- On May 24, 2018, at 3:03 AM, Michael Ellerman m...@ellerman.id.au wrote:
> Mathieu Desnoyers writes:
>> - On May 23, 2018, at 4:14 PM, Mathieu Desnoyers
>> mathieu.desnoy...@efficios.com wrote:
> ...
>>>
>>> Hi Boqun,
>>>
>>> I tried your patch in a ppc64 le environment, and it does n
On 28/05/2018 07:23, Song, HaiyanX wrote:
>
> Some regression and improvements is found by LKP-tools(linux kernel
> performance) on V9 patch series
> tested on Intel 4s Skylake platform.
Hi,
Thanks for reporting this benchmark results, but you mentioned the "V9 patch
series" while responding to
Hi Laurent,
Yes, these tests are done on V9 patch.
Best regards,
Haiyan Song
On Mon, May 28, 2018 at 09:51:34AM +0200, Laurent Dufour wrote:
> On 28/05/2018 07:23, Song, HaiyanX wrote:
> >
> > Some regression and improvements is found by LKP-tools(linux kernel
> > performance) on V9 patch ser
On 28/05/2018 10:22, Haiyan Song wrote:
> Hi Laurent,
>
> Yes, these tests are done on V9 patch.
Do you plan to give this V11 a run ?
>
>
> Best regards,
> Haiyan Song
>
> On Mon, May 28, 2018 at 09:51:34AM +0200, Laurent Dufour wrote:
>> On 28/05/2018 07:23, Song, HaiyanX wrote:
>>>
>>> Some
On Fri, 25 May 2018 21:19:17 +0530
"Aneesh Kumar K.V" wrote:
> When relaxing access (read -> read_write update), pte need to be marked
> invalid
> to handle a nest MMU bug. We also need to do a tlb flush after the pte is
> marked invalid before updating the pte with new access bits.
>
> We also
On Tue, 22 May 2018 14:42:09 +0530
"Aneesh Kumar K.V" wrote:
> We used wrong page size (mmu_virtual_psize) when doing a tlbflush after
> pte update. This patch update the flush to use hugetlb page size.
> The page size is derived from hugetlb hstate.
>
> Now that ptep_set_access_flags won't be c
This serie intends to optimise memcmp() and __clear_user() for PPC32
in the same spirit as already done on PPC64.
The first patch moves PPC32 specific functions from string.S into
a dedicated file named string_32.S
The second patch rewrites __clear_user() by using dcbz intruction
The third patch r
In preparation of optimisation patches, move PPC32 specific
memcmp() and __clear_user() into string_32.S
Signed-off-by: Christophe Leroy
---
arch/powerpc/lib/Makefile| 5 +--
arch/powerpc/lib/string.S| 61 -
arch/powerpc/lib/string_32.S | 72 +
Rewrite clear_user() on the same principle as memset(0), making use
of dcbz to clear complete cache lines.
This code is a copy/paste of memset(), with some modifications
in order to retrieve remaining number of bytes to be cleared,
as it needs to be returned in case of error.
On a MPC885, through
At the time being, memcmp() compares two chunks of memory
byte per byte.
This patch optimises the comparison by comparing word by word.
A small benchmark performed on an 8xx comparing two chuncks
of 512 bytes performed 10 times gives:
Before : 5852274 TB ticks
After: 1488638 TB ticks
This
On Fri, May 25, 2018 at 12:07:33PM +0800, wei.guo.si...@gmail.com wrote:
> _GLOBAL(memcmp)
> cmpdi cr1,r5,0
>
> - /* Use the short loop if both strings are not 8B aligned */
> - or r6,r3,r4
> + /* Use the short loop if the src/dst addresses are not
> + * with the sa
Full run would take one or two weeks depended on our resource available. Could
you pick some ones up, e.g. those have performance regression?
-Original Message-
From: owner-linux...@kvack.org [mailto:owner-linux...@kvack.org] On Behalf Of
Laurent Dufour
Sent: Monday, May 28, 2018 4:55 PM
On Fri, May 25, 2018 at 12:07:34PM +0800, wei.guo.si...@gmail.com wrote:
> + /* save and restore cr0 */
> + mfocrf r5,64
> + EXIT_VMX_OPS
> + mtocrf 64,r5
> + b .LcmpAB_lightweight
That's cr1, not cr0. You can use mcrf instead, it is cheaper (esp. if
you have it in a n
On Wed, 23 May 2018 21:20:59 +0200,
Joe Perches wrote:
>
> Convert the S_ symbolic permissions to their octal equivalents as
> using octal and not symbolic permissions is preferred by many as more
> readable.
>
> see: https://lkml.org/lkml/2016/8/2/1945
>
> Done with automated conversion via:
>
Hi Simon,
wei.guo.si...@gmail.com writes:
> diff --git a/arch/powerpc/lib/memcmp_64.S b/arch/powerpc/lib/memcmp_64.S
> index f20e883..4ba7bb6 100644
> --- a/arch/powerpc/lib/memcmp_64.S
> +++ b/arch/powerpc/lib/memcmp_64.S
> @@ -174,6 +235,13 @@ _GLOBAL(memcmp)
> blr
>
> .Llong:
> +#ifdef
The generic implementation of strlen() reads strings byte per byte.
This patch implements strlen() in assembly for PPC32 based on
a read of entire words, in the same spirit as what some other
arches and glibc do.
For long strings, the time spent in strlen is reduced by 50-60%
Signed-off-by: Chri
On Sat, 2018-05-26 at 20:45 -0700, Guenter Roeck wrote:
>
> I already have a patch, or at least one that does the trick for me.
> Getting qemu patched was not the problem. I just want to be sure that
> the problem is indeed a qemu problem.
Hey Guenter !
It's not quite the right patch though. The
Latest GCC version emit many warnings similar to the following one.
As arch/powerpc code is built with -Werror, this breaks build with
GCC 8.1
This patch inhibits that warning
Signed-off-by: Christophe Leroy
---
arch/powerpc/Makefile | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/p
On Mon, May 28, 2018 at 02:17:49PM +, Christophe Leroy wrote:
> Latest GCC version emit many warnings similar to the following one.
You didn't actually show an example?
This warning should detect serious problems, so don't disable it without
first investigating please.
What do you call "late
On 05/28/2018 02:17 PM, Christophe Leroy wrote:
Latest GCC version emit many warnings similar to the following one.
Oops, forgot to include the failure text:
In file included from arch/powerpc/kernel/syscalls.c:24:
./include/linux/syscalls.h:233:18: warning: 'sys_mmap2' alias between
functi
Le 28/05/2018 à 16:37, Segher Boessenkool a écrit :
On Mon, May 28, 2018 at 02:17:49PM +, Christophe Leroy wrote:
Latest GCC version emit many warnings similar to the following one.
You didn't actually show an example?
Yes I forgot:
In file included from arch/powerpc/kernel/syscalls.c
On Mon, May 28, 2018 at 02:39:37PM +, Christophe Leroy wrote:
> In file included from arch/powerpc/kernel/syscalls.c:24:
> ./include/linux/syscalls.h:233:18: warning: 'sys_mmap2' alias between
> functions of incompatible types 'long int(long unsigned int, size_t,
> long unsigned int, long u
We now have barrier_nospec as mitigation so print it in
cpu_show_spectre_v1 when enabled.
Signed-off-by: Michal Suchanek
---
arch/powerpc/kernel/security.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c
ind
On Mon, May 28, 2018 at 5:07 PM, Segher Boessenkool
wrote:
> On Mon, May 28, 2018 at 02:39:37PM +, Christophe Leroy wrote:
>> In file included from arch/powerpc/kernel/syscalls.c:24:
>> ./include/linux/syscalls.h:233:18: warning: 'sys_mmap2' alias between
>> functions of incompatible types 'lo
CC arch/powerpc/kernel/nvram_64.o
arch/powerpc/kernel/nvram_64.c: In function 'nvram_create_partition':
arch/powerpc/kernel/nvram_64.c:1042:2: error: 'strncpy' specified bound 12
equals destination size [-Werror=stringop-truncation]
strncpy(new_part->header.name, name, 12);
^~~~
Names of cpuidle states were being used for description of states
in POWER as no descriptions were added in device tree. This patch
reads description for idle states which have been added in device
tree.
The description for idle states in case of POWER can be printed
using "cpupower monitor -l" or
On Mon, May 28, 2018 at 12:34 AM, Al Viro wrote:
> make it use copy_{from,to}_user(), rather than access_ok() +
> __copy_...
>
> Signed-off-by: Al Viro
> ---
> arch/powerpc/kernel/ptrace.c | 22 +++---
> 1 file changed, 7 insertions(+), 15 deletions(-)
>
> diff --git a/arch/power
> Maybe this is just an RFC, but:
>
> CALL../arch/powerpc/kernel/systbl_chk.sh
> ../arch/powerpc/kernel/ptrace.c: In function ‘arch_ptrace’:
> ../arch/powerpc/kernel/ptrace.c:3086:4: error: expected ‘)’ before ‘return’
> return -EFAULT;
> ^~
and the same a few lines later. Wh
On Mon, 2018-05-28 at 16:37 +, Christophe Leroy wrote:
> CC arch/powerpc/kernel/nvram_64.o
> arch/powerpc/kernel/nvram_64.c: In function 'nvram_create_partition':
> arch/powerpc/kernel/nvram_64.c:1042:2: error: 'strncpy' specified bound 12
> equals destination size [-Werror=stringop-tru
This tests perf hardware breakpoints (ie PERF_TYPE_BREAKPOINT) on
powerpc.
Signed-off-by: Michael Neuling
---
v2:
- fix failure noticed by mpe
- rewrite to explicitly test all options (rather than randomly)
---
.../selftests/powerpc/ptrace/.gitignore | 1 +
.../testing/selftests/powe
On Fri, 2018-05-25 at 20:45 +0300, Michael S. Tsirkin wrote:
> On Thu, May 24, 2018 at 08:27:04AM +1000, Benjamin Herrenschmidt wrote:
> > On Wed, 2018-05-23 at 21:50 +0300, Michael S. Tsirkin wrote:
> >
> > > I re-read that discussion and I'm still unclear on the
> > > original question, since I
On Tue, 2018-05-29 at 09:48 +1000, Benjamin Herrenschmidt wrote:
> > Well it's not supposed to be much slower for the static case.
> >
> > vhost has a cache so should be fine.
> >
> > A while ago Paolo implemented a translation cache which should be
> > perfect for this case - most of the code go
Vaibhav Jain writes:
> From: Vaibhav Jain
>
> Currently we see a kernel-oops reported on Power-9 while attaching a
> context to an AFU, with radix-mode and sysfs attr 'prefault_mode' set
> to anything other than 'none'. The backtrace of the oops is of this
> form:
>
> Unable to handle kernel pag
"Aneesh Kumar K.V" writes:
> We used wrong page size (mmu_virtual_psize) when doing a tlbflush after
> pte update. This patch update the flush to use hugetlb page size.
> The page size is derived from hugetlb hstate.
This sounds bad. Or is it not for some reason?
Either way a Fixes tag would be
Abhishek Goel writes:
> @@ -215,7 +216,7 @@ static inline void add_powernv_state(int index, const
> char *name,
>u64 psscr_val, u64 psscr_mask)
> {
> strlcpy(powernv_states[index].name, name, CPUIDLE_NAME_LEN);
> - strlcpy(powernv_states[index].desc,
From: Hoeun Ryu
Make printk_safe_flush() safe in NMI context. And printk_safe_flush_on_panic()
is
folded into this function. The prototype of printk_safe_flush() is changed to
"void printk_safe_flush(bool panic)".
nmi_trigger_cpumask_backtrace() can be called in NMI context. For example the
f
Michael Ellerman writes:
> Vaibhav Jain writes:
>
>> From: Vaibhav Jain
>>
>> Currently we see a kernel-oops reported on Power-9 while attaching a
>> context to an AFU, with radix-mode and sysfs attr 'prefault_mode' set
>> to anything other than 'none'. The backtrace of the oops is of this
>> f
On Tue, 29 May 2018 11:33:56 +1000
Michael Ellerman wrote:
> "Aneesh Kumar K.V" writes:
>
> > We used wrong page size (mmu_virtual_psize) when doing a tlbflush after
> > pte update. This patch update the flush to use hugetlb page size.
> > The page size is derived from hugetlb hstate.
>
> Th
On 05/29/2018 07:39 AM, Nicholas Piggin wrote:
On Tue, 29 May 2018 11:33:56 +1000
Michael Ellerman wrote:
"Aneesh Kumar K.V" writes:
We used wrong page size (mmu_virtual_psize) when doing a tlbflush after
pte update. This patch update the flush to use hugetlb page size.
The page size is der
CC arch/powerpc/kernel/nvram_64.o
arch/powerpc/kernel/nvram_64.c: In function 'nvram_create_partition':
arch/powerpc/kernel/nvram_64.c:1042:2: error: 'strncpy' specified bound 12
equals destination size [-Werror=stringop-truncation]
strncpy(new_part->header.name, name, 12);
^~~~
module_frob_arch_sections() is called before the module is moved to its
final location. The function descriptor section addresses we are setting
here are thus invalid. Fix this by processing opd section during
module_finalize()
Fixes: 5633e85b2c313 ("powerpc64: Add .opd based function descriptor
Latest GCC version emit the following warnings
As arch/powerpc code is built with -Werror, this breaks build with
GCC 8.1
This patch inhibits those warnings
CC arch/powerpc/kernel/syscalls.o
In file included from arch/powerpc/kernel/syscalls.c:24:
./include/linux/syscalls.h:233:18: error:
42 matches
Mail list logo
