Hi Mathieu,
OK - unfortunately I too recognised that the first two patches were fixes that could be applied irrespective of the new work, so created a separate patch set that went to the main arm-kernel mailing list, with these two patches plus one other relating to the comment fix that was originally part of the default sink set. Feel free to ignore the repeats!
Regards
Mike
On Tue, 16 Jun 2020 at 20:19, Mathieu Poirier mathieu.poirier@linaro.org wrote:
On Wed, Jun 10, 2020 at 10:13:31AM +0100, Mike Leach wrote:
The counter value registers change during operation, however this change is not reflected in the values seen by the user in sysfs.
This fixes the issue by reading back the values on disable.
Signed-off-by: Mike Leach mike.leach@linaro.org
drivers/hwtracing/coresight/coresight-etm4x.c | 6 ++++++ 1 file changed, 6 insertions(+)
diff --git a/drivers/hwtracing/coresight/coresight-etm4x.c b/drivers/hwtracing/coresight/coresight-etm4x.c index 747afc875f91..60a5133d801c 100644 --- a/drivers/hwtracing/coresight/coresight-etm4x.c +++ b/drivers/hwtracing/coresight/coresight-etm4x.c @@ -507,6 +507,12 @@ static void etm4_disable_hw(void *info) readl_relaxed(drvdata->base + TRCSSCSRn(i)); }
/* read back the current counter values */for (i = 0; i < drvdata->nr_cntr; i++) {config->cntr_val[i] =readl_relaxed(drvdata->base + TRCCNTVRn(i));}I have applied patches 1 and 2 in this set - no need to send them again.
Thanks, Mathieu
coresight_disclaim_device_unlocked(drvdata->base); CS_LOCK(drvdata->base);-- 2.17.1