [<00000000e34d94f9>] llc_ui_sendmsg+0x10a/0x540 net/llc/af_llc.c:933 [<00000000de2de3fb>] sock_sendmsg_nosec net/socket.c:652 [inline] [<00000000de2de3fb>] sock_sendmsg+0x54/0x70 net/socket.c:671 [<000000008fe16e7a>] __sys_sendto+0x148/0x1f0 net/socket.c:1964 [...] The bug is that llc_sap_state_process() always takes an extra reference to the skb, but sometimes neither llc_sap_next_state() nor llc_sap_state_process() itself drops this reference. Fix it by changing llc_sap_next_state() to never consume a reference to the skb, rather than sometimes do so and sometimes not. Then remove the extra skb_get() and kfree_skb() from llc_sap_state_process(). Reported-by: syzbot+6bf095f9becf5efef645@syzkaller.appspotmail.com Reported-by: syzbot+31c16aa4202dace3812e@syzkaller.appspotmail.com Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Biggers Signed-off-by: Jakub Kicinski --- net/llc/llc_s_ac.c | 12 +++++++++--- net/llc/llc_sap.c | 23 ++++++++--------------- 2 files changed, 17 insertions(+), 18 deletions(-) diff --git a/net/llc/llc_s_ac.c b/net/llc/llc_s_ac.c index a94bd56bcac6..7ae4cc684d3a 100644 --- a/net/llc/llc_s_ac.c +++ b/net/llc/llc_s_ac.c @@ -58,8 +58,10 @@ int llc_sap_action_send_ui(struct llc_sap *sap, struct sk_buff *skb) ev->daddr.lsap, LLC_PDU_CMD); llc_pdu_init_as_ui_cmd(skb); rc = llc_mac_hdr_init(skb, ev->saddr.mac, ev->daddr.mac); - if (likely(!rc)) + if (likely(!rc)) { + skb_get(skb); rc = dev_queue_xmit(skb); + } return rc; } @@ -81,8 +83,10 @@ int llc_sap_action_send_xid_c(struct llc_sap *sap, struct sk_buff *skb) ev->daddr.lsap, LLC_PDU_CMD); llc_pdu_init_as_xid_cmd(skb, LLC_XID_NULL_CLASS_2, 0); rc = llc_mac_hdr_init(skb, ev->saddr.mac, ev->daddr.mac); - if (likely(!rc)) + if (likely(!rc)) { + skb_get(skb); rc = dev_queue_xmit(skb); + } return rc; } @@ -135,8 +139,10 @@ int llc_sap_action_send_test_c(struct llc_sap *sap, struct sk_buff *skb) ev->daddr.lsap, LLC_PDU_CMD); llc_pdu_init_as_test_cmd(skb); rc = llc_mac_hdr_init(skb, ev->saddr.mac, ev->daddr.mac); - if (likely(!rc)) + if (likely(!rc)) { + skb_get(skb); rc = dev_queue_xmit(skb); + } return rc; } diff --git a/net/llc/llc_sap.c b/net/llc/llc_sap.c index a7f7b8ff4729..be419062e19a 100644 --- a/net/llc/llc_sap.c +++ b/net/llc/llc_sap.c @@ -197,29 +197,22 @@ out: * After executing actions of the event, upper layer will be indicated * if needed(on receiving an UI frame). sk can be null for the * datalink_proto case. + * + * This function always consumes a reference to the skb. */ static void llc_sap_state_process(struct llc_sap *sap, struct sk_buff *skb) { struct llc_sap_state_ev *ev = llc_sap_ev(skb); - /* - * We have to hold the skb, because llc_sap_next_state - * will kfree it in the sending path and we need to - * look at the skb->cb, where we encode llc_sap_state_ev. - */ - skb_get(skb); ev->ind_cfm_flag = 0; llc_sap_next_state(sap, skb); - if (ev->ind_cfm_flag == LLC_IND) { - if (skb->sk->sk_state == TCP_LISTEN) - kfree_skb(skb); - else { - llc_save_primitive(skb->sk, skb, ev->prim); - /* queue skb to the user. */ - if (sock_queue_rcv_skb(skb->sk, skb)) - kfree_skb(skb); - } + if (ev->ind_cfm_flag == LLC_IND && skb->sk->sk_state != TCP_LISTEN) { + llc_save_primitive(skb->sk, skb, ev->prim); + + /* queue skb to the user. */ + if (sock_queue_rcv_skb(skb->sk, skb) == 0) + return; } kfree_skb(skb); } From b74555de21acd791f12c4a1aeaf653dd7ac21133 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Sun, 6 Oct 2019 14:24:25 -0700 Subject: [PATCH 030/145] llc: fix sk_buff leak in llc_conn_service() syzbot reported: BUG: memory leak unreferenced object 0xffff88811eb3de00 (size 224): comm "syz-executor559", pid 7315, jiffies 4294943019 (age 10.300s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 a0 38 24 81 88 ff ff 00 c0 f2 15 81 88 ff ff ..8$............ backtrace: [<000000008d1c66a1>] kmemleak_alloc_recursive include/linux/kmemleak.h:55 [inline] [<000000008d1c66a1>] slab_post_alloc_hook mm/slab.h:439 [inline] [<000000008d1c66a1>] slab_alloc_node mm/slab.c:3269 [inline] [<000000008d1c66a1>] kmem_cache_alloc_node+0x153/0x2a0 mm/slab.c:3579 [<00000000447d9496>] __alloc_skb+0x6e/0x210 net/core/skbuff.c:198 [<000000000cdbf82f>] alloc_skb include/linux/skbuff.h:1058 [inline] [<000000000cdbf82f>] llc_alloc_frame+0x66/0x110 net/llc/llc_sap.c:54 [<000000002418b52e>] llc_conn_ac_send_sabme_cmd_p_set_x+0x2f/0x140 net/llc/llc_c_ac.c:777 [<000000001372ae17>] llc_exec_conn_trans_actions net/llc/llc_conn.c:475 [inline] [<000000001372ae17>] llc_conn_service net/llc/llc_conn.c:400 [inline] [<000000001372ae17>] llc_conn_state_process+0x1ac/0x640 net/llc/llc_conn.c:75 [<00000000f27e53c1>] llc_establish_connection+0x110/0x170 net/llc/llc_if.c:109 [<00000000291b2ca0>] llc_ui_connect+0x10e/0x370 net/llc/af_llc.c:477 [<000000000f9c740b>] __sys_connect+0x11d/0x170 net/socket.c:1840 [...] The bug is that most callers of llc_conn_send_pdu() assume it consumes a reference to the skb, when actually due to commit b85ab56c3f81 ("llc: properly handle dev_queue_xmit() return value") it doesn't. Revert most of that commit, and instead make the few places that need llc_conn_send_pdu() to *not* consume a reference call skb_get() before. Fixes: b85ab56c3f81 ("llc: properly handle dev_queue_xmit() return value") Reported-by: syzbot+6b825a6494a04cc0e3f7@syzkaller.appspotmail.com Signed-off-by: Eric Biggers Signed-off-by: Jakub Kicinski --- include/net/llc_conn.h | 2 +- net/llc/llc_c_ac.c | 8 ++++++-- net/llc/llc_conn.c | 32 +++++++++----------------------- 3 files changed, 16 insertions(+), 26 deletions(-) diff --git a/include/net/llc_conn.h b/include/net/llc_conn.h index df528a623548..ea985aa7a6c5 100644 --- a/include/net/llc_conn.h +++ b/include/net/llc_conn.h @@ -104,7 +104,7 @@ void llc_sk_reset(struct sock *sk); /* Access to a connection */ int llc_conn_state_process(struct sock *sk, struct sk_buff *skb); -int llc_conn_send_pdu(struct sock *sk, struct sk_buff *skb); +void llc_conn_send_pdu(struct sock *sk, struct sk_buff *skb); void llc_conn_rtn_pdu(struct sock *sk, struct sk_buff *skb); void llc_conn_resend_i_pdu_as_cmd(struct sock *sk, u8 nr, u8 first_p_bit); void llc_conn_resend_i_pdu_as_rsp(struct sock *sk, u8 nr, u8 first_f_bit); diff --git a/net/llc/llc_c_ac.c b/net/llc/llc_c_ac.c index 4d78375f9872..647c0554d04c 100644 --- a/net/llc/llc_c_ac.c +++ b/net/llc/llc_c_ac.c @@ -372,6 +372,7 @@ int llc_conn_ac_send_i_cmd_p_set_1(struct sock *sk, struct sk_buff *skb) llc_pdu_init_as_i_cmd(skb, 1, llc->vS, llc->vR); rc = llc_mac_hdr_init(skb, llc->dev->dev_addr, llc->daddr.mac); if (likely(!rc)) { + skb_get(skb); llc_conn_send_pdu(sk, skb); llc_conn_ac_inc_vs_by_1(sk, skb); } @@ -389,7 +390,8 @@ static int llc_conn_ac_send_i_cmd_p_set_0(struct sock *sk, struct sk_buff *skb) llc_pdu_init_as_i_cmd(skb, 0, llc->vS, llc->vR); rc = llc_mac_hdr_init(skb, llc->dev->dev_addr, llc->daddr.mac); if (likely(!rc)) { - rc = llc_conn_send_pdu(sk, skb); + skb_get(skb); + llc_conn_send_pdu(sk, skb); llc_conn_ac_inc_vs_by_1(sk, skb); } return rc; @@ -406,6 +408,7 @@ int llc_conn_ac_send_i_xxx_x_set_0(struct sock *sk, struct sk_buff *skb) llc_pdu_init_as_i_cmd(skb, 0, llc->vS, llc->vR); rc = llc_mac_hdr_init(skb, llc->dev->dev_addr, llc->daddr.mac); if (likely(!rc)) { + skb_get(skb); llc_conn_send_pdu(sk, skb); llc_conn_ac_inc_vs_by_1(sk, skb); } @@ -916,7 +919,8 @@ static int llc_conn_ac_send_i_rsp_f_set_ackpf(struct sock *sk, llc_pdu_init_as_i_cmd(skb, llc->ack_pf, llc->vS, llc->vR); rc = llc_mac_hdr_init(skb, llc->dev->dev_addr, llc->daddr.mac); if (likely(!rc)) { - rc = llc_conn_send_pdu(sk, skb); + skb_get(skb); + llc_conn_send_pdu(sk, skb); llc_conn_ac_inc_vs_by_1(sk, skb); } return rc; diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c index 4ff89cb7c86f..ed2aca12460c 100644 --- a/net/llc/llc_conn.c +++ b/net/llc/llc_conn.c @@ -30,7 +30,7 @@ #endif static int llc_find_offset(int state, int ev_type); -static int llc_conn_send_pdus(struct sock *sk, struct sk_buff *skb); +static void llc_conn_send_pdus(struct sock *sk); static int llc_conn_service(struct sock *sk, struct sk_buff *skb); static int llc_exec_conn_trans_actions(struct sock *sk, struct llc_conn_state_trans *trans, @@ -193,11 +193,11 @@ out_skb_put: return rc; } -int llc_conn_send_pdu(struct sock *sk, struct sk_buff *skb) +void llc_conn_send_pdu(struct sock *sk, struct sk_buff *skb) { /* queue PDU to send to MAC layer */ skb_queue_tail(&sk->sk_write_queue, skb); - return llc_conn_send_pdus(sk, skb); + llc_conn_send_pdus(sk); } /** @@ -255,7 +255,7 @@ void llc_conn_resend_i_pdu_as_cmd(struct sock *sk, u8 nr, u8 first_p_bit) if (howmany_resend > 0) llc->vS = (llc->vS + 1) % LLC_2_SEQ_NBR_MODULO; /* any PDUs to re-send are queued up; start sending to MAC */ - llc_conn_send_pdus(sk, NULL); + llc_conn_send_pdus(sk); out:; } @@ -296,7 +296,7 @@ void llc_conn_resend_i_pdu_as_rsp(struct sock *sk, u8 nr, u8 first_f_bit) if (howmany_resend > 0) llc->vS = (llc->vS + 1) % LLC_2_SEQ_NBR_MODULO; /* any PDUs to re-send are queued up; start sending to MAC */ - llc_conn_send_pdus(sk, NULL); + llc_conn_send_pdus(sk); out:; } @@ -340,16 +340,12 @@ out: /** * llc_conn_send_pdus - Sends queued PDUs * @sk: active connection - * @hold_skb: the skb held by caller, or NULL if does not care * - * Sends queued pdus to MAC layer for transmission. When @hold_skb is - * NULL, always return 0. Otherwise, return 0 if @hold_skb is sent - * successfully, or 1 for failure. + * Sends queued pdus to MAC layer for transmission. */ -static int llc_conn_send_pdus(struct sock *sk, struct sk_buff *hold_skb) +static void llc_conn_send_pdus(struct sock *sk) { struct sk_buff *skb; - int ret = 0; while ((skb = skb_dequeue(&sk->sk_write_queue)) != NULL) { struct llc_pdu_sn *pdu = llc_pdu_sn_hdr(skb); @@ -361,20 +357,10 @@ static int llc_conn_send_pdus(struct sock *sk, struct sk_buff *hold_skb) skb_queue_tail(&llc_sk(sk)->pdu_unack_q, skb); if (!skb2) break; - dev_queue_xmit(skb2); - } else { - bool is_target = skb == hold_skb; - int rc; - - if (is_target) - skb_get(skb); - rc = dev_queue_xmit(skb); - if (is_target) - ret = rc; + skb = skb2; } + dev_queue_xmit(skb); } - - return ret; } /** From fc8d5db10cbe1338a52ebc74e7feab9276721774 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Sun, 6 Oct 2019 14:24:26 -0700 Subject: [PATCH 031/145] llc: fix another potential sk_buff leak in llc_ui_sendmsg() All callers of llc_conn_state_process() except llc_build_and_send_pkt() (via llc_ui_sendmsg() -> llc_ui_send_data()) assume that it always consumes a reference to the skb. Fix this caller to do the same. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Biggers Signed-off-by: Jakub Kicinski --- net/llc/af_llc.c | 34 ++++++++++++++++++++-------------- net/llc/llc_conn.c | 2 ++ net/llc/llc_if.c | 12 ++++++++---- 3 files changed, 30 insertions(+), 18 deletions(-) diff --git a/net/llc/af_llc.c b/net/llc/af_llc.c index 2017b7d780f5..c74f44dfaa22 100644 --- a/net/llc/af_llc.c +++ b/net/llc/af_llc.c @@ -113,22 +113,26 @@ static inline u8 llc_ui_header_len(struct sock *sk, struct sockaddr_llc *addr) * * Send data via reliable llc2 connection. * Returns 0 upon success, non-zero if action did not succeed. + * + * This function always consumes a reference to the skb. */ static int llc_ui_send_data(struct sock* sk, struct sk_buff *skb, int noblock) { struct llc_sock* llc = llc_sk(sk); - int rc = 0; if (unlikely(llc_data_accept_state(llc->state) || llc->remote_busy_flag || llc->p_flag)) { long timeout = sock_sndtimeo(sk, noblock); + int rc; rc = llc_ui_wait_for_busy_core(sk, timeout); + if (rc) { + kfree_skb(skb); + return rc; + } } - if (unlikely(!rc)) - rc = llc_build_and_send_pkt(sk, skb); - return rc; + return llc_build_and_send_pkt(sk, skb); } static void llc_ui_sk_init(struct socket *sock, struct sock *sk) @@ -899,7 +903,7 @@ static int llc_ui_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) DECLARE_SOCKADDR(struct sockaddr_llc *, addr, msg->msg_name); int flags = msg->msg_flags; int noblock = flags & MSG_DONTWAIT; - struct sk_buff *skb; + struct sk_buff *skb = NULL; size_t size = 0; int rc = -EINVAL, copied = 0, hdrlen; @@ -908,10 +912,10 @@ static int llc_ui_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) lock_sock(sk); if (addr) { if (msg->msg_namelen < sizeof(*addr)) - goto release; + goto out; } else { if (llc_ui_addr_null(&llc->addr)) - goto release; + goto out; addr = &llc->addr; } /* must bind connection to sap if user hasn't done it. */ @@ -919,7 +923,7 @@ static int llc_ui_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) /* bind to sap with null dev, exclusive. */ rc = llc_ui_autobind(sock, addr); if (rc) - goto release; + goto out; } hdrlen = llc->dev->hard_header_len + llc_ui_header_len(sk, addr); size = hdrlen + len; @@ -928,12 +932,12 @@ static int llc_ui_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) copied = size - hdrlen; rc = -EINVAL; if (copied < 0) - goto release; + goto out; release_sock(sk); skb = sock_alloc_send_skb(sk, size, noblock, &rc); lock_sock(sk); if (!skb) - goto release; + goto out; skb->dev = llc->dev; skb->protocol = llc_proto_type(addr->sllc_arphrd); skb_reserve(skb, hdrlen); @@ -943,29 +947,31 @@ static int llc_ui_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) if (sk->sk_type == SOCK_DGRAM || addr->sllc_ua) { llc_build_and_send_ui_pkt(llc->sap, skb, addr->sllc_mac, addr->sllc_sap); + skb = NULL; goto out; } if (addr->sllc_test) { llc_build_and_send_test_pkt(llc->sap, skb, addr->sllc_mac, addr->sllc_sap); + skb = NULL; goto out; } if (addr->sllc_xid) { llc_build_and_send_xid_pkt(llc->sap, skb, addr->sllc_mac, addr->sllc_sap); + skb = NULL; goto out; } rc = -ENOPROTOOPT; if (!(sk->sk_type == SOCK_STREAM && !addr->sllc_ua)) goto out; rc = llc_ui_send_data(sk, skb, noblock); + skb = NULL; out: - if (rc) { - kfree_skb(skb); -release: + kfree_skb(skb); + if (rc) dprintk("%s: failed sending from %02X to %02X: %d\n", __func__, llc->laddr.lsap, llc->daddr.lsap, rc); - } release_sock(sk); return rc ? : copied; } diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c index ed2aca12460c..0b0c6f12153b 100644 --- a/net/llc/llc_conn.c +++ b/net/llc/llc_conn.c @@ -55,6 +55,8 @@ int sysctl_llc2_busy_timeout = LLC2_BUSY_TIME * HZ; * (executing it's actions and changing state), upper layer will be * indicated or confirmed, if needed. Returns 0 for success, 1 for * failure. The socket lock has to be held before calling this function. + * + * This function always consumes a reference to the skb. */ int llc_conn_state_process(struct sock *sk, struct sk_buff *skb) { diff --git a/net/llc/llc_if.c b/net/llc/llc_if.c index 8db03c2d5440..ad6547736c21 100644 --- a/net/llc/llc_if.c +++ b/net/llc/llc_if.c @@ -38,6 +38,8 @@ * closed and -EBUSY when sending data is not permitted in this state or * LLC has send an I pdu with p bit set to 1 and is waiting for it's * response. + * + * This function always consumes a reference to the skb. */ int llc_build_and_send_pkt(struct sock *sk, struct sk_buff *skb) { @@ -46,20 +48,22 @@ int llc_build_and_send_pkt(struct sock *sk, struct sk_buff *skb) struct llc_sock *llc = llc_sk(sk); if (unlikely(llc->state == LLC_CONN_STATE_ADM)) - goto out; + goto out_free; rc = -EBUSY; if (unlikely(llc_data_accept_state(llc->state) || /* data_conn_refuse */ llc->p_flag)) { llc->failed_data_req = 1; - goto out; + goto out_free; } ev = llc_conn_ev(skb); ev->type = LLC_CONN_EV_TYPE_PRIM; ev->prim = LLC_DATA_PRIM; ev->prim_type = LLC_PRIM_TYPE_REQ; skb->dev = llc->dev; - rc = llc_conn_state_process(sk, skb); -out: + return llc_conn_state_process(sk, skb); + +out_free: + kfree_skb(skb); return rc; } From 36453c852816f19947ca482a595dffdd2efa4965 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Sun, 6 Oct 2019 14:24:27 -0700 Subject: [PATCH 032/145] llc: fix sk_buff refcounting in llc_conn_state_process() If llc_conn_state_process() sees that llc_conn_service() put the skb on a list, it will drop one fewer references to it. This is wrong because the current behavior is that llc_conn_service() never consumes a reference to the skb. The code also makes the number of skb references being dropped conditional on which of ind_prim and cfm_prim are nonzero, yet neither of these affects how many references are *acquired*. So there is extra code that tries to fix this up by sometimes taking another reference. Remove the unnecessary/broken refcounting logic and instead just add an skb_get() before the only two places where an extra reference is actually consumed. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Biggers Signed-off-by: Jakub Kicinski --- net/llc/llc_conn.c | 33 ++++++--------------------------- 1 file changed, 6 insertions(+), 27 deletions(-) diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c index 0b0c6f12153b..a79b739eb223 100644 --- a/net/llc/llc_conn.c +++ b/net/llc/llc_conn.c @@ -64,12 +64,6 @@ int llc_conn_state_process(struct sock *sk, struct sk_buff *skb) struct llc_sock *llc = llc_sk(skb->sk); struct llc_conn_state_ev *ev = llc_conn_ev(skb); - /* - * We have to hold the skb, because llc_conn_service will kfree it in - * the sending path and we need to look at the skb->cb, where we encode - * llc_conn_state_ev. - */ - skb_get(skb); ev->ind_prim = ev->cfm_prim = 0; /* * Send event to state machine @@ -77,21 +71,12 @@ int llc_conn_state_process(struct sock *sk, struct sk_buff *skb) rc = llc_conn_service(skb->sk, skb); if (unlikely(rc != 0)) { printk(KERN_ERR "%s: llc_conn_service failed\n", __func__); - goto out_kfree_skb; - } - - if (unlikely(!ev->ind_prim && !ev->cfm_prim)) { - /* indicate or confirm not required */ - if (!skb->next) - goto out_kfree_skb; goto out_skb_put; } - if (unlikely(ev->ind_prim && ev->cfm_prim)) /* Paranoia */ - skb_get(skb); - switch (ev->ind_prim) { case LLC_DATA_PRIM: + skb_get(skb); llc_save_primitive(sk, skb, LLC_DATA_PRIM); if (unlikely(sock_queue_rcv_skb(sk, skb))) { /* @@ -108,6 +93,7 @@ int llc_conn_state_process(struct sock *sk, struct sk_buff *skb) * skb->sk pointing to the newly created struct sock in * llc_conn_handler. -acme */ + skb_get(skb); skb_queue_tail(&sk->sk_receive_queue, skb); sk->sk_state_change(sk); break; @@ -123,7 +109,6 @@ int llc_conn_state_process(struct sock *sk, struct sk_buff *skb) sk->sk_state_change(sk); } } - kfree_skb(skb); sock_put(sk); break; case LLC_RESET_PRIM: @@ -132,14 +117,11 @@ int llc_conn_state_process(struct sock *sk, struct sk_buff *skb) * RESET is not being notified to upper layers for now */ printk(KERN_INFO "%s: received a reset ind!\n", __func__); - kfree_skb(skb); break; default: - if (ev->ind_prim) { + if (ev->ind_prim) printk(KERN_INFO "%s: received unknown %d prim!\n", __func__, ev->ind_prim); - kfree_skb(skb); - } /* No indication */ break; } @@ -181,15 +163,12 @@ int llc_conn_state_process(struct sock *sk, struct sk_buff *skb) printk(KERN_INFO "%s: received a reset conf!\n", __func__); break; default: - if (ev->cfm_prim) { + if (ev->cfm_prim) printk(KERN_INFO "%s: received unknown %d prim!\n", __func__, ev->cfm_prim); - break; - } - goto out_skb_put; /* No confirmation */ + /* No confirmation */ + break; } -out_kfree_skb: - kfree_skb(skb); out_skb_put: kfree_skb(skb); return rc; From fd418b01fe26c2430b1091675cceb3ab2b52e1e0 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Tue, 8 Oct 2019 15:10:44 +0200 Subject: [PATCH 033/145] selftests/bpf: Set rp_filter in test_flow_dissector Many distributions enable rp_filter. However, the flow dissector test generates packets that have 1.1.1.1 set as (inner) source address without this address being reachable. This causes the selftest to fail. The selftests should not assume a particular initial configuration. Switch off rp_filter. Fixes: 50b3ed57dee9 ("selftests/bpf: test bpf flow dissection") Signed-off-by: Jiri Benc Signed-off-by: Daniel Borkmann Acked-by: Petar Penkov Link: https://lore.kernel.org/bpf/513a298f53e99561d2f70b2e60e2858ea6cda754.1570539863.git.jbenc@redhat.com --- tools/testing/selftests/bpf/test_flow_dissector.sh | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/testing/selftests/bpf/test_flow_dissector.sh b/tools/testing/selftests/bpf/test_flow_dissector.sh index d23d4da66b83..e2d06191bd35 100755 --- a/tools/testing/selftests/bpf/test_flow_dissector.sh +++ b/tools/testing/selftests/bpf/test_flow_dissector.sh @@ -63,6 +63,9 @@ fi # Setup tc qdisc add dev lo ingress +echo 0 > /proc/sys/net/ipv4/conf/default/rp_filter +echo 0 > /proc/sys/net/ipv4/conf/all/rp_filter +echo 0 > /proc/sys/net/ipv4/conf/lo/rp_filter echo "Testing IPv4..." # Drops all IP/UDP packets coming from port 9 From 106c35dda32f8b63f88cad7433f1b8bb0056958a Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Tue, 8 Oct 2019 15:10:45 +0200 Subject: [PATCH 034/145] selftests/bpf: More compatible nc options in test_lwt_ip_encap Out of the three nc implementations widely in use, at least two (BSD netcat and nmap-ncat) do not support -l combined with -s. Modify the nc invocation to be accepted by all of them. Fixes: 17a90a788473 ("selftests/bpf: test that GSO works in lwt_ip_encap") Signed-off-by: Jiri Benc Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/9f177682c387f3f943bb64d849e6c6774df3c5b4.1570539863.git.jbenc@redhat.com --- tools/testing/selftests/bpf/test_lwt_ip_encap.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/bpf/test_lwt_ip_encap.sh b/tools/testing/selftests/bpf/test_lwt_ip_encap.sh index acf7a74f97cd..59ea56945e6c 100755 --- a/tools/testing/selftests/bpf/test_lwt_ip_encap.sh +++ b/tools/testing/selftests/bpf/test_lwt_ip_encap.sh @@ -314,15 +314,15 @@ test_gso() command -v nc >/dev/null 2>&1 || \ { echo >&2 "nc is not available: skipping TSO tests"; return; } - # listen on IPv*_DST, capture TCP into $TMPFILE + # listen on port 9000, capture TCP into $TMPFILE if [ "${PROTO}" == "IPv4" ] ; then IP_DST=${IPv4_DST} ip netns exec ${NS3} bash -c \ - "nc -4 -l -s ${IPv4_DST} -p 9000 > ${TMPFILE} &" + "nc -4 -l -p 9000 > ${TMPFILE} &" elif [ "${PROTO}" == "IPv6" ] ; then IP_DST=${IPv6_DST} ip netns exec ${NS3} bash -c \ - "nc -6 -l -s ${IPv6_DST} -p 9000 > ${TMPFILE} &" + "nc -6 -l -p 9000 > ${TMPFILE} &" RET=$? else echo " test_gso: unknown PROTO: ${PROTO}" From 0041412694eca70387aee4076254fbed8222700a Mon Sep 17 00:00:00 2001 From: Alex Vesker Date: Mon, 7 Oct 2019 16:13:25 +0300 Subject: [PATCH 035/145] net/mlx5: DR, Allow insertion of duplicate rules Duplicate rules were not allowed to be configured with SW steering. This restriction caused failures with the replace rule logic done by upper layers. This fix allows for multiple rules with the same match values, in such case the first inserted rules will match. Fixes: 41d07074154c ("net/mlx5: DR, Expose steering rule functionality") Signed-off-by: Alex Vesker Signed-off-by: Tariq Toukan Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/steering/dr_rule.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c index 4187f2b112b8..e8b656075c6f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c @@ -788,12 +788,10 @@ again: * it means that all the previous stes are the same, * if so, this rule is duplicated. */ - if (mlx5dr_ste_is_last_in_rule(nic_matcher, - matched_ste->ste_chain_location)) { - mlx5dr_info(dmn, "Duplicate rule inserted, aborting!!\n"); - return NULL; - } - return matched_ste; + if (!mlx5dr_ste_is_last_in_rule(nic_matcher, ste_location)) + return matched_ste; + + mlx5dr_dbg(dmn, "Duplicate rule inserted\n"); } if (!skip_rehash && dr_rule_need_enlarge_hash(cur_htbl, dmn, nic_dmn)) { From 6f96c3c6904c26cea9ca2726d5d8a9b0b8205b3c Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Mon, 7 Oct 2019 13:26:28 -0700 Subject: [PATCH 036/145] net_sched: fix backward compatibility for TCA_KIND Marcelo noticed a backward compatibility issue of TCA_KIND after we move from NLA_STRING to NLA_NUL_STRING, so it is probably too late to change it. Instead, to make everyone happy, we can just insert a NUL to terminate the string with nla_strlcpy() like we do for TC actions. Fixes: 62794fc4fbf5 ("net_sched: add max len check for TCA_KIND") Reported-by: Marcelo Ricardo Leitner Cc: Jamal Hadi Salim Cc: Jiri Pirko Signed-off-by: Cong Wang Reviewed-by: Marcelo Ricardo Leitner Signed-off-by: Jakub Kicinski --- net/sched/cls_api.c | 36 +++++++++++++++++++++++++++++++++--- net/sched/sch_api.c | 3 +-- 2 files changed, 34 insertions(+), 5 deletions(-) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 64584a1df425..8717c0b26c90 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -162,11 +162,22 @@ static inline u32 tcf_auto_prio(struct tcf_proto *tp) return TC_H_MAJ(first); } +static bool tcf_proto_check_kind(struct nlattr *kind, char *name) +{ + if (kind) + return nla_strlcpy(name, kind, IFNAMSIZ) >= IFNAMSIZ; + memset(name, 0, IFNAMSIZ); + return false; +} + static bool tcf_proto_is_unlocked(const char *kind) { const struct tcf_proto_ops *ops; bool ret; + if (strlen(kind) == 0) + return false; + ops = tcf_proto_lookup_ops(kind, false, NULL); /* On error return false to take rtnl lock. Proto lookup/create * functions will perform lookup again and properly handle errors. @@ -1843,6 +1854,7 @@ static int tc_new_tfilter(struct sk_buff *skb, struct nlmsghdr *n, { struct net *net = sock_net(skb->sk); struct nlattr *tca[TCA_MAX + 1]; + char name[IFNAMSIZ]; struct tcmsg *t; u32 protocol; u32 prio; @@ -1899,13 +1911,19 @@ replay: if (err) return err; + if (tcf_proto_check_kind(tca[TCA_KIND], name)) { + NL_SET_ERR_MSG(extack, "Specified TC filter name too long"); + err = -EINVAL; + goto errout; + } + /* Take rtnl mutex if rtnl_held was set to true on previous iteration, * block is shared (no qdisc found), qdisc is not unlocked, classifier * type is not specified, classifier is not unlocked. */ if (rtnl_held || (q && !(q->ops->cl_ops->flags & QDISC_CLASS_OPS_DOIT_UNLOCKED)) || - !tca[TCA_KIND] || !tcf_proto_is_unlocked(nla_data(tca[TCA_KIND]))) { + !tcf_proto_is_unlocked(name)) { rtnl_held = true; rtnl_lock(); } @@ -2063,6 +2081,7 @@ static int tc_del_tfilter(struct sk_buff *skb, struct nlmsghdr *n, { struct net *net = sock_net(skb->sk); struct nlattr *tca[TCA_MAX + 1]; + char name[IFNAMSIZ]; struct tcmsg *t; u32 protocol; u32 prio; @@ -2102,13 +2121,18 @@ static int tc_del_tfilter(struct sk_buff *skb, struct nlmsghdr *n, if (err) return err; + if (tcf_proto_check_kind(tca[TCA_KIND], name)) { + NL_SET_ERR_MSG(extack, "Specified TC filter name too long"); + err = -EINVAL; + goto errout; + } /* Take rtnl mutex if flushing whole chain, block is shared (no qdisc * found), qdisc is not unlocked, classifier type is not specified, * classifier is not unlocked. */ if (!prio || (q && !(q->ops->cl_ops->flags & QDISC_CLASS_OPS_DOIT_UNLOCKED)) || - !tca[TCA_KIND] || !tcf_proto_is_unlocked(nla_data(tca[TCA_KIND]))) { + !tcf_proto_is_unlocked(name)) { rtnl_held = true; rtnl_lock(); } @@ -2216,6 +2240,7 @@ static int tc_get_tfilter(struct sk_buff *skb, struct nlmsghdr *n, { struct net *net = sock_net(skb->sk); struct nlattr *tca[TCA_MAX + 1]; + char name[IFNAMSIZ]; struct tcmsg *t; u32 protocol; u32 prio; @@ -2252,12 +2277,17 @@ static int tc_get_tfilter(struct sk_buff *skb, struct nlmsghdr *n, if (err) return err; + if (tcf_proto_check_kind(tca[TCA_KIND], name)) { + NL_SET_ERR_MSG(extack, "Specified TC filter name too long"); + err = -EINVAL; + goto errout; + } /* Take rtnl mutex if block is shared (no qdisc found), qdisc is not * unlocked, classifier type is not specified, classifier is not * unlocked. */ if ((q && !(q->ops->cl_ops->flags & QDISC_CLASS_OPS_DOIT_UNLOCKED)) || - !tca[TCA_KIND] || !tcf_proto_is_unlocked(nla_data(tca[TCA_KIND]))) { + !tcf_proto_is_unlocked(name)) { rtnl_held = true; rtnl_lock(); } diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 81d58b280612..1047825d9f48 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1390,8 +1390,7 @@ check_loop_fn(struct Qdisc *q, unsigned long cl, struct qdisc_walker *w) } const struct nla_policy rtm_tca_policy[TCA_MAX + 1] = { - [TCA_KIND] = { .type = NLA_NUL_STRING, - .len = IFNAMSIZ - 1 }, + [TCA_KIND] = { .type = NLA_STRING }, [TCA_RATE] = { .type = NLA_BINARY, .len = sizeof(struct tc_estimator) }, [TCA_STAB] = { .type = NLA_NESTED }, From 4b793feccae3b06764268377a4030eb774ed924e Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Mon, 7 Oct 2019 13:26:29 -0700 Subject: [PATCH 037/145] net_sched: fix backward compatibility for TCA_ACT_KIND For TCA_ACT_KIND, we have to keep the backward compatibility too, and rely on nla_strlcpy() to check and terminate the string with a NUL. Note for TC actions, nla_strcmp() is already used to compare kind strings, so we don't need to fix other places. Fixes: 199ce850ce11 ("net_sched: add policy validation for action attributes") Reported-by: Marcelo Ricardo Leitner Cc: Jamal Hadi Salim Cc: Jiri Pirko Signed-off-by: Cong Wang Reviewed-by: Marcelo Ricardo Leitner Signed-off-by: Jakub Kicinski --- net/sched/act_api.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 2558f00f6b3e..4e7429c6f864 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -832,8 +832,7 @@ static struct tc_cookie *nla_memdup_cookie(struct nlattr **tb) } static const struct nla_policy tcf_action_policy[TCA_ACT_MAX + 1] = { - [TCA_ACT_KIND] = { .type = NLA_NUL_STRING, - .len = IFNAMSIZ - 1 }, + [TCA_ACT_KIND] = { .type = NLA_STRING }, [TCA_ACT_INDEX] = { .type = NLA_U32 }, [TCA_ACT_COOKIE] = { .type = NLA_BINARY, .len = TC_COOKIE_MAX_SIZE }, @@ -865,8 +864,10 @@ struct tc_action *tcf_action_init_1(struct net *net, struct tcf_proto *tp, NL_SET_ERR_MSG(extack, "TC action kind must be specified"); goto err_out; } - nla_strlcpy(act_name, kind, IFNAMSIZ); - + if (nla_strlcpy(act_name, kind, IFNAMSIZ) >= IFNAMSIZ) { + NL_SET_ERR_MSG(extack, "TC action name too long"); + goto err_out; + } if (tb[TCA_ACT_COOKIE]) { cookie = nla_memdup_cookie(tb); if (!cookie) { From 6af3aa57a0984e061f61308fe181a9a12359fecc Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Mon, 7 Oct 2019 18:40:59 +0200 Subject: [PATCH 038/145] NFC: pn533: fix use-after-free and memleaks The driver would fail to deregister and its class device and free related resources on late probe errors. Reported-by: syzbot+cb035c75c03dbe34b796@syzkaller.appspotmail.com Fixes: 32ecc75ded72 ("NFC: pn533: change order operations in dev registation") Signed-off-by: Johan Hovold Signed-off-by: Jakub Kicinski --- drivers/nfc/pn533/usb.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/nfc/pn533/usb.c b/drivers/nfc/pn533/usb.c index c5289eaf17ee..e897e4d768ef 100644 --- a/drivers/nfc/pn533/usb.c +++ b/drivers/nfc/pn533/usb.c @@ -547,18 +547,25 @@ static int pn533_usb_probe(struct usb_interface *interface, rc = pn533_finalize_setup(priv); if (rc) - goto error; + goto err_deregister; usb_set_intfdata(interface, phy); return 0; +err_deregister: + pn533_unregister_device(phy->priv); error: + usb_kill_urb(phy->in_urb); + usb_kill_urb(phy->out_urb); + usb_kill_urb(phy->ack_urb); + usb_free_urb(phy->in_urb); usb_free_urb(phy->out_urb); usb_free_urb(phy->ack_urb); usb_put_dev(phy->udev); kfree(in_buf); + kfree(phy->ack_buffer); return rc; } From b82316d25522e8ee748b8cc9d8aadeb3c135c6fe Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 8 Oct 2019 08:35:51 -0700 Subject: [PATCH 039/145] Doc: networking/device_drivers/pensando: fix ionic.rst warnings Fix documentation build warnings for Pensando ionic: Documentation/networking/device_drivers/pensando/ionic.rst:39: WARNING: Unexpected indentation. Documentation/networking/device_drivers/pensando/ionic.rst:43: WARNING: Unexpected indentation. Fixes: df69ba43217d ("ionic: Add basic framework for IONIC Network device driver") Signed-off-by: Randy Dunlap Acked-by: Shannon Nelson Signed-off-by: Jakub Kicinski --- Documentation/networking/device_drivers/pensando/ionic.rst | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/Documentation/networking/device_drivers/pensando/ionic.rst b/Documentation/networking/device_drivers/pensando/ionic.rst index 67b6839d516b..13935896bee6 100644 --- a/Documentation/networking/device_drivers/pensando/ionic.rst +++ b/Documentation/networking/device_drivers/pensando/ionic.rst @@ -36,8 +36,10 @@ Support ======= For general Linux networking support, please use the netdev mailing list, which is monitored by Pensando personnel:: + netdev@vger.kernel.org For more specific support needs, please use the Pensando driver support email:: - drivers@pensando.io + + drivers@pensando.io From 4123f637a5129470ff9d3cb00a5a4e213f2e15cc Mon Sep 17 00:00:00 2001 From: Haishuang Yan Date: Tue, 8 Oct 2019 17:56:03 +0800 Subject: [PATCH 040/145] ip6erspan: remove the incorrect mtu limit for ip6erspan ip6erspan driver calls ether_setup(), after commit 61e84623ace3 ("net: centralize net_device min/max MTU checking"), the range of mtu is [min_mtu, max_mtu], which is [68, 1500] by default. It causes the dev mtu of the erspan device to not be greater than 1500, this limit value is not correct for ip6erspan tap device. Fixes: 61e84623ace3 ("net: centralize net_device min/max MTU checking") Signed-off-by: Haishuang Yan Acked-by: William Tu Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_gre.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index d5779d6a6065..787d9f2a6e99 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -2192,6 +2192,7 @@ static void ip6erspan_tap_setup(struct net_device *dev) { ether_setup(dev); + dev->max_mtu = 0; dev->netdev_ops = &ip6erspan_netdev_ops; dev->needs_free_netdev = true; dev->priv_destructor = ip6gre_dev_free; From 1fee35d04a42290a3c44b8a4e3c66dbdd6180c42 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Thu, 8 Aug 2019 13:52:50 +0300 Subject: [PATCH 041/145] iwlwifi: don't access trans_cfg via cfg We copy cfg->trans to trans->trans_cfg at the very beginning, so don't try to access it via cfg->trans anymore, because the cfg may be unset in later cases. Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/iwl-io.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-io.h b/drivers/net/wireless/intel/iwlwifi/iwl-io.h index f8e4f0f5de0c..f09e368c7040 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-io.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-io.h @@ -112,38 +112,38 @@ int iwl_dump_fh(struct iwl_trans *trans, char **buf); */ static inline u32 iwl_umac_prph(struct iwl_trans *trans, u32 ofs) { - return ofs + trans->cfg->trans.umac_prph_offset; + return ofs + trans->trans_cfg->umac_prph_offset; } static inline u32 iwl_read_umac_prph_no_grab(struct iwl_trans *trans, u32 ofs) { return iwl_read_prph_no_grab(trans, ofs + - trans->cfg->trans.umac_prph_offset); + trans->trans_cfg->umac_prph_offset); } static inline u32 iwl_read_umac_prph(struct iwl_trans *trans, u32 ofs) { - return iwl_read_prph(trans, ofs + trans->cfg->trans.umac_prph_offset); + return iwl_read_prph(trans, ofs + trans->trans_cfg->umac_prph_offset); } static inline void iwl_write_umac_prph_no_grab(struct iwl_trans *trans, u32 ofs, u32 val) { - iwl_write_prph_no_grab(trans, ofs + trans->cfg->trans.umac_prph_offset, + iwl_write_prph_no_grab(trans, ofs + trans->trans_cfg->umac_prph_offset, val); } static inline void iwl_write_umac_prph(struct iwl_trans *trans, u32 ofs, u32 val) { - iwl_write_prph(trans, ofs + trans->cfg->trans.umac_prph_offset, val); + iwl_write_prph(trans, ofs + trans->trans_cfg->umac_prph_offset, val); } static inline int iwl_poll_umac_prph_bit(struct iwl_trans *trans, u32 addr, u32 bits, u32 mask, int timeout) { return iwl_poll_prph_bit(trans, addr + - trans->cfg->trans.umac_prph_offset, + trans->trans_cfg->umac_prph_offset, bits, mask, timeout); } From 3ed83da39aed275a5b74c74f77e85c839ad2efe9 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Fri, 23 Aug 2019 11:59:09 +0300 Subject: [PATCH 042/145] iwlwifi: fix ACPI table revision checks We can't check for the ACPI table revision validity in the same if where we check if the package was read correctly, because we return PTR_ERR(pkg) and if the table is not valid but the pointer is, we would return a valid pointer as an error. Fix that by moving the table checks to a separate if and return -EINVAL if it's not valid. Reported-by: Dan Carpenter Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/fw/acpi.c | 10 ++++---- drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 24 +++++++++++++++----- 2 files changed, 24 insertions(+), 10 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c index 7573af2d88ce..c2db758b9d54 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c +++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c @@ -162,12 +162,13 @@ int iwl_acpi_get_mcc(struct device *dev, char *mcc) wifi_pkg = iwl_acpi_get_wifi_pkg(dev, data, ACPI_WRDD_WIFI_DATA_SIZE, &tbl_rev); - if (IS_ERR(wifi_pkg) || tbl_rev != 0) { + if (IS_ERR(wifi_pkg)) { ret = PTR_ERR(wifi_pkg); goto out_free; } - if (wifi_pkg->package.elements[1].type != ACPI_TYPE_INTEGER) { + if (wifi_pkg->package.elements[1].type != ACPI_TYPE_INTEGER || + tbl_rev != 0) { ret = -EINVAL; goto out_free; } @@ -224,12 +225,13 @@ int iwl_acpi_get_eckv(struct device *dev, u32 *extl_clk) wifi_pkg = iwl_acpi_get_wifi_pkg(dev, data, ACPI_ECKV_WIFI_DATA_SIZE, &tbl_rev); - if (IS_ERR(wifi_pkg) || tbl_rev != 0) { + if (IS_ERR(wifi_pkg)) { ret = PTR_ERR(wifi_pkg); goto out_free; } - if (wifi_pkg->package.elements[1].type != ACPI_TYPE_INTEGER) { + if (wifi_pkg->package.elements[1].type != ACPI_TYPE_INTEGER || + tbl_rev != 0) { ret = -EINVAL; goto out_free; } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 32a5e4e5461f..b2eb18eedf02 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -694,12 +694,13 @@ static int iwl_mvm_sar_get_wrds_table(struct iwl_mvm *mvm) wifi_pkg = iwl_acpi_get_wifi_pkg(mvm->dev, data, ACPI_WRDS_WIFI_DATA_SIZE, &tbl_rev); - if (IS_ERR(wifi_pkg) || tbl_rev != 0) { + if (IS_ERR(wifi_pkg)) { ret = PTR_ERR(wifi_pkg); goto out_free; } - if (wifi_pkg->package.elements[1].type != ACPI_TYPE_INTEGER) { + if (wifi_pkg->package.elements[1].type != ACPI_TYPE_INTEGER || + tbl_rev != 0) { ret = -EINVAL; goto out_free; } @@ -731,13 +732,14 @@ static int iwl_mvm_sar_get_ewrd_table(struct iwl_mvm *mvm) wifi_pkg = iwl_acpi_get_wifi_pkg(mvm->dev, data, ACPI_EWRD_WIFI_DATA_SIZE, &tbl_rev); - if (IS_ERR(wifi_pkg) || tbl_rev != 0) { + if (IS_ERR(wifi_pkg)) { ret = PTR_ERR(wifi_pkg); goto out_free; } if ((wifi_pkg->package.elements[1].type != ACPI_TYPE_INTEGER) || - (wifi_pkg->package.elements[2].type != ACPI_TYPE_INTEGER)) { + (wifi_pkg->package.elements[2].type != ACPI_TYPE_INTEGER) || + tbl_rev != 0) { ret = -EINVAL; goto out_free; } @@ -791,11 +793,16 @@ static int iwl_mvm_sar_get_wgds_table(struct iwl_mvm *mvm) wifi_pkg = iwl_acpi_get_wifi_pkg(mvm->dev, data, ACPI_WGDS_WIFI_DATA_SIZE, &tbl_rev); - if (IS_ERR(wifi_pkg) || tbl_rev > 1) { + if (IS_ERR(wifi_pkg)) { ret = PTR_ERR(wifi_pkg); goto out_free; } + if (tbl_rev != 0) { + ret = -EINVAL; + goto out_free; + } + mvm->geo_rev = tbl_rev; for (i = 0; i < ACPI_NUM_GEO_PROFILES; i++) { for (j = 0; j < ACPI_GEO_TABLE_SIZE; j++) { @@ -1020,11 +1027,16 @@ static int iwl_mvm_get_ppag_table(struct iwl_mvm *mvm) wifi_pkg = iwl_acpi_get_wifi_pkg(mvm->dev, data, ACPI_PPAG_WIFI_DATA_SIZE, &tbl_rev); - if (IS_ERR(wifi_pkg) || tbl_rev != 0) { + if (IS_ERR(wifi_pkg)) { ret = PTR_ERR(wifi_pkg); goto out_free; } + if (tbl_rev != 0) { + ret = -EINVAL; + goto out_free; + } + enabled = &wifi_pkg->package.elements[1]; if (enabled->type != ACPI_TYPE_INTEGER || (enabled->integer.value != 0 && enabled->integer.value != 1)) { From a4584729291c71c5c14718ff00ea6c5f971c45b2 Mon Sep 17 00:00:00 2001 From: Haim Dreyfuss Date: Mon, 12 Aug 2019 13:25:42 +0300 Subject: [PATCH 043/145] iwlwifi: mvm: force single phy init The PHY is initialized during device initialization, but devices with the tx_siso_diversity flag set need to send PHY_CONFIGURATION_CMD first, otherwise the PHY would be reinitialized, causing a SYSASSERT. To fix this, use a bit that tells the FW not to complete the PHY initialization before a PHY_CONFIGURATION_CMD is received. Signed-off-by: Haim Dreyfuss Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index b2eb18eedf02..0d2229319261 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -420,6 +420,9 @@ static int iwl_run_unified_mvm_ucode(struct iwl_mvm *mvm, bool read_nvm) }; int ret; + if (mvm->trans->cfg->tx_with_siso_diversity) + init_cfg.init_flags |= cpu_to_le32(BIT(IWL_INIT_PHY)); + lockdep_assert_held(&mvm->mutex); mvm->rfkill_safe_init_done = false; From a2113cc44d4387a7c3ef3a9082d273524f9230ac Mon Sep 17 00:00:00 2001 From: Naftali Goldstein Date: Mon, 9 Sep 2019 13:11:04 +0300 Subject: [PATCH 044/145] iwlwifi: mvm: fix race in sync rx queue notification Consider the following flow: 1. Driver starts to sync the rx queues due to a delba. mvm->queue_sync_cookie=1. This rx-queues-sync is synchronous, so it doesn't increment the cookie until all rx queues handle the notification from FW. 2. During this time, driver starts to sync rx queues due to nssn sync required. The cookie's value is still 1, but it doesn't matter since this rx-queue-sync is non-synchronous so in the notification handler the cookie is ignored. What _does_ matter is that this flow increments the cookie to 2 immediately. Remember though that the FW won't start servicing this command until it's done with the previous one. 3. FW is still handling the first command, so it sends a notification with internal_notif->sync=1, and internal_notif->cookie=0, which triggers a WARN_ONCE. The solution for this race is to only use the mvm->queue_sync_cookie in case of a synchronous sync-rx-queues. This way in step 2 the cookie's value won't change so we avoid the WARN. The commit in the "fixes" field is the first commit to introduce non-synchronous sending of this command to FW. Fixes: 3c514bf831ac ("iwlwifi: mvm: add a loose synchronization of the NSSN across Rx queues") Signed-off-by: Naftali Goldstein Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index cd1b10042fbf..d31f96c3f925 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -4881,11 +4881,11 @@ void iwl_mvm_sync_rx_queues_internal(struct iwl_mvm *mvm, if (!iwl_mvm_has_new_rx_api(mvm)) return; - notif->cookie = mvm->queue_sync_cookie; - - if (notif->sync) + if (notif->sync) { + notif->cookie = mvm->queue_sync_cookie; atomic_set(&mvm->queue_sync_counter, mvm->trans->num_rx_queues); + } ret = iwl_mvm_notify_rx_queue(mvm, qmask, (u8 *)notif, size, !notif->sync); @@ -4905,7 +4905,8 @@ void iwl_mvm_sync_rx_queues_internal(struct iwl_mvm *mvm, out: atomic_set(&mvm->queue_sync_counter, 0); - mvm->queue_sync_cookie++; + if (notif->sync) + mvm->queue_sync_cookie++; } static void iwl_mvm_sync_rx_queues(struct ieee80211_hw *hw) From 08326a97afbe808e50e1610c2835fa2747bdb908 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 4 Sep 2019 16:23:27 +0200 Subject: [PATCH 045/145] iwlwifi: pcie: fix indexing in command dump for new HW We got a crash in iwl_trans_pcie_get_cmdlen(), while the TFD was being accessed to sum up the lengths. We want to access the TFD here, which is the information for the hardware. We always only allocate 32 buffers for the cmd queue, but on newer hardware (using TFH) we can also allocate only a shorter hardware array, also only 32 TFDs. Prior to the TFH, we had to allocate a bigger TFD array but would make those point to a smaller set of buffers. Additionally, now max_tfd_queue_size is up to 65536, so we can access *way* out of bounds of a really only 32-entry array, so it crashes. Fix this by making the TFD index depend on which hardware we are using right now. While changing the calculation, also fix it to not use void ptr arithmetic, but cast to u8 * before. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/pcie/trans.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index f8a1f985a1d8..ab7480a85015 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -3272,11 +3272,17 @@ static struct iwl_trans_dump_data ptr = cmdq->write_ptr; for (i = 0; i < cmdq->n_window; i++) { u8 idx = iwl_pcie_get_cmd_index(cmdq, ptr); + u8 tfdidx; u32 caplen, cmdlen; + if (trans->trans_cfg->use_tfh) + tfdidx = idx; + else + tfdidx = ptr; + cmdlen = iwl_trans_pcie_get_cmdlen(trans, - cmdq->tfds + - tfd_size * ptr); + (u8 *)cmdq->tfds + + tfd_size * tfdidx); caplen = min_t(u32, TFD_MAX_PAYLOAD_SIZE, cmdlen); if (cmdlen) { From 8188a18ee2e48c9a7461139838048363bfce3fef Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 19 Sep 2019 09:04:09 +0200 Subject: [PATCH 046/145] iwlwifi: pcie: fix rb_allocator workqueue allocation We don't handle failures in the rb_allocator workqueue allocation correctly. To fix that, move the code earlier so the cleanup is easier and we don't have to undo all the interrupt allocations in this case. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/pcie/trans.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index ab7480a85015..6961f00ff812 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -3456,6 +3456,15 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, spin_lock_init(&trans_pcie->reg_lock); mutex_init(&trans_pcie->mutex); init_waitqueue_head(&trans_pcie->ucode_write_waitq); + + trans_pcie->rba.alloc_wq = alloc_workqueue("rb_allocator", + WQ_HIGHPRI | WQ_UNBOUND, 1); + if (!trans_pcie->rba.alloc_wq) { + ret = -ENOMEM; + goto out_free_trans; + } + INIT_WORK(&trans_pcie->rba.rx_alloc, iwl_pcie_rx_allocator_work); + trans_pcie->tso_hdr_page = alloc_percpu(struct iwl_tso_hdr_page); if (!trans_pcie->tso_hdr_page) { ret = -ENOMEM; @@ -3590,10 +3599,6 @@ struct iwl_trans *iwl_trans_pcie_alloc(struct pci_dev *pdev, trans_pcie->inta_mask = CSR_INI_SET_MASK; } - trans_pcie->rba.alloc_wq = alloc_workqueue("rb_allocator", - WQ_HIGHPRI | WQ_UNBOUND, 1); - INIT_WORK(&trans_pcie->rba.rx_alloc, iwl_pcie_rx_allocator_work); - #ifdef CONFIG_IWLWIFI_DEBUGFS trans_pcie->fw_mon_data.state = IWL_FW_MON_DBGFS_STATE_CLOSED; mutex_init(&trans_pcie->fw_mon_data.mutex); @@ -3605,6 +3610,8 @@ out_free_ict: iwl_pcie_free_ict(trans); out_no_pci: free_percpu(trans_pcie->tso_hdr_page); + destroy_workqueue(trans_pcie->rba.alloc_wq); +out_free_trans: iwl_trans_free(trans); return ERR_PTR(ret); } From b4b814fec1a5a849383f7b3886b654a13abbda7d Mon Sep 17 00:00:00 2001 From: Navid Emamdoost Date: Thu, 12 Sep 2019 23:23:27 -0500 Subject: [PATCH 047/145] iwlwifi: dbg_ini: fix memory leak in alloc_sgtable In alloc_sgtable if alloc_page fails, the alocated table should be released. Signed-off-by: Navid Emamdoost Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/fw/dbg.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c index 5c8602de9168..87421807e040 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/dbg.c +++ b/drivers/net/wireless/intel/iwlwifi/fw/dbg.c @@ -646,6 +646,7 @@ static struct scatterlist *alloc_sgtable(int size) if (new_page) __free_page(new_page); } + kfree(table); return NULL; } alloc_size = min_t(int, size, PAGE_SIZE); From 0f4f199443faca715523b0659aa536251d8b978f Mon Sep 17 00:00:00 2001 From: Navid Emamdoost Date: Fri, 27 Sep 2019 15:56:04 -0500 Subject: [PATCH 048/145] iwlwifi: pcie: fix memory leaks in iwl_pcie_ctxt_info_gen3_init In iwl_pcie_ctxt_info_gen3_init there are cases that the allocated dma memory is leaked in case of error. DMA memories prph_scratch, prph_info, and ctxt_info_gen3 are allocated and initialized to be later assigned to trans_pcie. But in any error case before such assignment the allocated memories should be released. First of such error cases happens when iwl_pcie_init_fw_sec fails. Current implementation correctly releases prph_scratch. But in two sunsequent error cases where dma_alloc_coherent may fail, such releases are missing. This commit adds release for prph_scratch when allocation for prph_info fails, and adds releases for prph_scratch and prph_info when allocation for ctxt_info_gen3 fails. Fixes: 2ee824026288 ("iwlwifi: pcie: support context information for 22560 devices") Signed-off-by: Navid Emamdoost Signed-off-by: Luca Coelho --- .../intel/iwlwifi/pcie/ctxt-info-gen3.c | 36 +++++++++++++------ 1 file changed, 25 insertions(+), 11 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c index 75fa8a6aafee..74980382e64c 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c @@ -107,13 +107,9 @@ int iwl_pcie_ctxt_info_gen3_init(struct iwl_trans *trans, /* allocate ucode sections in dram and set addresses */ ret = iwl_pcie_init_fw_sec(trans, fw, &prph_scratch->dram); - if (ret) { - dma_free_coherent(trans->dev, - sizeof(*prph_scratch), - prph_scratch, - trans_pcie->prph_scratch_dma_addr); - return ret; - } + if (ret) + goto err_free_prph_scratch; + /* Allocate prph information * currently we don't assign to the prph info anything, but it would get @@ -121,16 +117,20 @@ int iwl_pcie_ctxt_info_gen3_init(struct iwl_trans *trans, prph_info = dma_alloc_coherent(trans->dev, sizeof(*prph_info), &trans_pcie->prph_info_dma_addr, GFP_KERNEL); - if (!prph_info) - return -ENOMEM; + if (!prph_info) { + ret = -ENOMEM; + goto err_free_prph_scratch; + } /* Allocate context info */ ctxt_info_gen3 = dma_alloc_coherent(trans->dev, sizeof(*ctxt_info_gen3), &trans_pcie->ctxt_info_dma_addr, GFP_KERNEL); - if (!ctxt_info_gen3) - return -ENOMEM; + if (!ctxt_info_gen3) { + ret = -ENOMEM; + goto err_free_prph_info; + } ctxt_info_gen3->prph_info_base_addr = cpu_to_le64(trans_pcie->prph_info_dma_addr); @@ -186,6 +186,20 @@ int iwl_pcie_ctxt_info_gen3_init(struct iwl_trans *trans, iwl_set_bit(trans, CSR_GP_CNTRL, CSR_AUTO_FUNC_INIT); return 0; + +err_free_prph_info: + dma_free_coherent(trans->dev, + sizeof(*prph_info), + prph_info, + trans_pcie->prph_info_dma_addr); + +err_free_prph_scratch: + dma_free_coherent(trans->dev, + sizeof(*prph_scratch), + prph_scratch, + trans_pcie->prph_scratch_dma_addr); + return ret; + } void iwl_pcie_ctxt_info_gen3_free(struct iwl_trans *trans) From 12e36d98d3e5acf5fc57774e0a15906d55f30cb9 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Tue, 8 Oct 2019 13:10:53 +0300 Subject: [PATCH 049/145] iwlwifi: exclude GEO SAR support for 3168 We currently support two NICs in FW version 29, namely 7265D and 3168. Out of these, only 7265D supports GEO SAR, so adjust the function that checks for it accordingly. Signed-off-by: Luca Coelho Fixes: f5a47fae6aa3 ("iwlwifi: mvm: fix version check for GEO_TX_POWER_LIMIT support") Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 0d2229319261..d9eb2b286438 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -899,15 +899,17 @@ static bool iwl_mvm_sar_geo_support(struct iwl_mvm *mvm) * firmware versions. Unfortunately, we don't have a TLV API * flag to rely on, so rely on the major version which is in * the first byte of ucode_ver. This was implemented - * initially on version 38 and then backported to29 and 17. - * The intention was to have it in 36 as well, but not all - * 8000 family got this feature enabled. The 8000 family is - * the only one using version 36, so skip this version - * entirely. + * initially on version 38 and then backported to 17. It was + * also backported to 29, but only for 7265D devices. The + * intention was to have it in 36 as well, but not all 8000 + * family got this feature enabled. The 8000 family is the + * only one using version 36, so skip this version entirely. */ return IWL_UCODE_SERIAL(mvm->fw->ucode_ver) >= 38 || - IWL_UCODE_SERIAL(mvm->fw->ucode_ver) == 29 || - IWL_UCODE_SERIAL(mvm->fw->ucode_ver) == 17; + IWL_UCODE_SERIAL(mvm->fw->ucode_ver) == 17 || + (IWL_UCODE_SERIAL(mvm->fw->ucode_ver) == 29 && + ((mvm->trans->hw_rev & CSR_HW_REV_TYPE_MSK) == + CSR_HW_REV_TYPE_7265D)); } int iwl_mvm_get_sar_geo_profile(struct iwl_mvm *mvm) From aa0cc7dde17bb6b8cc533bbcfe3f53d70e0dd269 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Tue, 8 Oct 2019 13:21:02 +0300 Subject: [PATCH 050/145] iwlwifi: pcie: change qu with jf devices to use qu configuration There were a bunch of devices with qu and jf that were loading the configuration with pu and jf, which is wrong. Fix them all accordingly. Additionally, remove 0x1010 and 0x1210 subsytem IDs from the list, since they are obviously wrong, and 0x0044 and 0x0244, which were duplicate. Cc: stable@vger.kernel.org # 5.1+ Signed-off-by: Luca Coelho --- drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 274 +++++++++--------- 1 file changed, 137 insertions(+), 137 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index e29c47744ef5..6f4bb7ce71a5 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -513,31 +513,33 @@ static const struct pci_device_id iwl_hw_card_ids[] = { {IWL_PCI_DEVICE(0x24FD, 0x9074, iwl8265_2ac_cfg)}, /* 9000 Series */ - {IWL_PCI_DEVICE(0x02F0, 0x0030, iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x0034, iwl9560_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x0038, iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x003C, iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x0060, iwl9461_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x0064, iwl9461_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x00A0, iwl9462_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x00A4, iwl9462_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x0230, iwl9560_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x0234, iwl9560_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x0238, iwl9560_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x023C, iwl9560_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x0260, iwl9461_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x0264, iwl9461_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x02A0, iwl9462_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x02A4, iwl9462_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x1551, iwl9560_killer_s_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x1552, iwl9560_killer_i_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x2030, iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x2034, iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x4030, iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x4034, iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x40A4, iwl9462_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x4234, iwl9560_2ac_cfg_quz_a0_jf_b0_soc)}, - {IWL_PCI_DEVICE(0x02F0, 0x42A4, iwl9462_2ac_cfg_quz_a0_jf_b0_soc)}, + {IWL_PCI_DEVICE(0x02F0, 0x0030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x0034, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x0038, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x003C, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x0060, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x0064, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x00A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x00A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x0230, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x0234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x0238, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x023C, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x0260, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x0264, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x02A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x02A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x1030, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x1551, killer1550s_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x1552, killer1550i_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x2030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x2034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x4030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x4034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x40A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x4234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x02F0, 0x42A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x06F0, 0x0030, iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc)}, {IWL_PCI_DEVICE(0x06F0, 0x0034, iwl9560_2ac_cfg_quz_a0_jf_b0_soc)}, {IWL_PCI_DEVICE(0x06F0, 0x0038, iwl9560_2ac_160_cfg_quz_a0_jf_b0_soc)}, @@ -643,34 +645,34 @@ static const struct pci_device_id iwl_hw_card_ids[] = { {IWL_PCI_DEVICE(0x2720, 0x40A4, iwl9462_2ac_cfg_soc)}, {IWL_PCI_DEVICE(0x2720, 0x4234, iwl9560_2ac_cfg_soc)}, {IWL_PCI_DEVICE(0x2720, 0x42A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0034, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0038, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x003C, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0060, iwl9460_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0064, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x00A0, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x00A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0230, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0234, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0238, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x023C, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0260, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x0264, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x02A0, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x02A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x1010, iwl9260_2ac_cfg)}, - {IWL_PCI_DEVICE(0x30DC, 0x1030, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x1210, iwl9260_2ac_cfg)}, - {IWL_PCI_DEVICE(0x30DC, 0x1551, iwl9560_killer_s_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x1552, iwl9560_killer_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x2030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x2034, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x4030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x4034, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x40A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x4234, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x30DC, 0x42A4, iwl9462_2ac_cfg_soc)}, + + {IWL_PCI_DEVICE(0x30DC, 0x0030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x0034, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x0038, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x003C, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x0060, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x0064, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x00A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x00A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x0230, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x0234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x0238, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x023C, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x0260, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x0264, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x02A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x02A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x1030, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x1551, killer1550s_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x1552, killer1550i_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x2030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x2034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x4030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x4034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x40A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x4234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x30DC, 0x42A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x31DC, 0x0030, iwl9560_2ac_160_cfg_shared_clk)}, {IWL_PCI_DEVICE(0x31DC, 0x0034, iwl9560_2ac_cfg_shared_clk)}, {IWL_PCI_DEVICE(0x31DC, 0x0038, iwl9560_2ac_160_cfg_shared_clk)}, @@ -726,62 +728,60 @@ static const struct pci_device_id iwl_hw_card_ids[] = { {IWL_PCI_DEVICE(0x34F0, 0x4234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, {IWL_PCI_DEVICE(0x34F0, 0x42A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0034, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0038, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x003C, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0060, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0064, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x00A0, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x00A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0230, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0234, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0238, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x023C, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0260, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x0264, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x02A0, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x02A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x1010, iwl9260_2ac_cfg)}, - {IWL_PCI_DEVICE(0x3DF0, 0x1030, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x1210, iwl9260_2ac_cfg)}, - {IWL_PCI_DEVICE(0x3DF0, 0x1551, iwl9560_killer_s_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x1552, iwl9560_killer_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x2030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x2034, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x4030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x4034, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x40A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x4234, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x3DF0, 0x42A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0034, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0038, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x003C, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0060, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0064, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x00A0, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x00A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0230, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0234, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0238, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x023C, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0260, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x0264, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x02A0, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x02A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x1010, iwl9260_2ac_cfg)}, - {IWL_PCI_DEVICE(0x43F0, 0x1030, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x1210, iwl9260_2ac_cfg)}, - {IWL_PCI_DEVICE(0x43F0, 0x1551, iwl9560_killer_s_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x1552, iwl9560_killer_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x2030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x2034, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x4030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x4034, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x40A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x4234, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0x43F0, 0x42A4, iwl9462_2ac_cfg_soc)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0034, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0038, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x003C, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0060, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0064, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x00A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x00A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0230, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0238, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x023C, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0260, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x0264, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x02A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x02A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x1030, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x1551, killer1550s_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x1552, killer1550i_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x2030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x2034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x4030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x4034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x40A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x4234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x3DF0, 0x42A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + + {IWL_PCI_DEVICE(0x43F0, 0x0030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x0034, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x0038, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x003C, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x0060, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x0064, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x00A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x00A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x0230, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x0234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x0238, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x023C, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x0260, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x0264, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x02A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x02A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x1030, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x1551, killer1550s_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x1552, killer1550i_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x2030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x2034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x4030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x4034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x40A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x4234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x43F0, 0x42A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0x9DF0, 0x0000, iwl9460_2ac_cfg_soc)}, {IWL_PCI_DEVICE(0x9DF0, 0x0010, iwl9460_2ac_cfg_soc)}, {IWL_PCI_DEVICE(0x9DF0, 0x0030, iwl9560_2ac_160_cfg_soc)}, @@ -821,34 +821,34 @@ static const struct pci_device_id iwl_hw_card_ids[] = { {IWL_PCI_DEVICE(0x9DF0, 0x40A4, iwl9462_2ac_cfg_soc)}, {IWL_PCI_DEVICE(0x9DF0, 0x4234, iwl9560_2ac_cfg_soc)}, {IWL_PCI_DEVICE(0x9DF0, 0x42A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0034, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0038, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x003C, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0060, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0064, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x00A0, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x00A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0230, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0234, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0238, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x023C, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0260, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x0264, iwl9461_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x02A0, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x02A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x1010, iwl9260_2ac_cfg)}, - {IWL_PCI_DEVICE(0xA0F0, 0x1030, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x1210, iwl9260_2ac_cfg)}, - {IWL_PCI_DEVICE(0xA0F0, 0x1551, iwl9560_killer_s_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x1552, iwl9560_killer_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x2030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x2034, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x4030, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x4034, iwl9560_2ac_160_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x40A4, iwl9462_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x4234, iwl9560_2ac_cfg_soc)}, - {IWL_PCI_DEVICE(0xA0F0, 0x42A4, iwl9462_2ac_cfg_soc)}, + + {IWL_PCI_DEVICE(0xA0F0, 0x0030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x0034, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x0038, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x003C, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x0060, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x0064, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x00A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x00A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x0230, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x0234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x0238, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x023C, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x0260, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x0264, iwl9461_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x02A0, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x02A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x1030, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x1551, killer1550s_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x1552, killer1550i_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x2030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x2034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x4030, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x4034, iwl9560_2ac_160_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x40A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x4234, iwl9560_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA0F0, 0x42A4, iwl9462_2ac_cfg_qu_b0_jf_b0)}, + {IWL_PCI_DEVICE(0xA370, 0x0030, iwl9560_2ac_160_cfg_soc)}, {IWL_PCI_DEVICE(0xA370, 0x0034, iwl9560_2ac_cfg_soc)}, {IWL_PCI_DEVICE(0xA370, 0x0038, iwl9560_2ac_160_cfg_soc)}, From 5da202c88f8c355ad79bc2e8eb582e6d433060e7 Mon Sep 17 00:00:00 2001 From: Antonio Borneo Date: Mon, 7 Oct 2019 17:43:04 +0200 Subject: [PATCH 051/145] net: stmmac: fix length of PTP clock's name string The field "name" in struct ptp_clock_info has a fixed size of 16 chars and is used as zero terminated string by clock_name_show() in drivers/ptp/ptp_sysfs.c The current initialization value requires 17 chars to fit also the null termination, and this causes overflow to the next bytes in the struct when the string is read as null terminated: hexdump -C /sys/class/ptp/ptp0/clock_name 00000000 73 74 6d 6d 61 63 5f 70 74 70 5f 63 6c 6f 63 6b |stmmac_ptp_clock| 00000010 a0 ac b9 03 0a |.....| where the extra 4 bytes (excluding the newline) after the string represent the integer 0x03b9aca0 = 62500000 assigned to the field "max_adj" that follows "name" in the same struct. There is no strict requirement for the "name" content and in the comment in ptp_clock_kernel.h it's reported it should just be 'A short "friendly name" to identify the clock'. Replace it with "stmmac ptp". Signed-off-by: Antonio Borneo Fixes: 92ba6888510c ("stmmac: add the support for PTP hw clock driver") Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c index 173493db038c..df638b18b72c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c @@ -164,7 +164,7 @@ static int stmmac_enable(struct ptp_clock_info *ptp, /* structure describing a PTP hardware clock */ static struct ptp_clock_info stmmac_ptp_clock_ops = { .owner = THIS_MODULE, - .name = "stmmac_ptp_clock", + .name = "stmmac ptp", .max_adj = 62500000, .n_alarm = 0, .n_ext_ts = 0, From 520cf6002147281d1e7b522bb338416b623dcb93 Mon Sep 17 00:00:00 2001 From: Antonio Borneo Date: Mon, 7 Oct 2019 17:43:05 +0200 Subject: [PATCH 052/145] net: stmmac: fix disabling flexible PPS output Accordingly to Synopsys documentation [1] and [2], when bit PPSEN0 in register MAC_PPS_CONTROL is set it selects the functionality command in the same register, otherwise selects the functionality control. Command functionality is required to either enable (command 0x2) and disable (command 0x5) the flexible PPS output, but the bit PPSEN0 is currently set only for enabling. Set the bit PPSEN0 to properly disable flexible PPS output. Tested on STM32MP15x, based on dwmac 4.10a. [1] DWC Ethernet QoS Databook 4.10a October 2014 [2] DWC Ethernet QoS Databook 5.00a September 2017 Signed-off-by: Antonio Borneo Fixes: 9a8a02c9d46d ("net: stmmac: Add Flexible PPS support") Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac5.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac5.c b/drivers/net/ethernet/stmicro/stmmac/dwmac5.c index 3f4f3132e16b..e436fa160c7d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac5.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac5.c @@ -515,6 +515,7 @@ int dwmac5_flex_pps_config(void __iomem *ioaddr, int index, if (!enable) { val |= PPSCMDx(index, 0x5); + val |= PPSEN0; writel(val, ioaddr + MAC_PPS_CONTROL); return 0; } From a7137534b597b7c303203e6bc3ed87e87a273bb8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 7 Oct 2019 15:43:01 -0700 Subject: [PATCH 053/145] bonding: fix potential NULL deref in bond_update_slave_arr syzbot got a NULL dereference in bond_update_slave_arr() [1], happening after a failure to allocate bond->slave_arr A workqueue (bond_slave_arr_handler) is supposed to retry the allocation later, but if the slave is removed before the workqueue had a chance to complete, bond->slave_arr can still be NULL. [1] Failed to build slave-array. kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] SMP KASAN PTI Modules linked in: Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:bond_update_slave_arr.cold+0xc6/0x198 drivers/net/bonding/bond_main.c:4039 RSP: 0018:ffff88018fe33678 EFLAGS: 00010246 RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffc9000290b000 RDX: 0000000000000000 RSI: ffffffff82b63037 RDI: ffff88019745ea20 RBP: ffff88018fe33760 R08: ffff880170754280 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 R13: ffff88019745ea00 R14: 0000000000000000 R15: ffff88018fe338b0 FS: 00007febd837d700(0000) GS:ffff8801dad00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000004540a0 CR3: 00000001c242e005 CR4: 00000000001626f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: [] __bond_release_one+0x43e/0x500 drivers/net/bonding/bond_main.c:1923 [] bond_release drivers/net/bonding/bond_main.c:2039 [inline] [] bond_do_ioctl+0x416/0x870 drivers/net/bonding/bond_main.c:3562 [] dev_ifsioc+0x6f4/0x940 net/core/dev_ioctl.c:328 [] dev_ioctl+0x1b8/0xc70 net/core/dev_ioctl.c:495 [] sock_do_ioctl+0x1bd/0x300 net/socket.c:1088 [] sock_ioctl+0x300/0x5d0 net/socket.c:1196 [] vfs_ioctl fs/ioctl.c:47 [inline] [] file_ioctl fs/ioctl.c:501 [inline] [] do_vfs_ioctl+0xacb/0x1300 fs/ioctl.c:688 [] SYSC_ioctl fs/ioctl.c:705 [inline] [] SyS_ioctl+0xb6/0xe0 fs/ioctl.c:696 [] do_syscall_64+0x528/0x770 arch/x86/entry/common.c:305 [] entry_SYSCALL_64_after_hwframe+0x42/0xb7 Fixes: ee6377147409 ("bonding: Simplify the xmit function for modes that use xmit_hash") Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Mahesh Bandewar Signed-off-by: Jakub Kicinski --- drivers/net/bonding/bond_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 931d9d935686..21d8fcc83c9c 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4039,7 +4039,7 @@ out: * this to-be-skipped slave to send a packet out. */ old_arr = rtnl_dereference(bond->slave_arr); - for (idx = 0; idx < old_arr->count; idx++) { + for (idx = 0; old_arr != NULL && idx < old_arr->count; idx++) { if (skipslave == old_arr->arr[idx]) { old_arr->arr[idx] = old_arr->arr[old_arr->count-1]; From 819be8108fded0b9e710bbbf81193e52f7bab2f7 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 8 Oct 2019 19:09:23 +0800 Subject: [PATCH 054/145] sctp: add chunks to sk_backlog when the newsk sk_socket is not set This patch is to fix a NULL-ptr deref in selinux_socket_connect_helper: [...] kasan: GPF could be caused by NULL-ptr deref or user memory access [...] RIP: 0010:selinux_socket_connect_helper+0x94/0x460 [...] Call Trace: [...] selinux_sctp_bind_connect+0x16a/0x1d0 [...] security_sctp_bind_connect+0x58/0x90 [...] sctp_process_asconf+0xa52/0xfd0 [sctp] [...] sctp_sf_do_asconf+0x785/0x980 [sctp] [...] sctp_do_sm+0x175/0x5a0 [sctp] [...] sctp_assoc_bh_rcv+0x285/0x5b0 [sctp] [...] sctp_backlog_rcv+0x482/0x910 [sctp] [...] __release_sock+0x11e/0x310 [...] release_sock+0x4f/0x180 [...] sctp_accept+0x3f9/0x5a0 [sctp] [...] inet_accept+0xe7/0x720 It was caused by that the 'newsk' sk_socket was not set before going to security sctp hook when processing asconf chunk with SCTP_PARAM_ADD_IP or SCTP_PARAM_SET_PRIMARY: inet_accept()-> sctp_accept(): lock_sock(): lock listening 'sk' do_softirq(): sctp_rcv(): <-- [1] asconf chunk arrives and enqueued in 'sk' backlog sctp_sock_migrate(): set asoc's sk to 'newsk' release_sock(): sctp_backlog_rcv(): lock 'newsk' sctp_process_asconf() <-- [2] unlock 'newsk' sock_graft(): set sk_socket <-- [3] As it shows, at [1] the asconf chunk would be put into the listening 'sk' backlog, as accept() was holding its sock lock. Then at [2] asconf would get processed with 'newsk' as asoc's sk had been set to 'newsk'. However, 'newsk' sk_socket is not set until [3], while selinux_sctp_bind_connect() would deref it, then kernel crashed. Here to fix it by adding the chunk to sk_backlog until newsk sk_socket is set when .accept() is done. Note that sk->sk_socket can be NULL when the sock is closed, so SOCK_DEAD flag is also needed to check in sctp_newsk_ready(). Thanks to Ondrej for reviewing the code. Fixes: d452930fd3b9 ("selinux: Add SCTP support") Reported-by: Ying Xu Suggested-by: Marcelo Ricardo Leitner Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Acked-by: Neil Horman Signed-off-by: Jakub Kicinski --- include/net/sctp/sctp.h | 5 +++++ net/sctp/input.c | 12 +++++++++--- 2 files changed, 14 insertions(+), 3 deletions(-) diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 5d60f13d2347..3ab5c6bbb90b 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -610,4 +610,9 @@ static inline __u32 sctp_min_frag_point(struct sctp_sock *sp, __u16 datasize) return sctp_mtu_payload(sp, SCTP_DEFAULT_MINSEGMENT, datasize); } +static inline bool sctp_newsk_ready(const struct sock *sk) +{ + return sock_flag(sk, SOCK_DEAD) || sk->sk_socket; +} + #endif /* __net_sctp_h__ */ diff --git a/net/sctp/input.c b/net/sctp/input.c index 5a070fb5b278..f2771375bfc0 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -243,7 +243,7 @@ int sctp_rcv(struct sk_buff *skb) bh_lock_sock(sk); } - if (sock_owned_by_user(sk)) { + if (sock_owned_by_user(sk) || !sctp_newsk_ready(sk)) { if (sctp_add_backlog(sk, skb)) { bh_unlock_sock(sk); sctp_chunk_free(chunk); @@ -321,7 +321,7 @@ int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) local_bh_disable(); bh_lock_sock(sk); - if (sock_owned_by_user(sk)) { + if (sock_owned_by_user(sk) || !sctp_newsk_ready(sk)) { if (sk_add_backlog(sk, skb, sk->sk_rcvbuf)) sctp_chunk_free(chunk); else @@ -336,7 +336,13 @@ int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) if (backloged) return 0; } else { - sctp_inq_push(inqueue, chunk); + if (!sctp_newsk_ready(sk)) { + if (!sk_add_backlog(sk, skb, sk->sk_rcvbuf)) + return 0; + sctp_chunk_free(chunk); + } else { + sctp_inq_push(inqueue, chunk); + } } done: From 9db74e51ec08d75c3e97f851299bc25e645cc5ad Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 8 Oct 2019 08:39:04 -0700 Subject: [PATCH 055/145] phylink: fix kernel-doc warnings Fix kernel-doc warnings in phylink.c: ../drivers/net/phy/phylink.c:595: warning: Function parameter or member 'config' not described in 'phylink_create' ../drivers/net/phy/phylink.c:595: warning: Excess function parameter 'ndev' description in 'phylink_create' Fixes: 8796c8923d9c ("phylink: add documentation for kernel APIs") Signed-off-by: Randy Dunlap Cc: Russell King Signed-off-by: Jakub Kicinski --- drivers/net/phy/phylink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index a5a57ca94c1a..20e2ebe458f2 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -576,7 +576,7 @@ static int phylink_register_sfp(struct phylink *pl, /** * phylink_create() - create a phylink instance - * @ndev: a pointer to the &struct net_device + * @config: a pointer to the target &struct phylink_config * @fwnode: a pointer to a &struct fwnode_handle describing the network * interface * @iface: the desired link mode defined by &typedef phy_interface_t From b528965bcc827dad32a8d21745feaacfc76c9703 Mon Sep 17 00:00:00 2001 From: Alexandra Winter Date: Tue, 8 Oct 2019 18:21:06 +0200 Subject: [PATCH 056/145] s390/qeth: Fix error handling during VNICC initialization Smatch discovered the use of uninitialized variable sup_cmds in error paths. Fixes: caa1f0b10d18 ("s390/qeth: add VNICC enable/disable support") Signed-off-by: Alexandra Winter Signed-off-by: Julian Wiedmann Signed-off-by: Jakub Kicinski --- drivers/s390/net/qeth_l2_main.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index b8799cd3e7aa..de62a9ccc882 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -2021,10 +2021,10 @@ static bool qeth_l2_vnicc_recover_char(struct qeth_card *card, u32 vnicc, static void qeth_l2_vnicc_init(struct qeth_card *card) { u32 *timeout = &card->options.vnicc.learning_timeout; + bool enable, error = false; unsigned int chars_len, i; unsigned long chars_tmp; u32 sup_cmds, vnicc; - bool enable, error; QETH_CARD_TEXT(card, 2, "vniccini"); /* reset rx_bcast */ @@ -2045,7 +2045,10 @@ static void qeth_l2_vnicc_init(struct qeth_card *card) chars_len = sizeof(card->options.vnicc.sup_chars) * BITS_PER_BYTE; for_each_set_bit(i, &chars_tmp, chars_len) { vnicc = BIT(i); - qeth_l2_vnicc_query_cmds(card, vnicc, &sup_cmds); + if (qeth_l2_vnicc_query_cmds(card, vnicc, &sup_cmds)) { + sup_cmds = 0; + error = true; + } if (!(sup_cmds & IPA_VNICC_SET_TIMEOUT) || !(sup_cmds & IPA_VNICC_GET_TIMEOUT)) card->options.vnicc.getset_timeout_sup &= ~vnicc; @@ -2054,8 +2057,8 @@ static void qeth_l2_vnicc_init(struct qeth_card *card) card->options.vnicc.set_char_sup &= ~vnicc; } /* enforce assumed default values and recover settings, if changed */ - error = qeth_l2_vnicc_recover_timeout(card, QETH_VNICC_LEARNING, - timeout); + error |= qeth_l2_vnicc_recover_timeout(card, QETH_VNICC_LEARNING, + timeout); chars_tmp = card->options.vnicc.wanted_chars ^ QETH_VNICC_DEFAULT; chars_tmp |= QETH_VNICC_BRIDGE_INVISIBLE; chars_len = sizeof(card->options.vnicc.wanted_chars) * BITS_PER_BYTE; From be40a86c319706f90caca144343c64743c32b953 Mon Sep 17 00:00:00 2001 From: Alexandra Winter Date: Tue, 8 Oct 2019 18:21:07 +0200 Subject: [PATCH 057/145] s390/qeth: Fix initialization of vnicc cmd masks during set online Without this patch, a command bit in the supported commands mask is only ever set to unsupported during set online. If a command is ever marked as unsupported (e.g. because of error during qeth_l2_vnicc_query_cmds), subsequent successful initialization (offline/online) would not bring it back. Fixes: caa1f0b10d18 ("s390/qeth: add VNICC enable/disable support") Signed-off-by: Alexandra Winter Signed-off-by: Julian Wiedmann Signed-off-by: Jakub Kicinski --- drivers/s390/net/qeth_l2_main.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index de62a9ccc882..bd8143e51747 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -2049,11 +2049,15 @@ static void qeth_l2_vnicc_init(struct qeth_card *card) sup_cmds = 0; error = true; } - if (!(sup_cmds & IPA_VNICC_SET_TIMEOUT) || - !(sup_cmds & IPA_VNICC_GET_TIMEOUT)) + if ((sup_cmds & IPA_VNICC_SET_TIMEOUT) && + (sup_cmds & IPA_VNICC_GET_TIMEOUT)) + card->options.vnicc.getset_timeout_sup |= vnicc; + else card->options.vnicc.getset_timeout_sup &= ~vnicc; - if (!(sup_cmds & IPA_VNICC_ENABLE) || - !(sup_cmds & IPA_VNICC_DISABLE)) + if ((sup_cmds & IPA_VNICC_ENABLE) && + (sup_cmds & IPA_VNICC_DISABLE)) + card->options.vnicc.set_char_sup |= vnicc; + else card->options.vnicc.set_char_sup &= ~vnicc; } /* enforce assumed default values and recover settings, if changed */ From a954380acde6adf584de76c41e1d520097820dd5 Mon Sep 17 00:00:00 2001 From: Vinicius Costa Gomes Date: Tue, 8 Oct 2019 16:20:07 -0700 Subject: [PATCH 058/145] net: taprio: Fix returning EINVAL when configuring without flags When configuring a taprio instance if "flags" is not specified (or it's zero), taprio currently replies with an "Invalid argument" error. So, set the return value to zero after we are done with all the checks. Fixes: 9c66d1564676 ("taprio: Add support for hardware offloading") Signed-off-by: Vinicius Costa Gomes Acked-by: Vladimir Oltean Signed-off-by: Jakub Kicinski --- net/sched/sch_taprio.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c index 68b543f85a96..6719a65169d4 100644 --- a/net/sched/sch_taprio.c +++ b/net/sched/sch_taprio.c @@ -1341,6 +1341,10 @@ static int taprio_parse_clockid(struct Qdisc *sch, struct nlattr **tb, NL_SET_ERR_MSG(extack, "Specifying a 'clockid' is mandatory"); goto out; } + + /* Everything went ok, return success. */ + err = 0; + out: return err; } From 11c9a7d38af524217efb7a176ad322b97ac2f163 Mon Sep 17 00:00:00 2001 From: YueHaibing Date: Wed, 9 Oct 2019 11:10:52 +0800 Subject: [PATCH 059/145] act_mirred: Fix mirred_init_module error handling If tcf_register_action failed, mirred_device_notifier should be unregistered. Fixes: 3b87956ea645 ("net sched: fix race in mirred device removal") Signed-off-by: YueHaibing Signed-off-by: Jakub Kicinski --- net/sched/act_mirred.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 9ce073a05414..08923b21e566 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -484,7 +484,11 @@ static int __init mirred_init_module(void) return err; pr_info("Mirror/redirect action on\n"); - return tcf_register_action(&act_mirred_ops, &mirred_net_ops); + err = tcf_register_action(&act_mirred_ops, &mirred_net_ops); + if (err) + unregister_netdevice_notifier(&mirred_device_notifier); + + return err; } static void __exit mirred_cleanup_module(void) From e0ae2c578d3909e60e9448207f5d83f785f1129f Mon Sep 17 00:00:00 2001 From: Daniele Palmas Date: Wed, 9 Oct 2019 11:07:18 +0200 Subject: [PATCH 060/145] net: usb: qmi_wwan: add Telit 0x1050 composition MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds support for Telit FN980 0x1050 composition 0x1050: tty, adb, rmnet, tty, tty, tty, tty Signed-off-by: Daniele Palmas Acked-by: Bjørn Mork Signed-off-by: Jakub Kicinski --- drivers/net/usb/qmi_wwan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index 3d77cd402ba9..596428ec71df 100644 --- a/drivers/net/usb/qmi_wwan.c +++ b/drivers/net/usb/qmi_wwan.c @@ -1327,6 +1327,7 @@ static const struct usb_device_id products[] = { {QMI_FIXED_INTF(0x2357, 0x0201, 4)}, /* TP-LINK HSUPA Modem MA180 */ {QMI_FIXED_INTF(0x2357, 0x9000, 4)}, /* TP-LINK MA260 */ {QMI_QUIRK_SET_DTR(0x1bc7, 0x1040, 2)}, /* Telit LE922A */ + {QMI_QUIRK_SET_DTR(0x1bc7, 0x1050, 2)}, /* Telit FN980 */ {QMI_FIXED_INTF(0x1bc7, 0x1100, 3)}, /* Telit ME910 */ {QMI_FIXED_INTF(0x1bc7, 0x1101, 3)}, /* Telit ME910 dual modem */ {QMI_FIXED_INTF(0x1bc7, 0x1200, 5)}, /* Telit LE920 */ From 993e4c929a073595d22c85f59082f0c387e31c21 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Wed, 9 Oct 2019 11:19:10 +0200 Subject: [PATCH 061/145] netns: fix NLM_F_ECHO mechanism for RTM_NEWNSID The flag NLM_F_ECHO aims to reply to the user the message notified to all listeners. It was not the case with the command RTM_NEWNSID, let's fix this. Fixes: 0c7aecd4bde4 ("netns: add rtnl cmd to add and get peer netns ids") Reported-by: Guillaume Nault Signed-off-by: Nicolas Dichtel Acked-by: Guillaume Nault Tested-by: Guillaume Nault Signed-off-by: Jakub Kicinski --- net/core/net_namespace.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index a0e0d298c991..6d3e4821b02d 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -245,7 +245,8 @@ static int __peernet2id(struct net *net, struct net *peer) return __peernet2id_alloc(net, peer, &no); } -static void rtnl_net_notifyid(struct net *net, int cmd, int id); +static void rtnl_net_notifyid(struct net *net, int cmd, int id, u32 portid, + struct nlmsghdr *nlh); /* This function returns the id of a peer netns. If no id is assigned, one will * be allocated and returned. */ @@ -268,7 +269,7 @@ int peernet2id_alloc(struct net *net, struct net *peer) id = __peernet2id_alloc(net, peer, &alloc); spin_unlock_bh(&net->nsid_lock); if (alloc && id >= 0) - rtnl_net_notifyid(net, RTM_NEWNSID, id); + rtnl_net_notifyid(net, RTM_NEWNSID, id, 0, NULL); if (alive) put_net(peer); return id; @@ -532,7 +533,7 @@ static void unhash_nsid(struct net *net, struct net *last) idr_remove(&tmp->netns_ids, id); spin_unlock_bh(&tmp->nsid_lock); if (id >= 0) - rtnl_net_notifyid(tmp, RTM_DELNSID, id); + rtnl_net_notifyid(tmp, RTM_DELNSID, id, 0, NULL); if (tmp == last) break; } @@ -764,7 +765,8 @@ static int rtnl_net_newid(struct sk_buff *skb, struct nlmsghdr *nlh, err = alloc_netid(net, peer, nsid); spin_unlock_bh(&net->nsid_lock); if (err >= 0) { - rtnl_net_notifyid(net, RTM_NEWNSID, err); + rtnl_net_notifyid(net, RTM_NEWNSID, err, NETLINK_CB(skb).portid, + nlh); err = 0; } else if (err == -ENOSPC && nsid >= 0) { err = -EEXIST; @@ -1051,9 +1053,12 @@ end: return err < 0 ? err : skb->len; } -static void rtnl_net_notifyid(struct net *net, int cmd, int id) +static void rtnl_net_notifyid(struct net *net, int cmd, int id, u32 portid, + struct nlmsghdr *nlh) { struct net_fill_args fillargs = { + .portid = portid, + .seq = nlh ? nlh->nlmsg_seq : 0, .cmd = cmd, .nsid = id, }; @@ -1068,7 +1073,7 @@ static void rtnl_net_notifyid(struct net *net, int cmd, int id) if (err < 0) goto err_out; - rtnl_notify(msg, net, 0, RTNLGRP_NSID, NULL, 0); + rtnl_notify(msg, net, portid, RTNLGRP_NSID, nlh, 0); return; err_out: From e37542ba111f3974dc622ae0a21c1787318de500 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Oct 2019 09:19:13 -0700 Subject: [PATCH 062/145] netfilter: conntrack: avoid possible false sharing As hinted by KCSAN, we need at least one READ_ONCE() to prevent a compiler optimization. More details on : https://github.com/google/ktsan/wiki/READ_ONCE-and-WRITE_ONCE#it-may-improve-performance sysbot report : BUG: KCSAN: data-race in __nf_ct_refresh_acct / __nf_ct_refresh_acct read to 0xffff888123eb4f08 of 4 bytes by interrupt on cpu 0: __nf_ct_refresh_acct+0xd4/0x1b0 net/netfilter/nf_conntrack_core.c:1796 nf_ct_refresh_acct include/net/netfilter/nf_conntrack.h:201 [inline] nf_conntrack_tcp_packet+0xd40/0x3390 net/netfilter/nf_conntrack_proto_tcp.c:1161 nf_conntrack_handle_packet net/netfilter/nf_conntrack_core.c:1633 [inline] nf_conntrack_in+0x410/0xaa0 net/netfilter/nf_conntrack_core.c:1727 ipv4_conntrack_in+0x27/0x40 net/netfilter/nf_conntrack_proto.c:178 nf_hook_entry_hookfn include/linux/netfilter.h:135 [inline] nf_hook_slow+0x83/0x160 net/netfilter/core.c:512 nf_hook include/linux/netfilter.h:260 [inline] NF_HOOK include/linux/netfilter.h:303 [inline] ip_rcv+0x12f/0x1a0 net/ipv4/ip_input.c:523 __netif_receive_skb_one_core+0xa7/0xe0 net/core/dev.c:5004 __netif_receive_skb+0x37/0xf0 net/core/dev.c:5118 netif_receive_skb_internal+0x59/0x190 net/core/dev.c:5208 napi_skb_finish net/core/dev.c:5671 [inline] napi_gro_receive+0x28f/0x330 net/core/dev.c:5704 receive_buf+0x284/0x30b0 drivers/net/virtio_net.c:1061 virtnet_receive drivers/net/virtio_net.c:1323 [inline] virtnet_poll+0x436/0x7d0 drivers/net/virtio_net.c:1428 napi_poll net/core/dev.c:6352 [inline] net_rx_action+0x3ae/0xa50 net/core/dev.c:6418 __do_softirq+0x115/0x33f kernel/softirq.c:292 write to 0xffff888123eb4f08 of 4 bytes by task 7191 on cpu 1: __nf_ct_refresh_acct+0xfb/0x1b0 net/netfilter/nf_conntrack_core.c:1797 nf_ct_refresh_acct include/net/netfilter/nf_conntrack.h:201 [inline] nf_conntrack_tcp_packet+0xd40/0x3390 net/netfilter/nf_conntrack_proto_tcp.c:1161 nf_conntrack_handle_packet net/netfilter/nf_conntrack_core.c:1633 [inline] nf_conntrack_in+0x410/0xaa0 net/netfilter/nf_conntrack_core.c:1727 ipv4_conntrack_local+0xbe/0x130 net/netfilter/nf_conntrack_proto.c:200 nf_hook_entry_hookfn include/linux/netfilter.h:135 [inline] nf_hook_slow+0x83/0x160 net/netfilter/core.c:512 nf_hook include/linux/netfilter.h:260 [inline] __ip_local_out+0x1f7/0x2b0 net/ipv4/ip_output.c:114 ip_local_out+0x31/0x90 net/ipv4/ip_output.c:123 __ip_queue_xmit+0x3a8/0xa40 net/ipv4/ip_output.c:532 ip_queue_xmit+0x45/0x60 include/net/ip.h:236 __tcp_transmit_skb+0xdeb/0x1cd0 net/ipv4/tcp_output.c:1158 __tcp_send_ack+0x246/0x300 net/ipv4/tcp_output.c:3685 tcp_send_ack+0x34/0x40 net/ipv4/tcp_output.c:3691 tcp_cleanup_rbuf+0x130/0x360 net/ipv4/tcp.c:1575 Reported by Kernel Concurrency Sanitizer on: CPU: 1 PID: 7191 Comm: syz-fuzzer Not tainted 5.3.0+ #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: cc16921351d8 ("netfilter: conntrack: avoid same-timeout update") Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Jozsef Kadlecsik Cc: Florian Westphal Acked-by: Pablo Neira Ayuso Signed-off-by: Jakub Kicinski --- net/netfilter/nf_conntrack_core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 0c63120b2db2..5cd610b547e0 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1792,8 +1792,8 @@ void __nf_ct_refresh_acct(struct nf_conn *ct, if (nf_ct_is_confirmed(ct)) extra_jiffies += nfct_time_stamp; - if (ct->timeout != extra_jiffies) - ct->timeout = extra_jiffies; + if (READ_ONCE(ct->timeout) != extra_jiffies) + WRITE_ONCE(ct->timeout, extra_jiffies); acct: if (do_acct) nf_ct_acct_update(ct, ctinfo, skb->len); From 4ffdd22e49f47db543906d75453a0048a53071ab Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Oct 2019 09:20:02 -0700 Subject: [PATCH 063/145] tun: remove possible false sharing in tun_flow_update() As mentioned in https://github.com/google/ktsan/wiki/READ_ONCE-and-WRITE_ONCE#it-may-improve-performance a C compiler can legally transform if (e->queue_index != queue_index) e->queue_index = queue_index; to : e->queue_index = queue_index; Note that the code using jiffies has no issue, since jiffies has volatile attribute. if (e->updated != jiffies) e->updated = jiffies; Fixes: 83b1bc122cab ("tun: align write-heavy flow entry members to a cache line") Signed-off-by: Eric Dumazet Cc: Zhang Yu Cc: Wang Li Cc: Li RongQing Acked-by: Jason Wang Signed-off-by: Jakub Kicinski --- drivers/net/tun.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 812dc3a65efb..a8d3141582a5 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -526,8 +526,8 @@ static void tun_flow_update(struct tun_struct *tun, u32 rxhash, e = tun_flow_find(head, rxhash); if (likely(e)) { /* TODO: keep queueing to old queue until it's empty? */ - if (e->queue_index != queue_index) - e->queue_index = queue_index; + if (READ_ONCE(e->queue_index) != queue_index) + WRITE_ONCE(e->queue_index, queue_index); if (e->updated != jiffies) e->updated = jiffies; sock_rps_record_flow_hash(e->rps_rxhash); From 503978aca46124cd714703e180b9c8292ba50ba7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Oct 2019 12:55:53 -0700 Subject: [PATCH 064/145] net: avoid possible false sharing in sk_leave_memory_pressure() As mentioned in https://github.com/google/ktsan/wiki/READ_ONCE-and-WRITE_ONCE#it-may-improve-performance a C compiler can legally transform : if (memory_pressure && *memory_pressure) *memory_pressure = 0; to : if (memory_pressure) *memory_pressure = 0; Fixes: 0604475119de ("tcp: add TCPMemoryPressuresChrono counter") Fixes: 180d8cd942ce ("foundations of per-cgroup memory pressure controlling.") Fixes: 3ab224be6d69 ("[NET] CORE: Introducing new memory accounting interface.") Signed-off-by: Eric Dumazet Signed-off-by: Jakub Kicinski --- net/core/sock.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/core/sock.c b/net/core/sock.c index fac2b4d80de5..50647a10fdb7 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2334,8 +2334,8 @@ static void sk_leave_memory_pressure(struct sock *sk) } else { unsigned long *memory_pressure = sk->sk_prot->memory_pressure; - if (memory_pressure && *memory_pressure) - *memory_pressure = 0; + if (memory_pressure && READ_ONCE(*memory_pressure)) + WRITE_ONCE(*memory_pressure, 0); } } From 60b173ca3d1cd1782bd0096dc17298ec242f6fb1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Oct 2019 14:51:20 -0700 Subject: [PATCH 065/145] net: add {READ|WRITE}_ONCE() annotations on ->rskq_accept_head reqsk_queue_empty() is called from inet_csk_listen_poll() while other cpus might write ->rskq_accept_head value. Use {READ|WRITE}_ONCE() to avoid compiler tricks and potential KCSAN splats. Fixes: fff1f3001cc5 ("tcp: add a spinlock to protect struct request_sock_queue") Signed-off-by: Eric Dumazet Signed-off-by: Jakub Kicinski --- drivers/xen/pvcalls-back.c | 2 +- include/net/request_sock.h | 4 ++-- net/ipv4/inet_connection_sock.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/xen/pvcalls-back.c b/drivers/xen/pvcalls-back.c index 69a626b0e594..c57c71b7d53d 100644 --- a/drivers/xen/pvcalls-back.c +++ b/drivers/xen/pvcalls-back.c @@ -775,7 +775,7 @@ static int pvcalls_back_poll(struct xenbus_device *dev, mappass->reqcopy = *req; icsk = inet_csk(mappass->sock->sk); queue = &icsk->icsk_accept_queue; - data = queue->rskq_accept_head != NULL; + data = READ_ONCE(queue->rskq_accept_head) != NULL; if (data) { mappass->reqcopy.cmd = 0; ret = 0; diff --git a/include/net/request_sock.h b/include/net/request_sock.h index fd178d58fa84..cf8b33213bbc 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -185,7 +185,7 @@ void reqsk_fastopen_remove(struct sock *sk, struct request_sock *req, static inline bool reqsk_queue_empty(const struct request_sock_queue *queue) { - return queue->rskq_accept_head == NULL; + return READ_ONCE(queue->rskq_accept_head) == NULL; } static inline struct request_sock *reqsk_queue_remove(struct request_sock_queue *queue, @@ -197,7 +197,7 @@ static inline struct request_sock *reqsk_queue_remove(struct request_sock_queue req = queue->rskq_accept_head; if (req) { sk_acceptq_removed(parent); - queue->rskq_accept_head = req->dl_next; + WRITE_ONCE(queue->rskq_accept_head, req->dl_next); if (queue->rskq_accept_head == NULL) queue->rskq_accept_tail = NULL; } diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index a9183543ca30..dbcf34ec8dd2 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -934,7 +934,7 @@ struct sock *inet_csk_reqsk_queue_add(struct sock *sk, req->sk = child; req->dl_next = NULL; if (queue->rskq_accept_head == NULL) - queue->rskq_accept_head = req; + WRITE_ONCE(queue->rskq_accept_head, req); else queue->rskq_accept_tail->dl_next = req; queue->rskq_accept_tail = req; From 1f142c17d19a5618d5a633195a46f2c8be9bf232 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Oct 2019 15:10:15 -0700 Subject: [PATCH 066/145] tcp: annotate lockless access to tcp_memory_pressure tcp_memory_pressure is read without holding any lock, and its value could be changed on other cpus. Use READ_ONCE() to annotate these lockless reads. The write side is already using atomic ops. Fixes: b8da51ebb1aa ("tcp: introduce tcp_under_memory_pressure()") Signed-off-by: Eric Dumazet Signed-off-by: Jakub Kicinski --- include/net/tcp.h | 2 +- net/ipv4/tcp.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index c9a3f9688223..88e63d64c698 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -258,7 +258,7 @@ static inline bool tcp_under_memory_pressure(const struct sock *sk) mem_cgroup_under_socket_pressure(sk->sk_memcg)) return true; - return tcp_memory_pressure; + return READ_ONCE(tcp_memory_pressure); } /* * The next routines deal with comparing 32 bit unsigned ints diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index f98a1882e537..888c92b63f5a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -326,7 +326,7 @@ void tcp_enter_memory_pressure(struct sock *sk) { unsigned long val; - if (tcp_memory_pressure) + if (READ_ONCE(tcp_memory_pressure)) return; val = jiffies; @@ -341,7 +341,7 @@ void tcp_leave_memory_pressure(struct sock *sk) { unsigned long val; - if (!tcp_memory_pressure) + if (!READ_ONCE(tcp_memory_pressure)) return; val = xchg(&tcp_memory_pressure, 0); if (val) From 8265792bf8871acc2d00fd03883d830e2249d395 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Oct 2019 15:21:13 -0700 Subject: [PATCH 067/145] net: silence KCSAN warnings around sk_add_backlog() calls sk_add_backlog() callers usually read sk->sk_rcvbuf without owning the socket lock. This means sk_rcvbuf value can be changed by other cpus, and KCSAN complains. Add READ_ONCE() annotations to document the lockless nature of these reads. Note that writes over sk_rcvbuf should also use WRITE_ONCE(), but this will be done in separate patches to ease stable backports (if we decide this is relevant for stable trees). BUG: KCSAN: data-race in tcp_add_backlog / tcp_recvmsg write to 0xffff88812ab369f8 of 8 bytes by interrupt on cpu 1: __sk_add_backlog include/net/sock.h:902 [inline] sk_add_backlog include/net/sock.h:933 [inline] tcp_add_backlog+0x45a/0xcc0 net/ipv4/tcp_ipv4.c:1737 tcp_v4_rcv+0x1aba/0x1bf0 net/ipv4/tcp_ipv4.c:1925 ip_protocol_deliver_rcu+0x51/0x470 net/ipv4/ip_input.c:204 ip_local_deliver_finish+0x110/0x140 net/ipv4/ip_input.c:231 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_local_deliver+0x133/0x210 net/ipv4/ip_input.c:252 dst_input include/net/dst.h:442 [inline] ip_rcv_finish+0x121/0x160 net/ipv4/ip_input.c:413 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_rcv+0x18f/0x1a0 net/ipv4/ip_input.c:523 __netif_receive_skb_one_core+0xa7/0xe0 net/core/dev.c:5004 __netif_receive_skb+0x37/0xf0 net/core/dev.c:5118 netif_receive_skb_internal+0x59/0x190 net/core/dev.c:5208 napi_skb_finish net/core/dev.c:5671 [inline] napi_gro_receive+0x28f/0x330 net/core/dev.c:5704 receive_buf+0x284/0x30b0 drivers/net/virtio_net.c:1061 virtnet_receive drivers/net/virtio_net.c:1323 [inline] virtnet_poll+0x436/0x7d0 drivers/net/virtio_net.c:1428 napi_poll net/core/dev.c:6352 [inline] net_rx_action+0x3ae/0xa50 net/core/dev.c:6418 read to 0xffff88812ab369f8 of 8 bytes by task 7271 on cpu 0: tcp_recvmsg+0x470/0x1a30 net/ipv4/tcp.c:2047 inet_recvmsg+0xbb/0x250 net/ipv4/af_inet.c:838 sock_recvmsg_nosec net/socket.c:871 [inline] sock_recvmsg net/socket.c:889 [inline] sock_recvmsg+0x92/0xb0 net/socket.c:885 sock_read_iter+0x15f/0x1e0 net/socket.c:967 call_read_iter include/linux/fs.h:1864 [inline] new_sync_read+0x389/0x4f0 fs/read_write.c:414 __vfs_read+0xb1/0xc0 fs/read_write.c:427 vfs_read fs/read_write.c:461 [inline] vfs_read+0x143/0x2c0 fs/read_write.c:446 ksys_read+0xd5/0x1b0 fs/read_write.c:587 __do_sys_read fs/read_write.c:597 [inline] __se_sys_read fs/read_write.c:595 [inline] __x64_sys_read+0x4c/0x60 fs/read_write.c:595 do_syscall_64+0xcf/0x2f0 arch/x86/entry/common.c:296 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 7271 Comm: syz-fuzzer Not tainted 5.3.0+ #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Signed-off-by: Eric Dumazet Reported-by: syzbot Signed-off-by: Jakub Kicinski --- net/core/sock.c | 2 +- net/ipv4/tcp_ipv4.c | 2 +- net/llc/llc_conn.c | 2 +- net/sctp/input.c | 6 +++--- net/tipc/socket.c | 6 +++--- net/x25/x25_dev.c | 2 +- 6 files changed, 10 insertions(+), 10 deletions(-) diff --git a/net/core/sock.c b/net/core/sock.c index 50647a10fdb7..1cf06934da50 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -522,7 +522,7 @@ int __sk_receive_skb(struct sock *sk, struct sk_buff *skb, rc = sk_backlog_rcv(sk, skb); mutex_release(&sk->sk_lock.dep_map, 1, _RET_IP_); - } else if (sk_add_backlog(sk, skb, sk->sk_rcvbuf)) { + } else if (sk_add_backlog(sk, skb, READ_ONCE(sk->sk_rcvbuf))) { bh_unlock_sock(sk); atomic_inc(&sk->sk_drops); goto discard_and_relse; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index bf124b1742df..492bf6a6b023 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1644,7 +1644,7 @@ int tcp_v4_early_demux(struct sk_buff *skb) bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) { - u32 limit = sk->sk_rcvbuf + sk->sk_sndbuf; + u32 limit = READ_ONCE(sk->sk_rcvbuf) + READ_ONCE(sk->sk_sndbuf); struct skb_shared_info *shinfo; const struct tcphdr *th; struct tcphdr *thtail; diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c index a79b739eb223..7b620acaca9e 100644 --- a/net/llc/llc_conn.c +++ b/net/llc/llc_conn.c @@ -813,7 +813,7 @@ void llc_conn_handler(struct llc_sap *sap, struct sk_buff *skb) else { dprintk("%s: adding to backlog...\n", __func__); llc_set_backlog_type(skb, LLC_PACKET); - if (sk_add_backlog(sk, skb, sk->sk_rcvbuf)) + if (sk_add_backlog(sk, skb, READ_ONCE(sk->sk_rcvbuf))) goto drop_unlock; } out: diff --git a/net/sctp/input.c b/net/sctp/input.c index f2771375bfc0..2277981559d0 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -322,7 +322,7 @@ int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) bh_lock_sock(sk); if (sock_owned_by_user(sk) || !sctp_newsk_ready(sk)) { - if (sk_add_backlog(sk, skb, sk->sk_rcvbuf)) + if (sk_add_backlog(sk, skb, READ_ONCE(sk->sk_rcvbuf))) sctp_chunk_free(chunk); else backloged = 1; @@ -337,7 +337,7 @@ int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) return 0; } else { if (!sctp_newsk_ready(sk)) { - if (!sk_add_backlog(sk, skb, sk->sk_rcvbuf)) + if (!sk_add_backlog(sk, skb, READ_ONCE(sk->sk_rcvbuf))) return 0; sctp_chunk_free(chunk); } else { @@ -364,7 +364,7 @@ static int sctp_add_backlog(struct sock *sk, struct sk_buff *skb) struct sctp_ep_common *rcvr = chunk->rcvr; int ret; - ret = sk_add_backlog(sk, skb, sk->sk_rcvbuf); + ret = sk_add_backlog(sk, skb, READ_ONCE(sk->sk_rcvbuf)); if (!ret) { /* Hold the assoc/ep while hanging on the backlog queue. * This way, we know structures we need will not disappear diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 3b9f8cc328f5..7c736cfec57f 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -2119,13 +2119,13 @@ static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *skb) struct tipc_msg *hdr = buf_msg(skb); if (unlikely(msg_in_group(hdr))) - return sk->sk_rcvbuf; + return READ_ONCE(sk->sk_rcvbuf); if (unlikely(!msg_connected(hdr))) - return sk->sk_rcvbuf << msg_importance(hdr); + return READ_ONCE(sk->sk_rcvbuf) << msg_importance(hdr); if (likely(tsk->peer_caps & TIPC_BLOCK_FLOWCTL)) - return sk->sk_rcvbuf; + return READ_ONCE(sk->sk_rcvbuf); return FLOWCTL_MSG_LIM; } diff --git a/net/x25/x25_dev.c b/net/x25/x25_dev.c index 5c111bc3c8ea..00e782335cb0 100644 --- a/net/x25/x25_dev.c +++ b/net/x25/x25_dev.c @@ -55,7 +55,7 @@ static int x25_receive_data(struct sk_buff *skb, struct x25_neigh *nb) if (!sock_owned_by_user(sk)) { queued = x25_process_rx_frame(sk, skb); } else { - queued = !sk_add_backlog(sk, skb, sk->sk_rcvbuf); + queued = !sk_add_backlog(sk, skb, READ_ONCE(sk->sk_rcvbuf)); } bh_unlock_sock(sk); sock_put(sk); From eac66402d1c342f07ff38f8d631ff95eb7ad3220 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Oct 2019 15:32:35 -0700 Subject: [PATCH 068/145] net: annotate sk->sk_rcvlowat lockless reads sock_rcvlowat() or int_sk_rcvlowat() might be called without the socket lock for example from tcp_poll(). Use READ_ONCE() to document the fact that other cpus might change sk->sk_rcvlowat under us and avoid KCSAN splats. Use WRITE_ONCE() on write sides too. Signed-off-by: Eric Dumazet Signed-off-by: Jakub Kicinski --- include/net/sock.h | 4 +++- net/core/filter.c | 2 +- net/core/sock.c | 2 +- net/ipv4/tcp.c | 2 +- net/sched/em_meta.c | 2 +- 5 files changed, 7 insertions(+), 5 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 2c53f1a1d905..79f54e1f8827 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2271,7 +2271,9 @@ static inline long sock_sndtimeo(const struct sock *sk, bool noblock) static inline int sock_rcvlowat(const struct sock *sk, int waitall, int len) { - return (waitall ? len : min_t(int, sk->sk_rcvlowat, len)) ? : 1; + int v = waitall ? len : min_t(int, READ_ONCE(sk->sk_rcvlowat), len); + + return v ?: 1; } /* Alas, with timeout socket operations are not restartable. diff --git a/net/core/filter.c b/net/core/filter.c index ed6563622ce3..a50c0b6846f2 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -4274,7 +4274,7 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, case SO_RCVLOWAT: if (val < 0) val = INT_MAX; - sk->sk_rcvlowat = val ? : 1; + WRITE_ONCE(sk->sk_rcvlowat, val ? : 1); break; case SO_MARK: if (sk->sk_mark != val) { diff --git a/net/core/sock.c b/net/core/sock.c index 1cf06934da50..b7c5c6ea51ba 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -974,7 +974,7 @@ set_rcvbuf: if (sock->ops->set_rcvlowat) ret = sock->ops->set_rcvlowat(sk, val); else - sk->sk_rcvlowat = val ? : 1; + WRITE_ONCE(sk->sk_rcvlowat, val ? : 1); break; case SO_RCVTIMEO_OLD: diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 888c92b63f5a..8781a92ea4b6 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1699,7 +1699,7 @@ int tcp_set_rcvlowat(struct sock *sk, int val) else cap = sock_net(sk)->ipv4.sysctl_tcp_rmem[2] >> 1; val = min(val, cap); - sk->sk_rcvlowat = val ? : 1; + WRITE_ONCE(sk->sk_rcvlowat, val ? : 1); /* Check if we need to signal EPOLLIN right now */ tcp_data_ready(sk); diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index 82bd14e7ac93..4c9122fc35c9 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -554,7 +554,7 @@ META_COLLECTOR(int_sk_rcvlowat) *err = -1; return; } - dst->value = sk->sk_rcvlowat; + dst->value = READ_ONCE(sk->sk_rcvlowat); } META_COLLECTOR(int_sk_rcvtimeo) From 70c2655849a25431f31b505a07fe0c861e5e41fb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Oct 2019 15:41:03 -0700 Subject: [PATCH 069/145] net: silence KCSAN warnings about sk->sk_backlog.len reads sk->sk_backlog.len can be written by BH handlers, and read from process contexts in a lockless way. Note the write side should also use WRITE_ONCE() or a variant. We need some agreement about the best way to do this. syzbot reported : BUG: KCSAN: data-race in tcp_add_backlog / tcp_grow_window.isra.0 write to 0xffff88812665f32c of 4 bytes by interrupt on cpu 1: sk_add_backlog include/net/sock.h:934 [inline] tcp_add_backlog+0x4a0/0xcc0 net/ipv4/tcp_ipv4.c:1737 tcp_v4_rcv+0x1aba/0x1bf0 net/ipv4/tcp_ipv4.c:1925 ip_protocol_deliver_rcu+0x51/0x470 net/ipv4/ip_input.c:204 ip_local_deliver_finish+0x110/0x140 net/ipv4/ip_input.c:231 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_local_deliver+0x133/0x210 net/ipv4/ip_input.c:252 dst_input include/net/dst.h:442 [inline] ip_rcv_finish+0x121/0x160 net/ipv4/ip_input.c:413 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_rcv+0x18f/0x1a0 net/ipv4/ip_input.c:523 __netif_receive_skb_one_core+0xa7/0xe0 net/core/dev.c:5004 __netif_receive_skb+0x37/0xf0 net/core/dev.c:5118 netif_receive_skb_internal+0x59/0x190 net/core/dev.c:5208 napi_skb_finish net/core/dev.c:5671 [inline] napi_gro_receive+0x28f/0x330 net/core/dev.c:5704 receive_buf+0x284/0x30b0 drivers/net/virtio_net.c:1061 virtnet_receive drivers/net/virtio_net.c:1323 [inline] virtnet_poll+0x436/0x7d0 drivers/net/virtio_net.c:1428 napi_poll net/core/dev.c:6352 [inline] net_rx_action+0x3ae/0xa50 net/core/dev.c:6418 read to 0xffff88812665f32c of 4 bytes by task 7292 on cpu 0: tcp_space include/net/tcp.h:1373 [inline] tcp_grow_window.isra.0+0x6b/0x480 net/ipv4/tcp_input.c:413 tcp_event_data_recv+0x68f/0x990 net/ipv4/tcp_input.c:717 tcp_rcv_established+0xbfe/0xf50 net/ipv4/tcp_input.c:5618 tcp_v4_do_rcv+0x381/0x4e0 net/ipv4/tcp_ipv4.c:1542 sk_backlog_rcv include/net/sock.h:945 [inline] __release_sock+0x135/0x1e0 net/core/sock.c:2427 release_sock+0x61/0x160 net/core/sock.c:2943 tcp_recvmsg+0x63b/0x1a30 net/ipv4/tcp.c:2181 inet_recvmsg+0xbb/0x250 net/ipv4/af_inet.c:838 sock_recvmsg_nosec net/socket.c:871 [inline] sock_recvmsg net/socket.c:889 [inline] sock_recvmsg+0x92/0xb0 net/socket.c:885 sock_read_iter+0x15f/0x1e0 net/socket.c:967 call_read_iter include/linux/fs.h:1864 [inline] new_sync_read+0x389/0x4f0 fs/read_write.c:414 __vfs_read+0xb1/0xc0 fs/read_write.c:427 vfs_read fs/read_write.c:461 [inline] vfs_read+0x143/0x2c0 fs/read_write.c:446 Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 7292 Comm: syz-fuzzer Not tainted 5.3.0+ #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Signed-off-by: Eric Dumazet Reported-by: syzbot Signed-off-by: Jakub Kicinski --- include/net/tcp.h | 3 ++- net/core/sock.c | 2 +- net/sctp/diag.c | 2 +- net/tipc/socket.c | 2 +- 4 files changed, 5 insertions(+), 4 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index 88e63d64c698..35f6f7e0fdc2 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1380,7 +1380,8 @@ static inline int tcp_win_from_space(const struct sock *sk, int space) /* Note: caller must be prepared to deal with negative returns */ static inline int tcp_space(const struct sock *sk) { - return tcp_win_from_space(sk, sk->sk_rcvbuf - sk->sk_backlog.len - + return tcp_win_from_space(sk, sk->sk_rcvbuf - + READ_ONCE(sk->sk_backlog.len) - atomic_read(&sk->sk_rmem_alloc)); } diff --git a/net/core/sock.c b/net/core/sock.c index b7c5c6ea51ba..2a053999df11 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3210,7 +3210,7 @@ void sk_get_meminfo(const struct sock *sk, u32 *mem) mem[SK_MEMINFO_FWD_ALLOC] = sk->sk_forward_alloc; mem[SK_MEMINFO_WMEM_QUEUED] = sk->sk_wmem_queued; mem[SK_MEMINFO_OPTMEM] = atomic_read(&sk->sk_omem_alloc); - mem[SK_MEMINFO_BACKLOG] = sk->sk_backlog.len; + mem[SK_MEMINFO_BACKLOG] = READ_ONCE(sk->sk_backlog.len); mem[SK_MEMINFO_DROPS] = atomic_read(&sk->sk_drops); } diff --git a/net/sctp/diag.c b/net/sctp/diag.c index fc9a4c6629ce..0851166b9175 100644 --- a/net/sctp/diag.c +++ b/net/sctp/diag.c @@ -175,7 +175,7 @@ static int inet_sctp_diag_fill(struct sock *sk, struct sctp_association *asoc, mem[SK_MEMINFO_FWD_ALLOC] = sk->sk_forward_alloc; mem[SK_MEMINFO_WMEM_QUEUED] = sk->sk_wmem_queued; mem[SK_MEMINFO_OPTMEM] = atomic_read(&sk->sk_omem_alloc); - mem[SK_MEMINFO_BACKLOG] = sk->sk_backlog.len; + mem[SK_MEMINFO_BACKLOG] = READ_ONCE(sk->sk_backlog.len); mem[SK_MEMINFO_DROPS] = atomic_read(&sk->sk_drops); if (nla_put(skb, INET_DIAG_SKMEMINFO, sizeof(mem), &mem) < 0) diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 7c736cfec57f..f8bbc4aab213 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -3790,7 +3790,7 @@ int tipc_sk_dump(struct sock *sk, u16 dqueues, char *buf) i += scnprintf(buf + i, sz - i, " %d", sk->sk_sndbuf); i += scnprintf(buf + i, sz - i, " | %d", sk_rmem_alloc_get(sk)); i += scnprintf(buf + i, sz - i, " %d", sk->sk_rcvbuf); - i += scnprintf(buf + i, sz - i, " | %d\n", sk->sk_backlog.len); + i += scnprintf(buf + i, sz - i, " | %d\n", READ_ONCE(sk->sk_backlog.len)); if (dqueues & TIPC_DUMP_SK_SNDQ) { i += scnprintf(buf + i, sz - i, "sk_write_queue: "); From 4ebcb113edcc498888394821bca2e60ef89c6ff3 Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Wed, 9 Oct 2019 20:55:48 +0200 Subject: [PATCH 070/145] r8169: fix jumbo packet handling on resume from suspend Mariusz reported that invalid packets are sent after resume from suspend if jumbo packets are active. It turned out that his BIOS resets chip settings to non-jumbo on resume. Most chip settings are re-initialized on resume from suspend by calling rtl_hw_start(), so let's add configuring jumbo to this function. There's nothing wrong with the commit marked as fixed, it's just the first one where the patch applies cleanly. Fixes: 7366016d2d4c ("r8169: read common register for PCI commit") Reported-by: Mariusz Bialonczyk Tested-by: Mariusz Bialonczyk Signed-off-by: Heiner Kallweit Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/r8169_main.c | 35 +++++++---------------- 1 file changed, 11 insertions(+), 24 deletions(-) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 74f81fe03810..350b0d949611 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -4146,6 +4146,14 @@ static void rtl_hw_jumbo_disable(struct rtl8169_private *tp) rtl_lock_config_regs(tp); } +static void rtl_jumbo_config(struct rtl8169_private *tp, int mtu) +{ + if (mtu > ETH_DATA_LEN) + rtl_hw_jumbo_enable(tp); + else + rtl_hw_jumbo_disable(tp); +} + DECLARE_RTL_COND(rtl_chipcmd_cond) { return RTL_R8(tp, ChipCmd) & CmdReset; @@ -4442,11 +4450,6 @@ static void rtl8168g_set_pause_thresholds(struct rtl8169_private *tp, static void rtl_hw_start_8168bb(struct rtl8169_private *tp) { RTL_W8(tp, Config3, RTL_R8(tp, Config3) & ~Beacon_en); - - if (tp->dev->mtu <= ETH_DATA_LEN) { - rtl_tx_performance_tweak(tp, PCI_EXP_DEVCTL_READRQ_4096B | - PCI_EXP_DEVCTL_NOSNOOP_EN); - } } static void rtl_hw_start_8168bef(struct rtl8169_private *tp) @@ -4462,9 +4465,6 @@ static void __rtl_hw_start_8168cp(struct rtl8169_private *tp) RTL_W8(tp, Config3, RTL_R8(tp, Config3) & ~Beacon_en); - if (tp->dev->mtu <= ETH_DATA_LEN) - rtl_tx_performance_tweak(tp, PCI_EXP_DEVCTL_READRQ_4096B); - rtl_disable_clock_request(tp); } @@ -4490,9 +4490,6 @@ static void rtl_hw_start_8168cp_2(struct rtl8169_private *tp) rtl_set_def_aspm_entry_latency(tp); RTL_W8(tp, Config3, RTL_R8(tp, Config3) & ~Beacon_en); - - if (tp->dev->mtu <= ETH_DATA_LEN) - rtl_tx_performance_tweak(tp, PCI_EXP_DEVCTL_READRQ_4096B); } static void rtl_hw_start_8168cp_3(struct rtl8169_private *tp) @@ -4503,9 +4500,6 @@ static void rtl_hw_start_8168cp_3(struct rtl8169_private *tp) /* Magic. */ RTL_W8(tp, DBG_REG, 0x20); - - if (tp->dev->mtu <= ETH_DATA_LEN) - rtl_tx_performance_tweak(tp, PCI_EXP_DEVCTL_READRQ_4096B); } static void rtl_hw_start_8168c_1(struct rtl8169_private *tp) @@ -4611,9 +4605,6 @@ static void rtl_hw_start_8168e_1(struct rtl8169_private *tp) rtl_ephy_init(tp, e_info_8168e_1); - if (tp->dev->mtu <= ETH_DATA_LEN) - rtl_tx_performance_tweak(tp, PCI_EXP_DEVCTL_READRQ_4096B); - rtl_disable_clock_request(tp); /* Reset tx FIFO pointer */ @@ -4636,9 +4627,6 @@ static void rtl_hw_start_8168e_2(struct rtl8169_private *tp) rtl_ephy_init(tp, e_info_8168e_2); - if (tp->dev->mtu <= ETH_DATA_LEN) - rtl_tx_performance_tweak(tp, PCI_EXP_DEVCTL_READRQ_4096B); - rtl_eri_write(tp, 0xc0, ERIAR_MASK_0011, 0x0000); rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000); rtl_set_fifo_size(tp, 0x10, 0x10, 0x02, 0x06); @@ -5485,6 +5473,8 @@ static void rtl_hw_start(struct rtl8169_private *tp) rtl_set_rx_tx_desc_registers(tp); rtl_lock_config_regs(tp); + rtl_jumbo_config(tp, tp->dev->mtu); + /* Initially a 10 us delay. Turned it into a PCI commit. - FR */ RTL_R16(tp, CPlusCmd); RTL_W8(tp, ChipCmd, CmdTxEnb | CmdRxEnb); @@ -5498,10 +5488,7 @@ static int rtl8169_change_mtu(struct net_device *dev, int new_mtu) { struct rtl8169_private *tp = netdev_priv(dev); - if (new_mtu > ETH_DATA_LEN) - rtl_hw_jumbo_enable(tp); - else - rtl_hw_jumbo_disable(tp); + rtl_jumbo_config(tp, new_mtu); dev->mtu = new_mtu; netdev_update_features(dev); From 2168da4594040529f6a6cb413c22668a4402f83c Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Wed, 9 Oct 2019 12:18:31 -0700 Subject: [PATCH 071/145] net: update net_dim documentation after rename Commit 8960b38932be ("linux/dim: Rename externally used net_dim members") renamed the net_dim API, removing the "net_" prefix from the structures and functions. The patch didn't update the net_dim.txt documentation file. Fix the documentation so that its examples match the current code. Fixes: 8960b38932be ("linux/dim: Rename externally used net_dim members", 2019-06-25) Fixes: c002bd529d71 ("linux/dim: Rename externally exposed macros", 2019-06-25) Fixes: 4f75da3666c0 ("linux/dim: Move implementation to .c files") Cc: Tal Gilboa Signed-off-by: Jacob Keller Signed-off-by: Jakub Kicinski --- Documentation/networking/net_dim.txt | 36 ++++++++++++++-------------- 1 file changed, 18 insertions(+), 18 deletions(-) diff --git a/Documentation/networking/net_dim.txt b/Documentation/networking/net_dim.txt index 9cb31c5e2dcd..9bdb7d5a3ba3 100644 --- a/Documentation/networking/net_dim.txt +++ b/Documentation/networking/net_dim.txt @@ -92,16 +92,16 @@ under some conditions. Part III: Registering a Network Device to DIM ============================================== -Net DIM API exposes the main function net_dim(struct net_dim *dim, -struct net_dim_sample end_sample). This function is the entry point to the Net +Net DIM API exposes the main function net_dim(struct dim *dim, +struct dim_sample end_sample). This function is the entry point to the Net DIM algorithm and has to be called every time the driver would like to check if it should change interrupt moderation parameters. The driver should provide two -data structures: struct net_dim and struct net_dim_sample. Struct net_dim +data structures: struct dim and struct dim_sample. Struct dim describes the state of DIM for a specific object (RX queue, TX queue, other queues, etc.). This includes the current selected profile, previous data samples, the callback function provided by the driver and more. -Struct net_dim_sample describes a data sample, which will be compared to the -data sample stored in struct net_dim in order to decide on the algorithm's next +Struct dim_sample describes a data sample, which will be compared to the +data sample stored in struct dim in order to decide on the algorithm's next step. The sample should include bytes, packets and interrupts, measured by the driver. @@ -110,9 +110,9 @@ main net_dim() function. The recommended method is to call net_dim() on each interrupt. Since Net DIM has a built-in moderation and it might decide to skip iterations under certain conditions, there is no need to moderate the net_dim() calls as well. As mentioned above, the driver needs to provide an object of type -struct net_dim to the net_dim() function call. It is advised for each entity -using Net DIM to hold a struct net_dim as part of its data structure and use it -as the main Net DIM API object. The struct net_dim_sample should hold the latest +struct dim to the net_dim() function call. It is advised for each entity +using Net DIM to hold a struct dim as part of its data structure and use it +as the main Net DIM API object. The struct dim_sample should hold the latest bytes, packets and interrupts count. No need to perform any calculations, just include the raw data. @@ -132,19 +132,19 @@ usage is not complete but it should make the outline of the usage clear. my_driver.c: -#include +#include /* Callback for net DIM to schedule on a decision to change moderation */ void my_driver_do_dim_work(struct work_struct *work) { - /* Get struct net_dim from struct work_struct */ - struct net_dim *dim = container_of(work, struct net_dim, - work); + /* Get struct dim from struct work_struct */ + struct dim *dim = container_of(work, struct dim, + work); /* Do interrupt moderation related stuff */ ... /* Signal net DIM work is done and it should move to next iteration */ - dim->state = NET_DIM_START_MEASURE; + dim->state = DIM_START_MEASURE; } /* My driver's interrupt handler */ @@ -152,13 +152,13 @@ int my_driver_handle_interrupt(struct my_driver_entity *my_entity, ...) { ... /* A struct to hold current measured data */ - struct net_dim_sample dim_sample; + struct dim_sample dim_sample; ... /* Initiate data sample struct with current data */ - net_dim_sample(my_entity->events, - my_entity->packets, - my_entity->bytes, - &dim_sample); + dim_update_sample(my_entity->events, + my_entity->packets, + my_entity->bytes, + &dim_sample); /* Call net DIM */ net_dim(&my_entity->dim, dim_sample); ... From 29ee2701529e1905c0e948688f9688c68c8d4ea4 Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Thu, 10 Oct 2019 10:16:09 +0200 Subject: [PATCH 072/145] net/smc: fix SMCD link group creation with VLAN id If creation of an SMCD link group with VLAN id fails, the initial smc_ism_get_vlan() step has to be reverted as well. Fixes: c6ba7c9ba43d ("net/smc: add base infrastructure for SMC-D and ISM") Signed-off-by: Ursula Braun Signed-off-by: Karsten Graul Signed-off-by: Jakub Kicinski --- net/smc/smc_core.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 4ca50ddf8d16..88556f0251ab 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -213,7 +213,7 @@ static int smc_lgr_create(struct smc_sock *smc, struct smc_init_info *ini) lgr = kzalloc(sizeof(*lgr), GFP_KERNEL); if (!lgr) { rc = SMC_CLC_DECL_MEM; - goto out; + goto ism_put_vlan; } lgr->is_smcd = ini->is_smcd; lgr->sync_err = 0; @@ -289,6 +289,9 @@ clear_llc_lnk: smc_llc_link_clear(lnk); free_lgr: kfree(lgr); +ism_put_vlan: + if (ini->is_smcd && ini->vlan_id) + smc_ism_put_vlan(ini->ism_dev, ini->vlan_id); out: if (rc < 0) { if (rc == -ENOMEM) From 882dcfe5a1785c20f45820cbe6fec4b8b647c946 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Thu, 10 Oct 2019 10:16:10 +0200 Subject: [PATCH 073/145] net/smc: receive returns without data smc_cdc_rxed_any_close_or_senddone() is used as an end condition for the receive loop. This conflicts with smc_cdc_msg_recv_action() which could run in parallel and set the bits checked by smc_cdc_rxed_any_close_or_senddone() before the receive is processed. In that case we could return from receive with no data, although data is available. The same applies to smc_rx_wait(). Fix this by checking for RCV_SHUTDOWN only, which is set in smc_cdc_msg_recv_action() after the receive was actually processed. Fixes: 952310ccf2d8 ("smc: receive data from RMBE") Reviewed-by: Ursula Braun Signed-off-by: Karsten Graul Signed-off-by: Jakub Kicinski --- net/smc/smc_rx.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/net/smc/smc_rx.c b/net/smc/smc_rx.c index 413a6abf227e..000002642288 100644 --- a/net/smc/smc_rx.c +++ b/net/smc/smc_rx.c @@ -211,8 +211,7 @@ int smc_rx_wait(struct smc_sock *smc, long *timeo, rc = sk_wait_event(sk, timeo, sk->sk_err || sk->sk_shutdown & RCV_SHUTDOWN || - fcrit(conn) || - smc_cdc_rxed_any_close_or_senddone(conn), + fcrit(conn), &wait); remove_wait_queue(sk_sleep(sk), &wait); sk_clear_bit(SOCKWQ_ASYNC_WAITDATA, sk); @@ -310,7 +309,6 @@ int smc_rx_recvmsg(struct smc_sock *smc, struct msghdr *msg, smc_rx_update_cons(smc, 0); if (sk->sk_shutdown & RCV_SHUTDOWN || - smc_cdc_rxed_any_close_or_senddone(conn) || conn->local_tx_ctrl.conn_state_flags.peer_conn_abort) break; From 107529e31a87acd475ff6a0f82745821b8f70fec Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Thu, 10 Oct 2019 10:16:11 +0200 Subject: [PATCH 074/145] net/smc: receive pending data after RCV_SHUTDOWN smc_rx_recvmsg() first checks if data is available, and then if RCV_SHUTDOWN is set. There is a race when smc_cdc_msg_recv_action() runs in between these 2 checks, receives data and sets RCV_SHUTDOWN. In that case smc_rx_recvmsg() would return from receive without to process the available data. Fix that with a final check for data available if RCV_SHUTDOWN is set. Move the check for data into a function and call it twice. And use the existing helper smc_rx_data_available(). Fixes: 952310ccf2d8 ("smc: receive data from RMBE") Reviewed-by: Ursula Braun Signed-off-by: Karsten Graul Signed-off-by: Jakub Kicinski --- net/smc/smc_rx.c | 25 ++++++++++++++++++++----- 1 file changed, 20 insertions(+), 5 deletions(-) diff --git a/net/smc/smc_rx.c b/net/smc/smc_rx.c index 000002642288..97e8369002d7 100644 --- a/net/smc/smc_rx.c +++ b/net/smc/smc_rx.c @@ -261,6 +261,18 @@ static int smc_rx_recv_urg(struct smc_sock *smc, struct msghdr *msg, int len, return -EAGAIN; } +static bool smc_rx_recvmsg_data_available(struct smc_sock *smc) +{ + struct smc_connection *conn = &smc->conn; + + if (smc_rx_data_available(conn)) + return true; + else if (conn->urg_state == SMC_URG_VALID) + /* we received a single urgent Byte - skip */ + smc_rx_update_cons(smc, 0); + return false; +} + /* smc_rx_recvmsg - receive data from RMBE * @msg: copy data to receive buffer * @pipe: copy data to pipe if set - indicates splice() call @@ -302,15 +314,18 @@ int smc_rx_recvmsg(struct smc_sock *smc, struct msghdr *msg, if (read_done >= target || (pipe && read_done)) break; - if (atomic_read(&conn->bytes_to_rcv)) + if (smc_rx_recvmsg_data_available(smc)) goto copy; - else if (conn->urg_state == SMC_URG_VALID) - /* we received a single urgent Byte - skip */ - smc_rx_update_cons(smc, 0); if (sk->sk_shutdown & RCV_SHUTDOWN || - conn->local_tx_ctrl.conn_state_flags.peer_conn_abort) + conn->local_tx_ctrl.conn_state_flags.peer_conn_abort) { + /* smc_cdc_msg_recv_action() could have run after + * above smc_rx_recvmsg_data_available() + */ + if (smc_rx_recvmsg_data_available(smc)) + goto copy; break; + } if (read_done) { if (sk->sk_err || From f0308fb0708078d6c1d8a4d533941a7a191af634 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 10 Oct 2019 15:52:34 +0100 Subject: [PATCH 075/145] rxrpc: Fix possible NULL pointer access in ICMP handling If an ICMP packet comes in on the UDP socket backing an AF_RXRPC socket as the UDP socket is being shut down, rxrpc_error_report() may get called to deal with it after sk_user_data on the UDP socket has been cleared, leading to a NULL pointer access when this local endpoint record gets accessed. Fix this by just returning immediately if sk_user_data was NULL. The oops looks like the following: #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page ... RIP: 0010:rxrpc_error_report+0x1bd/0x6a9 ... Call Trace: ? sock_queue_err_skb+0xbd/0xde ? __udp4_lib_err+0x313/0x34d __udp4_lib_err+0x313/0x34d icmp_unreach+0x1ee/0x207 icmp_rcv+0x25b/0x28f ip_protocol_deliver_rcu+0x95/0x10e ip_local_deliver+0xe9/0x148 __netif_receive_skb_one_core+0x52/0x6e process_backlog+0xdc/0x177 net_rx_action+0xf9/0x270 __do_softirq+0x1b6/0x39a ? smpboot_register_percpu_thread+0xce/0xce run_ksoftirqd+0x1d/0x42 smpboot_thread_fn+0x19e/0x1b3 kthread+0xf1/0xf6 ? kthread_delayed_work_timer_fn+0x83/0x83 ret_from_fork+0x24/0x30 Fixes: 17926a79320a ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both") Reported-by: syzbot+611164843bd48cc2190c@syzkaller.appspotmail.com Signed-off-by: David Howells Signed-off-by: David S. Miller --- net/rxrpc/peer_event.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index c97ebdc043e4..61451281d74a 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -151,6 +151,9 @@ void rxrpc_error_report(struct sock *sk) struct rxrpc_peer *peer; struct sk_buff *skb; + if (unlikely(!local)) + return; + _enter("%p{%d}", sk, local->debug_id); /* Clear the outstanding error value on the socket so that it doesn't From d983ea6f16b835dcde2ee9a58a1e764ce68bfccc Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 10 Oct 2019 20:17:38 -0700 Subject: [PATCH 076/145] tcp: add rcu protection around tp->fastopen_rsk Both tcp_v4_err() and tcp_v6_err() do the following operations while they do not own the socket lock : fastopen = tp->fastopen_rsk; snd_una = fastopen ? tcp_rsk(fastopen)->snt_isn : tp->snd_una; The problem is that without appropriate barrier, the compiler might reload tp->fastopen_rsk and trigger a NULL deref. request sockets are protected by RCU, we can simply add the missing annotations and barriers to solve the issue. Fixes: 168a8f58059a ("tcp: TCP Fast Open Server - main code path") Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/tcp.h | 6 +++--- net/core/request_sock.c | 2 +- net/ipv4/inet_connection_sock.c | 4 ++-- net/ipv4/tcp.c | 11 ++++++++--- net/ipv4/tcp_fastopen.c | 2 +- net/ipv4/tcp_input.c | 13 +++++++++---- net/ipv4/tcp_ipv4.c | 4 ++-- net/ipv4/tcp_minisocks.c | 2 +- net/ipv4/tcp_output.c | 2 +- net/ipv4/tcp_timer.c | 11 ++++++----- net/ipv6/tcp_ipv6.c | 2 +- 11 files changed, 35 insertions(+), 24 deletions(-) diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 99617e528ea2..668e25a76d69 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -393,7 +393,7 @@ struct tcp_sock { /* fastopen_rsk points to request_sock that resulted in this big * socket. Used to retransmit SYNACKs etc. */ - struct request_sock *fastopen_rsk; + struct request_sock __rcu *fastopen_rsk; u32 *saved_syn; }; @@ -447,8 +447,8 @@ static inline struct tcp_timewait_sock *tcp_twsk(const struct sock *sk) static inline bool tcp_passive_fastopen(const struct sock *sk) { - return (sk->sk_state == TCP_SYN_RECV && - tcp_sk(sk)->fastopen_rsk != NULL); + return sk->sk_state == TCP_SYN_RECV && + rcu_access_pointer(tcp_sk(sk)->fastopen_rsk) != NULL; } static inline void fastopen_queue_tune(struct sock *sk, int backlog) diff --git a/net/core/request_sock.c b/net/core/request_sock.c index c9bb00008528..f35c2e998406 100644 --- a/net/core/request_sock.c +++ b/net/core/request_sock.c @@ -96,7 +96,7 @@ void reqsk_fastopen_remove(struct sock *sk, struct request_sock *req, fastopenq = &inet_csk(lsk)->icsk_accept_queue.fastopenq; - tcp_sk(sk)->fastopen_rsk = NULL; + RCU_INIT_POINTER(tcp_sk(sk)->fastopen_rsk, NULL); spin_lock_bh(&fastopenq->lock); fastopenq->qlen--; tcp_rsk(req)->tfo_listener = false; diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index dbcf34ec8dd2..eb30fc1770de 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -906,7 +906,7 @@ static void inet_child_forget(struct sock *sk, struct request_sock *req, percpu_counter_inc(sk->sk_prot->orphan_count); if (sk->sk_protocol == IPPROTO_TCP && tcp_rsk(req)->tfo_listener) { - BUG_ON(tcp_sk(child)->fastopen_rsk != req); + BUG_ON(rcu_access_pointer(tcp_sk(child)->fastopen_rsk) != req); BUG_ON(sk != req->rsk_listener); /* Paranoid, to prevent race condition if @@ -915,7 +915,7 @@ static void inet_child_forget(struct sock *sk, struct request_sock *req, * Also to satisfy an assertion in * tcp_v4_destroy_sock(). */ - tcp_sk(child)->fastopen_rsk = NULL; + RCU_INIT_POINTER(tcp_sk(child)->fastopen_rsk, NULL); } inet_csk_destroy_sock(child); } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 8781a92ea4b6..c59d0bd29c5c 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -543,7 +543,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) /* Connected or passive Fast Open socket? */ if (state != TCP_SYN_SENT && - (state != TCP_SYN_RECV || tp->fastopen_rsk)) { + (state != TCP_SYN_RECV || rcu_access_pointer(tp->fastopen_rsk))) { int target = sock_rcvlowat(sk, 0, INT_MAX); if (tp->urg_seq == tp->copied_seq && @@ -2487,7 +2487,10 @@ adjudge_to_death: } if (sk->sk_state == TCP_CLOSE) { - struct request_sock *req = tcp_sk(sk)->fastopen_rsk; + struct request_sock *req; + + req = rcu_dereference_protected(tcp_sk(sk)->fastopen_rsk, + lockdep_sock_is_held(sk)); /* We could get here with a non-NULL req if the socket is * aborted (e.g., closed with unread data) before 3WHS * finishes. @@ -3831,8 +3834,10 @@ EXPORT_SYMBOL(tcp_md5_hash_key); void tcp_done(struct sock *sk) { - struct request_sock *req = tcp_sk(sk)->fastopen_rsk; + struct request_sock *req; + req = rcu_dereference_protected(tcp_sk(sk)->fastopen_rsk, + lockdep_sock_is_held(sk)); if (sk->sk_state == TCP_SYN_SENT || sk->sk_state == TCP_SYN_RECV) TCP_INC_STATS(sock_net(sk), TCP_MIB_ATTEMPTFAILS); diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 3fd451271a70..a915ade0c818 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -253,7 +253,7 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk, */ tp = tcp_sk(child); - tp->fastopen_rsk = req; + rcu_assign_pointer(tp->fastopen_rsk, req); tcp_rsk(req)->tfo_listener = true; /* RFC1323: The window in SYN & SYN/ACK segments is never diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 3578357abe30..5f9b102c3b55 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2666,7 +2666,7 @@ static void tcp_process_loss(struct sock *sk, int flag, int num_dupack, struct tcp_sock *tp = tcp_sk(sk); bool recovered = !before(tp->snd_una, tp->high_seq); - if ((flag & FLAG_SND_UNA_ADVANCED || tp->fastopen_rsk) && + if ((flag & FLAG_SND_UNA_ADVANCED || rcu_access_pointer(tp->fastopen_rsk)) && tcp_try_undo_loss(sk, false)) return; @@ -2990,7 +2990,7 @@ void tcp_rearm_rto(struct sock *sk) /* If the retrans timer is currently being used by Fast Open * for SYN-ACK retrans purpose, stay put. */ - if (tp->fastopen_rsk) + if (rcu_access_pointer(tp->fastopen_rsk)) return; if (!tp->packets_out) { @@ -6087,6 +6087,8 @@ reset_and_undo: static void tcp_rcv_synrecv_state_fastopen(struct sock *sk) { + struct request_sock *req; + tcp_try_undo_loss(sk, false); /* Reset rtx states to prevent spurious retransmits_timed_out() */ @@ -6096,7 +6098,9 @@ static void tcp_rcv_synrecv_state_fastopen(struct sock *sk) /* Once we leave TCP_SYN_RECV or TCP_FIN_WAIT_1, * we no longer need req so release it. */ - reqsk_fastopen_remove(sk, tcp_sk(sk)->fastopen_rsk, false); + req = rcu_dereference_protected(tcp_sk(sk)->fastopen_rsk, + lockdep_sock_is_held(sk)); + reqsk_fastopen_remove(sk, req, false); /* Re-arm the timer because data may have been sent out. * This is similar to the regular data transmission case @@ -6171,7 +6175,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) tcp_mstamp_refresh(tp); tp->rx_opt.saw_tstamp = 0; - req = tp->fastopen_rsk; + req = rcu_dereference_protected(tp->fastopen_rsk, + lockdep_sock_is_held(sk)); if (req) { bool req_stolen; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 492bf6a6b023..ffa366099eb2 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -478,7 +478,7 @@ int tcp_v4_err(struct sk_buff *icmp_skb, u32 info) icsk = inet_csk(sk); tp = tcp_sk(sk); /* XXX (TFO) - tp->snd_una should be ISN (tcp_create_openreq_child() */ - fastopen = tp->fastopen_rsk; + fastopen = rcu_dereference(tp->fastopen_rsk); snd_una = fastopen ? tcp_rsk(fastopen)->snt_isn : tp->snd_una; if (sk->sk_state != TCP_LISTEN && !between(seq, snd_una, tp->snd_nxt)) { @@ -2121,7 +2121,7 @@ void tcp_v4_destroy_sock(struct sock *sk) if (inet_csk(sk)->icsk_bind_hash) inet_put_port(sk); - BUG_ON(tp->fastopen_rsk); + BUG_ON(rcu_access_pointer(tp->fastopen_rsk)); /* If socket is aborted during connect operation */ tcp_free_fastopen_req(tp); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index bb140a5db8c0..5401dbd39c8f 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -541,7 +541,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, newtp->rx_opt.mss_clamp = req->mss; tcp_ecn_openreq_child(newtp, req); newtp->fastopen_req = NULL; - newtp->fastopen_rsk = NULL; + RCU_INIT_POINTER(newtp->fastopen_rsk, NULL); __TCP_INC_STATS(sock_net(sk), TCP_MIB_PASSIVEOPENS); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index fec6d67bfd14..84ae4d1449ea 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2482,7 +2482,7 @@ bool tcp_schedule_loss_probe(struct sock *sk, bool advancing_rto) /* Don't do any loss probe on a Fast Open connection before 3WHS * finishes. */ - if (tp->fastopen_rsk) + if (rcu_access_pointer(tp->fastopen_rsk)) return false; early_retrans = sock_net(sk)->ipv4.sysctl_tcp_early_retrans; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 05be564414e9..dd5a6317a801 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -386,15 +386,13 @@ abort: tcp_write_err(sk); * Timer for Fast Open socket to retransmit SYNACK. Note that the * sk here is the child socket, not the parent (listener) socket. */ -static void tcp_fastopen_synack_timer(struct sock *sk) +static void tcp_fastopen_synack_timer(struct sock *sk, struct request_sock *req) { struct inet_connection_sock *icsk = inet_csk(sk); int max_retries = icsk->icsk_syn_retries ? : sock_net(sk)->ipv4.sysctl_tcp_synack_retries + 1; /* add one more retry for fastopen */ struct tcp_sock *tp = tcp_sk(sk); - struct request_sock *req; - req = tcp_sk(sk)->fastopen_rsk; req->rsk_ops->syn_ack_timeout(req); if (req->num_timeout >= max_retries) { @@ -435,11 +433,14 @@ void tcp_retransmit_timer(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct net *net = sock_net(sk); struct inet_connection_sock *icsk = inet_csk(sk); + struct request_sock *req; - if (tp->fastopen_rsk) { + req = rcu_dereference_protected(tp->fastopen_rsk, + lockdep_sock_is_held(sk)); + if (req) { WARN_ON_ONCE(sk->sk_state != TCP_SYN_RECV && sk->sk_state != TCP_FIN_WAIT1); - tcp_fastopen_synack_timer(sk); + tcp_fastopen_synack_timer(sk, req); /* Before we receive ACK to our SYN-ACK don't retransmit * anything else (e.g., data or FIN segments). */ diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index e3d9f4559c99..45a95e032bdf 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -406,7 +406,7 @@ static int tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, tp = tcp_sk(sk); /* XXX (TFO) - tp->snd_una should be ISN (tcp_create_openreq_child() */ - fastopen = tp->fastopen_rsk; + fastopen = rcu_dereference(tp->fastopen_rsk); snd_una = fastopen ? tcp_rsk(fastopen)->snt_isn : tp->snd_una; if (sk->sk_state != TCP_LISTEN && !between(seq, snd_una, tp->snd_nxt)) { From dba7d9b8c739df27ff3a234c81d6c6b23e3986fa Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 10 Oct 2019 20:17:39 -0700 Subject: [PATCH 077/145] tcp: annotate tp->rcv_nxt lockless reads There are few places where we fetch tp->rcv_nxt while this field can change from IRQ or other cpu. We need to add READ_ONCE() annotations, and also make sure write sides use corresponding WRITE_ONCE() to avoid store-tearing. Note that tcp_inq_hint() was already using READ_ONCE(tp->rcv_nxt) syzbot reported : BUG: KCSAN: data-race in tcp_poll / tcp_queue_rcv write to 0xffff888120425770 of 4 bytes by interrupt on cpu 0: tcp_rcv_nxt_update net/ipv4/tcp_input.c:3365 [inline] tcp_queue_rcv+0x180/0x380 net/ipv4/tcp_input.c:4638 tcp_rcv_established+0xbf1/0xf50 net/ipv4/tcp_input.c:5616 tcp_v4_do_rcv+0x381/0x4e0 net/ipv4/tcp_ipv4.c:1542 tcp_v4_rcv+0x1a03/0x1bf0 net/ipv4/tcp_ipv4.c:1923 ip_protocol_deliver_rcu+0x51/0x470 net/ipv4/ip_input.c:204 ip_local_deliver_finish+0x110/0x140 net/ipv4/ip_input.c:231 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_local_deliver+0x133/0x210 net/ipv4/ip_input.c:252 dst_input include/net/dst.h:442 [inline] ip_rcv_finish+0x121/0x160 net/ipv4/ip_input.c:413 NF_HOOK include/linux/netfilter.h:305 [inline] NF_HOOK include/linux/netfilter.h:299 [inline] ip_rcv+0x18f/0x1a0 net/ipv4/ip_input.c:523 __netif_receive_skb_one_core+0xa7/0xe0 net/core/dev.c:5004 __netif_receive_skb+0x37/0xf0 net/core/dev.c:5118 netif_receive_skb_internal+0x59/0x190 net/core/dev.c:5208 napi_skb_finish net/core/dev.c:5671 [inline] napi_gro_receive+0x28f/0x330 net/core/dev.c:5704 receive_buf+0x284/0x30b0 drivers/net/virtio_net.c:1061 read to 0xffff888120425770 of 4 bytes by task 7254 on cpu 1: tcp_stream_is_readable net/ipv4/tcp.c:480 [inline] tcp_poll+0x204/0x6b0 net/ipv4/tcp.c:554 sock_poll+0xed/0x250 net/socket.c:1256 vfs_poll include/linux/poll.h:90 [inline] ep_item_poll.isra.0+0x90/0x190 fs/eventpoll.c:892 ep_send_events_proc+0x113/0x5c0 fs/eventpoll.c:1749 ep_scan_ready_list.constprop.0+0x189/0x500 fs/eventpoll.c:704 ep_send_events fs/eventpoll.c:1793 [inline] ep_poll+0xe3/0x900 fs/eventpoll.c:1930 do_epoll_wait+0x162/0x180 fs/eventpoll.c:2294 __do_sys_epoll_pwait fs/eventpoll.c:2325 [inline] __se_sys_epoll_pwait fs/eventpoll.c:2311 [inline] __x64_sys_epoll_pwait+0xcd/0x170 fs/eventpoll.c:2311 do_syscall_64+0xcf/0x2f0 arch/x86/entry/common.c:296 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Reported by Kernel Concurrency Sanitizer on: CPU: 1 PID: 7254 Comm: syz-fuzzer Not tainted 5.3.0+ #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Signed-off-by: Eric Dumazet Reported-by: syzbot Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 4 ++-- net/ipv4/tcp_diag.c | 2 +- net/ipv4/tcp_input.c | 6 +++--- net/ipv4/tcp_ipv4.c | 3 ++- net/ipv4/tcp_minisocks.c | 7 +++++-- net/ipv6/tcp_ipv6.c | 3 ++- 6 files changed, 15 insertions(+), 10 deletions(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index c59d0bd29c5c..883ee863db43 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -477,7 +477,7 @@ static void tcp_tx_timestamp(struct sock *sk, u16 tsflags) static inline bool tcp_stream_is_readable(const struct tcp_sock *tp, int target, struct sock *sk) { - return (tp->rcv_nxt - tp->copied_seq >= target) || + return (READ_ONCE(tp->rcv_nxt) - tp->copied_seq >= target) || (sk->sk_prot->stream_memory_read ? sk->sk_prot->stream_memory_read(sk) : false); } @@ -2935,7 +2935,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, else if (tp->repair_queue == TCP_SEND_QUEUE) tp->write_seq = val; else if (tp->repair_queue == TCP_RECV_QUEUE) - tp->rcv_nxt = val; + WRITE_ONCE(tp->rcv_nxt, val); else err = -EINVAL; break; diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 81a8221d650a..cd219161f106 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -26,7 +26,7 @@ static void tcp_diag_get_info(struct sock *sk, struct inet_diag_msg *r, } else if (sk->sk_type == SOCK_STREAM) { const struct tcp_sock *tp = tcp_sk(sk); - r->idiag_rqueue = max_t(int, tp->rcv_nxt - tp->copied_seq, 0); + r->idiag_rqueue = max_t(int, READ_ONCE(tp->rcv_nxt) - tp->copied_seq, 0); r->idiag_wqueue = tp->write_seq - tp->snd_una; } if (info) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5f9b102c3b55..5b7c8768ed5f 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3362,7 +3362,7 @@ static void tcp_rcv_nxt_update(struct tcp_sock *tp, u32 seq) sock_owned_by_me((struct sock *)tp); tp->bytes_received += delta; - tp->rcv_nxt = seq; + WRITE_ONCE(tp->rcv_nxt, seq); } /* Update our send window. @@ -5932,7 +5932,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, /* Ok.. it's good. Set up sequence numbers and * move to established. */ - tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1; + WRITE_ONCE(tp->rcv_nxt, TCP_SKB_CB(skb)->seq + 1); tp->rcv_wup = TCP_SKB_CB(skb)->seq + 1; /* RFC1323: The window in SYN & SYN/ACK segments is @@ -6035,7 +6035,7 @@ discard: tp->tcp_header_len = sizeof(struct tcphdr); } - tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1; + WRITE_ONCE(tp->rcv_nxt, TCP_SKB_CB(skb)->seq + 1); tp->copied_seq = tp->rcv_nxt; tp->rcv_wup = TCP_SKB_CB(skb)->seq + 1; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index ffa366099eb2..5089dd6bee0f 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2455,7 +2455,8 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i) /* Because we don't lock the socket, * we might find a transient negative value. */ - rx_queue = max_t(int, tp->rcv_nxt - tp->copied_seq, 0); + rx_queue = max_t(int, READ_ONCE(tp->rcv_nxt) - + tp->copied_seq, 0); seq_printf(f, "%4d: %08X:%04X %08X:%04X %02X %08X:%08X %02X:%08lX " "%08X %5u %8d %lu %d %pK %lu %lu %u %u %d", diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 5401dbd39c8f..adc6ce486a38 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -462,6 +462,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, struct tcp_request_sock *treq = tcp_rsk(req); struct inet_connection_sock *newicsk; struct tcp_sock *oldtp, *newtp; + u32 seq; if (!newsk) return NULL; @@ -475,8 +476,10 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, /* Now setup tcp_sock */ newtp->pred_flags = 0; - newtp->rcv_wup = newtp->copied_seq = - newtp->rcv_nxt = treq->rcv_isn + 1; + seq = treq->rcv_isn + 1; + newtp->rcv_wup = seq; + newtp->copied_seq = seq; + WRITE_ONCE(newtp->rcv_nxt, seq); newtp->segs_in = 1; newtp->snd_sml = newtp->snd_una = diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 45a95e032bdf..89ea0a7018b5 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1895,7 +1895,8 @@ static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i) /* Because we don't lock the socket, * we might find a transient negative value. */ - rx_queue = max_t(int, tp->rcv_nxt - tp->copied_seq, 0); + rx_queue = max_t(int, READ_ONCE(tp->rcv_nxt) - + tp->copied_seq, 0); seq_printf(seq, "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X " From 7db48e983930285b765743ebd665aecf9850582b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 10 Oct 2019 20:17:40 -0700 Subject: [PATCH 078/145] tcp: annotate tp->copied_seq lockless reads There are few places where we fetch tp->copied_seq while this field can change from IRQ or other cpu. We need to add READ_ONCE() annotations, and also make sure write sides use corresponding WRITE_ONCE() to avoid store-tearing. Note that tcp_inq_hint() was already using READ_ONCE(tp->copied_seq) Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 20 ++++++++++---------- net/ipv4/tcp_diag.c | 3 ++- net/ipv4/tcp_input.c | 6 +++--- net/ipv4/tcp_ipv4.c | 2 +- net/ipv4/tcp_minisocks.c | 2 +- net/ipv4/tcp_output.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 7 files changed, 19 insertions(+), 18 deletions(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 883ee863db43..c322ad071e17 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -477,7 +477,7 @@ static void tcp_tx_timestamp(struct sock *sk, u16 tsflags) static inline bool tcp_stream_is_readable(const struct tcp_sock *tp, int target, struct sock *sk) { - return (READ_ONCE(tp->rcv_nxt) - tp->copied_seq >= target) || + return (READ_ONCE(tp->rcv_nxt) - READ_ONCE(tp->copied_seq) >= target) || (sk->sk_prot->stream_memory_read ? sk->sk_prot->stream_memory_read(sk) : false); } @@ -546,7 +546,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) (state != TCP_SYN_RECV || rcu_access_pointer(tp->fastopen_rsk))) { int target = sock_rcvlowat(sk, 0, INT_MAX); - if (tp->urg_seq == tp->copied_seq && + if (tp->urg_seq == READ_ONCE(tp->copied_seq) && !sock_flag(sk, SOCK_URGINLINE) && tp->urg_data) target++; @@ -607,7 +607,7 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg) unlock_sock_fast(sk, slow); break; case SIOCATMARK: - answ = tp->urg_data && tp->urg_seq == tp->copied_seq; + answ = tp->urg_data && tp->urg_seq == READ_ONCE(tp->copied_seq); break; case SIOCOUTQ: if (sk->sk_state == TCP_LISTEN) @@ -1668,9 +1668,9 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, sk_eat_skb(sk, skb); if (!desc->count) break; - tp->copied_seq = seq; + WRITE_ONCE(tp->copied_seq, seq); } - tp->copied_seq = seq; + WRITE_ONCE(tp->copied_seq, seq); tcp_rcv_space_adjust(sk); @@ -1819,7 +1819,7 @@ static int tcp_zerocopy_receive(struct sock *sk, out: up_read(¤t->mm->mmap_sem); if (length) { - tp->copied_seq = seq; + WRITE_ONCE(tp->copied_seq, seq); tcp_rcv_space_adjust(sk); /* Clean up data we have read: This will do ACK frames. */ @@ -2117,7 +2117,7 @@ found_ok_skb: if (urg_offset < used) { if (!urg_offset) { if (!sock_flag(sk, SOCK_URGINLINE)) { - ++*seq; + WRITE_ONCE(*seq, *seq + 1); urg_hole++; offset++; used--; @@ -2139,7 +2139,7 @@ found_ok_skb: } } - *seq += used; + WRITE_ONCE(*seq, *seq + used); copied += used; len -= used; @@ -2166,7 +2166,7 @@ skip_copy: found_fin_ok: /* Process the FIN. */ - ++*seq; + WRITE_ONCE(*seq, *seq + 1); if (!(flags & MSG_PEEK)) sk_eat_skb(sk, skb); break; @@ -2588,7 +2588,7 @@ int tcp_disconnect(struct sock *sk, int flags) __kfree_skb(sk->sk_rx_skb_cache); sk->sk_rx_skb_cache = NULL; } - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); tp->urg_data = 0; tcp_write_queue_purge(sk); tcp_fastopen_active_disable_ofo_check(sk); diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index cd219161f106..66273c8a55c2 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -26,7 +26,8 @@ static void tcp_diag_get_info(struct sock *sk, struct inet_diag_msg *r, } else if (sk->sk_type == SOCK_STREAM) { const struct tcp_sock *tp = tcp_sk(sk); - r->idiag_rqueue = max_t(int, READ_ONCE(tp->rcv_nxt) - tp->copied_seq, 0); + r->idiag_rqueue = max_t(int, READ_ONCE(tp->rcv_nxt) - + READ_ONCE(tp->copied_seq), 0); r->idiag_wqueue = tp->write_seq - tp->snd_una; } if (info) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5b7c8768ed5f..a30aae3a6a18 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5961,7 +5961,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, /* Remember, tcp_poll() does not lock socket! * Change state from SYN-SENT only after copied_seq * is initialized. */ - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); smc_check_reset_syn(tp); @@ -6036,7 +6036,7 @@ discard: } WRITE_ONCE(tp->rcv_nxt, TCP_SKB_CB(skb)->seq + 1); - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); tp->rcv_wup = TCP_SKB_CB(skb)->seq + 1; /* RFC1323: The window in SYN & SYN/ACK segments is @@ -6216,7 +6216,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) tcp_try_undo_spurious_syn(sk); tp->retrans_stamp = 0; tcp_init_transfer(sk, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); } smp_mb(); tcp_set_state(sk, TCP_ESTABLISHED); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 5089dd6bee0f..39560f482e0b 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2456,7 +2456,7 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i) * we might find a transient negative value. */ rx_queue = max_t(int, READ_ONCE(tp->rcv_nxt) - - tp->copied_seq, 0); + READ_ONCE(tp->copied_seq), 0); seq_printf(f, "%4d: %08X:%04X %08X:%04X %02X %08X:%08X %02X:%08lX " "%08X %5u %8d %lu %d %pK %lu %lu %u %u %d", diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index adc6ce486a38..c4731d26ab4a 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -478,7 +478,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, seq = treq->rcv_isn + 1; newtp->rcv_wup = seq; - newtp->copied_seq = seq; + WRITE_ONCE(newtp->copied_seq, seq); WRITE_ONCE(newtp->rcv_nxt, seq); newtp->segs_in = 1; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 84ae4d1449ea..7dda12720169 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -3433,7 +3433,7 @@ static void tcp_connect_init(struct sock *sk) else tp->rcv_tstamp = tcp_jiffies32; tp->rcv_wup = tp->rcv_nxt; - tp->copied_seq = tp->rcv_nxt; + WRITE_ONCE(tp->copied_seq, tp->rcv_nxt); inet_csk(sk)->icsk_rto = tcp_timeout_init(sk); inet_csk(sk)->icsk_retransmits = 0; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 89ea0a7018b5..a62c7042fc4a 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1896,7 +1896,7 @@ static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i) * we might find a transient negative value. */ rx_queue = max_t(int, READ_ONCE(tp->rcv_nxt) - - tp->copied_seq, 0); + READ_ONCE(tp->copied_seq), 0); seq_printf(seq, "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X " From 0f31746452e6793ad6271337438af8f4defb8940 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 10 Oct 2019 20:17:41 -0700 Subject: [PATCH 079/145] tcp: annotate tp->write_seq lockless reads There are few places where we fetch tp->write_seq while this field can change from IRQ or other cpu. We need to add READ_ONCE() annotations, and also make sure write sides use corresponding WRITE_ONCE() to avoid store-tearing. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 2 +- net/ipv4/tcp.c | 20 ++++++++++++-------- net/ipv4/tcp_diag.c | 2 +- net/ipv4/tcp_ipv4.c | 21 ++++++++++++--------- net/ipv4/tcp_minisocks.c | 2 +- net/ipv4/tcp_output.c | 4 ++-- net/ipv6/tcp_ipv6.c | 13 +++++++------ 7 files changed, 36 insertions(+), 28 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index 35f6f7e0fdc2..8e7c3f6801a9 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1917,7 +1917,7 @@ static inline u32 tcp_notsent_lowat(const struct tcp_sock *tp) static inline bool tcp_stream_memory_free(const struct sock *sk, int wake) { const struct tcp_sock *tp = tcp_sk(sk); - u32 notsent_bytes = tp->write_seq - tp->snd_nxt; + u32 notsent_bytes = READ_ONCE(tp->write_seq) - tp->snd_nxt; return (notsent_bytes << wake) < tcp_notsent_lowat(tp); } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index c322ad071e17..96dd65cbeb85 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -616,7 +616,7 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg) if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) answ = 0; else - answ = tp->write_seq - tp->snd_una; + answ = READ_ONCE(tp->write_seq) - tp->snd_una; break; case SIOCOUTQNSD: if (sk->sk_state == TCP_LISTEN) @@ -625,7 +625,7 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg) if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) answ = 0; else - answ = tp->write_seq - tp->snd_nxt; + answ = READ_ONCE(tp->write_seq) - tp->snd_nxt; break; default: return -ENOIOCTLCMD; @@ -1035,7 +1035,7 @@ new_segment: sk->sk_wmem_queued += copy; sk_mem_charge(sk, copy); skb->ip_summed = CHECKSUM_PARTIAL; - tp->write_seq += copy; + WRITE_ONCE(tp->write_seq, tp->write_seq + copy); TCP_SKB_CB(skb)->end_seq += copy; tcp_skb_pcount_set(skb, 0); @@ -1362,7 +1362,7 @@ new_segment: if (!copied) TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_PSH; - tp->write_seq += copy; + WRITE_ONCE(tp->write_seq, tp->write_seq + copy); TCP_SKB_CB(skb)->end_seq += copy; tcp_skb_pcount_set(skb, 0); @@ -2562,6 +2562,7 @@ int tcp_disconnect(struct sock *sk, int flags) struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); int old_state = sk->sk_state; + u32 seq; if (old_state != TCP_CLOSE) tcp_set_state(sk, TCP_CLOSE); @@ -2604,9 +2605,12 @@ int tcp_disconnect(struct sock *sk, int flags) tp->srtt_us = 0; tp->mdev_us = jiffies_to_usecs(TCP_TIMEOUT_INIT); tp->rcv_rtt_last_tsecr = 0; - tp->write_seq += tp->max_window + 2; - if (tp->write_seq == 0) - tp->write_seq = 1; + + seq = tp->write_seq + tp->max_window + 2; + if (!seq) + seq = 1; + WRITE_ONCE(tp->write_seq, seq); + icsk->icsk_backoff = 0; tp->snd_cwnd = 2; icsk->icsk_probes_out = 0; @@ -2933,7 +2937,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, if (sk->sk_state != TCP_CLOSE) err = -EPERM; else if (tp->repair_queue == TCP_SEND_QUEUE) - tp->write_seq = val; + WRITE_ONCE(tp->write_seq, val); else if (tp->repair_queue == TCP_RECV_QUEUE) WRITE_ONCE(tp->rcv_nxt, val); else diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 66273c8a55c2..549506162dde 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -28,7 +28,7 @@ static void tcp_diag_get_info(struct sock *sk, struct inet_diag_msg *r, r->idiag_rqueue = max_t(int, READ_ONCE(tp->rcv_nxt) - READ_ONCE(tp->copied_seq), 0); - r->idiag_wqueue = tp->write_seq - tp->snd_una; + r->idiag_wqueue = READ_ONCE(tp->write_seq) - tp->snd_una; } if (info) tcp_get_info(sk, info); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 39560f482e0b..6be568334848 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -164,9 +164,11 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp) * without appearing to create any others. */ if (likely(!tp->repair)) { - tp->write_seq = tcptw->tw_snd_nxt + 65535 + 2; - if (tp->write_seq == 0) - tp->write_seq = 1; + u32 seq = tcptw->tw_snd_nxt + 65535 + 2; + + if (!seq) + seq = 1; + WRITE_ONCE(tp->write_seq, seq); tp->rx_opt.ts_recent = tcptw->tw_ts_recent; tp->rx_opt.ts_recent_stamp = tcptw->tw_ts_recent_stamp; } @@ -253,7 +255,7 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) tp->rx_opt.ts_recent = 0; tp->rx_opt.ts_recent_stamp = 0; if (likely(!tp->repair)) - tp->write_seq = 0; + WRITE_ONCE(tp->write_seq, 0); } inet->inet_dport = usin->sin_port; @@ -291,10 +293,11 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) if (likely(!tp->repair)) { if (!tp->write_seq) - tp->write_seq = secure_tcp_seq(inet->inet_saddr, - inet->inet_daddr, - inet->inet_sport, - usin->sin_port); + WRITE_ONCE(tp->write_seq, + secure_tcp_seq(inet->inet_saddr, + inet->inet_daddr, + inet->inet_sport, + usin->sin_port)); tp->tsoffset = secure_tcp_ts_off(sock_net(sk), inet->inet_saddr, inet->inet_daddr); @@ -2461,7 +2464,7 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i) seq_printf(f, "%4d: %08X:%04X %08X:%04X %02X %08X:%08X %02X:%08lX " "%08X %5u %8d %lu %d %pK %lu %lu %u %u %d", i, src, srcp, dest, destp, state, - tp->write_seq - tp->snd_una, + READ_ONCE(tp->write_seq) - tp->snd_una, rx_queue, timer_active, jiffies_delta_to_clock_t(timer_expires - jiffies), diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index c4731d26ab4a..339944690329 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -498,7 +498,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, newtp->total_retrans = req->num_retrans; tcp_init_xmit_timers(newsk); - newtp->write_seq = newtp->pushed_seq = treq->snt_isn + 1; + WRITE_ONCE(newtp->write_seq, newtp->pushed_seq = treq->snt_isn + 1); if (sock_flag(newsk, SOCK_KEEPOPEN)) inet_csk_reset_keepalive_timer(newsk, diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 7dda12720169..c17c2a78809d 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1196,7 +1196,7 @@ static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb) struct tcp_sock *tp = tcp_sk(sk); /* Advance write_seq and place onto the write_queue. */ - tp->write_seq = TCP_SKB_CB(skb)->end_seq; + WRITE_ONCE(tp->write_seq, TCP_SKB_CB(skb)->end_seq); __skb_header_release(skb); tcp_add_write_queue_tail(sk, skb); sk->sk_wmem_queued += skb->truesize; @@ -3449,7 +3449,7 @@ static void tcp_connect_queue_skb(struct sock *sk, struct sk_buff *skb) __skb_header_release(skb); sk->sk_wmem_queued += skb->truesize; sk_mem_charge(sk, skb->truesize); - tp->write_seq = tcb->end_seq; + WRITE_ONCE(tp->write_seq, tcb->end_seq); tp->packets_out += tcp_skb_pcount(skb); } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index a62c7042fc4a..4804b6dc5e65 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -215,7 +215,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, !ipv6_addr_equal(&sk->sk_v6_daddr, &usin->sin6_addr)) { tp->rx_opt.ts_recent = 0; tp->rx_opt.ts_recent_stamp = 0; - tp->write_seq = 0; + WRITE_ONCE(tp->write_seq, 0); } sk->sk_v6_daddr = usin->sin6_addr; @@ -311,10 +311,11 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, if (likely(!tp->repair)) { if (!tp->write_seq) - tp->write_seq = secure_tcpv6_seq(np->saddr.s6_addr32, - sk->sk_v6_daddr.s6_addr32, - inet->inet_sport, - inet->inet_dport); + WRITE_ONCE(tp->write_seq, + secure_tcpv6_seq(np->saddr.s6_addr32, + sk->sk_v6_daddr.s6_addr32, + inet->inet_sport, + inet->inet_dport)); tp->tsoffset = secure_tcpv6_ts_off(sock_net(sk), np->saddr.s6_addr32, sk->sk_v6_daddr.s6_addr32); @@ -1907,7 +1908,7 @@ static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i) dest->s6_addr32[0], dest->s6_addr32[1], dest->s6_addr32[2], dest->s6_addr32[3], destp, state, - tp->write_seq - tp->snd_una, + READ_ONCE(tp->write_seq) - tp->snd_una, rx_queue, timer_active, jiffies_delta_to_clock_t(timer_expires - jiffies), From e0d694d638dba768b47be31c22e1a9b4f862f561 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 10 Oct 2019 20:17:42 -0700 Subject: [PATCH 080/145] tcp: annotate tp->snd_nxt lockless reads There are few places where we fetch tp->snd_nxt while this field can change from IRQ or other cpu. We need to add READ_ONCE() annotations, and also make sure write sides use corresponding WRITE_ONCE() to avoid store-tearing. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 3 ++- net/ipv4/tcp.c | 3 ++- net/ipv4/tcp_minisocks.c | 6 ++++-- net/ipv4/tcp_output.c | 10 +++++----- 4 files changed, 13 insertions(+), 9 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index 8e7c3f6801a9..e1d08f69fd39 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1917,7 +1917,8 @@ static inline u32 tcp_notsent_lowat(const struct tcp_sock *tp) static inline bool tcp_stream_memory_free(const struct sock *sk, int wake) { const struct tcp_sock *tp = tcp_sk(sk); - u32 notsent_bytes = READ_ONCE(tp->write_seq) - tp->snd_nxt; + u32 notsent_bytes = READ_ONCE(tp->write_seq) - + READ_ONCE(tp->snd_nxt); return (notsent_bytes << wake) < tcp_notsent_lowat(tp); } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 96dd65cbeb85..652568750cb1 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -625,7 +625,8 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg) if ((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV)) answ = 0; else - answ = READ_ONCE(tp->write_seq) - tp->snd_nxt; + answ = READ_ONCE(tp->write_seq) - + READ_ONCE(tp->snd_nxt); break; default: return -ENOIOCTLCMD; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 339944690329..c802bc80c400 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -482,8 +482,10 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, WRITE_ONCE(newtp->rcv_nxt, seq); newtp->segs_in = 1; - newtp->snd_sml = newtp->snd_una = - newtp->snd_nxt = newtp->snd_up = treq->snt_isn + 1; + seq = treq->snt_isn + 1; + newtp->snd_sml = newtp->snd_una = seq; + WRITE_ONCE(newtp->snd_nxt, seq); + newtp->snd_up = seq; INIT_LIST_HEAD(&newtp->tsq_node); INIT_LIST_HEAD(&newtp->tsorted_sent_queue); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index c17c2a78809d..a115a991dfb5 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -67,7 +67,7 @@ static void tcp_event_new_data_sent(struct sock *sk, struct sk_buff *skb) struct tcp_sock *tp = tcp_sk(sk); unsigned int prior_packets = tp->packets_out; - tp->snd_nxt = TCP_SKB_CB(skb)->end_seq; + WRITE_ONCE(tp->snd_nxt, TCP_SKB_CB(skb)->end_seq); __skb_unlink(skb, &sk->sk_write_queue); tcp_rbtree_insert(&sk->tcp_rtx_queue, skb); @@ -3142,7 +3142,7 @@ void tcp_send_fin(struct sock *sk) * if FIN had been sent. This is because retransmit path * does not change tp->snd_nxt. */ - tp->snd_nxt++; + WRITE_ONCE(tp->snd_nxt, tp->snd_nxt + 1); return; } } else { @@ -3426,7 +3426,7 @@ static void tcp_connect_init(struct sock *sk) tp->snd_una = tp->write_seq; tp->snd_sml = tp->write_seq; tp->snd_up = tp->write_seq; - tp->snd_nxt = tp->write_seq; + WRITE_ONCE(tp->snd_nxt, tp->write_seq); if (likely(!tp->repair)) tp->rcv_nxt = 0; @@ -3586,11 +3586,11 @@ int tcp_connect(struct sock *sk) /* We change tp->snd_nxt after the tcp_transmit_skb() call * in order to make this packet get counted in tcpOutSegs. */ - tp->snd_nxt = tp->write_seq; + WRITE_ONCE(tp->snd_nxt, tp->write_seq); tp->pushed_seq = tp->write_seq; buff = tcp_send_head(sk); if (unlikely(buff)) { - tp->snd_nxt = TCP_SKB_CB(buff)->seq; + WRITE_ONCE(tp->snd_nxt, TCP_SKB_CB(buff)->seq); tp->pushed_seq = TCP_SKB_CB(buff)->seq; } TCP_INC_STATS(sock_net(sk), TCP_MIB_ACTIVEOPENS); From d9b55bf7b6788ec0bd1db1acefbc4feb1399144a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 10 Oct 2019 20:17:43 -0700 Subject: [PATCH 081/145] tcp: annotate tp->urg_seq lockless reads There two places where we fetch tp->urg_seq while this field can change from IRQ or other cpu. We need to add READ_ONCE() annotations, and also make sure write side use corresponding WRITE_ONCE() to avoid store-tearing. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 5 +++-- net/ipv4/tcp_input.c | 2 +- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 652568750cb1..577a8c6eef9f 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -546,7 +546,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) (state != TCP_SYN_RECV || rcu_access_pointer(tp->fastopen_rsk))) { int target = sock_rcvlowat(sk, 0, INT_MAX); - if (tp->urg_seq == READ_ONCE(tp->copied_seq) && + if (READ_ONCE(tp->urg_seq) == READ_ONCE(tp->copied_seq) && !sock_flag(sk, SOCK_URGINLINE) && tp->urg_data) target++; @@ -607,7 +607,8 @@ int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg) unlock_sock_fast(sk, slow); break; case SIOCATMARK: - answ = tp->urg_data && tp->urg_seq == READ_ONCE(tp->copied_seq); + answ = tp->urg_data && + READ_ONCE(tp->urg_seq) == READ_ONCE(tp->copied_seq); break; case SIOCOUTQ: if (sk->sk_state == TCP_LISTEN) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index a30aae3a6a18..16342e043ab3 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5356,7 +5356,7 @@ static void tcp_check_urg(struct sock *sk, const struct tcphdr *th) } tp->urg_data = TCP_URG_NOTYET; - tp->urg_seq = ptr; + WRITE_ONCE(tp->urg_seq, ptr); /* Disable header prediction. */ tp->pred_flags = 0; From ebb3b78db7bf842270a46fd4fe7cc45c78fa5ed6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 10 Oct 2019 20:17:44 -0700 Subject: [PATCH 082/145] tcp: annotate sk->sk_rcvbuf lockless reads For the sake of tcp_poll(), there are few places where we fetch sk->sk_rcvbuf while this field can change from IRQ or other cpu. We need to add READ_ONCE() annotations, and also make sure write sides use corresponding WRITE_ONCE() to avoid store-tearing. Note that other transports probably need similar fixes. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 4 ++-- include/trace/events/sock.h | 2 +- net/core/filter.c | 3 ++- net/core/skbuff.c | 2 +- net/core/sock.c | 5 +++-- net/ipv4/tcp.c | 4 ++-- net/ipv4/tcp_input.c | 7 ++++--- 7 files changed, 15 insertions(+), 12 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index e1d08f69fd39..ab4eb5eb5d07 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1380,14 +1380,14 @@ static inline int tcp_win_from_space(const struct sock *sk, int space) /* Note: caller must be prepared to deal with negative returns */ static inline int tcp_space(const struct sock *sk) { - return tcp_win_from_space(sk, sk->sk_rcvbuf - + return tcp_win_from_space(sk, READ_ONCE(sk->sk_rcvbuf) - READ_ONCE(sk->sk_backlog.len) - atomic_read(&sk->sk_rmem_alloc)); } static inline int tcp_full_space(const struct sock *sk) { - return tcp_win_from_space(sk, sk->sk_rcvbuf); + return tcp_win_from_space(sk, READ_ONCE(sk->sk_rcvbuf)); } extern void tcp_openreq_init_rwin(struct request_sock *req, diff --git a/include/trace/events/sock.h b/include/trace/events/sock.h index a0c4b8a30966..f720c32e7dfd 100644 --- a/include/trace/events/sock.h +++ b/include/trace/events/sock.h @@ -82,7 +82,7 @@ TRACE_EVENT(sock_rcvqueue_full, TP_fast_assign( __entry->rmem_alloc = atomic_read(&sk->sk_rmem_alloc); __entry->truesize = skb->truesize; - __entry->sk_rcvbuf = sk->sk_rcvbuf; + __entry->sk_rcvbuf = READ_ONCE(sk->sk_rcvbuf); ), TP_printk("rmem_alloc=%d truesize=%u sk_rcvbuf=%d", diff --git a/net/core/filter.c b/net/core/filter.c index a50c0b6846f2..7deceaeeed7b 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -4252,7 +4252,8 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, case SO_RCVBUF: val = min_t(u32, val, sysctl_rmem_max); sk->sk_userlocks |= SOCK_RCVBUF_LOCK; - sk->sk_rcvbuf = max_t(int, val * 2, SOCK_MIN_RCVBUF); + WRITE_ONCE(sk->sk_rcvbuf, + max_t(int, val * 2, SOCK_MIN_RCVBUF)); break; case SO_SNDBUF: val = min_t(u32, val, sysctl_wmem_max); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 529133611ea2..8c178703467b 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -4415,7 +4415,7 @@ static void skb_set_err_queue(struct sk_buff *skb) int sock_queue_err_skb(struct sock *sk, struct sk_buff *skb) { if (atomic_read(&sk->sk_rmem_alloc) + skb->truesize >= - (unsigned int)sk->sk_rcvbuf) + (unsigned int)READ_ONCE(sk->sk_rcvbuf)) return -ENOMEM; skb_orphan(skb); diff --git a/net/core/sock.c b/net/core/sock.c index 2a053999df11..8c8f61e70141 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -831,7 +831,8 @@ set_rcvbuf: * returning the value we actually used in getsockopt * is the most desirable behavior. */ - sk->sk_rcvbuf = max_t(int, val * 2, SOCK_MIN_RCVBUF); + WRITE_ONCE(sk->sk_rcvbuf, + max_t(int, val * 2, SOCK_MIN_RCVBUF)); break; case SO_RCVBUFFORCE: @@ -3204,7 +3205,7 @@ void sk_get_meminfo(const struct sock *sk, u32 *mem) memset(mem, 0, sizeof(*mem) * SK_MEMINFO_VARS); mem[SK_MEMINFO_RMEM_ALLOC] = sk_rmem_alloc_get(sk); - mem[SK_MEMINFO_RCVBUF] = sk->sk_rcvbuf; + mem[SK_MEMINFO_RCVBUF] = READ_ONCE(sk->sk_rcvbuf); mem[SK_MEMINFO_WMEM_ALLOC] = sk_wmem_alloc_get(sk); mem[SK_MEMINFO_SNDBUF] = sk->sk_sndbuf; mem[SK_MEMINFO_FWD_ALLOC] = sk->sk_forward_alloc; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 577a8c6eef9f..bc0481aa6633 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -451,7 +451,7 @@ void tcp_init_sock(struct sock *sk) icsk->icsk_sync_mss = tcp_sync_mss; sk->sk_sndbuf = sock_net(sk)->ipv4.sysctl_tcp_wmem[1]; - sk->sk_rcvbuf = sock_net(sk)->ipv4.sysctl_tcp_rmem[1]; + WRITE_ONCE(sk->sk_rcvbuf, sock_net(sk)->ipv4.sysctl_tcp_rmem[1]); sk_sockets_allocated_inc(sk); sk->sk_route_forced_caps = NETIF_F_GSO; @@ -1711,7 +1711,7 @@ int tcp_set_rcvlowat(struct sock *sk, int val) val <<= 1; if (val > sk->sk_rcvbuf) { - sk->sk_rcvbuf = val; + WRITE_ONCE(sk->sk_rcvbuf, val); tcp_sk(sk)->window_clamp = tcp_win_from_space(sk, val); } return 0; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 16342e043ab3..6995df20710a 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -483,8 +483,9 @@ static void tcp_clamp_window(struct sock *sk) !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) && !tcp_under_memory_pressure(sk) && sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)) { - sk->sk_rcvbuf = min(atomic_read(&sk->sk_rmem_alloc), - net->ipv4.sysctl_tcp_rmem[2]); + WRITE_ONCE(sk->sk_rcvbuf, + min(atomic_read(&sk->sk_rmem_alloc), + net->ipv4.sysctl_tcp_rmem[2])); } if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf) tp->rcv_ssthresh = min(tp->window_clamp, 2U * tp->advmss); @@ -648,7 +649,7 @@ void tcp_rcv_space_adjust(struct sock *sk) rcvbuf = min_t(u64, rcvwin * rcvmem, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]); if (rcvbuf > sk->sk_rcvbuf) { - sk->sk_rcvbuf = rcvbuf; + WRITE_ONCE(sk->sk_rcvbuf, rcvbuf); /* Make the window clamp follow along. */ tp->window_clamp = tcp_win_from_space(sk, rcvbuf); From e292f05e0df73f9fcc93329663936e1ded97a988 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 10 Oct 2019 20:17:45 -0700 Subject: [PATCH 083/145] tcp: annotate sk->sk_sndbuf lockless reads For the sake of tcp_poll(), there are few places where we fetch sk->sk_sndbuf while this field can change from IRQ or other cpu. We need to add READ_ONCE() annotations, and also make sure write sides use corresponding WRITE_ONCE() to avoid store-tearing. Note that other transports probably need similar fixes. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 18 +++++++++++------- net/core/filter.c | 3 ++- net/core/sock.c | 15 +++++++++------ net/ipv4/tcp.c | 2 +- net/ipv4/tcp_input.c | 3 ++- 5 files changed, 25 insertions(+), 16 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 79f54e1f8827..3d1e7502333e 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -883,7 +883,7 @@ static inline int sk_stream_min_wspace(const struct sock *sk) static inline int sk_stream_wspace(const struct sock *sk) { - return sk->sk_sndbuf - sk->sk_wmem_queued; + return READ_ONCE(sk->sk_sndbuf) - sk->sk_wmem_queued; } void sk_stream_write_space(struct sock *sk); @@ -1207,7 +1207,7 @@ static inline void sk_refcnt_debug_release(const struct sock *sk) static inline bool __sk_stream_memory_free(const struct sock *sk, int wake) { - if (sk->sk_wmem_queued >= sk->sk_sndbuf) + if (sk->sk_wmem_queued >= READ_ONCE(sk->sk_sndbuf)) return false; return sk->sk_prot->stream_memory_free ? @@ -2220,10 +2220,14 @@ static inline void sk_wake_async(const struct sock *sk, int how, int band) static inline void sk_stream_moderate_sndbuf(struct sock *sk) { - if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK)) { - sk->sk_sndbuf = min(sk->sk_sndbuf, sk->sk_wmem_queued >> 1); - sk->sk_sndbuf = max_t(u32, sk->sk_sndbuf, SOCK_MIN_SNDBUF); - } + u32 val; + + if (sk->sk_userlocks & SOCK_SNDBUF_LOCK) + return; + + val = min(sk->sk_sndbuf, sk->sk_wmem_queued >> 1); + + WRITE_ONCE(sk->sk_sndbuf, max_t(u32, val, SOCK_MIN_SNDBUF)); } struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp, @@ -2251,7 +2255,7 @@ bool sk_page_frag_refill(struct sock *sk, struct page_frag *pfrag); */ static inline bool sock_writeable(const struct sock *sk) { - return refcount_read(&sk->sk_wmem_alloc) < (sk->sk_sndbuf >> 1); + return refcount_read(&sk->sk_wmem_alloc) < (READ_ONCE(sk->sk_sndbuf) >> 1); } static inline gfp_t gfp_any(void) diff --git a/net/core/filter.c b/net/core/filter.c index 7deceaeeed7b..3fed5755494b 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -4258,7 +4258,8 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock, case SO_SNDBUF: val = min_t(u32, val, sysctl_wmem_max); sk->sk_userlocks |= SOCK_SNDBUF_LOCK; - sk->sk_sndbuf = max_t(int, val * 2, SOCK_MIN_SNDBUF); + WRITE_ONCE(sk->sk_sndbuf, + max_t(int, val * 2, SOCK_MIN_SNDBUF)); break; case SO_MAX_PACING_RATE: /* 32bit version */ if (val != ~0U) diff --git a/net/core/sock.c b/net/core/sock.c index 8c8f61e70141..cd075bc86407 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -785,7 +785,8 @@ set_sndbuf: */ val = min_t(int, val, INT_MAX / 2); sk->sk_userlocks |= SOCK_SNDBUF_LOCK; - sk->sk_sndbuf = max_t(int, val * 2, SOCK_MIN_SNDBUF); + WRITE_ONCE(sk->sk_sndbuf, + max_t(int, val * 2, SOCK_MIN_SNDBUF)); /* Wake up sending tasks if we upped the value. */ sk->sk_write_space(sk); break; @@ -2089,8 +2090,10 @@ EXPORT_SYMBOL(sock_i_ino); struct sk_buff *sock_wmalloc(struct sock *sk, unsigned long size, int force, gfp_t priority) { - if (force || refcount_read(&sk->sk_wmem_alloc) < sk->sk_sndbuf) { + if (force || + refcount_read(&sk->sk_wmem_alloc) < READ_ONCE(sk->sk_sndbuf)) { struct sk_buff *skb = alloc_skb(size, priority); + if (skb) { skb_set_owner_w(skb, sk); return skb; @@ -2191,7 +2194,7 @@ static long sock_wait_for_wmem(struct sock *sk, long timeo) break; set_bit(SOCK_NOSPACE, &sk->sk_socket->flags); prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); - if (refcount_read(&sk->sk_wmem_alloc) < sk->sk_sndbuf) + if (refcount_read(&sk->sk_wmem_alloc) < READ_ONCE(sk->sk_sndbuf)) break; if (sk->sk_shutdown & SEND_SHUTDOWN) break; @@ -2226,7 +2229,7 @@ struct sk_buff *sock_alloc_send_pskb(struct sock *sk, unsigned long header_len, if (sk->sk_shutdown & SEND_SHUTDOWN) goto failure; - if (sk_wmem_alloc_get(sk) < sk->sk_sndbuf) + if (sk_wmem_alloc_get(sk) < READ_ONCE(sk->sk_sndbuf)) break; sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk); @@ -2807,7 +2810,7 @@ static void sock_def_write_space(struct sock *sk) /* Do not wake up a writer until he can make "significant" * progress. --DaveM */ - if ((refcount_read(&sk->sk_wmem_alloc) << 1) <= sk->sk_sndbuf) { + if ((refcount_read(&sk->sk_wmem_alloc) << 1) <= READ_ONCE(sk->sk_sndbuf)) { wq = rcu_dereference(sk->sk_wq); if (skwq_has_sleeper(wq)) wake_up_interruptible_sync_poll(&wq->wait, EPOLLOUT | @@ -3207,7 +3210,7 @@ void sk_get_meminfo(const struct sock *sk, u32 *mem) mem[SK_MEMINFO_RMEM_ALLOC] = sk_rmem_alloc_get(sk); mem[SK_MEMINFO_RCVBUF] = READ_ONCE(sk->sk_rcvbuf); mem[SK_MEMINFO_WMEM_ALLOC] = sk_wmem_alloc_get(sk); - mem[SK_MEMINFO_SNDBUF] = sk->sk_sndbuf; + mem[SK_MEMINFO_SNDBUF] = READ_ONCE(sk->sk_sndbuf); mem[SK_MEMINFO_FWD_ALLOC] = sk->sk_forward_alloc; mem[SK_MEMINFO_WMEM_QUEUED] = sk->sk_wmem_queued; mem[SK_MEMINFO_OPTMEM] = atomic_read(&sk->sk_omem_alloc); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index bc0481aa6633..111853262972 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -450,7 +450,7 @@ void tcp_init_sock(struct sock *sk) icsk->icsk_sync_mss = tcp_sync_mss; - sk->sk_sndbuf = sock_net(sk)->ipv4.sysctl_tcp_wmem[1]; + WRITE_ONCE(sk->sk_sndbuf, sock_net(sk)->ipv4.sysctl_tcp_wmem[1]); WRITE_ONCE(sk->sk_rcvbuf, sock_net(sk)->ipv4.sysctl_tcp_rmem[1]); sk_sockets_allocated_inc(sk); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 6995df20710a..a2e52ad7cdab 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -359,7 +359,8 @@ static void tcp_sndbuf_expand(struct sock *sk) sndmem *= nr_segs * per_mss; if (sk->sk_sndbuf < sndmem) - sk->sk_sndbuf = min(sndmem, sock_net(sk)->ipv4.sysctl_tcp_wmem[2]); + WRITE_ONCE(sk->sk_sndbuf, + min(sndmem, sock_net(sk)->ipv4.sysctl_tcp_wmem[2])); } /* 2. Tuning advertised window (window_clamp, rcv_ssthresh) From ab4e846a82d0ae00176de19f2db3c5c64f8eb5f2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 10 Oct 2019 20:17:46 -0700 Subject: [PATCH 084/145] tcp: annotate sk->sk_wmem_queued lockless reads For the sake of tcp_poll(), there are few places where we fetch sk->sk_wmem_queued while this field can change from IRQ or other cpu. We need to add READ_ONCE() annotations, and also make sure write sides use corresponding WRITE_ONCE() to avoid store-tearing. sk_wmem_queued_add() helper is added so that we can in the future convert to ADD_ONCE() or equivalent if/when available. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 15 ++++++++++----- include/trace/events/sock.h | 2 +- net/core/datagram.c | 2 +- net/core/sock.c | 2 +- net/ipv4/inet_diag.c | 2 +- net/ipv4/tcp.c | 4 ++-- net/ipv4/tcp_output.c | 14 +++++++------- net/sched/em_meta.c | 2 +- 8 files changed, 24 insertions(+), 19 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index 3d1e7502333e..f69b58bff7e5 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -878,12 +878,17 @@ static inline bool sk_acceptq_is_full(const struct sock *sk) */ static inline int sk_stream_min_wspace(const struct sock *sk) { - return sk->sk_wmem_queued >> 1; + return READ_ONCE(sk->sk_wmem_queued) >> 1; } static inline int sk_stream_wspace(const struct sock *sk) { - return READ_ONCE(sk->sk_sndbuf) - sk->sk_wmem_queued; + return READ_ONCE(sk->sk_sndbuf) - READ_ONCE(sk->sk_wmem_queued); +} + +static inline void sk_wmem_queued_add(struct sock *sk, int val) +{ + WRITE_ONCE(sk->sk_wmem_queued, sk->sk_wmem_queued + val); } void sk_stream_write_space(struct sock *sk); @@ -1207,7 +1212,7 @@ static inline void sk_refcnt_debug_release(const struct sock *sk) static inline bool __sk_stream_memory_free(const struct sock *sk, int wake) { - if (sk->sk_wmem_queued >= READ_ONCE(sk->sk_sndbuf)) + if (READ_ONCE(sk->sk_wmem_queued) >= READ_ONCE(sk->sk_sndbuf)) return false; return sk->sk_prot->stream_memory_free ? @@ -1467,7 +1472,7 @@ DECLARE_STATIC_KEY_FALSE(tcp_tx_skb_cache_key); static inline void sk_wmem_free_skb(struct sock *sk, struct sk_buff *skb) { sock_set_flag(sk, SOCK_QUEUE_SHRUNK); - sk->sk_wmem_queued -= skb->truesize; + sk_wmem_queued_add(sk, -skb->truesize); sk_mem_uncharge(sk, skb->truesize); if (static_branch_unlikely(&tcp_tx_skb_cache_key) && !sk->sk_tx_skb_cache && !skb_cloned(skb)) { @@ -2014,7 +2019,7 @@ static inline int skb_copy_to_page_nocache(struct sock *sk, struct iov_iter *fro skb->len += copy; skb->data_len += copy; skb->truesize += copy; - sk->sk_wmem_queued += copy; + sk_wmem_queued_add(sk, copy); sk_mem_charge(sk, copy); return 0; } diff --git a/include/trace/events/sock.h b/include/trace/events/sock.h index f720c32e7dfd..51fe9f6719eb 100644 --- a/include/trace/events/sock.h +++ b/include/trace/events/sock.h @@ -115,7 +115,7 @@ TRACE_EVENT(sock_exceed_buf_limit, __entry->rmem_alloc = atomic_read(&sk->sk_rmem_alloc); __entry->sysctl_wmem = sk_get_wmem0(sk, prot); __entry->wmem_alloc = refcount_read(&sk->sk_wmem_alloc); - __entry->wmem_queued = sk->sk_wmem_queued; + __entry->wmem_queued = READ_ONCE(sk->sk_wmem_queued); __entry->kind = kind; ), diff --git a/net/core/datagram.c b/net/core/datagram.c index 4cc8dc5db2b7..c210fc116103 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -640,7 +640,7 @@ int __zerocopy_sg_from_iter(struct sock *sk, struct sk_buff *skb, skb->len += copied; skb->truesize += truesize; if (sk && sk->sk_type == SOCK_STREAM) { - sk->sk_wmem_queued += truesize; + sk_wmem_queued_add(sk, truesize); sk_mem_charge(sk, truesize); } else { refcount_add(truesize, &skb->sk->sk_wmem_alloc); diff --git a/net/core/sock.c b/net/core/sock.c index cd075bc86407..a515392ba84b 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3212,7 +3212,7 @@ void sk_get_meminfo(const struct sock *sk, u32 *mem) mem[SK_MEMINFO_WMEM_ALLOC] = sk_wmem_alloc_get(sk); mem[SK_MEMINFO_SNDBUF] = READ_ONCE(sk->sk_sndbuf); mem[SK_MEMINFO_FWD_ALLOC] = sk->sk_forward_alloc; - mem[SK_MEMINFO_WMEM_QUEUED] = sk->sk_wmem_queued; + mem[SK_MEMINFO_WMEM_QUEUED] = READ_ONCE(sk->sk_wmem_queued); mem[SK_MEMINFO_OPTMEM] = atomic_read(&sk->sk_omem_alloc); mem[SK_MEMINFO_BACKLOG] = READ_ONCE(sk->sk_backlog.len); mem[SK_MEMINFO_DROPS] = atomic_read(&sk->sk_drops); diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index bbb005eb5218..7dc79b973e6e 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -193,7 +193,7 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, if (ext & (1 << (INET_DIAG_MEMINFO - 1))) { struct inet_diag_meminfo minfo = { .idiag_rmem = sk_rmem_alloc_get(sk), - .idiag_wmem = sk->sk_wmem_queued, + .idiag_wmem = READ_ONCE(sk->sk_wmem_queued), .idiag_fmem = sk->sk_forward_alloc, .idiag_tmem = sk_wmem_alloc_get(sk), }; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 111853262972..b2ac4f074e2d 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -659,7 +659,7 @@ static void skb_entail(struct sock *sk, struct sk_buff *skb) tcb->sacked = 0; __skb_header_release(skb); tcp_add_write_queue_tail(sk, skb); - sk->sk_wmem_queued += skb->truesize; + sk_wmem_queued_add(sk, skb->truesize); sk_mem_charge(sk, skb->truesize); if (tp->nonagle & TCP_NAGLE_PUSH) tp->nonagle &= ~TCP_NAGLE_PUSH; @@ -1034,7 +1034,7 @@ new_segment: skb->len += copy; skb->data_len += copy; skb->truesize += copy; - sk->sk_wmem_queued += copy; + sk_wmem_queued_add(sk, copy); sk_mem_charge(sk, copy); skb->ip_summed = CHECKSUM_PARTIAL; WRITE_ONCE(tp->write_seq, tp->write_seq + copy); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index a115a991dfb5..0488607c5cd3 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1199,7 +1199,7 @@ static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb) WRITE_ONCE(tp->write_seq, TCP_SKB_CB(skb)->end_seq); __skb_header_release(skb); tcp_add_write_queue_tail(sk, skb); - sk->sk_wmem_queued += skb->truesize; + sk_wmem_queued_add(sk, skb->truesize); sk_mem_charge(sk, skb->truesize); } @@ -1333,7 +1333,7 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue, return -ENOMEM; /* We'll just try again later. */ skb_copy_decrypted(buff, skb); - sk->sk_wmem_queued += buff->truesize; + sk_wmem_queued_add(sk, buff->truesize); sk_mem_charge(sk, buff->truesize); nlen = skb->len - len - nsize; buff->truesize += nlen; @@ -1443,7 +1443,7 @@ int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len) if (delta_truesize) { skb->truesize -= delta_truesize; - sk->sk_wmem_queued -= delta_truesize; + sk_wmem_queued_add(sk, -delta_truesize); sk_mem_uncharge(sk, delta_truesize); sock_set_flag(sk, SOCK_QUEUE_SHRUNK); } @@ -1888,7 +1888,7 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, return -ENOMEM; skb_copy_decrypted(buff, skb); - sk->sk_wmem_queued += buff->truesize; + sk_wmem_queued_add(sk, buff->truesize); sk_mem_charge(sk, buff->truesize); buff->truesize += nlen; skb->truesize -= nlen; @@ -2152,7 +2152,7 @@ static int tcp_mtu_probe(struct sock *sk) nskb = sk_stream_alloc_skb(sk, probe_size, GFP_ATOMIC, false); if (!nskb) return -1; - sk->sk_wmem_queued += nskb->truesize; + sk_wmem_queued_add(sk, nskb->truesize); sk_mem_charge(sk, nskb->truesize); skb = tcp_send_head(sk); @@ -3222,7 +3222,7 @@ int tcp_send_synack(struct sock *sk) tcp_rtx_queue_unlink_and_free(skb, sk); __skb_header_release(nskb); tcp_rbtree_insert(&sk->tcp_rtx_queue, nskb); - sk->sk_wmem_queued += nskb->truesize; + sk_wmem_queued_add(sk, nskb->truesize); sk_mem_charge(sk, nskb->truesize); skb = nskb; } @@ -3447,7 +3447,7 @@ static void tcp_connect_queue_skb(struct sock *sk, struct sk_buff *skb) tcb->end_seq += skb->len; __skb_header_release(skb); - sk->sk_wmem_queued += skb->truesize; + sk_wmem_queued_add(sk, skb->truesize); sk_mem_charge(sk, skb->truesize); WRITE_ONCE(tp->write_seq, tcb->end_seq); tp->packets_out += tcp_skb_pcount(skb); diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index 4c9122fc35c9..3177dcb17316 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -446,7 +446,7 @@ META_COLLECTOR(int_sk_wmem_queued) *err = -1; return; } - dst->value = sk->sk_wmem_queued; + dst->value = READ_ONCE(sk->sk_wmem_queued); } META_COLLECTOR(int_sk_fwd_alloc) From c23936fad79ea2bea749db19b8921bb4dc524905 Mon Sep 17 00:00:00 2001 From: Alexandre Belloni Date: Thu, 10 Oct 2019 22:46:06 +0200 Subject: [PATCH 085/145] net: lpc_eth: avoid resetting twice __lpc_eth_shutdown is called after __lpc_eth_reset but it is already calling __lpc_eth_reset. Avoid resetting the IP twice. Signed-off-by: Alexandre Belloni Signed-off-by: David S. Miller --- drivers/net/ethernet/nxp/lpc_eth.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c index 141571e2ec11..544012a67221 100644 --- a/drivers/net/ethernet/nxp/lpc_eth.c +++ b/drivers/net/ethernet/nxp/lpc_eth.c @@ -1356,9 +1356,6 @@ static int lpc_eth_drv_probe(struct platform_device *pdev) if (!is_valid_ether_addr(ndev->dev_addr)) eth_hw_addr_random(ndev); - /* Reset the ethernet controller */ - __lpc_eth_reset(pldat); - /* then shut everything down to save power */ __lpc_eth_shutdown(pldat); From 11d49ce9f7946dfed4dcf5dbde865c78058b50ab Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A9dric=20Le=20Goater?= Date: Fri, 11 Oct 2019 07:52:54 +0200 Subject: [PATCH 086/145] net/ibmvnic: Fix EOI when running in XIVE mode. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit pSeries machines on POWER9 processors can run with the XICS (legacy) interrupt mode or with the XIVE exploitation interrupt mode. These interrupt contollers have different interfaces for interrupt management : XICS uses hcalls and XIVE loads and stores on a page. H_EOI being a XICS interface the enable_scrq_irq() routine can fail when the machine runs in XIVE mode. Fix that by calling the EOI handler of the interrupt chip. Fixes: f23e0643cd0b ("ibmvnic: Clear pending interrupt after device reset") Signed-off-by: Cédric Le Goater Signed-off-by: David S. Miller --- drivers/net/ethernet/ibm/ibmvnic.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 2b073a3c0b84..f59d9a8e35e2 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -2878,12 +2878,10 @@ static int enable_scrq_irq(struct ibmvnic_adapter *adapter, if (test_bit(0, &adapter->resetting) && adapter->reset_reason == VNIC_RESET_MOBILITY) { - u64 val = (0xff000000) | scrq->hw_irq; + struct irq_desc *desc = irq_to_desc(scrq->irq); + struct irq_chip *chip = irq_desc_get_chip(desc); - rc = plpar_hcall_norets(H_EOI, val); - if (rc) - dev_err(dev, "H_EOI FAILED irq 0x%llx. rc=%ld\n", - val, rc); + chip->irq_eoi(&desc->irq_data); } rc = plpar_hcall_norets(H_VIOCTL, adapter->vdev->unit_address, From 33902b4a4227877896dd9368ac10f4ca0d100de5 Mon Sep 17 00:00:00 2001 From: YueHaibing Date: Fri, 11 Oct 2019 17:46:53 +0800 Subject: [PATCH 087/145] netdevsim: Fix error handling in nsim_fib_init and nsim_fib_exit In nsim_fib_init(), if register_fib_notifier failed, nsim_fib_net_ops should be unregistered before return. In nsim_fib_exit(), unregister_fib_notifier should be called before nsim_fib_net_ops be unregistered, otherwise may cause use-after-free: BUG: KASAN: use-after-free in nsim_fib_event_nb+0x342/0x570 [netdevsim] Read of size 8 at addr ffff8881daaf4388 by task kworker/0:3/3499 CPU: 0 PID: 3499 Comm: kworker/0:3 Not tainted 5.3.0-rc7+ #30 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014 Workqueue: ipv6_addrconf addrconf_dad_work [ipv6] Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0xa9/0x10e lib/dump_stack.c:113 print_address_description+0x65/0x380 mm/kasan/report.c:351 __kasan_report+0x149/0x18d mm/kasan/report.c:482 kasan_report+0xe/0x20 mm/kasan/common.c:618 nsim_fib_event_nb+0x342/0x570 [netdevsim] notifier_call_chain+0x52/0xf0 kernel/notifier.c:95 __atomic_notifier_call_chain+0x78/0x140 kernel/notifier.c:185 call_fib_notifiers+0x30/0x60 net/core/fib_notifier.c:30 call_fib6_entry_notifiers+0xc1/0x100 [ipv6] fib6_add+0x92e/0x1b10 [ipv6] __ip6_ins_rt+0x40/0x60 [ipv6] ip6_ins_rt+0x84/0xb0 [ipv6] __ipv6_ifa_notify+0x4b6/0x550 [ipv6] ipv6_ifa_notify+0xa5/0x180 [ipv6] addrconf_dad_completed+0xca/0x640 [ipv6] addrconf_dad_work+0x296/0x960 [ipv6] process_one_work+0x5c0/0xc00 kernel/workqueue.c:2269 worker_thread+0x5c/0x670 kernel/workqueue.c:2415 kthread+0x1d7/0x200 kernel/kthread.c:255 ret_from_fork+0x3a/0x50 arch/x86/entry/entry_64.S:352 Allocated by task 3388: save_stack+0x19/0x80 mm/kasan/common.c:69 set_track mm/kasan/common.c:77 [inline] __kasan_kmalloc.constprop.3+0xa0/0xd0 mm/kasan/common.c:493 kmalloc include/linux/slab.h:557 [inline] kzalloc include/linux/slab.h:748 [inline] ops_init+0xa9/0x220 net/core/net_namespace.c:127 __register_pernet_operations net/core/net_namespace.c:1135 [inline] register_pernet_operations+0x1d4/0x420 net/core/net_namespace.c:1212 register_pernet_subsys+0x24/0x40 net/core/net_namespace.c:1253 nsim_fib_init+0x12/0x70 [netdevsim] veth_get_link_ksettings+0x2b/0x50 [veth] do_one_initcall+0xd4/0x454 init/main.c:939 do_init_module+0xe0/0x330 kernel/module.c:3490 load_module+0x3c2f/0x4620 kernel/module.c:3841 __do_sys_finit_module+0x163/0x190 kernel/module.c:3931 do_syscall_64+0x72/0x2e0 arch/x86/entry/common.c:296 entry_SYSCALL_64_after_hwframe+0x49/0xbe Freed by task 3534: save_stack+0x19/0x80 mm/kasan/common.c:69 set_track mm/kasan/common.c:77 [inline] __kasan_slab_free+0x130/0x180 mm/kasan/common.c:455 slab_free_hook mm/slub.c:1423 [inline] slab_free_freelist_hook mm/slub.c:1474 [inline] slab_free mm/slub.c:3016 [inline] kfree+0xe9/0x2d0 mm/slub.c:3957 ops_free net/core/net_namespace.c:151 [inline] ops_free_list.part.7+0x156/0x220 net/core/net_namespace.c:184 ops_free_list net/core/net_namespace.c:182 [inline] __unregister_pernet_operations net/core/net_namespace.c:1165 [inline] unregister_pernet_operations+0x221/0x2a0 net/core/net_namespace.c:1224 unregister_pernet_subsys+0x1d/0x30 net/core/net_namespace.c:1271 nsim_fib_exit+0x11/0x20 [netdevsim] nsim_module_exit+0x16/0x21 [netdevsim] __do_sys_delete_module kernel/module.c:1015 [inline] __se_sys_delete_module kernel/module.c:958 [inline] __x64_sys_delete_module+0x244/0x330 kernel/module.c:958 do_syscall_64+0x72/0x2e0 arch/x86/entry/common.c:296 entry_SYSCALL_64_after_hwframe+0x49/0xbe Reported-by: Hulk Robot Fixes: 59c84b9fcf42 ("netdevsim: Restore per-network namespace accounting for fib entries") Signed-off-by: YueHaibing Acked-by: Jakub Kicinski Signed-off-by: David S. Miller --- drivers/net/netdevsim/fib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/netdevsim/fib.c b/drivers/net/netdevsim/fib.c index f61d094746c0..1a251f76d09b 100644 --- a/drivers/net/netdevsim/fib.c +++ b/drivers/net/netdevsim/fib.c @@ -241,8 +241,8 @@ static struct pernet_operations nsim_fib_net_ops = { void nsim_fib_exit(void) { - unregister_pernet_subsys(&nsim_fib_net_ops); unregister_fib_notifier(&nsim_fib_nb); + unregister_pernet_subsys(&nsim_fib_net_ops); } int nsim_fib_init(void) @@ -258,6 +258,7 @@ int nsim_fib_init(void) err = register_fib_notifier(&nsim_fib_nb, nsim_fib_dump_inconsistent); if (err < 0) { pr_err("Failed to register fib notifier\n"); + unregister_pernet_subsys(&nsim_fib_net_ops); goto err_out; } From d79749f7716d9dc32fa2d5075f6ec29aac63c76d Mon Sep 17 00:00:00 2001 From: Miaoqing Pan Date: Thu, 29 Aug 2019 10:45:12 +0800 Subject: [PATCH 088/145] ath10k: fix latency issue for QCA988x (kvalo: cherry picked from commit 1340cc631bd00431e2f174525c971f119df9efa1 in wireless-drivers-next to wireless-drivers as this a frequently reported regression) Bad latency is found on QCA988x, the issue was introduced by commit 4504f0e5b571 ("ath10k: sdio: workaround firmware UART pin configuration bug"). If uart_pin_workaround is false, this change will set uart pin even if uart_print is false. Tested HW: QCA9880 Tested FW: 10.2.4-1.0-00037 Fixes: 4504f0e5b571 ("ath10k: sdio: workaround firmware UART pin configuration bug") Signed-off-by: Miaoqing Pan Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/ath10k/core.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index dc45d16e8d21..383d4fa555a8 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -2118,12 +2118,15 @@ static int ath10k_init_uart(struct ath10k *ar) return ret; } - if (!uart_print && ar->hw_params.uart_pin_workaround) { - ret = ath10k_bmi_write32(ar, hi_dbg_uart_txpin, - ar->hw_params.uart_pin); - if (ret) { - ath10k_warn(ar, "failed to set UART TX pin: %d", ret); - return ret; + if (!uart_print) { + if (ar->hw_params.uart_pin_workaround) { + ret = ath10k_bmi_write32(ar, hi_dbg_uart_txpin, + ar->hw_params.uart_pin); + if (ret) { + ath10k_warn(ar, "failed to set UART TX pin: %d", + ret); + return ret; + } } return 0; From 06b0d7fe7e5ff3ba4c7e265ef41135e8bcc232bb Mon Sep 17 00:00:00 2001 From: Igor Russkikh Date: Fri, 11 Oct 2019 13:45:19 +0000 Subject: [PATCH 089/145] net: aquantia: temperature retrieval fix MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Chip temperature is a two byte word, colocated internally with cable length data. We do all readouts from HW memory by dwords, thus we should clear extra high bytes, otherwise temperature output gets weird as soon as we attach a cable to the NIC. Fixes: 8f8940118654 ("net: aquantia: add infrastructure to readout chip temperature") Tested-by: Holger Hoffstätte Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- .../net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c index da726489e3c8..7bc51f8d6f2f 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c @@ -337,7 +337,7 @@ static int aq_fw2x_get_phy_temp(struct aq_hw_s *self, int *temp) /* Convert PHY temperature from 1/256 degree Celsius * to 1/1000 degree Celsius. */ - *temp = temp_res * 1000 / 256; + *temp = (temp_res & 0xFFFF) * 1000 / 256; return 0; } From ed4d81c4b3f28ccf624f11fd66f67aec5b58859c Mon Sep 17 00:00:00 2001 From: Igor Russkikh Date: Fri, 11 Oct 2019 13:45:20 +0000 Subject: [PATCH 090/145] net: aquantia: when cleaning hw cache it should be toggled >From HW specification to correctly reset HW caches (this is a required workaround when stopping the device), register bit should actually be toggled. It was previosly always just set. Due to the way driver stops HW this never actually caused any issues, but it still may, so cleaning this up. Fixes: 7a1bb49461b1 ("net: aquantia: fix potential IOMMU fault after driver unbind") Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- .../aquantia/atlantic/hw_atl/hw_atl_b0.c | 16 ++++++++++++++-- .../aquantia/atlantic/hw_atl/hw_atl_llh.c | 17 +++++++++++++++-- .../aquantia/atlantic/hw_atl/hw_atl_llh.h | 7 +++++-- .../atlantic/hw_atl/hw_atl_llh_internal.h | 19 +++++++++++++++++++ 4 files changed, 53 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index 30f7fc4c97ff..3459fadb7ddd 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -968,14 +968,26 @@ static int hw_atl_b0_hw_interrupt_moderation_set(struct aq_hw_s *self) static int hw_atl_b0_hw_stop(struct aq_hw_s *self) { + int err; + u32 val; + hw_atl_b0_hw_irq_disable(self, HW_ATL_B0_INT_MASK); /* Invalidate Descriptor Cache to prevent writing to the cached * descriptors and to the data pointer of those descriptors */ - hw_atl_rdm_rx_dma_desc_cache_init_set(self, 1); + hw_atl_rdm_rx_dma_desc_cache_init_tgl(self); - return aq_hw_err_from_flags(self); + err = aq_hw_err_from_flags(self); + + if (err) + goto err_exit; + + readx_poll_timeout_atomic(hw_atl_rdm_rx_dma_desc_cache_init_done_get, + self, val, val == 1, 1000U, 10000U); + +err_exit: + return err; } static int hw_atl_b0_hw_ring_tx_stop(struct aq_hw_s *self, diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c index 1149812ae463..6f340695e6bd 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c @@ -606,12 +606,25 @@ void hw_atl_rpb_rx_flow_ctl_mode_set(struct aq_hw_s *aq_hw, u32 rx_flow_ctl_mode HW_ATL_RPB_RX_FC_MODE_SHIFT, rx_flow_ctl_mode); } -void hw_atl_rdm_rx_dma_desc_cache_init_set(struct aq_hw_s *aq_hw, u32 init) +void hw_atl_rdm_rx_dma_desc_cache_init_tgl(struct aq_hw_s *aq_hw) { + u32 val; + + val = aq_hw_read_reg_bit(aq_hw, HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_ADR, + HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSK, + HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_SHIFT); + aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_ADR, HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSK, HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_SHIFT, - init); + val ^ 1); +} + +u32 hw_atl_rdm_rx_dma_desc_cache_init_done_get(struct aq_hw_s *aq_hw) +{ + return aq_hw_read_reg_bit(aq_hw, RDM_RX_DMA_DESC_CACHE_INIT_DONE_ADR, + RDM_RX_DMA_DESC_CACHE_INIT_DONE_MSK, + RDM_RX_DMA_DESC_CACHE_INIT_DONE_SHIFT); } void hw_atl_rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw, diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h index 0c37abbabca5..c3ee278c3747 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h @@ -313,8 +313,11 @@ void hw_atl_rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw, u32 rx_pkt_buff_size_per_tc, u32 buffer); -/* set rdm rx dma descriptor cache init */ -void hw_atl_rdm_rx_dma_desc_cache_init_set(struct aq_hw_s *aq_hw, u32 init); +/* toggle rdm rx dma descriptor cache init */ +void hw_atl_rdm_rx_dma_desc_cache_init_tgl(struct aq_hw_s *aq_hw); + +/* get rdm rx dma descriptor cache init done */ +u32 hw_atl_rdm_rx_dma_desc_cache_init_done_get(struct aq_hw_s *aq_hw); /* set rx xoff enable (per tc) */ void hw_atl_rpb_rx_xoff_en_per_tc_set(struct aq_hw_s *aq_hw, u32 rx_xoff_en_per_tc, diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h index c3febcdfa92e..35887ad89025 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h @@ -318,6 +318,25 @@ /* default value of bitfield rdm_desc_init_i */ #define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_DEFAULT 0x0 +/* rdm_desc_init_done_i bitfield definitions + * preprocessor definitions for the bitfield rdm_desc_init_done_i. + * port="pif_rdm_desc_init_done_i" + */ + +/* register address for bitfield rdm_desc_init_done_i */ +#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_ADR 0x00005a10 +/* bitmask for bitfield rdm_desc_init_done_i */ +#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_MSK 0x00000001U +/* inverted bitmask for bitfield rdm_desc_init_done_i */ +#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_MSKN 0xfffffffe +/* lower bit position of bitfield rdm_desc_init_done_i */ +#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_SHIFT 0U +/* width of bitfield rdm_desc_init_done_i */ +#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_WIDTH 1 +/* default value of bitfield rdm_desc_init_done_i */ +#define RDM_RX_DMA_DESC_CACHE_INIT_DONE_DEFAULT 0x0 + + /* rx int_desc_wrb_en bitfield definitions * preprocessor definitions for the bitfield "int_desc_wrb_en". * port="pif_rdm_int_desc_wrb_en_i" From d08b9a0a3ebdf71b0aabe576c7dd48e57e80e0f0 Mon Sep 17 00:00:00 2001 From: Dmitry Bogdanov Date: Fri, 11 Oct 2019 13:45:22 +0000 Subject: [PATCH 091/145] net: aquantia: do not pass lro session with invalid tcp checksum Individual descriptors on LRO TCP session should be checked for CRC errors. It was discovered that HW recalculates L4 checksums on LRO session and does not break it up on bad L4 csum. Thus, driver should aggregate HW LRO L4 statuses from all individual buffers of LRO session and drop packet if one of the buffers has bad L4 checksum. Fixes: f38f1ee8aeb2 ("net: aquantia: check rx csum for all packets in LRO session") Signed-off-by: Dmitry Bogdanov Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_ring.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c index 3901d7994ca1..76bdbe1596d6 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c @@ -313,6 +313,7 @@ int aq_ring_rx_clean(struct aq_ring_s *self, break; buff->is_error |= buff_->is_error; + buff->is_cso_err |= buff_->is_cso_err; } while (!buff_->is_eop); @@ -320,7 +321,7 @@ int aq_ring_rx_clean(struct aq_ring_s *self, err = 0; goto err_exit; } - if (buff->is_error) { + if (buff->is_error || buff->is_cso_err) { buff_ = buff; do { next_ = buff_->next, From 9f051db566da1e8110659ab4ab188af1c2510bb4 Mon Sep 17 00:00:00 2001 From: Dmitry Bogdanov Date: Fri, 11 Oct 2019 13:45:23 +0000 Subject: [PATCH 092/145] net: aquantia: correctly handle macvlan and multicast coexistence macvlan and multicast handling is now mixed up. The explicit issue is that macvlan interface gets broken (no traffic) after clearing MULTICAST flag on the real interface. We now do separate logic and consider both ALLMULTI and MULTICAST flags on the device. Fixes: 11ba961c9161 ("net: aquantia: Fix IFF_ALLMULTI flag functionality") Signed-off-by: Dmitry Bogdanov Signed-off-by: Igor Russkikh Signed-off-by: David S. Miller --- .../net/ethernet/aquantia/atlantic/aq_main.c | 4 +-- .../net/ethernet/aquantia/atlantic/aq_nic.c | 32 +++++++++---------- .../aquantia/atlantic/hw_atl/hw_atl_b0.c | 7 ++-- 3 files changed, 21 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_main.c b/drivers/net/ethernet/aquantia/atlantic/aq_main.c index b4a0fb281e69..bb65dd39f847 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_main.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_main.c @@ -194,9 +194,7 @@ static void aq_ndev_set_multicast_settings(struct net_device *ndev) { struct aq_nic_s *aq_nic = netdev_priv(ndev); - aq_nic_set_packet_filter(aq_nic, ndev->flags); - - aq_nic_set_multicast_list(aq_nic, ndev); + (void)aq_nic_set_multicast_list(aq_nic, ndev); } static int aq_ndo_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index 8f66e7817811..2a18439b36fb 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -631,9 +631,12 @@ err_exit: int aq_nic_set_multicast_list(struct aq_nic_s *self, struct net_device *ndev) { - unsigned int packet_filter = self->packet_filter; + const struct aq_hw_ops *hw_ops = self->aq_hw_ops; + struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg; + unsigned int packet_filter = ndev->flags; struct netdev_hw_addr *ha = NULL; unsigned int i = 0U; + int err = 0; self->mc_list.count = 0; if (netdev_uc_count(ndev) > AQ_HW_MULTICAST_ADDRESS_MAX) { @@ -641,29 +644,26 @@ int aq_nic_set_multicast_list(struct aq_nic_s *self, struct net_device *ndev) } else { netdev_for_each_uc_addr(ha, ndev) { ether_addr_copy(self->mc_list.ar[i++], ha->addr); - - if (i >= AQ_HW_MULTICAST_ADDRESS_MAX) - break; } } - if (i + netdev_mc_count(ndev) > AQ_HW_MULTICAST_ADDRESS_MAX) { - packet_filter |= IFF_ALLMULTI; - } else { - netdev_for_each_mc_addr(ha, ndev) { - ether_addr_copy(self->mc_list.ar[i++], ha->addr); - - if (i >= AQ_HW_MULTICAST_ADDRESS_MAX) - break; + cfg->is_mc_list_enabled = !!(packet_filter & IFF_MULTICAST); + if (cfg->is_mc_list_enabled) { + if (i + netdev_mc_count(ndev) > AQ_HW_MULTICAST_ADDRESS_MAX) { + packet_filter |= IFF_ALLMULTI; + } else { + netdev_for_each_mc_addr(ha, ndev) { + ether_addr_copy(self->mc_list.ar[i++], + ha->addr); + } } } if (i > 0 && i <= AQ_HW_MULTICAST_ADDRESS_MAX) { - packet_filter |= IFF_MULTICAST; self->mc_list.count = i; - self->aq_hw_ops->hw_multicast_list_set(self->aq_hw, - self->mc_list.ar, - self->mc_list.count); + err = hw_ops->hw_multicast_list_set(self->aq_hw, + self->mc_list.ar, + self->mc_list.count); } return aq_nic_set_packet_filter(self, packet_filter); } diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index 3459fadb7ddd..2ad3fa6316ce 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -818,14 +818,15 @@ static int hw_atl_b0_hw_packet_filter_set(struct aq_hw_s *self, cfg->is_vlan_force_promisc); hw_atl_rpfl2multicast_flr_en_set(self, - IS_FILTER_ENABLED(IFF_ALLMULTI), 0); + IS_FILTER_ENABLED(IFF_ALLMULTI) && + IS_FILTER_ENABLED(IFF_MULTICAST), 0); hw_atl_rpfl2_accept_all_mc_packets_set(self, - IS_FILTER_ENABLED(IFF_ALLMULTI)); + IS_FILTER_ENABLED(IFF_ALLMULTI) && + IS_FILTER_ENABLED(IFF_MULTICAST)); hw_atl_rpfl2broadcast_en_set(self, IS_FILTER_ENABLED(IFF_BROADCAST)); - cfg->is_mc_list_enabled = IS_FILTER_ENABLED(IFF_MULTICAST); for (i = HW_ATL_B0_MAC_MIN; i < HW_ATL_B0_MAC_MAX; ++i) hw_atl_rpfl2_uc_flr_en_set(self, From b0818f80c8c1bc215bba276bd61c216014fab23b Mon Sep 17 00:00:00 2001 From: Mahesh Bandewar Date: Fri, 11 Oct 2019 18:14:55 -0700 Subject: [PATCH 093/145] blackhole_netdev: fix syzkaller reported issue While invalidating the dst, we assign backhole_netdev instead of loopback device. However, this device does not have idev pointer and hence no ip6_ptr even if IPv6 is enabled. Possibly this has triggered the syzbot reported crash. The syzbot report does not have reproducer, however, this is the only device that doesn't have matching idev created. Crash instruction is : static inline bool ip6_ignore_linkdown(const struct net_device *dev) { const struct inet6_dev *idev = __in6_dev_get(dev); return !!idev->cnf.ignore_routes_with_linkdown; <= crash } Also ipv6 always assumes presence of idev and never checks for it being NULL (as does the above referenced code). So adding a idev for the blackhole_netdev to avoid this class of crashes in the future. Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 7 ++++++- net/ipv6/route.c | 15 ++++++--------- 2 files changed, 12 insertions(+), 10 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 34ccef18b40e..4c87594d1389 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -6996,7 +6996,7 @@ static struct rtnl_af_ops inet6_ops __read_mostly = { int __init addrconf_init(void) { - struct inet6_dev *idev; + struct inet6_dev *idev, *bdev; int i, err; err = ipv6_addr_label_init(); @@ -7036,10 +7036,14 @@ int __init addrconf_init(void) */ rtnl_lock(); idev = ipv6_add_dev(init_net.loopback_dev); + bdev = ipv6_add_dev(blackhole_netdev); rtnl_unlock(); if (IS_ERR(idev)) { err = PTR_ERR(idev); goto errlo; + } else if (IS_ERR(bdev)) { + err = PTR_ERR(bdev); + goto errlo; } ip6_route_init_special_entries(); @@ -7124,6 +7128,7 @@ void addrconf_cleanup(void) addrconf_ifdown(dev, 1); } addrconf_ifdown(init_net.loopback_dev, 2); + addrconf_ifdown(blackhole_netdev, 2); /* * Check hash table. diff --git a/net/ipv6/route.c b/net/ipv6/route.c index a63ff85fe141..742120728869 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -155,10 +155,9 @@ void rt6_uncached_list_del(struct rt6_info *rt) static void rt6_uncached_list_flush_dev(struct net *net, struct net_device *dev) { - struct net_device *loopback_dev = net->loopback_dev; int cpu; - if (dev == loopback_dev) + if (dev == net->loopback_dev) return; for_each_possible_cpu(cpu) { @@ -171,7 +170,7 @@ static void rt6_uncached_list_flush_dev(struct net *net, struct net_device *dev) struct net_device *rt_dev = rt->dst.dev; if (rt_idev->dev == dev) { - rt->rt6i_idev = in6_dev_get(loopback_dev); + rt->rt6i_idev = in6_dev_get(blackhole_netdev); in6_dev_put(rt_idev); } @@ -386,13 +385,11 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, { struct rt6_info *rt = (struct rt6_info *)dst; struct inet6_dev *idev = rt->rt6i_idev; - struct net_device *loopback_dev = - dev_net(dev)->loopback_dev; - if (idev && idev->dev != loopback_dev) { - struct inet6_dev *loopback_idev = in6_dev_get(loopback_dev); - if (loopback_idev) { - rt->rt6i_idev = loopback_idev; + if (idev && idev->dev != dev_net(dev)->loopback_dev) { + struct inet6_dev *ibdev = in6_dev_get(blackhole_netdev); + if (ibdev) { + rt->rt6i_idev = ibdev; in6_dev_put(idev); } } From 92696286f3bb37ba50e4bd8d1beb24afb759a799 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 11 Oct 2019 12:53:49 -0700 Subject: [PATCH 094/145] net: bcmgenet: Set phydev->dev_flags only for internal PHYs phydev->dev_flags is entirely dependent on the PHY device driver which is going to be used, setting the internal GENET PHY revision in those bits only makes sense when drivers/net/phy/bcm7xxx.c is the PHY driver being used. Fixes: 487320c54143 ("net: bcmgenet: communicate integrated PHY revision to PHY driver") Signed-off-by: Florian Fainelli Acked-by: Doug Berger Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/genet/bcmmii.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index 970e478a9017..94d1dd5d56bf 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -273,11 +273,12 @@ int bcmgenet_mii_probe(struct net_device *dev) struct bcmgenet_priv *priv = netdev_priv(dev); struct device_node *dn = priv->pdev->dev.of_node; struct phy_device *phydev; - u32 phy_flags; + u32 phy_flags = 0; int ret; /* Communicate the integrated PHY revision */ - phy_flags = priv->gphy_rev; + if (priv->internal_phy) + phy_flags = priv->gphy_rev; /* Initialize link state variables that bcmgenet_mii_setup() uses */ priv->old_link = -1; From f913eac8e555bc776565a21ce84bf2bcc3715cbb Mon Sep 17 00:00:00 2001 From: David Ahern Date: Fri, 11 Oct 2019 20:43:03 -0600 Subject: [PATCH 095/145] net: Update address for vrf and l3mdev in MAINTAINERS Use my kernel.org address for all entries in MAINTAINERS. Signed-off-by: David Ahern Signed-off-by: David S. Miller --- MAINTAINERS | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 55199ef7fa74..49e4b004917b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9126,7 +9126,7 @@ F: drivers/auxdisplay/ks0108.c F: include/linux/ks0108.h L3MDEV -M: David Ahern +M: David Ahern L: netdev@vger.kernel.org S: Maintained F: net/l3mdev @@ -17439,7 +17439,7 @@ F: include/linux/regulator/ K: regulator_get_optional VRF -M: David Ahern +M: David Ahern M: Shrijeet Mukherjee L: netdev@vger.kernel.org S: Maintained From ddc790e92b3afa4e366ffb41818cfcd19015031e Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Fri, 11 Oct 2019 21:03:33 -0700 Subject: [PATCH 096/145] net: ethernet: broadcom: have drivers select DIMLIB as needed MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit NET_VENDOR_BROADCOM is intended to control a kconfig menu only. It should not have anything to do with code generation. As such, it should not select DIMLIB for all drivers under NET_VENDOR_BROADCOM. Instead each driver that needs DIMLIB should select it (being the symbols SYSTEMPORT, BNXT, and BCMGENET). Link: https://lkml.kernel.org/r/alpine.DEB.2.21.1907021810220.13058@ramsan.of.borg/ Fixes: 4f75da3666c0 ("linux/dim: Move implementation to .c files") Reported-by: Geert Uytterhoeven Signed-off-by: Randy Dunlap Cc: Uwe Kleine-König Cc: Tal Gilboa Cc: Saeed Mahameed Cc: netdev@vger.kernel.org Cc: linux-rdma@vger.kernel.org Cc: "David S. Miller" Cc: Jakub Kicinski Cc: Doug Ledford Cc: Jason Gunthorpe Cc: Leon Romanovsky Cc: Or Gerlitz Cc: Sagi Grimberg Acked-by: Florian Fainelli Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/Kconfig | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig index e24f5d2b6afe..53055ce5dfd6 100644 --- a/drivers/net/ethernet/broadcom/Kconfig +++ b/drivers/net/ethernet/broadcom/Kconfig @@ -8,7 +8,6 @@ config NET_VENDOR_BROADCOM default y depends on (SSB_POSSIBLE && HAS_DMA) || PCI || BCM63XX || \ SIBYTE_SB1xxx_SOC - select DIMLIB ---help--- If you have a network (Ethernet) chipset belonging to this class, say Y. @@ -69,6 +68,7 @@ config BCMGENET select FIXED_PHY select BCM7XXX_PHY select MDIO_BCM_UNIMAC + select DIMLIB help This driver supports the built-in Ethernet MACs found in the Broadcom BCM7xxx Set Top Box family chipset. @@ -188,6 +188,7 @@ config SYSTEMPORT select MII select PHYLIB select FIXED_PHY + select DIMLIB help This driver supports the built-in Ethernet MACs found in the Broadcom BCM7xxx Set Top Box family chipset using an internal @@ -200,6 +201,7 @@ config BNXT select LIBCRC32C select NET_DEVLINK select PAGE_POOL + select DIMLIB ---help--- This driver supports Broadcom NetXtreme-C/E 10/25/40/50 gigabit Ethernet cards. To compile this driver as a module, choose M here: From b31141d390f1d47f1f3e1aef04909dc48e894611 Mon Sep 17 00:00:00 2001 From: Nishad Kamdar Date: Sat, 12 Oct 2019 17:48:56 +0530 Subject: [PATCH 097/145] net: dsa: microchip: Use the correct style for SPDX License Identifier This patch corrects the SPDX License Identifier style in header files related to Distributed Switch Architecture drivers for Microchip KSZ series switch support. For C header files Documentation/process/license-rules.rst mandates C-like comments (opposed to C source files where C++ style should be used) Changes made by using a script provided by Joe Perches here: https://lkml.org/lkml/2019/2/7/46. Suggested-by: Joe Perches Signed-off-by: Nishad Kamdar Signed-off-by: David S. Miller --- drivers/net/dsa/microchip/ksz9477_reg.h | 4 ++-- drivers/net/dsa/microchip/ksz_common.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/dsa/microchip/ksz9477_reg.h b/drivers/net/dsa/microchip/ksz9477_reg.h index 2938e892b631..16939f29faa5 100644 --- a/drivers/net/dsa/microchip/ksz9477_reg.h +++ b/drivers/net/dsa/microchip/ksz9477_reg.h @@ -1,5 +1,5 @@ -/* SPDX-License-Identifier: GPL-2.0 - * +/* SPDX-License-Identifier: GPL-2.0 */ +/* * Microchip KSZ9477 register definitions * * Copyright (C) 2017-2018 Microchip Technology Inc. diff --git a/drivers/net/dsa/microchip/ksz_common.h b/drivers/net/dsa/microchip/ksz_common.h index dd60d0837fc6..64c2677693d2 100644 --- a/drivers/net/dsa/microchip/ksz_common.h +++ b/drivers/net/dsa/microchip/ksz_common.h @@ -1,5 +1,5 @@ -/* SPDX-License-Identifier: GPL-2.0 - * Microchip switch driver common header +/* SPDX-License-Identifier: GPL-2.0 */ +/* Microchip switch driver common header * * Copyright (C) 2017-2019 Microchip Technology Inc. */ From a03681dd5d1bf7b0ed3aad3162f7810af4c4e05b Mon Sep 17 00:00:00 2001 From: Nishad Kamdar Date: Sat, 12 Oct 2019 18:42:28 +0530 Subject: [PATCH 098/145] net: cavium: Use the correct style for SPDX License Identifier This patch corrects the SPDX License Identifier style in header files related to Cavium Ethernet drivers. For C header files Documentation/process/license-rules.rst mandates C-like comments (opposed to C source files where C++ style should be used) Changes made by using a script provided by Joe Perches here: https://lkml.org/lkml/2019/2/7/46. Suggested-by: Joe Perches Signed-off-by: Nishad Kamdar Signed-off-by: David S. Miller --- drivers/net/ethernet/cavium/common/cavium_ptp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/cavium/common/cavium_ptp.h b/drivers/net/ethernet/cavium/common/cavium_ptp.h index be2bafc7beeb..a04eccbc78e8 100644 --- a/drivers/net/ethernet/cavium/common/cavium_ptp.h +++ b/drivers/net/ethernet/cavium/common/cavium_ptp.h @@ -1,4 +1,4 @@ -// SPDX-License-Identifier: GPL-2.0 +/* SPDX-License-Identifier: GPL-2.0 */ /* cavium_ptp.h - PTP 1588 clock on Cavium hardware * Copyright (c) 2003-2015, 2017 Cavium, Inc. */ From dedc5a08da07874c6e0d411e7f39c5c2cf137014 Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Sat, 12 Oct 2019 13:55:06 +0200 Subject: [PATCH 099/145] net: avoid errors when trying to pop MLPS header on non-MPLS packets the following script: # tc qdisc add dev eth0 clsact # tc filter add dev eth0 egress matchall action mpls pop implicitly makes the kernel drop all packets transmitted by eth0, if they don't have a MPLS header. This behavior is uncommon: other encapsulations (like VLAN) just let the packet pass unmodified. Since the result of MPLS 'pop' operation would be the same regardless of the presence / absence of MPLS header(s) in the original packet, we can let skb_mpls_pop() return 0 when dealing with non-MPLS packets. For the OVS use-case, this is acceptable because __ovs_nla_copy_actions() already ensures that MPLS 'pop' operation only occurs with packets having an MPLS Ethernet type (and there are no other callers in current code, so the semantic change should be ok). v2: better documentation of use-cases for skb_mpls_pop(), thanks to Simon Horman Fixes: 2a2ea50870ba ("net: sched: add mpls manipulation actions to TC") Reviewed-by: Simon Horman Acked-by: John Hurley Signed-off-by: Davide Caratti Signed-off-by: David S. Miller --- net/core/skbuff.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 8c178703467b..03b6809ebde4 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -5536,7 +5536,7 @@ int skb_mpls_pop(struct sk_buff *skb, __be16 next_proto) int err; if (unlikely(!eth_p_mpls(skb->protocol))) - return -EINVAL; + return 0; err = skb_ensure_writable(skb, skb->mac_len + MPLS_HLEN); if (unlikely(err)) From fa4e0f8855fcba600e0be2575ee29c69166f74bd Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Sat, 12 Oct 2019 13:55:07 +0200 Subject: [PATCH 100/145] net/sched: fix corrupted L2 header with MPLS 'push' and 'pop' actions the following script: # tc qdisc add dev eth0 clsact # tc filter add dev eth0 egress protocol ip matchall \ > action mpls push protocol mpls_uc label 0x355aa bos 1 causes corruption of all IP packets transmitted by eth0. On TC egress, we can't rely on the value of skb->mac_len, because it's 0 and a MPLS 'push' operation will result in an overwrite of the first 4 octets in the packet L2 header (e.g. the Destination Address if eth0 is an Ethernet); the same error pattern is present also in the MPLS 'pop' operation. Fix this error in act_mpls data plane, computing 'mac_len' as the difference between the network header and the mac header (when not at TC ingress), and use it in MPLS 'push'/'pop' core functions. v2: unbreak 'make htmldocs' because of missing documentation of 'mac_len' in skb_mpls_pop(), reported by kbuild test robot CC: Lorenzo Bianconi Fixes: 2a2ea50870ba ("net: sched: add mpls manipulation actions to TC") Reviewed-by: Simon Horman Acked-by: John Hurley Signed-off-by: Davide Caratti Signed-off-by: David S. Miller --- include/linux/skbuff.h | 5 +++-- net/core/skbuff.c | 19 +++++++++++-------- net/openvswitch/actions.c | 5 +++-- net/sched/act_mpls.c | 12 ++++++++---- 4 files changed, 25 insertions(+), 16 deletions(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 4351577b14d7..7914fdaf4226 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3510,8 +3510,9 @@ int skb_ensure_writable(struct sk_buff *skb, int write_len); int __skb_vlan_pop(struct sk_buff *skb, u16 *vlan_tci); int skb_vlan_pop(struct sk_buff *skb); int skb_vlan_push(struct sk_buff *skb, __be16 vlan_proto, u16 vlan_tci); -int skb_mpls_push(struct sk_buff *skb, __be32 mpls_lse, __be16 mpls_proto); -int skb_mpls_pop(struct sk_buff *skb, __be16 next_proto); +int skb_mpls_push(struct sk_buff *skb, __be32 mpls_lse, __be16 mpls_proto, + int mac_len); +int skb_mpls_pop(struct sk_buff *skb, __be16 next_proto, int mac_len); int skb_mpls_update_lse(struct sk_buff *skb, __be32 mpls_lse); int skb_mpls_dec_ttl(struct sk_buff *skb); struct sk_buff *pskb_extract(struct sk_buff *skb, int off, int to_copy, diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 03b6809ebde4..867e61df00db 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -5477,12 +5477,14 @@ static void skb_mod_eth_type(struct sk_buff *skb, struct ethhdr *hdr, * @skb: buffer * @mpls_lse: MPLS label stack entry to push * @mpls_proto: ethertype of the new MPLS header (expects 0x8847 or 0x8848) + * @mac_len: length of the MAC header * * Expects skb->data at mac header. * * Returns 0 on success, -errno otherwise. */ -int skb_mpls_push(struct sk_buff *skb, __be32 mpls_lse, __be16 mpls_proto) +int skb_mpls_push(struct sk_buff *skb, __be32 mpls_lse, __be16 mpls_proto, + int mac_len) { struct mpls_shim_hdr *lse; int err; @@ -5499,15 +5501,15 @@ int skb_mpls_push(struct sk_buff *skb, __be32 mpls_lse, __be16 mpls_proto) return err; if (!skb->inner_protocol) { - skb_set_inner_network_header(skb, skb->mac_len); + skb_set_inner_network_header(skb, mac_len); skb_set_inner_protocol(skb, skb->protocol); } skb_push(skb, MPLS_HLEN); memmove(skb_mac_header(skb) - MPLS_HLEN, skb_mac_header(skb), - skb->mac_len); + mac_len); skb_reset_mac_header(skb); - skb_set_network_header(skb, skb->mac_len); + skb_set_network_header(skb, mac_len); lse = mpls_hdr(skb); lse->label_stack_entry = mpls_lse; @@ -5526,29 +5528,30 @@ EXPORT_SYMBOL_GPL(skb_mpls_push); * * @skb: buffer * @next_proto: ethertype of header after popped MPLS header + * @mac_len: length of the MAC header * * Expects skb->data at mac header. * * Returns 0 on success, -errno otherwise. */ -int skb_mpls_pop(struct sk_buff *skb, __be16 next_proto) +int skb_mpls_pop(struct sk_buff *skb, __be16 next_proto, int mac_len) { int err; if (unlikely(!eth_p_mpls(skb->protocol))) return 0; - err = skb_ensure_writable(skb, skb->mac_len + MPLS_HLEN); + err = skb_ensure_writable(skb, mac_len + MPLS_HLEN); if (unlikely(err)) return err; skb_postpull_rcsum(skb, mpls_hdr(skb), MPLS_HLEN); memmove(skb_mac_header(skb) + MPLS_HLEN, skb_mac_header(skb), - skb->mac_len); + mac_len); __skb_pull(skb, MPLS_HLEN); skb_reset_mac_header(skb); - skb_set_network_header(skb, skb->mac_len); + skb_set_network_header(skb, mac_len); if (skb->dev && skb->dev->type == ARPHRD_ETHER) { struct ethhdr *hdr; diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 3572e11b6f21..1c77f520f474 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -165,7 +165,8 @@ static int push_mpls(struct sk_buff *skb, struct sw_flow_key *key, { int err; - err = skb_mpls_push(skb, mpls->mpls_lse, mpls->mpls_ethertype); + err = skb_mpls_push(skb, mpls->mpls_lse, mpls->mpls_ethertype, + skb->mac_len); if (err) return err; @@ -178,7 +179,7 @@ static int pop_mpls(struct sk_buff *skb, struct sw_flow_key *key, { int err; - err = skb_mpls_pop(skb, ethertype); + err = skb_mpls_pop(skb, ethertype, skb->mac_len); if (err) return err; diff --git a/net/sched/act_mpls.c b/net/sched/act_mpls.c index e168df0e008a..4cf6c553bb0b 100644 --- a/net/sched/act_mpls.c +++ b/net/sched/act_mpls.c @@ -55,7 +55,7 @@ static int tcf_mpls_act(struct sk_buff *skb, const struct tc_action *a, struct tcf_mpls *m = to_mpls(a); struct tcf_mpls_params *p; __be32 new_lse; - int ret; + int ret, mac_len; tcf_lastuse_update(&m->tcf_tm); bstats_cpu_update(this_cpu_ptr(m->common.cpu_bstats), skb); @@ -63,8 +63,12 @@ static int tcf_mpls_act(struct sk_buff *skb, const struct tc_action *a, /* Ensure 'data' points at mac_header prior calling mpls manipulating * functions. */ - if (skb_at_tc_ingress(skb)) + if (skb_at_tc_ingress(skb)) { skb_push_rcsum(skb, skb->mac_len); + mac_len = skb->mac_len; + } else { + mac_len = skb_network_header(skb) - skb_mac_header(skb); + } ret = READ_ONCE(m->tcf_action); @@ -72,12 +76,12 @@ static int tcf_mpls_act(struct sk_buff *skb, const struct tc_action *a, switch (p->tcfm_action) { case TCA_MPLS_ACT_POP: - if (skb_mpls_pop(skb, p->tcfm_proto)) + if (skb_mpls_pop(skb, p->tcfm_proto, mac_len)) goto drop; break; case TCA_MPLS_ACT_PUSH: new_lse = tcf_mpls_get_lse(NULL, p, !eth_p_mpls(skb->protocol)); - if (skb_mpls_push(skb, new_lse, p->tcfm_proto)) + if (skb_mpls_push(skb, new_lse, p->tcfm_proto, mac_len)) goto drop; break; case TCA_MPLS_ACT_MODIFY: From 4e3eff5beafa5c97cdf2af2181f46479941a7f45 Mon Sep 17 00:00:00 2001 From: MarkLee Date: Mon, 14 Oct 2019 15:15:17 +0800 Subject: [PATCH 101/145] net: ethernet: mediatek: Fix MT7629 missing GMII mode support In the original design, mtk_phy_connect function will set ge_mode=1 if phy-mode is GMII(PHY_INTERFACE_MODE_GMII) and then set the correct ge_mode to ETHSYS_SYSCFG0 register. This logic was broken after apply mediatek PHYLINK patch(Fixes tag), the new mtk_mac_config function will not set ge_mode=1 for GMII mode hence the final ETHSYS_SYSCFG0 setting will be incorrect for mt7629 GMII mode. This patch add the missing logic back to fix it. Fixes: b8fc9f30821e ("net: ethernet: mediatek: Add basic PHYLINK support") Signed-off-by: MarkLee Signed-off-by: David S. Miller --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index c61069340f4f..703adb96429e 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -261,6 +261,7 @@ static void mtk_mac_config(struct phylink_config *config, unsigned int mode, ge_mode = 0; switch (state->interface) { case PHY_INTERFACE_MODE_MII: + case PHY_INTERFACE_MODE_GMII: ge_mode = 1; break; case PHY_INTERFACE_MODE_REVMII: From 2618500dd370da413cb1f616111e1bd8d9f5f94f Mon Sep 17 00:00:00 2001 From: MarkLee Date: Mon, 14 Oct 2019 15:15:18 +0800 Subject: [PATCH 102/145] arm: dts: mediatek: Update mt7629 dts to reflect the latest dt-binding * Removes mediatek,physpeed property from dtsi that is useless in PHYLINK * Use the fixed-link property speed = <2500> to set the phy in 2.5Gbit. * Set gmac1 to gmii mode that connect to a internal gphy Signed-off-by: MarkLee Signed-off-by: David S. Miller --- arch/arm/boot/dts/mt7629-rfb.dts | 13 ++++++++++++- arch/arm/boot/dts/mt7629.dtsi | 2 -- 2 files changed, 12 insertions(+), 3 deletions(-) diff --git a/arch/arm/boot/dts/mt7629-rfb.dts b/arch/arm/boot/dts/mt7629-rfb.dts index 3621b7d2b22a..9980c10c6e29 100644 --- a/arch/arm/boot/dts/mt7629-rfb.dts +++ b/arch/arm/boot/dts/mt7629-rfb.dts @@ -66,9 +66,21 @@ pinctrl-1 = <&ephy_leds_pins>; status = "okay"; + gmac0: mac@0 { + compatible = "mediatek,eth-mac"; + reg = <0>; + phy-mode = "2500base-x"; + fixed-link { + speed = <2500>; + full-duplex; + pause; + }; + }; + gmac1: mac@1 { compatible = "mediatek,eth-mac"; reg = <1>; + phy-mode = "gmii"; phy-handle = <&phy0>; }; @@ -78,7 +90,6 @@ phy0: ethernet-phy@0 { reg = <0>; - phy-mode = "gmii"; }; }; }; diff --git a/arch/arm/boot/dts/mt7629.dtsi b/arch/arm/boot/dts/mt7629.dtsi index 9608bc2ccb3f..867b88103b9d 100644 --- a/arch/arm/boot/dts/mt7629.dtsi +++ b/arch/arm/boot/dts/mt7629.dtsi @@ -468,14 +468,12 @@ compatible = "mediatek,mt7629-sgmiisys", "syscon"; reg = <0x1b128000 0x3000>; #clock-cells = <1>; - mediatek,physpeed = "2500"; }; sgmiisys1: syscon@1b130000 { compatible = "mediatek,mt7629-sgmiisys", "syscon"; reg = <0x1b130000 0x3000>; #clock-cells = <1>; - mediatek,physpeed = "2500"; }; }; }; From cab209e571a9375f7dc6db69a6c40d2d98e57e3b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 14 Oct 2019 06:47:57 -0700 Subject: [PATCH 103/145] tcp: fix a possible lockdep splat in tcp_done() syzbot found that if __inet_inherit_port() returns an error, we call tcp_done() after inet_csk_prepare_forced_close(), meaning the socket lock is no longer held. We might fix this in a different way in net-next, but for 5.4 it seems safer to relax the lockdep check. Fixes: d983ea6f16b8 ("tcp: add rcu protection around tp->fastopen_rsk") Signed-off-by: Eric Dumazet Reported-by: syzbot Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b2ac4f074e2d..42187a3b82f4 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -3842,8 +3842,12 @@ void tcp_done(struct sock *sk) { struct request_sock *req; - req = rcu_dereference_protected(tcp_sk(sk)->fastopen_rsk, - lockdep_sock_is_held(sk)); + /* We might be called with a new socket, after + * inet_csk_prepare_forced_close() has been called + * so we can not use lockdep_sock_is_held(sk) + */ + req = rcu_dereference_protected(tcp_sk(sk)->fastopen_rsk, 1); + if (sk->sk_state == TCP_SYN_SENT || sk->sk_state == TCP_SYN_RECV) TCP_INC_STATS(sock_net(sk), TCP_MIB_ATTEMPTFAILS); From b790b5549bdf498ab0ecc5632610a9149532fa38 Mon Sep 17 00:00:00 2001 From: Nishad Kamdar Date: Mon, 14 Oct 2019 21:51:20 +0530 Subject: [PATCH 104/145] net: dsa: sja1105: Use the correct style for SPDX License Identifier This patch corrects the SPDX License Identifier style in header files related to Distributed Switch Architecture drivers for NXP SJA1105 series Ethernet switch support. It uses an expilict block comment for the SPDX License Identifier. Changes made by using a script provided by Joe Perches here: https://lkml.org/lkml/2019/2/7/46. Suggested-by: Joe Perches Signed-off-by: Nishad Kamdar Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/dsa/sja1105/sja1105.h | 4 ++-- drivers/net/dsa/sja1105/sja1105_dynamic_config.h | 4 ++-- drivers/net/dsa/sja1105/sja1105_ptp.h | 4 ++-- drivers/net/dsa/sja1105/sja1105_static_config.h | 4 ++-- drivers/net/dsa/sja1105/sja1105_tas.h | 4 ++-- 5 files changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/net/dsa/sja1105/sja1105.h b/drivers/net/dsa/sja1105/sja1105.h index e53e494c22e0..fbb564c3beb8 100644 --- a/drivers/net/dsa/sja1105/sja1105.h +++ b/drivers/net/dsa/sja1105/sja1105.h @@ -1,5 +1,5 @@ -/* SPDX-License-Identifier: GPL-2.0 - * Copyright (c) 2018, Sensor-Technik Wiedemann GmbH +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2018, Sensor-Technik Wiedemann GmbH * Copyright (c) 2018-2019, Vladimir Oltean */ #ifndef _SJA1105_H diff --git a/drivers/net/dsa/sja1105/sja1105_dynamic_config.h b/drivers/net/dsa/sja1105/sja1105_dynamic_config.h index 740dadf43f01..1fc0d13dc623 100644 --- a/drivers/net/dsa/sja1105/sja1105_dynamic_config.h +++ b/drivers/net/dsa/sja1105/sja1105_dynamic_config.h @@ -1,5 +1,5 @@ -/* SPDX-License-Identifier: GPL-2.0 - * Copyright (c) 2019, Vladimir Oltean +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2019, Vladimir Oltean */ #ifndef _SJA1105_DYNAMIC_CONFIG_H #define _SJA1105_DYNAMIC_CONFIG_H diff --git a/drivers/net/dsa/sja1105/sja1105_ptp.h b/drivers/net/dsa/sja1105/sja1105_ptp.h index af456b0a4d27..394e12a6ad59 100644 --- a/drivers/net/dsa/sja1105/sja1105_ptp.h +++ b/drivers/net/dsa/sja1105/sja1105_ptp.h @@ -1,5 +1,5 @@ -/* SPDX-License-Identifier: GPL-2.0 - * Copyright (c) 2019, Vladimir Oltean +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2019, Vladimir Oltean */ #ifndef _SJA1105_PTP_H #define _SJA1105_PTP_H diff --git a/drivers/net/dsa/sja1105/sja1105_static_config.h b/drivers/net/dsa/sja1105/sja1105_static_config.h index 7f87022a2d61..f4a5c5c04311 100644 --- a/drivers/net/dsa/sja1105/sja1105_static_config.h +++ b/drivers/net/dsa/sja1105/sja1105_static_config.h @@ -1,5 +1,5 @@ -/* SPDX-License-Identifier: BSD-3-Clause - * Copyright (c) 2016-2018, NXP Semiconductors +/* SPDX-License-Identifier: BSD-3-Clause */ +/* Copyright (c) 2016-2018, NXP Semiconductors * Copyright (c) 2018-2019, Vladimir Oltean */ #ifndef _SJA1105_STATIC_CONFIG_H diff --git a/drivers/net/dsa/sja1105/sja1105_tas.h b/drivers/net/dsa/sja1105/sja1105_tas.h index 0b803c30e640..0aad212d88b2 100644 --- a/drivers/net/dsa/sja1105/sja1105_tas.h +++ b/drivers/net/dsa/sja1105/sja1105_tas.h @@ -1,5 +1,5 @@ -/* SPDX-License-Identifier: GPL-2.0 - * Copyright (c) 2019, Vladimir Oltean +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2019, Vladimir Oltean */ #ifndef _SJA1105_TAS_H #define _SJA1105_TAS_H From 39f13ea2f61b439ebe0060393e9c39925c9ee28c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 14 Oct 2019 11:22:30 -0700 Subject: [PATCH 105/145] net: avoid potential infinite loop in tc_ctl_action() tc_ctl_action() has the ability to loop forever if tcf_action_add() returns -EAGAIN. This special case has been done in case a module needed to be loaded, but it turns out that tcf_add_notify() could also return -EAGAIN if the socket sk_rcvbuf limit is hit. We need to separate the two cases, and only loop for the module loading case. While we are at it, add a limit of 10 attempts since unbounded loops are always scary. syzbot repro was something like : socket(PF_NETLINK, SOCK_RAW|SOCK_NONBLOCK, NETLINK_ROUTE) = 3 write(3, ..., 38) = 38 setsockopt(3, SOL_SOCKET, SO_RCVBUF, [0], 4) = 0 sendmsg(3, {msg_name(0)=NULL, msg_iov(1)=[{..., 388}], msg_controllen=0, msg_flags=0x10}, ...) NMI backtrace for cpu 0 CPU: 0 PID: 1054 Comm: khungtaskd Not tainted 5.4.0-rc1+ #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x172/0x1f0 lib/dump_stack.c:113 nmi_cpu_backtrace.cold+0x70/0xb2 lib/nmi_backtrace.c:101 nmi_trigger_cpumask_backtrace+0x23b/0x28b lib/nmi_backtrace.c:62 arch_trigger_cpumask_backtrace+0x14/0x20 arch/x86/kernel/apic/hw_nmi.c:38 trigger_all_cpu_backtrace include/linux/nmi.h:146 [inline] check_hung_uninterruptible_tasks kernel/hung_task.c:205 [inline] watchdog+0x9d0/0xef0 kernel/hung_task.c:289 kthread+0x361/0x430 kernel/kthread.c:255 ret_from_fork+0x24/0x30 arch/x86/entry/entry_64.S:352 Sending NMI from CPU 0 to CPUs 1: NMI backtrace for cpu 1 CPU: 1 PID: 8859 Comm: syz-executor910 Not tainted 5.4.0-rc1+ #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:arch_local_save_flags arch/x86/include/asm/paravirt.h:751 [inline] RIP: 0010:lockdep_hardirqs_off+0x1df/0x2e0 kernel/locking/lockdep.c:3453 Code: 5c 08 00 00 5b 41 5c 41 5d 5d c3 48 c7 c0 58 1d f3 88 48 ba 00 00 00 00 00 fc ff df 48 c1 e8 03 80 3c 10 00 0f 85 d3 00 00 00 <48> 83 3d 21 9e 99 07 00 0f 84 b9 00 00 00 9c 58 0f 1f 44 00 00 f6 RSP: 0018:ffff8880a6f3f1b8 EFLAGS: 00000046 RAX: 1ffffffff11e63ab RBX: ffff88808c9c6080 RCX: 0000000000000000 RDX: dffffc0000000000 RSI: 0000000000000000 RDI: ffff88808c9c6914 RBP: ffff8880a6f3f1d0 R08: ffff88808c9c6080 R09: fffffbfff16be5d1 R10: fffffbfff16be5d0 R11: 0000000000000003 R12: ffffffff8746591f R13: ffff88808c9c6080 R14: ffffffff8746591f R15: 0000000000000003 FS: 00000000011e4880(0000) GS:ffff8880ae900000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffffffffff600400 CR3: 00000000a8920000 CR4: 00000000001406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: trace_hardirqs_off+0x62/0x240 kernel/trace/trace_preemptirq.c:45 __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:108 [inline] _raw_spin_lock_irqsave+0x6f/0xcd kernel/locking/spinlock.c:159 __wake_up_common_lock+0xc8/0x150 kernel/sched/wait.c:122 __wake_up+0xe/0x10 kernel/sched/wait.c:142 netlink_unlock_table net/netlink/af_netlink.c:466 [inline] netlink_unlock_table net/netlink/af_netlink.c:463 [inline] netlink_broadcast_filtered+0x705/0xb80 net/netlink/af_netlink.c:1514 netlink_broadcast+0x3a/0x50 net/netlink/af_netlink.c:1534 rtnetlink_send+0xdd/0x110 net/core/rtnetlink.c:714 tcf_add_notify net/sched/act_api.c:1343 [inline] tcf_action_add+0x243/0x370 net/sched/act_api.c:1362 tc_ctl_action+0x3b5/0x4bc net/sched/act_api.c:1410 rtnetlink_rcv_msg+0x463/0xb00 net/core/rtnetlink.c:5386 netlink_rcv_skb+0x177/0x450 net/netlink/af_netlink.c:2477 rtnetlink_rcv+0x1d/0x30 net/core/rtnetlink.c:5404 netlink_unicast_kernel net/netlink/af_netlink.c:1302 [inline] netlink_unicast+0x531/0x710 net/netlink/af_netlink.c:1328 netlink_sendmsg+0x8a5/0xd60 net/netlink/af_netlink.c:1917 sock_sendmsg_nosec net/socket.c:637 [inline] sock_sendmsg+0xd7/0x130 net/socket.c:657 ___sys_sendmsg+0x803/0x920 net/socket.c:2311 __sys_sendmsg+0x105/0x1d0 net/socket.c:2356 __do_sys_sendmsg net/socket.c:2365 [inline] __se_sys_sendmsg net/socket.c:2363 [inline] __x64_sys_sendmsg+0x78/0xb0 net/socket.c:2363 do_syscall_64+0xfa/0x760 arch/x86/entry/common.c:290 entry_SYSCALL_64_after_hwframe+0x49/0xbe RIP: 0033:0x440939 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Dumazet Reported-by: syzbot+cf0adbb9c28c8866c788@syzkaller.appspotmail.com Signed-off-by: David S. Miller --- net/sched/act_api.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 4e7429c6f864..69d4676a402f 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -1353,11 +1353,16 @@ static int tcf_action_add(struct net *net, struct nlattr *nla, struct netlink_ext_ack *extack) { size_t attr_size = 0; - int ret = 0; + int loop, ret; struct tc_action *actions[TCA_ACT_MAX_PRIO] = {}; - ret = tcf_action_init(net, NULL, nla, NULL, NULL, ovr, 0, actions, - &attr_size, true, extack); + for (loop = 0; loop < 10; loop++) { + ret = tcf_action_init(net, NULL, nla, NULL, NULL, ovr, 0, + actions, &attr_size, true, extack); + if (ret != -EAGAIN) + break; + } + if (ret < 0) return ret; ret = tcf_add_notify(net, n, actions, portid, attr_size, extack); @@ -1407,11 +1412,8 @@ static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, */ if (n->nlmsg_flags & NLM_F_REPLACE) ovr = 1; -replay: ret = tcf_action_add(net, tca[TCA_ACT_TAB], n, portid, ovr, extack); - if (ret == -EAGAIN) - goto replay; break; case RTM_DELACTION: ret = tca_action_gd(net, tca[TCA_ACT_TAB], n, From 28aa7c86c2b49f659c8460a89e53b506c45979bb Mon Sep 17 00:00:00 2001 From: Vinicius Costa Gomes Date: Mon, 14 Oct 2019 13:38:22 -0700 Subject: [PATCH 106/145] sched: etf: Fix ordering of packets with same txtime When a application sends many packets with the same txtime, they may be transmitted out of order (different from the order in which they were enqueued). This happens because when inserting elements into the tree, when the txtime of two packets are the same, the new packet is inserted at the left side of the tree, causing the reordering. The only effect of this change should be that packets with the same txtime will be transmitted in the order they are enqueued. The application in question (the AVTP GStreamer plugin, still in development) is sending video traffic, in which each video frame have a single presentation time, the problem is that when packetizing, multiple packets end up with the same txtime. The receiving side was rejecting packets because they were being received out of order. Fixes: 25db26a91364 ("net/sched: Introduce the ETF Qdisc") Reported-by: Ederson de Souza Signed-off-by: Vinicius Costa Gomes Signed-off-by: David S. Miller --- net/sched/sch_etf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/sch_etf.c b/net/sched/sch_etf.c index cebfb65d8556..b1da5589a0c6 100644 --- a/net/sched/sch_etf.c +++ b/net/sched/sch_etf.c @@ -177,7 +177,7 @@ static int etf_enqueue_timesortedlist(struct sk_buff *nskb, struct Qdisc *sch, parent = *p; skb = rb_to_skb(parent); - if (ktime_after(txtime, skb->tstamp)) { + if (ktime_compare(txtime, skb->tstamp) >= 0) { p = &parent->rb_right; leftmost = false; } else { From 63dfb7938b13fa2c2fbcb45f34d065769eb09414 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 15 Oct 2019 15:24:38 +0800 Subject: [PATCH 107/145] sctp: change sctp_prot .no_autobind with true syzbot reported a memory leak: BUG: memory leak, unreferenced object 0xffff888120b3d380 (size 64): backtrace: [...] slab_alloc mm/slab.c:3319 [inline] [...] kmem_cache_alloc+0x13f/0x2c0 mm/slab.c:3483 [...] sctp_bucket_create net/sctp/socket.c:8523 [inline] [...] sctp_get_port_local+0x189/0x5a0 net/sctp/socket.c:8270 [...] sctp_do_bind+0xcc/0x200 net/sctp/socket.c:402 [...] sctp_bindx_add+0x4b/0xd0 net/sctp/socket.c:497 [...] sctp_setsockopt_bindx+0x156/0x1b0 net/sctp/socket.c:1022 [...] sctp_setsockopt net/sctp/socket.c:4641 [inline] [...] sctp_setsockopt+0xaea/0x2dc0 net/sctp/socket.c:4611 [...] sock_common_setsockopt+0x38/0x50 net/core/sock.c:3147 [...] __sys_setsockopt+0x10f/0x220 net/socket.c:2084 [...] __do_sys_setsockopt net/socket.c:2100 [inline] It was caused by when sending msgs without binding a port, in the path: inet_sendmsg() -> inet_send_prepare() -> inet_autobind() -> .get_port/sctp_get_port(), sp->bind_hash will be set while bp->port is not. Later when binding another port by sctp_setsockopt_bindx(), a new bucket will be created as bp->port is not set. sctp's autobind is supposed to call sctp_autobind() where it does all things including setting bp->port. Since sctp_autobind() is called in sctp_sendmsg() if the sk is not yet bound, it should have skipped the auto bind. THis patch is to avoid calling inet_autobind() in inet_send_prepare() by changing sctp_prot .no_autobind with true, also remove the unused .get_port. Reported-by: syzbot+d44f7bbebdea49dbc84a@syzkaller.appspotmail.com Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- net/sctp/socket.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 939b8d2595bc..5ca0ec0e823c 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -9500,7 +9500,7 @@ struct proto sctp_prot = { .backlog_rcv = sctp_backlog_rcv, .hash = sctp_hash, .unhash = sctp_unhash, - .get_port = sctp_get_port, + .no_autobind = true, .obj_size = sizeof(struct sctp_sock), .useroffset = offsetof(struct sctp_sock, subscribe), .usersize = offsetof(struct sctp_sock, initmsg) - @@ -9542,7 +9542,7 @@ struct proto sctpv6_prot = { .backlog_rcv = sctp_backlog_rcv, .hash = sctp_hash, .unhash = sctp_unhash, - .get_port = sctp_get_port, + .no_autobind = true, .obj_size = sizeof(struct sctp6_sock), .useroffset = offsetof(struct sctp6_sock, sctp.subscribe), .usersize = offsetof(struct sctp6_sock, sctp.initmsg) - From 61c1d33daf7b5146f44d4363b3322f8cda6a6c43 Mon Sep 17 00:00:00 2001 From: Thomas Bogendoerfer Date: Tue, 15 Oct 2019 16:42:45 +0200 Subject: [PATCH 108/145] net: i82596: fix dma_alloc_attr for sni_82596 Commit 7f683b920479 ("i825xx: switch to switch to dma_alloc_attrs") switched dma allocation over to dma_alloc_attr, but didn't convert the SNI part to request consistent DMA memory. This broke sni_82596 since driver doesn't do dma_cache_sync for performance reasons. Fix this by using different DMA_ATTRs for lasi_82596 and sni_82596. Fixes: 7f683b920479 ("i825xx: switch to switch to dma_alloc_attrs") Signed-off-by: Thomas Bogendoerfer Signed-off-by: David S. Miller --- drivers/net/ethernet/i825xx/lasi_82596.c | 4 +++- drivers/net/ethernet/i825xx/lib82596.c | 4 ++-- drivers/net/ethernet/i825xx/sni_82596.c | 4 +++- 3 files changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/i825xx/lasi_82596.c b/drivers/net/ethernet/i825xx/lasi_82596.c index 211c5f74b4c8..aec7e98bcc85 100644 --- a/drivers/net/ethernet/i825xx/lasi_82596.c +++ b/drivers/net/ethernet/i825xx/lasi_82596.c @@ -96,6 +96,8 @@ #define OPT_SWAP_PORT 0x0001 /* Need to wordswp on the MPU port */ +#define LIB82596_DMA_ATTR DMA_ATTR_NON_CONSISTENT + #define DMA_WBACK(ndev, addr, len) \ do { dma_cache_sync((ndev)->dev.parent, (void *)addr, len, DMA_TO_DEVICE); } while (0) @@ -200,7 +202,7 @@ static int __exit lan_remove_chip(struct parisc_device *pdev) unregister_netdev (dev); dma_free_attrs(&pdev->dev, sizeof(struct i596_private), lp->dma, - lp->dma_addr, DMA_ATTR_NON_CONSISTENT); + lp->dma_addr, LIB82596_DMA_ATTR); free_netdev (dev); return 0; } diff --git a/drivers/net/ethernet/i825xx/lib82596.c b/drivers/net/ethernet/i825xx/lib82596.c index 1274ad24d6af..f9742af7f142 100644 --- a/drivers/net/ethernet/i825xx/lib82596.c +++ b/drivers/net/ethernet/i825xx/lib82596.c @@ -1065,7 +1065,7 @@ static int i82596_probe(struct net_device *dev) dma = dma_alloc_attrs(dev->dev.parent, sizeof(struct i596_dma), &lp->dma_addr, GFP_KERNEL, - DMA_ATTR_NON_CONSISTENT); + LIB82596_DMA_ATTR); if (!dma) { printk(KERN_ERR "%s: Couldn't get shared memory\n", __FILE__); return -ENOMEM; @@ -1087,7 +1087,7 @@ static int i82596_probe(struct net_device *dev) i = register_netdev(dev); if (i) { dma_free_attrs(dev->dev.parent, sizeof(struct i596_dma), - dma, lp->dma_addr, DMA_ATTR_NON_CONSISTENT); + dma, lp->dma_addr, LIB82596_DMA_ATTR); return i; } diff --git a/drivers/net/ethernet/i825xx/sni_82596.c b/drivers/net/ethernet/i825xx/sni_82596.c index 6eb6c2ff7f09..6436a98c5953 100644 --- a/drivers/net/ethernet/i825xx/sni_82596.c +++ b/drivers/net/ethernet/i825xx/sni_82596.c @@ -24,6 +24,8 @@ static const char sni_82596_string[] = "snirm_82596"; +#define LIB82596_DMA_ATTR 0 + #define DMA_WBACK(priv, addr, len) do { } while (0) #define DMA_INV(priv, addr, len) do { } while (0) #define DMA_WBACK_INV(priv, addr, len) do { } while (0) @@ -152,7 +154,7 @@ static int sni_82596_driver_remove(struct platform_device *pdev) unregister_netdev(dev); dma_free_attrs(dev->dev.parent, sizeof(struct i596_private), lp->dma, - lp->dma_addr, DMA_ATTR_NON_CONSISTENT); + lp->dma_addr, LIB82596_DMA_ATTR); iounmap(lp->ca); iounmap(lp->mpu_port); free_netdev (dev); From 539825a536040e09a2f26eb54bef206e8e8086b5 Mon Sep 17 00:00:00 2001 From: "Ben Dooks (Codethink)" Date: Tue, 15 Oct 2019 17:15:58 +0100 Subject: [PATCH 109/145] davinci_cpdma: make cpdma_chan_split_pool static The cpdma_chan_split_pool() function is not used outside of the driver, so make it static to avoid the following sparse warning: drivers/net/ethernet/ti/davinci_cpdma.c:725:5: warning: symbol 'cpdma_chan_split_pool' was not declared. Should it be static? Signed-off-by: Ben Dooks Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/davinci_cpdma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c index a65edd2770e6..37ba708ac781 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.c +++ b/drivers/net/ethernet/ti/davinci_cpdma.c @@ -722,7 +722,7 @@ static void cpdma_chan_set_descs(struct cpdma_ctlr *ctlr, * cpdma_chan_split_pool - Splits ctrl pool between all channels. * Has to be called under ctlr lock */ -int cpdma_chan_split_pool(struct cpdma_ctlr *ctlr) +static int cpdma_chan_split_pool(struct cpdma_ctlr *ctlr) { int tx_per_ch_desc = 0, rx_per_ch_desc = 0; int free_rx_num = 0, free_tx_num = 0; From bad28d889caea00bc9415c59fa6a433a72eeaf5e Mon Sep 17 00:00:00 2001 From: "Ben Dooks (Codethink)" Date: Tue, 15 Oct 2019 17:17:48 +0100 Subject: [PATCH 110/145] net: stmmac: make tc_flow_parsers static The tc_flow_parsers is not used outside of the driver, so make it static to avoid the following sparse warning: drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c:516:3: warning: symbol 'tc_flow_parsers' was not declared. Should it be static? Signed-off-by: Ben Dooks Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c index e231098061b6..f9a9a9d82233 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c @@ -510,7 +510,7 @@ static struct stmmac_flow_entry *tc_find_flow(struct stmmac_priv *priv, return NULL; } -struct { +static struct { int (*fn)(struct stmmac_priv *priv, struct flow_cls_offload *cls, struct stmmac_flow_entry *entry); } tc_flow_parsers[] = { From efb86fede98cdc70b674692ff617b1162f642c49 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 15 Oct 2019 10:45:47 -0700 Subject: [PATCH 111/145] net: bcmgenet: Fix RGMII_MODE_EN value for GENET v1/2/3 The RGMII_MODE_EN bit value was 0 for GENET versions 1 through 3, and became 6 for GENET v4 and above, account for that difference. Fixes: aa09677cba42 ("net: bcmgenet: add MDIO routines") Signed-off-by: Florian Fainelli Acked-by: Doug Berger Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/genet/bcmgenet.h | 1 + drivers/net/ethernet/broadcom/genet/bcmmii.c | 6 +++++- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h index 4a8fc03d82fd..dbc69d8fa05f 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h @@ -366,6 +366,7 @@ struct bcmgenet_mib_counters { #define EXT_PWR_DOWN_PHY_EN (1 << 20) #define EXT_RGMII_OOB_CTRL 0x0C +#define RGMII_MODE_EN_V123 (1 << 0) #define RGMII_LINK (1 << 4) #define OOB_DISABLE (1 << 5) #define RGMII_MODE_EN (1 << 6) diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index 94d1dd5d56bf..e7c291bf4ed1 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -258,7 +258,11 @@ int bcmgenet_mii_config(struct net_device *dev, bool init) */ if (priv->ext_phy) { reg = bcmgenet_ext_readl(priv, EXT_RGMII_OOB_CTRL); - reg |= RGMII_MODE_EN | id_mode_dis; + reg |= id_mode_dis; + if (GENET_IS_V1(priv) || GENET_IS_V2(priv) || GENET_IS_V3(priv)) + reg |= RGMII_MODE_EN_V123; + else + reg |= RGMII_MODE_EN; bcmgenet_ext_writel(priv, reg, EXT_RGMII_OOB_CTRL); } From 77b6d09f4ae66d42cd63b121af67780ae3d1a5e9 Mon Sep 17 00:00:00 2001 From: Valentin Vidic Date: Tue, 15 Oct 2019 22:20:20 +0200 Subject: [PATCH 112/145] net: usb: sr9800: fix uninitialized local variable Make sure res does not contain random value if the call to sr_read_cmd fails for some reason. Reported-by: syzbot+f1842130bbcfb335bac1@syzkaller.appspotmail.com Signed-off-by: Valentin Vidic Signed-off-by: David S. Miller --- drivers/net/usb/sr9800.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/usb/sr9800.c b/drivers/net/usb/sr9800.c index c5d4a0060124..681e0def6356 100644 --- a/drivers/net/usb/sr9800.c +++ b/drivers/net/usb/sr9800.c @@ -335,7 +335,7 @@ static void sr_set_multicast(struct net_device *net) static int sr_mdio_read(struct net_device *net, int phy_id, int loc) { struct usbnet *dev = netdev_priv(net); - __le16 res; + __le16 res = 0; mutex_lock(&dev->phy_mutex); sr_set_sw_mii(dev); From bd74708cd979f4934f0744055ce3b47da68733ce Mon Sep 17 00:00:00 2001 From: Mahesh Bandewar Date: Wed, 16 Oct 2019 00:04:38 -0700 Subject: [PATCH 113/145] Revert "blackhole_netdev: fix syzkaller reported issue" This reverts commit b0818f80c8c1bc215bba276bd61c216014fab23b. Started seeing weird behavior after this patch especially in the IPv6 code path. Haven't root caused it, but since this was applied to net branch, taking a precautionary measure to revert it and look / analyze those failures Revert this now and I'll send a better fix after analysing / fixing the weirdness observed. CC: Eric Dumazet CC: Wei Wang CC: David S. Miller Signed-off-by: Mahesh Bandewar Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 7 +------ net/ipv6/route.c | 15 +++++++++------ 2 files changed, 10 insertions(+), 12 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 4c87594d1389..34ccef18b40e 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -6996,7 +6996,7 @@ static struct rtnl_af_ops inet6_ops __read_mostly = { int __init addrconf_init(void) { - struct inet6_dev *idev, *bdev; + struct inet6_dev *idev; int i, err; err = ipv6_addr_label_init(); @@ -7036,14 +7036,10 @@ int __init addrconf_init(void) */ rtnl_lock(); idev = ipv6_add_dev(init_net.loopback_dev); - bdev = ipv6_add_dev(blackhole_netdev); rtnl_unlock(); if (IS_ERR(idev)) { err = PTR_ERR(idev); goto errlo; - } else if (IS_ERR(bdev)) { - err = PTR_ERR(bdev); - goto errlo; } ip6_route_init_special_entries(); @@ -7128,7 +7124,6 @@ void addrconf_cleanup(void) addrconf_ifdown(dev, 1); } addrconf_ifdown(init_net.loopback_dev, 2); - addrconf_ifdown(blackhole_netdev, 2); /* * Check hash table. diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 742120728869..a63ff85fe141 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -155,9 +155,10 @@ void rt6_uncached_list_del(struct rt6_info *rt) static void rt6_uncached_list_flush_dev(struct net *net, struct net_device *dev) { + struct net_device *loopback_dev = net->loopback_dev; int cpu; - if (dev == net->loopback_dev) + if (dev == loopback_dev) return; for_each_possible_cpu(cpu) { @@ -170,7 +171,7 @@ static void rt6_uncached_list_flush_dev(struct net *net, struct net_device *dev) struct net_device *rt_dev = rt->dst.dev; if (rt_idev->dev == dev) { - rt->rt6i_idev = in6_dev_get(blackhole_netdev); + rt->rt6i_idev = in6_dev_get(loopback_dev); in6_dev_put(rt_idev); } @@ -385,11 +386,13 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, { struct rt6_info *rt = (struct rt6_info *)dst; struct inet6_dev *idev = rt->rt6i_idev; + struct net_device *loopback_dev = + dev_net(dev)->loopback_dev; - if (idev && idev->dev != dev_net(dev)->loopback_dev) { - struct inet6_dev *ibdev = in6_dev_get(blackhole_netdev); - if (ibdev) { - rt->rt6i_idev = ibdev; + if (idev && idev->dev != loopback_dev) { + struct inet6_dev *loopback_idev = in6_dev_get(loopback_dev); + if (loopback_idev) { + rt->rt6i_idev = loopback_idev; in6_dev_put(idev); } } From 2ca4f6ca4562594ef161e4140c2a5e0e5282967b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 14 Oct 2019 06:04:38 -0700 Subject: [PATCH 114/145] rxrpc: use rcu protection while reading sk->sk_user_data We need to extend the rcu_read_lock() section in rxrpc_error_report() and use rcu_dereference_sk_user_data() instead of plain access to sk->sk_user_data to make sure all rules are respected. The compiler wont reload sk->sk_user_data at will, and RCU rules prevent memory beeing freed too soon. Fixes: f0308fb07080 ("rxrpc: Fix possible NULL pointer access in ICMP handling") Fixes: 17926a79320a ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both") Signed-off-by: Eric Dumazet Cc: David Howells Signed-off-by: David S. Miller --- net/rxrpc/peer_event.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 61451281d74a..48f67a9b1037 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -147,13 +147,16 @@ void rxrpc_error_report(struct sock *sk) { struct sock_exterr_skb *serr; struct sockaddr_rxrpc srx; - struct rxrpc_local *local = sk->sk_user_data; + struct rxrpc_local *local; struct rxrpc_peer *peer; struct sk_buff *skb; - if (unlikely(!local)) + rcu_read_lock(); + local = rcu_dereference_sk_user_data(sk); + if (unlikely(!local)) { + rcu_read_unlock(); return; - + } _enter("%p{%d}", sk, local->debug_id); /* Clear the outstanding error value on the socket so that it doesn't @@ -163,6 +166,7 @@ void rxrpc_error_report(struct sock *sk) skb = sock_dequeue_err_skb(sk); if (!skb) { + rcu_read_unlock(); _leave("UDP socket errqueue empty"); return; } @@ -170,11 +174,11 @@ void rxrpc_error_report(struct sock *sk) serr = SKB_EXT_ERR(skb); if (!skb->len && serr->ee.ee_origin == SO_EE_ORIGIN_TIMESTAMPING) { _leave("UDP empty message"); + rcu_read_unlock(); rxrpc_free_skb(skb, rxrpc_skb_freed); return; } - rcu_read_lock(); peer = rxrpc_lookup_peer_icmp_rcu(local, skb, &srx); if (peer && !rxrpc_get_peer_maybe(peer)) peer = NULL; From 3de5ae54712c75cf3c517a288e0a704784ec6cf5 Mon Sep 17 00:00:00 2001 From: Yonglong Liu Date: Wed, 16 Oct 2019 10:30:39 +0800 Subject: [PATCH 115/145] net: phy: Fix "link partner" information disappear issue Some drivers just call phy_ethtool_ksettings_set() to set the links, for those phy drivers that use genphy_read_status(), if autoneg is on, and the link is up, than execute "ethtool -s ethx autoneg on" will cause "link partner" information disappear. The call trace is phy_ethtool_ksettings_set()->phy_start_aneg() ->linkmode_zero(phydev->lp_advertising)->genphy_read_status(), the link didn't change, so genphy_read_status() just return, and phydev->lp_advertising is zero now. This patch moves the clear operation of lp_advertising from phy_start_aneg() to genphy_read_lpa()/genphy_c45_read_lpa(), and if autoneg on and autoneg not complete, just clear what the generic functions care about. Fixes: 88d6272acaaa ("net: phy: avoid unneeded MDIO reads in genphy_read_status") Signed-off-by: Yonglong Liu Reviewed-by: Heiner Kallweit Signed-off-by: David S. Miller --- drivers/net/phy/phy-c45.c | 2 ++ drivers/net/phy/phy.c | 3 --- drivers/net/phy/phy_device.c | 11 ++++++++++- 3 files changed, 12 insertions(+), 4 deletions(-) diff --git a/drivers/net/phy/phy-c45.c b/drivers/net/phy/phy-c45.c index 7935593debb1..a1caeee12236 100644 --- a/drivers/net/phy/phy-c45.c +++ b/drivers/net/phy/phy-c45.c @@ -323,6 +323,8 @@ int genphy_c45_read_pma(struct phy_device *phydev) { int val; + linkmode_zero(phydev->lp_advertising); + val = phy_read_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_CTRL1); if (val < 0) return val; diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 119e6f466056..105d389b58e7 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -572,9 +572,6 @@ int phy_start_aneg(struct phy_device *phydev) if (AUTONEG_DISABLE == phydev->autoneg) phy_sanitize_settings(phydev); - /* Invalidate LP advertising flags */ - linkmode_zero(phydev->lp_advertising); - err = phy_config_aneg(phydev); if (err < 0) goto out_unlock; diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 9d2bbb13293e..adb66a2fae18 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -1787,7 +1787,14 @@ int genphy_read_lpa(struct phy_device *phydev) { int lpa, lpagb; - if (phydev->autoneg == AUTONEG_ENABLE && phydev->autoneg_complete) { + if (phydev->autoneg == AUTONEG_ENABLE) { + if (!phydev->autoneg_complete) { + mii_stat1000_mod_linkmode_lpa_t(phydev->lp_advertising, + 0); + mii_lpa_mod_linkmode_lpa_t(phydev->lp_advertising, 0); + return 0; + } + if (phydev->is_gigabit_capable) { lpagb = phy_read(phydev, MII_STAT1000); if (lpagb < 0) @@ -1815,6 +1822,8 @@ int genphy_read_lpa(struct phy_device *phydev) return lpa; mii_lpa_mod_linkmode_lpa_t(phydev->lp_advertising, lpa); + } else { + linkmode_zero(phydev->lp_advertising); } return 0; From e497c20e203680aba9ccf7bb475959595908ca7e Mon Sep 17 00:00:00 2001 From: Biao Huang Date: Tue, 15 Oct 2019 11:24:44 +0800 Subject: [PATCH 116/145] net: stmmac: disable/enable ptp_ref_clk in suspend/resume flow disable ptp_ref_clk in suspend flow, and enable it in resume flow. Fixes: f573c0b9c4e0 ("stmmac: move stmmac_clk, pclk, clk_ptp_ref and stmmac_rst to platform structure") Signed-off-by: Biao Huang Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index c76a1336a451..8fa13d43b5bc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -4742,8 +4742,10 @@ int stmmac_suspend(struct device *dev) stmmac_mac_set(priv, priv->ioaddr, false); pinctrl_pm_select_sleep_state(priv->device); /* Disable clock in case of PWM is off */ - clk_disable(priv->plat->pclk); - clk_disable(priv->plat->stmmac_clk); + if (priv->plat->clk_ptp_ref) + clk_disable_unprepare(priv->plat->clk_ptp_ref); + clk_disable_unprepare(priv->plat->pclk); + clk_disable_unprepare(priv->plat->stmmac_clk); } mutex_unlock(&priv->lock); @@ -4806,8 +4808,10 @@ int stmmac_resume(struct device *dev) } else { pinctrl_pm_select_default_state(priv->device); /* enable the clk previously disabled */ - clk_enable(priv->plat->stmmac_clk); - clk_enable(priv->plat->pclk); + clk_prepare_enable(priv->plat->stmmac_clk); + clk_prepare_enable(priv->plat->pclk); + if (priv->plat->clk_ptp_ref) + clk_prepare_enable(priv->plat->clk_ptp_ref); /* reset the phy so that it's ready */ if (priv->mii) stmmac_mdio_reset(priv->mii); From af0de1303c4e8f44fadd7b4c593f09f22324b04f Mon Sep 17 00:00:00 2001 From: Oliver Neukum Date: Thu, 17 Oct 2019 15:25:47 +0200 Subject: [PATCH 117/145] usb: hso: obey DMA rules in tiocmget The serial state information must not be embedded into another data structure, as this interferes with cache handling for DMA on architectures without cache coherence.. That would result in data corruption on some architectures Allocating it separately. v2: fix syntax error Signed-off-by: Oliver Neukum Signed-off-by: David S. Miller --- drivers/net/usb/hso.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/drivers/net/usb/hso.c b/drivers/net/usb/hso.c index a505b2ab88b8..74849da031fa 100644 --- a/drivers/net/usb/hso.c +++ b/drivers/net/usb/hso.c @@ -186,7 +186,7 @@ struct hso_tiocmget { int intr_completed; struct usb_endpoint_descriptor *endp; struct urb *urb; - struct hso_serial_state_notification serial_state_notification; + struct hso_serial_state_notification *serial_state_notification; u16 prev_UART_state_bitmap; struct uart_icount icount; }; @@ -1432,7 +1432,7 @@ static int tiocmget_submit_urb(struct hso_serial *serial, usb_rcvintpipe(usb, tiocmget->endp-> bEndpointAddress & 0x7F), - &tiocmget->serial_state_notification, + tiocmget->serial_state_notification, sizeof(struct hso_serial_state_notification), tiocmget_intr_callback, serial, tiocmget->endp->bInterval); @@ -1479,7 +1479,7 @@ static void tiocmget_intr_callback(struct urb *urb) /* wIndex should be the USB interface number of the port to which the * notification applies, which should always be the Modem port. */ - serial_state_notification = &tiocmget->serial_state_notification; + serial_state_notification = tiocmget->serial_state_notification; if (serial_state_notification->bmRequestType != BM_REQUEST_TYPE || serial_state_notification->bNotification != B_NOTIFICATION || le16_to_cpu(serial_state_notification->wValue) != W_VALUE || @@ -2565,6 +2565,8 @@ static void hso_free_tiomget(struct hso_serial *serial) usb_free_urb(tiocmget->urb); tiocmget->urb = NULL; serial->tiocmget = NULL; + kfree(tiocmget->serial_state_notification); + tiocmget->serial_state_notification = NULL; kfree(tiocmget); } } @@ -2615,10 +2617,13 @@ static struct hso_device *hso_create_bulk_serial_device( num_urbs = 2; serial->tiocmget = kzalloc(sizeof(struct hso_tiocmget), GFP_KERNEL); + serial->tiocmget->serial_state_notification + = kzalloc(sizeof(struct hso_serial_state_notification), + GFP_KERNEL); /* it isn't going to break our heart if serial->tiocmget * allocation fails don't bother checking this. */ - if (serial->tiocmget) { + if (serial->tiocmget && serial->tiocmget->serial_state_notification) { tiocmget = serial->tiocmget; tiocmget->endp = hso_get_ep(interface, USB_ENDPOINT_XFER_INT, From 8398b375a9e3f5e4bba9bcdfed152a8a247dee01 Mon Sep 17 00:00:00 2001 From: Florin Chiculita Date: Wed, 16 Oct 2019 10:36:22 +0300 Subject: [PATCH 118/145] dpaa2-eth: add irq for the dpmac connect/disconnect event Add IRQ for the DPNI endpoint change event, resolving the issue when a dynamically created DPNI gets a randomly generated hw address when the endpoint is a DPMAC object. Signed-off-by: Florin Chiculita Signed-off-by: Ioana Ciornei Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c | 6 +++++- drivers/net/ethernet/freescale/dpaa2/dpni.h | 5 ++++- 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c index 162d7d8fb295..5acd734a216b 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c @@ -3306,6 +3306,9 @@ static irqreturn_t dpni_irq0_handler_thread(int irq_num, void *arg) if (status & DPNI_IRQ_EVENT_LINK_CHANGED) link_state_update(netdev_priv(net_dev)); + if (status & DPNI_IRQ_EVENT_ENDPOINT_CHANGED) + set_mac_addr(netdev_priv(net_dev)); + return IRQ_HANDLED; } @@ -3331,7 +3334,8 @@ static int setup_irqs(struct fsl_mc_device *ls_dev) } err = dpni_set_irq_mask(ls_dev->mc_io, 0, ls_dev->mc_handle, - DPNI_IRQ_INDEX, DPNI_IRQ_EVENT_LINK_CHANGED); + DPNI_IRQ_INDEX, DPNI_IRQ_EVENT_LINK_CHANGED | + DPNI_IRQ_EVENT_ENDPOINT_CHANGED); if (err < 0) { dev_err(&ls_dev->dev, "dpni_set_irq_mask(): %d\n", err); goto free_irq; diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni.h b/drivers/net/ethernet/freescale/dpaa2/dpni.h index fd583911b6c0..ee0711d06b3a 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpni.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpni.h @@ -133,9 +133,12 @@ int dpni_reset(struct fsl_mc_io *mc_io, */ #define DPNI_IRQ_INDEX 0 /** - * IRQ event - indicates a change in link state + * IRQ events: + * indicates a change in link state + * indicates a change in endpoint */ #define DPNI_IRQ_EVENT_LINK_CHANGED 0x00000001 +#define DPNI_IRQ_EVENT_ENDPOINT_CHANGED 0x00000002 int dpni_set_irq_enable(struct fsl_mc_io *mc_io, u32 cmd_flags, From a690af4f16f9bd1f976a4257e1529b6323a8d0e1 Mon Sep 17 00:00:00 2001 From: Ioana Radulescu Date: Wed, 16 Oct 2019 10:36:23 +0300 Subject: [PATCH 119/145] dpaa2-eth: Fix TX FQID values Depending on when MC connects the DPNI to a MAC, Tx FQIDs may not be available during probe time. Read the FQIDs each time the link goes up to avoid using invalid values. In case an error occurs or an invalid value is retrieved, fall back to QDID-based enqueueing. Fixes: 1fa0f68c9255 ("dpaa2-eth: Use FQ-based DPIO enqueue API") Signed-off-by: Ioana Radulescu Signed-off-by: Ioana Ciornei Signed-off-by: David S. Miller --- .../net/ethernet/freescale/dpaa2/dpaa2-eth.c | 44 +++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c index 5acd734a216b..19379bae0144 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c @@ -1235,6 +1235,8 @@ static void dpaa2_eth_set_rx_taildrop(struct dpaa2_eth_priv *priv, bool enable) priv->rx_td_enabled = enable; } +static void update_tx_fqids(struct dpaa2_eth_priv *priv); + static int link_state_update(struct dpaa2_eth_priv *priv) { struct dpni_link_state state = {0}; @@ -1261,6 +1263,7 @@ static int link_state_update(struct dpaa2_eth_priv *priv) goto out; if (state.up) { + update_tx_fqids(priv); netif_carrier_on(priv->net_dev); netif_tx_start_all_queues(priv->net_dev); } else { @@ -2533,6 +2536,47 @@ static int set_pause(struct dpaa2_eth_priv *priv) return 0; } +static void update_tx_fqids(struct dpaa2_eth_priv *priv) +{ + struct dpni_queue_id qid = {0}; + struct dpaa2_eth_fq *fq; + struct dpni_queue queue; + int i, j, err; + + /* We only use Tx FQIDs for FQID-based enqueue, so check + * if DPNI version supports it before updating FQIDs + */ + if (dpaa2_eth_cmp_dpni_ver(priv, DPNI_ENQUEUE_FQID_VER_MAJOR, + DPNI_ENQUEUE_FQID_VER_MINOR) < 0) + return; + + for (i = 0; i < priv->num_fqs; i++) { + fq = &priv->fq[i]; + if (fq->type != DPAA2_TX_CONF_FQ) + continue; + for (j = 0; j < dpaa2_eth_tc_count(priv); j++) { + err = dpni_get_queue(priv->mc_io, 0, priv->mc_token, + DPNI_QUEUE_TX, j, fq->flowid, + &queue, &qid); + if (err) + goto out_err; + + fq->tx_fqid[j] = qid.fqid; + if (fq->tx_fqid[j] == 0) + goto out_err; + } + } + + priv->enqueue = dpaa2_eth_enqueue_fq; + + return; + +out_err: + netdev_info(priv->net_dev, + "Error reading Tx FQID, fallback to QDID-based enqueue\n"); + priv->enqueue = dpaa2_eth_enqueue_qd; +} + /* Configure the DPNI object this interface is associated with */ static int setup_dpni(struct fsl_mc_device *ls_dev) { From c9ad4c1049f7e0e8d59e975963dda002af47d93e Mon Sep 17 00:00:00 2001 From: "Ben Dooks (Codethink)" Date: Wed, 16 Oct 2019 09:22:05 +0100 Subject: [PATCH 120/145] net: stmmac: fix argument to stmmac_pcs_ctrl_ane() The stmmac_pcs_ctrl_ane() expects a register address as argument 1, but for some reason the mac_device_info is being passed. Fix the warning (and possible bug) from sparse: drivers/net/ethernet/stmicro/stmmac/stmmac_main.c:2613:17: warning: incorrect type in argument 1 (different address spaces) drivers/net/ethernet/stmicro/stmmac/stmmac_main.c:2613:17: expected void [noderef] *ioaddr drivers/net/ethernet/stmicro/stmmac/stmmac_main.c:2613:17: got struct mac_device_info *hw Signed-off-by: Ben Dooks Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 8fa13d43b5bc..3dfd04e0506a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -2610,7 +2610,7 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) } if (priv->hw->pcs) - stmmac_pcs_ctrl_ane(priv, priv->hw, 1, priv->hw->ps, 0); + stmmac_pcs_ctrl_ane(priv, priv->ioaddr, 1, priv->hw->ps, 0); /* set TX and RX rings length */ stmmac_set_rings_length(priv); From 7f238ca984b9c6a89e2339c661440c775d617386 Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Wed, 16 Oct 2019 15:33:23 +0200 Subject: [PATCH 121/145] net: dsa: microchip: Do not reinit mutexes on KSZ87xx The KSZ87xx driver calls mutex_init() on mutexes already inited in ksz_common.c ksz_switch_register(). Do not do it twice, drop the reinitialization. Signed-off-by: Marek Vasut Cc: Andrew Lunn Cc: David S. Miller Cc: Florian Fainelli Cc: George McCollister Cc: Tristram Ha Cc: Woojung Huh Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/dsa/microchip/ksz8795.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/net/dsa/microchip/ksz8795.c b/drivers/net/dsa/microchip/ksz8795.c index a23d3ffdf0c4..24a5e99f7fd5 100644 --- a/drivers/net/dsa/microchip/ksz8795.c +++ b/drivers/net/dsa/microchip/ksz8795.c @@ -1224,10 +1224,6 @@ static int ksz8795_switch_init(struct ksz_device *dev) { int i; - mutex_init(&dev->stats_mutex); - mutex_init(&dev->alu_mutex); - mutex_init(&dev->vlan_mutex); - dev->ds->ops = &ksz8795_switch_ops; for (i = 0; i < ARRAY_SIZE(ksz8795_switch_chips); i++) { From 013572a236ef53cbd1e315e0acf2ee632cc816d4 Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Wed, 16 Oct 2019 15:33:24 +0200 Subject: [PATCH 122/145] net: dsa: microchip: Add shared regmap mutex The KSZ driver uses one regmap per register width (8/16/32), each with it's own lock, but accessing the same set of registers. In theory, it is possible to create a race condition between these regmaps, although the underlying bus (SPI or I2C) locking should assure nothing bad will really happen and the accesses would be correct. To make the driver do the right thing, add one single shared mutex for all the regmaps used by the driver instead. This assures that even if some future hardware is on a bus which does not serialize the accesses the same way SPI or I2C does, nothing bad will happen. Note that the status_mutex was unused and only initied, hence it was renamed and repurposed as the regmap mutex. Signed-off-by: Marek Vasut Cc: Andrew Lunn Cc: David S. Miller Cc: Florian Fainelli Cc: George McCollister Cc: Tristram Ha Cc: Woojung Huh Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/dsa/microchip/ksz8795_spi.c | 7 ++++--- drivers/net/dsa/microchip/ksz9477_i2c.c | 6 ++++-- drivers/net/dsa/microchip/ksz9477_spi.c | 6 ++++-- drivers/net/dsa/microchip/ksz_common.c | 2 +- drivers/net/dsa/microchip/ksz_common.h | 16 +++++++++++++++- 5 files changed, 28 insertions(+), 9 deletions(-) diff --git a/drivers/net/dsa/microchip/ksz8795_spi.c b/drivers/net/dsa/microchip/ksz8795_spi.c index d0f8153e86b7..8b00f8e6c02f 100644 --- a/drivers/net/dsa/microchip/ksz8795_spi.c +++ b/drivers/net/dsa/microchip/ksz8795_spi.c @@ -25,6 +25,7 @@ KSZ_REGMAP_TABLE(ksz8795, 16, SPI_ADDR_SHIFT, static int ksz8795_spi_probe(struct spi_device *spi) { + struct regmap_config rc; struct ksz_device *dev; int i, ret; @@ -33,9 +34,9 @@ static int ksz8795_spi_probe(struct spi_device *spi) return -ENOMEM; for (i = 0; i < ARRAY_SIZE(ksz8795_regmap_config); i++) { - dev->regmap[i] = devm_regmap_init_spi(spi, - &ksz8795_regmap_config - [i]); + rc = ksz8795_regmap_config[i]; + rc.lock_arg = &dev->regmap_mutex; + dev->regmap[i] = devm_regmap_init_spi(spi, &rc); if (IS_ERR(dev->regmap[i])) { ret = PTR_ERR(dev->regmap[i]); dev_err(&spi->dev, diff --git a/drivers/net/dsa/microchip/ksz9477_i2c.c b/drivers/net/dsa/microchip/ksz9477_i2c.c index 0b1e01f0873d..fdffd9e0c518 100644 --- a/drivers/net/dsa/microchip/ksz9477_i2c.c +++ b/drivers/net/dsa/microchip/ksz9477_i2c.c @@ -17,6 +17,7 @@ KSZ_REGMAP_TABLE(ksz9477, not_used, 16, 0, 0); static int ksz9477_i2c_probe(struct i2c_client *i2c, const struct i2c_device_id *i2c_id) { + struct regmap_config rc; struct ksz_device *dev; int i, ret; @@ -25,8 +26,9 @@ static int ksz9477_i2c_probe(struct i2c_client *i2c, return -ENOMEM; for (i = 0; i < ARRAY_SIZE(ksz9477_regmap_config); i++) { - dev->regmap[i] = devm_regmap_init_i2c(i2c, - &ksz9477_regmap_config[i]); + rc = ksz9477_regmap_config[i]; + rc.lock_arg = &dev->regmap_mutex; + dev->regmap[i] = devm_regmap_init_i2c(i2c, &rc); if (IS_ERR(dev->regmap[i])) { ret = PTR_ERR(dev->regmap[i]); dev_err(&i2c->dev, diff --git a/drivers/net/dsa/microchip/ksz9477_spi.c b/drivers/net/dsa/microchip/ksz9477_spi.c index f4198d6f72be..c5f64959a184 100644 --- a/drivers/net/dsa/microchip/ksz9477_spi.c +++ b/drivers/net/dsa/microchip/ksz9477_spi.c @@ -24,6 +24,7 @@ KSZ_REGMAP_TABLE(ksz9477, 32, SPI_ADDR_SHIFT, static int ksz9477_spi_probe(struct spi_device *spi) { + struct regmap_config rc; struct ksz_device *dev; int i, ret; @@ -32,8 +33,9 @@ static int ksz9477_spi_probe(struct spi_device *spi) return -ENOMEM; for (i = 0; i < ARRAY_SIZE(ksz9477_regmap_config); i++) { - dev->regmap[i] = devm_regmap_init_spi(spi, - &ksz9477_regmap_config[i]); + rc = ksz9477_regmap_config[i]; + rc.lock_arg = &dev->regmap_mutex; + dev->regmap[i] = devm_regmap_init_spi(spi, &rc); if (IS_ERR(dev->regmap[i])) { ret = PTR_ERR(dev->regmap[i]); dev_err(&spi->dev, diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c index b0b870f0c252..fe47180c908b 100644 --- a/drivers/net/dsa/microchip/ksz_common.c +++ b/drivers/net/dsa/microchip/ksz_common.c @@ -436,7 +436,7 @@ int ksz_switch_register(struct ksz_device *dev, } mutex_init(&dev->dev_mutex); - mutex_init(&dev->stats_mutex); + mutex_init(&dev->regmap_mutex); mutex_init(&dev->alu_mutex); mutex_init(&dev->vlan_mutex); diff --git a/drivers/net/dsa/microchip/ksz_common.h b/drivers/net/dsa/microchip/ksz_common.h index 64c2677693d2..a20ebb749377 100644 --- a/drivers/net/dsa/microchip/ksz_common.h +++ b/drivers/net/dsa/microchip/ksz_common.h @@ -47,7 +47,7 @@ struct ksz_device { const char *name; struct mutex dev_mutex; /* device access */ - struct mutex stats_mutex; /* status access */ + struct mutex regmap_mutex; /* regmap access */ struct mutex alu_mutex; /* ALU access */ struct mutex vlan_mutex; /* vlan access */ const struct ksz_dev_ops *dev_ops; @@ -290,6 +290,18 @@ static inline void ksz_pwrite32(struct ksz_device *dev, int port, int offset, ksz_write32(dev, dev->dev_ops->get_port_addr(port, offset), data); } +static inline void ksz_regmap_lock(void *__mtx) +{ + struct mutex *mtx = __mtx; + mutex_lock(mtx); +} + +static inline void ksz_regmap_unlock(void *__mtx) +{ + struct mutex *mtx = __mtx; + mutex_unlock(mtx); +} + /* Regmap tables generation */ #define KSZ_SPI_OP_RD 3 #define KSZ_SPI_OP_WR 2 @@ -314,6 +326,8 @@ static inline void ksz_pwrite32(struct ksz_device *dev, int port, int offset, .write_flag_mask = \ KSZ_SPI_OP_FLAG_MASK(KSZ_SPI_OP_WR, swp, \ regbits, regpad), \ + .lock = ksz_regmap_lock, \ + .unlock = ksz_regmap_unlock, \ .reg_format_endian = REGMAP_ENDIAN_BIG, \ .val_format_endian = REGMAP_ENDIAN_BIG \ } From 8b95599c55ed24b36cf44a4720067cfe67edbcb4 Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Wed, 16 Oct 2019 15:35:06 +0200 Subject: [PATCH 123/145] net: phy: micrel: Discern KSZ8051 and KSZ8795 PHYs The KSZ8051 PHY and the KSZ8794/KSZ8795/KSZ8765 switch share exactly the same PHY ID. Since KSZ8051 is higher in the ksphy_driver[] list of PHYs in the micrel PHY driver, it is used even with the KSZ87xx switch. This is wrong, since the KSZ8051 configures registers of the PHY which are not present on the simplified KSZ87xx switch PHYs and misconfigures other registers of the KSZ87xx switch PHYs. Fortunatelly, it is possible to tell apart the KSZ8051 PHY from the KSZ87xx switch by checking the Basic Status register Bit 0, which is read-only and indicates presence of the Extended Capability Registers. The KSZ8051 PHY has those registers while the KSZ87xx switch does not. This patch implements simple check for the presence of this bit for both the KSZ8051 PHY and KSZ87xx switch, to let both use the correct PHY driver instance. Fixes: 9d162ed69f51 ("net: phy: micrel: add support for KSZ8795") Signed-off-by: Marek Vasut Cc: Andrew Lunn Cc: David S. Miller Cc: Florian Fainelli Cc: George McCollister Cc: Heiner Kallweit Cc: Sean Nyekjaer Cc: Tristram Ha Cc: Woojung Huh Signed-off-by: David S. Miller --- drivers/net/phy/micrel.c | 40 ++++++++++++++++++++++++++++++++++++---- 1 file changed, 36 insertions(+), 4 deletions(-) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 2fea5541c35a..a0444e28c6e7 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -341,6 +341,35 @@ static int ksz8041_config_aneg(struct phy_device *phydev) return genphy_config_aneg(phydev); } +static int ksz8051_ksz8795_match_phy_device(struct phy_device *phydev, + const u32 ksz_phy_id) +{ + int ret; + + if ((phydev->phy_id & MICREL_PHY_ID_MASK) != ksz_phy_id) + return 0; + + ret = phy_read(phydev, MII_BMSR); + if (ret < 0) + return ret; + + /* KSZ8051 PHY and KSZ8794/KSZ8795/KSZ8765 switch share the same + * exact PHY ID. However, they can be told apart by the extended + * capability registers presence. The KSZ8051 PHY has them while + * the switch does not. + */ + ret &= BMSR_ERCAP; + if (ksz_phy_id == PHY_ID_KSZ8051) + return ret; + else + return !ret; +} + +static int ksz8051_match_phy_device(struct phy_device *phydev) +{ + return ksz8051_ksz8795_match_phy_device(phydev, PHY_ID_KSZ8051); +} + static int ksz8081_config_init(struct phy_device *phydev) { /* KSZPHY_OMSO_FACTORY_TEST is set at de-assertion of the reset line @@ -364,6 +393,11 @@ static int ksz8061_config_init(struct phy_device *phydev) return kszphy_config_init(phydev); } +static int ksz8795_match_phy_device(struct phy_device *phydev) +{ + return ksz8051_ksz8795_match_phy_device(phydev, PHY_ID_KSZ8795); +} + static int ksz9021_load_values_from_of(struct phy_device *phydev, const struct device_node *of_node, u16 reg, @@ -1017,8 +1051,6 @@ static struct phy_driver ksphy_driver[] = { .suspend = genphy_suspend, .resume = genphy_resume, }, { - .phy_id = PHY_ID_KSZ8051, - .phy_id_mask = MICREL_PHY_ID_MASK, .name = "Micrel KSZ8051", /* PHY_BASIC_FEATURES */ .driver_data = &ksz8051_type, @@ -1029,6 +1061,7 @@ static struct phy_driver ksphy_driver[] = { .get_sset_count = kszphy_get_sset_count, .get_strings = kszphy_get_strings, .get_stats = kszphy_get_stats, + .match_phy_device = ksz8051_match_phy_device, .suspend = genphy_suspend, .resume = genphy_resume, }, { @@ -1141,13 +1174,12 @@ static struct phy_driver ksphy_driver[] = { .suspend = genphy_suspend, .resume = genphy_resume, }, { - .phy_id = PHY_ID_KSZ8795, - .phy_id_mask = MICREL_PHY_ID_MASK, .name = "Micrel KSZ8795", /* PHY_BASIC_FEATURES */ .config_init = kszphy_config_init, .config_aneg = ksz8873mll_config_aneg, .read_status = ksz8873mll_read_status, + .match_phy_device = ksz8795_match_phy_device, .suspend = genphy_suspend, .resume = genphy_resume, }, { From 1d951ba3da67bbc7a9b0e05987e09552c2060e18 Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Wed, 16 Oct 2019 15:35:07 +0200 Subject: [PATCH 124/145] net: phy: micrel: Update KSZ87xx PHY name The KSZ8795 PHY ID is in fact used by KSZ8794/KSZ8795/KSZ8765 switches. Update the PHY ID and name to reflect that, as this family of switches is commonly refered to as KSZ87xx Signed-off-by: Marek Vasut Cc: Andrew Lunn Cc: David S. Miller Cc: Florian Fainelli Cc: George McCollister Cc: Heiner Kallweit Cc: Sean Nyekjaer Cc: Tristram Ha Cc: Woojung Huh Signed-off-by: David S. Miller --- drivers/net/phy/micrel.c | 4 ++-- include/linux/micrel_phy.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index a0444e28c6e7..63dedec0433d 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -395,7 +395,7 @@ static int ksz8061_config_init(struct phy_device *phydev) static int ksz8795_match_phy_device(struct phy_device *phydev) { - return ksz8051_ksz8795_match_phy_device(phydev, PHY_ID_KSZ8795); + return ksz8051_ksz8795_match_phy_device(phydev, PHY_ID_KSZ87XX); } static int ksz9021_load_values_from_of(struct phy_device *phydev, @@ -1174,7 +1174,7 @@ static struct phy_driver ksphy_driver[] = { .suspend = genphy_suspend, .resume = genphy_resume, }, { - .name = "Micrel KSZ8795", + .name = "Micrel KSZ87XX Switch", /* PHY_BASIC_FEATURES */ .config_init = kszphy_config_init, .config_aneg = ksz8873mll_config_aneg, diff --git a/include/linux/micrel_phy.h b/include/linux/micrel_phy.h index ad24554f11f9..75f880c25bb8 100644 --- a/include/linux/micrel_phy.h +++ b/include/linux/micrel_phy.h @@ -31,7 +31,7 @@ #define PHY_ID_KSZ886X 0x00221430 #define PHY_ID_KSZ8863 0x00221435 -#define PHY_ID_KSZ8795 0x00221550 +#define PHY_ID_KSZ87XX 0x00221550 #define PHY_ID_KSZ9477 0x00221631 From 595e0651d0296bad2491a4a29a7a43eae6328b02 Mon Sep 17 00:00:00 2001 From: Stefano Brivio Date: Wed, 16 Oct 2019 20:52:09 +0200 Subject: [PATCH 125/145] ipv4: Return -ENETUNREACH if we can't create route but saddr is valid ...instead of -EINVAL. An issue was found with older kernel versions while unplugging a NFS client with pending RPCs, and the wrong error code here prevented it from recovering once link is back up with a configured address. Incidentally, this is not an issue anymore since commit 4f8943f80883 ("SUNRPC: Replace direct task wakeups from softirq context"), included in 5.2-rc7, had the effect of decoupling the forwarding of this error by using SO_ERROR in xs_wake_error(), as pointed out by Benjamin Coddington. To the best of my knowledge, this isn't currently causing any further issue, but the error code doesn't look appropriate anyway, and we might hit this in other paths as well. In detail, as analysed by Gonzalo Siero, once the route is deleted because the interface is down, and can't be resolved and we return -EINVAL here, this ends up, courtesy of inet_sk_rebuild_header(), as the socket error seen by tcp_write_err(), called by tcp_retransmit_timer(). In turn, tcp_write_err() indirectly calls xs_error_report(), which wakes up the RPC pending tasks with a status of -EINVAL. This is then seen by call_status() in the SUN RPC implementation, which aborts the RPC call calling rpc_exit(), instead of handling this as a potentially temporary condition, i.e. as a timeout. Return -EINVAL only if the input parameters passed to ip_route_output_key_hash_rcu() are actually invalid (this is the case if the specified source address is multicast, limited broadcast or all zeroes), but return -ENETUNREACH in all cases where, at the given moment, the given source address doesn't allow resolving the route. While at it, drop the initialisation of err to -ENETUNREACH, which was added to __ip_route_output_key() back then by commit 0315e3827048 ("net: Fix behaviour of unreachable, blackhole and prohibit routes"), but actually had no effect, as it was, and is, overwritten by the fib_lookup() return code assignment, and anyway ignored in all other branches, including the if (fl4->saddr) one: I find this rather confusing, as it would look like -ENETUNREACH is the "default" error, while that statement has no effect. Also note that after commit fc75fc8339e7 ("ipv4: dont create routes on down devices"), we would get -ENETUNREACH if the device is down, but -EINVAL if the source address is specified and we can't resolve the route, and this appears to be rather inconsistent. Reported-by: Stefan Walter Analysed-by: Benjamin Coddington Analysed-by: Gonzalo Siero Signed-off-by: Stefano Brivio Signed-off-by: David S. Miller --- net/ipv4/route.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 14654876127e..5bc172abd143 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2470,14 +2470,17 @@ struct rtable *ip_route_output_key_hash_rcu(struct net *net, struct flowi4 *fl4, int orig_oif = fl4->flowi4_oif; unsigned int flags = 0; struct rtable *rth; - int err = -ENETUNREACH; + int err; if (fl4->saddr) { - rth = ERR_PTR(-EINVAL); if (ipv4_is_multicast(fl4->saddr) || ipv4_is_lbcast(fl4->saddr) || - ipv4_is_zeronet(fl4->saddr)) + ipv4_is_zeronet(fl4->saddr)) { + rth = ERR_PTR(-EINVAL); goto out; + } + + rth = ERR_PTR(-ENETUNREACH); /* I removed check for oif == dev_out->oif here. It was wrong for two reasons: From 5018c59607a511cdee743b629c76206d9c9e6d7b Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Wed, 16 Oct 2019 12:03:15 -0700 Subject: [PATCH 126/145] ipv4: fix race condition between route lookup and invalidation Jesse and Ido reported the following race condition: - Received packet A is forwarded and cached dst entry is taken from the nexthop ('nhc->nhc_rth_input'). Calls skb_dst_set() - Given Jesse has busy routers ("ingesting full BGP routing tables from multiple ISPs"), route is added / deleted and rt_cache_flush() is called - Received packet B tries to use the same cached dst entry from t0, but rt_cache_valid() is no longer true and it is replaced in rt_cache_route() by the newer one. This calls dst_dev_put() on the original dst entry which assigns the blackhole netdev to 'dst->dev' - dst_input(skb) is called on packet A and it is dropped due to 'dst->dev' being the blackhole netdev There are 2 issues in the v4 routing code: 1. A per-netns counter is used to do the validation of the route. That means whenever a route is changed in the netns, users of all routes in the netns needs to redo lookup. v6 has an implementation of only updating fn_sernum for routes that are affected. 2. When rt_cache_valid() returns false, rt_cache_route() is called to throw away the current cache, and create a new one. This seems unnecessary because as long as this route does not change, the route cache does not need to be recreated. To fully solve the above 2 issues, it probably needs quite some code changes and requires careful testing, and does not suite for net branch. So this patch only tries to add the deleted cached rt into the uncached list, so user could still be able to use it to receive packets until it's done. Fixes: 95c47f9cf5e0 ("ipv4: call dst_dev_put() properly") Signed-off-by: Wei Wang Reported-by: Ido Schimmel Reported-by: Jesse Hathaway Tested-by: Jesse Hathaway Acked-by: Martin KaFai Lau Cc: David Ahern Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- net/ipv4/route.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 5bc172abd143..621f83434b24 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1482,7 +1482,7 @@ static bool rt_cache_route(struct fib_nh_common *nhc, struct rtable *rt) prev = cmpxchg(p, orig, rt); if (prev == orig) { if (orig) { - dst_dev_put(&orig->dst); + rt_add_uncached_list(orig); dst_release(&orig->dst); } } else { From 0cc76d2b14ffcd00a99d6ec3b9ebad39c659701a Mon Sep 17 00:00:00 2001 From: Sean Wang Date: Thu, 17 Oct 2019 05:14:08 +0800 Subject: [PATCH 127/145] net: Update address for MediaTek ethernet driver in MAINTAINERS Update maintainers for MediaTek ethernet driver with Mark Lee. He is familiar with MediaTek mt762x series ethernet devices and will keep following maintenance from the vendor side. Signed-off-by: Sean Wang Signed-off-by: Mark Lee Signed-off-by: David S. Miller --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 49e4b004917b..60c8b337eb88 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10258,7 +10258,7 @@ MEDIATEK ETHERNET DRIVER M: Felix Fietkau M: John Crispin M: Sean Wang -M: Nelson Chang +M: Mark Lee L: netdev@vger.kernel.org S: Maintained F: drivers/net/ethernet/mediatek/ From 7de48402faa32298c3551ea32c76ccb4f9d3025d Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Wed, 16 Oct 2019 16:06:29 -0700 Subject: [PATCH 128/145] net: bcmgenet: don't set phydev->link from MAC When commit 28b2e0d2cd13 ("net: phy: remove parameter new_link from phy_mac_interrupt()") removed the new_link parameter it set the phydev->link state from the MAC before invoking phy_mac_interrupt(). However, once commit 88d6272acaaa ("net: phy: avoid unneeded MDIO reads in genphy_read_status") was added this initialization prevents the proper determination of the connection parameters by the function genphy_read_status(). This commit removes that initialization to restore the proper functionality. Fixes: 88d6272acaaa ("net: phy: avoid unneeded MDIO reads in genphy_read_status") Signed-off-by: Doug Berger Acked-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 12cb77ef1081..10d68017ff6c 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -2612,10 +2612,8 @@ static void bcmgenet_irq_task(struct work_struct *work) spin_unlock_irq(&priv->lock); /* Link UP/DOWN event */ - if (status & UMAC_IRQ_LINK_EVENT) { - priv->dev->phydev->link = !!(status & UMAC_IRQ_LINK_UP); + if (status & UMAC_IRQ_LINK_EVENT) phy_mac_interrupt(priv->dev->phydev); - } } /* bcmgenet_isr1: handle Rx and Tx priority queues */ From fe586b823372a9f43f90e2c6aa0573992ce7ccb7 Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Wed, 16 Oct 2019 16:06:30 -0700 Subject: [PATCH 129/145] net: phy: bcm7xxx: define soft_reset for 40nm EPHY The internal 40nm EPHYs use a "Workaround for putting the PHY in IDDQ mode." These PHYs require a soft reset to restore functionality after they are powered back up. This commit defines the soft_reset function to use genphy_soft_reset during phy_init_hw to accommodate this. Fixes: 6e2d85ec0559 ("net: phy: Stop with excessive soft reset") Signed-off-by: Doug Berger Acked-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/bcm7xxx.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c index 8fc33867e524..af8eabe7a6d4 100644 --- a/drivers/net/phy/bcm7xxx.c +++ b/drivers/net/phy/bcm7xxx.c @@ -572,6 +572,7 @@ static int bcm7xxx_28nm_probe(struct phy_device *phydev) .name = _name, \ /* PHY_BASIC_FEATURES */ \ .flags = PHY_IS_INTERNAL, \ + .soft_reset = genphy_soft_reset, \ .config_init = bcm7xxx_config_init, \ .suspend = bcm7xxx_suspend, \ .resume = bcm7xxx_config_init, \ From 1f515486275a08a17a2c806b844cca18f7de5b34 Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Wed, 16 Oct 2019 16:06:31 -0700 Subject: [PATCH 130/145] net: bcmgenet: soft reset 40nm EPHYs before MAC init It turns out that the "Workaround for putting the PHY in IDDQ mode" used by the internal EPHYs on 40nm Set-Top Box chips when powering down puts the interface to the GENET MAC in a state that can cause subsequent MAC resets to be incomplete. Rather than restore the forced soft reset when powering up internal PHYs, this commit moves the invocation of phy_init_hw earlier in the MAC initialization sequence to just before the MAC reset in the open and resume functions. This allows the interface to be stable and allows the MAC resets to be successful. The bcmgenet_mii_probe() function is split in two to accommodate this. The new function bcmgenet_mii_connect() handles the first half of the functionality before the MAC initialization, and the bcmgenet_mii_config() function is extended to provide the remaining PHY configuration following the MAC initialization. Fixes: 484bfa1507bf ("Revert "net: bcmgenet: Software reset EPHY after power on"") Signed-off-by: Doug Berger Acked-by: Florian Fainelli Signed-off-by: David S. Miller --- .../net/ethernet/broadcom/genet/bcmgenet.c | 28 +++-- .../net/ethernet/broadcom/genet/bcmgenet.h | 2 +- drivers/net/ethernet/broadcom/genet/bcmmii.c | 114 ++++++++---------- 3 files changed, 70 insertions(+), 74 deletions(-) diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 10d68017ff6c..f0937c650e3c 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -2872,6 +2872,12 @@ static int bcmgenet_open(struct net_device *dev) if (priv->internal_phy) bcmgenet_power_up(priv, GENET_POWER_PASSIVE); + ret = bcmgenet_mii_connect(dev); + if (ret) { + netdev_err(dev, "failed to connect to PHY\n"); + goto err_clk_disable; + } + /* take MAC out of reset */ bcmgenet_umac_reset(priv); @@ -2881,6 +2887,12 @@ static int bcmgenet_open(struct net_device *dev) reg = bcmgenet_umac_readl(priv, UMAC_CMD); priv->crc_fwd_en = !!(reg & CMD_CRC_FWD); + ret = bcmgenet_mii_config(dev, true); + if (ret) { + netdev_err(dev, "unsupported PHY\n"); + goto err_disconnect_phy; + } + bcmgenet_set_hw_addr(priv, dev->dev_addr); if (priv->internal_phy) { @@ -2896,7 +2908,7 @@ static int bcmgenet_open(struct net_device *dev) ret = bcmgenet_init_dma(priv); if (ret) { netdev_err(dev, "failed to initialize DMA\n"); - goto err_clk_disable; + goto err_disconnect_phy; } /* Always enable ring 16 - descriptor ring */ @@ -2919,25 +2931,19 @@ static int bcmgenet_open(struct net_device *dev) goto err_irq0; } - ret = bcmgenet_mii_probe(dev); - if (ret) { - netdev_err(dev, "failed to connect to PHY\n"); - goto err_irq1; - } - bcmgenet_netif_start(dev); netif_tx_start_all_queues(dev); return 0; -err_irq1: - free_irq(priv->irq1, priv); err_irq0: free_irq(priv->irq0, priv); err_fini_dma: bcmgenet_dma_teardown(priv); bcmgenet_fini_dma(priv); +err_disconnect_phy: + phy_disconnect(dev->phydev); err_clk_disable: if (priv->internal_phy) bcmgenet_power_down(priv, GENET_POWER_PASSIVE); @@ -3618,6 +3624,8 @@ static int bcmgenet_resume(struct device *d) if (priv->internal_phy) bcmgenet_power_up(priv, GENET_POWER_PASSIVE); + phy_init_hw(dev->phydev); + bcmgenet_umac_reset(priv); init_umac(priv); @@ -3626,8 +3634,6 @@ static int bcmgenet_resume(struct device *d) if (priv->wolopts) clk_disable_unprepare(priv->clk_wol); - phy_init_hw(dev->phydev); - /* Speed settings must be restored */ bcmgenet_mii_config(priv->dev, false); diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h index dbc69d8fa05f..7fbf573d8d52 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h @@ -720,8 +720,8 @@ GENET_IO_MACRO(rbuf, GENET_RBUF_OFF); /* MDIO routines */ int bcmgenet_mii_init(struct net_device *dev); +int bcmgenet_mii_connect(struct net_device *dev); int bcmgenet_mii_config(struct net_device *dev, bool init); -int bcmgenet_mii_probe(struct net_device *dev); void bcmgenet_mii_exit(struct net_device *dev); void bcmgenet_phy_power_set(struct net_device *dev, bool enable); void bcmgenet_mii_setup(struct net_device *dev); diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index e7c291bf4ed1..17bb8d60a157 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -173,6 +173,46 @@ static void bcmgenet_moca_phy_setup(struct bcmgenet_priv *priv) bcmgenet_fixed_phy_link_update); } +int bcmgenet_mii_connect(struct net_device *dev) +{ + struct bcmgenet_priv *priv = netdev_priv(dev); + struct device_node *dn = priv->pdev->dev.of_node; + struct phy_device *phydev; + u32 phy_flags = 0; + int ret; + + /* Communicate the integrated PHY revision */ + if (priv->internal_phy) + phy_flags = priv->gphy_rev; + + /* Initialize link state variables that bcmgenet_mii_setup() uses */ + priv->old_link = -1; + priv->old_speed = -1; + priv->old_duplex = -1; + priv->old_pause = -1; + + if (dn) { + phydev = of_phy_connect(dev, priv->phy_dn, bcmgenet_mii_setup, + phy_flags, priv->phy_interface); + if (!phydev) { + pr_err("could not attach to PHY\n"); + return -ENODEV; + } + } else { + phydev = dev->phydev; + phydev->dev_flags = phy_flags; + + ret = phy_connect_direct(dev, phydev, bcmgenet_mii_setup, + priv->phy_interface); + if (ret) { + pr_err("could not attach to PHY\n"); + return -ENODEV; + } + } + + return 0; +} + int bcmgenet_mii_config(struct net_device *dev, bool init) { struct bcmgenet_priv *priv = netdev_priv(dev); @@ -266,71 +306,21 @@ int bcmgenet_mii_config(struct net_device *dev, bool init) bcmgenet_ext_writel(priv, reg, EXT_RGMII_OOB_CTRL); } - if (init) + if (init) { + linkmode_copy(phydev->advertising, phydev->supported); + + /* The internal PHY has its link interrupts routed to the + * Ethernet MAC ISRs. On GENETv5 there is a hardware issue + * that prevents the signaling of link UP interrupts when + * the link operates at 10Mbps, so fallback to polling for + * those versions of GENET. + */ + if (priv->internal_phy && !GENET_IS_V5(priv)) + phydev->irq = PHY_IGNORE_INTERRUPT; + dev_info(kdev, "configuring instance for %s\n", phy_name); - - return 0; -} - -int bcmgenet_mii_probe(struct net_device *dev) -{ - struct bcmgenet_priv *priv = netdev_priv(dev); - struct device_node *dn = priv->pdev->dev.of_node; - struct phy_device *phydev; - u32 phy_flags = 0; - int ret; - - /* Communicate the integrated PHY revision */ - if (priv->internal_phy) - phy_flags = priv->gphy_rev; - - /* Initialize link state variables that bcmgenet_mii_setup() uses */ - priv->old_link = -1; - priv->old_speed = -1; - priv->old_duplex = -1; - priv->old_pause = -1; - - if (dn) { - phydev = of_phy_connect(dev, priv->phy_dn, bcmgenet_mii_setup, - phy_flags, priv->phy_interface); - if (!phydev) { - pr_err("could not attach to PHY\n"); - return -ENODEV; - } - } else { - phydev = dev->phydev; - phydev->dev_flags = phy_flags; - - ret = phy_connect_direct(dev, phydev, bcmgenet_mii_setup, - priv->phy_interface); - if (ret) { - pr_err("could not attach to PHY\n"); - return -ENODEV; - } } - /* Configure port multiplexer based on what the probed PHY device since - * reading the 'max-speed' property determines the maximum supported - * PHY speed which is needed for bcmgenet_mii_config() to configure - * things appropriately. - */ - ret = bcmgenet_mii_config(dev, true); - if (ret) { - phy_disconnect(dev->phydev); - return ret; - } - - linkmode_copy(phydev->advertising, phydev->supported); - - /* The internal PHY has its link interrupts routed to the - * Ethernet MAC ISRs. On GENETv5 there is a hardware issue - * that prevents the signaling of link UP interrupts when - * the link operates at 10Mbps, so fallback to polling for - * those versions of GENET. - */ - if (priv->internal_phy && !GENET_IS_V5(priv)) - dev->phydev->irq = PHY_IGNORE_INTERRUPT; - return 0; } From 25382b991d252aed961cd434176240f9de6bb15f Mon Sep 17 00:00:00 2001 From: Doug Berger Date: Wed, 16 Oct 2019 16:06:32 -0700 Subject: [PATCH 131/145] net: bcmgenet: reset 40nm EPHY on energy detect The EPHY integrated into the 40nm Set-Top Box devices can falsely detect energy when connected to a disabled peer interface. When the peer interface is enabled the EPHY will detect and report the link as active, but on occasion may get into a state where it is not able to exchange data with the connected GENET MAC. This issue has not been observed when the link parameters are auto-negotiated; however, it has been observed with a manually configured link. It has been empirically determined that issuing a soft reset to the EPHY when energy is detected prevents it from getting into this bad state. Fixes: 1c1008c793fa ("net: bcmgenet: add main driver file") Signed-off-by: Doug Berger Acked-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index f0937c650e3c..0f138280315a 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -2018,6 +2018,8 @@ static void bcmgenet_link_intr_enable(struct bcmgenet_priv *priv) */ if (priv->internal_phy) { int0_enable |= UMAC_IRQ_LINK_EVENT; + if (GENET_IS_V1(priv) || GENET_IS_V2(priv) || GENET_IS_V3(priv)) + int0_enable |= UMAC_IRQ_PHY_DET_R; } else if (priv->ext_phy) { int0_enable |= UMAC_IRQ_LINK_EVENT; } else if (priv->phy_interface == PHY_INTERFACE_MODE_MOCA) { @@ -2611,9 +2613,14 @@ static void bcmgenet_irq_task(struct work_struct *work) priv->irq0_stat = 0; spin_unlock_irq(&priv->lock); + if (status & UMAC_IRQ_PHY_DET_R && + priv->dev->phydev->autoneg != AUTONEG_ENABLE) + phy_init_hw(priv->dev->phydev); + /* Link UP/DOWN event */ if (status & UMAC_IRQ_LINK_EVENT) phy_mac_interrupt(priv->dev->phydev); + } /* bcmgenet_isr1: handle Rx and Tx priority queues */ @@ -2708,7 +2715,7 @@ static irqreturn_t bcmgenet_isr0(int irq, void *dev_id) } /* all other interested interrupts handled in bottom half */ - status &= UMAC_IRQ_LINK_EVENT; + status &= (UMAC_IRQ_LINK_EVENT | UMAC_IRQ_PHY_DET_R); if (status) { /* Save irq status for bottom-half processing. */ spin_lock_irqsave(&priv->lock, flags); From 9669fffc1415bb0c30e5d2ec98a8e1c3a418cb9c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 16 Oct 2019 18:00:56 -0700 Subject: [PATCH 132/145] net: ensure correct skb->tstamp in various fragmenters Thomas found that some forwarded packets would be stuck in FQ packet scheduler because their skb->tstamp contained timestamps far in the future. We thought we addressed this point in commit 8203e2d844d3 ("net: clear skb->tstamp in forwarding paths") but there is still an issue when/if a packet needs to be fragmented. In order to meet EDT requirements, we have to make sure all fragments get the original skb->tstamp. Note that this original skb->tstamp should be zero in forwarding path, but might have a non zero value in output path if user decided so. Fixes: fb420d5d91c1 ("tcp/fq: move back to CLOCK_MONOTONIC") Signed-off-by: Eric Dumazet Reported-by: Thomas Bartschies Signed-off-by: David S. Miller --- net/bridge/netfilter/nf_conntrack_bridge.c | 3 +++ net/ipv4/ip_output.c | 3 +++ net/ipv6/ip6_output.c | 3 +++ net/ipv6/netfilter.c | 3 +++ 4 files changed, 12 insertions(+) diff --git a/net/bridge/netfilter/nf_conntrack_bridge.c b/net/bridge/netfilter/nf_conntrack_bridge.c index 8842798c29e6..506d6141e44e 100644 --- a/net/bridge/netfilter/nf_conntrack_bridge.c +++ b/net/bridge/netfilter/nf_conntrack_bridge.c @@ -33,6 +33,7 @@ static int nf_br_ip_fragment(struct net *net, struct sock *sk, { int frag_max_size = BR_INPUT_SKB_CB(skb)->frag_max_size; unsigned int hlen, ll_rs, mtu; + ktime_t tstamp = skb->tstamp; struct ip_frag_state state; struct iphdr *iph; int err; @@ -80,6 +81,7 @@ static int nf_br_ip_fragment(struct net *net, struct sock *sk, if (iter.frag) ip_fraglist_prepare(skb, &iter); + skb->tstamp = tstamp; err = output(net, sk, data, skb); if (err || !iter.frag) break; @@ -104,6 +106,7 @@ slow_path: goto blackhole; } + skb2->tstamp = tstamp; err = output(net, sk, data, skb2); if (err) goto blackhole; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 28fca408812c..814b9b8882a0 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -771,6 +771,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, struct rtable *rt = skb_rtable(skb); unsigned int mtu, hlen, ll_rs; struct ip_fraglist_iter iter; + ktime_t tstamp = skb->tstamp; struct ip_frag_state state; int err = 0; @@ -846,6 +847,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, ip_fraglist_prepare(skb, &iter); } + skb->tstamp = tstamp; err = output(net, sk, skb); if (!err) @@ -900,6 +902,7 @@ slow_path: /* * Put this fragment into the sending queue. */ + skb2->tstamp = tstamp; err = output(net, sk, skb2); if (err) goto fail; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index edadee4a7e76..71827b56c006 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -768,6 +768,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, inet6_sk(skb->sk) : NULL; struct ip6_frag_state state; unsigned int mtu, hlen, nexthdr_offset; + ktime_t tstamp = skb->tstamp; int hroom, err = 0; __be32 frag_id; u8 *prevhdr, nexthdr = 0; @@ -855,6 +856,7 @@ int ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, if (iter.frag) ip6_fraglist_prepare(skb, &iter); + skb->tstamp = tstamp; err = output(net, sk, skb); if (!err) IP6_INC_STATS(net, ip6_dst_idev(&rt->dst), @@ -913,6 +915,7 @@ slow_path: /* * Put this fragment into the sending queue. */ + frag->tstamp = tstamp; err = output(net, sk, frag); if (err) goto fail; diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index a9bff556d3b2..409e79b84a83 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -119,6 +119,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, struct sk_buff *)) { int frag_max_size = BR_INPUT_SKB_CB(skb)->frag_max_size; + ktime_t tstamp = skb->tstamp; struct ip6_frag_state state; u8 *prevhdr, nexthdr = 0; unsigned int mtu, hlen; @@ -183,6 +184,7 @@ int br_ip6_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, if (iter.frag) ip6_fraglist_prepare(skb, &iter); + skb->tstamp = tstamp; err = output(net, sk, data, skb); if (err || !iter.frag) break; @@ -215,6 +217,7 @@ slow_path: goto blackhole; } + skb2->tstamp = tstamp; err = output(net, sk, data, skb2); if (err) goto blackhole; From 2e978795bb4c14293bf6ecf32621d32529706aef Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Thu, 17 Oct 2019 10:11:03 +0300 Subject: [PATCH 133/145] mlxsw: spectrum_trap: Push Ethernet header before reporting trap devlink maintains packets and bytes statistics for each trap. Since eth_type_trans() was called to set the skb's protocol, the data pointer no longer points to the start of the packet and the bytes accounting is off by 14 bytes. Fix this by pushing the skb's data pointer to the start of the packet. Fixes: b5ce611fd96e ("mlxsw: spectrum: Add devlink-trap support") Reported-by: Alex Kushnarov Tested-by: Alex Kushnarov Acked-by: Jiri Pirko Signed-off-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c index 899450b28621..7c03b661ae7e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c @@ -99,6 +99,7 @@ static void mlxsw_sp_rx_drop_listener(struct sk_buff *skb, u8 local_port, devlink = priv_to_devlink(mlxsw_sp->core); in_devlink_port = mlxsw_core_port_devlink_port_get(mlxsw_sp->core, local_port); + skb_push(skb, ETH_HLEN); devlink_trap_report(devlink, skb, trap_ctx, in_devlink_port); consume_skb(skb); } From ec3359b685db834c249953db6ac5717bf5cde425 Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Thu, 17 Oct 2019 14:44:02 +0200 Subject: [PATCH 134/145] vsock/virtio: send a credit update when buffer size is changed When the user application set a new buffer size value, we should update the remote peer about this change, since it uses this information to calculate the credit available. Signed-off-by: Stefano Garzarella Signed-off-by: David S. Miller --- net/vmw_vsock/virtio_transport_common.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index a666ef8fc54e..db127a69f5c3 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -458,6 +458,9 @@ void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val) vvs->buf_size_max = val; vvs->buf_size = val; vvs->buf_alloc = val; + + virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_STREAM, + NULL); } EXPORT_SYMBOL_GPL(virtio_transport_set_buffer_size); From ae6fcfbf5f03de3407b809aaee319330d3dc7f8b Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Thu, 17 Oct 2019 14:44:03 +0200 Subject: [PATCH 135/145] vsock/virtio: discard packets if credit is not respected If the remote peer doesn't respect the credit information (buf_alloc, fwd_cnt), sending more data than it can send, we should drop the packets to prevent a malicious peer from using all of our memory. This is patch follows the VIRTIO spec: "VIRTIO_VSOCK_OP_RW data packets MUST only be transmitted when the peer has sufficient free buffer space for the payload" Signed-off-by: Stefano Garzarella Signed-off-by: David S. Miller --- net/vmw_vsock/virtio_transport_common.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index db127a69f5c3..481f7f8a1655 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -204,10 +204,14 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, return virtio_transport_get_ops()->send_pkt(pkt); } -static void virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs, +static bool virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs, struct virtio_vsock_pkt *pkt) { + if (vvs->rx_bytes + pkt->len > vvs->buf_alloc) + return false; + vvs->rx_bytes += pkt->len; + return true; } static void virtio_transport_dec_rx_pkt(struct virtio_vsock_sock *vvs, @@ -879,14 +883,18 @@ virtio_transport_recv_enqueue(struct vsock_sock *vsk, struct virtio_vsock_pkt *pkt) { struct virtio_vsock_sock *vvs = vsk->trans; - bool free_pkt = false; + bool can_enqueue, free_pkt = false; pkt->len = le32_to_cpu(pkt->hdr.len); pkt->off = 0; spin_lock_bh(&vvs->rx_lock); - virtio_transport_inc_rx_pkt(vvs, pkt); + can_enqueue = virtio_transport_inc_rx_pkt(vvs, pkt); + if (!can_enqueue) { + free_pkt = true; + goto out; + } /* Try to copy small packets into the buffer of last packet queued, * to avoid wasting memory queueing the entire buffer with a small From 38b4fe320119859c11b1dc06f6b4987a16344fa1 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Thu, 17 Oct 2019 21:29:26 +0200 Subject: [PATCH 136/145] net: usb: lan78xx: Connect PHY before registering MAC As soon as the netdev is registers, the kernel can start using the interface. If the driver connects the MAC to the PHY after the netdev is registered, there is a race condition where the interface can be opened without having the PHY connected. Change the order to close this race condition. Fixes: 92571a1aae40 ("lan78xx: Connect phy early") Reported-by: Daniel Wagner Signed-off-by: Andrew Lunn Tested-by: Daniel Wagner Signed-off-by: David S. Miller --- drivers/net/usb/lan78xx.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c index 58f5a219fb65..62948098191f 100644 --- a/drivers/net/usb/lan78xx.c +++ b/drivers/net/usb/lan78xx.c @@ -3782,10 +3782,14 @@ static int lan78xx_probe(struct usb_interface *intf, /* driver requires remote-wakeup capability during autosuspend. */ intf->needs_remote_wakeup = 1; + ret = lan78xx_phy_init(dev); + if (ret < 0) + goto out4; + ret = register_netdev(netdev); if (ret != 0) { netif_err(dev, probe, netdev, "couldn't register the device\n"); - goto out4; + goto out5; } usb_set_intfdata(intf, dev); @@ -3798,14 +3802,10 @@ static int lan78xx_probe(struct usb_interface *intf, pm_runtime_set_autosuspend_delay(&udev->dev, DEFAULT_AUTOSUSPEND_DELAY); - ret = lan78xx_phy_init(dev); - if (ret < 0) - goto out5; - return 0; out5: - unregister_netdev(netdev); + phy_disconnect(netdev->phydev); out4: usb_free_urb(dev->urb_intr); out3: From 580a05f9d4ada3bfb689140d0efec1efdb8a48da Mon Sep 17 00:00:00 2001 From: Yonglong Liu Date: Fri, 18 Oct 2019 11:42:59 +0800 Subject: [PATCH 137/145] net: hns3: fix mis-counting IRQ vector numbers issue Currently, the num_msi_left means the vector numbers of NIC, but if the PF supported RoCE, it contains the vector numbers of NIC and RoCE(Not expected). This may cause interrupts lost in some case, because of the NIC module used the vector resources which belongs to RoCE. This patch adds a new variable num_nic_msi to store the vector numbers of NIC, and adjust the default TQP numbers and rss_size according to the value of num_nic_msi. Fixes: 46a3df9f9718 ("net: hns3: Add HNS3 Acceleration Engine & Compatibility Layer Support") Signed-off-by: Yonglong Liu Signed-off-by: Huazhong Tan Signed-off-by: David S. Miller --- drivers/net/ethernet/hisilicon/hns3/hnae3.h | 2 ++ .../hisilicon/hns3/hns3pf/hclge_main.c | 21 +++++++++++++- .../hisilicon/hns3/hns3pf/hclge_main.h | 1 + .../ethernet/hisilicon/hns3/hns3pf/hclge_tm.c | 11 ++++++-- .../hisilicon/hns3/hns3vf/hclgevf_main.c | 28 +++++++++++++++++-- .../hisilicon/hns3/hns3vf/hclgevf_main.h | 1 + 6 files changed, 58 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h index c4b7bf851a28..75ccc1e7076b 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h +++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h @@ -32,6 +32,8 @@ #define HNAE3_MOD_VERSION "1.0" +#define HNAE3_MIN_VECTOR_NUM 2 /* first one for misc, another for IO */ + /* Device IDs */ #define HNAE3_DEV_ID_GE 0xA220 #define HNAE3_DEV_ID_25GE 0xA221 diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index fd7f94372ff0..e02e01bd9eff 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -906,6 +906,9 @@ static int hclge_query_pf_resource(struct hclge_dev *hdev) hnae3_get_field(__le16_to_cpu(req->pf_intr_vector_number), HCLGE_PF_VEC_NUM_M, HCLGE_PF_VEC_NUM_S); + /* nic's msix numbers is always equals to the roce's. */ + hdev->num_nic_msi = hdev->num_roce_msi; + /* PF should have NIC vectors and Roce vectors, * NIC vectors are queued before Roce vectors. */ @@ -915,6 +918,15 @@ static int hclge_query_pf_resource(struct hclge_dev *hdev) hdev->num_msi = hnae3_get_field(__le16_to_cpu(req->pf_intr_vector_number), HCLGE_PF_VEC_NUM_M, HCLGE_PF_VEC_NUM_S); + + hdev->num_nic_msi = hdev->num_msi; + } + + if (hdev->num_nic_msi < HNAE3_MIN_VECTOR_NUM) { + dev_err(&hdev->pdev->dev, + "Just %u msi resources, not enough for pf(min:2).\n", + hdev->num_nic_msi); + return -EINVAL; } return 0; @@ -1507,6 +1519,10 @@ static int hclge_assign_tqp(struct hclge_vport *vport, u16 num_tqps) kinfo->rss_size = min_t(u16, hdev->rss_size_max, vport->alloc_tqps / hdev->tm_info.num_tc); + /* ensure one to one mapping between irq and queue at default */ + kinfo->rss_size = min_t(u16, kinfo->rss_size, + (hdev->num_nic_msi - 1) / hdev->tm_info.num_tc); + return 0; } @@ -2285,7 +2301,8 @@ static int hclge_init_msi(struct hclge_dev *hdev) int vectors; int i; - vectors = pci_alloc_irq_vectors(pdev, 1, hdev->num_msi, + vectors = pci_alloc_irq_vectors(pdev, HNAE3_MIN_VECTOR_NUM, + hdev->num_msi, PCI_IRQ_MSI | PCI_IRQ_MSIX); if (vectors < 0) { dev_err(&pdev->dev, @@ -2300,6 +2317,7 @@ static int hclge_init_msi(struct hclge_dev *hdev) hdev->num_msi = vectors; hdev->num_msi_left = vectors; + hdev->base_msi_vector = pdev->irq; hdev->roce_base_vector = hdev->base_msi_vector + hdev->roce_base_msix_offset; @@ -3903,6 +3921,7 @@ static int hclge_get_vector(struct hnae3_handle *handle, u16 vector_num, int alloc = 0; int i, j; + vector_num = min_t(u16, hdev->num_nic_msi - 1, vector_num); vector_num = min(hdev->num_msi_left, vector_num); for (j = 0; j < vector_num; j++) { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h index 3e9574a9e22d..c3d56b872ed7 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h @@ -763,6 +763,7 @@ struct hclge_dev { u32 base_msi_vector; u16 *vector_status; int *vector_irq; + u16 num_nic_msi; /* Num of nic vectors for this PF */ u16 num_roce_msi; /* Num of roce vectors for this PF */ int roce_base_vector; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c index 9f0e35f27789..62399cc1c5a6 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c @@ -537,9 +537,16 @@ static void hclge_tm_vport_tc_info_update(struct hclge_vport *vport) kinfo->rss_size = kinfo->req_rss_size; } else if (kinfo->rss_size > max_rss_size || (!kinfo->req_rss_size && kinfo->rss_size < max_rss_size)) { + /* if user not set rss, the rss_size should compare with the + * valid msi numbers to ensure one to one map between tqp and + * irq as default. + */ + if (!kinfo->req_rss_size) + max_rss_size = min_t(u16, max_rss_size, + (hdev->num_nic_msi - 1) / + kinfo->num_tc); + /* Set to the maximum specification value (max_rss_size). */ - dev_info(&hdev->pdev->dev, "rss changes from %d to %d\n", - kinfo->rss_size, max_rss_size); kinfo->rss_size = max_rss_size; } diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index e3090b3dab1d..7d7e712691b9 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -411,6 +411,13 @@ static int hclgevf_knic_setup(struct hclgevf_dev *hdev) kinfo->tqp[i] = &hdev->htqp[i].q; } + /* after init the max rss_size and tqps, adjust the default tqp numbers + * and rss size with the actual vector numbers + */ + kinfo->num_tqps = min_t(u16, hdev->num_nic_msix - 1, kinfo->num_tqps); + kinfo->rss_size = min_t(u16, kinfo->num_tqps / kinfo->num_tc, + kinfo->rss_size); + return 0; } @@ -502,6 +509,7 @@ static int hclgevf_get_vector(struct hnae3_handle *handle, u16 vector_num, int alloc = 0; int i, j; + vector_num = min_t(u16, hdev->num_nic_msix - 1, vector_num); vector_num = min(hdev->num_msi_left, vector_num); for (j = 0; j < vector_num; j++) { @@ -2246,13 +2254,14 @@ static int hclgevf_init_msi(struct hclgevf_dev *hdev) int vectors; int i; - if (hnae3_get_bit(hdev->ae_dev->flag, HNAE3_DEV_SUPPORT_ROCE_B)) + if (hnae3_dev_roce_supported(hdev)) vectors = pci_alloc_irq_vectors(pdev, hdev->roce_base_msix_offset + 1, hdev->num_msi, PCI_IRQ_MSIX); else - vectors = pci_alloc_irq_vectors(pdev, 1, hdev->num_msi, + vectors = pci_alloc_irq_vectors(pdev, HNAE3_MIN_VECTOR_NUM, + hdev->num_msi, PCI_IRQ_MSI | PCI_IRQ_MSIX); if (vectors < 0) { @@ -2268,6 +2277,7 @@ static int hclgevf_init_msi(struct hclgevf_dev *hdev) hdev->num_msi = vectors; hdev->num_msi_left = vectors; + hdev->base_msi_vector = pdev->irq; hdev->roce_base_vector = pdev->irq + hdev->roce_base_msix_offset; @@ -2533,7 +2543,7 @@ static int hclgevf_query_vf_resource(struct hclgevf_dev *hdev) req = (struct hclgevf_query_res_cmd *)desc.data; - if (hnae3_get_bit(hdev->ae_dev->flag, HNAE3_DEV_SUPPORT_ROCE_B)) { + if (hnae3_dev_roce_supported(hdev)) { hdev->roce_base_msix_offset = hnae3_get_field(__le16_to_cpu(req->msixcap_localid_ba_rocee), HCLGEVF_MSIX_OFT_ROCEE_M, @@ -2542,6 +2552,9 @@ static int hclgevf_query_vf_resource(struct hclgevf_dev *hdev) hnae3_get_field(__le16_to_cpu(req->vf_intr_vector_number), HCLGEVF_VEC_NUM_M, HCLGEVF_VEC_NUM_S); + /* nic's msix numbers is always equals to the roce's. */ + hdev->num_nic_msix = hdev->num_roce_msix; + /* VF should have NIC vectors and Roce vectors, NIC vectors * are queued before Roce vectors. The offset is fixed to 64. */ @@ -2551,6 +2564,15 @@ static int hclgevf_query_vf_resource(struct hclgevf_dev *hdev) hdev->num_msi = hnae3_get_field(__le16_to_cpu(req->vf_intr_vector_number), HCLGEVF_VEC_NUM_M, HCLGEVF_VEC_NUM_S); + + hdev->num_nic_msix = hdev->num_msi; + } + + if (hdev->num_nic_msix < HNAE3_MIN_VECTOR_NUM) { + dev_err(&hdev->pdev->dev, + "Just %u msi resources, not enough for vf(min:2).\n", + hdev->num_nic_msix); + return -EINVAL; } return 0; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h index bdde3afc286b..2b8d6bc6d224 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h @@ -270,6 +270,7 @@ struct hclgevf_dev { u16 num_msi; u16 num_msi_left; u16 num_msi_used; + u16 num_nic_msix; /* Num of nic vectors for this VF */ u16 num_roce_msix; /* Num of roce vectors for this VF */ u16 roce_base_msix_offset; int roce_base_vector; From 3d5c1a037d37392a6859afbde49be5ba6a70a6b3 Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Fri, 18 Oct 2019 09:45:49 +0200 Subject: [PATCH 138/145] xen/netback: fix error path of xenvif_connect_data() xenvif_connect_data() calls module_put() in case of error. This is wrong as there is no related module_get(). Remove the superfluous module_put(). Fixes: 279f438e36c0a7 ("xen-netback: Don't destroy the netdev until the vif is shut down") Cc: # 3.12 Signed-off-by: Juergen Gross Reviewed-by: Paul Durrant Reviewed-by: Wei Liu Signed-off-by: David S. Miller --- drivers/net/xen-netback/interface.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c index 240f762b3749..103ed00775eb 100644 --- a/drivers/net/xen-netback/interface.c +++ b/drivers/net/xen-netback/interface.c @@ -719,7 +719,6 @@ err_unmap: xenvif_unmap_frontend_data_rings(queue); netif_napi_del(&queue->napi); err: - module_put(THIS_MODULE); return err; } From bd310aca442fcc72731b7acb65d32d05c956d56b Mon Sep 17 00:00:00 2001 From: Michael Tretter Date: Fri, 18 Oct 2019 16:11:43 +0200 Subject: [PATCH 139/145] macb: propagate errors when getting optional clocks The tx_clk, rx_clk, and tsu_clk are optional. Currently the macb driver marks clock as not available if it receives an error when trying to get a clock. This is wrong, because a clock controller might return -EPROBE_DEFER if a clock is not available, but will eventually become available. In these cases, the driver would probe successfully but will never be able to adjust the clocks, because the clocks were not available during probe, but became available later. For example, the clock controller for the ZynqMP is implemented in the PMU firmware and the clocks are only available after the firmware driver has been probed. Use devm_clk_get_optional() in instead of devm_clk_get() to get the optional clock and propagate all errors to the calling function. Signed-off-by: Michael Tretter Acked-by: Nicolas Ferre Tested-by: Nicolas Ferre Signed-off-by: David S. Miller --- drivers/net/ethernet/cadence/macb_main.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 8e8d557901a9..1e1b774e1953 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -3405,17 +3405,17 @@ static int macb_clk_init(struct platform_device *pdev, struct clk **pclk, return err; } - *tx_clk = devm_clk_get(&pdev->dev, "tx_clk"); + *tx_clk = devm_clk_get_optional(&pdev->dev, "tx_clk"); if (IS_ERR(*tx_clk)) - *tx_clk = NULL; + return PTR_ERR(*tx_clk); - *rx_clk = devm_clk_get(&pdev->dev, "rx_clk"); + *rx_clk = devm_clk_get_optional(&pdev->dev, "rx_clk"); if (IS_ERR(*rx_clk)) - *rx_clk = NULL; + return PTR_ERR(*rx_clk); - *tsu_clk = devm_clk_get(&pdev->dev, "tsu_clk"); + *tsu_clk = devm_clk_get_optional(&pdev->dev, "tsu_clk"); if (IS_ERR(*tsu_clk)) - *tsu_clk = NULL; + return PTR_ERR(*tsu_clk); err = clk_prepare_enable(*pclk); if (err) { From a7fa12d15855904aff1716e1fc723c03ba38c5cc Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 18 Oct 2019 09:16:57 -0700 Subject: [PATCH 140/145] net: netem: fix error path for corrupted GSO frames To corrupt a GSO frame we first perform segmentation. We then proceed using the first segment instead of the full GSO skb and requeue the rest of the segments as separate packets. If there are any issues with processing the first segment we still want to process the rest, therefore we jump to the finish_segs label. Commit 177b8007463c ("net: netem: fix backlog accounting for corrupted GSO frames") started using the pointer to the first segment in the "rest of segments processing", but as mentioned above the first segment may had already been freed at this point. Backlog corrections for parent qdiscs have to be adjusted. Fixes: 177b8007463c ("net: netem: fix backlog accounting for corrupted GSO frames") Reported-by: kbuild test robot Reported-by: Dan Carpenter Reported-by: Ben Hutchings Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_netem.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 0e44039e729c..942eb17f413c 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -509,6 +509,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb)) { qdisc_drop(skb, sch, to_free); + skb = NULL; goto finish_segs; } @@ -593,9 +594,10 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, finish_segs: if (segs) { unsigned int len, last_len; - int nb = 0; + int nb; - len = skb->len; + len = skb ? skb->len : 0; + nb = skb ? 1 : 0; while (segs) { skb2 = segs->next; @@ -612,7 +614,8 @@ finish_segs: } segs = skb2; } - qdisc_tree_reduce_backlog(sch, -nb, prev_len - len); + /* Parent qdiscs accounted for 1 skb of size @prev_len */ + qdisc_tree_reduce_backlog(sch, -(nb - 1), -(len - prev_len)); } return NET_XMIT_SUCCESS; } From e0ad032e144731a5928f2d75e91c2064ba1a764c Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 18 Oct 2019 09:16:58 -0700 Subject: [PATCH 141/145] net: netem: correct the parent's backlog when corrupted packet was dropped If packet corruption failed we jump to finish_segs and return NET_XMIT_SUCCESS. Seeing success will make the parent qdisc increment its backlog, that's incorrect - we need to return NET_XMIT_DROP. Fixes: 6071bd1aa13e ("netem: Segment GSO packets on enqueue") Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- net/sched/sch_netem.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 942eb17f413c..42e557d48e4e 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -616,6 +616,8 @@ finish_segs: } /* Parent qdiscs accounted for 1 skb of size @prev_len */ qdisc_tree_reduce_backlog(sch, -(nb - 1), -(len - prev_len)); + } else if (!skb) { + return NET_XMIT_DROP; } return NET_XMIT_SUCCESS; } From 3d00cf2fbb61212f47a3cf838be51c921366c937 Mon Sep 17 00:00:00 2001 From: Chenwandun Date: Fri, 18 Oct 2019 18:20:37 +0800 Subject: [PATCH 142/145] net: aquantia: add an error handling in aq_nic_set_multicast_list add an error handling in aq_nic_set_multicast_list, it may not work when hw_multicast_list_set error; and at the same time it will remove gcc Wunused-but-set-variable warning. Signed-off-by: Chenwandun Reviewed-by: Igor Russkikh Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index 2a18439b36fb..137c1de4c6ec 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -664,6 +664,8 @@ int aq_nic_set_multicast_list(struct aq_nic_s *self, struct net_device *ndev) err = hw_ops->hw_multicast_list_set(self->aq_hw, self->mc_list.ar, self->mc_list.count); + if (err < 0) + return err; } return aq_nic_set_packet_filter(self, packet_filter); } From 05908d72cc8fa39058cd9535db2833089cc3df6f Mon Sep 17 00:00:00 2001 From: Mans Rullgard Date: Fri, 18 Oct 2019 17:56:58 +0100 Subject: [PATCH 143/145] net: ethernet: dwmac-sun8i: show message only when switching to promisc Printing the info message every time more than the max number of mac addresses are requested generates unnecessary log spam. Showing it only when the hw is not already in promiscous mode is equally informative without being annoying. Signed-off-by: Mans Rullgard Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c index f97a4096f8fc..ddcc191febdb 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-sun8i.c @@ -651,7 +651,8 @@ static void sun8i_dwmac_set_filter(struct mac_device_info *hw, } } } else { - netdev_info(dev, "Too many address, switching to promiscuous\n"); + if (!(readl(ioaddr + EMAC_RX_FRM_FLT) & EMAC_FRM_FLT_RXALL)) + netdev_info(dev, "Too many address, switching to promiscuous\n"); v = EMAC_FRM_FLT_RXALL; } From 50c7d2ba9de20f60a2d527ad6928209ef67e4cdd Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Fri, 18 Oct 2019 17:02:46 -0400 Subject: [PATCH 144/145] net: dsa: fix switch tree list If there are multiple switch trees on the device, only the last one will be listed, because the arguments of list_add_tail are swapped. Fixes: 83c0afaec7b7 ("net: dsa: Add new binding implementation") Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/dsa2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 73002022c9d8..716d265ba8ca 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -46,7 +46,7 @@ static struct dsa_switch_tree *dsa_tree_alloc(int index) dst->index = index; INIT_LIST_HEAD(&dst->list); - list_add_tail(&dsa_tree_list, &dst->list); + list_add_tail(&dst->list, &dsa_tree_list); kref_init(&dst->refcount); From 2a06b8982f8f2f40d03a3daf634676386bd84dbc Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 18 Oct 2019 15:20:05 -0700 Subject: [PATCH 145/145] net: reorder 'struct net' fields to avoid false sharing Intel test robot reported a ~7% regression on TCP_CRR tests that they bisected to the cited commit. Indeed, every time a new TCP socket is created or deleted, the atomic counter net->count is touched (via get_net(net) and put_net(net) calls) So cpus might have to reload a contended cache line in net_hash_mix(net) calls. We need to reorder 'struct net' fields to move @hash_mix in a read mostly cache line. We move in the first cache line fields that can be dirtied often. We probably will have to address in a followup patch the __randomize_layout that was added in linux-4.13, since this might break our placement choices. Fixes: 355b98553789 ("netns: provide pure entropy for net_hash_mix()") Signed-off-by: Eric Dumazet Reported-by: kernel test robot Signed-off-by: David S. Miller --- include/net/net_namespace.h | 25 +++++++++++++++++-------- 1 file changed, 17 insertions(+), 8 deletions(-) diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index f8712bbeb2e0..4c2cd9378699 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -52,6 +52,9 @@ struct bpf_prog; #define NETDEV_HASHENTRIES (1 << NETDEV_HASHBITS) struct net { + /* First cache line can be often dirtied. + * Do not place here read-mostly fields. + */ refcount_t passive; /* To decide when the network * namespace should be freed. */ @@ -60,7 +63,13 @@ struct net { */ spinlock_t rules_mod_lock; - u32 hash_mix; + unsigned int dev_unreg_count; + + unsigned int dev_base_seq; /* protected by rtnl_mutex */ + int ifindex; + + spinlock_t nsid_lock; + atomic_t fnhe_genid; struct list_head list; /* list of network namespaces */ struct list_head exit_list; /* To linked to call pernet exit @@ -76,11 +85,11 @@ struct net { #endif struct user_namespace *user_ns; /* Owning user namespace */ struct ucounts *ucounts; - spinlock_t nsid_lock; struct idr netns_ids; struct ns_common ns; + struct list_head dev_base_head; struct proc_dir_entry *proc_net; struct proc_dir_entry *proc_net_stat; @@ -93,17 +102,18 @@ struct net { struct uevent_sock *uevent_sock; /* uevent socket */ - struct list_head dev_base_head; struct hlist_head *dev_name_head; struct hlist_head *dev_index_head; - unsigned int dev_base_seq; /* protected by rtnl_mutex */ - int ifindex; - unsigned int dev_unreg_count; + /* Note that @hash_mix can be read millions times per second, + * it is critical that it is on a read_mostly cache line. + */ + u32 hash_mix; + + struct net_device *loopback_dev; /* The loopback */ /* core fib_rules */ struct list_head rules_ops; - struct net_device *loopback_dev; /* The loopback */ struct netns_core core; struct netns_mib mib; struct netns_packet packet; @@ -171,7 +181,6 @@ struct net { struct sock *crypto_nlsk; #endif struct sock *diag_nlsk; - atomic_t fnhe_genid; } __randomize_layout; #include