Re: [PATCH v10 1/8] coresight: etm4x: fix inconsistencies with sysfs configuration

[Yeoreum Yun <yeoreum.yun@arm.com>]

On Fri, Sep 11, 2026 at 11:21:44AM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [High] Hardware counter values (`cntr_val`) read back during trace disable are saved to the temporary runtime `active_config` instead of the sysfs-visible `config` structure, permanently losing the post-session hardware state.
> --
> 
> commit 293039b149bb540c924b83cd181604cbff99de4c
> Author: Yeoreum Yun <yeoreum.yun@arm.com>
> 
> coresight: etm4x: fix inconsistencies with sysfs configuration
> 
> This commit addresses inconsistencies in ETM4x sysfs configuration by
> separating the configuration into an active_config for the current session
> and a config for sysfs. It updates the enable/disable paths to use the
> active session configuration to prevent corruption from concurrent access.
> 
> > diff --git a/drivers/hwtracing/coresight/coresight-etm4x-core.c b/drivers/hwtracing/coresight/coresight-etm4x-core.c
> > --- a/drivers/hwtracing/coresight/coresight-etm4x-core.c
> > +++ b/drivers/hwtracing/coresight/coresight-etm4x-core.c
> 
> [ ... ]
> 
> > @@ -1044,7 +1054,7 @@ static void etm4_disable_trace_unit(struct etmv4_drvdata *drvdata)
> >  static void etm4_disable_hw(struct etmv4_drvdata *drvdata)
> >  {
> >  	u32 control;
> > -	struct etmv4_config *config = &drvdata->config;
> > +	struct etmv4_config *config = &drvdata->active_config;
> >  	struct coresight_device *csdev = drvdata->csdev;
> >  	struct csdev_access *csa = &csdev->access;
> >  	int i;
> 
> [Severity: High]
> Does this change cause the post-session hardware counter values to be lost?
> 
> Later in etm4_disable_hw(), the hardware counter values are read back from
> the device:
> 
> 	/* read back the current counter values */
> 	for (i = 0; i < caps->nr_cntr; i++) {
> 		config->cntr_val[i] =
> 			etm4x_relaxed_read32(csa, TRCCNTVRn(i));
> 	}
> 
> Because config now points to drvdata->active_config, the updated counter
> values are saved to the temporary active session configuration instead of
> the primary sysfs configuration.
> 
> When a user subsequently reads the counter value via the sysfs interface
> in cntr_val_show():
> 
> 	struct etmv4_config *config = &drvdata->config;
> 	...
> 	val = config->cntr_val[idx];
> 
> they will read from the static drvdata->config, which never receives these
> read-back values and will instead return the initial configuration value.
> Should the read-back hardware state be synchronized back to drvdata->config
> when the trace is disabled?
> 
> -- 
> Sashiko AI review · https://sashiko.dev/#/patchset/20260911-separate_etm_cfg_v2-v10-0-1b715d95927a@arm.com?part=1

The same for etm3. I'll fix it in next round.

-- 
Sincerely,
Yeoreum Yun