Hi,
> This enables us to share the same page table code for
> both radix and hash. Radix use a hardware defined big endian
> page table
This is measurably worse (a little over 2% on POWER8) on a futex
microbenchmark:
#define _GNU_SOURCE
#include
#include
#include
#define ITERATIONS 1000
From: Anton Blanchard
In both __giveup_fpu() and __giveup_altivec() we make two modifications
to tsk->thread.regs->msr. gcc decides to do a read/modify/write of
each change, so we end up with a load hit store:
ld r9,264(r10)
rldicl r9,r9,50,1
rotldi r9,r9,14
From: Anton Blanchard
In setup_sigcontext(), we set current->thread.vrsave then use it
straight after. Since current is hidden from the compiler via inline
assembly, it cannot optimise this and we end up with a load hit store.
Fix this by using a temporary.
Signed-off-by: Anton Blanchard
---
From: Anton Blanchard
In many cases we disable interrupts right before calling
find_linux_pte_or_hugepte().
find_linux_pte_or_hugepte() first checks interrupts are disabled
before calling __find_linux_pte_or_hugepte():
if (!arch_irqs_disabled()) {
pr_info("%s called with
On Sun, 2016-05-29 at 21:03 +1000, Anton Blanchard wrote:
> Hi,
>
> >
> > This enables us to share the same page table code for
> > both radix and hash. Radix use a hardware defined big endian
> > page table
> This is measurably worse (a little over 2% on POWER8) on a futex
> microbenchmark:
Tha
On Sun, 2016-05-29 at 22:03 +1000, Anton Blanchard wrote:
> From: Anton Blanchard
>
> In setup_sigcontext(), we set current->thread.vrsave then use it
> straight after. Since current is hidden from the compiler via inline
> assembly, it cannot optimise this and we end up with a load hit store.
I
Hi,
> On Sun, 2016-05-29 at 22:03 +1000, Anton Blanchard wrote:
> > From: Anton Blanchard
> >
> > In setup_sigcontext(), we set current->thread.vrsave then use it
> > straight after. Since current is hidden from the compiler via inline
> > assembly, it cannot optimise this and we end up with a l
Hi,
On 03/05/2016 at 12:05:34 +0200, Arnd Bergmann wrote :
> On Tuesday 03 May 2016 09:24:18 Alexandre Belloni wrote:
> > Hi Arnd,
> >
> > I see you didn't copy Greg on that series (that may explain his
> > confusion on the previous patch), do you expect me to take it
> > through the RTC tree? Th
Hi Ben,
> That is surprising, do we have any idea what specifically increases
> the overhead so significantly ? Does gcc know about ldbrx/stdbrx ? I
> notice in our io.h for example we still do manual ld/std + swap
> because old processors didn't know these, we should fix that for
> CONFIG_POWER8
On Mon, 2016-05-30 at 09:08 +1000, Anton Blanchard via Linuxppc-dev wrote:
> > That is surprising, do we have any idea what specifically increases
> > the overhead so significantly ? Does gcc know about ldbrx/stdbrx ? I
> > notice in our io.h for example we still do manual ld/std + swap
> > because
Anton Blanchard writes:
> From: Anton Blanchard
>
> In many cases we disable interrupts right before calling
> find_linux_pte_or_hugepte().
>
> find_linux_pte_or_hugepte() first checks interrupts are disabled
> before calling __find_linux_pte_or_hugepte():
>
> if (!arch_irqs_disabled())
MPIC was only used by Power3 which is now unsupported, so drop support
for MPIC. XICS is now the only supported interrupt controller for
pSeries so make the XICS functions generic.
Signed-off-by: Rashmica Gupta
---
arch/powerpc/platforms/pseries/setup.c | 77 +++---
1
MPIC was only used by Power3 which is now unsupported, so remove MPIC
code. XICS is now the only supported interrupt controller for
pSeries so do some cleanups too.
Signed-off-by: Rashmica Gupta
---
arch/powerpc/platforms/pseries/pseries.h | 6 ++
arch/powerpc/platforms/pseries/setup.c |
MPIC was only used by Power3 which is now unsupported, so remove MPIC
code. XICS is now the only supported interrupt controller for
pSeries so do some cleanups too.
Signed-off-by: Rashmica Gupta
---
arch/powerpc/platforms/pseries/kexec.c | 23 +--
arch/powerpc/platforms/pse
MPIC was only used by Power3 which is now unsupported, so remove MPIC
code.
Signed-off-by: Rashmica Gupta
---
arch/powerpc/platforms/pseries/hotplug-cpu.c | 13 -
1 file changed, 13 deletions(-)
diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c
b/arch/powerpc/platforms/pser
MPIC was only used by Power3 which is now unsupported, so remove MPIC
code. XICS is now the only supported interrupt controller for
pSeries so do some cleanups too.
Signed-off-by: Rashmica Gupta
---
arch/powerpc/platforms/pseries/event_sources.c | 53 +++---
1 file changed, 1
16 matches
Mail list logo