Re: [PATCH v7 0/3] perf arm-spe: Add support for synthetic events

[James Clark]

Hi Will and Leo,

I've tested this on an Arm N1 machine and it looks good to me.


James

On 26/05/2020 20:54, Will Deacon wrote:
> On Tue, May 26, 2020 at 06:43:37PM +0800, Leo Yan wrote:
>> On Tue, May 26, 2020 at 11:26:03AM +0100, Will Deacon wrote:
>>> On Fri, May 22, 2020 at 11:09:19AM +0800, Leo Yan wrote:
>>>> On Mon, May 04, 2020 at 07:56:22PM +0800, Leo Yan wrote:
>>>>> This patch set is to support synthetic events with enabling Arm SPE
>>>>> decoder.  Since before Xiaojun Tan (Hisilicon) and James Clark (Arm)
>>>>> have contributed much for this task, so this patch set is based on their
>>>>> privous work and polish for the version 7.
>>>>>
>>>>> The main work in this version is to polished the core patch "perf
>>>>> arm-spe: Support synthetic events", e.g. rewrite the code to calculate
>>>>> ip, packet generation for multiple types (L1 data cache, Last level
>>>>> cache, TLB, remote access, etc).  It also heavily refactors code for
>>>>> data structure and program flow, which removed unused fields in
>>>>> structure and polished the program flow to achieve neat code as
>>>>> possible.
>>>>>
>>>>> This patch set has been checked with checkpatch.pl, though it leaves
>>>>> several warnings, but these warnings are delibarately kept after
>>>>> reviewing.  Some warnings ask to add maintainer (so far it's not
>>>>> necessary), and some warnings complaint for patch 02 "perf auxtrace:
>>>>> Add four itrace options" for the text format, since need to keep the
>>>>> consistency with the same code format in the source code, this is why
>>>>> this patch doesn't get rid of checkpatch warnings.
>>>>
>>>> Gentle ping ...
>>>>
>>>> It would be appreciate if can get some review for this patch set.
>>>
>>> I was hoping that James Clark would have a look, since he was the last
>>> person to go near the userspace side of SPE.
>>
>> Yes, I have offline synced with James and James has verified this
>> patch set at his side.
>>
>> I don't want to rush to ask Arnaldo to merge patches, so just
>> want to get wider reviewing if possible; otherwise, I will rebase this
>> patch set and resend to ML.
> 
> One thing that might be useful is if James could offer his Tested-by or
> Acked-by on the public mailing list. Neither Arnaldo nor I have details
> about your offline sync!
> 
> Will
>