在 2022/11/15 上午11:14, Yicong Yang 写道:
From: Anshuman Khandual
The entire scheme of deferred TLB flush in reclaim path rests on the
fact that the cost to refill TLB entries is less than flushing out
individual entries by sending IPI to remote CPUs. But architecture
can have different ways to e
在 2022/11/15 上午11:14, Yicong Yang 写道:
From: Barry Song
on x86, batched and deferred tlb shootdown has lead to 90%
performance increase on tlb shootdown. on arm64, HW can do
tlb shootdown without software IPI. But sync tlbi is still
quite expensive.
Even running a simplest program which require
On Tue, Nov 15, 2022 at 04:07:53AM +0100, Thomas Weißschuh wrote:
> On 2022-11-14 16:16-0800, Dmitry Torokhov wrote:
> > On Tue, Nov 15, 2022 at 12:54:41AM +0100, Thomas Weißschuh wrote:
> >> Cc Franz who wrote the driver originally.
> >> (I hope I got the correct one)
> >>
> >> Hi Dmitry,
> >>
>
> We're currently working on a feature in chromium that uses pkeys for
> in-process isolation. Being able to use the pkey state in the seccomp
> filter would be pretty useful for this. For example, it would allow
> us to enforce that no code outside the isolated thread would ever
> map/mprotect
On 2022/11/14 22:19, Anshuman Khandual wrote:
>
>
> On 11/14/22 14:16, Yicong Yang wrote:
>> On 2022/11/14 11:29, Anshuman Khandual wrote:
>>>
>>> On 10/28/22 13:42, Yicong Yang wrote:
+static inline bool arch_tlbbatch_should_defer(struct mm_struct *mm)
+{
+ /*
+ * TLB batc
From: Anshuman Khandual
The entire scheme of deferred TLB flush in reclaim path rests on the
fact that the cost to refill TLB entries is less than flushing out
individual entries by sending IPI to remote CPUs. But architecture
can have different ways to evaluate that. Hence apart from checking
TT
From: Yicong Yang
Though ARM64 has the hardware to do tlb shootdown, the hardware
broadcasting is not free.
A simplest micro benchmark shows even on snapdragon 888 with only
8 cores, the overhead for ptep_clear_flush is huge even for paging
out one page mapped by only one process:
5.36% a.out
From: Barry Song
on x86, batched and deferred tlb shootdown has lead to 90%
performance increase on tlb shootdown. on arm64, HW can do
tlb shootdown without software IPI. But sync tlbi is still
quite expensive.
Even running a simplest program which requires swapout can
prove this is true,
#incl
On 2022-11-14 16:16-0800, Dmitry Torokhov wrote:
> On Tue, Nov 15, 2022 at 12:54:41AM +0100, Thomas Weißschuh wrote:
>> Cc Franz who wrote the driver originally.
>> (I hope I got the correct one)
>>
>> Hi Dmitry,
>>
>> On 2022-11-14 10:33-0800, Dmitry Torokhov wrote:
>>> On Sun, Nov 13, 2022 at 0
On 2022-11-14, Christophe Leroy wrote:
Le 28/09/2022 à 07:25, Fangrui Song a écrit :
On Sat, Sep 10, 2022 at 12:53 AM Fangrui Song wrote:
On 2022-08-30, Fangrui Song wrote:
The actual intention is that no dynamic relocation exists. However, some
GNU ld ports produce unneeded R_*_NONE. (If
On Tue, Nov 15, 2022 at 12:54:41AM +0100, Thomas Weißschuh wrote:
> Cc Franz who wrote the driver originally.
> (I hope I got the correct one)
>
> Hi Dmitry,
>
> On 2022-11-14 10:33-0800, Dmitry Torokhov wrote:
> > On Sun, Nov 13, 2022 at 04:30:22AM +0100, Thomas Weißschuh wrote:
> >> There shoul
Cc Franz who wrote the driver originally.
(I hope I got the correct one)
Hi Dmitry,
On 2022-11-14 10:33-0800, Dmitry Torokhov wrote:
> On Sun, Nov 13, 2022 at 04:30:22AM +0100, Thomas Weißschuh wrote:
>> There should be no need to automatically load this driver on *all*
>> machines with a keyboar
https://bugzilla.kernel.org/show_bug.cgi?id=216095
Erhard F. (erhar...@mailbox.org) changed:
What|Removed |Added
Attachment #301576|0 |1
is obsolete|
https://bugzilla.kernel.org/show_bug.cgi?id=216095
--- Comment #5 from Erhard F. (erhar...@mailbox.org) ---
Created attachment 303182
--> https://bugzilla.kernel.org/attachment.cgi?id=303182&action=edit
dmesg (6.0-rc1, PowerMac G5 11,2)
Still shows up in 6.1-rc5:
[...]
sysfs: cannot create dup
On 11/10/22 04:58, Greg Kroah-Hartman wrote:
On Wed, Nov 09, 2022 at 03:10:37PM -0500, Nayna wrote:
On 11/9/22 08:46, Greg Kroah-Hartman wrote:
On Sun, Nov 06, 2022 at 04:07:42PM -0500, Nayna Jain wrote:
securityfs is meant for Linux security subsystems to expose policies/logs
or any other i
https://bugzilla.kernel.org/show_bug.cgi?id=216156
Erhard F. (erhar...@mailbox.org) changed:
What|Removed |Added
Attachment #301654|0 |1
is obsolete|
https://bugzilla.kernel.org/show_bug.cgi?id=216156
Erhard F. (erhar...@mailbox.org) changed:
What|Removed |Added
Attachment #301653|0 |1
is obsolete|
On Mon, Nov 14, 2022 at 01:08:03PM -0500, Sean Anderson wrote:
> On 11/14/22 12:23, Vladimir Oltean wrote:
> > On Thu, Nov 10, 2022 at 11:56:15AM -0500, Sean Anderson wrote:
> >> these will probably be in device trees for a year before the kernel
> >> starts using them. But once that is done, we ar
On Mon, Nov 14, 2022 at 12:21:33PM +, Christophe Leroy wrote:
> Gentle ping ... Can this fix be merged ?
Thanks for the reminder. Looks good, I'll go ahead and take it.
--
Josh
The pci_bus_for_each_resource_p() hides the iterator loop since
it may be not used otherwise. With this, we may drop that iterator
variable definition.
Signed-off-by: Andy Shevchenko
Acked-by: Dominik Brodowski
---
drivers/pcmcia/rsrc_nonstatic.c | 9 +++--
drivers/pcmcia/yenta_socket.c |
From: Mika Westerberg
Instead of open-coding it everywhere introduce a tiny helper that can be
used to iterate over each resource of a PCI device, and convert the most
obvious users into it.
While at it drop doubled empty line before pdev_sort_resources().
No functional changes intended.
Sugge
The pci_bus_for_each_resource_p() hides the iterator loop since
it may be not used otherwise. With this, we may drop that iterator
variable definition.
Signed-off-by: Andy Shevchenko
---
drivers/eisa/pci_eisa.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/eisa/
Refactor pci_bus_for_each_resource() in the same way as it's done in
pci_dev_for_each_resource() case. This will allow to hide iterator
inside the loop, where it's not used otherwise.
No functional changes intended.
Signed-off-by: Andy Shevchenko
---
.clang-format | 1 +
d
Provide two new helper macros to iterate over PCI device resources and
convert users.
Looking at it, refactor existing pci_bus_for_each_resource() and convert
users accordingly.
This applies on top of this patch Mika sent out earlier:
https://lore.kernel.org/r/20221114115953.40236-1-mika.westerb.
Hi Thomas,
On Sun, Nov 13, 2022 at 04:30:22AM +0100, Thomas Weißschuh wrote:
> There should be no need to automatically load this driver on *all*
> machines with a keyboard.
>
> This driver is of very limited utility and has to be enabled by the user
> explicitly anyway.
> Furthermore its own hea
On 11/14/22 12:23, Vladimir Oltean wrote:
> On Thu, Nov 10, 2022 at 11:56:15AM -0500, Sean Anderson wrote:
>> these will probably be in device trees for a year before the kernel
>> starts using them. But once that is done, we are free to require them.
>
> Sorry, you need to propose something that
This patch enables objtool --mcount on powerpc, and adds implementation
specific to powerpc.
Tested-by: Naveen N. Rao
Reviewed-by: Naveen N. Rao
Reviewed-by: Christophe Leroy
Acked-by: Josh Poimboeuf
Signed-off-by: Sathvika Vasireddy
---
arch/powerpc/Kconfig | 1 +
This patch adds [stub] implementations for required functions, inorder
to enable objtool build on powerpc.
Tested-by: Naveen N. Rao
Reviewed-by: Naveen N. Rao
Acked-by: Josh Poimboeuf
Signed-off-by: Sathvika Vasireddy
[Christophe Leroy: powerpc: Add missing asm/asm.h for objtool,
Use local var
Add architecture specific function to look for relocation records
pointing to architecture specific symbols.
Suggested-by: Christophe Leroy
Tested-by: Naveen N. Rao
Reviewed-by: Naveen N. Rao
Reviewed-by: Christophe Leroy
Acked-by: Josh Poimboeuf
Signed-off-by: Sathvika Vasireddy
---
tools/
Make relocation types architecture specific.
Tested-by: Naveen N. Rao
Reviewed-by: Naveen N. Rao
Reviewed-by: Christophe Leroy
Acked-by: Peter Zijlstra (Intel)
Acked-by: Josh Poimboeuf
Signed-off-by: Sathvika Vasireddy
---
tools/objtool/arch/x86/include/arch/elf.h | 2 ++
tools/objtool/chec
Call add_special_section_alts() only when stackval or orc or uaccess or
noinstr options are passed to objtool.
Tested-by: Naveen N. Rao
Reviewed-by: Naveen N. Rao
Reviewed-by: Christophe Leroy
Acked-by: Josh Poimboeuf
Signed-off-by: Sathvika Vasireddy
---
tools/objtool/check.c | 8 +---
Some architectures (powerpc) may not support ftrace locations being nop'ed
out at build time. Introduce CONFIG_HAVE_OBJTOOL_NOP_MCOUNT for objtool, as
a means for architectures to enable nop'ing of ftrace locations. Add --mnop
as an option to objtool --mcount, to indicate support for the same.
Als
From: Christophe Leroy
In order to allow using objtool on cross-built kernels,
determine size of long from elf data instead of using
sizeof(long) at build time.
For the time being this covers only mcount.
Tested-by: Naveen N. Rao
Reviewed-by: Naveen N. Rao
Acked-by: Josh Poimboeuf
Acked-by:
From: Christophe Leroy
Some architectures like powerpc support both endianness, it's
therefore not possible to fix the endianness via arch/endianness.h
because there is no easy way to get the target endianness at
build time.
Use the endianness recorded in the file objtool is working on.
Tested-
From: Christophe Leroy
find_insn() will return NULL in case of failure. Check insn in order
to avoid a kernel Oops for NULL pointer dereference.
Tested-by: Naveen N. Rao
Reviewed-by: Naveen N. Rao
Acked-by: Josh Poimboeuf
Acked-by: Peter Zijlstra (Intel)
Signed-off-by: Christophe Leroy
---
Do not run objtool on VDSO files, by using OBJECT_FILES_NON_STANDARD.
Suggested-by: Christophe Leroy
Tested-by: Naveen N. Rao
Reviewed-by: Naveen N. Rao
Reviewed-by: Christophe Leroy
Acked-by: Josh Poimboeuf
Signed-off-by: Sathvika Vasireddy
---
arch/powerpc/kernel/vdso/Makefile | 2 ++
1 f
From: Christophe Leroy
Fix several annotations in assembly files on PPC32.
Tested-by: Naveen N. Rao
Reviewed-by: Naveen N. Rao
Acked-by: Josh Poimboeuf
Signed-off-by: Christophe Leroy
[Sathvika Vasireddy: Changed subject line from "objtool/powerpc: Activate
objtool on PPC32" to "powerpc: Fix
With objtool enabled, below warnings are seen when trying to build:
drivers/crypto/vmx/aesp8-ppc.o: warning: objtool: aes_p8_set_encrypt_key+0x44:
unannotated intra-function call
drivers/crypto/vmx/aesp8-ppc.o: warning: objtool: .text+0x2448: unannotated
intra-function call
drivers/crypto/vmx/aes
objtool throws the following unannotated intra-function call warnings:
arch/powerpc/kernel/entry_64.o: warning: objtool: .text+0x4: unannotated
intra-function call
arch/powerpc/kvm/book3s_hv_rmhandlers.o: warning: objtool: .text+0xe64:
unannotated intra-function call
arch/powerpc/kvm/book3s_hv_rm
Objtool throws unannotated intra-function call warnings in the following
assembly files:
arch/powerpc/kernel/vector.o: warning: objtool: .text+0x53c: unannotated
intra-function call
arch/powerpc/kvm/book3s_hv_rmhandlers.o: warning: objtool: .text+0x60:
unannotated intra-function call
arch/power
In a subsequent patch, we would want to annotate powerpc assembly functions
with SYM_FUNC_START_LOCAL macro. This macro depends on __ALIGN macro.
The default expansion of __ALIGN macro is:
#define __ALIGN .align 4,0x90
So, override __ALIGN and __ALIGN_STR macros to use the same align
Commit 1e688dd2a3d675 ("powerpc/bug: Provide better flexibility to
WARN_ON/__WARN_FLAGS() with asm goto") updated __WARN_FLAGS() to use asm
goto, and added a call to 'unreachable()' after the asm goto for optimal
code generation. With CONFIG_OBJTOOL enabled, 'annotate_unreachable()'
statement in 'u
This patchset enables and implements objtool --mcount
option on powerpc. This applies atop powerpc/merge branch.
Changelog:
v6:
* Patch 06/16 - Add annotations to assembly files
arch/powerpc/kernel/cpu_setup_e500.S and
arch/powerpc/kernel/head_85xx.S
v5:
* Patch 02/16
On Mon, Nov 14, 2022 at 05:45:56PM +0100, Jason A. Donenfeld wrote:
> This is a simple mechanical transformation done by:
>
> @@
> expression E;
> @@
> - prandom_u32_max
> + get_random_u32_below
> (E)
>
> Reviewed-by: Greg Kroah-Hartman
> Acked-by: Darrick J. Wong # for xfs
> Reviewed-by: Seo
Le 14/11/2022 à 15:47, Hari Bathini a écrit :
> Hi Christophe,
>
> On 11/11/22 4:55 pm, Christophe Leroy wrote:
>> Le 10/11/2022 à 19:43, Hari Bathini a écrit :
>>> Most BPF programs are small, but they consume a page each. For systems
>>> with busy traffic and many BPF programs, this may also a
On Wed, 2 Nov 2022 18:29:41 +
Valentin Schneider wrote:
> This is incomplete, just looking at arm64 there's more IPI types that aren't
> covered:
>
> IPI_CPU_STOP,
> IPI_CPU_CRASH_STOP,
> IPI_TIMER,
> IPI_WAKEUP,
>
> ... But it feels like a good starting point.
For the tracing po
On Thu, Nov 10, 2022 at 11:56:15AM -0500, Sean Anderson wrote:
> these will probably be in device trees for a year before the kernel
> starts using them. But once that is done, we are free to require them.
Sorry, you need to propose something that is not "we can break compatibility
with today's de
These cases were done with this Coccinelle:
@@
expression H;
expression L;
@@
- (get_random_u32_below(H) + L)
+ get_random_u32_between(L, H + L)
@@
expression H;
expression L;
expression E;
@@
get_random_u32_between(L,
H
- + E
- - E
)
@@
expression H;
expression L;
expression E;
@@
get_r
These cases were done with this Coccinelle:
@@
expression E;
identifier I;
@@
- do {
... when != I
- I = get_random_u32();
... when != I
- } while (I > E);
+ I = get_random_u32_below(E + 1);
@@
expression E;
identifier I;
@@
- do {
... when != I
- I = get_random_
This is a simple mechanical transformation done by:
@@
expression E;
@@
- prandom_u32_max
+ get_random_u32_below
(E)
Reviewed-by: Greg Kroah-Hartman
Acked-by: Darrick J. Wong # for xfs
Reviewed-by: SeongJae Park # for damon
Reviewed-by: Jason Gunthorpe # for infiniband
Signed-off-by: Jason
Hey everyone,
[Changes v1->v2: Jakub asked me to send a v2 around now in the current
cycle to minimize potential conflicts with the net tree. So this is
rebased and stripped down to just the treewide parts. As well, the
removal function is deferred until the end of 6.2 in a separate pull so
th
It is not reliable to check for CON_ENABLED in order to identify if a
console is registered. Use console_is_registered() instead.
Signed-off-by: John Ogness
Reviewed-by: Petr Mladek
---
drivers/tty/hvc/hvc_console.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers
This is v4 of a series to prepare for threaded/atomic
printing. v3 is here [0]. This series focuses on reducing the
scope of the BKL console_lock. It achieves this by switching to
SRCU and a dedicated mutex for console list iteration and
modification, respectively. The console_lock will no longer
o
This adds compile-time options that allow the EH lock hint bit to be
enabled or disabled, and adds some new options that may or may not
help matters. To help with experimentation and tuning.
Signed-off-by: Nicholas Piggin
---
arch/powerpc/include/asm/qspinlock.h | 61 ++--
With the update primitive that clears the tail if it matches, and is
tolerant of other queueing activity on the lock, there is no longer a
significant reason to keep the large !maybe_stealers special case, so
remove it.
Signed-off-by: Nicholas Piggin
---
arch/powerpc/lib/qspinlock.c | 124 ++
Locking by the MCS queue head must clear the tail CPU if there are
no more queue entries left, and it has to deal with concurrent lock
stealing. Implementing these with cmpxchg style updates leaves the
possibility for unnecessary failure when the lock word changes.
Implement this instead within on
Using a cmpxchg-style trylock in the lock stealing code opens the
possibility for failures due to the lock word changing, even if it
could have been stolen.
Instead, use the stealing trylock which will do the right thing
and succeed unless the lwarx finds it locked or mustq.
Signed-off-by: Nichol
This series (at least the first 3 patches) will get folded into the main
series, but I will to post incremental changes first to make it easier
to see the change.
The 3rd patch is now quite an ugly mess, but once it is folded in then
everything should become simpler because that special case will
The VPA should unregister when offlining a CPU. Otherwise there could be a
short window where 2 CPUs could share the same VPA.
This happens because the hypervisor is still keeping the VPA attached to
the vCPU even if it became offline.
Here is a potential situation:
1. remove proc A,
2. add pro
During the comparative study of cpuidle governors, it is noticed that the
menu governor does not select CEDE state in some scenarios even though when
the sleep duration of the CPU exceeds the target residency of the CEDE idle
state this is because the CPU exits the snooze "polling" state when snooz
On 13/11/22 6:47 pm, Christophe Leroy wrote:
Le 10/11/2022 à 19:43, Hari Bathini a écrit :
bpf_arch_text_copy is used to dump JITed binary to RX page, allowing
multiple BPF programs to share the same page. Using patch_instruction
to implement it.
Using patch_instruction() is nice for a quic
Hi Christophe,
On 11/11/22 4:55 pm, Christophe Leroy wrote:
Le 10/11/2022 à 19:43, Hari Bathini a écrit :
Most BPF programs are small, but they consume a page each. For systems
with busy traffic and many BPF programs, this may also add significant
pressure on instruction TLB. High iTLB pressure
On 11/14/22 14:16, Yicong Yang wrote:
> On 2022/11/14 11:29, Anshuman Khandual wrote:
>>
>> On 10/28/22 13:42, Yicong Yang wrote:
>>> +static inline bool arch_tlbbatch_should_defer(struct mm_struct *mm)
>>> +{
>>> + /*
>>> +* TLB batched flush is proved to be beneficial for systems with la
Le 13/11/2022 à 10:13, wangkail...@jari.cn a écrit :
Fix the following coccicheck warning:
arch/powerpc/xmon/ppc-opc.c:957:67-68: WARNING: Use ARRAY_SIZE
arch/powerpc/xmon/ppc-opc.c:7280:24-25: WARNING: Use ARRAY_SIZE
arch/powerpc/xmon/ppc-opc.c:6972:25-26: WARNING: Use ARRAY_SIZE
arch/powerp
From: Athira Rajeev
[ Upstream commit ad353b710c7493df3d4fc2d3a51819126bed2e81 ]
'perf stat' with CSV output option prints an extra empty string as first
field in metrics output line. Sample output below:
# ./perf stat -x, --per-socket -a -C 1 ls
S0,1,1.78,msec,cpu-clock,178514
From: Athira Rajeev
[ Upstream commit ad353b710c7493df3d4fc2d3a51819126bed2e81 ]
'perf stat' with CSV output option prints an extra empty string as first
field in metrics output line. Sample output below:
# ./perf stat -x, --per-socket -a -C 1 ls
S0,1,1.78,msec,cpu-clock,178514
Le 18/05/2022 à 12:07, Paul Mackerras a écrit :
> From: Daniel Axtens
>
> Signed-off-by: Paul Mackerras
This change doesn't apply, can you rebase ?
Thanks
Christophe
> ---
> Documentation/dev-tools/kasan.rst | 7 ++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/
From: Athira Rajeev
[ Upstream commit ad353b710c7493df3d4fc2d3a51819126bed2e81 ]
'perf stat' with CSV output option prints an extra empty string as first
field in metrics output line. Sample output below:
# ./perf stat -x, --per-socket -a -C 1 ls
S0,1,1.78,msec,cpu-clock,178514
Le 23/05/2022 à 01:27, Andrew Donnellan a écrit :
> On 21/5/22 21:11, Julia Lawall wrote:
>> Spelling mistake (triple letters) in comment.
>> Detected with the help of Coccinelle.
>>
>> Signed-off-by: Julia Lawall
>
> Acked-by: Andrew Donnellan
>
>
How should apply that change ?
Thanks
Chris
Should this be applied by Bjorn as most recent changes to that file ?
Le 14/06/2022 à 12:12, Andrew Donnellan a écrit :
> On Sun, 2022-06-05 at 10:00 +0400, Miaoqian Lin wrote:
>> of_get_next_parent() returns a node pointer with refcount
>> incremented,
>> we should use of_node_put() on it when n
Le 28/09/2022 à 07:25, Fangrui Song a écrit :
> On Sat, Sep 10, 2022 at 12:53 AM Fangrui Song wrote:
>>
>> On 2022-08-30, Fangrui Song wrote:
>>> The actual intention is that no dynamic relocation exists. However, some
>>> GNU ld ports produce unneeded R_*_NONE. (If a port is not care enough to
Gentle ping ... Can this fix be merged ?
Thanks
Le 30/09/2022 à 12:09, Naveen N. Rao a écrit :
> Thadeu Lima de Souza Cascardo wrote:
>> On Tue, Sep 27, 2022 at 01:22:11PM +0530, Srikar Dronamraju wrote:
>>> Commit 1d1a0e7c5100 ("scripts/faddr2line: Fix overlapping text
>>> section failures")
>>
Le 26/10/2022 à 06:27, Samuel Holland a écrit :
> On 9/21/22 00:17, Christophe Leroy wrote:
>>
>> Instead of the patch from Michael, could you try by replacing
>> preempt_enable() by preempt_enable_no_resched() in exit_vmx_usercopy() ?
>
> I finally got a chance to test this, and the simpler fix
Le 28/10/2022 à 07:05, Dmitry Torokhov a écrit :
> On Mon, Sep 26, 2022 at 11:03:25PM -0700, Dmitry Torokhov wrote:
>> This switches PIKA Warp away from legacy gpio API and to newer gpiod
>> API, so that we can eventually deprecate the former.
>>
>> Because LEDs are normally driven by leds-gpio d
On 14/11/2022 12:46, Yang Yingliang wrote:
Hi,
On 2022/11/14 19:23, Frederic Barrat wrote:
On 11/11/2022 15:59, Yang Yingliang wrote:
If device_register() returns error in ocxl_file_register_afu(),
the name allocated by dev_set_name() need be freed. As comment
of device_register() says, i
Hi,
On 2022/11/14 19:23, Frederic Barrat wrote:
On 11/11/2022 15:59, Yang Yingliang wrote:
If device_register() returns error in ocxl_file_register_afu(),
the name allocated by dev_set_name() need be freed. As comment
of device_register() says, it should use put_device() to give
up the refere
On 11/11/2022 15:59, Yang Yingliang wrote:
If device_register() returns error in ocxl_file_register_afu(),
the name allocated by dev_set_name() need be freed. As comment
of device_register() says, it should use put_device() to give
up the reference in the error path. So fix this by calling
put
> Add the current value of an architecture specific protection keys
> register (currently PKRU on x86) to data available for seccomp-bpf
> programs to work on. This allows filters based on the currently
> enabled protection keys.
We're currently working on a feature in chromium that uses pkeys for
During the comparative study of cpuidle governors, it is noticed that the
menu governor does not select CEDE state in some scenarios even though when
the sleep duration of the CPU exceeds the target residency of the CEDE idle
state this is because the CPU exits the snooze "polling" state when snooz
> On 09-Nov-2022, at 3:53 PM, Athira Rajeev wrote:
>
>
>
>> On 09-Nov-2022, at 2:27 AM, Arnaldo Carvalho de Melo wrote:
>>
>> Em Wed, Nov 02, 2022 at 02:07:06PM +0530, Athira Rajeev escreveu:
>>>
>>>
On 18-Oct-2022, at 2:26 PM, Athira Rajeev
wrote:
In perf stat wit
In perf stat with CSV output option, number of fields
in metrics output is not matching with number of fields
in other event output lines.
Sample output below after applying patch to fix
printing os->prefix.
# ./perf stat -x, --per-socket -a -C 1 ls
S0,1,1.89,msec,cpu-clock,188769
On 2022/11/14 11:29, Anshuman Khandual wrote:
>
>
> On 10/28/22 13:42, Yicong Yang wrote:
>> +static inline bool arch_tlbbatch_should_defer(struct mm_struct *mm)
>> +{
>> +/*
>> + * TLB batched flush is proved to be beneficial for systems with large
>> + * number of CPUs, especially s
在 2022/10/28 下午4:12, Yicong Yang 写道:
From: Barry Song
on x86, batched and deferred tlb shootdown has lead to 90%
performance increase on tlb shootdown. on arm64, HW can do
tlb shootdown without software IPI. But sync tlbi is still
quite expensive.
Even running a simplest program which requir
83 matches
Mail list logo