* [PATCH] powerpc/powernv: Fix BE PE number to opal_pci_eeh_freeze_clear()
@ 2016-08-01 7:03 Gavin Shan
2016-08-01 10:50 ` Michael Ellerman
0 siblings, 1 reply; 4+ messages in thread
From: Gavin Shan @ 2016-08-01 7:03 UTC (permalink / raw)
To: linuxppc-dev; +Cc: paulus, benh, mpe, ruscur, Gavin Shan
The PE number (@frozen_pe_no), filled by opal_pci_next_error() is
in big-endian format. Frozen PE number in CPU-endian should be passed
to opal_pci_eeh_freeze_clear() when clearing the frozen state if the
PE is invalid one.
This passes CPU-endian PE number to opal_pci_eeh_freeze_clear() and
it should be part of commit <0f36db77643b> ("powerpc/eeh: Fix wrong
printed PE number").
Cc: stable@vger.kernel.org # v4.3+
Suggested-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
---
arch/powerpc/platforms/powernv/eeh-powernv.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index 86544ea..75363d9 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -1538,7 +1538,7 @@ static int pnv_eeh_next_error(struct eeh_pe **pe)
/* Try best to clear it */
opal_pci_eeh_freeze_clear(phb->opal_id,
- frozen_pe_no,
+ be64_to_cpu(frozen_pe_no),
OPAL_EEH_ACTION_CLEAR_FREEZE_ALL);
ret = EEH_NEXT_ERR_NONE;
} else if ((*pe)->state & EEH_PE_ISOLATED ||
--
2.1.0
^ permalink raw reply related [flat|nested] 4+ messages in thread* Re: [PATCH] powerpc/powernv: Fix BE PE number to opal_pci_eeh_freeze_clear()
2016-08-01 7:03 [PATCH] powerpc/powernv: Fix BE PE number to opal_pci_eeh_freeze_clear() Gavin Shan
@ 2016-08-01 10:50 ` Michael Ellerman
2016-08-01 23:59 ` Gavin Shan
2016-08-02 3:54 ` Gavin Shan
0 siblings, 2 replies; 4+ messages in thread
From: Michael Ellerman @ 2016-08-01 10:50 UTC (permalink / raw)
To: Gavin Shan, linuxppc-dev; +Cc: paulus, benh, ruscur, Gavin Shan
Gavin Shan <gwshan@linux.vnet.ibm.com> writes:
> The PE number (@frozen_pe_no), filled by opal_pci_next_error() is
> in big-endian format. Frozen PE number in CPU-endian should be passed
> to opal_pci_eeh_freeze_clear() when clearing the frozen state if the
> PE is invalid one.
>
> This passes CPU-endian PE number to opal_pci_eeh_freeze_clear() and
> it should be part of commit <0f36db77643b> ("powerpc/eeh: Fix wrong
> printed PE number").
Which commit added the broken usage?
Looks like 2a485ad7c88d ("powerpc/powernv: Drop PHB operation
next_error()") ?
> Cc: stable@vger.kernel.org # v4.3+
> Suggested-by: Paul Mackerras <paulus@samba.org>
> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
> ---
> arch/powerpc/platforms/powernv/eeh-powernv.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
> index 86544ea..75363d9 100644
> --- a/arch/powerpc/platforms/powernv/eeh-powernv.c
> +++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
> @@ -1538,7 +1538,7 @@ static int pnv_eeh_next_error(struct eeh_pe **pe)
>
> /* Try best to clear it */
> opal_pci_eeh_freeze_clear(phb->opal_id,
> - frozen_pe_no,
> + be64_to_cpu(frozen_pe_no),
Sparse knew about this:
arch/powerpc/platforms/powernv/eeh-powernv.c:1541:41: warning: incorrect type in argument 2 (different base types)
arch/powerpc/platforms/powernv/eeh-powernv.c:1541:41: expected unsigned long long [unsigned] [usertype] pe_number
arch/powerpc/platforms/powernv/eeh-powernv.c:1541:41: got restricted __be64 [addressable] [usertype] frozen_pe_no
And finds some other problems in that file:
arch/powerpc/platforms/powernv/eeh-powernv.c:875:23: warning: constant 0x8000000000000000 is so big it is unsigned long
arch/powerpc/platforms/powernv/eeh-powernv.c:1309:21: warning: restricted __be16 degrades to integer
cheers
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] powerpc/powernv: Fix BE PE number to opal_pci_eeh_freeze_clear()
2016-08-01 10:50 ` Michael Ellerman
@ 2016-08-01 23:59 ` Gavin Shan
2016-08-02 3:54 ` Gavin Shan
1 sibling, 0 replies; 4+ messages in thread
From: Gavin Shan @ 2016-08-01 23:59 UTC (permalink / raw)
To: Michael Ellerman; +Cc: Gavin Shan, linuxppc-dev, paulus, benh, ruscur
On Mon, Aug 01, 2016 at 08:50:58PM +1000, Michael Ellerman wrote:
>Gavin Shan <gwshan@linux.vnet.ibm.com> writes:
>
>> The PE number (@frozen_pe_no), filled by opal_pci_next_error() is
>> in big-endian format. Frozen PE number in CPU-endian should be passed
>> to opal_pci_eeh_freeze_clear() when clearing the frozen state if the
>> PE is invalid one.
>>
>> This passes CPU-endian PE number to opal_pci_eeh_freeze_clear() and
>> it should be part of commit <0f36db77643b> ("powerpc/eeh: Fix wrong
>> printed PE number").
>
>Which commit added the broken usage?
>
>Looks like 2a485ad7c88d ("powerpc/powernv: Drop PHB operation
>next_error()") ?
>
It is commit 71b540adffd9 ("powerpc/powernv: Don't escalate non-existing frozen PE").
As commit 0f36db77643b was merged to 4.3 kernel and the changes included in this
patch should be part of the commit. It's why I marked this with 4.3+
>> Cc: stable@vger.kernel.org # v4.3+
>> Suggested-by: Paul Mackerras <paulus@samba.org>
>> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
>> ---
>> arch/powerpc/platforms/powernv/eeh-powernv.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
>> index 86544ea..75363d9 100644
>> --- a/arch/powerpc/platforms/powernv/eeh-powernv.c
>> +++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
>> @@ -1538,7 +1538,7 @@ static int pnv_eeh_next_error(struct eeh_pe **pe)
>>
>> /* Try best to clear it */
>> opal_pci_eeh_freeze_clear(phb->opal_id,
>> - frozen_pe_no,
>> + be64_to_cpu(frozen_pe_no),
>
>Sparse knew about this:
>
> arch/powerpc/platforms/powernv/eeh-powernv.c:1541:41: warning: incorrect type in argument 2 (different base types)
> arch/powerpc/platforms/powernv/eeh-powernv.c:1541:41: expected unsigned long long [unsigned] [usertype] pe_number
> arch/powerpc/platforms/powernv/eeh-powernv.c:1541:41: got restricted __be64 [addressable] [usertype] frozen_pe_no
>
Ok, thanks. I never run sparse before and I will learn from you how to run it.
>And finds some other problems in that file:
>
> arch/powerpc/platforms/powernv/eeh-powernv.c:875:23: warning: constant 0x8000000000000000 is so big it is unsigned long
> arch/powerpc/platforms/powernv/eeh-powernv.c:1309:21: warning: restricted __be16 degrades to integer
>
Ok. I will fix it with separate patches.
Thanks,
Gavin
>cheers
>
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] powerpc/powernv: Fix BE PE number to opal_pci_eeh_freeze_clear()
2016-08-01 10:50 ` Michael Ellerman
2016-08-01 23:59 ` Gavin Shan
@ 2016-08-02 3:54 ` Gavin Shan
1 sibling, 0 replies; 4+ messages in thread
From: Gavin Shan @ 2016-08-02 3:54 UTC (permalink / raw)
To: Michael Ellerman; +Cc: Gavin Shan, linuxppc-dev, paulus, benh, ruscur
On Mon, Aug 01, 2016 at 08:50:58PM +1000, Michael Ellerman wrote:
>Gavin Shan <gwshan@linux.vnet.ibm.com> writes:
>
>> The PE number (@frozen_pe_no), filled by opal_pci_next_error() is
>> in big-endian format. Frozen PE number in CPU-endian should be passed
>> to opal_pci_eeh_freeze_clear() when clearing the frozen state if the
>> PE is invalid one.
>>
>> This passes CPU-endian PE number to opal_pci_eeh_freeze_clear() and
>> it should be part of commit <0f36db77643b> ("powerpc/eeh: Fix wrong
>> printed PE number").
>
>Which commit added the broken usage?
>
>Looks like 2a485ad7c88d ("powerpc/powernv: Drop PHB operation
>next_error()") ?
>
>> Cc: stable@vger.kernel.org # v4.3+
>> Suggested-by: Paul Mackerras <paulus@samba.org>
>> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com>
>> ---
>> arch/powerpc/platforms/powernv/eeh-powernv.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
>> index 86544ea..75363d9 100644
>> --- a/arch/powerpc/platforms/powernv/eeh-powernv.c
>> +++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
>> @@ -1538,7 +1538,7 @@ static int pnv_eeh_next_error(struct eeh_pe **pe)
>>
>> /* Try best to clear it */
>> opal_pci_eeh_freeze_clear(phb->opal_id,
>> - frozen_pe_no,
>> + be64_to_cpu(frozen_pe_no),
>
>Sparse knew about this:
>
> arch/powerpc/platforms/powernv/eeh-powernv.c:1541:41: warning: incorrect type in argument 2 (different base types)
> arch/powerpc/platforms/powernv/eeh-powernv.c:1541:41: expected unsigned long long [unsigned] [usertype] pe_number
> arch/powerpc/platforms/powernv/eeh-powernv.c:1541:41: got restricted __be64 [addressable] [usertype] frozen_pe_no
>
>And finds some other problems in that file:
>
> arch/powerpc/platforms/powernv/eeh-powernv.c:875:23: warning: constant 0x8000000000000000 is so big it is unsigned long
> arch/powerpc/platforms/powernv/eeh-powernv.c:1309:21: warning: restricted __be16 degrades to integer
>
Michael, please ignore this one. A updated version will be included in the
series fixing the warning reported from sparse on PowerNV PCI related files.
I will post the series shortly.
Thanks,
Gavin
>cheers
>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2016-08-02 3:54 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-08-01 7:03 [PATCH] powerpc/powernv: Fix BE PE number to opal_pci_eeh_freeze_clear() Gavin Shan
2016-08-01 10:50 ` Michael Ellerman
2016-08-01 23:59 ` Gavin Shan
2016-08-02 3:54 ` Gavin Shan
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).