On Fri, 28 Nov 2025 11:55:12 +0000, James Clark wrote:
Do some cleanups then expand the timestamp format attribute from 1 bit to 4 bits for ETMv4 in Perf mode. The current interval is too high for most use cases, and particularly on the FVP the number of timestamps generated is excessive. This change not only still allows disabling or enabling timestamps, but also allows the interval to be configured.
The old bit is kept deprecated and undocumented for now. There are known broken versions of Perf that don't read the format attribute positions from sysfs and instead hard code the timestamp bit. We can leave the old bit in the driver until we need the bit for another feature or enough time has passed that these old Perfs are unlikely to be used.
[...]
Applied, thanks!
[01/13] coresight: Change syncfreq to be a u8 (no commit info) [02/13] coresight: Repack struct etmv4_drvdata (no commit info) [03/13] coresight: Refactor etm4_config_timestamp_event() (no commit info) [04/13] coresight: Hide unused ETMv3 format attributes (no commit info) [05/13] coresight: Define format attributes with GEN_PMU_FORMAT_ATTR() (no commit info) [06/13] coresight: Interpret ETMv3 config with ATTR_CFG_GET_FLD() (no commit info) [07/13] coresight: Don't reject unrecognized ETMv3 format attributes (no commit info) [08/13] coresight: Interpret perf config with ATTR_CFG_GET_FLD() (no commit info) [09/13] coresight: Interpret ETMv4 config with ATTR_CFG_GET_FLD() (no commit info) [10/13] coresight: Remove misleading definitions (no commit info) [11/13] coresight: Prepare to allow setting the timestamp interval (no commit info) [12/13] coresight: Extend width of timestamp format attribute (no commit info) [13/13] coresight: docs: Document etm4x timestamp interval option (no commit info)
Best regards,