The Documentation/memory-barriers.txt document requires that atomic
operations that return a value act as a memory barrier both before
and after the actual atomic operation.
Our current implementation doesn't guarantee this. More specifically,
while a load following the isync can not be issued before stwcx. has
completed, that completion doesn't architecturally means that the
result of stwcx. is visible to other processors (or any previous stores
for that matter) (typically, the other processors L1 caches can still
hold the old value).
This has caused an actual crash in RCU torture testing on Power 7
This fixes it by changing those atomic ops to use new macros instead
of RELEASE/ACQUIRE barriers, called ATOMIC_ENTRY and ATMOIC_EXIT barriers,
which are then defined respectively to lwsync and sync.
I haven't had a chance to measure the performance impact (or rather
what I measured with kernel compiles is in the noise, I yet have to
find a more precise benchmark)
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Recent binutils refuses to assemble AltiVec opcodes when in e500/SPE
mode, as some of those opcodes alias the "SPE" instructions. This
triggers an ancient binutils version check even when building a kernel
with CONFIG_ALTIVEC disabled.
In theory, the check could be conditionalized on CONFIG_ALTIVEC, but in
practice it has long outlived its utility. It is virtually impossible
to find binutils older than 2.12.1 (released 2002) in the wild anymore.
Even ancient RedHat Enterprise Linux 4 has binutils-2.14.
To fix the kernel build when done natively on e500 systems with this new
binutils, the test is simply removed.
Signed-off-by: Kyle Moffett <Kyle.D.Moffett@boeing.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
With the introduction of CONFIG_PPC_ADV_DEBUG_REGS user space debug is
broken on Book-E 64-bit parts that support delayed debug events. When
switch_booke_debug_regs() sets DBCR0 we'll start getting debug events as
MSR_DE is also set and we aren't able to handle debug events from kernel
space.
We can remove the hack that always enables MSR_DE and loads up DBCR0 and
just utilize switch_booke_debug_regs() to get user space debug working
again.
We still need to handle critical/debug exception stacks & proper
save/restore of state for those exception levles to support debug events
from kernel space like we have on 32-bit.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
All of DebugException is already protected by CONFIG_PPC_ADV_DEBUG_REGS
there is no need to have another such ifdef inside the function.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We had an existing ifdef for 4xx & BOOKE processors that got changed to
CONFIG_PPC_ADV_DEBUG_REGS. The define has nothing to do with
CONFIG_PPC_ADV_DEBUG_REGS. The define really should be:
#if defined(CONFIG_4xx) || defined(CONFIG_BOOKE)
and not
#ifdef CONFIG_PPC_ADV_DEBUG_REGS
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Mark stats timer as deferrable: punctuality in waking the stats timer
callback doesn't matter much, as it is responsible only to avoid
integer wraparound.
We need at least 1 other timer to fire within 17s (fully loaded 1Gbps)
to avoid wrap-arounds. Desired period is still 10s.
Signed-off-by: David Decotigny <david.decotigny@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds code to update the stats counter for dropped RX frames.
Signed-off-by: David Decotigny <david.decotigny@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit implements the ndo_get_stats64() API for forcedeth. Since
hardware stats are being updated from different contexts (process and
timer), this commit adds synchronization. For software stats, it
relies on the u64_stats_sync.h API.
Tested:
- 16-way SMP x86_64 ->
RX bytes:7244556582 (7.2 GB) TX bytes:181904254 (181.9 MB)
- pktgen + loopback: identical rx_bytes/tx_bytes and rx_packets/tx_packets
Signed-off-by: David Decotigny <david.decotigny@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds a new module parameter "debug_tx_timeout" to silence most
debug messages in case of TX timeout. These messages don't provide a
signal/noise ratio high enough for production systems and, with ~30kB
logged each time, they tend to add to a cascade effect if the system
is already under stress (memory pressure, disk, etc.).
By default, the parameter is clear, meaning that only a single warning
will be reported.
Signed-off-by: David Decotigny <david.decotigny@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds a few kernel messages to indicate whether PCIe interrupts
are signaled with MSI or MSI-X.
Signed-off-by: David Decotigny <david.decotigny@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds the /sys/class/net/DEV/queues/Q/tx_timeout attribute
containing the total number of timeout events on the given queue. It
is always available with CONFIG_SYSFS, independently of
CONFIG_RPS/XPS.
Credits to Stephen Hemminger for a preliminary version of this patch.
Tested:
without CONFIG_SYSFS (compilation only)
with sysfs and without CONFIG_RPS & CONFIG_XPS
with sysfs and without CONFIG_RPS
with sysfs and without CONFIG_XPS
with defaults
Signed-off-by: David Decotigny <david.decotigny@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit fixes following warning:
net/core/net-sysfs.c:921:6: warning: symbol 'numa_node' shadows an earlier one
include/linux/topology.h:222:1: originally declared here
Signed-off-by: David Decotigny <david.decotigny@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This change makes sure that tx_packets/rx_bytes ifconfig counters are
updated even on NICs that don't provide hardware support for these
stats: they are now updated in software. For the sake of consistency,
we also now have tx_bytes updated in software (hardware counters
include ethernet CRC, and software doesn't account for it).
This reverts parts of:
- "forcedeth: statistics optimization" (21828163b2)
- "forcedeth: Improve stats counters" (0bdfea8ba8)
- "forcedeth: remove unneeded stats updates" (4687f3f364)
Tested:
pktgen + loopback (http://patchwork.ozlabs.org/patch/124698/)
reports identical tx_packets/rx_packets and tx_bytes/rx_bytes.
Signed-off-by: David Decotigny <david.decotigny@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 898bdf2cb4)
Save some punctuation by using bool type's property equivalent to
doubled negation operator.
Reported-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
takes vfsmount and relative path, does lookup within that vfsmount
(possibly triggering automounts) and returns the result as root
of subtree suitable for return by ->mount() (i.e. a reference to
dentry and an active reference to its superblock grabbed, superblock
locked exclusive).
btrfs and nfs switched to it instead of open-coding the sucker.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add missing spaces around multiplication operator.
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rather than generating a different RSS key on each boot, just use
a predetermined value that will map same flow to same value on
every device.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The default Tx ring size for the sky2 driver is quite large and could
cause excess buffer bloat for many users. The minimum ring size
possible and still allow handling the worst case packet on 64bit platforms
is 38 which gets rounded up to a power of 2. But most packets only require
a couple of ring elements.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The code is clearer if the up/down functions are renamed to
open/close like other drivers. Purely syntax change.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A couple of the reset and setup paths have possible PCI posting issues.
When setting registers, a read is necessary to force the writes to complete.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are several problems with recent change to how IRQ's are setup.
* synchronize_irq in sky2_shutdown would hang because there
was no IRQ setup.
* when device was set to down, some IRQ bits left enabled so a
hardware error would produce IRQ with no handler
* quick link on Optima chip set was enabled without handler
* suspend/resume would leave IRQ on with no handler if device
was down
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We are checking whether the MCR0_HASHEN bit is set using a logical and
instead of bitwise and, fix that.
Signed-off-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reduce the number of #defines, use the normal #define from if_ether.h
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reduce the number of #defines, use the normal #define from if_ether.h
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reduce the number of #defines, use the normal #define from if_ether.h
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reduce the number of #defines, use the normal #define from if_ether.h
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reduce the number of #defines, use the normal #define from if_ether.h
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If there is an architecture-specific random number generator we use
it to acquire randomness one "long" at a time. We should put these
random words into consecutive words in the result buffer - not just
overwrite the first word again and again.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/4ec4061010261a4cb0@agluck-desktop.sc.intel.com
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Translating between ethtool advertisement settings and MII
advertisements are common operations for ethernet drivers. This patch
adds a set of helper functions that implements the conversion. The
patch then modifies a couple of the drivers to use the new functions.
Signed-off-by: Matt Carlson <mcarlson@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since multiple team instances are putting defined options into their
option list, during register each option must be cloned before added
into list. This resolves uncool memory corruptions when using multiple
teams.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
No need to have spinlock for this purpose. So convert this to mutex and
avoid current schedule while atomic problems in netlink code.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
dev_queue_xmit() consumes its own skb, so the call to dev_kfree_skb()
in lowpan_fragment_xmit() is a double free.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Alexander Smirnov <alex.bluesman.smirnov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use V1 version of query link status command for Lancer.
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To enable VLAN promiscous mode, the HW interface should be created
with VLAN promiscous capability in Lancer. Add this capability during
creation of the HW interface.
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implemented eeprom dump using ethtool feature for Lancer.
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement register dump using ethtool for Lancer.
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Lancer uses V1 version of TXQ create. This command needs interface
id for TX queue creation. Rearrange code such that tx queue create
is after interface create. As TXQ create is now called after MCC
ring create use MCC instead of MBOX.
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
My old email account at linux-foundation is no longer usable after
the LF breakin.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Only distinct use is checking if NETIF_F_NOCACHE_COPY should be
enabled by default. The check heuristics is altered a bit here,
so it hits other people than before. The default shouldn't be
trusted for performance-critical cases anyway.
For all other uses NETIF_F_NO_CSUM is equivalent to NETIF_F_HW_CSUM.
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
v2: changed loop in ethtool_set_features() per Ben's suggestion
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
Define feature values by bit position instead of direct 2**i values
and force the values to be of type netdev_features_t.
Cleaned and extended from patch by Mahesh Bandewar <maheshb@google.com>:
+ added netdev_features_t casts
+ included bits under NETIF_F_GSO_MASK
+ moved feature #defines out of struct net_device definition
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
v2: add couple missing conversions in drivers
split unexporting netdev_fix_features()
implemented %pNF
convert sock::sk_route_(no?)caps
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move features definitions to separate header so that linux/skbuff.h won't
need to include linux/netdevice.h after netdev_features_t is introduced.
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>