[Zesty,1/1] powerpc/perf: Avoid spurious PMU interrupts after idle

Submitted by Rodrigo R. Galvao on Aug. 9, 2017, 1:44 p.m.

Details

Message ID 1502286296-27932-1-git-send-email-rosattig@linux.vnet.ibm.com
State New
Headers show

Commit Message

Rodrigo R. Galvao Aug. 9, 2017, 1:44 p.m.
From: Nicholas Piggin <npiggin@gmail.com>

BugLink: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1709352

POWER9 DD2 can see spurious PMU interrupts after state-loss idle in
some conditions.

A solution is to save and reload MMCR0 over state-loss idle.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Acked-by: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
Tested-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
(cherry picked from commit 101dd590a7fa37954540cf3149a1c502c0acc524)
Signed-off-by: Rodrigo R. Galvao <rosattig@linux.vnet.ibm.com>
---
 arch/powerpc/kernel/idle_book3s.S | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

Comments

Seth Forshee Aug. 11, 2017, 3:31 p.m.
On Wed, Aug 09, 2017 at 10:44:56AM -0300, Rodrigo R. Galvao wrote:
> From: Nicholas Piggin <npiggin@gmail.com>
> 
> BugLink: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1709352
> 
> POWER9 DD2 can see spurious PMU interrupts after state-loss idle in
> some conditions.
> 
> A solution is to save and reload MMCR0 over state-loss idle.
> 
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> Acked-by: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
> Tested-by: Anton Blanchard <anton@samba.org>
> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> (cherry picked from commit 101dd590a7fa37954540cf3149a1c502c0acc524)
> Signed-off-by: Rodrigo R. Galvao <rosattig@linux.vnet.ibm.com>

This patch had some context differences when I tried to apply it to
4.12 so I cherry picked it directly. That makes me wonder whether you
had to fix it up for zesty though; if so the "cherry picked" should be
changed to "backported".

Otherwise the patch looks fine to me.

Acked-by: Seth Forshee <seth.forshee@canonical.com>
Kleber Souza Aug. 15, 2017, 12:24 p.m.
On 08/09/17 15:44, Rodrigo R. Galvao wrote:
> From: Nicholas Piggin <npiggin@gmail.com>
> 
> BugLink: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1709352
> 
> POWER9 DD2 can see spurious PMU interrupts after state-loss idle in
> some conditions.
> 
> A solution is to save and reload MMCR0 over state-loss idle.
> 
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> Acked-by: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
> Tested-by: Anton Blanchard <anton@samba.org>
> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> (cherry picked from commit 101dd590a7fa37954540cf3149a1c502c0acc524)
> Signed-off-by: Rodrigo R. Galvao <rosattig@linux.vnet.ibm.com>
> ---
>  arch/powerpc/kernel/idle_book3s.S | 15 ++++++++++++++-
>  1 file changed, 14 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/powerpc/kernel/idle_book3s.S b/arch/powerpc/kernel/idle_book3s.S
> index fb762ee..e003fb9 100644
> --- a/arch/powerpc/kernel/idle_book3s.S
> +++ b/arch/powerpc/kernel/idle_book3s.S
> @@ -29,6 +29,7 @@
>   * Use unused space in the interrupt stack to save and restore
>   * registers for winkle support.
>   */
> +#define _MMCR0	GPR0
>  #define _SDR1	GPR3
>  #define _RPR	GPR4
>  #define _SPURR	GPR5
> @@ -282,6 +283,14 @@ power_enter_stop:
>  	b 	pnv_wakeup_noloss
>  
>  .Lhandle_esl_ec_set:
> +	/*
> +	 * POWER9 DD2 can incorrectly set PMAO when waking up after a
> +	 * state-loss idle. Saving and restoring MMCR0 over idle is a
> +	 * workaround.
> +	 */
> +	mfspr	r4,SPRN_MMCR0
> +	std	r4,_MMCR0(r1)
> +
>  /*
>   * Check if the requested state is a deep idle state.
>   */
> @@ -434,10 +443,14 @@ FTR_SECTION_ELSE_NESTED(70)
>  ALT_FTR_SECTION_END_NESTED_IFSET(CPU_FTR_POWER9_DD1, 70)
>  	/*
>  	 * Workaround for POWER9, if we lost resources, the ERAT
> -	 * might have been mixed up and needs flushing.
> +	 * might have been mixed up and needs flushing. We also need
> +	 * to reload MMCR0 (see comment above).
>  	 */
>  	blt	cr3,1f
>  	PPC_INVALIDATE_ERAT
> +	ld	r1,PACAR1(r13)
> +	ld	r4,_MMCR0(r1)
> +	mtspr	SPRN_MMCR0,r4
>  1:
>  	/*
>  	 * POWER ISA 3. Use PSSCR to determine if we
> 

Clean cherry-pick (applies cleanly to zesty/master-next), scope limited
to POWER9.

Acked-by: Kleber Sacilotto de Souza <kleber.souza@canonical.com>
Kleber Souza Aug. 23, 2017, 3:49 p.m.
Applied to zesty/master-next branch fixing the BugLink to the format
"http://bugs.launchpad.net/bugs/<bug-id>".

Thanks.

Patch hide | download patch | download mbox

diff --git a/arch/powerpc/kernel/idle_book3s.S b/arch/powerpc/kernel/idle_book3s.S
index fb762ee..e003fb9 100644
--- a/arch/powerpc/kernel/idle_book3s.S
+++ b/arch/powerpc/kernel/idle_book3s.S
@@ -29,6 +29,7 @@ 
  * Use unused space in the interrupt stack to save and restore
  * registers for winkle support.
  */
+#define _MMCR0	GPR0
 #define _SDR1	GPR3
 #define _RPR	GPR4
 #define _SPURR	GPR5
@@ -282,6 +283,14 @@  power_enter_stop:
 	b 	pnv_wakeup_noloss
 
 .Lhandle_esl_ec_set:
+	/*
+	 * POWER9 DD2 can incorrectly set PMAO when waking up after a
+	 * state-loss idle. Saving and restoring MMCR0 over idle is a
+	 * workaround.
+	 */
+	mfspr	r4,SPRN_MMCR0
+	std	r4,_MMCR0(r1)
+
 /*
  * Check if the requested state is a deep idle state.
  */
@@ -434,10 +443,14 @@  FTR_SECTION_ELSE_NESTED(70)
 ALT_FTR_SECTION_END_NESTED_IFSET(CPU_FTR_POWER9_DD1, 70)
 	/*
 	 * Workaround for POWER9, if we lost resources, the ERAT
-	 * might have been mixed up and needs flushing.
+	 * might have been mixed up and needs flushing. We also need
+	 * to reload MMCR0 (see comment above).
 	 */
 	blt	cr3,1f
 	PPC_INVALIDATE_ERAT
+	ld	r1,PACAR1(r13)
+	ld	r4,_MMCR0(r1)
+	mtspr	SPRN_MMCR0,r4
 1:
 	/*
 	 * POWER ISA 3. Use PSSCR to determine if we