* [PATCH] Add fast little-endian switch system call
@ 2008-04-28 3:52 Paul Mackerras
[not found] ` <18453.18943.906693.409245-UYQwCShxghk5kJ7NmlRacFaTQe2KTcn/@public.gmane.org>
0 siblings, 1 reply; 19+ messages in thread
From: Paul Mackerras @ 2008-04-28 3:52 UTC (permalink / raw)
To: linuxppc-dev
This adds a system call on 64-bit platforms for switching between
little-endian and big-endian modes that is much faster than doing a
prctl call. This system call is handled as a special case right at
the start of the system call entry code, and because it is a special
case, it uses a system call number which is out of the range of
normal system calls, namely 0x1ebe.
Measurements with lmbench on a 4.2GHz POWER6 showed no measurable
change in the speed of normal system calls with this patch.
Switching endianness with this new system call takes around 60ns on a
4.2GHz POWER6, compared with around 300ns to switch endian mode with a
prctl. This can provide a significant performance advantage for
emulators for little-endian architectures that want to switch between
big-endian and little-endian mode frequently, e.g. because they are
generating instructions sequences on the fly and they want to run
those sequences in little-endian mode.
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 215973a..2eb49a7 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -239,6 +239,10 @@ instruction_access_slb_pSeries:
.globl system_call_pSeries
system_call_pSeries:
HMT_MEDIUM
+BEGIN_FTR_SECTION
+ cmpdi r0,0x1ebe
+ beq- 1f
+END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE)
mr r9,r13
mfmsr r10
mfspr r13,SPRN_SPRG3
@@ -253,6 +257,13 @@ system_call_pSeries:
rfid
b . /* prevent speculative execution */
+/* Fast LE/BE switch system call */
+1: mfspr r12,SPRN_SRR1
+ xori r12,r12,MSR_LE
+ mtspr SPRN_SRR1,r12
+ rfid /* return to userspace */
+ b .
+
STD_EXCEPTION_PSERIES(0xd00, single_step)
STD_EXCEPTION_PSERIES(0xe00, trap_0e)
^ permalink raw reply related [flat|nested] 19+ messages in thread[parent not found: <18453.18943.906693.409245-UYQwCShxghk5kJ7NmlRacFaTQe2KTcn/@public.gmane.org>]
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-28 14:43 ` Christoph Hellwig 0 siblings, 0 replies; 19+ messages in thread From: Christoph Hellwig @ 2008-04-28 14:43 UTC (permalink / raw) To: Paul Mackerras Cc: linuxppc-dev-mnsaURCQ41sdnm+yROfE0A, mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w, linux-arch-u79uwXL29TY76Z2rM5mHXA Please see Michael Kerrisk on userspace ABI updates. A nice little manpage for this gimmick would be helpful, and maybe help other platforms that want one aswell to implement the same API. On Mon, Apr 28, 2008 at 01:52:31PM +1000, Paul Mackerras wrote: > This adds a system call on 64-bit platforms for switching between > little-endian and big-endian modes that is much faster than doing a > prctl call. This system call is handled as a special case right at > the start of the system call entry code, and because it is a special > case, it uses a system call number which is out of the range of > normal system calls, namely 0x1ebe. > > Measurements with lmbench on a 4.2GHz POWER6 showed no measurable > change in the speed of normal system calls with this patch. > > Switching endianness with this new system call takes around 60ns on a > 4.2GHz POWER6, compared with around 300ns to switch endian mode with a > prctl. This can provide a significant performance advantage for > emulators for little-endian architectures that want to switch between > big-endian and little-endian mode frequently, e.g. because they are > generating instructions sequences on the fly and they want to run > those sequences in little-endian mode. > > Signed-off-by: Paul Mackerras <paulus-eUNUBHrolfbYtjvyW6yDsg@public.gmane.org> > --- > > diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S > index 215973a..2eb49a7 100644 > --- a/arch/powerpc/kernel/head_64.S > +++ b/arch/powerpc/kernel/head_64.S > @@ -239,6 +239,10 @@ instruction_access_slb_pSeries: > .globl system_call_pSeries > system_call_pSeries: > HMT_MEDIUM > +BEGIN_FTR_SECTION > + cmpdi r0,0x1ebe > + beq- 1f > +END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) Am I missing something here or does this add a branch for every normal syscall? > mr r9,r13 > mfmsr r10 > mfspr r13,SPRN_SPRG3 > @@ -253,6 +257,13 @@ system_call_pSeries: > rfid > b . /* prevent speculative execution */ > > +/* Fast LE/BE switch system call */ > +1: mfspr r12,SPRN_SRR1 > + xori r12,r12,MSR_LE > + mtspr SPRN_SRR1,r12 > + rfid /* return to userspace */ > + b . > + > STD_EXCEPTION_PSERIES(0xd00, single_step) > STD_EXCEPTION_PSERIES(0xe00, trap_0e) > > _______________________________________________ > Linuxppc-dev mailing list > Linuxppc-dev-mnsaURCQ41sdnm+yROfE0A@public.gmane.org > https://ozlabs.org/mailman/listinfo/linuxppc-dev ---end quoted text--- ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-28 14:43 ` Christoph Hellwig 0 siblings, 0 replies; 19+ messages in thread From: Christoph Hellwig @ 2008-04-28 14:43 UTC (permalink / raw) To: Paul Mackerras; +Cc: linux-arch, linuxppc-dev, mtk.manpages Please see Michael Kerrisk on userspace ABI updates. A nice little manpage for this gimmick would be helpful, and maybe help other platforms that want one aswell to implement the same API. On Mon, Apr 28, 2008 at 01:52:31PM +1000, Paul Mackerras wrote: > This adds a system call on 64-bit platforms for switching between > little-endian and big-endian modes that is much faster than doing a > prctl call. This system call is handled as a special case right at > the start of the system call entry code, and because it is a special > case, it uses a system call number which is out of the range of > normal system calls, namely 0x1ebe. > > Measurements with lmbench on a 4.2GHz POWER6 showed no measurable > change in the speed of normal system calls with this patch. > > Switching endianness with this new system call takes around 60ns on a > 4.2GHz POWER6, compared with around 300ns to switch endian mode with a > prctl. This can provide a significant performance advantage for > emulators for little-endian architectures that want to switch between > big-endian and little-endian mode frequently, e.g. because they are > generating instructions sequences on the fly and they want to run > those sequences in little-endian mode. > > Signed-off-by: Paul Mackerras <paulus@samba.org> > --- > > diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S > index 215973a..2eb49a7 100644 > --- a/arch/powerpc/kernel/head_64.S > +++ b/arch/powerpc/kernel/head_64.S > @@ -239,6 +239,10 @@ instruction_access_slb_pSeries: > .globl system_call_pSeries > system_call_pSeries: > HMT_MEDIUM > +BEGIN_FTR_SECTION > + cmpdi r0,0x1ebe > + beq- 1f > +END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) Am I missing something here or does this add a branch for every normal syscall? > mr r9,r13 > mfmsr r10 > mfspr r13,SPRN_SPRG3 > @@ -253,6 +257,13 @@ system_call_pSeries: > rfid > b . /* prevent speculative execution */ > > +/* Fast LE/BE switch system call */ > +1: mfspr r12,SPRN_SRR1 > + xori r12,r12,MSR_LE > + mtspr SPRN_SRR1,r12 > + rfid /* return to userspace */ > + b . > + > STD_EXCEPTION_PSERIES(0xd00, single_step) > STD_EXCEPTION_PSERIES(0xe00, trap_0e) > > _______________________________________________ > Linuxppc-dev mailing list > Linuxppc-dev@ozlabs.org > https://ozlabs.org/mailman/listinfo/linuxppc-dev ---end quoted text--- ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-28 14:43 ` Christoph Hellwig 0 siblings, 0 replies; 19+ messages in thread From: Christoph Hellwig @ 2008-04-28 14:43 UTC (permalink / raw) To: Paul Mackerras; +Cc: linuxppc-dev, mtk.manpages, linux-arch Please see Michael Kerrisk on userspace ABI updates. A nice little manpage for this gimmick would be helpful, and maybe help other platforms that want one aswell to implement the same API. On Mon, Apr 28, 2008 at 01:52:31PM +1000, Paul Mackerras wrote: > This adds a system call on 64-bit platforms for switching between > little-endian and big-endian modes that is much faster than doing a > prctl call. This system call is handled as a special case right at > the start of the system call entry code, and because it is a special > case, it uses a system call number which is out of the range of > normal system calls, namely 0x1ebe. > > Measurements with lmbench on a 4.2GHz POWER6 showed no measurable > change in the speed of normal system calls with this patch. > > Switching endianness with this new system call takes around 60ns on a > 4.2GHz POWER6, compared with around 300ns to switch endian mode with a > prctl. This can provide a significant performance advantage for > emulators for little-endian architectures that want to switch between > big-endian and little-endian mode frequently, e.g. because they are > generating instructions sequences on the fly and they want to run > those sequences in little-endian mode. > > Signed-off-by: Paul Mackerras <paulus@samba.org> > --- > > diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S > index 215973a..2eb49a7 100644 > --- a/arch/powerpc/kernel/head_64.S > +++ b/arch/powerpc/kernel/head_64.S > @@ -239,6 +239,10 @@ instruction_access_slb_pSeries: > .globl system_call_pSeries > system_call_pSeries: > HMT_MEDIUM > +BEGIN_FTR_SECTION > + cmpdi r0,0x1ebe > + beq- 1f > +END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) Am I missing something here or does this add a branch for every normal syscall? > mr r9,r13 > mfmsr r10 > mfspr r13,SPRN_SPRG3 > @@ -253,6 +257,13 @@ system_call_pSeries: > rfid > b . /* prevent speculative execution */ > > +/* Fast LE/BE switch system call */ > +1: mfspr r12,SPRN_SRR1 > + xori r12,r12,MSR_LE > + mtspr SPRN_SRR1,r12 > + rfid /* return to userspace */ > + b . > + > STD_EXCEPTION_PSERIES(0xd00, single_step) > STD_EXCEPTION_PSERIES(0xe00, trap_0e) > > _______________________________________________ > Linuxppc-dev mailing list > Linuxppc-dev@ozlabs.org > https://ozlabs.org/mailman/listinfo/linuxppc-dev ---end quoted text--- ^ permalink raw reply [flat|nested] 19+ messages in thread
[parent not found: <20080428144332.GA17109-jcswGhMUV9g@public.gmane.org>]
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-28 15:42 ` Michael Kerrisk 0 siblings, 0 replies; 19+ messages in thread From: Michael Kerrisk @ 2008-04-28 15:42 UTC (permalink / raw) To: Christoph Hellwig Cc: Paul Mackerras, linuxppc-dev-mnsaURCQ41sdnm+yROfE0A, mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w, linux-arch-u79uwXL29TY76Z2rM5mHXA On Mon, Apr 28, 2008 at 4:43 PM, Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org> wrote: > Please see Michael Kerrisk on userspace ABI updates. A nice little > manpage for this gimmick would be helpful, and maybe help other > platforms that want one aswell to implement the same API. Thanks Chrsitoph. I'm not on any of these lists at the moment. Paul -- is this syscall defintely going in? Could you write a short description for userland programmers? I'll do the grotty *roff stuff. > On Mon, Apr 28, 2008 at 01:52:31PM +1000, Paul Mackerras wrote: > > This adds a system call on 64-bit platforms for switching between > > little-endian and big-endian modes that is much faster than doing a > > prctl call. This system call is handled as a special case right at > > the start of the system call entry code, and because it is a special > > case, it uses a system call number which is out of the range of > > normal system calls, namely 0x1ebe. > > > > Measurements with lmbench on a 4.2GHz POWER6 showed no measurable > > change in the speed of normal system calls with this patch. > > > > Switching endianness with this new system call takes around 60ns on a > > 4.2GHz POWER6, compared with around 300ns to switch endian mode with a > > prctl. This can provide a significant performance advantage for > > emulators for little-endian architectures that want to switch between > > big-endian and little-endian mode frequently, e.g. because they are > > generating instructions sequences on the fly and they want to run > > those sequences in little-endian mode. > > > > Signed-off-by: Paul Mackerras <paulus-eUNUBHrolfbYtjvyW6yDsg@public.gmane.org> > > --- > > > > diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S > > index 215973a..2eb49a7 100644 > > --- a/arch/powerpc/kernel/head_64.S > > +++ b/arch/powerpc/kernel/head_64.S > > @@ -239,6 +239,10 @@ instruction_access_slb_pSeries: > > .globl system_call_pSeries > > system_call_pSeries: > > HMT_MEDIUM > > +BEGIN_FTR_SECTION > > + cmpdi r0,0x1ebe > > + beq- 1f > > +END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) > > Am I missing something here or does this add a branch for every normal > syscall? > > > mr r9,r13 > > mfmsr r10 > > mfspr r13,SPRN_SPRG3 > > @@ -253,6 +257,13 @@ system_call_pSeries: > > rfid > > b . /* prevent speculative execution */ > > > > +/* Fast LE/BE switch system call */ > > +1: mfspr r12,SPRN_SRR1 > > + xori r12,r12,MSR_LE > > + mtspr SPRN_SRR1,r12 > > + rfid /* return to userspace */ > > + b . > > + > > STD_EXCEPTION_PSERIES(0xd00, single_step) > > STD_EXCEPTION_PSERIES(0xe00, trap_0e) > > > > _______________________________________________ > > Linuxppc-dev mailing list > > Linuxppc-dev-mnsaURCQ41sdnm+yROfE0A@public.gmane.org > > https://ozlabs.org/mailman/listinfo/linuxppc-dev > ---end quoted text--- > -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Found a bug? http://www.kernel.org/doc/man-pages/reporting_bugs.html ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-28 15:42 ` Michael Kerrisk 0 siblings, 0 replies; 19+ messages in thread From: Michael Kerrisk @ 2008-04-28 15:42 UTC (permalink / raw) To: Christoph Hellwig; +Cc: linux-arch, linuxppc-dev, Paul Mackerras, mtk.manpages On Mon, Apr 28, 2008 at 4:43 PM, Christoph Hellwig <hch@lst.de> wrote: > Please see Michael Kerrisk on userspace ABI updates. A nice little > manpage for this gimmick would be helpful, and maybe help other > platforms that want one aswell to implement the same API. Thanks Chrsitoph. I'm not on any of these lists at the moment. Paul -- is this syscall defintely going in? Could you write a short description for userland programmers? I'll do the grotty *roff stuff. > On Mon, Apr 28, 2008 at 01:52:31PM +1000, Paul Mackerras wrote: > > This adds a system call on 64-bit platforms for switching between > > little-endian and big-endian modes that is much faster than doing a > > prctl call. This system call is handled as a special case right at > > the start of the system call entry code, and because it is a special > > case, it uses a system call number which is out of the range of > > normal system calls, namely 0x1ebe. > > > > Measurements with lmbench on a 4.2GHz POWER6 showed no measurable > > change in the speed of normal system calls with this patch. > > > > Switching endianness with this new system call takes around 60ns on a > > 4.2GHz POWER6, compared with around 300ns to switch endian mode with a > > prctl. This can provide a significant performance advantage for > > emulators for little-endian architectures that want to switch between > > big-endian and little-endian mode frequently, e.g. because they are > > generating instructions sequences on the fly and they want to run > > those sequences in little-endian mode. > > > > Signed-off-by: Paul Mackerras <paulus@samba.org> > > --- > > > > diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S > > index 215973a..2eb49a7 100644 > > --- a/arch/powerpc/kernel/head_64.S > > +++ b/arch/powerpc/kernel/head_64.S > > @@ -239,6 +239,10 @@ instruction_access_slb_pSeries: > > .globl system_call_pSeries > > system_call_pSeries: > > HMT_MEDIUM > > +BEGIN_FTR_SECTION > > + cmpdi r0,0x1ebe > > + beq- 1f > > +END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) > > Am I missing something here or does this add a branch for every normal > syscall? > > > mr r9,r13 > > mfmsr r10 > > mfspr r13,SPRN_SPRG3 > > @@ -253,6 +257,13 @@ system_call_pSeries: > > rfid > > b . /* prevent speculative execution */ > > > > +/* Fast LE/BE switch system call */ > > +1: mfspr r12,SPRN_SRR1 > > + xori r12,r12,MSR_LE > > + mtspr SPRN_SRR1,r12 > > + rfid /* return to userspace */ > > + b . > > + > > STD_EXCEPTION_PSERIES(0xd00, single_step) > > STD_EXCEPTION_PSERIES(0xe00, trap_0e) > > > > _______________________________________________ > > Linuxppc-dev mailing list > > Linuxppc-dev@ozlabs.org > > https://ozlabs.org/mailman/listinfo/linuxppc-dev > ---end quoted text--- > -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Found a bug? http://www.kernel.org/doc/man-pages/reporting_bugs.html ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-28 15:42 ` Michael Kerrisk 0 siblings, 0 replies; 19+ messages in thread From: Michael Kerrisk @ 2008-04-28 15:42 UTC (permalink / raw) To: Christoph Hellwig; +Cc: Paul Mackerras, linuxppc-dev, mtk.manpages, linux-arch On Mon, Apr 28, 2008 at 4:43 PM, Christoph Hellwig <hch@lst.de> wrote: > Please see Michael Kerrisk on userspace ABI updates. A nice little > manpage for this gimmick would be helpful, and maybe help other > platforms that want one aswell to implement the same API. Thanks Chrsitoph. I'm not on any of these lists at the moment. Paul -- is this syscall defintely going in? Could you write a short description for userland programmers? I'll do the grotty *roff stuff. > On Mon, Apr 28, 2008 at 01:52:31PM +1000, Paul Mackerras wrote: > > This adds a system call on 64-bit platforms for switching between > > little-endian and big-endian modes that is much faster than doing a > > prctl call. This system call is handled as a special case right at > > the start of the system call entry code, and because it is a special > > case, it uses a system call number which is out of the range of > > normal system calls, namely 0x1ebe. > > > > Measurements with lmbench on a 4.2GHz POWER6 showed no measurable > > change in the speed of normal system calls with this patch. > > > > Switching endianness with this new system call takes around 60ns on a > > 4.2GHz POWER6, compared with around 300ns to switch endian mode with a > > prctl. This can provide a significant performance advantage for > > emulators for little-endian architectures that want to switch between > > big-endian and little-endian mode frequently, e.g. because they are > > generating instructions sequences on the fly and they want to run > > those sequences in little-endian mode. > > > > Signed-off-by: Paul Mackerras <paulus@samba.org> > > --- > > > > diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S > > index 215973a..2eb49a7 100644 > > --- a/arch/powerpc/kernel/head_64.S > > +++ b/arch/powerpc/kernel/head_64.S > > @@ -239,6 +239,10 @@ instruction_access_slb_pSeries: > > .globl system_call_pSeries > > system_call_pSeries: > > HMT_MEDIUM > > +BEGIN_FTR_SECTION > > + cmpdi r0,0x1ebe > > + beq- 1f > > +END_FTR_SECTION_IFSET(CPU_FTR_REAL_LE) > > Am I missing something here or does this add a branch for every normal > syscall? > > > mr r9,r13 > > mfmsr r10 > > mfspr r13,SPRN_SPRG3 > > @@ -253,6 +257,13 @@ system_call_pSeries: > > rfid > > b . /* prevent speculative execution */ > > > > +/* Fast LE/BE switch system call */ > > +1: mfspr r12,SPRN_SRR1 > > + xori r12,r12,MSR_LE > > + mtspr SPRN_SRR1,r12 > > + rfid /* return to userspace */ > > + b . > > + > > STD_EXCEPTION_PSERIES(0xd00, single_step) > > STD_EXCEPTION_PSERIES(0xe00, trap_0e) > > > > _______________________________________________ > > Linuxppc-dev mailing list > > Linuxppc-dev@ozlabs.org > > https://ozlabs.org/mailman/listinfo/linuxppc-dev > ---end quoted text--- > -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Found a bug? http://www.kernel.org/doc/man-pages/reporting_bugs.html ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 2:46 ` Paul Mackerras 0 siblings, 0 replies; 19+ messages in thread From: Paul Mackerras @ 2008-04-29 2:46 UTC (permalink / raw) To: Christoph Hellwig Cc: linuxppc-dev-mnsaURCQ41sdnm+yROfE0A, mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w, linux-arch-u79uwXL29TY76Z2rM5mHXA Christoph Hellwig writes: > Am I missing something here or does this add a branch for every normal > syscall? It does, but the impact is so small as to be unmeasurable with lmbench, even on the null syscall measurement. The overhead of the easily-predicted not-taken branch is completely swamped by the amount of time that the sc and rfid instructions take. I had it under a config option at one point but then decided not to bother with that when I couldn't measure any difference. Paul. ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 2:46 ` Paul Mackerras 0 siblings, 0 replies; 19+ messages in thread From: Paul Mackerras @ 2008-04-29 2:46 UTC (permalink / raw) To: Christoph Hellwig; +Cc: linux-arch, linuxppc-dev, mtk.manpages Christoph Hellwig writes: > Am I missing something here or does this add a branch for every normal > syscall? It does, but the impact is so small as to be unmeasurable with lmbench, even on the null syscall measurement. The overhead of the easily-predicted not-taken branch is completely swamped by the amount of time that the sc and rfid instructions take. I had it under a config option at one point but then decided not to bother with that when I couldn't measure any difference. Paul. ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 2:46 ` Paul Mackerras 0 siblings, 0 replies; 19+ messages in thread From: Paul Mackerras @ 2008-04-29 2:46 UTC (permalink / raw) To: Christoph Hellwig; +Cc: linuxppc-dev, mtk.manpages, linux-arch Christoph Hellwig writes: > Am I missing something here or does this add a branch for every normal > syscall? It does, but the impact is so small as to be unmeasurable with lmbench, even on the null syscall measurement. The overhead of the easily-predicted not-taken branch is completely swamped by the amount of time that the sc and rfid instructions take. I had it under a config option at one point but then decided not to bother with that when I couldn't measure any difference. Paul. ^ permalink raw reply [flat|nested] 19+ messages in thread
[parent not found: <18454.35824.527711.355488-UYQwCShxghk5kJ7NmlRacFaTQe2KTcn/@public.gmane.org>]
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 18:40 ` Wolfgang Denk 0 siblings, 0 replies; 19+ messages in thread From: Wolfgang Denk @ 2008-04-29 18:40 UTC (permalink / raw) To: Paul Mackerras Cc: Christoph Hellwig, linux-arch-u79uwXL29TY76Z2rM5mHXA, linuxppc-dev-mnsaURCQ41sdnm+yROfE0A, mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w In message <18454.35824.527711.355488-UYQwCShxghk5kJ7NmlRacFaTQe2KTcn/@public.gmane.org> you wrote: > > > Am I missing something here or does this add a branch for every normal > > syscall? > > It does, but the impact is so small as to be unmeasurable with > lmbench, even on the null syscall measurement. The overhead of the > easily-predicted not-taken branch is completely swamped by the amount > of time that the sc and rfid instructions take. I had it under a > config option at one point but then decided not to bother with that > when I couldn't measure any difference. This probably depends a bit on the performance of the system in question. Did you measure it - for example - on a 50 MHz MPC850 ? Best regards, Wolfgang Denk -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd-ynQEQJNshbs@public.gmane.org To get something done, a committee should consist of no more than three men, two of them absent. ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 18:40 ` Wolfgang Denk 0 siblings, 0 replies; 19+ messages in thread From: Wolfgang Denk @ 2008-04-29 18:40 UTC (permalink / raw) To: Paul Mackerras; +Cc: linux-arch, linuxppc-dev, Christoph Hellwig, mtk.manpages In message <18454.35824.527711.355488@cargo.ozlabs.ibm.com> you wrote: > > > Am I missing something here or does this add a branch for every normal > > syscall? > > It does, but the impact is so small as to be unmeasurable with > lmbench, even on the null syscall measurement. The overhead of the > easily-predicted not-taken branch is completely swamped by the amount > of time that the sc and rfid instructions take. I had it under a > config option at one point but then decided not to bother with that > when I couldn't measure any difference. This probably depends a bit on the performance of the system in question. Did you measure it - for example - on a 50 MHz MPC850 ? Best regards, Wolfgang Denk -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de To get something done, a committee should consist of no more than three men, two of them absent. ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 18:40 ` Wolfgang Denk 0 siblings, 0 replies; 19+ messages in thread From: Wolfgang Denk @ 2008-04-29 18:40 UTC (permalink / raw) To: Paul Mackerras; +Cc: Christoph Hellwig, linux-arch, linuxppc-dev, mtk.manpages In message <18454.35824.527711.355488@cargo.ozlabs.ibm.com> you wrote: > > > Am I missing something here or does this add a branch for every normal > > syscall? > > It does, but the impact is so small as to be unmeasurable with > lmbench, even on the null syscall measurement. The overhead of the > easily-predicted not-taken branch is completely swamped by the amount > of time that the sc and rfid instructions take. I had it under a > config option at one point but then decided not to bother with that > when I couldn't measure any difference. This probably depends a bit on the performance of the system in question. Did you measure it - for example - on a 50 MHz MPC850 ? Best regards, Wolfgang Denk -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de To get something done, a committee should consist of no more than three men, two of them absent. ^ permalink raw reply [flat|nested] 19+ messages in thread
[parent not found: <20080429184047.47CB7247B4-C2Gvrrd9BC/j/ljBK/0BTg@public.gmane.org>]
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 18:46 ` Christoph Hellwig 0 siblings, 0 replies; 19+ messages in thread From: Christoph Hellwig @ 2008-04-29 18:46 UTC (permalink / raw) To: Wolfgang Denk Cc: Paul Mackerras, Christoph Hellwig, linux-arch-u79uwXL29TY76Z2rM5mHXA, linuxppc-dev-mnsaURCQ41sdnm+yROfE0A, mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w On Tue, Apr 29, 2008 at 08:40:47PM +0200, Wolfgang Denk wrote: > This probably depends a bit on the performance of the system in > question. Did you measure it - for example - on a 50 MHz MPC850 ? You got a 64bit kernel to run on a MPC850? wow :) Not sure what the slowest supported 64bit cpu is (RS64-II?), but Paul might be right and in this case it really doesn't matter. ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 18:46 ` Christoph Hellwig 0 siblings, 0 replies; 19+ messages in thread From: Christoph Hellwig @ 2008-04-29 18:46 UTC (permalink / raw) To: Wolfgang Denk Cc: linux-arch, linuxppc-dev, Paul Mackerras, Christoph Hellwig, mtk.manpages On Tue, Apr 29, 2008 at 08:40:47PM +0200, Wolfgang Denk wrote: > This probably depends a bit on the performance of the system in > question. Did you measure it - for example - on a 50 MHz MPC850 ? You got a 64bit kernel to run on a MPC850? wow :) Not sure what the slowest supported 64bit cpu is (RS64-II?), but Paul might be right and in this case it really doesn't matter. ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 18:46 ` Christoph Hellwig 0 siblings, 0 replies; 19+ messages in thread From: Christoph Hellwig @ 2008-04-29 18:46 UTC (permalink / raw) To: Wolfgang Denk Cc: Paul Mackerras, Christoph Hellwig, linux-arch, linuxppc-dev, mtk.manpages On Tue, Apr 29, 2008 at 08:40:47PM +0200, Wolfgang Denk wrote: > This probably depends a bit on the performance of the system in > question. Did you measure it - for example - on a 50 MHz MPC850 ? You got a 64bit kernel to run on a MPC850? wow :) Not sure what the slowest supported 64bit cpu is (RS64-II?), but Paul might be right and in this case it really doesn't matter. ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 21:16 ` Paul Mackerras 0 siblings, 0 replies; 19+ messages in thread From: Paul Mackerras @ 2008-04-29 21:16 UTC (permalink / raw) To: Wolfgang Denk Cc: Christoph Hellwig, linux-arch-u79uwXL29TY76Z2rM5mHXA, linuxppc-dev-mnsaURCQ41sdnm+yROfE0A, mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w Wolfgang Denk writes: > This probably depends a bit on the performance of the system in > question. Did you measure it - for example - on a 50 MHz MPC850 ? The patch only affects arch/powerpc/kernel/entry_64.S. So no, I didn't measure it on a 50MHz MPC850, or indeed any 32-bit system. :) Paul. ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 21:16 ` Paul Mackerras 0 siblings, 0 replies; 19+ messages in thread From: Paul Mackerras @ 2008-04-29 21:16 UTC (permalink / raw) To: Wolfgang Denk; +Cc: linux-arch, linuxppc-dev, Christoph Hellwig, mtk.manpages Wolfgang Denk writes: > This probably depends a bit on the performance of the system in > question. Did you measure it - for example - on a 50 MHz MPC850 ? The patch only affects arch/powerpc/kernel/entry_64.S. So no, I didn't measure it on a 50MHz MPC850, or indeed any 32-bit system. :) Paul. ^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] Add fast little-endian switch system call @ 2008-04-29 21:16 ` Paul Mackerras 0 siblings, 0 replies; 19+ messages in thread From: Paul Mackerras @ 2008-04-29 21:16 UTC (permalink / raw) To: Wolfgang Denk; +Cc: Christoph Hellwig, linux-arch, linuxppc-dev, mtk.manpages Wolfgang Denk writes: > This probably depends a bit on the performance of the system in > question. Did you measure it - for example - on a 50 MHz MPC850 ? The patch only affects arch/powerpc/kernel/entry_64.S. So no, I didn't measure it on a 50MHz MPC850, or indeed any 32-bit system. :) Paul. ^ permalink raw reply [flat|nested] 19+ messages in thread
end of thread, other threads:[~2008-04-29 21:16 UTC | newest]
Thread overview: 19+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-04-28 3:52 [PATCH] Add fast little-endian switch system call Paul Mackerras
[not found] ` <18453.18943.906693.409245-UYQwCShxghk5kJ7NmlRacFaTQe2KTcn/@public.gmane.org>
2008-04-28 14:43 ` Christoph Hellwig
2008-04-28 14:43 ` Christoph Hellwig
2008-04-28 14:43 ` Christoph Hellwig
[not found] ` <20080428144332.GA17109-jcswGhMUV9g@public.gmane.org>
2008-04-28 15:42 ` Michael Kerrisk
2008-04-28 15:42 ` Michael Kerrisk
2008-04-28 15:42 ` Michael Kerrisk
2008-04-29 2:46 ` Paul Mackerras
2008-04-29 2:46 ` Paul Mackerras
2008-04-29 2:46 ` Paul Mackerras
[not found] ` <18454.35824.527711.355488-UYQwCShxghk5kJ7NmlRacFaTQe2KTcn/@public.gmane.org>
2008-04-29 18:40 ` Wolfgang Denk
2008-04-29 18:40 ` Wolfgang Denk
2008-04-29 18:40 ` Wolfgang Denk
[not found] ` <20080429184047.47CB7247B4-C2Gvrrd9BC/j/ljBK/0BTg@public.gmane.org>
2008-04-29 18:46 ` Christoph Hellwig
2008-04-29 18:46 ` Christoph Hellwig
2008-04-29 18:46 ` Christoph Hellwig
2008-04-29 21:16 ` Paul Mackerras
2008-04-29 21:16 ` Paul Mackerras
2008-04-29 21:16 ` Paul Mackerras
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.