Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1640178 > unrolled thread
| Started by | Andrew Jeffery <andrew@aj.id.au> |
|---|---|
| First post | 2017-05-12 06:00 +0200 |
| Last post | 2017-05-18 19:10 +0200 |
| Articles | 2 — 2 participants |
Back to article view | Back to linux.kernel
[PATCH] powerpc: Tweak copy selection parameter in __copy_tofrom_user_power7() Andrew Jeffery <andrew@aj.id.au> - 2017-05-12 06:00 +0200
Re: [PATCH] powerpc: Tweak copy selection parameter in __copy_tofrom_user_power7() Anton Blanchard <anton@samba.org> - 2017-05-18 19:10 +0200
| From | Andrew Jeffery <andrew@aj.id.au> |
|---|---|
| Date | 2017-05-12 06:00 +0200 |
| Subject | [PATCH] powerpc: Tweak copy selection parameter in __copy_tofrom_user_power7() |
| Message-ID | <tG9Ue-75O-5@gated-at.bofh.it> |
Experiments with the netperf benchmark indicated that the size selecting VMX-based copies in __copy_tofrom_user_power7() was suboptimal on POWER8. Measurements showed that parity was in the neighbourhood of 3328 bytes, rather than greater than 4096. The change gives a 1.5-2.0% improvement in performance for 4096-byte buffers, reducing the relative time spent in __copy_tofrom_user_power7() from approximately 7% to approximately 5% in the TCP_RR benchmark. Signed-off-by: Andrew Jeffery <andrew@aj.id.au> --- arch/powerpc/lib/copyuser_power7.S | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/lib/copyuser_power7.S b/arch/powerpc/lib/copyuser_power7.S index a24b4039352c..706b7cc19846 100644 --- a/arch/powerpc/lib/copyuser_power7.S +++ b/arch/powerpc/lib/copyuser_power7.S @@ -82,14 +82,14 @@ _GLOBAL(__copy_tofrom_user_power7) #ifdef CONFIG_ALTIVEC cmpldi r5,16 - cmpldi cr1,r5,4096 + cmpldi cr1,r5,3328 std r3,-STACKFRAMESIZE+STK_REG(R31)(r1) std r4,-STACKFRAMESIZE+STK_REG(R30)(r1) std r5,-STACKFRAMESIZE+STK_REG(R29)(r1) blt .Lshort_copy - bgt cr1,.Lvmx_copy + bge cr1,.Lvmx_copy #else cmpldi r5,16 -- 2.9.3
[toc] | [next] | [standalone]
| From | Anton Blanchard <anton@samba.org> |
|---|---|
| Date | 2017-05-18 19:10 +0200 |
| Subject | Re: [PATCH] powerpc: Tweak copy selection parameter in __copy_tofrom_user_power7() |
| Message-ID | <tIx61-5FV-1@gated-at.bofh.it> |
| In reply to | #1640178 |
Hi Andrew, > Experiments with the netperf benchmark indicated that the size > selecting VMX-based copies in __copy_tofrom_user_power7() was > suboptimal on POWER8. Measurements showed that parity was in the > neighbourhood of 3328 bytes, rather than greater than 4096. The > change gives a 1.5-2.0% improvement in performance for 4096-byte > buffers, reducing the relative time spent in > __copy_tofrom_user_power7() from approximately 7% to approximately 5% > in the TCP_RR benchmark. Nice work! All our context switch optimisations we've made over the last year has likely moved the break even point for this. Acked-by: Anton Blanchard <anton@samba.org> Anton > Signed-off-by: Andrew Jeffery <andrew@aj.id.au> > --- > arch/powerpc/lib/copyuser_power7.S | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/arch/powerpc/lib/copyuser_power7.S > b/arch/powerpc/lib/copyuser_power7.S index a24b4039352c..706b7cc19846 > 100644 --- a/arch/powerpc/lib/copyuser_power7.S > +++ b/arch/powerpc/lib/copyuser_power7.S > @@ -82,14 +82,14 @@ > _GLOBAL(__copy_tofrom_user_power7) > #ifdef CONFIG_ALTIVEC > cmpldi r5,16 > - cmpldi cr1,r5,4096 > + cmpldi cr1,r5,3328 > > std r3,-STACKFRAMESIZE+STK_REG(R31)(r1) > std r4,-STACKFRAMESIZE+STK_REG(R30)(r1) > std r5,-STACKFRAMESIZE+STK_REG(R29)(r1) > > blt .Lshort_copy > - bgt cr1,.Lvmx_copy > + bge cr1,.Lvmx_copy > #else > cmpldi r5,16 >
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web