Re: [PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2016-01-06 Thread Will Deacon
Hi Andrew, On Tue, Dec 22, 2015 at 03:32:19PM -0800, Andrew Pinski wrote: > On Tue, Dec 21, 2015 at 5:43 AM, Arnd Bergmann wrote: > > On Monday 21 December 2015, Will Deacon wrote: > >> On Sat, Dec 19, 2015 at 04:11:18PM -0800, Andrew Pinski wrote: > I think it is the prefetching. ThunderX T88

Re: [PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2016-01-06 Thread Will Deacon
Hi Andrew, On Tue, Dec 22, 2015 at 03:32:19PM -0800, Andrew Pinski wrote: > On Tue, Dec 21, 2015 at 5:43 AM, Arnd Bergmann wrote: > > On Monday 21 December 2015, Will Deacon wrote: > >> On Sat, Dec 19, 2015 at 04:11:18PM -0800, Andrew Pinski wrote: > I think it is the prefetching.

Re: [PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2015-12-22 Thread Andrew Pinski
On Tue, Dec 21, 2015 at 5:43 AM, Arnd Bergmann wrote: > > On Monday 21 December 2015, Will Deacon wrote: >> On Sat, Dec 19, 2015 at 04:11:18PM -0800, Andrew Pinski wrote: >> > Adding a check for the cache line size is not much overhead. >> > Special case 128 byte cache line size. >> > This

Re: [PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2015-12-22 Thread Andrew Pinski
On Tue, Dec 21, 2015 at 5:43 AM, Arnd Bergmann wrote: > > On Monday 21 December 2015, Will Deacon wrote: >> On Sat, Dec 19, 2015 at 04:11:18PM -0800, Andrew Pinski wrote: >> > Adding a check for the cache line size is not much overhead. >> > Special case 128 byte cache line size.

Re: [PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2015-12-21 Thread Arnd Bergmann
On Monday 21 December 2015, Will Deacon wrote: > On Sat, Dec 19, 2015 at 04:11:18PM -0800, Andrew Pinski wrote: > > Adding a check for the cache line size is not much overhead. > > Special case 128 byte cache line size. > > This improves copy_page by 85% on ThunderX compared to the > > original

Re: [PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2015-12-21 Thread Will Deacon
On Sat, Dec 19, 2015 at 04:11:18PM -0800, Andrew Pinski wrote: > Adding a check for the cache line size is not much overhead. > Special case 128 byte cache line size. > This improves copy_page by 85% on ThunderX compared to the > original implementation. So this patch seems to: - Align the

Re: [PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2015-12-21 Thread Arnd Bergmann
On Monday 21 December 2015, Will Deacon wrote: > On Sat, Dec 19, 2015 at 04:11:18PM -0800, Andrew Pinski wrote: > > Adding a check for the cache line size is not much overhead. > > Special case 128 byte cache line size. > > This improves copy_page by 85% on ThunderX compared to the > > original

Re: [PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2015-12-21 Thread Will Deacon
On Sat, Dec 19, 2015 at 04:11:18PM -0800, Andrew Pinski wrote: > Adding a check for the cache line size is not much overhead. > Special case 128 byte cache line size. > This improves copy_page by 85% on ThunderX compared to the > original implementation. So this patch seems to: - Align the

[PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2015-12-19 Thread Andrew Pinski
Adding a check for the cache line size is not much overhead. Special case 128 byte cache line size. This improves copy_page by 85% on ThunderX compared to the original implementation. For LMBench, it improves between 4-10%. Signed-off-by: Andrew Pinski --- arch/arm64/lib/copy_page.S | 39

[PATCH] ARM64: Improve copy_page for 128 cache line sizes.

2015-12-19 Thread Andrew Pinski
Adding a check for the cache line size is not much overhead. Special case 128 byte cache line size. This improves copy_page by 85% on ThunderX compared to the original implementation. For LMBench, it improves between 4-10%. Signed-off-by: Andrew Pinski ---