Re: [PATCH v8 3/3] perf vendor events arm64: Add "_" to the event name, which starts with a number

2021-03-08 Thread nakamura.s...@fujitsu.com
Hi, John > On 05/03/2021 10:18, Shunsuke Nakamura wrote: > > The perf parser does not support event names that start with a number. > > This patch adds "_" to the event name, which starts with a number. > > > > Signed-off-by: Shunsuke Nakamura > > I think that this can be folded into patch 2/3,

Re: [PATCH v7 2/4] perf tools: Add lexical definition of event name

2021-03-07 Thread nakamura.s...@fujitsu.com
nge the event name. > > hum, I'm not sure what's the final decision in here.. are you keeping > numbers in the start? would be much easier if you did not ;-) I don't keep the event name starting with a number. v8 patch serase changed the event name starting with a number to start with "_". https://lore.kernel.org/linux-arm-kernel/20210305101841.3133721-1-nakamura.s...@fujitsu.com/ Best Regards Shunsuke

Re: [PATCH v7 2/4] perf tools: Add lexical definition of event name

2021-03-05 Thread nakamura.s...@fujitsu.com
Hi, John > > From: Shunsuke Nakamura > > > > Add the lexical definition of event name so that the numbers are > > recognizable. > > > > Sorry for the hassle, but it's not clear that we got a definitive > conclusion on this patch from v6 series: > > https://lore.kernel.org/lkml/YCuo01Wxxe%2F

Re: [PATCH v5 0/4] perf vendor events: Support PMU events for A64FX

2021-02-10 Thread nakamura.s...@fujitsu.com
Hi, John >> I will resend based on tmp.perf/core. > > Actually it is now on perf/core, so that is a better baseline. I will resend based on perf/core. > > I will send it in HTML format. > > Hmmm, that's not better. I think that the mail servers reject html. > > Please consult Documentation/proc

RE: [PATCH v5 0/4] perf vendor events: Support PMU events for A64FX

2021-02-07 Thread nakamura.s...@fujitsu.com
Hi, John > > > Apart from that, I think that we're a bit uncertain about patch 3/4 > > What are your concerns? > > I think it's okay for perf to read a new event code with a number at the > beginning. > > The impact of this fix is on {name} and later rules. > parse_events.l uses {name} only in on

RE: [PATCH v5 0/4] perf vendor events: Support PMU events for A64FX

2021-02-05 Thread nakamura.s...@fujitsu.com
does not misrecognize the {name_tag}, and I think there is no problem. Best Regards Shunsuke > -Original Message- > From: linux-arm-kernel On > Behalf Of nakamura.s...@fujitsu.com > Sent: Wednesday, February 3, 2021 4:02 PM > To: 'John Garry' ; w...@kerne

RE: [PATCH v5 0/4] perf vendor events: Support PMU events for A64FX

2021-02-02 Thread nakamura.s...@fujitsu.com
Hi, John > Just some things to note: > - please pick up and apply reviewed-by tags granted from previous reviews if > explicitly granted, subject to making any trivial changes requested If any > significant changes are made, then don't pick up theses tags or drop them (if > they > were picked up

RE: [PATCH v4 4/4] perf vendor events: Add Fujitsu A64FX V1.2 pmu event

2021-02-02 Thread nakamura.s...@fujitsu.com
(I missed to include the quotation in the text. Will resend. Sorry about that.) Hi, John > About the subject, include "arm64", like other commits in same arch path > > about "V1.2", we use wildcard for matching MIDR, so will these JSONs really > just > patch V1.2? V1.2 is version of the specifi

RE: [PATCH v4 3/4] perf tools: Fix lexical definition of event name

2021-02-02 Thread nakamura.s...@fujitsu.com
(I missed to include the quotation in the text. Will resend. Sorry about that.) Hi, John > > Fix the lexical definition of event name so that the numbers are > > recognizable. > > > > A64FX defines an event name that starts with a number. > > - 0inst_commit > > - 1inst_commit > > - 2inst_c

RE: [PATCH v4 2/4] perf vendor events: Add L2I_TLB, SVE, SIMD, and FP events

2021-02-02 Thread nakamura.s...@fujitsu.com
(I missed to include the quotation in the text. Will resend. Sorry about that.) Hi, John > I think that you will need to fix the subject, like: > > perf vendor events arm64: Add more common and uarch events > > > Add the following events. > > It would be good to mention your reference document

RE: [PATCH v4 4/4] perf vendor events: Add Fujitsu A64FX V1.2 pmu event

2021-02-02 Thread nakamura.s...@fujitsu.com
Hi, John V1.2 is version of the specification sheet. It is not the version of A64FX. I remove the version from the subject because it will be misunderstood. I checked the mail, there was a space before <. https://lore.kernel.org/linux-arm-kernel/2021020310.3231469-5-nakamura.s...@jp.fujitsu.c

RE: [PATCH v4 3/4] perf tools: Fix lexical definition of event name

2021-02-02 Thread nakamura.s...@fujitsu.com
Hi, John I have applied this patch and confirmed that the event can be specified correctly. Event names are defined in the A64FX specification. I want to deal with it at that time if there is a problem. So I would like to resend the v5 patch and wait for advice from others. Best Regards Shunsuk

RE: [PATCH v4 2/4] perf vendor events: Add L2I_TLB, SVE, SIMD, and FP events

2021-02-02 Thread nakamura.s...@fujitsu.com
Hi, John Thanks for review. I will fix it. Best Regards Shunsuke

RE: [PATCH 2/4] perf vendor events arm64: Add common and uarch event JSON

2021-01-27 Thread nakamura.s...@fujitsu.com
Hi ,john I would like to make a comment because there is excess or deficiency in comma. There are no differences in pmu-events.c. > +{ > +"PublicDescription": "Attributable Bus access", > +"EventCode": "0x19", > +"EventName": "BUS_ACCESS", > +"BriefDescription"

RE: [PATCH v2 3/3] perf vendor events: Add Fujitsu A64FX V1.2 pmu event

2021-01-26 Thread nakamura.s...@fujitsu.com
fradead.org; linux-kernel@vger.kernel.org > Subject: Re: [PATCH v2 3/3] perf vendor events: Add Fujitsu A64FX V1.2 pmu > event > > On 22/01/2021 08:37, nakamura.s...@fujitsu.com wrote: > >> I think that we need to the same here as the IMPDEF recommended > >> even

RE: [PATCH v2 3/3] perf vendor events: Add Fujitsu A64FX V1.2 pmu event

2021-01-22 Thread nakamura.s...@fujitsu.com
Hi John > I think that we need to the same here as the IMPDEF recommended events - add a > common JSON to reduce the duplication. > > I had done work on this for current CPUs, but never got it finished. Let me > check > the status. If you let me know the result, I will resend the patch accordin

RE: [PATCH v2 1/3] perf vendor events: Add cache refill and DCZVA events

2021-01-22 Thread nakamura.s...@fujitsu.com
Hi, John, Shaokun > > 在 2021/1/21 18:54, Shunsuke Nakamura 写道: > >> Adds L1 data cache refill prefetch, L2 data cache refill prefetch, > >> and DCZVA instruction events. > > > > A silly question, Does Arm define these events? I checked Arm ARM > > document(DDI0487Fc) that these event numbers are