On Tue, Feb 13, 2024 at 8:01 PM Barry Song <21cn...@gmail.com> wrote:
>
> Hi Alex, Valentin,
>
>
> On Sun, Feb 11, 2024 at 12:37 AM wrote:
> >
> > From: Alex Shi
> >
> > SD_CLUSTER shares the CPU resources like llc tags or l2 cache, that's
>
ches.
So I feel this patch isn't precise.
>
> Suggested-by: Valentin Schneider
> Signed-off-by: Alex Shi
> Cc: linux-ker...@vger.kernel.org
> Cc: linuxppc-dev@lists.ozlabs.org
> Cc: Miaohe Lin
> Cc: Barry Song
> Cc: Mark Rutland
> Cc: Frederic Weisbecker
On Tue, Jul 4, 2023 at 10:36 PM Yicong Yang wrote:
>
> On 2023/6/30 1:26, Catalin Marinas wrote:
> > On Thu, Jun 29, 2023 at 05:31:36PM +0100, Catalin Marinas wrote:
> >> On Thu, May 18, 2023 at 02:59:34PM +0800, Yicong Yang wrote:
> >>> From: Barry Song
> &g
On Tue, Jan 10, 2023 at 1:19 AM Catalin Marinas wrote:
>
> On Sun, Jan 08, 2023 at 06:48:41PM +0800, Barry Song wrote:
> > On Fri, Jan 6, 2023 at 2:15 AM Catalin Marinas
> > wrote:
> > > On Thu, Nov 17, 2022 at 04:26:48PM +0800, Yicong Yang wrote:
> > > >
On Fri, Jan 6, 2023 at 2:15 AM Catalin Marinas wrote:
>
> On Thu, Nov 17, 2022 at 04:26:48PM +0800, Yicong Yang wrote:
> > It is tested on 4,8,128 CPU platforms and shows to be beneficial on
> > large systems but may not have improvement on small systems like on
> > a 4 CPU platform. So make ARCH_
On Sat, Oct 29, 2022 at 2:11 AM Punit Agrawal
wrote:
>
> Yicong Yang writes:
>
> > On 2022/10/27 22:19, Punit Agrawal wrote:
> >>
> >> [ Apologies for chiming in late in the conversation ]
> >>
> >> Anshuman Khandual writes:
> >>
>
On Thu, Oct 27, 2022 at 11:42 PM Anshuman Khandual
wrote:
>
>
>
> On 9/28/22 05:53, Barry Song wrote:
> > On Tue, Sep 27, 2022 at 10:15 PM Yicong Yang wrote:
> >>
> >> On 2022/9/27 14:16, Anshuman Khandual wrote:
> >>> [...]
> >>>
>
On Fri, Oct 28, 2022 at 3:19 AM Punit Agrawal
wrote:
>
>
> [ Apologies for chiming in late in the conversation ]
>
> Anshuman Khandual writes:
>
> > On 9/28/22 05:53, Barry Song wrote:
> >> On Tue, Sep 27, 2022 at 10:15 PM Yicong Yang wrote:
> >>>
On Tue, Sep 27, 2022 at 10:15 PM Yicong Yang wrote:
>
> On 2022/9/27 14:16, Anshuman Khandual wrote:
> > [...]
> >
> > On 9/21/22 14:13, Yicong Yang wrote:
> >> +static inline bool arch_tlbbatch_should_defer(struct mm_struct *mm)
> >> +{
> >> +/* for small systems with small number of CPUs, TL
..@linux.vnet.ibm.com/]
> Signed-off-by: Yicong Yang
> [Rebase and fix incorrect return value type]
> Reviewed-by: Kefeng Wang
> Reviewed-by: Anshuman Khandual
> ---
Reviewed-by: Barry Song
> arch/x86/include/asm/tlbflush.h | 12
> mm/rmap.c
On Wed, Sep 21, 2022 at 6:53 PM Anshuman Khandual
wrote:
>
>
> On 8/22/22 13:51, Yicong Yang wrote:
> > +static inline void arch_tlbbatch_add_mm(struct arch_tlbflush_unmap_batch
> > *batch,
> > + struct mm_struct *mm,
> > + u
On Wed, Sep 21, 2022 at 1:50 PM Barry Song <21cn...@gmail.com> wrote:
>
> On Tue, Sep 20, 2022 at 8:45 PM Anshuman Khandual
> wrote:
> >
> >
> >
> > On 9/20/22 09:09, Barry Song wrote:
> > > On Tue, Sep 20, 2022 at 3:00 PM Anshuman Khandual
>
On Tue, Sep 20, 2022 at 8:45 PM Anshuman Khandual
wrote:
>
>
>
> On 9/20/22 09:09, Barry Song wrote:
> > On Tue, Sep 20, 2022 at 3:00 PM Anshuman Khandual
> > wrote:
> >>
> >>
> >> On 8/22/22 13:51, Yicong Yang wrote:
> >>> +sta
On Tue, Sep 20, 2022 at 3:00 PM Anshuman Khandual
wrote:
>
>
> On 8/22/22 13:51, Yicong Yang wrote:
> > +static inline bool arch_tlbbatch_should_defer(struct mm_struct *mm)
> > +{
> > + return true;
> > +}
>
> This needs to be conditional on systems, where there will be performance
> improveme
On Mon, Sep 19, 2022 at 4:53 PM Barry Song <21cn...@gmail.com> wrote:
>
> On Mon, Sep 19, 2022 at 4:24 PM Anshuman Khandual
> wrote:
> >
> >
> >
> > On 9/15/22 12:12, Barry Song wrote:
> > > On Thu, Sep 15, 2022 at 6:07 PM Anshuman Khandual
> &
On Mon, Sep 19, 2022 at 4:24 PM Anshuman Khandual
wrote:
>
>
>
> On 9/15/22 12:12, Barry Song wrote:
> > On Thu, Sep 15, 2022 at 6:07 PM Anshuman Khandual
> > wrote:
> >>
> >>
> >>
> >> On 9/9/22 11:05, Barry Song wrote:
> &g
On Thu, Sep 15, 2022 at 6:07 PM Anshuman Khandual
wrote:
>
>
>
> On 9/9/22 11:05, Barry Song wrote:
> > On Fri, Sep 9, 2022 at 5:24 PM Anshuman Khandual
> > wrote:
> >>
> >>
> >>
> >> On 8/22/22 13:51, Yicong Yang wrote:
> >
On Fri, Sep 9, 2022 at 5:24 PM Anshuman Khandual
wrote:
>
>
>
> On 8/22/22 13:51, Yicong Yang wrote:
> > From: Barry Song
> >
> > on x86, batched and deferred tlb shootdown has lead to 90%
> > performance increase on tlb shootdown. on arm64, HW can do
> &g
On Fri, Sep 9, 2022 at 4:51 PM Anshuman Khandual
wrote:
>
>
>
> On 8/22/22 13:51, Yicong Yang wrote:
> > From: Barry Song
> >
> > Add uaddr to tlbbatch APIs so that platforms like ARM64 are
>
> I guess 'uaddr' refers to a virtual address from the p
On Fri, Sep 9, 2022 at 4:26 PM Anshuman Khandual
wrote:
>
>
>
> On 8/22/22 13:51, Yicong Yang wrote:
> > From: Barry Song
> >
> > This reverts commit 6bfef171d0d74cb050112e0e49feb20bfddf7f42.
> >
> > I was wrong. Though ARM64 has hardware TLB flu
On Tue, Jul 19, 2022 at 1:28 AM Yicong Yang wrote:
>
> On 2022/7/14 12:51, Barry Song wrote:
> > On Thu, Jul 14, 2022 at 3:29 PM Xin Hao wrote:
> >>
> >> Hi barry.
> >>
> >> I do some test on Kunpeng arm64 machine use Unixbench.
> >>
>
pts (1 concurrent) 109229.0 lpm (60.0 s, 1
> samples)
> System Benchmarks Partial Index BASELINE RESULTINDEX
> Shell Scripts (1 concurrent) 42.4 109229.0 25761.6
>
On Tue, Jul 12, 2022 at 1:35 AM Kefeng Wang wrote:
>
> Hi Barry，
>
> On 2022/7/11 11:46, Barry Song wrote:
> > From: Barry Song
> >
> > Platforms like ARM64 have hareware TLB shootdown broadcast. They
> > don't maintain mm_cpumask but just send tlbi
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
From: Barry Song
Add uaddr to tlbbatch APIs so that platforms like ARM64 are
able to apply this on their specific hardware features. For
ARM64, this could be sending tlbi into hardware queues for
the page with this particular uaddr.
Cc: Thomas Gleixner
Cc: Ingo Molnar
Cc: Borislav Petkov
Cc
From: Barry Song
Platforms like ARM64 have hareware TLB shootdown broadcast. They
don't maintain mm_cpumask but just send tlbi and related sync
instructions for TLB flush. task's mm_cpumask is normally empty
in this case. We also allow deferred TLB flush on this kind of
platforms.
Sig
From: Barry Song
This reverts commit 6bfef171d0d74cb050112e0e49feb20bfddf7f42.
I was wrong. Though ARM64 has hardware TLB flush, but it is not free
and it is still expensive.
We still have a good chance to enable batched and deferred TLB flush
on ARM64 for memory reclamation. A possible way is
242.425242-1-21cn...@gmail.com/
Barry Song (4):
Revert "Documentation/features: mark BATCHED_UNMAP_TLB_FLUSH doesn't
apply to ARM64"
mm: rmap: Allow platforms without mm_cpumask to defer TLB flush
mm: rmap: Extend tlbbatch APIs to fit new platforms
arm64: support batch
the patch has no
side effects and doesn't change the existing behaviour. But it has side effects
by changing the default N to Y on a couple of platforms.
> Suggested-by: Barry Song <21cn...@gmail.com>
> Signed-off-by: Valentin Schneider
> ---
> a
risc-v and arm64 support numa=off by common arch_numa_init()
in drivers/base/arch_numa.c. x86, ppc, mips, sparc support it
by arch-level early_param.
numa=off is widely used in linux distributions. it is better
to document it.
Signed-off-by: Barry Song
---
Documentation/admin-guide/kernel
as you are already in a tasklet, it is unnecessary to call spin_lock_bh.
Signed-off-by: Barry Song <21cn...@gmail.com>
---
drivers/dma/fsldma.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 1117b51..9d360a3 100644
ter Korsgaard
> Cc: Tony Lindgren
> Cc: Olof Johansson
> Cc: Vitaly Wool
> Cc: Guan Xuetao
> Cc: Barry Song
> Cc: "Mark M. Hoffman"
> Cc: linux-...@vger.kernel.org
> Cc: linuxppc-dev@lists.ozlabs.org
> Cc: linux-o...@vger.kernel.org
> Cc: linux-arm-ke
2012/6/14 Wolfram Sang :
> Currently, every driver has to do it on its own, but it should be done
> in the core, like we already do with board_info structs.
>
> Signed-off-by: Wolfram Sang
Acked-by: Barry Song
> ---
>
> Based on v3.5-rc2. Only build tested, I don
On Mon, Jun 21, 2010 at 7:20 PM, Anton Vorontsov wrote:
> On Mon, Jun 21, 2010 at 06:31:44PM +0800, Barry Song wrote:
>> On Mon, Jun 21, 2010 at 3:39 PM, Anton Vorontsov
>> wrote:
>> > On Mon, Jun 21, 2010 at 03:22:48PM +0800, Barry Song wrote:
>> >> On
On Mon, Jun 21, 2010 at 3:39 PM, Anton Vorontsov wrote:
> On Mon, Jun 21, 2010 at 03:22:48PM +0800, Barry Song wrote:
>> On Mon, Jun 21, 2010 at 3:15 PM, Anton Vorontsov
>> wrote:
>> > On Mon, Jun 21, 2010 at 11:27:31AM +0800, Barry Song wrote:
>> > [...]
>
On Mon, Jun 21, 2010 at 3:15 PM, Anton Vorontsov wrote:
> On Mon, Jun 21, 2010 at 11:27:31AM +0800, Barry Song wrote:
> [...]
>> > How about we add a non_jedec flag in platform_data, if the flag is 1, we
>> > let the detection pass even though the ID is 0? Otherwise,
On Mon, Jun 21, 2010 at 10:42 AM, Song, Barry wrote:
>
>
>>-Original Message-
>>From: uclinux-dist-devel-boun...@blackfin.uclinux.org
>>[mailto:uclinux-dist-devel-boun...@blackfin.uclinux.org] On
>>Behalf Of Anton Vorontsov
>>Sent: Friday, June 18
On Wed, Aug 19, 2009 at 5:46 AM, Anton Vorontsov
wrote:
>
> Previosly the driver always tried JEDEC probing, assuming that non-JEDEC
> chips will return '0'. But truly non-JEDEC chips (like CAT25) won't do
> that, their behaviour on RDID command is undefined, so the driver should
> not call jedec_
On Fri, Dec 18, 2009 at 12:08 PM, Zhiyong Wu wrote:
> HI,
>
> linux-2.6.32 is compiled on a p6 machine with RH5.4 OS and KVM option is
> enable.
>
> When rebooting this machine, a crash takes place such as:
>
> Loading ramdisk...
> ramdisk loaded at 0170, size: 2700 Kbytes
> OF stdout device
39 matches
Mail list logo
