| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Pull NFS client fixes from Anna Schumaker:
- Fix /proc/PID/io read_bytes accounting
- Fix setting NLM file_lock start and end during decoding testargs
- Fix timing for setting access cache timestamps
* tag 'nfs-for-6.3-2' of git://git.linux-nfs.org/projects/anna/linux-nfs:
NFS: Correct timing for assigning access cache timestamp
lockd: set file_lock start and end when decoding nlm4 testargs
NFS: Fix /proc/PID/io read_bytes for buffered reads
|
|
This reverts commit ee892ea83d99610fa33bea612de058e0955eec3a.
It was accidentally picked up for backporting. Revert.
Cc: Jani Nikula <[email protected]>
Cc: Rodrigo Vivi <[email protected]>
Signed-off-by: Ashutosh Dixit <[email protected]>
Signed-off-by: Jani Nikula <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
|
|
Geoff Levand says:
====================
net/ps3_gelic_net: DMA related fixes
v9: Make rx_skb_size local to gelic_descr_prepare_rx.
v8: Add more cpu_to_be32 calls.
v7: Remove all cleanups, sync to spider net.
v6: Reworked and cleaned up patches.
v5: Some additional patch cleanups.
v4: More patch cleanups.
v3: Cleaned up patches as requested.
====================
Signed-off-by: David S. Miller <[email protected]>
|
|
The current Gelic Etherenet driver was checking the return value of its
dma_map_single call, and not using the dma_mapping_error() routine.
Fixes runtime problems like these:
DMA-API: ps3_gelic_driver sb_05: device driver failed to check map error
WARNING: CPU: 0 PID: 0 at kernel/dma/debug.c:1027 .check_unmap+0x888/0x8dc
Fixes: 02c1889166b4 ("ps3: gigabit ethernet driver for PS3, take3")
Reviewed-by: Alexander Duyck <[email protected]>
Signed-off-by: Geoff Levand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
The Gelic Ethernet device needs to have the RX sk_buffs aligned to
GELIC_NET_RXBUF_ALIGN, and also the length of the RX sk_buffs must
be a multiple of GELIC_NET_RXBUF_ALIGN.
The current Gelic Ethernet driver was not allocating sk_buffs large
enough to allow for this alignment.
Also, correct the maximum and minimum MTU sizes, and add a new
preprocessor macro for the maximum frame size, GELIC_NET_MAX_FRAME.
Fixes various randomly occurring runtime network errors.
Fixes: 02c1889166b4 ("ps3: gigabit ethernet driver for PS3, take3")
Signed-off-by: Geoff Levand <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
clang with W=1 reports
drivers/net/usb/plusb.c:65:1: error:
unused function 'pl_clear_QuickLink_features' [-Werror,-Wunused-function]
pl_clear_QuickLink_features(struct usbnet *dev, int val)
^
This static function is not used, so remove it.
Signed-off-by: Tom Rix <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Packet length retrieved from descriptor may be larger than
the actual socket buffer length. In such case the cloned
skb passed up the network stack will leak kernel memory contents.
Additionally prevent integer underflow when size is less than
ETH_FCS_LEN.
Fixes: 55d7de9de6c3 ("Microchip's LAN7800 family USB 2/3 to 10/100/1000 Ethernet device driver")
Signed-off-by: Szymon Heidrich <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
In emac_probe, &adpt->work_thread is bound with
emac_work_thread. Then it will be started by timeout
handler emac_tx_timeout or a IRQ handler emac_isr.
If we remove the driver which will call emac_remove
to make cleanup, there may be a unfinished work.
The possible sequence is as follows:
Fix it by finishing the work before cleanup in the emac_remove
and disable timeout response.
CPU0 CPU1
|emac_work_thread
emac_remove |
free_netdev |
kfree(netdev); |
|emac_reinit_locked
|emac_mac_down
|//use netdev
Fixes: b9b17debc69d ("net: emac: emac gigabit ethernet controller driver")
Signed-off-by: Zheng Wang <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
We collect the software statistics counters for RX bytes (reported to
/proc/net/dev and to ethtool -S $dev | grep 'rx_bytes: ") at a time when
skb->len has already been adjusted by the eth_type_trans() ->
skb_pull_inline(skb, ETH_HLEN) call to exclude the L2 header.
This means that when connecting 2 DSA interfaces back to back and
sending 1 packet with length 100, the sending interface will report
tx_bytes as incrementing by 100, and the receiving interface will report
rx_bytes as incrementing by 86.
Since accounting for that in scripts is quirky and is something that
would be DSA-specific behavior (requiring users to know that they are
running on a DSA interface in the first place), the proposal is that we
treat it as a bug and fix it.
This design bug has always existed in DSA, according to my analysis:
commit 91da11f870f0 ("net: Distributed Switch Architecture protocol
support") also updates skb->dev->stats.rx_bytes += skb->len after the
eth_type_trans() call. Technically, prior to Florian's commit
a86d8becc3f0 ("net: dsa: Factor bottom tag receive functions"), each and
every vendor-specific tagging protocol driver open-coded the same bug,
until the buggy code was consolidated into something resembling what can
be seen now. So each and every driver should have its own Fixes: tag,
because of their different histories until the convergence point.
I'm not going to do that, for the sake of simplicity, but just blame the
oldest appearance of buggy code.
There are 2 ways to fix the problem. One is the obvious way, and the
other is how I ended up doing it. Obvious would have been to move
dev_sw_netstats_rx_add() one line above eth_type_trans(), and below
skb_push(skb, ETH_HLEN). But DSA processing is not as simple as that.
We count the bytes after removing everything DSA-related from the
packet, to emulate what the packet's length was, on the wire, when the
user port received it.
When eth_type_trans() executes, dsa_untag_bridge_pvid() has not run yet,
so in case the switch driver requests this behavior - commit
412a1526d067 ("net: dsa: untag the bridge pvid from rx skbs") has the
details - the obvious variant of the fix wouldn't have worked, because
the positioning there would have also counted the not-yet-stripped VLAN
header length, something which is absent from the packet as seen on the
wire (there it may be untagged, whereas software will see it as
PVID-tagged).
Fixes: f613ed665bb3 ("net: dsa: Add support for 64-bit statistics")
Signed-off-by: Vladimir Oltean <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
When we change the M/N values seamlessly during a fastset we should
also update the vblank timestamping stuff to make sure the vblank
timestamp corrections/guesstimations come out exact.
Note that only crtc_clock and framedur_ns can actually end up
changing here during fastsets. Everything else we touch can
only change during full modesets.
Technically we should try to do this exactly at the start of
vblank, but that would require some kind of double buffering
scheme. Let's skip that for now and just update things right
after the commit has been submitted to the hardware. This
means the information will be properly up to date when the
vblank irq handler goes to work. Only if someone ends up
querying some vblanky stuff in between the commit and start
of vblank may we see a slight discrepancy.
Also this same problem really exists for the DRRS downclocking
stuff. But as that is supposed to be more or less transparent
to the user, and it only drops to low gear after a long delay
(1 sec currently) we probably don't have to worry about it.
Any time something is actively submitting updates DRRS will
remain in high gear and so the timestamping constants will
match the hardware state.
Reviewed-by: Jani Nikula <[email protected]>
Reviewed-by: Mitul Golani <[email protected]>
Fixes: e6f29923c048 ("drm/i915: Allow M/N change during fastset on bdw+")
Signed-off-by: Ville Syrjälä <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit 8cb1f95cca68421b08333175719fdd3615372ca8)
Signed-off-by: Jani Nikula <[email protected]>
|
|
Use hex format so that it is easier to decode.
Fixes: fe5979665f64 ("drm/i915/debugfs: Add perf_limit_reasons in debugfs")
Signed-off-by: Vinay Belgaumkar <[email protected]>
Reviewed-by: Ashutosh Dixit <[email protected]>
Signed-off-by: John Harrison <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit 5e008ba67cb80084e99b40ccd46f9029ae421632)
Signed-off-by: Jani Nikula <[email protected]>
|
|
Probe pseudo errors should be injected only in places where real errors
can be encountered, otherwise unwinding code can be broken.
Placing intel_uc_init_late before i915_inject_probe_error violated
this rule, resulting in following bug:
__intel_gt_disable:655 GEM_BUG_ON(intel_gt_pm_is_awake(gt))
Fixes: 481d458caede ("drm/i915/guc: Add golden context to GuC ADS")
Acked-by: Nirmoy Das <[email protected]>
Reviewed-by: Andi Shyti <[email protected]>
Signed-off-by: Andrzej Hajda <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit c4252a11131c7f27a158294241466e2a4e7ff94e)
Signed-off-by: Jani Nikula <[email protected]>
|
|
debug_active_activate() expected ref->count to be zero
which is not true anymore as __i915_active_activate() calls
debug_active_activate() after incrementing the count.
v2: No need to check for "ref->count == 1" as __i915_active_activate()
already make sure of that(Janusz).
References: https://gitlab.freedesktop.org/drm/intel/-/issues/6733
Fixes: 04240e30ed06 ("drm/i915: Skip taking acquire mutex for no ref->active callback")
Cc: Chris Wilson <[email protected]>
Cc: Tvrtko Ursulin <[email protected]>
Cc: Thomas Hellström <[email protected]>
Cc: Andi Shyti <[email protected]>
Cc: [email protected]
Cc: Janusz Krzysztofik <[email protected]>
Cc: <[email protected]> # v5.10+
Signed-off-by: Nirmoy Das <[email protected]>
Reviewed-by: Janusz Krzysztofik <[email protected]>
Reviewed-by: Andrzej Hajda <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit bfad380c542438a9b642f8190b7fd37bc77e2723)
Signed-off-by: Jani Nikula <[email protected]>
|
|
Error captures are tagged with an 'ecode'. This is a pseduo-unique magic
number that is meant to distinguish similar seeming bugs with
different underlying signatures. It is a combination of two ring state
registers. Unfortunately, the register state being used is only valid
in execlist mode. In GuC mode, the register state exists in a separate
list of arbitrary register address/value pairs rather than the named
entry structure. So, search through that list to find the two exciting
registers and copy them over to the structure's named members.
v2: if else if instead of if if (Alan)
Signed-off-by: John Harrison <[email protected]>
Reviewed-by: Alan Previn <[email protected]>
Fixes: a6f0f9cf330a ("drm/i915/guc: Plumb GuC-capture into gpu_coredump")
Cc: Alan Previn <[email protected]>
Cc: Umesh Nerlige Ramappa <[email protected]>
Cc: Lucas De Marchi <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Cc: Rodrigo Vivi <[email protected]>
Cc: Tvrtko Ursulin <[email protected]>
Cc: Matt Roper <[email protected]>
Cc: Aravind Iddamsetty <[email protected]>
Cc: Michael Cheng <[email protected]>
Cc: Matthew Brost <[email protected]>
Cc: Bruce Chang <[email protected]>
Cc: Daniele Ceraolo Spurio <[email protected]>
Cc: Matthew Auld <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit 9724ecdbb9ddd6da3260e4a442574b90fc75188a)
Signed-off-by: Jani Nikula <[email protected]>
|
|
The Wa_14017073508 require to send Media Busy/Idle mailbox while
accessing Media tile. As of now it is getting handled while __gt_unpark,
__gt_park. But there are various corner cases where forcewakes are taken
without __gt_unpark i.e. without sending Busy Mailbox especially during
register reads. Forcewakes are taken without busy mailbox leads to
GPU HANG. So bringing mailbox calls under forcewake calls are no feasible
option as forcewake calls are atomic and mailbox calls are blocking.
The issue already fixed in B step so disabling MC6 on A step and
reverting previous commit which handles Wa_14017073508
Fixes: 8f70f1ec587d ("drm/i915/mtl: Add Wa_14017073508 for SAMedia")
Cc: Rodrigo Vivi <[email protected]>
Signed-off-by: Badal Nilawar <[email protected]>
Reviewed-by: Rodrigo Vivi <[email protected]>
Signed-off-by: Anshuman Gupta <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit 038a24835ab68f341eaa7a0e3bcc6ce0f9b22e17)
Signed-off-by: Jani Nikula <[email protected]>
|
|
intel_crtc_prepare_cleared_state() is unintentionally losing
the "inherited" flag. This will happen if intel_initial_commit()
is forced to go through the full modeset calculations for
whatever reason.
Afterwards the first real commit from userspace will not get
forced to the full modeset path, and thus eg. audio state may
not get recomputed properly. So if the monitor was already
enabled during boot audio will not work until userspace itself
does an explicit full modeset.
Cc: [email protected]
Tested-by: Lee Shawn C <[email protected]>
Signed-off-by: Ville Syrjälä <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Uma Shankar <[email protected]>
(cherry picked from commit 2553bacaf953b48c59357f5a622282bc0c45adae)
Signed-off-by: Jani Nikula <[email protected]>
|
|
lock the fbdev obj before calling into
i915_vma_pin_iomap(). This helps to solve below :
<7>[ 93.563308] i915 0000:00:02.0: [drm:intelfb_create [i915]] no BIOS fb, allocating a new one
<4>[ 93.581844] ------------[ cut here ]------------
<4>[ 93.581855] WARNING: CPU: 12 PID: 625 at drivers/gpu/drm/i915/gem/i915_gem_pages.c:424 i915_gem_object_pin_map+0x152/0x1c0 [i915]
Fixes: f0b6b01b3efe ("drm/i915: Add ww context to intel_dpt_pin, v2.")
Cc: Chris Wilson <[email protected]>
Cc: Matthew Auld <[email protected]>
Cc: Maarten Lankhorst <[email protected]>
Signed-off-by: Tejas Upadhyay <[email protected]>
Signed-off-by: Radhakrishna Sripada <[email protected]>
Reviewed-by: Andi Shyti <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit 561b31acfd65502a2cda2067513240fc57ccdbdc)
Signed-off-by: Jani Nikula <[email protected]>
|
|
The commit 2357f2b271ad ("drm/i915/mtl: Initial display workarounds")
extended the workaround Wa_16015201720 to MTL. However the registers
that the original WA implemented moved for MTL.
Implement the workaround with the correct register.
v3: Skip clock gating for pipe C, D DMC's and fix the title
Fixes: 2357f2b271ad ("drm/i915/mtl: Initial display workarounds")
Cc: Matt Atwood <[email protected]>
Cc: Lucas De Marchi <[email protected]>
Signed-off-by: Radhakrishna Sripada <[email protected]>
Reviewed-by: Matt Roper <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit 0188be507b973e36f637ba010a369057c8cb7282)
Signed-off-by: Jani Nikula <[email protected]>
|
|
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull tracing fixes from Steven Rostedt:
- Fix setting affinity of hwlat threads in containers
Using sched_set_affinity() has unwanted side effects when being
called within a container. Use set_cpus_allowed_ptr() instead
- Fix per cpu thread management of the hwlat tracer:
- Do not start per_cpu threads if one is already running for the CPU
- When starting per_cpu threads, do not clear the kthread variable
as it may already be set to running per cpu threads
- Fix return value for test_gen_kprobe_cmd()
On error the return value was overwritten by being set to the result
of the call from kprobe_event_delete(), which would likely succeed,
and thus have the function return success
- Fix splice() reads from the trace file that was broken by commit
36e2c7421f02 ("fs: don't allow splice read/write without explicit
ops")
- Remove obsolete and confusing comment in ring_buffer.c
The original design of the ring buffer used struct page flags for
tricks to optimize, which was shortly removed due to them being
tricks. But a comment for those tricks remained
- Set local functions and variables to static
* tag 'trace-v6.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing/hwlat: Replace sched_setaffinity with set_cpus_allowed_ptr
ring-buffer: remove obsolete comment for free_buffer_page()
tracing: Make splice_read available again
ftrace: Set direct_ops storage-class-specifier to static
trace/hwlat: Do not start per-cpu thread if it is already running
trace/hwlat: Do not wipe the contents of per-cpu thread data
tracing/osnoise: set several trace_osnoise.c variables storage-class-specifier to static
tracing: Fix wrong return in kprobe_event_gen_test.c
|
|
There is a problem with the behavior of hwlat in a container,
resulting in incorrect output. A warning message is generated:
"cpumask changed while in round-robin mode, switching to mode none",
and the tracing_cpumask is ignored. This issue arises because
the kernel thread, hwlatd, is not a part of the container, and
the function sched_setaffinity is unable to locate it using its PID.
Additionally, the task_struct of hwlatd is already known.
Ultimately, the function set_cpus_allowed_ptr achieves
the same outcome as sched_setaffinity, but employs task_struct
instead of PID.
Test case:
# cd /sys/kernel/tracing
# echo 0 > tracing_on
# echo round-robin > hwlat_detector/mode
# echo hwlat > current_tracer
# unshare --fork --pid bash -c 'echo 1 > tracing_on'
# dmesg -c
Actual behavior:
[573502.809060] hwlat_detector: cpumask changed while in round-robin mode, switching to mode none
Link: https://lore.kernel.org/linux-trace-kernel/[email protected]
Cc: Masami Hiramatsu <[email protected]>
Fixes: 0330f7aa8ee63 ("tracing: Have hwlat trace migrate across tracing_cpumask CPUs")
Signed-off-by: Costa Shulyupin <[email protected]>
Acked-by: Daniel Bristot de Oliveira <[email protected]>
Signed-off-by: Steven Rostedt (Google) <[email protected]>
|
|
The comment refers to mm/slob.c which is being removed. It comes from
commit ed56829cb319 ("ring_buffer: reset buffer page when freeing") and
according to Steven the borrowed code was a page mapcount and mapping
reset, which was later removed by commit e4c2ce82ca27 ("ring_buffer:
allocate buffer page pointer"). Thus the comment is not accurate anyway,
remove it.
Link: https://lore.kernel.org/linux-trace-kernel/[email protected]
Cc: Masami Hiramatsu <[email protected]>
Cc: Ingo Molnar <[email protected]>
Reported-by: Mike Rapoport <[email protected]>
Suggested-by: Steven Rostedt (Google) <[email protected]>
Fixes: e4c2ce82ca27 ("ring_buffer: allocate buffer page pointer")
Signed-off-by: Vlastimil Babka <[email protected]>
Reviewed-by: Mukesh Ojha <[email protected]>
Signed-off-by: Steven Rostedt (Google) <[email protected]>
|
|
Since the commit 36e2c7421f02 ("fs: don't allow splice read/write
without explicit ops") is applied to the kernel, splice() and
sendfile() calls on the trace file (/sys/kernel/debug/tracing
/trace) return EINVAL.
This patch restores these system calls by initializing splice_read
in file_operations of the trace file. This patch only enables such
functionalities for the read case.
Link: https://lore.kernel.org/linux-trace-kernel/[email protected]
Cc: [email protected]
Fixes: 36e2c7421f02 ("fs: don't allow splice read/write without explicit ops")
Signed-off-by: Sung-hun Kim <[email protected]>
Signed-off-by: Steven Rostedt (Google) <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Pull tty/serial driver fixes from Greg KH:
"Here are some small tty and serial driver fixes for 6.3-rc3 to resolve
some reported issues.
They include:
- 8250 driver Kconfig issue pointed out by you that showed up in -rc1
- qcom-geni serial driver fixes
- various 8250 driver fixes for reported problems
- fsl_lpuart driver fixes
- serdev fix for regression in -rc1
- vt.c bugfix
All have been in linux-next for over a week with no reported problems"
* tag 'tty-6.3-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
tty: vt: protect KD_FONT_OP_GET_TALL from unbound access
serial: qcom-geni: drop bogus uart_write_wakeup()
serial: qcom-geni: fix mapping of empty DMA buffer
serial: qcom-geni: fix DMA mapping leak on shutdown
serial: qcom-geni: fix console shutdown hang
serdev: Set fwnode for serdev devices
tty: serial: fsl_lpuart: fix race on RX DMA shutdown
serial: 8250_pci1xxxx: Disable SERIAL_8250_PCI1XXXX config by default
serial: 8250_fsl: fix handle_irq locking
serial: 8250_em: Fix UART port type
serial: 8250: ASPEED_VUART: select REGMAP instead of depending on it
tty: serial: fsl_lpuart: skip waiting for transmission complete when UARTCTRL_SBK is asserted
Revert "tty: serial: fsl_lpuart: adjust SERIAL_FSL_LPUART_CONSOLE config dependency"
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char/misc driver fixes from Greg KH:
"Here are a few small char/misc/other driver subsystem patches to
resolve reported problems for 6.3-rc3.
Included in here are:
- Interconnect driver fixes for reported problems
- Memory driver fixes for reported problems
- nvmem core fix
- firmware driver fix for reported problem
All of these have been in linux-next for a while with no reported
issues"
* tag 'char-misc-6.3-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (23 commits)
memory: tegra30-emc: fix interconnect registration race
memory: tegra20-emc: fix interconnect registration race
memory: tegra124-emc: fix interconnect registration race
memory: tegra: fix interconnect registration race
interconnect: exynos: drop redundant link destroy
interconnect: exynos: fix registration race
interconnect: exynos: fix node leak in probe PM QoS error path
interconnect: qcom: msm8974: fix registration race
interconnect: qcom: rpmh: fix registration race
interconnect: qcom: rpmh: fix probe child-node error handling
interconnect: qcom: rpm: fix registration race
nvmem: core: return -ENOENT if nvmem cell is not found
firmware: xilinx: don't make a sleepable memory allocation from an atomic context
interconnect: qcom: rpm: fix probe child-node error handling
interconnect: qcom: osm-l3: fix registration race
interconnect: imx: fix registration race
interconnect: fix provider registration API
interconnect: fix icc_provider_del() error handling
interconnect: fix mem leak when freeing nodes
interconnect: qcom: qcm2290: Fix MASTER_SNOC_BIMC_NRT
...
|
|
percpu_counter_sum_all() is now redundant as the race condition it
was invented to handle is now dealt with by percpu_counter_sum()
directly and all users of percpu_counter_sum_all() have been
removed.
Remove it.
This effectively reverts the changes made in f689054aace2
("percpu_counter: add percpu_counter_sum_all interface") except for
the cpumask iteration that fixes percpu_counter_sum() made earlier
in this series.
Signed-off-by: Dave Chinner <[email protected]>
Reviewed-by: Darrick J. Wong <[email protected]>
Signed-off-by: Darrick J. Wong <[email protected]>
|
|
This effectively reverts the change made in commit f689054aace2
("percpu_counter: add percpu_counter_sum_all interface") as the
race condition percpu_counter_sum_all() was invented to avoid is
now handled directly in percpu_counter_sum() and nobody needs to
care about summing racing with cpu unplug anymore.
Signed-off-by: Dave Chinner <[email protected]>
Reviewed-by: Darrick J. Wong <[email protected]>
Signed-off-by: Darrick J. Wong <[email protected]>
|
|
In commit f689054aace2 ("percpu_counter: add percpu_counter_sum_all
interface") a race condition between a cpu dying and
percpu_counter_sum() iterating online CPUs was identified. The
solution was to iterate all possible CPUs for summation via
percpu_counter_sum_all().
We recently had a percpu_counter_sum() call in XFS trip over this
same race condition and it fired a debug assert because the
filesystem was unmounting and the counter *should* be zero just
before we destroy it. That was reported here:
https://lore.kernel.org/linux-kernel/[email protected]/
likely as a result of running generic/648 which exercises
filesystems in the presence of CPU online/offline events.
The solution to use percpu_counter_sum_all() is an awful one. We
use percpu counters and percpu_counter_sum() for accurate and
reliable threshold detection for space management, so a summation
race condition during these operations can result in overcommit of
available space and that may result in filesystem shutdowns.
As percpu_counter_sum_all() iterates all possible CPUs rather than
just those online or even those present, the mask can include CPUs
that aren't even installed in the machine, or in the case of
machines that can hot-plug CPU capable nodes, even have physical
sockets present in the machine.
Fundamentally, this race condition is caused by the CPU being
offlined being removed from the cpu_online_mask before the notifier
that cleans up per-cpu state is run. Hence percpu_counter_sum() will
not sum the count for a cpu currently being taken offline,
regardless of whether the notifier has run or not. This is
the root cause of the bug.
The percpu counter notifier iterates all the registered counters,
locks the counter and moves the percpu count to the global sum.
This is serialised against other operations that move the percpu
counter to the global sum as well as percpu_counter_sum() operations
that sum the percpu counts while holding the counter lock.
Hence the notifier is safe to run concurrently with sum operations,
and the only thing we actually need to care about is that
percpu_counter_sum() iterates dying CPUs. That's trivial to do,
and when there are no CPUs dying, it has no addition overhead except
for a cpumask_or() operation.
This change makes percpu_counter_sum() always do the right thing in
the presence of CPU hot unplug events and makes
percpu_counter_sum_all() unnecessary. This, in turn, means that
filesystems like XFS, ext4, and btrfs don't have to work out when
they should use percpu_counter_sum() vs percpu_counter_sum_all() in
their space accounting algorithms
Signed-off-by: Dave Chinner <[email protected]>
Reviewed-by: Darrick J. Wong <[email protected]>
Signed-off-by: Darrick J. Wong <[email protected]>
|
|
Equivalent of for_each_cpu_and, except it ORs the two masks together
so it iterates all the CPUs present in either mask.
Signed-off-by: Dave Chinner <[email protected]>
Reviewed-by: Darrick J. Wong <[email protected]>
Signed-off-by: Darrick J. Wong <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull RAS fix from Borislav Petkov:
- Flush out logged errors immediately after MCA banks configuration
changes over sysfs have been done instead of waiting until something
else triggers the workqueue later - another error or the polling
interval cycle is reached
* tag 'ras_urgent_for_v6.3_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mce: Make sure logged MCEs are processed after sysfs update
|
|
Back in the 6.2-rc1 days, Eric Whitney reported a fstests regression in
ext4 against generic/454. The cause of this test failure was the
unfortunate combination of setting an xattr name containing UTF8 encoded
emoji, an xattr hash function that accepted a char pointer with no
explicit signedness, signed type extension of those chars to an int, and
the 6.2 build tools maintainers deciding to mandate -funsigned-char
across the board. As a result, the ondisk extended attribute structure
written out by 6.1 and 6.2 were not the same.
This discrepancy, in fact, had been noticeable if a filesystem with such
an xattr were moved between any two architectures that don't employ the
same signedness of a raw "char" declaration. The only reason anyone
noticed is that x86 gcc defaults to signed, and no such -funsigned-char
update was made to e2fsprogs, so e2fsck immediately started reporting
data corruption.
After a day and a half of discussing how to handle this use case (xattrs
with bit 7 set anywhere in the name) without breaking existing users,
Linus merged his own patch and didn't tell the maintainer. None of the
ext4 developers realized this until AUTOSEL announced that the commit
had been backported to stable.
In the end, this problem could have been detected much earlier if there
had been any useful tests of hash function(s) in use inside ext4 to make
sure that they always produce the same outputs given the same inputs.
The XFS dirent/xattr name hash takes a uint8_t*, so I don't think it's
vulnerable to this problem. However, let's avoid all this drama by
adding our own self test to check that the da hash produces the same
outputs for a static pile of inputs on various platforms. This enables
us to fix any breakage that may result in a controlled fashion. The
buffer and test data are identical to the patches submitted to xfsprogs.
Link: https://lore.kernel.org/linux-ext4/Y8bpkm3jA3bDm3eL@debian-BULLSEYE-live-builder-AMD64/
Link: https://lore.kernel.org/linux-xfs/ZBUKCRR7xvIqPrpX@destitution/T/#md38272cc684e2c0d61494435ccbb91f022e8dee4
Signed-off-by: Darrick J. Wong <[email protected]>
Reviewed-by: Dave Chinner <[email protected]>
|
|
There are now five separate space allocator interfaces exposed to the
rest of XFS for five different strategies to find space. Add
tracepoints for each of them so that I can tell from a trace dump
exactly which ones got called and what happened underneath them. Add a
sixth so it's more obvious if an allocation actually happened.
Signed-off-by: Darrick J. Wong <[email protected]>
Reviewed-by: Dave Chinner <[email protected]>
|
|
Callers of xfs_alloc_vextent_iterate_ags that pass in the TRYLOCK flag
want us to perform a non-blocking scan of the AGs for free space. There
are no ordering constraints for non-blocking AGF lock acquisition, so
the scan can freely start over at AG 0 even when minimum_agno > 0.
This manifests fairly reliably on xfs/294 on 6.3-rc2 with the parent
pointer patchset applied and the realtime volume enabled. I observed
the following sequence as part of an xfs_dir_createname call:
0. Fragment the free space, then allocate nearly all the free space in
all AGs except AG 0.
1. Create a directory in AG 2 and let it grow for a while.
2. Try to allocate 2 blocks to expand the dirent part of a directory.
The space will be allocated out of AG 0, but the allocation will not
be contiguous. This (I think) activates the LOWMODE allocator.
3. The bmapi call decides to convert from extents to bmbt format and
tries to allocate 1 block. This allocation request calls
xfs_alloc_vextent_start_ag with the inode number, which starts the
scan at AG 2. We ignore AG 0 (with all its free space) and instead
scrape AG 2 and 3 for more space. We find one block, but this now
kicks t_highest_agno to 3.
4. The createname call decides it needs to split the dabtree. It tries
to allocate even more space with xfs_alloc_vextent_start_ag, but now
we're constrained to AG 3, and we don't find the space. The
createname returns ENOSPC and the filesystem shuts down.
This change fixes the problem by making the trylock scan wrap around to
AG 0 if it doesn't like the AGs that it finds. Since the current
transaction itself holds AGF 0, the trylock of AGF 0 will succeed, and
we take space from the AG that has plenty.
Signed-off-by: Darrick J. Wong <[email protected]>
Reviewed-by: Dave Chinner <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Borislav Petkov:
- Check whether sibling events have been deactivated before adding them
to groups
- Update the proper event time tracking variable depending on the event
type
- Fix a memory overwrite issue due to using the wrong function argument
when outputting perf events
* tag 'perf_urgent_for_v6.3_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf: Fix check before add_event_to_groups() in perf_group_detach()
perf: fix perf_event_context->time
perf/core: Fix perf_output_begin parameter is incorrectly invoked in perf_event_bpf_output
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Borislav Petkov:
"There's a little bit more 'movement' in there for my taste but it
needs to happen and should make the code better after it.
- Check cmdline_find_option()'s return value before further
processing
- Clear temporary storage in the resctrl code to prevent access to an
unexistent MSR
- Add a simple throttling mechanism to protect the hypervisor from
potentially malicious SEV guests issuing requests in rapid
succession.
In order to not jeopardize the sanity of everyone involved in
maintaining this code, the request issuing side has received a
cleanup, split in more or less trivial, small and digestible
pieces. Otherwise, the code was threatening to become an
unmaintainable mess.
Therefore, that cleanup is marked indirectly also for stable so
that there's no differences between the upstream code and the
stable variant when it comes down to backporting more there"
* tag 'x86_urgent_for_v6.3_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mm: Fix use of uninitialized buffer in sme_enable()
x86/resctrl: Clear staged_config[] before and after it is used
virt/coco/sev-guest: Add throttling awareness
virt/coco/sev-guest: Convert the sw_exit_info_2 checking to a switch-case
virt/coco/sev-guest: Do some code style cleanups
virt/coco/sev-guest: Carve out the request issuing logic into a helper
virt/coco/sev-guest: Remove the disable_vmpck label in handle_guest_request()
virt/coco/sev-guest: Simplify extended guest request handling
virt/coco/sev-guest: Check SEV_SNP attribute at probe time
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4
Pull ext4 fix from Ted Ts'o:
"Fix a double unlock bug on an error path in ext4, found by smatch and
syzkaller"
* tag 'ext4_for_linus_urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
ext4: fix possible double unlock when moving a directory
|
|
smatch reports this warning
kernel/trace/ftrace.c:2594:19: warning:
symbol 'direct_ops' was not declared. Should it be static?
The variable direct_ops is only used in ftrace.c, so it should be static
Link: https://lore.kernel.org/linux-trace-kernel/[email protected]
Signed-off-by: Tom Rix <[email protected]>
Acked-by: Masami Hiramatsu (Google) <[email protected]>
Signed-off-by: Steven Rostedt (Google) <[email protected]>
|
|
The hwlatd tracer will end up starting multiple per-cpu threads with
the following script:
#!/bin/sh
cd /sys/kernel/debug/tracing
echo 0 > tracing_on
echo hwlat > current_tracer
echo per-cpu > hwlat_detector/mode
echo 100000 > hwlat_detector/width
echo 200000 > hwlat_detector/window
echo 1 > tracing_on
To fix the issue, check if the hwlatd thread for the cpu is already
running, before starting a new one. Along with the previous patch, this
avoids running multiple instances of the same CPU thread on the system.
Link: https://lore.kernel.org/all/[email protected]/
Link: https://lkml.kernel.org/r/[email protected]
Cc: [email protected]
Fixes: f46b16520a087 ("trace/hwlat: Implement the per-cpu mode")
Signed-off-by: Tero Kristo <[email protected]>
Acked-by: Daniel Bristot de Oliveira <[email protected]>
Signed-off-by: Steven Rostedt (Google) <[email protected]>
|
|
Do not wipe the contents of the per-cpu kthread data when starting the
tracer, as this will completely forget about already running instances
and can later start new additional per-cpu threads.
Link: https://lore.kernel.org/all/[email protected]/
Link: https://lkml.kernel.org/r/[email protected]
Cc: [email protected]
Fixes: f46b16520a087 ("trace/hwlat: Implement the per-cpu mode")
Signed-off-by: Tero Kristo <[email protected]>
Acked-by: Daniel Bristot de Oliveira <[email protected]>
Signed-off-by: Steven Rostedt (Google) <[email protected]>
|
|
storage-class-specifier to static
smatch reports several similar warnings
kernel/trace/trace_osnoise.c:220:1: warning:
symbol '__pcpu_scope_per_cpu_osnoise_var' was not declared. Should it be static?
kernel/trace/trace_osnoise.c:243:1: warning:
symbol '__pcpu_scope_per_cpu_timerlat_var' was not declared. Should it be static?
kernel/trace/trace_osnoise.c:335:14: warning:
symbol 'interface_lock' was not declared. Should it be static?
kernel/trace/trace_osnoise.c:2242:5: warning:
symbol 'timerlat_min_period' was not declared. Should it be static?
kernel/trace/trace_osnoise.c:2243:5: warning:
symbol 'timerlat_max_period' was not declared. Should it be static?
These variables are only used in trace_osnoise.c, so it should be static
Link: https://lore.kernel.org/linux-trace-kernel/[email protected]
Signed-off-by: Tom Rix <[email protected]>
Acked-by: Masami Hiramatsu (Google) <[email protected]>
Acked-by: Daniel Bristot de Oliveira <[email protected]>
Signed-off-by: Steven Rostedt (Google) <[email protected]>
|
|
Overwriting the error code with the deletion result may cause the
function to return 0 despite encountering an error. Commit b111545d26c0
("tracing: Remove the useless value assignment in
test_create_synth_event()") solves a similar issue by
returning the original error code, so this patch does the same.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Link: https://lore.kernel.org/linux-trace-kernel/[email protected]
Signed-off-by: Anton Gusev <[email protected]>
Reviewed-by: Steven Rostedt (Google) <[email protected]>
Acked-by: Masami Hiramatsu (Google) <[email protected]>
Signed-off-by: Steven Rostedt (Google) <[email protected]>
|
|
The cooling levels array is supposed to prevent the system fans from
being configured below a 20% duty cycle as otherwise some of them get
stuck at 0 RPM.
Due to an off-by-one error, the last element in the array was not
initialized, causing it to be set to zero, which in turn lead to fans
being configured with a 0% duty cycle in maximum cooling state.
Since commit 332fdf951df8 ("mlxsw: thermal: Fix out-of-bounds memory
accesses") the contents of the array are static. Therefore, instead of
fixing the initialization of the array, simply remove it and adjust
thermal_cooling_device_ops::set_cur_state() so that the configured duty
cycle is never set below 20%.
Before:
# cat /sys/class/thermal/thermal_zone0/cdev0/type
mlxsw_fan
# echo 10 > /sys/class/thermal/thermal_zone0/cdev0/cur_state
# cat /sys/class/hwmon/hwmon0/name
mlxsw
# cat /sys/class/hwmon/hwmon0/pwm1
0
After:
# cat /sys/class/thermal/thermal_zone0/cdev0/type
mlxsw_fan
# echo 10 > /sys/class/thermal/thermal_zone0/cdev0/cur_state
# cat /sys/class/hwmon/hwmon0/name
mlxsw
# cat /sys/class/hwmon/hwmon0/pwm1
255
This bug was uncovered when the thermal subsystem repeatedly tried to
configure the cooling devices to their maximum state due to another
issue [1]. This resulted in the fans being stuck at 0 RPM, which
eventually lead to the system undergoing thermal shutdown.
[1] https://lore.kernel.org/netdev/ZA3CFNhU4AbtsP4G@shredder/
Fixes: a421ce088ac8 ("mlxsw: core: Extend cooling device with cooling levels")
Signed-off-by: Ido Schimmel <[email protected]>
Reviewed-by: Vadim Pasternak <[email protected]>
Signed-off-by: Petr Machata <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Andrew reports that the SFF modules on one of the ZII platforms do not
indicate link up due to the SFP code believing that LOS indicating that
there is no signal being received from the remote end, but in fact the
LOS signal is showing that there is signal.
What makes SFF modules different from SFPs is they typically have an
inverted LOS, which uncovered this issue. When we read the hardware
state, we mask it with state_hw_mask so we ignore anything we're not
interested in. However, we don't re-read when state_hw_mask changes,
leading to sfp->state being stale.
Arrange for a software poll of the module state after we have parsed
the EEPROM in sfp_sm_mod_probe() and updated state_*_mask. This will
generate any necessary events for signal changes for the state
machine as well as updating sfp->state.
Reported-by: Andrew Lunn <[email protected]>
Tested-by: Andrew Lunn <[email protected]>
Fixes: 8475c4b70b04 ("net: sfp: re-implement soft state polling setup")
Signed-off-by: Russell King (Oracle) <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Currently DMA address width is either read from a RO device register
or force set from the platform data. This breaks DMA when the host DMA
address width is <=32it but the device is >32bit.
Right now the driver may decide to use a 2nd DMA descriptor for
another buffer (happens in case of TSO xmit) assuming that 32bit
addressing is used due to platform configuration but the device will
still use both descriptor addresses as one address.
This can be observed with the Intel EHL platform driver that sets
32bit for addr64 but the MAC reports 40bit. The TX queue gets stuck in
case of TCP with iptables NAT configuration on TSO packets.
The logic should be like this: Whatever we do on the host side (memory
allocation GFP flags) should happen with the host DMA width, whenever
we decide how to set addresses on the device registers we must use the
device DMA address width.
This patch renames the platform address width field from addr64 (term
used in device datasheet) to host_addr and uses this value exclusively
for host side operations while all chip operations consider the device
DMA width as read from the device register.
Fixes: 7cfc4486e7ea ("stmmac: intel: Configure EHL PSE0 GbE and PSE1 GbE to 32 bits DMA addressing")
Signed-off-by: Jochen Henneberg <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Florian Fainelli says:
====================
ACPI/DT mdiobus module owner fixes
This patch series fixes wrong mdiobus module ownership for MDIO buses
registered from DT or ACPI.
Thanks Maxime for providing the first patch and making me see that ACPI
also had the same issue.
Changes in v2:
- fixed missing kdoc in the first patch
====================
Signed-off-by: David S. Miller <[email protected]>
|
|
Bus ownership is wrong when using acpi_mdiobus_register() to register an
mdio bus. That function is not inline, so when it calls
mdiobus_register() the wrong THIS_MODULE value is captured.
CC: Maxime Bizon <[email protected]>
Fixes: 803ca24d2f92 ("net: mdio: Add ACPI support code for mdio")
Signed-off-by: Florian Fainelli <[email protected]>
Reviewed-by: Simon Horman <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Bus ownership is wrong when using of_mdiobus_register() to register an mdio
bus. That function is not inline, so when it calls mdiobus_register() the wrong
THIS_MODULE value is captured.
Signed-off-by: Maxime Bizon <[email protected]>
Fixes: 90eff9096c01 ("net: phy: Allow splitting MDIO bus/device support from PHYs")
[florian: fix kdoc, added Fixes tag]
Signed-off-by: Florian Fainelli <[email protected]>
Reviewed-by: Simon Horman <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
In the phy_disconnect() -> phy_stop() path, we will be forcibly setting
the PHY state machine to PHY_HALTED. This invalidates the old_state !=
phydev->state condition in phy_state_machine() such that we will neither
display the state change for debugging, nor will we invoke the
link_change_notify() callback.
Factor the code by introducing phy_process_state_change(), and ensure
that we process the state change from phy_stop() as well.
Fixes: 5c5f626bcace ("net: phy: improve handling link_change_notify callback")
Signed-off-by: Florian Fainelli <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2023-03-16 (igb, igbvf, igc)
This series contains updates to igb, igbvf, and igc drivers.
Lin Ma removes rtnl_lock() when disabling SRIOV on remove which was
causing deadlock on igb.
Akihiko Odaki delays enabling of SRIOV on igb to prevent early messages
that could get ignored and clears MAC address when PF returns nack on
reset; indicating no MAC address was assigned for igbvf.
Gaosheng Cui frees IRQs in error path for igbvf.
Akashi Takahiro fixes logic on checking TAPRIO gate support for igc.
====================
Signed-off-by: David S. Miller <[email protected]>
|
|
In xirc2ps_probe, the local->tx_timeout_task was bounded
with xirc2ps_tx_timeout_task. When timeout occurs,
it will call xirc_tx_timeout->schedule_work to start the
work.
When we call xirc2ps_detach to remove the driver, there
may be a sequence as follows:
Stop responding to timeout tasks and complete scheduled
tasks before cleanup in xirc2ps_detach, which will fix
the problem.
CPU0 CPU1
|xirc2ps_tx_timeout_task
xirc2ps_detach |
free_netdev |
kfree(dev); |
|
| do_reset
| //use dev
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Zheng Wang <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|