When possible use dev_err_probe help to properly deal with the
PROBE_DEFER error, the benefit is that DEFER issue will be logged
in the devices_deferred debugfs file.
Using dev_err_probe() can reduce code size, and the error value
gets printed.
Signed-off-by: Lucas Tanure <tanureal@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220117160830.709403-5-tanureal@opensource.cirrus.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The test key now needs to be manually held when calling
cs35l41_register_errata_patch, after patch:
Add the missing function calls to this driver.
Fixes: f517ba4924 ("ASoC: cs35l41: Add support for hibernate memory retention mode")
Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Signed-off-by: Lucas Tanure <tanureal@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220117160830.709403-2-tanureal@opensource.cirrus.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
regmap_register_patch can't be used to apply the probe sequence as a
patch is already registers with the regmap by
cs35l41_register_errata_patch and only a single patch can be attached to
a single regmap. The driver doesn't currently rely on a cache sync to
re-apply this probe sequence so simply switch it to a multi write.
Fixes: 7b2f3eb492 ("ALSA: hda: cs35l41: Add support for CS35L41 in HDA systems")
Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Signed-off-by: Lucas Tanure <tanureal@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220117160830.709403-1-tanureal@opensource.cirrus.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
When the target is a pid, not started by 'perf stat' we need to disable
the events, and in that case there is no need to setup affinities as we
use a dummy CPU map, with just one entry set to -1.
So stop doing it to avoid this needless call to sched_getaffinity():
# strace -ke sched_getaffinity perf stat -e cycles -p 241957 sleep 1
<SNIP>
sched_getaffinity(0, 512, [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]) = 8
> /usr/lib64/libc-2.33.so(sched_getaffinity@@GLIBC_2.3.4+0x1a) [0xe6eea]
> /var/home/acme/bin/perf(affinity__setup+0x6a) [0x532a2a]
> /var/home/acme/bin/perf(__evlist__disable.constprop.0+0x27) [0x4b9827]
> /var/home/acme/bin/perf(cmd_stat+0x29b5) [0x431725]
> /var/home/acme/bin/perf(run_builtin+0x6a) [0x4a2cfa]
> /var/home/acme/bin/perf(main+0x612) [0x40f8c2]
> /usr/lib64/libc-2.33.so(__libc_start_main+0xd4) [0x27b74]
> /var/home/acme/bin/perf(_start+0x2d) [0x40fadd]
<SNIP>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20220117160931.1191712-5-acme@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When the target is a pid, not started by 'perf stat' we need to enable
the events, and in that case there is no need to setup affinities as we
use a dummy CPU map, with just one entry set to -1.
So stop doing it to avoid this needless call to sched_getaffinity():
# strace -ke sched_getaffinity perf stat -e cycles -p 241957 sleep 1
<SNIP>
sched_getaffinity(0, 512, [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31]) = 8
> /usr/lib64/libc-2.33.so(sched_getaffinity@@GLIBC_2.3.4+0x1a) [0xe6eea]
> /var/home/acme/bin/perf(affinity__setup+0x6a) [0x5329ca]
> /var/home/acme/bin/perf(__evlist__enable.constprop.0+0x23) [0x4b9693]
> /var/home/acme/bin/perf(enable_counters+0x14d) [0x42de5d]
> /var/home/acme/bin/perf(cmd_stat+0x2358) [0x4310c8]
> /var/home/acme/bin/perf(run_builtin+0x6a) [0x4a2cfa]
> /var/home/acme/bin/perf(main+0x612) [0x40f8c2]
> /usr/lib64/libc-2.33.so(__libc_start_main+0xd4) [0x27b74]
> /var/home/acme/bin/perf(_start+0x2d) [0x40fadd]
<SNIP>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20220117160931.1191712-4-acme@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
I.e. the simple:
$ perf stat sleep 1
Uses a dummy CPU map and thus there is no need to setup/cleanup
affinities to avoid IPIs, etc.
With this we're down to a sched_getaffinity() call, in the libnuma
initialization, that probably can be removed in a followup patch.
Acked-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20220117160931.1191712-3-acme@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Just like with free(), NULL is checked to avoid having all callers do
it.
Its convenient for when not using affinity setup/cleanup for dummy CPU
maps, i.e. CPU maps for pid targets.
Acked-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20220117160931.1191712-2-acme@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Because of commit bf794bf52a ("powerpc/kprobes: Fix kallsyms
lookup across powerpc ABIv1 and ABIv2"), in ppc64 ABIv1, our perf
command eliminates the need to use the prefix "." at the symbol name.
But when the command "perf probe -a schedule" is executed on ppc64
ABIv1, it obtains two symbol address information through /proc/kallsyms,
for example:
cat /proc/kallsyms | grep -w schedule
c000000000657020 T .schedule
c000000000d4fdb8 D schedule
The symbol "D schedule" is not a function symbol, and perf will print:
"p:probe/schedule _text+13958584"Failed to write event: Invalid argument
Therefore, when searching symbols from map and adding probe point for
them, a symbol type check is added. If the type of symbol is not a
function, skip it.
Fixes: bf794bf52a ("powerpc/kprobes: Fix kallsyms lookup across powerpc ABIv1 and ABIv2")
Signed-off-by: Zechuan Chen <chenzechuan1@huawei.com>
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jianlin Lv <Jianlin.Lv@arm.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Cc: Yang Jihong <yangjihong1@huawei.com>
Link: https://lore.kernel.org/r/20211228111338.218602-1-chenzechuan1@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Now that have_bytes is never modified, we can simplify this function.
First, we move the check for negative entropy_count to be first. That
ensures that subsequent reads of this will be non-negative. Then,
have_bytes and ibytes can be folded into their one use site in the
min_t() function.
Suggested-by: Dominik Brodowski <linux@dominikbrodowski.net>
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
This is an old driver that has seen a lot of different eras of kernel
coding style. In an effort to make it easier to code for, unify the
coding style around the current norm, by accepting some of -- but
certainly not all of -- the suggestions from clang-format. This should
remove ambiguity in coding style, especially with regards to spacing,
when code is being changed or amended. Consequently it also makes code
review easier on the eyes, following one uniform style rather than
several.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
This gets rid of another abstraction we no longer need. It would be nice
if we could instead make pool an array rather than a pointer, but the
latent entropy plugin won't be able to do its magic in that case. So
instead we put all accesses to the input pool's actual data through the
input_pool_data array directly.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
The entropy estimator is calculated in terms of 1/8 bits, which means
there are various constants where things are shifted by 3. Move these
into our pool info enum with the other relevant constants. While we're
at it, move an English assertion about sizes into a proper BUILD_BUG_ON
so that the compiler can ensure this invariant.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
The other pool constants are prepended with POOL_, but not these last
ones. Rename them. This will then let us move them into the enum in the
following commit.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
We already had the POOL_* constants, so deduplicate the older INPUT_POOL
ones. As well, fold EXTRACT_SIZE into the poolinfo enum, since it's
related.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
We no longer have an output pool. Rather, we have just a wakeup bits
threshold for /dev/random reads, presumably so that processes don't
hang. This value, random_write_wakeup_bits, is configurable anyway. So
all the no longer usefully named OUTPUT_POOL constants were doing was
setting a reasonable default for random_write_wakeup_bits. This commit
gets rid of the constants and just puts it all in the default value of
random_write_wakeup_bits.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Originally, the RNG used several pools, so having things abstracted out
over a generic entropy_store object made sense. These days, there's only
one input pool, and then an uneven mix of usage via the abstraction and
usage via &input_pool. Rather than this uneasy mixture, just get rid of
the abstraction entirely and have things always use the global. This
simplifies the code and makes reading it a bit easier.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
This argument is always set to zero, as a result of us not caring about
keeping a certain amount reserved in the pool these days. So just remove
it and cleanup the function signatures.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
There were a few things added under the "if (fips_enabled)" banner,
which never really got completed, and the FIPS people anyway are
choosing a different direction. Rather than keep around this halfbaked
code, get rid of it so that we can focus on a single design of the RNG
rather than two designs.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Rather than using the userspace type, __uXX, switch to using uXX. And
rather than using variously chosen `char *` or `unsigned char *`, use
`u8 *` uniformly for things that aren't strings, in the case where we
are doing byte-by-byte traversal.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Now that we're only using one polynomial, we can cleanup its
representation into constants, instead of passing around pointers
dynamically to select different polynomials. This improves the codegen
and makes the code a bit more straightforward.
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
With SHA-1 no longer being used for anything performance oriented, and
also soon to be phased out entirely, we can make up for the space added
by unrolled BLAKE2s by simply re-rolling SHA-1. Since SHA-1 is so much
more complex, re-rolling it more or less takes care of the code size
added by BLAKE2s. And eventually, hopefully we'll see SHA-1 removed
entirely from most small kernel builds.
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Ard Biesheuvel <ardb@kernel.org>
Tested-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Basically nobody should use blake2s in an HMAC construction; it already
has a keyed variant. But unfortunately for historical reasons, Noise,
used by WireGuard, uses HKDF quite strictly, which means we have to use
this. Because this really shouldn't be used by others, this commit moves
it into wireguard's noise.c locally, so that kernels that aren't using
WireGuard don't get this superfluous code baked in. On m68k systems,
this shaves off ~314 bytes.
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Tested-by: Geert Uytterhoeven <geert@linux-m68k.org>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Commit 6048fdcc5f ("lib/crypto: blake2s: include as built-in") took
away a number of prompt texts from other crypto libraries. This makes
values flip from built-in to module when oldconfig runs, and causes
problems when these crypto libs need to be built in for thingslike
BIG_KEYS.
Fixes: 6048fdcc5f ("lib/crypto: blake2s: include as built-in")
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: linux-crypto@vger.kernel.org
Signed-off-by: Justin M. Forbes <jforbes@fedoraproject.org>
[Jason: - moved menu into submenu of lib/ instead of root menu
- fixed chacha sub-dependencies for CONFIG_CRYPTO]
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
New support:
- DMA_MEMCPY_SG support is bought back as we have a user in Xilinx driver
- Support for TI J721S2 SoC in k3-udma driver
- Support for Ingenic MDMA and BDMA in the JZ4760
- Support for Renesas r8a779f0 dmac
Updates:
- We are finally getting rid of slave_id, so this brings in the changes
across tree for that using tag dmaengine_topic_slave_id_removal_5.17
- updates for idxd driver
- at_xdmac driver cleanup
-----BEGIN PGP SIGNATURE-----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=1KYH
-----END PGP SIGNATURE-----
Merge tag 'dmaengine-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine
Pull dmaengine updates from Vinod Koul:
"A bunch of new support and few updates to drivers:
New support:
- DMA_MEMCPY_SG support is bought back as we have a user in Xilinx
driver
- Support for TI J721S2 SoC in k3-udma driver
- Support for Ingenic MDMA and BDMA in the JZ4760
- Support for Renesas r8a779f0 dmac
Updates:
- We are finally getting rid of slave_id, so this brings in the
changes across tree for that
- updates for idxd driver
- at_xdmac driver cleanup"
* tag 'dmaengine-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine: (60 commits)
dt-bindings: dma-controller: Split interrupt fields in example
dmaengine: pch_dma: Remove usage of the deprecated "pci-dma-compat.h" API
dmaengine: at_xdmac: Fix race over irq_status
dmaengine: at_xdmac: Remove a level of indentation in at_xdmac_tasklet()
dmaengine: at_xdmac: Fix at_xdmac_lld struct definition
dmaengine: at_xdmac: Fix lld view setting
dmaengine: at_xdmac: Remove a level of indentation in at_xdmac_advance_work()
dmaengine: at_xdmac: Fix concurrency over xfers_list
dmaengine: at_xdmac: Move the free desc to the tail of the desc list
dmaengine: at_xdmac: Fix race for the tx desc callback
dmaengine: at_xdmac: Fix concurrency over chan's completed_cookie
dmaengine: at_xdmac: Print debug message after realeasing the lock
dmaengine: at_xdmac: Start transfer for cyclic channels in issue_pending
dmaengine: at_xdmac: Don't start transactions at tx_submit level
dmaengine: idxd: deprecate token sysfs attributes for read buffers
dmaengine: idxd: change bandwidth token to read buffers
dmaengine: idxd: fix wq settings post wq disable
dmaengine: idxd: change MSIX allocation based on per wq activation
dmaengine: idxd: fix descriptor flushing locking
dmaengine: idxd: embed irq_entry in idxd_wq struct
...
Add a flag PIN_VALIDATE, to indicate we don't need to pin and only
protected by the object lock.
This removes the need to unpin, which is done by just releasing the
lock.
eb_reserve is slightly reworked for readability, but the same steps
are still done:
- First pass pins with NONBLOCK.
- Second pass unbinds all objects first, then pins.
- Third pass is only called when not all objects are softpinned, and
unbinds all objects, then calls i915_gem_evict_vm(), then pins.
Changes since v1:
- Split out eb_reserve() into separate functions for readability.
Changes since v2:
- Make batch buffer mappable on platforms where only GGTT is available,
to prevent moving the batch buffer during relocations.
Changes since v3:
- Preserve current behavior for batch buffer, instead be cautious when
calling i915_gem_object_ggtt_pin_ww, and re-use the current batch vma
if it's inside ggtt and map-and-fenceable.
- Remove impossible condition check from eb_reserve. (Matt)
Changes since v5:
- Do not even temporarily pin, just call i915_gem_evict_vm() and mark
all vma's as unpinned.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220114132320.109030-7-maarten.lankhorst@linux.intel.com
Now that we require the object lock for all ops, some code handling
race conditions can be removed.
This is required to not take short-term pins inside execbuf.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220114132320.109030-6-maarten.lankhorst@linux.intel.com
We want to remove more members of i915_vma, which requires the locking to
be held more often.
Start requiring gem object lock for i915_vma_unbind, as it's one of the
callers that may unpin pages.
Some special care is needed when evicting, because the last reference to
the object may be held by the VMA, so after __i915_vma_unbind, vma may be
garbage, and we need to cache vma->obj before unlocking.
Changes since v1:
- Make trylock failing a WARN. (Matt)
- Remove double i915_vma_wait_for_bind() (Matt)
- Move atomic_set to right before mutex_unlock(), to make it more clear
they belong together. (Matt)
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220114132320.109030-5-maarten.lankhorst@linux.intel.com
Because we will start to require the obj->resv lock for unbinding,
ensure these vma eviction utility functions also take the lock.
This requires some function signature changes, to ensure that the
ww context is passed around, but is mostly straightforward.
Previously this was split up into several patches, but reworking
should allow for easier bisection.
Changes since v1:
- Handle evicting dead objects better.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220114132320.109030-4-maarten.lankhorst@linux.intel.com
since commit 2279f540ea ("sched/deadline: Fix priority
inheritance with multiple scheduling classes"), we should not
keep it here.
Signed-off-by: Hui Su <suhui_kernel@163.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Daniel Bristot de Oliveira <bristot@redhat.com>
Link: https://lore.kernel.org/r/20220107095254.GA49258@localhost.localdomain
For PREEMPT/DYNAMIC_PREEMPT the *_unlock() will already trigger a
preemption, no point in then calling preempt_schedule_common()
*again*.
Use _cond_resched() instead, since this is a NOP for the preemptible
configs while it provide a preemption point for the others.
Reported-by: xuhaifeng <xuhaifeng@oppo.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/YcGnvDEYBwOiV0cR@hirez.programming.kicks-ass.net
Quieten all kernel-doc warnings in kernel/sched/fair.c:
kernel/sched/fair.c:3663: warning: No description found for return value of 'update_cfs_rq_load_avg'
kernel/sched/fair.c:8601: warning: No description found for return value of 'asym_smt_can_pull_tasks'
kernel/sched/fair.c:8673: warning: Function parameter or member 'sds' not described in 'update_sg_lb_stats'
kernel/sched/fair.c:9483: warning: contents before sections
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20211218055900.2704-1-rdunlap@infradead.org
There are two types of forced idle time: forced idle time from cookie'd
task and forced idle time form uncookie'd task. The forced idle time from
uncookie'd task is actually caused by the cookie'd task in runqueue
indirectly, and it's more accurate to measure the capacity loss with the
sum of both.
Assuming cpu x and cpu y are a pair of SMT siblings, consider the
following scenarios:
1.There's a cookie'd task running on cpu x, and there're 4 uncookie'd
tasks running on cpu y. For cpu x, there will be 80% forced idle time
(from uncookie'd task); for cpu y, there will be 20% forced idle time
(from cookie'd task).
2.There's a uncookie'd task running on cpu x, and there're 4 cookie'd
tasks running on cpu y. For cpu x, there will be 80% forced idle time
(from cookie'd task); for cpu y, there will be 20% forced idle time
(from uncookie'd task).
The scenario1 can recurrent by stress-ng(scenario2 can recurrent similary):
(cookie'd)taskset -c x stress-ng -c 1 -l 100
(uncookie'd)taskset -c y stress-ng -c 4 -l 100
In the above two scenarios, the total capacity loss is 1 cpu, but in
scenario1, the cookie'd forced idle time tells us 20% cpu capacity loss, in
scenario2, the cookie'd forced idle time tells us 80% cpu capacity loss,
which are not accurate. It'll be more accurate to measure with cookie'd
forced idle time and uncookie'd forced idle time.
Signed-off-by: Cruz Zhao <CruzZhao@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Josh Don <joshdon@google.com>
Link: https://lore.kernel.org/r/1641894961-9241-2-git-send-email-CruzZhao@linux.alibaba.com
Similarly to util_avg and util_sum, don't sync load_sum with the low
bound of load_avg but only ensure that load_sum stays in the correct range.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Link: https://lkml.kernel.org/r/20220111134659.24961-5-vincent.guittot@linaro.org
Similarly to util_avg and util_sum, don't sync runnable_sum with the low
bound of runnable_avg but only ensure that runnable_sum stays in the
correct range.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Link: https://lkml.kernel.org/r/20220111134659.24961-4-vincent.guittot@linaro.org
Rick reported performance regressions in bugzilla because of cpu frequency
being lower than before:
https://bugzilla.kernel.org/show_bug.cgi?id=215045
He bisected the problem to:
commit 1c35b07e6d ("sched/fair: Ensure _sum and _avg values stay consistent")
This commit forces util_sum to be synced with the new util_avg after
removing the contribution of a task and before the next periodic sync. By
doing so util_sum is rounded to its lower bound and might lost up to
LOAD_AVG_MAX-1 of accumulated contribution which has not yet been
reflected in util_avg.
update_tg_cfs_util() is not the only place where we round util_sum and
lost some accumulated contributions that are not already reflected in
util_avg. Modify update_tg_cfs_util() and detach_entity_load_avg() to not
sync util_sum with the new util_avg. Instead of always setting util_sum to
the low bound of util_avg, which can significantly lower the utilization,
we propagate the difference. In addition, we also check that cfs's util_sum
always stays above the lower bound for a given util_avg as it has been
observed that sched_entity's util_sum is sometimes above cfs one.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Link: https://lkml.kernel.org/r/20220111134659.24961-3-vincent.guittot@linaro.org
Rick reported performance regressions in bugzilla because of cpu frequency
being lower than before:
https://bugzilla.kernel.org/show_bug.cgi?id=215045
He bisected the problem to:
commit 1c35b07e6d ("sched/fair: Ensure _sum and _avg values stay consistent")
This commit forces util_sum to be synced with the new util_avg after
removing the contribution of a task and before the next periodic sync. By
doing so util_sum is rounded to its lower bound and might lost up to
LOAD_AVG_MAX-1 of accumulated contribution which has not yet been
reflected in util_avg.
Instead of always setting util_sum to the low bound of util_avg, which can
significantly lower the utilization of root cfs_rq after propagating the
change down into the hierarchy, we revert the change of util_sum and
propagate the difference.
In addition, we also check that cfs's util_sum always stays above the
lower bound for a given util_avg as it has been observed that
sched_entity's util_sum is sometimes above cfs one.
Fixes: 1c35b07e6d ("sched/fair: Ensure _sum and _avg values stay consistent")
Reported-by: Rick Yiu <rickyiu@google.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Link: https://lkml.kernel.org/r/20220111134659.24961-2-vincent.guittot@linaro.org
With write operation on psi files replacing old trigger with a new one,
the lifetime of its waitqueue is totally arbitrary. Overwriting an
existing trigger causes its waitqueue to be freed and pending poll()
will stumble on trigger->event_wait which was destroyed.
Fix this by disallowing to redefine an existing psi trigger. If a write
operation is used on a file descriptor with an already existing psi
trigger, the operation will fail with EBUSY error.
Also bypass a check for psi_disabled in the psi_trigger_destroy as the
flag can be flipped after the trigger is created, leading to a memory
leak.
Fixes: 0e94682b73 ("psi: introduce psi monitor")
Reported-by: syzbot+cdb5dd11c97cc532efad@syzkaller.appspotmail.com
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Analyzed-by: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Eric Biggers <ebiggers@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20220111232309.1786347-1-surenb@google.com
Some hypervisors support Arch LBR, but without the LBR XSAVE support.
The current Arch LBR init code prints a warning when the xsave size (0) is
unexpected. Avoid printing the warning for the "no LBR XSAVE" case.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20211215204029.150686-1-ak@linux.intel.com
Current ADL uncore code only supports the legacy IMC (memory controller)
free-running counters. Besides the free-running counters, ADL also
supports several general purpose-counters.
The general-purpose counters can also be accessed via MMIO but in a
different location. Factor out __uncore_imc_init_box() with offset as a
parameter. The function can be shared between ADL and TGL.
The event format and the layout of the control registers are a little
bit different from other uncore counters.
The intel_generic_uncore_mmio_enable_event() can be shared with client
IMC uncore. Expose the function.
Add more PCI IDs for ADL machines.
Fixes: 772ed05f3c ("perf/x86/intel/uncore: Add Alder Lake support")
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1642111554-118524-1-git-send-email-kan.liang@linux.intel.com
Using static_branch to replace the LBR INFO flags to optimize the LBR
INFO parsing.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lkml.kernel.org/r/1641315077-96661-2-git-send-email-peterz@infradead.org
The Goldmont plus and Tremont have LBR format V7. The V7 has LBR_INFO,
which is the same as LBR format V5. But V7 doesn't support TSX.
Without the patch, the associated misprediction and cycles information
in the LBR_INFO may be lost on a Goldmont plus platform.
For Tremont, the patch only impacts the non-PEBS events. Because of the
adaptive PEBS, the LBR_INFO is always processed for a PEBS event.
Currently, two different ways are used to check the LBR capabilities,
which make the codes complex and confusing.
For the LBR format V4 and earlier, the global static lbr_desc array is
used to store the flags for the LBR capabilities in each LBR format.
For LBR format V5 and V6, the current code checks the version number
for the LBR capabilities.
There are common LBR capabilities among LBR format versions. Several
flags for the LBR capabilities are introduced into the struct x86_pmu.
The flags, which can be shared among LBR formats, are used to check
the LBR capabilities. Add intel_pmu_lbr_init() to set the flags
accordingly at boot time.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lkml.kernel.org/r/1641315077-96661-1-git-send-email-peterz@infradead.org
The RAPL events exposed under /sys/devices/power/events should only reflect
what the underlying hardware actually support. This is how it works on Intel
RAPL and Intel core/uncore PMUs in general.
But on AMD, this was not the case. All possible RAPL events were advertised.
This is what it showed on an AMD Fam17h:
$ ls /sys/devices/power/events/
energy-cores energy-gpu energy-pkg energy-psys
energy-ram energy-cores.scale energy-gpu.scale energy-pkg.scale
energy-psys.scale energy-ram.scale energy-cores.unit energy-gpu.unit
energy-pkg.unit energy-psys.unit energy-ram.unit
Yet, on AMD Fam17h, only energy-pkg is supported.
This patch fixes the problem. Given the way perf_msr_probe() works, the
amd_rapl_msrs[] table has to have all entries filled out and in particular
the group field, otherwise perf_msr_probe() defaults to making the event
visible.
With the patch applied, the kernel now only shows was is actually supported:
$ ls /sys/devices/power/events/
energy-pkg energy-pkg.scale energy-pkg.unit
The patch also uses the RAPL_MSR_MASK because only the 32-bits LSB of the
RAPL counters are relevant when reading power consumption.
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220105185659.643355-1-eranian@google.com
The user recently report a perf issue in the ICX platform, when test by
perf event “uncore_imc_x/cas_count_write”,the write bandwidth is always
very small (only 0.38MB/s), it is caused by the wrong "umask" for the
"cas_count_write" event. When double-checking, find "cas_count_read"
also is wrong.
The public document for ICX uncore:
3rd Gen Intel® Xeon® Processor Scalable Family, Codename Ice Lake,Uncore
Performance Monitoring Reference Manual, Revision 1.00, May 2021
On 2.4.7, it defines Unit Masks for CAS_COUNT:
RD b00001111
WR b00110000
So corrected both "cas_count_read" and "cas_count_write" for ICX.
Old settings:
hswep_uncore_imc_events
INTEL_UNCORE_EVENT_DESC(cas_count_read, "event=0x04,umask=0x03")
INTEL_UNCORE_EVENT_DESC(cas_count_write, "event=0x04,umask=0x0c")
New settings:
snr_uncore_imc_events
INTEL_UNCORE_EVENT_DESC(cas_count_read, "event=0x04,umask=0x0f")
INTEL_UNCORE_EVENT_DESC(cas_count_write, "event=0x04,umask=0x30")
Fixes: 2b3b76b5ec ("perf/x86/intel/uncore: Add Ice Lake server uncore support")
Signed-off-by: Zhengjun Xing <zhengjun.xing@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20211223144826.841267-1-zhengjun.xing@linux.intel.com
For some Alder Lake machine with all E-cores disabled in a BIOS, the
below warning may be triggered.
[ 2.010766] hw perf events fixed 5 > max(4), clipping!
Current perf code relies on the CPUID leaf 0xA and leaf 7.EDX[15] to
calculate the number of the counters and follow the below assumption.
For a hybrid configuration, the leaf 7.EDX[15] (X86_FEATURE_HYBRID_CPU)
is set. The leaf 0xA only enumerate the common counters. Linux perf has
to manually add the extra GP counters and fixed counters for P-cores.
For a non-hybrid configuration, the X86_FEATURE_HYBRID_CPU should not
be set. The leaf 0xA enumerates all counters.
However, that's not the case when all E-cores are disabled in a BIOS.
Although there are only P-cores in the system, the leaf 7.EDX[15]
(X86_FEATURE_HYBRID_CPU) is still set. But the leaf 0xA is updated
to enumerate all counters of P-cores. The inconsistency triggers the
warning.
Several software ways were considered to handle the inconsistency.
- Drop the leaf 0xA and leaf 7.EDX[15] CPUID enumeration support.
Hardcode the number of counters. This solution may be a problem for
virtualization. A hypervisor cannot control the number of counters
in a Linux guest via changing the guest CPUID enumeration anymore.
- Find another CPUID bit that is also updated with E-cores disabled.
There may be a problem in the virtualization environment too. Because
a hypervisor may disable the feature/CPUID bit.
- The P-cores have a maximum of 8 GP counters and 4 fixed counters on
ADL. The maximum number can be used to detect the case.
This solution is implemented in this patch.
Fixes: ee72a94ea4 ("perf/x86/intel: Fix fixed counter check warning for some Alder Lake")
Reported-by: Damjan Marion (damarion) <damarion@cisco.com>
Reported-by: Chan Edison <edison_chan_gz@hotmail.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Damjan Marion (damarion) <damarion@cisco.com>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/1641925238-149288-1-git-send-email-kan.liang@linux.intel.com
Time readers that cannot take locks (due to NMI etc..) currently make
use of perf_event::shadow_ctx_time, which, for that event gives:
time' = now + (time - timestamp)
or, alternatively arranged:
time' = time + (now - timestamp)
IOW, the progression of time since the last time the shadow_ctx_time
was updated.
There's problems with this:
A) the shadow_ctx_time is per-event, even though the ctx_time it
reflects is obviously per context. The direct concequence of this
is that the context needs to iterate all events all the time to
keep the shadow_ctx_time in sync.
B) even with the prior point, the context itself might not be active
meaning its time should not advance to begin with.
C) shadow_ctx_time isn't consistently updated when ctx_time is
There are 3 users of this stuff, that suffer differently from this:
- calc_timer_values()
- perf_output_read()
- perf_event_update_userpage() /* A */
- perf_event_read_local() /* A,B */
In particular, perf_output_read() doesn't suffer at all, because it's
sample driven and hence only relevant when the event is actually
running.
This same was supposed to be true for perf_event_update_userpage(),
after all self-monitoring implies the context is active *HOWEVER*, as
per commit f792565326 ("perf/core: fix userpage->time_enabled of
inactive events") this goes wrong when combined with counter
overcommit, in that case those events that do not get scheduled when
the context becomes active (task events typically) miss out on the
EVENT_TIME update and ENABLED time is inflated (for a little while)
with the time the context was inactive. Once the event gets rotated
in, this gets corrected, leading to a non-monotonic timeflow.
perf_event_read_local() made things even worse, it can request time at
any point, suffering all the problems perf_event_update_userpage()
does and more. Because while perf_event_update_userpage() is limited
by the context being active, perf_event_read_local() users have no
such constraint.
Therefore, completely overhaul things and do away with
perf_event::shadow_ctx_time. Instead have regular context time updates
keep track of this offset directly and provide perf_event_time_now()
to complement perf_event_time().
perf_event_time_now() will, in adition to being context wide, also
take into account if the context is active. For inactive context, it
will not advance time.
This latter property means the cgroup perf_cgroup_info context needs
to grow addition state to track this.
Additionally, since all this is strictly per-cpu, we can use barrier()
to order context activity vs context time.
Fixes: 7d9285e82d ("perf/bpf: Extend the perf_event_read_local() interface, a.k.a. "bpf: perf event change needed for subsequent bpf helpers"")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Song Liu <song@kernel.org>
Tested-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/r/YcB06DasOBtU0b00@hirez.programming.kicks-ass.net
i915_gem_evict_vm will need to be able to evict objects that are
locked by the current ctx. By testing if the current context already
locked the object, we can do this correctly. This allows us to
evict the entire vm even if we already hold some objects' locks.
Previously, this was spread over several commits, but it makes
more sense to commit the changes to i915_gem_evict_vm separately
from the changes to i915_gem_evict_something() and
i915_gem_evict_for_node().
Changes since v1:
- Handle evicting dead objects better.
Changes since v2:
- Use for_i915_gem_ww in igt_evict_vm. (Thomas)
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
[mlankhorst: Fix up doc warning.]
Link: https://patchwork.freedesktop.org/patch/msgid/20220117075604.131477-1-maarten.lankhorst@linux.intel.com
Now that we cannot unbind kill the currently locked object directly
because we're removing short term pinning, we may have to unbind the
object from gtt manually, using a i915_gem_evict_vm() call.
Changes since v1:
- Remove -ENOSPC warning, can still happen with concurrent mmaps
where we can't unbind the other mmap because of the lock held.
This fixes the gem_mmap_gtt@cpuset tests.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220114132320.109030-2-maarten.lankhorst@linux.intel.com
