Re: linux-next: build failure after merge of the tip tree

From: Mike Galbraith
Date: Sat Jan 18 2014 - 04:48:34 EST


On Fri, 2014-01-17 at 14:45 +1100, Stephen Rothwell wrote:
> Hi all,
>
> On Thu, 16 Jan 2014 14:51:08 -0800 "H. Peter Anvin" <hpa@xxxxxxxxx> wrote:
> >
> > On 01/16/2014 02:34 PM, Stephen Rothwell wrote:
> > >
> > > On Thu, 16 Jan 2014 23:25:36 +0100 Peter Zijlstra
> > > <peterz@xxxxxxxxxxxxx> wrote:
> > >>
> > >> On Fri, Jan 17, 2014 at 07:46:28AM +1100, Stephen Rothwell
> > >> wrote:
> > >>>
> > >>> On Thu, 16 Jan 2014 13:19:55 +0100 Peter Zijlstra
> > >>> <peterz@xxxxxxxxxxxxx> wrote:
> > >>>>
> > >>>> I think the below ought to work
> > >>>
> > >>> To be clear, all you did was replace the body of mwait_idle()
> > >>> with
> > >>>
> > >>> mwait_idle_with_hints(0, 0);
> > >>
> > >> Pretty much, and add the asm/mwait.h include, otherwise you'll
> > >> end up with a compile fail.
> > >>
> > >>> (and the comment above it)? I need to apply in incremental
> > >>> patch in the merge commit.
> > >>
> > >> I don't think I touched the comment at all.
> > >
> >
> > In retrospect this bit probably should have gone through the idle
> > tree. That was my bad, I need to coordinate with Len better.
>
> So this is what I added as a merge fix patch. Someone just needs to make
> sure Linus gets this when the latter of the tow trees gets merged.

I hope it doesn't look quite like that, next-20140117 is -ENOBOOT on
Q6600 box. See below for an alternative.

> From: Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx>
> Date: Fri, 17 Jan 2014 14:42:06 +1100
> Subject: [PATCH] x86 idle: mwait_idle merge update
>
> Signed-off-by: Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx>
> ---
> arch/x86/kernel/process.c | 14 ++------------
> 1 file changed, 2 insertions(+), 12 deletions(-)
>
> diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
> index db471a87fdd8..4da840f01561 100644
> --- a/arch/x86/kernel/process.c
> +++ b/arch/x86/kernel/process.c
> @@ -28,6 +28,7 @@
> #include <asm/fpu-internal.h>
> #include <asm/debugreg.h>
> #include <asm/nmi.h>
> +#include <asm/mwait.h>
>
> /*
> * per-CPU TSS segments. Threads are completely 'soft' on Linux,
> @@ -427,18 +428,7 @@ static int prefer_mwait_c1_over_halt(const struct cpuinfo_x86 *c)
>
> static void mwait_idle(void)
> {
> - if (!need_resched()) {
> - if (this_cpu_has(X86_FEATURE_CLFLUSH_MONITOR))
> - clflush((void *)&current_thread_info()->flags);
> -
> - __monitor((void *)&current_thread_info()->flags, 0, 0);
> - smp_mb();
> - if (!need_resched())
> - __sti_mwait(0, 0);
> - else
> - local_irq_enable();
> - } else
> - local_irq_enable();
> + mwait_idle_with_hints(0, 0);
> }
>
> void select_idle_routine(const struct cpuinfo_x86 *c)
> --
> 1.8.5.2

idle: kill unnecessary mwait_idle() resched IPIs

Set/clear polling instead.

Q6600, pipe-test scheduling cross core:
3.8.13 487.2 KHz 1.000
3.13.0-master 415.5 KHz .852
3.13.0-master+ 415.2 KHz .852 + restore mwait_idle
3.13.0-master++ 488.5 KHz 1.002 + restore mwait_idle + IPI fix
3.13.0-next-20140117 -ENOBOOT
3.13.0-next-20140117+ 531.4 KHz 1.090 + IPI fix

Signed-off-by: Mike Galbraith <bitbucket@xxxxxxxxx>
---
arch/x86/include/asm/processor.h | 8 ++++++++
arch/x86/kernel/process.c | 10 +++++++---
2 files changed, 15 insertions(+), 3 deletions(-)

--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -723,6 +723,14 @@ static inline void sync_core(void)
#endif
}

+static inline void __sti_mwait(unsigned long eax, unsigned long ecx)
+{
+ trace_hardirqs_on();
+ /* "mwait %eax, %ecx;" */
+ asm volatile("sti; .byte 0x0f, 0x01, 0xc9;"
+ :: "a" (eax), "c" (ecx));
+}
+
extern void select_idle_routine(const struct cpuinfo_x86 *c);
extern void init_amd_e400_c1e_mask(void);

--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -28,6 +28,7 @@
#include <asm/fpu-internal.h>
#include <asm/debugreg.h>
#include <asm/nmi.h>
+#include <asm/mwait.h>

/*
* per-CPU TSS segments. Threads are completely 'soft' on Linux,
@@ -427,18 +428,21 @@ static int prefer_mwait_c1_over_halt(con

static void mwait_idle(void)
{
- if (!need_resched()) {
- if (this_cpu_has(X86_FEATURE_CLFLUSH_MONITOR))
+ if (!current_set_polling_and_test()) {
+ if (static_cpu_has(X86_FEATURE_CLFLUSH_MONITOR)) {
+ mb();
clflush((void *)&current_thread_info()->flags);
+ mb();
+ }

__monitor((void *)&current_thread_info()->flags, 0, 0);
- smp_mb();
if (!need_resched())
__sti_mwait(0, 0);
else
local_irq_enable();
} else
local_irq_enable();
+ current_clr_polling();
}

void select_idle_routine(const struct cpuinfo_x86 *c)


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/