mirror of
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/
synced 2026-11-15 19:56:03 +00:00
1,403,262 commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
|
|
323cc36ef6 |
Chinese translation docs for 6.15-rc1
This is the Chinese translation subtree for 6.15-rc1. It just
includes few changes:
- Chinese disclaimer change
- add a new translation doc: snp-tdx-threat-model
- fix a typo
Above patches are tested by 'make htmldocs/pdfdocs'
Signed-off-by: Alex Shi <alexs@kernel.org>
-----BEGIN PGP SIGNATURE-----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=lEYN
-----END PGP SIGNATURE-----
Merge tag 'chinese-doc-6.15-rc1' of gitolite.kernel.org:pub/scm/linux/kernel/git/alexs/linux into docs-mw
Chinese translation docs for 6.15-rc1
This is the Chinese translation subtree for 6.15-rc1. It just
includes few changes:
- Chinese disclaimer change
- add a new translation doc: snp-tdx-threat-model
- fix a typo
Above patches are tested by 'make htmldocs/pdfdocs'
Signed-off-by: Alex Shi <alexs@kernel.org>
|
||
|
|
1794d7ab34 |
wifi: mt76: mt7996: fix locking in mt7996_mac_sta_rc_work()
The 'continue' statements need to be under spinlock, since
the spinlock needs to be held as a loop invariant.
Fixes:
|
||
|
|
a0aff75e15 |
bluetooth pull request for net:
- hci_event: Fix connection regression between LE and non-LE adapters - Fix error code in chan_alloc_skb_cb() -----BEGIN PGP SIGNATURE----- iQJNBAABCAA3FiEE7E6oRXp8w05ovYr/9JCA4xAyCykFAmfUWpAZHGx1aXoudm9u LmRlbnR6QGludGVsLmNvbQAKCRD0kIDjEDILKVf5D/9sOWdQHyD5O3JUmG9Z9RP5 CWsuJMHIqNlx+DWo//rxjD5iyzMozIutW8DS7zEW+Ndct++hn2VnM2BVqmLN5c81 QqTRPJzULOTz1wpckCZ5dfpEAliXL57YUrtrsefWNEVGooBSNblkCvNjsa1Xq6BQ FmQ49avxxs4lxNBxA5vLqNBuWKqH1oeH6Q2shL350iX8qYiCk1+1cozA4LvDP+lE 2qiOirYjRaarql7rcU+SC9PSt+cVQ/0OCRsOu2n0dbrjRn5KAWsq2+zEh36dP0Ks x+tmyYQWJk6Cyb+z/N2VN2NWpIIl9OiY3PSOCfNVD43MZ/RIk8QZgW+UjapbXShO 1dDFOA4LLI5c2gNg4Px/lnNTd7bBU0smrv0AQ+CD/9zlDd0TJtbvtFCx/rYPEH8n sY2tzplfYeBLbbrHeaN0RHR0GQW1IYHcNTjctNeFcJBewKAAX2m7JnylWsG/ZCw9 qW+c5qv6L+sGCHi92AQrxc0GU3/Pdl6t1YUow0nHFPu+7Elb3T/KOkaIWRauiAA1 C6cKkp+iMyTb0Q6beh3s3FgF24MmEIBSwGAUrworvQm5+tzC5l75cv7mSQtwGMR0 8V7x0e77zaTZ9ncgvloAwu5M472TvM/HdVkG/oODH5ab19kTiUOPHN2CtNeLpFo4 pI8fKTzFtNSQISpVyARw5A== =+VUv -----END PGP SIGNATURE----- Merge tag 'for-net-2025-03-14' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth Luiz Augusto von Dentz says: ==================== bluetooth pull request for net: - hci_event: Fix connection regression between LE and non-LE adapters - Fix error code in chan_alloc_skb_cb() * tag 'for-net-2025-03-14' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth: Bluetooth: hci_event: Fix connection regression between LE and non-LE adapters Bluetooth: Fix error code in chan_alloc_skb_cb() ==================== Link: https://patch.msgid.link/20250314163847.110069-1-luiz.dentz@gmail.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
90c8021155 |
mt76 patches for 6.15
- preparation for mt7996 mlo support - fixes -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iF0EABECAB0WIQR10Rp9kadxD0kAQu/XfRQdAqdu9QUCZ9rK6QAKCRDXfRQdAqdu 9WA9AKCTVeaXEtZN3ORdnFKhz5CpLBWo7QCcDb6PRaSE/jazna7/yc6KKpmiU4c= =9Kaf -----END PGP SIGNATURE----- Merge tag 'mt76-next-2025-03-19' of https://github.com/nbd168/wireless Felix Fietkau says: ==================== mt76 patches for 6.15 - preparation for mt7996 mlo support - fixes ==================== Signed-off-by: Johannes Berg <johannes.berg@intel.com> |
||
|
|
3d97da0ee6 |
net: macb: Add __nonstring annotations for unterminated strings
When a character array without a terminating NUL character has a static
initializer, GCC 15's -Wunterminated-string-initialization will only
warn if the array lacks the "nonstring" attribute[1]. Mark the arrays
with __nonstring to correctly identify the char array as "not a C string"
and thereby eliminate the warning:
In file included from ../drivers/net/ethernet/cadence/macb_main.c:42:
../drivers/net/ethernet/cadence/macb.h:1070:35: warning: initializer-string for array of 'char' truncates NUL terminator but destination lacks 'nonstring' attribute (33 chars into 32 available) [-Wunterminated-string-initialization]
1070 | GEM_STAT_TITLE(TX1519CNT, "tx_greater_than_1518_byte_frames"),
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/net/ethernet/cadence/macb.h:1050:24: note: in definition of macro 'GEM_STAT_TITLE_BITS'
1050 | .stat_string = title, \
| ^~~~~
../drivers/net/ethernet/cadence/macb.h:1070:9: note: in expansion of macro 'GEM_STAT_TITLE'
1070 | GEM_STAT_TITLE(TX1519CNT, "tx_greater_than_1518_byte_frames"),
| ^~~~~~~~~~~~~~
../drivers/net/ethernet/cadence/macb.h:1097:35: warning: initializer-string for array of 'char' truncates NUL terminator but destination lacks 'nonstring' attribute (33 chars into 32 available) [-Wunterminated-string-initialization]
1097 | GEM_STAT_TITLE(RX1519CNT, "rx_greater_than_1518_byte_frames"),
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../drivers/net/ethernet/cadence/macb.h:1050:24: note: in definition of macro 'GEM_STAT_TITLE_BITS'
1050 | .stat_string = title, \
| ^~~~~
../drivers/net/ethernet/cadence/macb.h:1097:9: note: in expansion of macro 'GEM_STAT_TITLE'
1097 | GEM_STAT_TITLE(RX1519CNT, "rx_greater_than_1518_byte_frames"),
| ^~~~~~~~~~~~~~
Since these strings are copied with memcpy() they do not need to be
NUL terminated, and can use __nonstring:
memcpy(p, gem_statistics[i].stat_string,
ETH_GSTRING_LEN);
Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117178 [1]
Signed-off-by: Kees Cook <kees@kernel.org>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20250312200700.make.521-kees@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
||
|
|
8904eeb9de |
Merge branch 'netconsole-allow-selection-of-egress-interface-via-mac-address'
Uday Shankar says: ==================== netconsole: allow selection of egress interface via MAC address This series adds support for selecting a netconsole egress interface by specifying the MAC address (in place of the interface name) in the boot/module parameter. Signed-off-by: Uday Shankar <ushankar@purestorage.com> ==================== Link: https://patch.msgid.link/20250312-netconsole-v6-0-3437933e79b8@purestorage.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
f8a10bed32 |
netconsole: allow selection of egress interface via MAC address
Currently, netconsole has two methods of configuration - module parameter and configfs. The former interface allows for netconsole activation earlier during boot (by specifying the module parameter on the kernel command line), so it is preferred for debugging issues which arise before userspace is up/the configfs interface can be used. The module parameter syntax requires specifying the egress interface name. This requirement makes it hard to use for a couple reasons: - The egress interface name can be hard or impossible to predict. For example, installing a new network card in a system can change the interface names assigned by the kernel. - When constructing the module parameter, one may have trouble determining the original (kernel-assigned) name of the interface (which is the name that should be given to netconsole) if some stable interface naming scheme is in effect. A human can usually look at kernel logs to determine the original name, but this is very painful if automation is constructing the parameter. For these reasons, allow selection of the egress interface via MAC address when configuring netconsole using the module parameter. Update the netconsole documentation with an example of the new syntax. Selection of egress interface by MAC address via configfs is far less interesting (since when this interface can be used, one should be able to easily convert between MAC address and interface name), so it is left unimplemented. Signed-off-by: Uday Shankar <ushankar@purestorage.com> Reviewed-by: Breno Leitao <leitao@debian.org> Tested-by: Breno Leitao <leitao@debian.org> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250312-netconsole-v6-2-3437933e79b8@purestorage.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
6d6c1ba782 |
net, treewide: define and use MAC_ADDR_STR_LEN
There are a few places in the tree which compute the length of the string representation of a MAC address as 3 * ETH_ALEN - 1. Define a constant for this and use it where relevant. No functionality changes are expected. Signed-off-by: Uday Shankar <ushankar@purestorage.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Reviewed-by: Breno Leitao <leitao@debian.org> Reviewed-by: Simon Horman <horms@verge.net.au> Link: https://patch.msgid.link/20250312-netconsole-v6-1-3437933e79b8@purestorage.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
a7f2e10ecd |
hwmon fixes for v6.14-rc8/v6.14
- Fix an entry in MAINTAINERS to avoid sending hwmon review requests
to the i2c mailing list.
- Fix an out-of-bounds access in nct6775 driver
-----BEGIN PGP SIGNATURE-----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=aD6V
-----END PGP SIGNATURE-----
Merge tag 'hwmon-fixes-for-v6.14-rc8/6.14' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging
Pull hwmon fixes from Guenter Roeck:
- Fix an entry in MAINTAINERS to avoid sending hwmon review requests to
the i2c mailing list
- Fix an out-of-bounds access in nct6775 driver
* tag 'hwmon-fixes-for-v6.14-rc8/6.14' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
hwmon: (nct6775-core) Fix out of bounds access for NCT679{8,9}
MAINTAINERS: correct list and scope of LTC4286 HARDWARE MONITOR
|
||
|
|
c9cb135bc6 |
net: stmmac: dwc-qos-eth: use devm_kzalloc() for AXI data
Everywhere else in the driver uses devm_kzalloc() when allocating the
AXI data, so there is no kfree() of this structure. However,
dwc-qos-eth uses kzalloc(), which leads to this memory being leaked.
Switch to use devm_kzalloc().
Fixes:
|
||
|
|
cb83f4b965 |
gpu: host1x: Do not assume that a NULL domain means no DMA IOMMU
Previously with tegra-smmu, even with CONFIG_IOMMU_DMA, the default domain could have been left as NULL. The NULL domain is specially recognized by host1x_iommu_attach() as meaning it is not the DMA domain and should be replaced with the special shared domain. This happened prior to the below commit because tegra-smmu was using the NULL domain to mean IDENTITY. Now that the domain is properly labled the test in DRM doesn't see NULL. Check for IDENTITY as well to enable the special domains. This is the same issue and basic fix as seen in commit |
||
|
|
34e5ededf4 |
r8169: switch away from deprecated pcim_iomap_table
Avoid using deprecated pcim_iomap_table by switching to pcim_iomap_region. Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/a36b4cf3-c792-40fa-8164-5dc9d5f14dd0@gmail.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
eb6a0803c9 |
mips: export pci_iounmap()
I added this function in an earlier patch, but the missing export caused a build failure Reported-by: Nathan Chancellor <nathan@kernel.org> Signed-off-by: Arnd Bergmann <arnd@arndb.de> |
||
|
|
502a668fad |
ASoC: ops: Apply platform_max after deciding control type
It doesn't really make sense for the type of a control to change based on the platform_max field. platform_max allows a specific system to limit values of a control for safety but it seems reasonable the control type should remain the same between different systems, even if it is restricted down to just two values. Move the application of platform_max to after control type determination in soc_info_volsw(). Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com> Link: https://patch.msgid.link/20250319175123.3835849-4-ckeepax@opensource.cirrus.com Signed-off-by: Mark Brown <broonie@kernel.org> |
||
|
|
94dfe71f0a |
ASoC: ops: Remove some unnecessary local variables
Remove some local variables that aren't adding much in terms of clarity or space saving. Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com> Link: https://patch.msgid.link/20250319175123.3835849-3-ckeepax@opensource.cirrus.com Signed-off-by: Mark Brown <broonie@kernel.org> |
||
|
|
1e3cd64a29 |
ASoC: ops: Factor out common code from get callbacks
There are only two differences between snd_soc_get_volsw() and snd_soc_get_volsw_sx(). The maximum field is handled differently, and snd_soc_get_volsw() supports double controls with both values in the same register. Factor out the common code into a new helper and pass in the appropriate max value such that it is handled correctly for each control. Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com> Link: https://patch.msgid.link/20250319175123.3835849-2-ckeepax@opensource.cirrus.com Signed-off-by: Mark Brown <broonie@kernel.org> |
||
|
|
05b815fc50 |
Merge branch 'net-bring-back-dev_addr_sem'
Stanislav Fomichev says: ==================== net: bring back dev_addr_sem Kohei reports an issue with dev_addr_sem conversion to netdev instance lock in [0]. Based on the discussion, switching to netdev instance lock to protect the address might not work for the devices that are not using netdev ops lock. Bring dev_addr_sem instance lock back but fix the ordering. 0: https://lore.kernel.org/netdev/20250308203835.60633-2-enjuk@amazon.com ==================== Link: https://patch.msgid.link/20250312190513.1252045-1-sdf@fomichev.me Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
6dd132516f |
net: reorder dev_addr_sem lock
Lockdep complains about circular lock in 1 -> 2 -> 3 (see below).
Change the lock ordering to be:
- rtnl_lock
- dev_addr_sem
- netdev_ops (only for lower devices!)
- team_lock (or other per-upper device lock)
1. rtnl_lock -> netdev_ops -> dev_addr_sem
rtnl_setlink
rtnl_lock
do_setlink IFLA_ADDRESS on lower
netdev_ops
dev_addr_sem
2. rtnl_lock -> team_lock -> netdev_ops
rtnl_newlink
rtnl_lock
do_setlink IFLA_MASTER on lower
do_set_master
team_add_slave
team_lock
team_port_add
dev_set_mtu
netdev_ops
3. rtnl_lock -> dev_addr_sem -> team_lock
rtnl_newlink
rtnl_lock
do_setlink IFLA_ADDRESS on upper
dev_addr_sem
netif_set_mac_address
team_set_mac_address
team_lock
4. rtnl_lock -> netdev_ops -> dev_addr_sem
rtnl_lock
dev_ifsioc
dev_set_mac_address_user
__tun_chr_ioctl
rtnl_lock
dev_set_mac_address_user
tap_ioctl
rtnl_lock
dev_set_mac_address_user
dev_set_mac_address_user
netdev_lock_ops
netif_set_mac_address_user
dev_addr_sem
v2:
- move lock reorder to happen after kmalloc (Kuniyuki)
Cc: Kohei Enju <enjuk@amazon.com>
Fixes:
|
||
|
|
8033d2aef5 |
Revert "net: replace dev_addr_sem with netdev instance lock"
This reverts commit |
||
|
|
756f88ff9c |
tc-tests: Update tc police action tests for tc buffer size rounding fixes.
Before tc's recent change to fix rounding errors, several tests which specified a burst size of "1m" would translate back to being 1048574 bytes (2b less than 1Mb). sprint_size prints this as "1024Kb". With the tc fix, the burst size is instead correctly reported as 1048576 bytes (precisely 1Mb), which sprint_size prints as "1Mb". This updates the expected output in the tests' matchPattern values to accept either the old or the new output. Signed-off-by: Jonathan Lennox <jonathan.lennox@8x8.com> Link: https://patch.msgid.link/20250312174804.313107-1-jonathan.lennox@8x8.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
9730763f47 |
block: correct locking order for protecting blk-wbt parameters
The commit ' |
||
|
|
89ed5fa3b5 |
block: release q->elevator_lock in ioc_qos_write
The ioc_qos_write method acquires q->elevator_lock to protect
updates to blk-wbt parameters. Once these updates are complete,
the lock should be released before returning from ioc_qos_write.
However, in one code path, the release of q->elevator_lock was
mistakenly omitted, potentially leading to a lock leak. This commit
fixes the issue by ensuring that q->elevator_lock is properly
released in all return paths of ioc_qos_write.
Fixes:
|
||
|
|
d3c58b656c |
net: stmmac: dwmac-rk: Provide FIFO sizes for DWMAC 1000
The DWMAC 1000 DMA capabilities register does not provide actual FIFO sizes, nor does the driver really care. If they are not provided via some other means, the driver will work fine, only disallowing changing the MTU setting. Provide the FIFO sizes through the driver's platform data to enable MTU changes. The FIFO sizes are confirmed to be the same across RK3288, RK3328, RK3399 and PX30, based on their respective manuals. It is likely that Rockchip synthesized their DWMAC 1000 with the same parameters on all their chips that have it. Signed-off-by: Chen-Yu Tsai <wens@csie.org> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Link: https://patch.msgid.link/20250312163426.2178314-1-wens@kernel.org Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
0efc297a3c |
cgroup/rstat: avoid disabling irqs for O(num_cpu)
cgroup_rstat_flush_locked() grabs the irq safe cgroup_rstat_lock while iterating all possible cpus. It only drops the lock if there is scheduler or spin lock contention. If neither, then interrupts can be disabled for a long time. On large machines this can disable interrupts for a long enough time to drop network packets. On 400+ CPU machines I've seen interrupt disabled for over 40 msec. Prevent rstat from disabling interrupts while processing all possible cpus. Instead drop and reacquire cgroup_rstat_lock for each cpu. This approach was previously discussed in https://lore.kernel.org/lkml/ZBz%2FV5a7%2F6PZeM7S@slm.duckdns.org/, though this was in the context of an non-irq rstat spin lock. Benchmark this change with: 1) a single stat_reader process with 400 threads, each reading a test memcg's memory.stat repeatedly for 10 seconds. 2) 400 memory hog processes running in the test memcg and repeatedly charging memory until oom killed. Then they repeat charging and oom killing. v6.14-rc6 with CONFIG_IRQSOFF_TRACER with stat_reader and hogs, finds interrupts are disabled by rstat for 45341 usec: # => started at: _raw_spin_lock_irq # => ended at: cgroup_rstat_flush # # # _------=> CPU# # / _-----=> irqs-off/BH-disabled # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / _-=> migrate-disable # ||||| / delay # cmd pid |||||| time | caller # \ / |||||| \ | / stat_rea-96532 52d.... 0us*: _raw_spin_lock_irq stat_rea-96532 52d.... 45342us : cgroup_rstat_flush stat_rea-96532 52d.... 45342us : tracer_hardirqs_on <-cgroup_rstat_flush stat_rea-96532 52d.... 45343us : <stack trace> => memcg1_stat_format => memory_stat_format => memory_stat_show => seq_read_iter => vfs_read => ksys_read => do_syscall_64 => entry_SYSCALL_64_after_hwframe With this patch the CONFIG_IRQSOFF_TRACER doesn't find rstat to be the longest holder. The longest irqs-off holder has irqs disabled for 4142 usec, a huge reduction from previous 45341 usec rstat finding. Running stat_reader memory.stat reader for 10 seconds: - without memory hogs: 9.84M accesses => 12.7M accesses - with memory hogs: 9.46M accesses => 11.1M accesses The throughput of memory.stat access improves. The mode of memory.stat access latency after grouping by of 2 buckets: - without memory hogs: 64 usec => 16 usec - with memory hogs: 64 usec => 8 usec The memory.stat latency improves. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Greg Thelen <gthelen@google.com> Tested-by: Greg Thelen <gthelen@google.com> Acked-by: Michal Koutný <mkoutny@suse.com> Reviewed-by: Yosry Ahmed <yosry.ahmed@linux.dev> Signed-off-by: Tejun Heo <tj@kernel.org> |
||
|
|
acf10a8c0b |
selftests: drv-net: use defer in the ping test
Make sure the test cleans up after itself. The XDP off statements
at the end of the test may not be reached.
Fixes:
|
||
|
|
f749448ce9 |
Merge branch 'net-mlx5-hw-steering-cleanups'
Tariq Toukan says: ==================== net/mlx5: HW Steering cleanups This short series by Yevgeny contains several small HW Steering cleanups: - Patch 1: removing unused FW commands - Patch 2: using list_move() instead of list_del/add - Patch 3: printing the unsupported combination of match fields ==================== Link: https://patch.msgid.link/1741780194-137519-1-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
8389f2de90 |
net/mlx5: HWS, log the unsupported mask in definer
If a user requested to match on an unsupported combination of fields, print the unsupported combination in the error message. Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com> Reviewed-by: Vlad Dogaru <vdogaru@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/1741780194-137519-4-git-send-email-tariqt@nvidia.com Reviewed-by: Dawid Osuchowski <dawid.osuchowski@linux.intel.com> Reviewed-by: Michal Kubiak <michal.kubiak@intel.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
1a403ad383 |
net/mlx5: HWS, use list_move() instead of del/add
Wherever applicable, use list_move function instead of list_del + list_add. Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com> Reviewed-by: Vlad Dogaru <vdogaru@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/1741780194-137519-3-git-send-email-tariqt@nvidia.com Reviewed-by: Dawid Osuchowski <dawid.osuchowski@linux.intel.com> Reviewed-by: Michal Kubiak <michal.kubiak@intel.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
eae1389ab2 |
net/mlx5: HWS, remove unused code for alias flow tables
Alias flow tables are not in use by HWS - remove the unused code. Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com> Reviewed-by: Vlad Dogaru <vdogaru@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/1741780194-137519-2-git-send-email-tariqt@nvidia.com Reviewed-by: Dawid Osuchowski <dawid.osuchowski@linux.intel.com> Reviewed-by: Michal Kubiak <michal.kubiak@intel.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
7602eb8734 |
Merge branch 'net-stmmac-deprecate-snps-en-tx-lpi-clockgating-property'
Russell King says: ==================== net: stmmac: deprecate "snps,en-tx-lpi-clockgating" property This series deprecates the "snps,en-tx-lpi-clockgating" property for stmmac. MII Transmit clock gating, where the MAC hardware supports gating this clock, is a function of the connected PHY capabilities, which it reports through its status register. GMAC versions that support transmit clock gating twiddle the LPITCSE bit accordingly in the LPI control/status register, which is handled by the GMAC core specific code. So, "snps,en-tx-lpi-clockgating" not something that is a GMAC property, but is a work-around for phylib not providing an interface to determine whether the PHY allows the transmit clock to be disabled. This series converts the two SoCs that make use of this property (which, I hasten to add, is set in the SoC code) to use the PHY capability bit instead of a DT property, then removes the DT property from the .dtsi, deprecates it in the snps,dwmac binding, and finally in the stmmac code. I am expecting some discussion on how to merge this, as I think the order in which these changes is made is important - we don't want to deprecate the old way until the new code has landed. ==================== Link: https://patch.msgid.link/Z9FVHEf3uUqtKzyt@shell.armlinux.org.uk Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
cf0a96de39 |
net: stmmac: deprecate "snps,en-tx-lpi-clockgating" property
Whether the MII transmit clock can be stopped is primarily a property of the PHY (there is a capability bit that should be checked first.) Whether the MAC is capable of stopping the transmit clock is a separate issue, but this is already handled by the core DesignWare MAC code. Therefore, snps,en-tx-lpi-clockgating is technically incorrect, and this commit adds a warning should a DT be encountered with the property present. However, we keep backwards compatibility. Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> Link: https://patch.msgid.link/E1tsIUK-005vGk-H7@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
a62b7901d3 |
dt-bindings: deprecate "snps,en-tx-lpi-clockgating" property
Whether the MII transmit clock can be stopped is primarily a property of the PHY (there is a capability bit that should be checked first.) Whether the MAC is capable of stopping the transmit clock is a separate issue, but this is already handled by the core DesignWare MAC code. Therefore, snps,en-tx-lpi-clockgating is technically incorrect, so this commit deprecates the property in the binding. Acked-by: Rob Herring (Arm) <robh@kernel.org> Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://patch.msgid.link/E1tsIUF-005vGd-C5@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
50a84bbc7e |
ARM: dts: stm32: remove "snps,en-tx-lpi-clockgating" property
Whether the MII transmit clock can be stopped is primarily a property of the PHY (there is a capability bit that should be checked first.) Whether the MAC is capable of stopping the transmit clock is a separate issue, but this is already handled by the core DesignWare MAC code. As commit "net: stmmac: stm32: use PHY capability for TX clock stop" adds the flag to use the PHY capability, remove the DT property that is now unecessary. Cc: Samin Guo <samin.guo@starfivetech.com> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://patch.msgid.link/E1tsIUA-005vGX-8A@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
637af286f9 |
riscv: dts: starfive: remove "snps,en-tx-lpi-clockgating" property
Whether the MII transmit clock can be stopped is primarily a property of the PHY (there is a capability bit that should be checked first.) Whether the MAC is capable of stopping the transmit clock is a separate issue, but this is already handled by the core DesignWare MAC code. As commit "net: stmmac: starfive: use PHY capability for TX clock stop" adds the flag to use the PHY capability, remove the DT property that is now unecessary. Cc: Samin Guo <samin.guo@starfivetech.com> Acked-by: Conor Dooley <conor.dooley@microchip.com> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://patch.msgid.link/E1tsIU5-005vGR-4c@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
a5bc19e2ab |
net: stmmac: stm32: use PHY capability for TX clock stop
Whether the MII transmit clock can be stopped is primarily a property of the PHY (there is a capability bit that should be checked first.) Whether the MAC is capable of stopping the transmit clock is a separate issue, but this is already handled by the core DesignWare MAC code. Add the flag to allow the stmmac core to use the PHY capability. Cc: Christophe Roullier <christophe.roullier@st.com> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://patch.msgid.link/E1tsIU0-005vGL-17@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
5f250bd72a |
net: stmmac: starfive: use PHY capability for TX clock stop
Whether the MII transmit clock can be stopped is primarily a property of the PHY (there is a capability bit that should be checked first.) Whether the MAC is capable of stopping the transmit clock is a separate issue, but this is already handled by the core DesignWare MAC code. Add the flag to allow the stmmac core to use the PHY capability. Cc: Samin Guo <samin.guo@starfivetech.com> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://patch.msgid.link/E1tsITu-005vGF-TM@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
0c1f1eb654 |
net: stmmac: allow platforms to use PHY tx clock stop capability
Allow platform glue to instruct stmmac to make use of the PHY transmit clock stop capability when deciding whether to allow the transmit clock from the DWMAC core to be stopped. Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Link: https://patch.msgid.link/E1tsITp-005vG9-Px@rmk-PC.armlinux.org.uk Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
4df2ebfca6 |
Merge tag 'ieee802154-for-net-next-2025-03-10' of git://git.kernel.org/pub/scm/linux/kernel/git/wpan/wpan-next
Stefan Schmidt says: ==================== pull-request: ieee802154-next 2025-03-10 An update from ieee802154 for your *net-next* tree: Andy Shevchenko reworked the ca8210 driver to use the gpiod API and fixed a few problems of the driver along the way. * tag 'ieee802154-for-net-next-2025-03-10' of git://git.kernel.org/pub/scm/linux/kernel/git/wpan/wpan-next: dt-bindings: ieee802154: ca8210: Update polarity of the reset pin ieee802154: ca8210: Switch to using gpiod API ieee802154: ca8210: Get platform data via dev_get_platdata() ieee802154: ca8210: Use proper setters and getters for bitwise types ==================== Link: https://patch.msgid.link/20250310185752.2683890-1-stefan@datenfreihafen.org Signed-off-by: Paolo Abeni <pabeni@redhat.com> |
||
|
|
565d065acd |
ata: ahci: simplify init function
This patch moves all the IRQ vector allocations into a single function. Instead of having the allocations spread out over two separate call sites everything will be handled in ahci_init_irq. Also a direct call into pci(m)_intx will be removed. The main part of this change is done by adding a PCI_IRQ_INTX flag into an already existing pci_alloc_irq_vectors invocation. In the current implementation of the pci_alloc_irq_vectors is the sequence of calls msi-x -> msi -> legacy irq and whatever there succeeds stops the call chain. That makes it impossible to merge all instances into as a single call to pci_alloc_irq_vectors since the order of calls there is: multiple msi-x a single msi a single msi-x a legacy irq. The two last steps can be merged into a single one which are the msi-x and legacy irq option. When PCI_IRQ_INTX flag is set the pci_alloc_irq_vectors succeeds in almost all cases - that makes it possible to convert ahci_init_irq(msi) into a void function. The exception is when dev->irq is zero then the pci_alloc_irq_vectors may return with an error code also pci_intx isn't called from pci_alloc_irq_vectors and thus certain pci calls aren't performed. That's just a negligible issue as later in ahci_init_one the (zero) value of dev->irq is via pci_irq_vector assigned to hpriv->irq. That value is then later tested in ahci_host_activate->ata_host_activate where it is welcomed with a WARN_ON message and fails with setting up irq and then the probe function (ahci_init_one) fails. The special zero value's meaning is that polling mode is being be set up which isn't the case. No functional change. Signed-off-by: Tomas Henzl <thenzl@redhat.com> Link: https://lore.kernel.org/r/20250319155030.16410-1-thenzl@redhat.com Signed-off-by: Niklas Cassel <cassel@kernel.org> |
||
|
|
5f14404bfa |
io_uring/cmd: don't expose entire cmd async data
io_uring needs private bits in cmd's ->async_data, and they should never be exposed to drivers as it'd certainly be abused. Leave struct io_uring_cmd_data for the drivers but wrap it into a structure. It's a prep patch and doesn't do anything useful yet. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/20250319061251.21452-3-sidong.yang@furiosa.ai Signed-off-by: Jens Axboe <axboe@kernel.dk> |
||
|
|
575e7b0629 |
io_uring: rename the data cmd cache
Pick a more descriptive name for the cmd async data cache. Signed-off-by: Pavel Begunkov <asml.silence@gmail.com> Link: https://lore.kernel.org/r/20250319061251.21452-2-sidong.yang@furiosa.ai Signed-off-by: Jens Axboe <axboe@kernel.dk> |
||
|
|
6ffb9017e9 |
Merge branch 'resilient-queued-spin-lock'
Kumar Kartikeya Dwivedi says: ==================== Resilient Queued Spin Lock Changelog: ---------- v3 -> v4 v4: https://lore.kernel.org/bpf/20250303152305.3195648-1-memxor@gmail.com * Fix bisectability problem by reordering locktorture commit before Makefile commit. * Add EXPORT_SYMBOL_GPL to all used symbols and variables by consumers. * Skip BPF selftest when nrprocs < 2. * Fix kdoc to describe return value for res_spin_lock, slowpath. * Move kernel/locking/rqspinlock.{c,h} to kernel/bpf/rqspinlock.{c,h}. v2 -> v3 v2: https://lore.kernel.org/bpf/20250206105435.2159977-1-memxor@gmail.com * Add ifdef's to fallback to Ankur's patch when it gets in, until then copy-paste the implementation. * Change the meaning of RES_DEF_TIMEOUT from two critical section lengths to one for clarity, and use RES_DEF_TIMEOUT * 2 where needed. * Use NSEC_PER_SEC as timeout for TAS fallback. * Add Closes: tags for known syzbot reports. * Change timeout for TAS fallback to 1 second. * Fix more kernel test robot errors. * More comments about smp_wmb in release_held_lock_entry interaction. * Change RES_NR_HELD to 31. * Address comments from Peter, Eduard, Alexei. v1 -> v2 v1: https://lore.kernel.org/bpf/20250107140004.2732830-1-memxor@gmail.com * Address nits from Waiman and Peter * Fix arm64 WFE bug pointed out by Peter. * Fix incorrect memory ordering in release_held_lock_entry, and document subtleties. Explain why release is sufficient in unlock but not in release_held_lock_entry. * Remove dependence on CONFIG_QUEUED_SPINLOCKS and introduce a test-and-set fallback when queued spinlock support is missing on an architecture. * Enforce FIFO ordering for BPF program spin unlocks. * Address comments from Eduard on verifier plumbing. * Add comments as suggested by Waiman. * Refactor paravirt TAS lock to use the implemented TAS fallback. * Use rqspinlock_t as the type throughout so that it can be replaced with a non-qspinlock type in case of fallback. * Testing and benchmarking on arm64, added numbers to the cover letter. * Fix kernel test robot errors. * Fix a BPF selftest bug leading to spurious failures on arm64. Introduction ------------ This patch set introduces Resilient Queued Spin Lock (or rqspinlock with res_spin_lock() and res_spin_unlock() APIs). This is a qspinlock variant which recovers the kernel from a stalled state when the lock acquisition path cannot make forward progress. This can occur when a lock acquisition attempt enters a deadlock situation (e.g. AA, or ABBA), or more generally, when the owner of the lock (which we’re trying to acquire) isn’t making forward progress. The cover letter provides an overview of the motivation, design, and alternative approaches. We then provide evaluation numbers showcasing that while rqspinlock incurs overhead, the performance of rqspinlock approaches that of the normal qspinlock used by the kernel. The evaluations for rqspinlock were performed by replacing the default qspinlock implementation with it and booting the kernel to run the experiments. Support for locktorture is also included with numbers in this series. The cover letter's design section provides an overview of the algorithmic approach. A technical document describing the implementation in more detail is available here: https://github.com/kkdwivedi/rqspinlock/blob/main/rqspinlock.pdf We have a WIP TLA+ proof for liveness and mutual exclusion of rqspinlock built on top of the qspinlock TLA+ proof from Catalin Marinas [3]. We will share more details and the links in the near future. Motivation ---------- In regular kernel code, usage of locks is assumed to be correct, so as to avoid deadlocks and stalls by construction, however, the same is not true for BPF programs. Users write normal C code and the in-kernel eBPF runtime ensures the safety of the kernel by rejecting unsafe programs. Users can upload programs that use locks in an improper fashion, and may cause deadlocks when these programs run inside the kernel. The verifier is responsible for rejecting such programs from being loaded into the kernel. Until now, the eBPF verifier ensured deadlock safety by only permitting one lock acquisition at a time, and by preventing any functions to be called from within the critical section. Additionally, only a few restricted program types are allowed to call spin locks. As the usage of eBPF grows (e.g. with sched_ext) beyond its conventional application in networking, tracing, and security, the limitations on locking are becoming a bottleneck for users. The rqspinlock implementation allows us to permit more flexible locking patterns in BPF programs, without limiting them to the subset that can be proven safe statically (which is fairly small, and requires complex static analysis), while ensuring that the kernel will recover in case we encounter a locking violation at runtime. We make a tradeoff here by accepting programs that may potentially have deadlocks, and recover the kernel quickly at runtime to ensure availability. Additionally, eBPF programs attached to different parts of the kernel can introduce new control flow into the kernel, which increases the likelihood of deadlocks in code not written to handle reentrancy. There have been multiple syzbot reports surfacing deadlocks in internal kernel code due to the diverse ways in which eBPF programs can be attached to different parts of the kernel. By switching the BPF subsystem’s lock usage to rqspinlock, all of these issues can be mitigated at runtime. This spin lock implementation allows BPF maps to become safer and remove mechanisms that have fallen short in assuring safety when nesting programs in arbitrary ways in the same context or across different contexts. The red diffs due to patches 16-18 demonstrate this simplification. > kernel/bpf/hashtab.c | 102 ++++++++++++++++++++++++++++++++--------------------------... > kernel/bpf/lpm_trie.c | 25 ++++++++++++++----------- > kernel/bpf/percpu_freelist.c | 113 +++++++++++++++++++++++++---------------------------------... > kernel/bpf/percpu_freelist.h | 4 ++-- > 4 files changed, 73 insertions(+), 171 deletions(-) Design ------ Deadlocks mostly manifest as stalls in the waiting loops of the qspinlock slow path. Thus, using stalls as a signal for deadlocks avoids introducing cost to the normal fast path, and ensures bounded termination of the waiting loop. Our recovery algorithm is focused on terminating the waiting loops of the qspinlock algorithm when it gets stuck, and implementing bespoke recovery procedures for each class of waiter to restore the lock to a usable state. Deadlock detection is the main mechanism used to provide faster recovery, with the timeout mechanism acting as a final line of defense. Deadlock Detection ~~~~~~~~~~~~~~~~~~ We handle two cases of deadlocks: AA deadlocks (attempts to acquire the same lock again), and ABBA deadlocks (attempts to acquire two locks in the opposite order from two distinct threads). Variants of ABBA deadlocks may be encountered with more than two locks being held in the incorrect order. These are not diagnosed explicitly, as they reduce to ABBA deadlocks. Deadlock detection is triggered immediately when beginning the waiting loop of a lock slow path. While timeouts ensure that any waiting loops in the locking slow path terminate and return to the caller, it can be excessively long in some situations. While the default timeout is short (0.5s), a stall for this duration inside the kernel can set off alerts for latency-critical services with strict SLOs. Ideally, the kernel should recover from an undesired state of the lock as soon as possible. A multi-step strategy is used to recover the kernel from waiting loops in the locking algorithm which may fail to terminate in a bounded amount of time. * Each CPU maintains a table of held locks. Entries are inserted and removed upon entry into lock, and exit from unlock, respectively. * Deadlock detection for AA locks is thus simple: we have an AA deadlock if we find a held lock entry for the lock we’re attempting to acquire on the same CPU. * During deadlock detection for ABBA, we search through the tables of all other CPUs to find situations where we are holding a lock the remote CPU is attempting to acquire, and they are holding a lock we are attempting to acquire. Upon encountering such a condition, we report an ABBA deadlock. * We divide the duration between entry time point into the waiting loop and the timeout time point into intervals of 1 ms, and perform deadlock detection until timeout happens. Upon entry into the slow path, and then completion of each 1 ms interval, we perform detection of both AA and ABBA deadlocks. In the event that deadlock detection yields a positive result, the recovery happens sooner than the timeout. Otherwise, it happens as a last resort upon completion of the timeout. Timeouts ~~~~~~~~ Timeouts act as final line of defense against stalls for waiting loops. The ‘ktime_get_mono_fast_ns’ function is used to poll for the current time, and it is compared to the timestamp indicating the end time in the waiter loop. Each waiting loop is instrumented to check an extra condition using a macro. Internally, the macro implementation amortizes the checking of the timeout to avoid sampling the clock in every iteration. Precisely, the timeout checks are invoked every 64k iterations. Recovery ~~~~~~~~ There is extensive literature in academia on designing locks that support timeouts [0][1], as timeouts can be used as a proxy for detecting the presence of deadlocks and recovering from them, without maintaining explicit metadata to construct a waits-for relationship between two threads at runtime. In case of rqspinlock, the key simplification in our algorithm comes from the fact that upon a timeout, waiters always leave the queue in FIFO order. As such, the timeout is only enforced by the head of the wait queue, while other waiters rely on the head to signal them when a timeout has occurred and when they need to exit. We don’t have to implement complex algorithms and do not need extra synchronization for waiters in the middle of the queue timing out before their predecessor or successor, unlike previous approaches [0][1]. There are three forms of waiters in the original queued spin lock algorithm. The first is the waiter which acquires the pending bit and spins on the lock word without forming a wait queue. The second is the head waiter that is the first waiter heading the wait queue. The third form is of all the non-head waiters queued behind the head, waiting to be signalled through their MCS node to overtake the responsibility of the head. In rqspinlock's recovery algorithm, we are concerned with the second and third kind. First, we augment the waiting loop of the head of the wait queue with a timeout. When this timeout happens, all waiters part of the wait queue will abort their lock acquisition attempts. This happens in three steps. * First, the head breaks out of its loop waiting for pending and locked bits to turn to 0, and non-head waiters break out of their MCS node spin (more on that later). * Next, every waiter (head or non-head) attempts to check whether they are also the tail waiter, in such a case they attempt to zero out the tail word and allow a new queue to be built up for this lock. If they succeed, they have no one to signal next in the queue to stop spinning. * Otherwise, they signal the MCS node of the next waiter to break out of its spin and try resetting the tail word back to 0. This goes on until the tail waiter is found. In case of races, the new tail will be responsible for performing the same task, as the old tail will then fail to reset the tail word and wait for its next pointer to be updated before it signals the new tail to do the same. Timeout Bound ~~~~~~~~~~~~~ The timeout is applied by two types of waiters: the pending bit waiter and the wait queue head waiter. As such, for the pending waiter, only the lock owner is ahead of it, and for the wait queue head waiter, only the lock owner and the pending waiter take precedence in executing their critical sections. We define the timeout value to span at most 1 critical section length, and then use the appropriate value (default, or default x 2) depending on if we are the pending waiter or head of wait queue. Therefore, the waiting loop wait can span at most 2 critical section lengths, and thus, it is unaffected by the amount of contention or the number of CPUs on the host. Non-head waiters simply wait for the wait queue head to signal them on a timeout. In Meta's production, we have noticed uncore PMU reads and SMIs consuming tens of msecs. While these events are rare, a 0.25 second timeout should absorb such tail events and not raise false alarms for timeouts. We will continue monitoring this in production and adjust the timeout if necessary in the future. More details of the recovery algorithm is described in patch 9 and a detailed description is available at [2]. Alternatives ------------ Lockdep: We do not rely on the lockdep facility for reporting violations for primarily two reasons: * Overhead: The lockdep infrastructure can add significant overhead to the lock acquisition path, and is not recommended for use in production due to this reason. While the report is more useful and exhaustive, the overhead can be prohibitive, especially as BPF programs run in hot paths of the kernel. Moreover, it also increases the size of the lock word to store extra metadata, which is not feasible for BPF spin locks that are 4-bytes in size today (similar to qspinlock). * Debug Tool: Lockdep is intended to be used as a debugging facility, providing extra context to the user about the locking violations occurring during runtime. It is always turned off on all production kernels, therefore isn’t available most of the time. We require a mechanism for detecting common variants of deadlocks that is always available in production kernels and never turned off. At the same time, it must not introduce overhead in terms of time (for the slow path) and memory (for the lock word size). Evaluation ---------- We run benchmarks that stress locking scalability and perform comparison against the baseline (qspinlock). For the rqspinlock case, we replace the default qspinlock with it in the kernel, such that all spin locks in the kernel use the rqspinlock slow path. As such, benchmarks that stress kernel spin locks end up exercising rqspinlock. Evaluation setup ~~~~~~~~~~~~~~~~ We set the CPU governor to performance for all experiments. Note: Numbers for arm64 have been obtained without the no-WFE fallback in this series, to perform a fair comparison with the WFE using qspinlock baseline. x86_64: Intel Xeon Platinum 8468 (Sapphire Rapids) 96 cores (48 x 2 sockets) 2 threads per core, 0-95, siblings from 96-191 2 NUMA nodes (every 48 cores), 2 LLCs (every 48 cores), 1 LLC per NUMA node Hyperthreading enabled arm64: Ampere Max Neoverse-N1 256-Core Processor 256 cores (128 cores x 2 sockets) 1 thread per core 2 NUMA nodes (every 128 cores), 1 L2 per core (256 instances), no shared L3 No hyperthreading available The locktorture experiment is run for 30 seconds. Average of 25 runs is used for will-it-scale, after an initial warm up. More information on the locks contended in the will-it-scale experiments is available in the evaluation section of the CNA paper, in table 1 [4]. Legend: QL - qspinlock (avg. throughput) RQL - rqspinlock (avg. throughput) Results ~~~~~~~ locktorture - x86_64 Threads QL RQL Speedup ----------------------------------------------- 1 46910437 45057327 0.96 2 29871063 25085034 0.84 4 |
||
|
|
60ba5b3ed7 |
selftests/bpf: Add tests for rqspinlock
Introduce selftests that trigger AA, ABBA deadlocks, and test the edge case where the held locks table runs out of entries, since we then fallback to the timeout as the final line of defense. Also exercise verifier's AA detection where applicable. Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20250316040541.108729-26-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> |
||
|
|
ea21771c07 |
bpf: Maintain FIFO property for rqspinlock unlock
Since out-of-order unlocks are unsupported for rqspinlock, and irqsave variants enforce strict FIFO ordering anyway, make the same change for normal non-irqsave variants, such that FIFO ordering is enforced. Two new verifier state fields (active_lock_id, active_lock_ptr) are used to denote the top of the stack, and prev_id and prev_ptr are ascertained whenever popping the topmost entry through an unlock. Take special care to make these fields part of the state comparison in refsafe. Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20250316040541.108729-25-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> |
||
|
|
0de2046137 |
bpf: Implement verifier support for rqspinlock
Introduce verifier-side support for rqspinlock kfuncs. The first step is allowing bpf_res_spin_lock type to be defined in map values and allocated objects, so BTF-side is updated with a new BPF_RES_SPIN_LOCK field to recognize and validate. Any object cannot have both bpf_spin_lock and bpf_res_spin_lock, only one of them (and at most one of them per-object, like before) must be present. The bpf_res_spin_lock can also be used to protect objects that require lock protection for their kfuncs, like BPF rbtree and linked list. The verifier plumbing to simulate success and failure cases when calling the kfuncs is done by pushing a new verifier state to the verifier state stack which will verify the failure case upon calling the kfunc. The path where success is indicated creates all lock reference state and IRQ state (if necessary for irqsave variants). In the case of failure, the state clears the registers r0-r5, sets the return value, and skips kfunc processing, proceeding to the next instruction. When marking the return value for success case, the value is marked as 0, and for the failure case as [-MAX_ERRNO, -1]. Then, in the program, whenever user checks the return value as 'if (ret)' or 'if (ret < 0)' the verifier never traverses such branches for success cases, and would be aware that the lock is not held in such cases. We push the kfunc state in check_kfunc_call whenever rqspinlock kfuncs are invoked. We introduce a kfunc_class state to avoid mixing lock irqrestore kfuncs with IRQ state created by bpf_local_irq_save. With all this infrastructure, these kfuncs become usable in programs while satisfying all safety properties required by the kernel. Acked-by: Eduard Zingerman <eddyz87@gmail.com> Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20250316040541.108729-24-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> |
||
|
|
97eb35f3ad |
bpf: Introduce rqspinlock kfuncs
Introduce four new kfuncs, bpf_res_spin_lock, and bpf_res_spin_unlock,
and their irqsave/irqrestore variants, which wrap the rqspinlock APIs.
bpf_res_spin_lock returns a conditional result, depending on whether the
lock was acquired (NULL is returned when lock acquisition succeeds,
non-NULL upon failure). The memory pointed to by the returned pointer
upon failure can be dereferenced after the NULL check to obtain the
error code.
Instead of using the old bpf_spin_lock type, introduce a new type with
the same layout, and the same alignment, but a different name to avoid
type confusion.
Preemption is disabled upon successful lock acquisition, however IRQs
are not. Special kfuncs can be introduced later to allow disabling IRQs
when taking a spin lock. Resilient locks are safe against AA deadlocks,
hence not disabling IRQs currently does not allow violation of kernel
safety.
__irq_flag annotation is used to accept IRQ flags for the IRQ-variants,
with the same semantics as existing bpf_local_irq_{save, restore}.
These kfuncs will require additional verifier-side support in subsequent
commits, to allow programs to hold multiple locks at the same time.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20250316040541.108729-23-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
||
|
|
47979314c0 |
bpf: Convert lpm_trie.c to rqspinlock
Convert all LPM trie usage of raw_spinlock to rqspinlock. Note that rcu_dereference_protected in trie_delete_elem is switched over to plain rcu_dereference, the RCU read lock should be held from BPF program side or eBPF syscall path, and the trie->lock is just acquired before the dereference. It is not clear the reason the protected variant was used from the commit history, but the above reasoning makes sense so switch over. Closes: https://lore.kernel.org/lkml/000000000000adb08b061413919e@google.com Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20250316040541.108729-22-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> |
||
|
|
f2ac0e5d1c |
bpf: Convert percpu_freelist.c to rqspinlock
Convert the percpu_freelist.c code to use rqspinlock, and remove the extralist fallback and trylock-based acquisitions to avoid deadlocks. Key thing to note is the retained while (true) loop to search through other CPUs when failing to push a node due to locking errors. This retains the behavior of the old code, where it would keep trying until it would be able to successfully push the node back into the freelist of a CPU. Technically, we should start iteration for this loop from raw_smp_processor_id() + 1, but to avoid hitting the edge of nr_cpus, we skip execution in the loop body instead. Closes: https://lore.kernel.org/bpf/CAPPBnEa1_pZ6W24+WwtcNFvTUHTHO7KUmzEbOcMqxp+m2o15qQ@mail.gmail.com Closes: https://lore.kernel.org/bpf/CAPPBnEYm+9zduStsZaDnq93q1jPLqO-PiKX9jy0MuL8LCXmCrQ@mail.gmail.com Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20250316040541.108729-21-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> |
||
|
|
4fa8d68aa5 |
bpf: Convert hashtab.c to rqspinlock
Convert hashtab.c from raw_spinlock to rqspinlock, and drop the hashed per-cpu counter crud from the code base which is no longer necessary. Closes: https://lore.kernel.org/bpf/675302fd.050a0220.2477f.0004.GAE@google.com Closes: https://lore.kernel.org/bpf/000000000000b3e63e061eed3f6b@google.com Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20250316040541.108729-20-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> |
||
|
|
a6884f6f1d |
rqspinlock: Add locktorture support
Introduce locktorture support for rqspinlock using the newly added macros as the first in-kernel user and consumer. Guard the code with CONFIG_BPF_SYSCALL ifdef since rqspinlock is not available otherwise. Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/r/20250316040541.108729-19-memxor@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> |