* Re: [PATCH] powerpc: Fix alignment of secondary cpu spin vars
From: Olof Johansson @ 2014-01-03 8:12 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: chzigotzky, linuxppc-dev, linux-kernel@vger.kernel.org,
Anton Blanchard
In-Reply-To: <CAOesGMgsY0R6KBXXH_ibTBYb1i_auWESwnoDkvi-esb58Pc+2A@mail.gmail.com>
On Thu, Jan 02, 2014 at 11:56:04PM -0800, Olof Johansson wrote:
> This makes things interesting though. The BE/LE trampoline code
> assumes at least 3 consecutive instructions. What was the reasoning
> behind entering the kernel LE instead of keeping the old boot protocol
> and just switching to LE once kernel is loaded? Is it actually used on
> some platforms or is this just a theoretical thing?
Actually, adding a little hack that zeroes out the memory once we're done
executing it will work just fine too. I know this is sort of icky, but maybe
it'll be good enough for now?
Of course, main worry is that this is just hiding some latent NULL deref in
the kernel now... :-/
-Olof
--- 8< --- 8< --- 8< --- 8< --- 8< --- 8< --- 8< --- 8< --- 8< ---
>From 4d003186cae546900cefc9e51b0ed4e65f775be1 Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Fri, 3 Jan 2014 00:09:28 -0800
Subject: [PATCH] powerpc: set some low memory contents to 0 early
The little-endian code adds some code path to __start, which essentially ends
up adding memory contents in low memory that didn't use to be there.
That seems to have triggered a latent bug, either in firmware or kernel, where
the 64-bit word located at physical address 8 needs to be 0.
The simple hack for this right now is to write it to 0 after we're done
executing it, which is what this patch does. Unfortunately I no longer
seem to have a working JTAG setup nor firmware sources, so debugging
this down to root cause might be more trouble than it's worth given the
relatively simple workaround.
Signed-off-by: Olof Johansson <olof@lixom.net>
---
arch/powerpc/kernel/head_64.S | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 2ae41ab..437d8bd 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -69,6 +69,13 @@ _GLOBAL(__start)
/* NOP this out unconditionally */
BEGIN_FTR_SECTION
FIXUP_ENDIAN
+ /* Hack for PWRficient platforms: Due to CFE(?) bug, the 64-bit
+ * word at 0x8 needs to be set to 0. Patch it up here once we're
+ * done executing it (we can be lazy and avoid invalidating
+ * icache)
+ */
+ li r0,0
+ std 0,8(0)
b .__start_initialization_multiplatform
END_FTR_SECTION(0, 1)
--
1.7.10.4
^ permalink raw reply related
* Re: [PATCH] powerpc: Fix alignment of secondary cpu spin vars
From: Olof Johansson @ 2014-01-03 7:56 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: Olof Johansson, chzigotzky, linuxppc-dev,
linux-kernel@vger.kernel.org, Anton Blanchard
In-Reply-To: <CAOesGMhHw-G-T7ZfNsVU92ftOucCNe9shNJtA7SWzqX=dQBQng@mail.gmail.com>
On Sat, Dec 28, 2013 at 1:05 PM, Olof Johansson <olof@lixom.net> wrote:
> Sigh, it's not this after all. I did a clean build with this applied
> and still see failures. Something else is (also?) going on here.
Ok, so after some more digging I actually think that this isn't about
the new code added as much as it is about having more code in low
memory.
Before, there were only two instuctions in __start:
b .__start_initialization_multiplatform
trap
Now, there's a whole bunch:
c000000000000000 <.__start>:
c000000000000000: 08 00 00 48 tdi 0,r0,72
c000000000000004: 48 00 00 24 b c000000000000028 <.__start+0x28>
c000000000000008: 05 00 9f 42 .long 0x5009f42
c00000000000000c: a6 02 48 7d lhzu r16,18557(r2)
c000000000000010: 1c 00 4a 39 mulli r0,r0,19001
c000000000000014: a6 00 60 7d lhzu r16,24701(0)
c000000000000018: 01 00 6b 69 .long 0x1006b69
c00000000000001c: a6 03 5a 7d lhzu r16,23165(r3)
c000000000000020: a6 03 7b 7d lhzu r16,31613(r3)
c000000000000024: 24 00 00 4c dozi r0,r0,76
c000000000000028: 48 00 95 84 b c0000000000095ac
<.__start_initialization_multiplatform>
c00000000000002c: 7f e0 00 08 trap
And indeed, by replacing some of the LE hand-converted code with 0x0,
it seems that what's really making things blow up here is that 0x8-0xc
contain something else than 0x0.
Where/why this comes from I'm less certain of -- and since I seem to
no longer have a usable JTAG setup, I can't break in and see where the
code gets stuck and call paths, etc. So it's pure speculation, but I'm
guessing it's a null pointer dereference somewhere with a chained
pointer as the second member in a struct, i.e. with NULL the stray
null ptr deref does no harm.
Since it doesn't seem to impact pSeries, there's a chance that the bug
is in firmware, not in the kernel, since this seems to happen during
fairly early boot, i.e. possibly while grabbing the DT contents out.
This makes things interesting though. The BE/LE trampoline code
assumes at least 3 consecutive instructions. What was the reasoning
behind entering the kernel LE instead of keeping the old boot protocol
and just switching to LE once kernel is loaded? Is it actually used on
some platforms or is this just a theoretical thing?
-Olof
^ permalink raw reply
* Re: [PATCH -V2] powerpc: thp: Fix crash on mremap
From: Aneesh Kumar K.V @ 2014-01-03 5:28 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: aarcange, linuxppc-dev, paulus, Kirill A. Shutemov, linux-mm
In-Reply-To: <1388665786.4373.48.camel@pasglop>
Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:
> On Thu, 2014-01-02 at 16:22 +0530, Aneesh Kumar K.V wrote:
>> > Just use config option directly:
>> >
>> > if (new_ptl != old_ptl ||
>> > IS_ENABLED(CONFIG_ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW))
>>
>>
>> I didn't like that. I found the earlier one easier for reading.
>> If you and others strongly feel about this, I can redo the patch.
>> Please let me know
>
> Yes, use IS_ENABLED, no need to have two indirections of #define's
>
> Another option is to have
>
> if (pmd_move_must_withdraw(new,old)) {
> }
>
> With in a generic header:
>
> #ifndef pmd_move_must_withdraw
> static inline bool pmd_move_must_withdraw(spinlock_t *new_ptl, ...)
> {
> return new_ptl != old_ptl;
> }
> #endif
>
> And in powerpc:
>
> static inline bool pmd_move_must_withdraw(spinlock_t *new_ptl, ...)
> {
> return true;
> }
> #define pmd_move_must_withdraw pmd_move_must_withdraw
This is better i guess. It is also in-line with rest of transparent
hugepage functions. I will do this.
-aneesh
^ permalink raw reply
* Re: commit e38c0a1f breaks powerpc boards with uli1575 chip
From: Scott Wood @ 2014-01-03 0:04 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: devicetree@vger.kernel.org, Arnd Bergmann, Dmitry Krivoschekov,
Nikita Yushchenko, Thierry Reding, linux-kernel, Rob Herring,
Grant Likely, linuxppc-dev, Alexey Lugovskoy
In-Reply-To: <1388373200.4373.25.camel@pasglop>
On Mon, 2013-12-30 at 14:13 +1100, Benjamin Herrenschmidt wrote:
> On Thu, 2013-12-19 at 08:42 +0400, Nikita Yushchenko wrote:
> > No, this does not help.
> >
> > I've dumped the actual content of 'range' and 'addr' at the failure
> > point
> > (i.e. ar point that returns error with e38c0a1f but passes without
> > e38c0a1f ):
> >
> > OF: default map, cp=0, s=10000, da=70
> > range: 01 00 00 00 00 00 00 00 00 00 00 00
> > addr: 00 00 00 00 00 00 00 00 00 00 00 70
>
> Something that has a #address-cells larger than 2, or more generally,
> an address field that contains more than a single number, must have
> a specific translation backend, like we have for PCI.
>
> This is a bit annoying but originates from the original OFW stuff on
> which this stuff is based where the bus node would provide the methods
> for translation.
I can maybe see that for PCI which has a special encoding, but why is it
always needed? E.g. if Freescale localbus had a 64-bit offset instead
of 32-bit, the child nodes would have 3 address cells, but
straightforward use of ranges would bring it down to 2 for the final
physical address. Existing localbus nodes already have "an address
field that contains more than a single number"; it's just a simple
enough encoding that it works to treat it as if it were a single large
number.
-Scott
^ permalink raw reply
* [PATCH] powerpc/fsl-booke: Use SPRN_SPRGn rather than mfsprg/mtsprg
From: Scott Wood @ 2014-01-02 22:37 UTC (permalink / raw)
To: Benjamin Herrenschmidt
Cc: Scott Wood, Anton Vorontsov, linuxppc-dev, Dongsheng Wang
This fixes a build break that was probably introduced with the removal
of -Wa,-me500 (commit f49596a4cf4753d13951608f24f939a59fdcc653), where
the assembler refuses to recognize SPRG4-7 with a generic PPC target.
Signed-off-by: Scott Wood <scottwood@freescale.com>
Cc: Dongsheng Wang <dongsheng.wang@freescale.com>
Cc: Anton Vorontsov <avorontsov@mvista.com>
---
Dongsheng, please test.
---
arch/powerpc/kernel/swsusp_booke.S | 32 ++++++++++++++++----------------
1 file changed, 16 insertions(+), 16 deletions(-)
diff --git a/arch/powerpc/kernel/swsusp_booke.S b/arch/powerpc/kernel/swsusp_booke.S
index 0f20405..553c140 100644
--- a/arch/powerpc/kernel/swsusp_booke.S
+++ b/arch/powerpc/kernel/swsusp_booke.S
@@ -74,21 +74,21 @@ _GLOBAL(swsusp_arch_suspend)
bne 1b
/* Save SPRGs */
- mfsprg r4,0
+ mfspr r4,SPRN_SPRG0
stw r4,SL_SPRG0(r11)
- mfsprg r4,1
+ mfspr r4,SPRN_SPRG1
stw r4,SL_SPRG1(r11)
- mfsprg r4,2
+ mfspr r4,SPRN_SPRG2
stw r4,SL_SPRG2(r11)
- mfsprg r4,3
+ mfspr r4,SPRN_SPRG3
stw r4,SL_SPRG3(r11)
- mfsprg r4,4
+ mfspr r4,SPRN_SPRG4
stw r4,SL_SPRG4(r11)
- mfsprg r4,5
+ mfspr r4,SPRN_SPRG5
stw r4,SL_SPRG5(r11)
- mfsprg r4,6
+ mfspr r4,SPRN_SPRG6
stw r4,SL_SPRG6(r11)
- mfsprg r4,7
+ mfspr r4,SPRN_SPRG7
stw r4,SL_SPRG7(r11)
/* Call the low level suspend stuff (we should probably have made
@@ -150,21 +150,21 @@ _GLOBAL(swsusp_arch_resume)
bl _tlbil_all
lwz r4,SL_SPRG0(r11)
- mtsprg 0,r4
+ mtspr SPRN_SPRG0,r4
lwz r4,SL_SPRG1(r11)
- mtsprg 1,r4
+ mtspr SPRN_SPRG1,r4
lwz r4,SL_SPRG2(r11)
- mtsprg 2,r4
+ mtspr SPRN_SPRG2,r4
lwz r4,SL_SPRG3(r11)
- mtsprg 3,r4
+ mtspr SPRN_SPRG3,r4
lwz r4,SL_SPRG4(r11)
- mtsprg 4,r4
+ mtspr SPRN_SPRG4,r4
lwz r4,SL_SPRG5(r11)
- mtsprg 5,r4
+ mtspr SPRN_SPRG5,r4
lwz r4,SL_SPRG6(r11)
- mtsprg 6,r4
+ mtspr SPRN_SPRG6,r4
lwz r4,SL_SPRG7(r11)
- mtsprg 7,r4
+ mtspr SPRN_SPRG7,r4
/* restore the MSR */
lwz r3,SL_MSR(r11)
--
1.8.3.2
^ permalink raw reply related
* Re: Where to post problems with linux-ppc for Freescale processors such as the p1020?
From: John Clark @ 2014-01-02 20:35 UTC (permalink / raw)
To: Kumar Gala; +Cc: linuxppc-dev
In-Reply-To: <EA9C3A7C-0F5A-4FB5-8DD1-CC6DF4489CA9@kernel.crashing.org>
On Jan 2, 2014, at 12:21 PM, Kumar Gala <galak@kernel.crashing.org> =
wrote:
>=20
> On Dec 31, 2013, at 2:35 PM, jeclark2006 <jeclark2006@aim.com> wrote:
>=20
>> As the subject says... where to post problems with ppc & linux that =
does not
>> involve 'apple' implementations.
>>=20
>> The 'embedded' list appears to be pretty much dead.
>=20
> This is now the list for any ppc linux issues for embedded, server, =
etc.
>=20
> - k
Thanks for the reply. I think I subscribed to 'linuxppc-dev', but as yet =
no 'confirmation' email has shown up.
Is there a moderator who needs to approve subscription?
Thanks,
John Clark.
^ permalink raw reply
* Re: Where to post problems with linux-ppc for Freescale processors such as the p1020?
From: Kumar Gala @ 2014-01-02 20:21 UTC (permalink / raw)
To: jeclark2006; +Cc: linuxppc-dev
In-Reply-To: <1388522127641-79016.post@n7.nabble.com>
On Dec 31, 2013, at 2:35 PM, jeclark2006 <jeclark2006@aim.com> wrote:
> As the subject says... where to post problems with ppc & linux that =
does not
> involve 'apple' implementations.
>=20
> The 'embedded' list appears to be pretty much dead.
This is now the list for any ppc linux issues for embedded, server, etc.
- k
^ permalink raw reply
* Re: [PATCH] block/ps3: Remove obsolete reference to MTD
From: Jens Axboe @ 2014-01-02 20:14 UTC (permalink / raw)
To: Geoff Levand, Geert Uytterhoeven
Cc: cbe-oss-dev, linuxppc-dev, linux-kernel, Jim Paris
In-Reply-To: <1388688375.32559.9.camel@smoke>
On 01/02/2014 11:46 AM, Geoff Levand wrote:
> On Mon, 2013-12-30 at 10:07 +0100, Geert Uytterhoeven wrote:
>> The ps3vram driver is a plain block device driver since commit
>> f507cd22035fdadd5dbb476dd05e9e7ee21c3b84 ("ps3/block: Replace mtd/ps3vram
>> by block/ps3vram").
>>
>> Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
>> ---
>> drivers/block/ps3vram.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/block/ps3vram.c b/drivers/block/ps3vram.c
>> index 06a2e53e5f37..313ee641ea10 100644
>> --- a/drivers/block/ps3vram.c
>> +++ b/drivers/block/ps3vram.c
>> @@ -1,5 +1,5 @@
>> /*
>> - * ps3vram - Use extra PS3 video ram as MTD block device.
>> + * ps3vram - Use extra PS3 video ram as block device.
>> *
>> * Copyright 2009 Sony Corporation
>> *
>
> Looks good.
>
> Jens, could you please apply.
>
> Acked-by: Geoff Levand <geoff@infradead.org>
Yup will do, thanks.
--
Jens Axboe
^ permalink raw reply
* Re: [PATCH] block/ps3: Remove obsolete reference to MTD
From: Geoff Levand @ 2014-01-02 18:46 UTC (permalink / raw)
To: Geert Uytterhoeven, Jens Axboe
Cc: cbe-oss-dev, linuxppc-dev, linux-kernel, Jim Paris
In-Reply-To: <1388394463-25798-1-git-send-email-geert@linux-m68k.org>
On Mon, 2013-12-30 at 10:07 +0100, Geert Uytterhoeven wrote:
> The ps3vram driver is a plain block device driver since commit
> f507cd22035fdadd5dbb476dd05e9e7ee21c3b84 ("ps3/block: Replace mtd/ps3vram
> by block/ps3vram").
>
> Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
> ---
> drivers/block/ps3vram.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/block/ps3vram.c b/drivers/block/ps3vram.c
> index 06a2e53e5f37..313ee641ea10 100644
> --- a/drivers/block/ps3vram.c
> +++ b/drivers/block/ps3vram.c
> @@ -1,5 +1,5 @@
> /*
> - * ps3vram - Use extra PS3 video ram as MTD block device.
> + * ps3vram - Use extra PS3 video ram as block device.
> *
> * Copyright 2009 Sony Corporation
> *
Looks good.
Jens, could you please apply.
Acked-by: Geoff Levand <geoff@infradead.org>
^ permalink raw reply
* Re: [PATCH -V2] powerpc: thp: Fix crash on mremap
From: Benjamin Herrenschmidt @ 2014-01-02 12:29 UTC (permalink / raw)
To: Aneesh Kumar K.V
Cc: aarcange, linuxppc-dev, paulus, Kirill A. Shutemov, linux-mm
In-Reply-To: <87zjneodtw.fsf@linux.vnet.ibm.com>
On Thu, 2014-01-02 at 16:22 +0530, Aneesh Kumar K.V wrote:
> > Just use config option directly:
> >
> > if (new_ptl != old_ptl ||
> > IS_ENABLED(CONFIG_ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW))
>
>
> I didn't like that. I found the earlier one easier for reading.
> If you and others strongly feel about this, I can redo the patch.
> Please let me know
Yes, use IS_ENABLED, no need to have two indirections of #define's
Another option is to have
if (pmd_move_must_withdraw(new,old)) {
}
With in a generic header:
#ifndef pmd_move_must_withdraw
static inline bool pmd_move_must_withdraw(spinlock_t *new_ptl, ...)
{
return new_ptl != old_ptl;
}
#endif
And in powerpc:
static inline bool pmd_move_must_withdraw(spinlock_t *new_ptl, ...)
{
return true;
}
#define pmd_move_must_withdraw pmd_move_must_withdraw
Cheers,
Ben.
^ permalink raw reply
* [PATCH] powerpc/powernv: Increase candidate fw image size
From: Vasant Hegde @ 2014-01-02 11:30 UTC (permalink / raw)
To: linuxppc-dev
At present we assume candidate image is <= 256MB. But in P8,
candidate image size can go up to 750MB. Hence increasing
candidate image max size to 1GB.
Signed-off-by: Vasant Hegde <hegdevasant@linux.vnet.ibm.com>
---
arch/powerpc/platforms/powernv/opal-flash.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/platforms/powernv/opal-flash.c b/arch/powerpc/platforms/powernv/opal-flash.c
index 4aeae4f..16e571b 100644
--- a/arch/powerpc/platforms/powernv/opal-flash.c
+++ b/arch/powerpc/platforms/powernv/opal-flash.c
@@ -76,8 +76,8 @@
/* Validate buffer size */
#define VALIDATE_BUF_SIZE 4096
-/* XXX: Assume candidate image size is <= 256MB */
-#define MAX_IMAGE_SIZE 0x10000000
+/* XXX: Assume candidate image size is <= 1GB */
+#define MAX_IMAGE_SIZE 0x40000000
/* Flash sg list version */
#define SG_LIST_VERSION (1UL)
^ permalink raw reply related
* RE: [PATCH -V2] powerpc: thp: Fix crash on mremap
From: Aneesh Kumar K.V @ 2014-01-02 10:52 UTC (permalink / raw)
To: Kirill A. Shutemov
Cc: aarcange, linux-mm, paulus, linuxppc-dev, kirill.shutemov
In-Reply-To: <20140102094124.04D76E0090@blue.fi.intel.com>
"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> writes:
> Aneesh Kumar K.V wrote:
>> From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
>>
>> This patch fix the below crash
>>
>> NIP [c00000000004cee4] .__hash_page_thp+0x2a4/0x440
>> LR [c0000000000439ac] .hash_page+0x18c/0x5e0
>> ...
>> Call Trace:
>> [c000000736103c40] [00001ffffb000000] 0x1ffffb000000(unreliable)
>> [437908.479693] [c000000736103d50] [c0000000000439ac] .hash_page+0x18c/0x5e0
>> [437908.479699] [c000000736103e30] [c00000000000924c] .do_hash_page+0x4c/0x58
>>
>> On ppc64 we use the pgtable for storing the hpte slot information and
>> store address to the pgtable at a constant offset (PTRS_PER_PMD) from
>> pmd. On mremap, when we switch the pmd, we need to withdraw and deposit
>> the pgtable again, so that we find the pgtable at PTRS_PER_PMD offset
>> from new pmd.
>>
>> We also want to move the withdraw and deposit before the set_pmd so
>> that, when page fault find the pmd as trans huge we can be sure that
>> pgtable can be located at the offset.
>>
>> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
>> ---
>> Changes from V1:
>> * limit the withraw/deposit to only ppc64
>>
>> arch/Kconfig | 3 +++
>> arch/powerpc/platforms/Kconfig.cputype | 1 +
>> include/linux/huge_mm.h | 6 ++++++
>> mm/huge_memory.c | 21 ++++++++++++---------
>> 4 files changed, 22 insertions(+), 9 deletions(-)
>>
>> diff --git a/arch/Kconfig b/arch/Kconfig
>> index f1cf895c040f..3759e70a649d 100644
>> --- a/arch/Kconfig
>> +++ b/arch/Kconfig
>> @@ -371,6 +371,9 @@ config HAVE_IRQ_TIME_ACCOUNTING
>> config HAVE_ARCH_TRANSPARENT_HUGEPAGE
>> bool
>>
>> +config ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW
>
> I don't like name of the option, but can't find any better... :(
>
>> + bool
>> +
>> config HAVE_ARCH_SOFT_DIRTY
>> bool
>>
>> diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
>> index bca2465a9c34..5f83b4334e5f 100644
>> --- a/arch/powerpc/platforms/Kconfig.cputype
>> +++ b/arch/powerpc/platforms/Kconfig.cputype
>> @@ -71,6 +71,7 @@ config PPC_BOOK3S_64
>> select PPC_FPU
>> select PPC_HAVE_PMU_SUPPORT
>> select SYS_SUPPORTS_HUGETLBFS
>> + select ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW
>> select HAVE_ARCH_TRANSPARENT_HUGEPAGE if PPC_64K_PAGES
>>
>> config PPC_BOOK3E_64
>> diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
>> index 91672e2deec3..836242a738a5 100644
>> --- a/include/linux/huge_mm.h
>> +++ b/include/linux/huge_mm.h
>> @@ -230,4 +230,10 @@ static inline int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_str
>>
>> #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
>>
>> +#ifdef CONFIG_ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW
>> +#define ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW 1
>> +#else
>> +#define ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW 0
>> +#endif
>> +
>
> Just use config option directly:
>
> if (new_ptl != old_ptl ||
> IS_ENABLED(CONFIG_ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW))
I didn't like that. I found the earlier one easier for reading.
If you and others strongly feel about this, I can redo the patch. Please let me know
> ...
>
>
> Otherwise, looks good:
>
> Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
>
>> #endif /* _LINUX_HUGE_MM_H */
>> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
>> index 7de1bf85f683..32006b51d102 100644
>> --- a/mm/huge_memory.c
>> +++ b/mm/huge_memory.c
>> @@ -1505,19 +1505,22 @@ int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
>> spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
>> pmd = pmdp_get_and_clear(mm, old_addr, old_pmd);
>> VM_BUG_ON(!pmd_none(*new_pmd));
>> - set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
>> - if (new_ptl != old_ptl) {
>> + /*
>> + * Archs like ppc64 use pgtable to store per pmd
>> + * specific information. So when we switch the pmd,
>> + * we should also withdraw and deposit the pgtable
>> + *
>> + * With split pmd lock we also need to move preallocated
>> + * PTE page table if new_pmd is on different PMD page table.
>> + */
>> + if (new_ptl != old_ptl || ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW) {
>> pgtable_t pgtable;
>> -
>> - /*
>> - * Move preallocated PTE page table if new_pmd is on
>> - * different PMD page table.
>> - */
>> pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
>> pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
>> -
>> - spin_unlock(new_ptl);
>> }
>> + set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
>> + if (new_ptl != old_ptl)
>> + spin_unlock(new_ptl);
>> spin_unlock(old_ptl);
>> }
>> out:
>> --
>> 1.8.3.2
>
> --
> Kirill A. Shutemov
^ permalink raw reply
* RE: [PATCH -V2] powerpc: thp: Fix crash on mremap
From: Kirill A. Shutemov @ 2014-01-02 9:41 UTC (permalink / raw)
To: Aneesh Kumar K.V
Cc: aarcange, linux-mm, paulus, Aneesh Kumar K.V, linuxppc-dev,
kirill.shutemov
In-Reply-To: <1388654266-5195-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com>
Aneesh Kumar K.V wrote:
> From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
>
> This patch fix the below crash
>
> NIP [c00000000004cee4] .__hash_page_thp+0x2a4/0x440
> LR [c0000000000439ac] .hash_page+0x18c/0x5e0
> ...
> Call Trace:
> [c000000736103c40] [00001ffffb000000] 0x1ffffb000000(unreliable)
> [437908.479693] [c000000736103d50] [c0000000000439ac] .hash_page+0x18c/0x5e0
> [437908.479699] [c000000736103e30] [c00000000000924c] .do_hash_page+0x4c/0x58
>
> On ppc64 we use the pgtable for storing the hpte slot information and
> store address to the pgtable at a constant offset (PTRS_PER_PMD) from
> pmd. On mremap, when we switch the pmd, we need to withdraw and deposit
> the pgtable again, so that we find the pgtable at PTRS_PER_PMD offset
> from new pmd.
>
> We also want to move the withdraw and deposit before the set_pmd so
> that, when page fault find the pmd as trans huge we can be sure that
> pgtable can be located at the offset.
>
> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
> ---
> Changes from V1:
> * limit the withraw/deposit to only ppc64
>
> arch/Kconfig | 3 +++
> arch/powerpc/platforms/Kconfig.cputype | 1 +
> include/linux/huge_mm.h | 6 ++++++
> mm/huge_memory.c | 21 ++++++++++++---------
> 4 files changed, 22 insertions(+), 9 deletions(-)
>
> diff --git a/arch/Kconfig b/arch/Kconfig
> index f1cf895c040f..3759e70a649d 100644
> --- a/arch/Kconfig
> +++ b/arch/Kconfig
> @@ -371,6 +371,9 @@ config HAVE_IRQ_TIME_ACCOUNTING
> config HAVE_ARCH_TRANSPARENT_HUGEPAGE
> bool
>
> +config ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW
I don't like name of the option, but can't find any better... :(
> + bool
> +
> config HAVE_ARCH_SOFT_DIRTY
> bool
>
> diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
> index bca2465a9c34..5f83b4334e5f 100644
> --- a/arch/powerpc/platforms/Kconfig.cputype
> +++ b/arch/powerpc/platforms/Kconfig.cputype
> @@ -71,6 +71,7 @@ config PPC_BOOK3S_64
> select PPC_FPU
> select PPC_HAVE_PMU_SUPPORT
> select SYS_SUPPORTS_HUGETLBFS
> + select ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW
> select HAVE_ARCH_TRANSPARENT_HUGEPAGE if PPC_64K_PAGES
>
> config PPC_BOOK3E_64
> diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
> index 91672e2deec3..836242a738a5 100644
> --- a/include/linux/huge_mm.h
> +++ b/include/linux/huge_mm.h
> @@ -230,4 +230,10 @@ static inline int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_str
>
> #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
>
> +#ifdef CONFIG_ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW
> +#define ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW 1
> +#else
> +#define ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW 0
> +#endif
> +
Just use config option directly:
if (new_ptl != old_ptl ||
IS_ENABLED(CONFIG_ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW))
...
Otherwise, looks good:
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> #endif /* _LINUX_HUGE_MM_H */
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 7de1bf85f683..32006b51d102 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -1505,19 +1505,22 @@ int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
> spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
> pmd = pmdp_get_and_clear(mm, old_addr, old_pmd);
> VM_BUG_ON(!pmd_none(*new_pmd));
> - set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
> - if (new_ptl != old_ptl) {
> + /*
> + * Archs like ppc64 use pgtable to store per pmd
> + * specific information. So when we switch the pmd,
> + * we should also withdraw and deposit the pgtable
> + *
> + * With split pmd lock we also need to move preallocated
> + * PTE page table if new_pmd is on different PMD page table.
> + */
> + if (new_ptl != old_ptl || ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW) {
> pgtable_t pgtable;
> -
> - /*
> - * Move preallocated PTE page table if new_pmd is on
> - * different PMD page table.
> - */
> pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
> pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
> -
> - spin_unlock(new_ptl);
> }
> + set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
> + if (new_ptl != old_ptl)
> + spin_unlock(new_ptl);
> spin_unlock(old_ptl);
> }
> out:
> --
> 1.8.3.2
--
Kirill A. Shutemov
^ permalink raw reply
* [PATCH -V2] powerpc: thp: Fix crash on mremap
From: Aneesh Kumar K.V @ 2014-01-02 9:17 UTC (permalink / raw)
To: benh, paulus, aarcange, kirill.shutemov
Cc: linux-mm, linuxppc-dev, Aneesh Kumar K.V
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
This patch fix the below crash
NIP [c00000000004cee4] .__hash_page_thp+0x2a4/0x440
LR [c0000000000439ac] .hash_page+0x18c/0x5e0
...
Call Trace:
[c000000736103c40] [00001ffffb000000] 0x1ffffb000000(unreliable)
[437908.479693] [c000000736103d50] [c0000000000439ac] .hash_page+0x18c/0x5e0
[437908.479699] [c000000736103e30] [c00000000000924c] .do_hash_page+0x4c/0x58
On ppc64 we use the pgtable for storing the hpte slot information and
store address to the pgtable at a constant offset (PTRS_PER_PMD) from
pmd. On mremap, when we switch the pmd, we need to withdraw and deposit
the pgtable again, so that we find the pgtable at PTRS_PER_PMD offset
from new pmd.
We also want to move the withdraw and deposit before the set_pmd so
that, when page fault find the pmd as trans huge we can be sure that
pgtable can be located at the offset.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
Changes from V1:
* limit the withraw/deposit to only ppc64
arch/Kconfig | 3 +++
arch/powerpc/platforms/Kconfig.cputype | 1 +
include/linux/huge_mm.h | 6 ++++++
mm/huge_memory.c | 21 ++++++++++++---------
4 files changed, 22 insertions(+), 9 deletions(-)
diff --git a/arch/Kconfig b/arch/Kconfig
index f1cf895c040f..3759e70a649d 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -371,6 +371,9 @@ config HAVE_IRQ_TIME_ACCOUNTING
config HAVE_ARCH_TRANSPARENT_HUGEPAGE
bool
+config ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW
+ bool
+
config HAVE_ARCH_SOFT_DIRTY
bool
diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
index bca2465a9c34..5f83b4334e5f 100644
--- a/arch/powerpc/platforms/Kconfig.cputype
+++ b/arch/powerpc/platforms/Kconfig.cputype
@@ -71,6 +71,7 @@ config PPC_BOOK3S_64
select PPC_FPU
select PPC_HAVE_PMU_SUPPORT
select SYS_SUPPORTS_HUGETLBFS
+ select ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW
select HAVE_ARCH_TRANSPARENT_HUGEPAGE if PPC_64K_PAGES
config PPC_BOOK3E_64
diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
index 91672e2deec3..836242a738a5 100644
--- a/include/linux/huge_mm.h
+++ b/include/linux/huge_mm.h
@@ -230,4 +230,10 @@ static inline int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_str
#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
+#ifdef CONFIG_ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW
+#define ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW 1
+#else
+#define ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW 0
+#endif
+
#endif /* _LINUX_HUGE_MM_H */
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 7de1bf85f683..32006b51d102 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1505,19 +1505,22 @@ int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
pmd = pmdp_get_and_clear(mm, old_addr, old_pmd);
VM_BUG_ON(!pmd_none(*new_pmd));
- set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
- if (new_ptl != old_ptl) {
+ /*
+ * Archs like ppc64 use pgtable to store per pmd
+ * specific information. So when we switch the pmd,
+ * we should also withdraw and deposit the pgtable
+ *
+ * With split pmd lock we also need to move preallocated
+ * PTE page table if new_pmd is on different PMD page table.
+ */
+ if (new_ptl != old_ptl || ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW) {
pgtable_t pgtable;
-
- /*
- * Move preallocated PTE page table if new_pmd is on
- * different PMD page table.
- */
pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
-
- spin_unlock(new_ptl);
}
+ set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
+ if (new_ptl != old_ptl)
+ spin_unlock(new_ptl);
spin_unlock(old_ptl);
}
out:
--
1.8.3.2
^ permalink raw reply related
* Re: [PATCH v2 3/9] POWERPC: pseries: cpuidle: use the common cpuidle_[un]register() routines
From: Deepthi Dharwar @ 2014-01-02 5:42 UTC (permalink / raw)
To: Bartlomiej Zolnierkiewicz
Cc: linux-samsung-soc, linux-pm, daniel.lezcano, rjw, linux-kernel,
kyungmin.park, linuxppc-dev, lenb
In-Reply-To: <1387565251-7051-4-git-send-email-b.zolnierkie@samsung.com>
On 12/21/2013 12:17 AM, Bartlomiej Zolnierkiewicz wrote:
> It is now possible to use the common cpuidle_[un]register() routines
> (instead of open-coding them) so do it.
>
> Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
> Cc: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
> ---
> arch/powerpc/platforms/pseries/processor_idle.c | 57 ++-----------------------
> 1 file changed, 3 insertions(+), 54 deletions(-)
>
> diff --git a/arch/powerpc/platforms/pseries/processor_idle.c b/arch/powerpc/platforms/pseries/processor_idle.c
> index 8aa8c40..94134a5 100644
> --- a/arch/powerpc/platforms/pseries/processor_idle.c
> +++ b/arch/powerpc/platforms/pseries/processor_idle.c
> @@ -28,7 +28,6 @@ struct cpuidle_driver pseries_idle_driver = {
> #define MAX_IDLE_STATE_COUNT 2
>
> static int max_idle_state = MAX_IDLE_STATE_COUNT - 1;
> -static struct cpuidle_device __percpu *pseries_cpuidle_devices;
> static struct cpuidle_state *cpuidle_state_table;
>
> static inline void idle_loop_prolog(unsigned long *in_purr)
> @@ -191,7 +190,7 @@ static int pseries_cpuidle_add_cpu_notifier(struct notifier_block *n,
> {
> int hotcpu = (unsigned long)hcpu;
> struct cpuidle_device *dev =
> - per_cpu_ptr(pseries_cpuidle_devices, hotcpu);
> + per_cpu_ptr(cpuidle_devices, hotcpu);
>
> if (dev && cpuidle_get_driver()) {
> switch (action) {
> @@ -248,48 +247,6 @@ static int pseries_cpuidle_driver_init(void)
> return 0;
> }
>
> -/* pseries_idle_devices_uninit(void)
> - * unregister cpuidle devices and de-allocate memory
> - */
> -static void pseries_idle_devices_uninit(void)
> -{
> - int i;
> - struct cpuidle_device *dev;
> -
> - for_each_possible_cpu(i) {
> - dev = per_cpu_ptr(pseries_cpuidle_devices, i);
> - cpuidle_unregister_device(dev);
> - }
> -
> - free_percpu(pseries_cpuidle_devices);
> - return;
> -}
> -
> -/* pseries_idle_devices_init()
> - * allocate, initialize and register cpuidle device
> - */
> -static int pseries_idle_devices_init(void)
> -{
> - int i;
> - struct cpuidle_device *dev;
> -
> - pseries_cpuidle_devices = alloc_percpu(struct cpuidle_device);
> - if (pseries_cpuidle_devices == NULL)
> - return -ENOMEM;
> -
> - for_each_possible_cpu(i) {
> - dev = per_cpu_ptr(pseries_cpuidle_devices, i);
> - dev->cpu = i;
> - if (cpuidle_register_device(dev)) {
> - printk(KERN_DEBUG \
> - "cpuidle_register_device %d failed!\n", i);
> - return -EIO;
> - }
> - }
> -
> - return 0;
> -}
> -
> /*
> * pseries_idle_probe()
> * Choose state table for shared versus dedicated partition
> @@ -325,19 +282,12 @@ static int __init pseries_processor_idle_init(void)
> return retval;
>
> pseries_cpuidle_driver_init();
> - retval = cpuidle_register_driver(&pseries_idle_driver);
> + retval = cpuidle_register(&pseries_idle_driver, NULL);
> if (retval) {
> printk(KERN_DEBUG "Registration of pseries driver failed.\n");
> return retval;
> }
>
> - retval = pseries_idle_devices_init();
> - if (retval) {
> - pseries_idle_devices_uninit();
> - cpuidle_unregister_driver(&pseries_idle_driver);
> - return retval;
> - }
> -
> register_cpu_notifier(&setup_hotplug_notifier);
> printk(KERN_DEBUG "pseries_idle_driver registered\n");
>
> @@ -348,8 +298,7 @@ static void __exit pseries_processor_idle_exit(void)
> {
>
> unregister_cpu_notifier(&setup_hotplug_notifier);
> - pseries_idle_devices_uninit();
> - cpuidle_unregister_driver(&pseries_idle_driver);
> + cpuidle_unregister(&pseries_idle_driver);
>
> return;
> }
>
^ permalink raw reply
* Re: [PATCH v2 2/9] POWERPC: pseries: cpuidle: remove superfluous dev->state_count initialization
From: Deepthi Dharwar @ 2014-01-02 5:41 UTC (permalink / raw)
To: Bartlomiej Zolnierkiewicz
Cc: linux-samsung-soc, linux-pm, daniel.lezcano, rjw, linux-kernel,
kyungmin.park, linuxppc-dev, lenb
In-Reply-To: <1387565251-7051-3-git-send-email-b.zolnierkie@samsung.com>
On 12/21/2013 12:17 AM, Bartlomiej Zolnierkiewicz wrote:
> pseries cpuidle driver sets dev->state_count to drv->state_count so
> the default dev->state_count initialization in cpuidle_enable_device()
> (called from cpuidle_register_device()) can be used instead.
>
> Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
> Cc: Deepthi Dharwar <deepthi@linux.vnet.ibm.com>
> ---
> arch/powerpc/platforms/pseries/processor_idle.c | 2 --
> 1 file changed, 2 deletions(-)
>
> diff --git a/arch/powerpc/platforms/pseries/processor_idle.c b/arch/powerpc/platforms/pseries/processor_idle.c
> index a166e38..8aa8c40 100644
> --- a/arch/powerpc/platforms/pseries/processor_idle.c
> +++ b/arch/powerpc/platforms/pseries/processor_idle.c
> @@ -271,7 +271,6 @@ static void pseries_idle_devices_uninit(void)
> static int pseries_idle_devices_init(void)
> {
> int i;
> - struct cpuidle_driver *drv = &pseries_idle_driver;
> struct cpuidle_device *dev;
>
> pseries_cpuidle_devices = alloc_percpu(struct cpuidle_device);
> @@ -280,7 +279,6 @@ static int pseries_idle_devices_init(void)
>
> for_each_possible_cpu(i) {
> dev = per_cpu_ptr(pseries_cpuidle_devices, i);
> - dev->state_count = drv->state_count;
> dev->cpu = i;
> if (cpuidle_register_device(dev)) {
> printk(KERN_DEBUG \
>
^ permalink raw reply
* Re: [PATCH 1/2] power7, perf: Make some new raw event codes available in sysfs
From: Anshuman Khandual @ 2014-01-02 5:05 UTC (permalink / raw)
To: Michael Ellerman; +Cc: linuxppc-dev, mikey, sukadev
In-Reply-To: <1387331962.19507.4.camel@concordia>
On 12/18/2013 07:29 AM, Michael Ellerman wrote:
> On Fri, 2013-12-13 at 10:00 +0530, Anshuman Khandual wrote:
>> On 12/13/2013 08:20 AM, Michael Ellerman wrote:
>>> On Wed, 2013-10-16 at 11:22 +0530, Anshuman Khandual wrote:
>>>> This patch adds some more raw event codes into the existing list
>>>> of event codes present in power7-events-list.h file. This tries
>>>> to complete the list of events supported in Power7 and matches
>>>> the raw event list with libpfm4 library.
>>>
>>> It's a bit annoying, but you also need to update the "ABI" document:
>>
>> What is annoying ? you need to be specific.
>
> It's annoying that we have to update the ABI document.
>
>>> Documentation/ABI/testing/sysfs-bus-event_source-devices-events
>>>
>>
>> The events listed under the following heading are events required to do
>> CPI analysis.
>
> No they are not, it doesn't say that anywhere in the file.
>
> It happens that the events in there *now* are the CPI events, but that's because
> they are the only ones that have been added.
>
>> "Description: POWER-systems specific performance monitoring event"
>>
>> /sys/devices/cpu/events/PM_1PLUS_PPC_CMPL
>> /sys/devices/cpu/events/PM_BRU_FIN
>> /sys/devices/cpu/events/PM_BR_MPRED
>> /sys/devices/cpu/events/PM_CMPLU_STALL
>
> All events that appear in /sys/devices/cpu/events on powerpc should be listed
> in the file.
Done, sent out the V2 version of the patch for POWER7 adding all the sysfs exported
raw events into the ABI documentation file.
^ permalink raw reply
* [PATCH V2] power7, perf: Make some new raw event codes available in sysfs
From: Anshuman Khandual @ 2014-01-02 5:02 UTC (permalink / raw)
To: Linux PPC dev; +Cc: Michael Ellerman
This patchset adds some missing event list for POWER7 PMU raw
events which are exported through sysfs interface. Also updates
the ABI documentation to add all the sysfs exported raw events.
Signed-off-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
---
.../testing/sysfs-bus-event_source-devices-events | 517 +++++++++++++++++++++
arch/powerpc/perf/power7-events-list.h | 10 +
2 files changed, 527 insertions(+)
diff --git a/Documentation/ABI/testing/sysfs-bus-event_source-devices-events b/Documentation/ABI/testing/sysfs-bus-event_source-devices-events
index 3c1cc24..7b40a3c 100644
--- a/Documentation/ABI/testing/sysfs-bus-event_source-devices-events
+++ b/Documentation/ABI/testing/sysfs-bus-event_source-devices-events
@@ -57,6 +57,523 @@ What: /sys/devices/cpu/events/PM_1PLUS_PPC_CMPL
/sys/devices/cpu/events/PM_LD_REF_L1
/sys/devices/cpu/events/PM_RUN_CYC
/sys/devices/cpu/events/PM_RUN_INST_CMPL
+ /sys/devices/cpu/events/PM_IC_DEMAND_L2_BR_ALL
+ /sys/devices/cpu/events/PM_GCT_UTIL_7_TO_10_SLOTS
+ /sys/devices/cpu/events/PM_PMC2_SAVED
+ /sys/devices/cpu/events/PM_VSU0_16FLOP
+ /sys/devices/cpu/events/PM_MRK_LSU_DERAT_MISS
+ /sys/devices/cpu/events/PM_MRK_ST_CMPL
+ /sys/devices/cpu/events/PM_NEST_PAIR3_ADD
+ /sys/devices/cpu/events/PM_L2_ST_DISP
+ /sys/devices/cpu/events/PM_L2_CASTOUT_MOD
+ /sys/devices/cpu/events/PM_ISEG
+ /sys/devices/cpu/events/PM_MRK_INST_TIMEO
+ /sys/devices/cpu/events/PM_L2_RCST_DISP_FAIL_ADDR
+ /sys/devices/cpu/events/PM_LSU1_DC_PREF_STREAM_CONFIRM
+ /sys/devices/cpu/events/PM_IERAT_WR_64K
+ /sys/devices/cpu/events/PM_MRK_DTLB_MISS_16M
+ /sys/devices/cpu/events/PM_IERAT_MISS
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_LMEM
+ /sys/devices/cpu/events/PM_FLOP
+ /sys/devices/cpu/events/PM_THRD_PRIO_4_5_CYC
+ /sys/devices/cpu/events/PM_BR_PRED_TA
+ /sys/devices/cpu/events/PM_EXT_INT
+ /sys/devices/cpu/events/PM_VSU_FSQRT_FDIV
+ /sys/devices/cpu/events/PM_MRK_LD_MISS_EXPOSED_CYC
+ /sys/devices/cpu/events/PM_LSU1_LDF
+ /sys/devices/cpu/events/PM_IC_WRITE_ALL
+ /sys/devices/cpu/events/PM_LSU0_SRQ_STFWD
+ /sys/devices/cpu/events/PM_PTEG_FROM_RL2L3_MOD
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L31_SHR
+ /sys/devices/cpu/events/PM_DATA_FROM_L21_MOD
+ /sys/devices/cpu/events/PM_VSU1_SCAL_DOUBLE_ISSUED
+ /sys/devices/cpu/events/PM_VSU0_8FLOP
+ /sys/devices/cpu/events/PM_POWER_EVENT1
+ /sys/devices/cpu/events/PM_DISP_CLB_HELD_BAL
+ /sys/devices/cpu/events/PM_VSU1_2FLOP
+ /sys/devices/cpu/events/PM_LWSYNC_HELD
+ /sys/devices/cpu/events/PM_PTEG_FROM_DL2L3_SHR
+ /sys/devices/cpu/events/PM_INST_FROM_L21_MOD
+ /sys/devices/cpu/events/PM_IERAT_XLATE_WR_16MPLUS
+ /sys/devices/cpu/events/PM_IC_REQ_ALL
+ /sys/devices/cpu/events/PM_DSLB_MISS
+ /sys/devices/cpu/events/PM_L3_MISS
+ /sys/devices/cpu/events/PM_LSU0_L1_PREF
+ /sys/devices/cpu/events/PM_VSU_SCALAR_SINGLE_ISSUED
+ /sys/devices/cpu/events/PM_LSU1_DC_PREF_STREAM_CONFIRM_STRIDE
+ /sys/devices/cpu/events/PM_L2_INST
+ /sys/devices/cpu/events/PM_VSU0_FRSP
+ /sys/devices/cpu/events/PM_FLUSH_DISP
+ /sys/devices/cpu/events/PM_PTEG_FROM_L2MISS
+ /sys/devices/cpu/events/PM_VSU1_DQ_ISSUED
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_DMEM
+ /sys/devices/cpu/events/PM_LSU_FLUSH_ULD
+ /sys/devices/cpu/events/PM_PTEG_FROM_LMEM
+ /sys/devices/cpu/events/PM_MRK_DERAT_MISS_16M
+ /sys/devices/cpu/events/PM_THRD_ALL_RUN_CYC
+ /sys/devices/cpu/events/PM_MEM0_PREFETCH_DISP
+ /sys/devices/cpu/events/PM_MRK_STALL_CMPLU_CYC_COUNT
+ /sys/devices/cpu/events/PM_DATA_FROM_DL2L3_MOD
+ /sys/devices/cpu/events/PM_VSU_FRSP
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L21_MOD
+ /sys/devices/cpu/events/PM_PMC1_OVERFLOW
+ /sys/devices/cpu/events/PM_VSU0_SINGLE
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_L3MISS
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_L31_SHR
+ /sys/devices/cpu/events/PM_VSU0_VECTOR_SP_ISSUED
+ /sys/devices/cpu/events/PM_VSU1_FEST
+ /sys/devices/cpu/events/PM_MRK_INST_DISP
+ /sys/devices/cpu/events/PM_VSU0_COMPLEX_ISSUED
+ /sys/devices/cpu/events/PM_LSU1_FLUSH_UST
+ /sys/devices/cpu/events/PM_FXU_IDLE
+ /sys/devices/cpu/events/PM_LSU0_FLUSH_ULD
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_DL2L3_MOD
+ /sys/devices/cpu/events/PM_LSU_LMQ_SRQ_EMPTY_ALL_CYC
+ /sys/devices/cpu/events/PM_LSU1_REJECT_LMQ_FULL
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_L21_MOD
+ /sys/devices/cpu/events/PM_INST_FROM_RL2L3_MOD
+ /sys/devices/cpu/events/PM_SHL_CREATED
+ /sys/devices/cpu/events/PM_L2_ST_HIT
+ /sys/devices/cpu/events/PM_DATA_FROM_DMEM
+ /sys/devices/cpu/events/PM_L3_LD_MISS
+ /sys/devices/cpu/events/PM_FXU1_BUSY_FXU0_IDLE
+ /sys/devices/cpu/events/PM_DISP_CLB_HELD_RES
+ /sys/devices/cpu/events/PM_L2_SN_SX_I_DONE
+ /sys/devices/cpu/events/PM_STCX_CMPL
+ /sys/devices/cpu/events/PM_VSU0_2FLOP
+ /sys/devices/cpu/events/PM_L3_PREF_MISS
+ /sys/devices/cpu/events/PM_LSU_SRQ_SYNC_CYC
+ /sys/devices/cpu/events/PM_LSU_REJECT_ERAT_MISS
+ /sys/devices/cpu/events/PM_L1_ICACHE_MISS
+ /sys/devices/cpu/events/PM_LSU1_FLUSH_SRQ
+ /sys/devices/cpu/events/PM_LD_REF_L1_LSU0
+ /sys/devices/cpu/events/PM_VSU0_FEST
+ /sys/devices/cpu/events/PM_VSU_VECTOR_SINGLE_ISSUED
+ /sys/devices/cpu/events/PM_FREQ_UP
+ /sys/devices/cpu/events/PM_DATA_FROM_LMEM
+ /sys/devices/cpu/events/PM_LSU1_LDX
+ /sys/devices/cpu/events/PM_PMC3_OVERFLOW
+ /sys/devices/cpu/events/PM_MRK_BR_MPRED
+ /sys/devices/cpu/events/PM_SHL_MATCH
+ /sys/devices/cpu/events/PM_MRK_BR_TAKEN
+ /sys/devices/cpu/events/PM_ISLB_MISS
+ /sys/devices/cpu/events/PM_DISP_HELD_THERMAL
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_RL2L3_SHR
+ /sys/devices/cpu/events/PM_LSU1_SRQ_STFWD
+ /sys/devices/cpu/events/PM_PTEG_FROM_DMEM
+ /sys/devices/cpu/events/PM_VSU_2FLOP
+ /sys/devices/cpu/events/PM_GCT_FULL_CYC
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L3_CYC
+ /sys/devices/cpu/events/PM_LSU_SRQ_S0_ALLOC
+ /sys/devices/cpu/events/PM_MRK_DERAT_MISS_4K
+ /sys/devices/cpu/events/PM_BR_MPRED_TA
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_L2MISS
+ /sys/devices/cpu/events/PM_DPU_HELD_POWER
+ /sys/devices/cpu/events/PM_MRK_VSU_FIN
+ /sys/devices/cpu/events/PM_LSU_SRQ_S0_VALID
+ /sys/devices/cpu/events/PM_GCT_EMPTY_CYC
+ /sys/devices/cpu/events/PM_IOPS_DISP
+ /sys/devices/cpu/events/PM_RUN_SPURR
+ /sys/devices/cpu/events/PM_PTEG_FROM_L21_MOD
+ /sys/devices/cpu/events/PM_VSU0_1FLOP
+ /sys/devices/cpu/events/PM_SNOOP_TLBIE
+ /sys/devices/cpu/events/PM_DATA_FROM_L3MISS
+ /sys/devices/cpu/events/PM_VSU_SINGLE
+ /sys/devices/cpu/events/PM_DTLB_MISS_16G
+ /sys/devices/cpu/events/PM_FLUSH
+ /sys/devices/cpu/events/PM_L2_LD_HIT
+ /sys/devices/cpu/events/PM_NEST_PAIR2_AND
+ /sys/devices/cpu/events/PM_VSU1_1FLOP
+ /sys/devices/cpu/events/PM_IC_PREF_REQ
+ /sys/devices/cpu/events/PM_L3_LD_HIT
+ /sys/devices/cpu/events/PM_DISP_HELD
+ /sys/devices/cpu/events/PM_L2_LD
+ /sys/devices/cpu/events/PM_LSU_FLUSH_SRQ
+ /sys/devices/cpu/events/PM_BC_PLUS_8_CONV
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L31_MOD_CYC
+ /sys/devices/cpu/events/PM_L2_RCST_BUSY_RC_FULL
+ /sys/devices/cpu/events/PM_TB_BIT_TRANS
+ /sys/devices/cpu/events/PM_THERMAL_MAX
+ /sys/devices/cpu/events/PM_LSU1_FLUSH_ULD
+ /sys/devices/cpu/events/PM_LSU1_REJECT_LHS
+ /sys/devices/cpu/events/PM_LSU_LRQ_S0_ALLOC
+ /sys/devices/cpu/events/PM_L3_CO_L31
+ /sys/devices/cpu/events/PM_POWER_EVENT4
+ /sys/devices/cpu/events/PM_DATA_FROM_L31_SHR
+ /sys/devices/cpu/events/PM_BR_UNCOND
+ /sys/devices/cpu/events/PM_LSU1_DC_PREF_STREAM_ALLOC
+ /sys/devices/cpu/events/PM_PMC4_REWIND
+ /sys/devices/cpu/events/PM_L2_RCLD_DISP
+ /sys/devices/cpu/events/PM_THRD_PRIO_2_3_CYC
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_L2MISS
+ /sys/devices/cpu/events/PM_IC_DEMAND_L2_BHT_REDIRECT
+ /sys/devices/cpu/events/PM_DATA_FROM_L31_SHR
+ /sys/devices/cpu/events/PM_IC_PREF_CANCEL_L2
+ /sys/devices/cpu/events/PM_MRK_FIN_STALL_CYC_COUNT
+ /sys/devices/cpu/events/PM_BR_PRED_CCACHE
+ /sys/devices/cpu/events/PM_GCT_UTIL_1_TO_2_SLOTS
+ /sys/devices/cpu/events/PM_MRK_ST_CMPL_INT
+ /sys/devices/cpu/events/PM_LSU_TWO_TABLEWALK_CYC
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L3MISS
+ /sys/devices/cpu/events/PM_LSU_SET_MPRED
+ /sys/devices/cpu/events/PM_FLUSH_DISP_TLBIE
+ /sys/devices/cpu/events/PM_VSU1_FCONV
+ /sys/devices/cpu/events/PM_DERAT_MISS_16G
+ /sys/devices/cpu/events/PM_INST_FROM_LMEM
+ /sys/devices/cpu/events/PM_IC_DEMAND_L2_BR_REDIRECT
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_L2
+ /sys/devices/cpu/events/PM_PTEG_FROM_L2
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L21_SHR_CYC
+ /sys/devices/cpu/events/PM_MRK_DTLB_MISS_4K
+ /sys/devices/cpu/events/PM_VSU0_FPSCR
+ /sys/devices/cpu/events/PM_VSU1_VECT_DOUBLE_ISSUED
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_RL2L3_MOD
+ /sys/devices/cpu/events/PM_MEM0_RQ_DISP
+ /sys/devices/cpu/events/PM_L2_LD_MISS
+ /sys/devices/cpu/events/PM_VMX_RESULT_SAT_1
+ /sys/devices/cpu/events/PM_L1_PREF
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_LMEM_CYC
+ /sys/devices/cpu/events/PM_GRP_IC_MISS_NONSPEC
+ /sys/devices/cpu/events/PM_PB_NODE_PUMP
+ /sys/devices/cpu/events/PM_SHL_MERGED
+ /sys/devices/cpu/events/PM_NEST_PAIR1_ADD
+ /sys/devices/cpu/events/PM_DATA_FROM_L3
+ /sys/devices/cpu/events/PM_LSU_FLUSH
+ /sys/devices/cpu/events/PM_LSU_SRQ_SYNC_COUNT
+ /sys/devices/cpu/events/PM_PMC2_OVERFLOW
+ /sys/devices/cpu/events/PM_LSU_LDF
+ /sys/devices/cpu/events/PM_POWER_EVENT3
+ /sys/devices/cpu/events/PM_DISP_WT
+ /sys/devices/cpu/events/PM_IC_BANK_CONFLICT
+ /sys/devices/cpu/events/PM_BR_MPRED_CR_TA
+ /sys/devices/cpu/events/PM_L2_INST_MISS
+ /sys/devices/cpu/events/PM_NEST_PAIR2_ADD
+ /sys/devices/cpu/events/PM_MRK_LSU_FLUSH
+ /sys/devices/cpu/events/PM_L2_LDST
+ /sys/devices/cpu/events/PM_INST_FROM_L31_SHR
+ /sys/devices/cpu/events/PM_VSU0_FIN
+ /sys/devices/cpu/events/PM_VSU1_FCONV
+ /sys/devices/cpu/events/PM_INST_FROM_RMEM
+ /sys/devices/cpu/events/PM_DISP_CLB_HELD_TLBIE
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_DMEM_CYC
+ /sys/devices/cpu/events/PM_BR_PRED_CR
+ /sys/devices/cpu/events/PM_LSU_REJECT
+ /sys/devices/cpu/events/PM_GCT_UTIL_3_TO_6_SLOTS
+ /sys/devices/cpu/events/PM_CMPLU_STALL_END_GCT_NOSLOT
+ /sys/devices/cpu/events/PM_LSU0_REJECT_LMQ_FULL
+ /sys/devices/cpu/events/PM_VSU_FEST
+ /sys/devices/cpu/events/PM_NEST_PAIR0_AND
+ /sys/devices/cpu/events/PM_PTEG_FROM_L3
+ /sys/devices/cpu/events/PM_POWER_EVENT2
+ /sys/devices/cpu/events/PM_IC_PREF_CANCEL_PAGE
+ /sys/devices/cpu/events/PM_VSU0_FSQRT_FDIV
+ /sys/devices/cpu/events/PM_MRK_GRP_CMPL
+ /sys/devices/cpu/events/PM_VSU0_SCAL_DOUBLE_ISSUED
+ /sys/devices/cpu/events/PM_GRP_DISP
+ /sys/devices/cpu/events/PM_LSU0_LDX
+ /sys/devices/cpu/events/PM_DATA_FROM_L2
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_RL2L3_MOD
+ /sys/devices/cpu/events/PM_VSU0_VECT_DOUBLE_ISSUED
+ /sys/devices/cpu/events/PM_VSU1_2FLOP_DOUBLE
+ /sys/devices/cpu/events/PM_THRD_PRIO_6_7_CYC
+ /sys/devices/cpu/events/PM_BC_PLUS_8_RSLV_TAKEN
+ /sys/devices/cpu/events/PM_BR_MPRED_CR
+ /sys/devices/cpu/events/PM_L3_CO_MEM
+ /sys/devices/cpu/events/PM_DATA_FROM_RL2L3_MOD
+ /sys/devices/cpu/events/PM_LSU_SRQ_FULL_CYC
+ /sys/devices/cpu/events/PM_TABLEWALK_CYC
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_RMEM
+ /sys/devices/cpu/events/PM_LSU_SRQ_STFWD
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_RMEM
+ /sys/devices/cpu/events/PM_FXU0_FIN
+ /sys/devices/cpu/events/PM_LSU1_L1_SW_PREF
+ /sys/devices/cpu/events/PM_PTEG_FROM_L31_MOD
+ /sys/devices/cpu/events/PM_PMC5_OVERFLOW
+ /sys/devices/cpu/events/PM_LD_REF_L1_LSU1
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_L21_SHR
+ /sys/devices/cpu/events/PM_DATA_FROM_RMEM
+ /sys/devices/cpu/events/PM_VSU0_SCAL_SINGLE_ISSUED
+ /sys/devices/cpu/events/PM_BR_MPRED_LSTACK
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_RL2L3_MOD_CYC
+ /sys/devices/cpu/events/PM_LSU0_FLUSH_UST
+ /sys/devices/cpu/events/PM_LSU_NCST
+ /sys/devices/cpu/events/PM_BR_TAKEN
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_LMEM
+ /sys/devices/cpu/events/PM_DTLB_MISS_4K
+ /sys/devices/cpu/events/PM_PMC4_SAVED
+ /sys/devices/cpu/events/PM_VSU1_PERMUTE_ISSUED
+ /sys/devices/cpu/events/PM_SLB_MISS
+ /sys/devices/cpu/events/PM_LSU1_FLUSH_LRQ
+ /sys/devices/cpu/events/PM_DTLB_MISS
+ /sys/devices/cpu/events/PM_VSU1_FRSP
+ /sys/devices/cpu/events/PM_VSU_VECTOR_DOUBLE_ISSUED
+ /sys/devices/cpu/events/PM_L2_CASTOUT_SHR
+ /sys/devices/cpu/events/PM_DATA_FROM_DL2L3_SHR
+ /sys/devices/cpu/events/PM_VSU1_STF
+ /sys/devices/cpu/events/PM_ST_FIN
+ /sys/devices/cpu/events/PM_PTEG_FROM_L21_SHR
+ /sys/devices/cpu/events/PM_L2_LOC_GUESS_WRONG
+ /sys/devices/cpu/events/PM_MRK_STCX_FAIL
+ /sys/devices/cpu/events/PM_LSU0_REJECT_LHS
+ /sys/devices/cpu/events/PM_IC_PREF_CANCEL_HIT
+ /sys/devices/cpu/events/PM_L3_PREF_BUSY
+ /sys/devices/cpu/events/PM_MRK_BRU_FIN
+ /sys/devices/cpu/events/PM_LSU1_NCLD
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_L31_MOD
+ /sys/devices/cpu/events/PM_LSU_NCLD
+ /sys/devices/cpu/events/PM_LSU_LDX
+ /sys/devices/cpu/events/PM_L2_LOC_GUESS_CORRECT
+ /sys/devices/cpu/events/PM_THRESH_TIMEO
+ /sys/devices/cpu/events/PM_L3_PREF_ST
+ /sys/devices/cpu/events/PM_DISP_CLB_HELD_SYNC
+ /sys/devices/cpu/events/PM_VSU_SIMPLE_ISSUED
+ /sys/devices/cpu/events/PM_VSU1_SINGLE
+ /sys/devices/cpu/events/PM_DATA_TABLEWALK_CYC
+ /sys/devices/cpu/events/PM_L2_RC_ST_DONE
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_L21_MOD
+ /sys/devices/cpu/events/PM_LARX_LSU1
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_RMEM
+ /sys/devices/cpu/events/PM_DISP_CLB_HELD
+ /sys/devices/cpu/events/PM_DERAT_MISS_4K
+ /sys/devices/cpu/events/PM_L2_RCLD_DISP_FAIL_ADDR
+ /sys/devices/cpu/events/PM_SEG_EXCEPTION
+ /sys/devices/cpu/events/PM_FLUSH_DISP_SB
+ /sys/devices/cpu/events/PM_L2_DC_INV
+ /sys/devices/cpu/events/PM_PTEG_FROM_DL2L3_MOD
+ /sys/devices/cpu/events/PM_DSEG
+ /sys/devices/cpu/events/PM_BR_PRED_LSTACK
+ /sys/devices/cpu/events/PM_VSU0_STF
+ /sys/devices/cpu/events/PM_LSU_FX_FIN
+ /sys/devices/cpu/events/PM_DERAT_MISS_16M
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_DL2L3_MOD
+ /sys/devices/cpu/events/PM_GCT_UTIL_11_PLUS_SLOTS
+ /sys/devices/cpu/events/PM_INST_FROM_L3
+ /sys/devices/cpu/events/PM_MRK_IFU_FIN
+ /sys/devices/cpu/events/PM_ITLB_MISS
+ /sys/devices/cpu/events/PM_VSU_STF
+ /sys/devices/cpu/events/PM_LSU_FLUSH_UST
+ /sys/devices/cpu/events/PM_L2_LDST_MISS
+ /sys/devices/cpu/events/PM_FXU1_FIN
+ /sys/devices/cpu/events/PM_SHL_DEALLOCATED
+ /sys/devices/cpu/events/PM_L2_SN_M_WR_DONE
+ /sys/devices/cpu/events/PM_LSU_REJECT_SET_MPRED
+ /sys/devices/cpu/events/PM_L3_PREF_LD
+ /sys/devices/cpu/events/PM_L2_SN_M_RD_DONE
+ /sys/devices/cpu/events/PM_MRK_DERAT_MISS_16G
+ /sys/devices/cpu/events/PM_VSU_FCONV
+ /sys/devices/cpu/events/PM_ANY_THRD_RUN_CYC
+ /sys/devices/cpu/events/PM_LSU_LMQ_FULL_CYC
+ /sys/devices/cpu/events/PM_MRK_LSU_REJECT_LHS
+ /sys/devices/cpu/events/PM_MRK_LD_MISS_L1_CYC
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L2_CYC
+ /sys/devices/cpu/events/PM_INST_IMC_MATCH_DISP
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_RMEM_CYC
+ /sys/devices/cpu/events/PM_VSU0_SIMPLE_ISSUED
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_RL2L3_SHR
+ /sys/devices/cpu/events/PM_VSU_FMA_DOUBLE
+ /sys/devices/cpu/events/PM_VSU_4FLOP
+ /sys/devices/cpu/events/PM_VSU1_FIN
+ /sys/devices/cpu/events/PM_NEST_PAIR1_AND
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_RL2L3_MOD
+ /sys/devices/cpu/events/PM_PTEG_FROM_RMEM
+ /sys/devices/cpu/events/PM_LSU_LRQ_S0_VALID
+ /sys/devices/cpu/events/PM_LSU0_LDF
+ /sys/devices/cpu/events/PM_FLUSH_COMPLETION
+ /sys/devices/cpu/events/PM_ST_MISS_L1
+ /sys/devices/cpu/events/PM_L2_NODE_PUMP
+ /sys/devices/cpu/events/PM_INST_FROM_DL2L3_SHR
+ /sys/devices/cpu/events/PM_MRK_STALL_CMPLU_CYC
+ /sys/devices/cpu/events/PM_VSU1_DENORM
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L31_SHR_CYC
+ /sys/devices/cpu/events/PM_NEST_PAIR0_ADD
+ /sys/devices/cpu/events/PM_INST_FROM_L3MISS
+ /sys/devices/cpu/events/PM_EE_OFF_EXT_INT
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_DMEM
+ /sys/devices/cpu/events/PM_INST_FROM_DL2L3_MOD
+ /sys/devices/cpu/events/PM_PMC6_OVERFLOW
+ /sys/devices/cpu/events/PM_VSU_2FLOP_DOUBLE
+ /sys/devices/cpu/events/PM_TLB_MISS
+ /sys/devices/cpu/events/PM_FXU_BUSY
+ /sys/devices/cpu/events/PM_L2_RCLD_DISP_FAIL_OTHER
+ /sys/devices/cpu/events/PM_LSU_REJECT_LMQ_FULL
+ /sys/devices/cpu/events/PM_IC_RELOAD_SHR
+ /sys/devices/cpu/events/PM_GRP_MRK
+ /sys/devices/cpu/events/PM_MRK_ST_NEST
+ /sys/devices/cpu/events/PM_VSU1_FSQRT_FDIV
+ /sys/devices/cpu/events/PM_LSU0_FLUSH_LRQ
+ /sys/devices/cpu/events/PM_LARX_LSU0
+ /sys/devices/cpu/events/PM_IBUF_FULL_CYC
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_DL2L3_SHR_CYC
+ /sys/devices/cpu/events/PM_LSU_DC_PREF_STREAM_ALLOC
+ /sys/devices/cpu/events/PM_GRP_MRK_CYC
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_RL2L3_SHR_CYC
+ /sys/devices/cpu/events/PM_L2_GLOB_GUESS_CORRECT
+ /sys/devices/cpu/events/PM_LSU_REJECT_LHS
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_LMEM
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_L3
+ /sys/devices/cpu/events/PM_FREQ_DOWN
+ /sys/devices/cpu/events/PM_PB_RETRY_NODE_PUMP
+ /sys/devices/cpu/events/PM_INST_FROM_RL2L3_SHR
+ /sys/devices/cpu/events/PM_MRK_INST_ISSUED
+ /sys/devices/cpu/events/PM_PTEG_FROM_L3MISS
+ /sys/devices/cpu/events/PM_RUN_PURR
+ /sys/devices/cpu/events/PM_MRK_GRP_IC_MISS
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L3
+ /sys/devices/cpu/events/PM_PTEG_FROM_RL2L3_SHR
+ /sys/devices/cpu/events/PM_LSU_FLUSH_LRQ
+ /sys/devices/cpu/events/PM_MRK_DERAT_MISS_64K
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_DL2L3_MOD
+ /sys/devices/cpu/events/PM_L2_ST_MISS
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_L21_SHR
+ /sys/devices/cpu/events/PM_LWSYNC
+ /sys/devices/cpu/events/PM_LSU0_DC_PREF_STREAM_CONFIRM_STRIDE
+ /sys/devices/cpu/events/PM_MRK_LSU_FLUSH_LRQ
+ /sys/devices/cpu/events/PM_INST_IMC_MATCH_CMPL
+ /sys/devices/cpu/events/PM_NEST_PAIR3_AND
+ /sys/devices/cpu/events/PM_PB_RETRY_SYS_PUMP
+ /sys/devices/cpu/events/PM_MRK_INST_FIN
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_DL2L3_SHR
+ /sys/devices/cpu/events/PM_INST_FROM_L31_MOD
+ /sys/devices/cpu/events/PM_MRK_DTLB_MISS_64K
+ /sys/devices/cpu/events/PM_LSU_FIN
+ /sys/devices/cpu/events/PM_MRK_LSU_REJECT
+ /sys/devices/cpu/events/PM_L2_CO_FAIL_BUSY
+ /sys/devices/cpu/events/PM_MEM0_WQ_DISP
+ /sys/devices/cpu/events/PM_DATA_FROM_L31_MOD
+ /sys/devices/cpu/events/PM_THERMAL_WARN
+ /sys/devices/cpu/events/PM_VSU0_4FLOP
+ /sys/devices/cpu/events/PM_BR_MPRED_CCACHE
+ /sys/devices/cpu/events/PM_L1_DEMAND_WRITE
+ /sys/devices/cpu/events/PM_FLUSH_BR_MPRED
+ /sys/devices/cpu/events/PM_MRK_DTLB_MISS_16G
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_DMEM
+ /sys/devices/cpu/events/PM_L2_RCST_DISP
+ /sys/devices/cpu/events/PM_LSU_PARTIAL_CDF
+ /sys/devices/cpu/events/PM_DISP_CLB_HELD_SB
+ /sys/devices/cpu/events/PM_VSU0_FMA_DOUBLE
+ /sys/devices/cpu/events/PM_FXU0_BUSY_FXU1_IDLE
+ /sys/devices/cpu/events/PM_IC_DEMAND_CYC
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L21_SHR
+ /sys/devices/cpu/events/PM_MRK_LSU_FLUSH_UST
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_L3MISS
+ /sys/devices/cpu/events/PM_VSU_DENORM
+ /sys/devices/cpu/events/PM_MRK_LSU_PARTIAL_CDF
+ /sys/devices/cpu/events/PM_INST_FROM_L21_SHR
+ /sys/devices/cpu/events/PM_IC_PREF_WRITE
+ /sys/devices/cpu/events/PM_BR_PRED
+ /sys/devices/cpu/events/PM_INST_FROM_DMEM
+ /sys/devices/cpu/events/PM_IC_PREF_CANCEL_ALL
+ /sys/devices/cpu/events/PM_LSU_DC_PREF_STREAM_CONFIRM
+ /sys/devices/cpu/events/PM_MRK_LSU_FLUSH_SRQ
+ /sys/devices/cpu/events/PM_MRK_FIN_STALL_CYC
+ /sys/devices/cpu/events/PM_L2_RCST_DISP_FAIL_OTHER
+ /sys/devices/cpu/events/PM_VSU1_DD_ISSUED
+ /sys/devices/cpu/events/PM_PTEG_FROM_L31_SHR
+ /sys/devices/cpu/events/PM_DATA_FROM_L21_SHR
+ /sys/devices/cpu/events/PM_LSU0_NCLD
+ /sys/devices/cpu/events/PM_VSU1_4FLOP
+ /sys/devices/cpu/events/PM_VSU1_8FLOP
+ /sys/devices/cpu/events/PM_VSU_8FLOP
+ /sys/devices/cpu/events/PM_LSU_LMQ_SRQ_EMPTY_CYC
+ /sys/devices/cpu/events/PM_DTLB_MISS_64K
+ /sys/devices/cpu/events/PM_THRD_CONC_RUN_INST
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_L2
+ /sys/devices/cpu/events/PM_PB_SYS_PUMP
+ /sys/devices/cpu/events/PM_VSU_FIN
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L31_MOD
+ /sys/devices/cpu/events/PM_THRD_PRIO_0_1_CYC
+ /sys/devices/cpu/events/PM_DERAT_MISS_64K
+ /sys/devices/cpu/events/PM_PMC2_REWIND
+ /sys/devices/cpu/events/PM_INST_FROM_L2
+ /sys/devices/cpu/events/PM_GRP_BR_MPRED_NONSPEC
+ /sys/devices/cpu/events/PM_INST_DISP
+ /sys/devices/cpu/events/PM_MEM0_RD_CANCEL_TOTAL
+ /sys/devices/cpu/events/PM_LSU0_DC_PREF_STREAM_CONFIRM
+ /sys/devices/cpu/events/PM_L1_DCACHE_RELOAD_VALID
+ /sys/devices/cpu/events/PM_VSU_SCALAR_DOUBLE_ISSUED
+ /sys/devices/cpu/events/PM_L3_PREF_HIT
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_L31_MOD
+ /sys/devices/cpu/events/PM_MRK_FXU_FIN
+ /sys/devices/cpu/events/PM_PMC4_OVERFLOW
+ /sys/devices/cpu/events/PM_MRK_PTEG_FROM_L3
+ /sys/devices/cpu/events/PM_LSU0_LMQ_LHR_MERGE
+ /sys/devices/cpu/events/PM_BTAC_HIT
+ /sys/devices/cpu/events/PM_L3_RD_BUSY
+ /sys/devices/cpu/events/PM_LSU0_L1_SW_PREF
+ /sys/devices/cpu/events/PM_INST_FROM_L2MISS
+ /sys/devices/cpu/events/PM_LSU0_DC_PREF_STREAM_ALLOC
+ /sys/devices/cpu/events/PM_L2_ST
+ /sys/devices/cpu/events/PM_VSU0_DENORM
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_DL2L3_SHR
+ /sys/devices/cpu/events/PM_BR_PRED_CR_TA
+ /sys/devices/cpu/events/PM_VSU0_FCONV
+ /sys/devices/cpu/events/PM_MRK_LSU_FLUSH_ULD
+ /sys/devices/cpu/events/PM_BTAC_MISS
+ /sys/devices/cpu/events/PM_MRK_LD_MISS_EXPOSED_CYC_COUNT
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L2
+ /sys/devices/cpu/events/PM_LSU_DCACHE_RELOAD_VALID
+ /sys/devices/cpu/events/PM_VSU_FMA
+ /sys/devices/cpu/events/PM_LSU0_FLUSH_SRQ
+ /sys/devices/cpu/events/PM_LSU1_L1_PREF
+ /sys/devices/cpu/events/PM_IOPS_CMPL
+ /sys/devices/cpu/events/PM_L2_SYS_PUMP
+ /sys/devices/cpu/events/PM_L2_RCLD_BUSY_RC_FULL
+ /sys/devices/cpu/events/PM_LSU_LMQ_S0_ALLOC
+ /sys/devices/cpu/events/PM_FLUSH_DISP_SYNC
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_DL2L3_MOD_CYC
+ /sys/devices/cpu/events/PM_L2_IC_INV
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L21_MOD_CYC
+ /sys/devices/cpu/events/PM_L3_PREF_LDST
+ /sys/devices/cpu/events/PM_LSU_SRQ_EMPTY_CYC
+ /sys/devices/cpu/events/PM_LSU_LMQ_S0_VALID
+ /sys/devices/cpu/events/PM_FLUSH_PARTIAL
+ /sys/devices/cpu/events/PM_VSU1_FMA_DOUBLE
+ /sys/devices/cpu/events/PM_1PLUS_PPC_DISP
+ /sys/devices/cpu/events/PM_DATA_FROM_L2MISS
+ /sys/devices/cpu/events/PM_SUSPENDED
+ /sys/devices/cpu/events/PM_VSU0_FMA
+ /sys/devices/cpu/events/PM_STCX_FAIL
+ /sys/devices/cpu/events/PM_VSU0_FSQRT_FDIV_DOUBLE
+ /sys/devices/cpu/events/PM_DC_PREF_DST
+ /sys/devices/cpu/events/PM_VSU1_SCAL_SINGLE_ISSUED
+ /sys/devices/cpu/events/PM_L3_HIT
+ /sys/devices/cpu/events/PM_L2_GLOB_GUESS_WRONG
+ /sys/devices/cpu/events/PM_MRK_DFU_FIN
+ /sys/devices/cpu/events/PM_INST_FROM_L1
+ /sys/devices/cpu/events/PM_IC_DEMAND_REQ
+ /sys/devices/cpu/events/PM_VSU1_FSQRT_FDIV_DOUBLE
+ /sys/devices/cpu/events/PM_VSU1_FMA
+ /sys/devices/cpu/events/PM_MRK_LD_MISS_L1
+ /sys/devices/cpu/events/PM_VSU0_2FLOP_DOUBLE
+ /sys/devices/cpu/events/PM_LSU_DC_PREF_STRIDED_STREAM_CONFIRM
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_L31_SHR
+ /sys/devices/cpu/events/PM_MRK_LSU_REJECT_ERAT_MISS
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_L2MISS
+ /sys/devices/cpu/events/PM_DATA_FROM_RL2L3_SHR
+ /sys/devices/cpu/events/PM_INST_FROM_PREF
+ /sys/devices/cpu/events/PM_VSU1_SQ
+ /sys/devices/cpu/events/PM_L2_LD_DISP
+ /sys/devices/cpu/events/PM_L2_DISP_ALL
+ /sys/devices/cpu/events/PM_THRD_GRP_CMPL_BOTH_CYC
+ /sys/devices/cpu/events/PM_VSU_FSQRT_FDIV_DOUBLE
+ /sys/devices/cpu/events/PM_INST_PTEG_FROM_DL2L3_SHR
+ /sys/devices/cpu/events/PM_VSU_1FLOP
+ /sys/devices/cpu/events/PM_HV_CYC
+ /sys/devices/cpu/events/PM_MRK_LSU_FIN
+ /sys/devices/cpu/events/PM_MRK_DATA_FROM_RL2L3_SHR
+ /sys/devices/cpu/events/PM_DTLB_MISS_16M
+ /sys/devices/cpu/events/PM_LSU1_LMQ_LHR_MERGE
+ /sys/devices/cpu/events/PM_IFU_FIN
+ /sys/devices/cpu/events/PM_1THRD_CON_RUN_INSTR
+ /sys/devices/cpu/events/PM_CMPLU_STALL_COUNT
+ /sys/devices/cpu/events/PM_MEM0_PB_RD_CL
+ /sys/devices/cpu/events/PM_THRD_1_RUN_CYC
+ /sys/devices/cpu/events/PM_THRD_2_CONC_RUN_INSTR
+ /sys/devices/cpu/events/PM_THRD_2_RUN_CYC
+ /sys/devices/cpu/events/PM_THRD_3_CONC_RUN_INST
+ /sys/devices/cpu/events/PM_THRD_3_RUN_CYC
+ /sys/devices/cpu/events/PM_THRD_4_CONC_RUN_INST
+ /sys/devices/cpu/events/PM_THRD_4_RUN_CYC
Date: 2013/01/08
diff --git a/arch/powerpc/perf/power7-events-list.h b/arch/powerpc/perf/power7-events-list.h
index 687790a..64f13d9 100644
--- a/arch/powerpc/perf/power7-events-list.h
+++ b/arch/powerpc/perf/power7-events-list.h
@@ -546,3 +546,13 @@ EVENT(PM_MRK_DATA_FROM_RL2L3_SHR, 0x1d04c)
EVENT(PM_DTLB_MISS_16M, 0x4c05e)
EVENT(PM_LSU1_LMQ_LHR_MERGE, 0x0d09a)
EVENT(PM_IFU_FIN, 0x40066)
+EVENT(PM_1THRD_CON_RUN_INSTR, 0x30062)
+EVENT(PM_CMPLU_STALL_COUNT, 0x4000B)
+EVENT(PM_MEM0_PB_RD_CL, 0x30083)
+EVENT(PM_THRD_1_RUN_CYC, 0x10060)
+EVENT(PM_THRD_2_CONC_RUN_INSTR, 0x40062)
+EVENT(PM_THRD_2_RUN_CYC, 0x20060)
+EVENT(PM_THRD_3_CONC_RUN_INST, 0x10062)
+EVENT(PM_THRD_3_RUN_CYC, 0x30060)
+EVENT(PM_THRD_4_CONC_RUN_INST, 0x20062)
+EVENT(PM_THRD_4_RUN_CYC, 0x40060)
--
1.7.11.7
^ permalink raw reply related
* Re: [PATCH] powerpc: thp: Fix crash on mremap
From: Kirill A. Shutemov @ 2014-01-02 2:19 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Aneesh Kumar K.V
Cc: aarcange, linuxppc-dev, paulus, kirill.shutemov, linux-mm
In-Reply-To: <1388572145.4373.41.camel@pasglop>
On Wed, Jan 01, 2014 at 09:29:05PM +1100, Benjamin Herrenschmidt wrote:
> On Wed, 2014-01-01 at 15:23 +0530, Aneesh Kumar K.V wrote:
> > From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
> >
> > This patch fix the below crash
> >
> > NIP [c00000000004cee4] .__hash_page_thp+0x2a4/0x440
> > LR [c0000000000439ac] .hash_page+0x18c/0x5e0
> > ...
> > Call Trace:
> > [c000000736103c40] [00001ffffb000000] 0x1ffffb000000(unreliable)
> > [437908.479693] [c000000736103d50] [c0000000000439ac] .hash_page+0x18c/0x5e0
> > [437908.479699] [c000000736103e30] [c00000000000924c] .do_hash_page+0x4c/0x58
> >
> > On ppc64 we use the pgtable for storing the hpte slot information and
> > store address to the pgtable at a constant offset (PTRS_PER_PMD) from
> > pmd. On mremap, when we switch the pmd, we need to withdraw and deposit
> > the pgtable again, so that we find the pgtable at PTRS_PER_PMD offset
> > from new pmd.
> >
> > We also want to move the withdraw and deposit before the set_pmd so
> > that, when page fault find the pmd as trans huge we can be sure that
> > pgtable can be located at the offset.
> >
> > Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
> > ---
> > NOTE:
> > For other archs we would just be removing the pgtable from the list and adding it back.
> > I didn't find an easy way to make it not do that without lots of #ifdef around. Any
> > suggestion around that is welcome.
>
> What about
>
> - if (new_ptl != old_ptl) {
> + if (new_ptl != old_ptl || ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW) {
>
> Or something similar ?
Looks sane to me. Or something with IS_ENABLED(), if needed.
>
> Cheers,
> Ben.
>
> > mm/huge_memory.c | 21 ++++++++++-----------
> > 1 file changed, 10 insertions(+), 11 deletions(-)
> >
> > diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> > index 7de1bf85f683..eb2e60d9ba45 100644
> > --- a/mm/huge_memory.c
> > +++ b/mm/huge_memory.c
> > @@ -1500,24 +1500,23 @@ int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
> > */
> > ret = __pmd_trans_huge_lock(old_pmd, vma, &old_ptl);
> > if (ret == 1) {
> > + pgtable_t pgtable;
> > +
> > new_ptl = pmd_lockptr(mm, new_pmd);
> > if (new_ptl != old_ptl)
> > spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
> > pmd = pmdp_get_and_clear(mm, old_addr, old_pmd);
> > VM_BUG_ON(!pmd_none(*new_pmd));
> > + /*
> > + * Archs like ppc64 use pgtable to store per pmd
> > + * specific information. So when we switch the pmd,
> > + * we should also withdraw and deposit the pgtable
> > + */
> > + pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
> > + pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
> > set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
> > - if (new_ptl != old_ptl) {
> > - pgtable_t pgtable;
> > -
> > - /*
> > - * Move preallocated PTE page table if new_pmd is on
> > - * different PMD page table.
> > - */
Please don't lose the comment.
> > - pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
> > - pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
> > -
> > + if (new_ptl != old_ptl)
> > spin_unlock(new_ptl);
> > - }
> > spin_unlock(old_ptl);
> > }
> > out:
>
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
--
Kirill A. Shutemov
^ permalink raw reply
* Re: [PATCH] powerpc: thp: Fix crash on mremap
From: Benjamin Herrenschmidt @ 2014-01-01 10:29 UTC (permalink / raw)
To: Aneesh Kumar K.V
Cc: aarcange, linuxppc-dev, paulus, kirill.shutemov, linux-mm
In-Reply-To: <1388570027-22933-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com>
On Wed, 2014-01-01 at 15:23 +0530, Aneesh Kumar K.V wrote:
> From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
>
> This patch fix the below crash
>
> NIP [c00000000004cee4] .__hash_page_thp+0x2a4/0x440
> LR [c0000000000439ac] .hash_page+0x18c/0x5e0
> ...
> Call Trace:
> [c000000736103c40] [00001ffffb000000] 0x1ffffb000000(unreliable)
> [437908.479693] [c000000736103d50] [c0000000000439ac] .hash_page+0x18c/0x5e0
> [437908.479699] [c000000736103e30] [c00000000000924c] .do_hash_page+0x4c/0x58
>
> On ppc64 we use the pgtable for storing the hpte slot information and
> store address to the pgtable at a constant offset (PTRS_PER_PMD) from
> pmd. On mremap, when we switch the pmd, we need to withdraw and deposit
> the pgtable again, so that we find the pgtable at PTRS_PER_PMD offset
> from new pmd.
>
> We also want to move the withdraw and deposit before the set_pmd so
> that, when page fault find the pmd as trans huge we can be sure that
> pgtable can be located at the offset.
>
> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
> ---
> NOTE:
> For other archs we would just be removing the pgtable from the list and adding it back.
> I didn't find an easy way to make it not do that without lots of #ifdef around. Any
> suggestion around that is welcome.
What about
- if (new_ptl != old_ptl) {
+ if (new_ptl != old_ptl || ARCH_THP_MOVE_PMD_ALWAYS_WITHDRAW) {
Or something similar ?
Cheers,
Ben.
> mm/huge_memory.c | 21 ++++++++++-----------
> 1 file changed, 10 insertions(+), 11 deletions(-)
>
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 7de1bf85f683..eb2e60d9ba45 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -1500,24 +1500,23 @@ int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
> */
> ret = __pmd_trans_huge_lock(old_pmd, vma, &old_ptl);
> if (ret == 1) {
> + pgtable_t pgtable;
> +
> new_ptl = pmd_lockptr(mm, new_pmd);
> if (new_ptl != old_ptl)
> spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
> pmd = pmdp_get_and_clear(mm, old_addr, old_pmd);
> VM_BUG_ON(!pmd_none(*new_pmd));
> + /*
> + * Archs like ppc64 use pgtable to store per pmd
> + * specific information. So when we switch the pmd,
> + * we should also withdraw and deposit the pgtable
> + */
> + pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
> + pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
> set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
> - if (new_ptl != old_ptl) {
> - pgtable_t pgtable;
> -
> - /*
> - * Move preallocated PTE page table if new_pmd is on
> - * different PMD page table.
> - */
> - pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
> - pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
> -
> + if (new_ptl != old_ptl)
> spin_unlock(new_ptl);
> - }
> spin_unlock(old_ptl);
> }
> out:
^ permalink raw reply
* [PATCH] powerpc: thp: Fix crash on mremap
From: Aneesh Kumar K.V @ 2014-01-01 9:53 UTC (permalink / raw)
To: benh, paulus, aarcange, kirill.shutemov
Cc: linux-mm, linuxppc-dev, Aneesh Kumar K.V
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
This patch fix the below crash
NIP [c00000000004cee4] .__hash_page_thp+0x2a4/0x440
LR [c0000000000439ac] .hash_page+0x18c/0x5e0
...
Call Trace:
[c000000736103c40] [00001ffffb000000] 0x1ffffb000000(unreliable)
[437908.479693] [c000000736103d50] [c0000000000439ac] .hash_page+0x18c/0x5e0
[437908.479699] [c000000736103e30] [c00000000000924c] .do_hash_page+0x4c/0x58
On ppc64 we use the pgtable for storing the hpte slot information and
store address to the pgtable at a constant offset (PTRS_PER_PMD) from
pmd. On mremap, when we switch the pmd, we need to withdraw and deposit
the pgtable again, so that we find the pgtable at PTRS_PER_PMD offset
from new pmd.
We also want to move the withdraw and deposit before the set_pmd so
that, when page fault find the pmd as trans huge we can be sure that
pgtable can be located at the offset.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
NOTE:
For other archs we would just be removing the pgtable from the list and adding it back.
I didn't find an easy way to make it not do that without lots of #ifdef around. Any
suggestion around that is welcome.
mm/huge_memory.c | 21 ++++++++++-----------
1 file changed, 10 insertions(+), 11 deletions(-)
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 7de1bf85f683..eb2e60d9ba45 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1500,24 +1500,23 @@ int move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
*/
ret = __pmd_trans_huge_lock(old_pmd, vma, &old_ptl);
if (ret == 1) {
+ pgtable_t pgtable;
+
new_ptl = pmd_lockptr(mm, new_pmd);
if (new_ptl != old_ptl)
spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
pmd = pmdp_get_and_clear(mm, old_addr, old_pmd);
VM_BUG_ON(!pmd_none(*new_pmd));
+ /*
+ * Archs like ppc64 use pgtable to store per pmd
+ * specific information. So when we switch the pmd,
+ * we should also withdraw and deposit the pgtable
+ */
+ pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
+ pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
set_pmd_at(mm, new_addr, new_pmd, pmd_mksoft_dirty(pmd));
- if (new_ptl != old_ptl) {
- pgtable_t pgtable;
-
- /*
- * Move preallocated PTE page table if new_pmd is on
- * different PMD page table.
- */
- pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
- pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
-
+ if (new_ptl != old_ptl)
spin_unlock(new_ptl);
- }
spin_unlock(old_ptl);
}
out:
--
1.8.3.2
^ permalink raw reply related
* [PATCH V2] time/cpuidle: Support in tick broadcast framework for archs without external clock device
From: Preeti U Murthy @ 2014-01-01 5:49 UTC (permalink / raw)
To: peterz, fweisbec, paul.gortmaker, paulus, mingo, shangw,
rafael.j.wysocki, galak, benh, paulmck, arnd, linux-pm, rostedt,
michael, john.stultz, tglx, chenhui.zhao, deepthi, r58472, geoff,
linux-kernel, srivatsa.bhat, schwidefsky, svaidy, linuxppc-dev
On some architectures, in certain CPU deep idle states the local timers stop.
An external clock device is used to wakeup these CPUs. The kernel support for the
wakeup of these CPUs is provided by the tick broadcast framework by using the
external clock device as the wakeup source.
However not all implementations of architectures provide such an external
clock device such as some PowerPC ones. This patch includes support in the
broadcast framework to handle the wakeup of the CPUs in deep idle states on such
systems by queuing a hrtimer on one of the CPUs, meant to handle the wakeup of
CPUs in deep idle states. This CPU is identified as the bc_cpu.
Each time the hrtimer expires, it is reprogrammed for the next wakeup of the
CPUs in deep idle state after handling broadcast. However when a CPU is about
to enter deep idle state with its wakeup time earlier than the time at which
the hrtimer is currently programmed, it *becomes the new bc_cpu* and restarts
the hrtimer on itself. This way the job of doing broadcast is handed around to
the CPUs that ask for the earliest wakeup just before entering deep idle
state. This is consistent with what happens in cases where an external clock
device is present. The smp affinity of this clock device is set to the CPU
with the earliest wakeup.
The important point here is that the bc_cpu cannot enter deep idle state
since it has a hrtimer queued to wakeup the other CPUs in deep idle. Hence it
cannot have its local timer stopped. Therefore for such a CPU, the
BROADCAST_ENTER notification has to fail implying that it cannot enter deep
idle state. On architectures where an external clock device is present, all
CPUs can enter deep idle.
During hotplug of the bc_cpu, the job of doing a broadcast is assigned to the
first cpu in the broadcast mask. This newly nominated bc_cpu is woken up by
an IPI so as to queue the above mentioned hrtimer on itself.
Changes from V1:https://lkml.org/lkml/2013/12/12/687
If idle states exist when the local timers of CPUs stop and
there is no external clock device to handle their wakeups the kernel switches
the tick mode to periodic so as to prevent the CPUs from entering such idle
states altogether. Therefore include an additional check consistent
with this patch, where if an external clock device does not exist, queue a
hrtimer to handle wakeups. If this also fails, only then switch the tick mode to
periodic.
Signed-off-by: Preeti U Murthy <preeti@linux.vnet.ibm.com>
---
include/linux/clockchips.h | 4 -
kernel/time/clockevents.c | 8 +-
kernel/time/tick-broadcast.c | 180 ++++++++++++++++++++++++++++++++++++++----
kernel/time/tick-internal.h | 8 +-
4 files changed, 173 insertions(+), 27 deletions(-)
diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index 493aa02..bbda37b 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -186,9 +186,9 @@ static inline int tick_check_broadcast_expired(void) { return 0; }
#endif
#ifdef CONFIG_GENERIC_CLOCKEVENTS
-extern void clockevents_notify(unsigned long reason, void *arg);
+extern int clockevents_notify(unsigned long reason, void *arg);
#else
-static inline void clockevents_notify(unsigned long reason, void *arg) {}
+static inline int clockevents_notify(unsigned long reason, void *arg) {}
#endif
#else /* CONFIG_GENERIC_CLOCKEVENTS_BUILD */
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index 086ad60..bbbd671 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -525,11 +525,11 @@ void clockevents_resume(void)
/**
* clockevents_notify - notification about relevant events
*/
-void clockevents_notify(unsigned long reason, void *arg)
+int clockevents_notify(unsigned long reason, void *arg)
{
struct clock_event_device *dev, *tmp;
unsigned long flags;
- int cpu;
+ int cpu, ret = 0;
raw_spin_lock_irqsave(&clockevents_lock, flags);
@@ -542,11 +542,12 @@ void clockevents_notify(unsigned long reason, void *arg)
case CLOCK_EVT_NOTIFY_BROADCAST_ENTER:
case CLOCK_EVT_NOTIFY_BROADCAST_EXIT:
- tick_broadcast_oneshot_control(reason);
+ ret = tick_broadcast_oneshot_control(reason);
break;
case CLOCK_EVT_NOTIFY_CPU_DYING:
tick_handover_do_timer(arg);
+ tick_handover_bc_cpu(arg);
break;
case CLOCK_EVT_NOTIFY_SUSPEND:
@@ -585,6 +586,7 @@ void clockevents_notify(unsigned long reason, void *arg)
break;
}
raw_spin_unlock_irqrestore(&clockevents_lock, flags);
+ return ret;
}
EXPORT_SYMBOL_GPL(clockevents_notify);
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 9532690..1755984 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -20,6 +20,7 @@
#include <linux/sched.h>
#include <linux/smp.h>
#include <linux/module.h>
+#include <linux/slab.h>
#include "tick-internal.h"
@@ -35,6 +36,11 @@ static cpumask_var_t tmpmask;
static DEFINE_RAW_SPINLOCK(tick_broadcast_lock);
static int tick_broadcast_force;
+static struct hrtimer *bc_hrtimer;
+static int bc_cpu = -1;
+static ktime_t bc_next_wakeup;
+static int hrtimer_initialized = 0;
+
#ifdef CONFIG_TICK_ONESHOT
static void tick_broadcast_clear_oneshot(int cpu);
#else
@@ -528,6 +534,20 @@ static int tick_broadcast_set_event(struct clock_event_device *bc, int cpu,
return ret;
}
+static void tick_broadcast_set_next_wakeup(int cpu, ktime_t expires, int force)
+{
+ struct clock_event_device *bc;
+
+ bc = tick_broadcast_device.evtdev;
+
+ if (bc) {
+ tick_broadcast_set_event(bc, cpu, expires, force);
+ } else {
+ hrtimer_start(bc_hrtimer, expires, HRTIMER_MODE_ABS_PINNED);
+ bc_cpu = cpu;
+ }
+}
+
int tick_resume_broadcast_oneshot(struct clock_event_device *bc)
{
clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT);
@@ -558,15 +578,13 @@ void tick_check_oneshot_broadcast(int cpu)
/*
* Handle oneshot mode broadcasting
*/
-static void tick_handle_oneshot_broadcast(struct clock_event_device *dev)
+static int tick_oneshot_broadcast(void)
{
struct tick_device *td;
ktime_t now, next_event;
int cpu, next_cpu = 0;
- raw_spin_lock(&tick_broadcast_lock);
-again:
- dev->next_event.tv64 = KTIME_MAX;
+ bc_next_wakeup.tv64 = KTIME_MAX;
next_event.tv64 = KTIME_MAX;
cpumask_clear(tmpmask);
now = ktime_get();
@@ -620,34 +638,89 @@ again:
* in the event mask
*/
if (next_event.tv64 != KTIME_MAX) {
- /*
- * Rearm the broadcast device. If event expired,
- * repeat the above
- */
- if (tick_broadcast_set_event(dev, next_cpu, next_event, 0))
+ bc_next_wakeup = next_event;
+ }
+
+ return next_cpu;
+}
+
+/*
+ * Handler in oneshot mode for the external clock device
+ */
+static void tick_handle_oneshot_broadcast(struct clock_event_device *dev)
+{
+ int next_cpu;
+
+ raw_spin_lock(&tick_broadcast_lock);
+
+again: next_cpu = tick_oneshot_broadcast();
+ /*
+ * Rearm the broadcast device. If event expired,
+ * repeat the above
+ */
+ if (bc_next_wakeup.tv64 != KTIME_MAX)
+ if (tick_broadcast_set_event(dev, next_cpu, bc_next_wakeup, 0))
goto again;
+
+ raw_spin_unlock(&tick_broadcast_lock);
+}
+
+/*
+ * Handler in oneshot mode for the hrtimer queued when there is no external
+ * clock device.
+ */
+static enum hrtimer_restart handle_broadcast(struct hrtimer *hrtmr)
+{
+ ktime_t now, interval;
+
+ raw_spin_lock(&tick_broadcast_lock);
+ tick_oneshot_broadcast();
+
+ now = ktime_get();
+
+ if (bc_next_wakeup.tv64 != KTIME_MAX) {
+ interval = ktime_sub(bc_next_wakeup, now);
+ hrtimer_forward_now(bc_hrtimer, interval);
+ raw_spin_unlock(&tick_broadcast_lock);
+ return HRTIMER_RESTART;
}
raw_spin_unlock(&tick_broadcast_lock);
+ return HRTIMER_NORESTART;
+}
+
+/* The CPU could be asked to take over from the previous bc_cpu,
+ * if it is being hotplugged out.
+ */
+static void tick_broadcast_exit_check(int cpu)
+{
+ if (cpu == bc_cpu)
+ hrtimer_start(bc_hrtimer, bc_next_wakeup,
+ HRTIMER_MODE_ABS_PINNED);
+}
+
+static int can_enter_broadcast(int cpu)
+{
+ return cpu != bc_cpu;
}
/*
* Powerstate information: The system enters/leaves a state, where
* affected devices might stop
*/
-void tick_broadcast_oneshot_control(unsigned long reason)
+int tick_broadcast_oneshot_control(unsigned long reason)
{
- struct clock_event_device *bc, *dev;
+ struct clock_event_device *dev;
struct tick_device *td;
unsigned long flags;
ktime_t now;
- int cpu;
+ int cpu, ret = 0;
/*
* Periodic mode does not care about the enter/exit of power
* states
*/
if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC)
- return;
+ return ret;
/*
* We are called with preemtion disabled from the depth of the
@@ -658,9 +731,8 @@ void tick_broadcast_oneshot_control(unsigned long reason)
dev = td->evtdev;
if (!(dev->features & CLOCK_EVT_FEAT_C3STOP))
- return;
+ return ret;
- bc = tick_broadcast_device.evtdev;
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
if (reason == CLOCK_EVT_NOTIFY_BROADCAST_ENTER) {
@@ -676,12 +748,22 @@ void tick_broadcast_oneshot_control(unsigned long reason)
* woken by the IPI right away.
*/
if (!cpumask_test_cpu(cpu, tick_broadcast_force_mask) &&
- dev->next_event.tv64 < bc->next_event.tv64)
- tick_broadcast_set_event(bc, cpu, dev->next_event, 1);
+ dev->next_event.tv64 < bc_next_wakeup.tv64) {
+ bc_next_wakeup = dev->next_event;
+ tick_broadcast_set_next_wakeup(cpu, dev->next_event, 1);
+ }
+
+ if (!can_enter_broadcast(cpu)) {
+ cpumask_test_and_clear_cpu(cpu, tick_broadcast_oneshot_mask);
+ clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT);
+ ret = 1;
+ }
}
} else {
if (cpumask_test_and_clear_cpu(cpu, tick_broadcast_oneshot_mask)) {
clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT);
+
+ tick_broadcast_exit_check(cpu);
/*
* The cpu which was handling the broadcast
* timer marked this cpu in the broadcast
@@ -746,6 +828,7 @@ void tick_broadcast_oneshot_control(unsigned long reason)
}
out:
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
+ return ret;
}
/*
@@ -821,18 +904,56 @@ void tick_broadcast_switch_to_oneshot(void)
{
struct clock_event_device *bc;
unsigned long flags;
+ int cpu = smp_processor_id();
raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
+ bc_next_wakeup.tv64 = KTIME_MAX;
+
tick_broadcast_device.mode = TICKDEV_MODE_ONESHOT;
bc = tick_broadcast_device.evtdev;
- if (bc)
+ if (bc) {
tick_broadcast_setup_oneshot(bc);
+ bc_next_wakeup = bc->next_event;
+ } else if (hrtimer_initialized){
+
+ /*
+ * There may be CPUs waiting for periodic broadcast. We need
+ * to set the oneshot bits for those and program the hrtimer
+ * to fire at the next tick period.
+ */
+ cpumask_copy(tmpmask, tick_broadcast_mask);
+ cpumask_clear_cpu(cpu, tmpmask);
+ cpumask_or(tick_broadcast_oneshot_mask,
+ tick_broadcast_oneshot_mask, tmpmask);
+
+ if (!cpumask_empty(tmpmask)) {
+ tick_broadcast_init_next_event(tmpmask,
+ tick_next_period);
+ hrtimer_start(bc_hrtimer, tick_next_period, HRTIMER_MODE_ABS_PINNED);
+ bc_next_wakeup = tick_next_period;
+ }
+ }
raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
}
+void tick_handover_bc_cpu(int *cpup)
+{
+ struct tick_device *td;
+
+ if (*cpup == bc_cpu) {
+ int cpu = cpumask_first(tick_broadcast_oneshot_mask);
+
+ bc_cpu = (cpu < nr_cpu_ids) ? cpu : -1;
+ if (bc_cpu != -1) {
+ td = &per_cpu(tick_cpu_device, bc_cpu);
+ td->evtdev->broadcast(cpumask_of(bc_cpu));
+ }
+ }
+}
+
/*
* Remove a dead CPU from broadcasting
*/
@@ -868,8 +989,29 @@ int tick_broadcast_oneshot_active(void)
bool tick_broadcast_oneshot_available(void)
{
struct clock_event_device *bc = tick_broadcast_device.evtdev;
+ bool ret = true;
+ unsigned long flags;
- return bc ? bc->features & CLOCK_EVT_FEAT_ONESHOT : false;
+ raw_spin_lock_irqsave(&tick_broadcast_lock, flags);
+
+ if (bc) {
+ ret = bc->features & CLOCK_EVT_FEAT_ONESHOT;
+ } else if (!hrtimer_initialized) {
+ /* An alternative to tick_broadcast_device on archs which do not have
+ * an external device
+ */
+ bc_hrtimer = kmalloc(sizeof(*bc_hrtimer), GFP_NOWAIT);
+ if (!bc_hrtimer) {
+ ret = false;
+ goto out;
+ }
+ hrtimer_init(bc_hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED);
+ bc_hrtimer->function = handle_broadcast;
+ hrtimer_initialized = 1;
+ }
+
+out: raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags);
+ return ret;
}
#endif
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index 18e71f7..1f73032 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -46,23 +46,25 @@ extern int tick_switch_to_oneshot(void (*handler)(struct clock_event_device *));
extern void tick_resume_oneshot(void);
# ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST
extern void tick_broadcast_setup_oneshot(struct clock_event_device *bc);
-extern void tick_broadcast_oneshot_control(unsigned long reason);
+extern int tick_broadcast_oneshot_control(unsigned long reason);
extern void tick_broadcast_switch_to_oneshot(void);
extern void tick_shutdown_broadcast_oneshot(unsigned int *cpup);
extern int tick_resume_broadcast_oneshot(struct clock_event_device *bc);
extern int tick_broadcast_oneshot_active(void);
extern void tick_check_oneshot_broadcast(int cpu);
+extern void tick_handover_bc_cpu(int *cpup);
bool tick_broadcast_oneshot_available(void);
# else /* BROADCAST */
static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
{
BUG();
}
-static inline void tick_broadcast_oneshot_control(unsigned long reason) { }
+static inline int tick_broadcast_oneshot_control(unsigned long reason) { }
static inline void tick_broadcast_switch_to_oneshot(void) { }
static inline void tick_shutdown_broadcast_oneshot(unsigned int *cpup) { }
static inline int tick_broadcast_oneshot_active(void) { return 0; }
static inline void tick_check_oneshot_broadcast(int cpu) { }
+static inline void tick_handover_bc_cpu(int *cpup) {}
static inline bool tick_broadcast_oneshot_available(void) { return true; }
# endif /* !BROADCAST */
@@ -87,7 +89,7 @@ static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc)
{
BUG();
}
-static inline void tick_broadcast_oneshot_control(unsigned long reason) { }
+static inline int tick_broadcast_oneshot_control(unsigned long reason) { }
static inline void tick_shutdown_broadcast_oneshot(unsigned int *cpup) { }
static inline int tick_resume_broadcast_oneshot(struct clock_event_device *bc)
{
^ permalink raw reply related
* Re: [PATCH 1/4] powerpc/eeh: Add restore_bars operation
From: Gavin Shan @ 2014-01-01 3:29 UTC (permalink / raw)
To: Benjamin Herrenschmidt; +Cc: linuxppc-dev, Gavin Shan
In-Reply-To: <1388182804.4373.17.camel@pasglop>
On Sat, Dec 28, 2013 at 09:20:04AM +1100, Benjamin Herrenschmidt wrote:
>On Wed, 2013-12-25 at 16:58 +0800, Gavin Shan wrote:
>> After reset on the specific PE or PHB, we never configure AER
>> correctly on PowerNV platform. We needn't care it on pSeries
>> platform. The patch introduces additional EEH operation eeh_ops::
>> restore_bars() so that we have chance to configure AER correctly
>> for PowerNV platform.
>
>Why call it "restore_bars" if it restores something else (in this case
>AER) ?
>
>I would call it "restore_config" instead... Also rather than adding
>the knowledge of what AER bit works or not for the device, would it
>make sense to instead have a FW call to re-init the device ?
>
>Otherwise, you introduce duplication between Linux and Firmware with
>the risk of getting out of sync...
>
Thanks for your comments, Ben. It's reasonable to have name "restore_config"
and I'll introduce a FW call in next revision as you suggested :-)
Thanks,
Gavin
>> Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
>> ---
>> arch/powerpc/include/asm/eeh.h | 1 +
>> arch/powerpc/kernel/eeh_pe.c | 3 +++
>> arch/powerpc/platforms/pseries/eeh_pseries.c | 4 +++-
>> 3 files changed, 7 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/powerpc/include/asm/eeh.h b/arch/powerpc/include/asm/eeh.h
>> index d3e5e9b..4b709bf 100644
>> --- a/arch/powerpc/include/asm/eeh.h
>> +++ b/arch/powerpc/include/asm/eeh.h
>> @@ -157,6 +157,7 @@ struct eeh_ops {
>> int (*read_config)(struct device_node *dn, int where, int size, u32 *val);
>> int (*write_config)(struct device_node *dn, int where, int size, u32 val);
>> int (*next_error)(struct eeh_pe **pe);
>> + void (*restore_bars)(struct device_node *dn);
>> };
>>
>> extern struct eeh_ops *eeh_ops;
>> diff --git a/arch/powerpc/kernel/eeh_pe.c b/arch/powerpc/kernel/eeh_pe.c
>> index f945053..19eb95a 100644
>> --- a/arch/powerpc/kernel/eeh_pe.c
>> +++ b/arch/powerpc/kernel/eeh_pe.c
>> @@ -737,6 +737,9 @@ static void *eeh_restore_one_device_bars(void *data, void *flag)
>> else
>> eeh_restore_device_bars(edev, dn);
>>
>> + if (eeh_ops->restore_bars)
>> + eeh_ops->restore_bars(dn);
>> +
>> return NULL;
>> }
>>
>> diff --git a/arch/powerpc/platforms/pseries/eeh_pseries.c b/arch/powerpc/platforms/pseries/eeh_pseries.c
>> index ccb633e..623adaf 100644
>> --- a/arch/powerpc/platforms/pseries/eeh_pseries.c
>> +++ b/arch/powerpc/platforms/pseries/eeh_pseries.c
>> @@ -689,7 +689,9 @@ static struct eeh_ops pseries_eeh_ops = {
>> .get_log = pseries_eeh_get_log,
>> .configure_bridge = pseries_eeh_configure_bridge,
>> .read_config = pseries_eeh_read_config,
>> - .write_config = pseries_eeh_write_config
>> + .write_config = pseries_eeh_write_config,
>> + .next_error = NULL,
>> + .restore_bars = NULL
>> };
>>
>> /**
>
>
^ permalink raw reply
* Where to post problems with linux-ppc for Freescale processors such as the p1020?
From: jeclark2006 @ 2013-12-31 20:35 UTC (permalink / raw)
To: linuxppc-dev
As the subject says... where to post problems with ppc & linux that does not
involve 'apple' implementations.
The 'embedded' list appears to be pretty much dead.
--
View this message in context: http://linuxppc.10917.n7.nabble.com/Where-to-post-problems-with-linux-ppc-for-Freescale-processors-such-as-the-p1020-tp79016.html
Sent from the linuxppc-dev mailing list archive at Nabble.com.
^ permalink raw reply
* RE: [PATCH] ASoC: fsl_sai: Fix one bug for hardware limitation.
From: Li.Xiubo @ 2013-12-31 3:29 UTC (permalink / raw)
To: Mark Brown
Cc: alsa-devel@alsa-project.org, lgirdwood@gmail.com, tiwai@suse.de,
linux-kernel@vger.kernel.org, timur@tabi.org, perex@perex.cz,
shawn.guo@linaro.org, linuxppc-dev@lists.ozlabs.org
In-Reply-To: <20131230121520.GR31886@sirena.org.uk>
Hi Mark,
> Subject: Re: [PATCH] ASoC: fsl_sai: Fix one bug for hardware limitation.
>=20
> On Thu, Dec 26, 2013 at 10:57:22AM +0800, Xiubo Li wrote:
> > This is maybe one bug or a limitation of the hardware that the {T,R}CR2=
's
> > Synchronous Mode bits must be set as late as possible, or the SAI devic=
e
> > maybe hanged up, and there has not any explaination about this limitati=
on
> > in the SAI Data Sheet.
>=20
> If they really should be set as late as possible then I'd expect them to
> be being set in the trigger function, that's what gets called as the
> audio actually starts streaming. This is moving it to the startup
> function which is one of the first things that happens during audio
> stream setup so presumably makes things worse.
>=20
Yes, it is.
If everything is OK, I will send another new patch later.
Thanks,
---
Best Regards,
Xiubo
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox