Re: [PATCH] ioat: remove chanerr mask setting for IOAT v3.x

[Dave Jiang <dave.jiang@intel.com>]

On 11/26/2012 09:56 PM, Dan Williams wrote:
> On Fri, Nov 16, 2012 at 3:26 PM, Dave Jiang <dave.jiang@intel.com> wrote:
>> The CHANERRMSK_INT register should be 0. The existing code set a value
>> for a workaround to address a pre-silicon bug on the Intel 5520 IO hub that
>> has been fixed when the hardware was released. There is no need for this
>> code.
>>
>> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
>> ---
>>
>>  drivers/dma/ioat/dma_v3.c |    8 ++------
>>  1 file changed, 2 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/dma/ioat/dma_v3.c b/drivers/dma/ioat/dma_v3.c
>> index f7f1dc6..60c15c4 100644
>> --- a/drivers/dma/ioat/dma_v3.c
>> +++ b/drivers/dma/ioat/dma_v3.c
>> @@ -1126,12 +1126,8 @@ static int ioat3_reset_hw(struct ioat_chan_common *chan)
>>         chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
>>         writel(chanerr, chan->reg_base + IOAT_CHANERR_OFFSET);
>>
>> -       /* -= IOAT ver.3 workarounds =- */
>> -       /* Write CHANERRMSK_INT with 3E07h to mask out the errors
>> -        * that can cause stability issues for IOAT ver.3, and clear any
>> -        * pending errors
>> -        */
>> -       pci_write_config_dword(pdev, IOAT_PCI_CHANERRMASK_INT_OFFSET, 0x3e07);
>> +       /* clearn CHANERRMASK_INT and clear any pending errors */
>> +       pci_write_config_dword(pdev, IOAT_PCI_CHANERRMASK_INT_OFFSET, 0);
> Can we get rid of this write / register access altogether?  Then the
> driver will load on systems where extended pci configuration space is
> not available.

So this entire thing does read and write to a mask register and the
error register. Were you wanting both removed? I think we may still want
to clear the error register. If that's the case, it doesn't get around
the extended PCI config space issue.