From: Leo Yan <leo.yan@arm.com>
To: Suzuki K Poulose <suzuki.poulose@arm.com>
Cc: Mike Leach <mike.leach@linaro.org>,
James Clark <james.clark@linaro.org>,
Anshuman Khandual <anshuman.khandual@arm.com>,
Yeoreum Yun <yeoreum.yun@arm.com>, Will Deacon <will@kernel.org>,
Mark Rutland <mark.rutland@arm.com>,
Tamas Petz <tamas.petz@arm.com>,
Tamas Zsoldos <tamas.zsoldos@arm.com>,
Arnaldo Carvalho de Melo <acme@kernel.org>,
Namhyung Kim <namhyung@kernel.org>, Jiri Olsa <jolsa@kernel.org>,
Ian Rogers <irogers@google.com>,
Adrian Hunter <adrian.hunter@intel.com>,
coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org
Subject: Re: [PATCH 07/19] coresight: trbe: Refactor AUX flag setting
Date: Tue, 2 Dec 2025 14:21:55 +0000 [thread overview]
Message-ID: <20251202142155.GB724103@e132581.arm.com> (raw)
In-Reply-To: <e315b489-2ebf-4ab6-8a60-347f1b1dca16@arm.com>
On Tue, Dec 02, 2025 at 11:15:47AM +0000, Suzuki K Poulose wrote:
[...]
> > @@ -687,6 +668,23 @@ static enum trbe_fault_action trbe_get_fault_act(struct perf_output_handle *hand
> > goto out_fatal;
> > }
> > + /*
> > + * Mark the buffer to indicate that there was a WRAP event by
> > + * setting the COLLISION flag. This indicates to the user that
> > + * the TRBE trace collection was stopped without stopping the
> > + * ETE and thus there might be some amount of trace that was
> > + * lost between the time the WRAP was detected and the IRQ
> > + * was consumed by the CPU.
> > + *
> > + * Setting the TRUNCATED flag would move the event to STOPPED
> > + * state unnecessarily, even when there is space left in the
> > + * ring buffer. Using the COLLISION flag doesn't have this side
> > + * effect. We only set TRUNCATED flag when there is no space
> > + * left in the ring buffer.
> > + */
> > + if (!is_trbe_running(trbsr))
>
> Is this really required ? There is a WARN_ON(is_trbe_running(trbsr))
> above ?
The WARN_ON() will reports warning for running case, here it sets
COLLISION flag for not running case, which is not conflict.
This works for the Fill mode (stop collection), we always set the
COLLISION flag for it. Once we enable trigger and wrap modes, due to
the trace continues to run, we don't set the flag for these modes.
> > + perf_aux_output_flag(handle, PERF_AUX_FLAG_COLLISION);
>
> Also, setting this would unnecessarily mark COLLISION while stopping the
> TRBE buffer normally (when called from arm_trbe_update_buffer), when
> there is no WRAP event ?
Good point. I should refactor arm_trbe_update_buffer(), it firstly
reads out TRBSR, stop trace afterwards, and then call
trbe_get_fault_act().
In this case, it is in running mode so COLLISION will not be set for
normal disable flow.
I will add a new patch for this and place it before this one.
> > +
> > if (is_trbe_wrap(trbsr))
> > return TRBE_FAULT_ACT_WRAP;
>
> Couldn't this be :
>
> /* Move the big fat comment here */
> if (is_trbe_wrap(trbsr)) {
> perf_aux_output_flag(handle, PERF_AUX_FLAG_COLLISION);
> return TRBE_FAULT_ACT_WRAP;
> }
It deliberately de-couples the COLLISION flag from the WRAP event.
In later change, a WRAP event only means buffer wrap around, it does
mean the trace has been stopped. The AUX flag is only set for
stopping case.
Thanks,
Leo
next prev parent reply other threads:[~2025-12-02 14:21 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-01 11:21 [PATCH 00/19] coresight: trbe: Support trigger and circle buffer modes Leo Yan
2025-12-01 11:21 ` [PATCH 01/19] coresight: trbe: Use helpers for checking errata Leo Yan
2025-12-04 12:08 ` Anshuman Khandual
2025-12-01 11:21 ` [PATCH 02/19] coresight: trbe: Remove redundant disable operation Leo Yan
2025-12-04 12:25 ` Anshuman Khandual
2025-12-01 11:21 ` [PATCH 03/19] coresight: trbe: Remove buffer disabling in trbe_handle_overflow() Leo Yan
2025-12-04 12:31 ` Anshuman Khandual
2025-12-01 11:21 ` [PATCH 04/19] coresight: trbe: Remove set_trbe_disabled() from the enable flow Leo Yan
2025-12-04 12:43 ` Anshuman Khandual
2025-12-04 13:25 ` Leo Yan
2025-12-01 11:21 ` [PATCH 05/19] coresight: trbe: Refactor status clearing Leo Yan
2025-12-04 12:57 ` Anshuman Khandual
2025-12-09 15:29 ` Leo Yan
2025-12-01 11:21 ` [PATCH 06/19] coresight: trbe: Refactor syndrome decoding Leo Yan
2025-12-02 11:06 ` Suzuki K Poulose
2025-12-02 14:24 ` Leo Yan
2025-12-09 13:17 ` James Clark
2025-12-09 16:06 ` Leo Yan
2025-12-05 4:10 ` Anshuman Khandual
2025-12-09 15:57 ` Leo Yan
2025-12-01 11:21 ` [PATCH 07/19] coresight: trbe: Refactor AUX flag setting Leo Yan
2025-12-02 11:15 ` Suzuki K Poulose
2025-12-02 14:21 ` Leo Yan [this message]
2025-12-09 13:37 ` James Clark
2025-12-10 15:43 ` Leo Yan
2025-12-12 14:50 ` James Clark
2025-12-12 15:27 ` Leo Yan
2025-12-12 15:52 ` James Clark
2025-12-01 11:21 ` [PATCH 08/19] coresight: trbe: Use PERF_AUX_FLAG_PARTIAL instead of PERF_AUX_FLAG_COLLISION Leo Yan
2025-12-05 4:28 ` Anshuman Khandual
2025-12-09 13:40 ` James Clark
2025-12-10 16:19 ` Leo Yan
2025-12-01 11:21 ` [PATCH 09/19] coresight: trbe: Add fault action argument to trbe_handle_overflow() Leo Yan
2025-12-01 11:22 ` [PATCH 10/19] coresight: trbe: Always check fault action when updating buffer Leo Yan
2025-12-02 12:00 ` Suzuki K Poulose
2025-12-01 11:22 ` [PATCH 11/19] coresight: trbe: Apply overwrite erratum for only wrap event Leo Yan
2025-12-02 12:05 ` Suzuki K Poulose
2025-12-02 16:56 ` Leo Yan
2025-12-02 17:12 ` Leo Yan
2025-12-01 11:22 ` [PATCH 12/19] coresight: trbe: Calculate size for buffer wrapping Leo Yan
2025-12-01 11:22 ` [PATCH 13/19] coresight: trbe: Remove misleading comment Leo Yan
2025-12-01 11:22 ` [PATCH 14/19] coresight: trbe: Refactor compute_trbe_buffer_limit() Leo Yan
2025-12-01 11:22 ` [PATCH 15/19] coresight: trbe: Add static key for bypassing trigger mode Leo Yan
2025-12-02 12:10 ` Suzuki K Poulose
2025-12-01 11:22 ` [PATCH 16/19] coresight: trbe: Support " Leo Yan
2025-12-01 11:22 ` [PATCH 17/19] coresight: trbe: Enable circle mode for snapshot Leo Yan
2025-12-01 11:22 ` [PATCH 18/19] coresight: trbe: Add kunit tests Leo Yan
2025-12-01 11:22 ` [PATCH 19/19] perf: cs-etm: Set watermark for AUX trace Leo Yan
2025-12-05 4:48 ` Anshuman Khandual
2025-12-09 14:54 ` James Clark
2025-12-10 2:22 ` Anshuman Khandual
2025-12-05 4:53 ` [PATCH 00/19] coresight: trbe: Support trigger and circle buffer modes Anshuman Khandual
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20251202142155.GB724103@e132581.arm.com \
--to=leo.yan@arm.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=anshuman.khandual@arm.com \
--cc=coresight@lists.linaro.org \
--cc=irogers@google.com \
--cc=james.clark@linaro.org \
--cc=jolsa@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mike.leach@linaro.org \
--cc=namhyung@kernel.org \
--cc=suzuki.poulose@arm.com \
--cc=tamas.petz@arm.com \
--cc=tamas.zsoldos@arm.com \
--cc=will@kernel.org \
--cc=yeoreum.yun@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).