| Message ID | 20230116124928.5440-1-mike.leach@linaro.org (mailing list archive) |
|---|---|
| Headers | show |
| Series | coresight: Add new API to allocate trace source ID values | expand |
Hi Arnaldo, On 16/01/2023 12:49, Mike Leach wrote: > The current method for allocating trace source ID values to sources is > to use a fixed algorithm for CPU based sources of (cpu_num * 2 + 0x10). > The STM is allocated ID 0x1. > > This fixed algorithm is used in both the CoreSight driver code, and by > perf when writing the trace metadata in the AUXTRACE_INFO record. > > The method needs replacing as currently:- > 1. It is inefficient in using available IDs. > 2. Does not scale to larger systems with many cores and the algorithm > has no limits so will generate invalid trace IDs for cpu number > 44. > > Additionally requirements to allocate additional system IDs on some > systems have been seen. > > This patch set introduces an API that allows the allocation of trace IDs > in a dynamic manner. > > Architecturally reserved IDs are never allocated, and the system is > limited to allocating only valid IDs. > > Each of the current trace sources ETM3.x, ETM4.x and STM is updated to use > the new API. > > For the ETMx.x devices IDs are allocated on certain events > a) When using sysfs, an ID will be allocated on hardware enable, or a read of > sysfs TRCTRACEID register and freed when the sysfs reset is written. > > b) When using perf, ID is allocated on during setup AUX event, and freed on > event free. IDs are communicated using the AUX_OUTPUT_HW_ID packet. > The ID allocator is notified when perf sessions start and stop > so CPU based IDs are kept constant throughout any perf session. > > > Note: This patchset breaks some backward compatibility for perf record and > perf report. > > The version of the AUXTRACE_INFO has been updated to reflect the fact that > the trace source IDs are generated differently. This will > mean older versions of perf report cannot decode the newer file. > > Appies to coresight/next > > Changes since v6: > 1) Export perf_report_aux_output_id() Acked by Peter, > 2) Update to ETMv3 docs requested by suzuki > 3) rebased and re-tested on coresight/next (6.2-rc2) > > Changes since v5: (requested by suzuki) > 1) Prefer odd ID values for system IDs to avoid overlap with legacy CPU IDs > 2) Some style changes > > Changes since v4: > 1) update to ensure that compiling after each individual patch added still > works - ie. git bisect not broken through the patchset.. > > 2) Revision to some of the now redundant code in cs-etm (James) > > 3) Comments and other minor fixes requested by Suzuki. > > Changes since v3: > 1) Fixed aarch32 build error in ETM3.x driver. > Reported-by: kernel test robot <lkp@intel.com> > > Changes since v2: > 1) Improved backward compatibility: (requested by James) > > Using the new version of perf on an old kernel will generate a usable file > legacy metadata values are set by the new perf and will be used if mew > ID packets are not present in the file. > > Using an older version of perf / simpleperf on an updated kernel may still > work. The trace ID allocator has been updated to use the legacy ID values > where possible, so generated file and used trace IDs will match up to the > point where the legacy algorithm is broken anyway. > > 2) Various changes to the ID allocator and ID packet format. > (suggested by Suzuki) > > 3) per CPU ID info in allocator now stored as atomic type to allow a passive read > without taking the allocator spinlock. perf flow now allocates and releases ID > values in setup_aux / free_event. Device enable and event enable use the passive > read to set the allocated values. This simplifies the locking mechanisms on the > perf run and fixes issues that arose with locking dependencies. > > Changes since v1: > (after feedback & discussion with Mathieu & Suzuki). > > 1) API has changed. The global trace ID map is managed internally, so it > is no longer passed in to the API functions. > > 2) perf record does not use sysfs to find the trace IDs. These are now > output as AUX_OUTPUT_HW_ID events. The drivers, perf record, and perf report > have been updated accordingly to generate and handle these events. > > Mike Leach (15): > coresight: trace-id: Add API to dynamically assign Trace ID values > coresight: Remove obsolete Trace ID unniqueness checks > coresight: perf: traceid: Add perf ID allocation and notifiers > coresight: stm: Update STM driver to use Trace ID API > coresight: etm4x: Update ETM4 driver to use Trace ID API > coresight: etm3x: Update ETM3 driver to use Trace ID API > coresight: etmX.X: stm: Remove trace_id() callback > coresight: trace id: Remove legacy get trace ID function. > perf: cs-etm: Move mapping of Trace ID and cpu into helper function > perf: cs-etm: Update record event to use new Trace ID protocol > kernel: events: Export perf_report_aux_output_id() > perf: cs-etm: Handle PERF_RECORD_AUX_OUTPUT_HW_ID packet > coresight: events: PERF_RECORD_AUX_OUTPUT_HW_ID used for Trace ID > coresight: trace-id: Add debug & test macros to Trace ID allocation > coresight: etm3x: docs: Alter sysfs documentation for trace id updates I have pushed the kernel patches on this series to coresight tree next[0]. I will be sending them out to Greg for v6.3. Please could you queue the perf tool changes in the series ? i.e., Patches 9,10 and 12. [0] https://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux.git/log/?h=next Kind regards Suzuki
Hi Arnaldo Gentle ping. On 19/01/2023 12:00, Suzuki K Poulose wrote: > Hi Arnaldo, > > On 16/01/2023 12:49, Mike Leach wrote: >> The current method for allocating trace source ID values to sources is >> to use a fixed algorithm for CPU based sources of (cpu_num * 2 + 0x10). >> The STM is allocated ID 0x1. >> >> This fixed algorithm is used in both the CoreSight driver code, and by >> perf when writing the trace metadata in the AUXTRACE_INFO record. >> >> The method needs replacing as currently:- >> 1. It is inefficient in using available IDs. >> 2. Does not scale to larger systems with many cores and the algorithm >> has no limits so will generate invalid trace IDs for cpu number > 44. >> >> Additionally requirements to allocate additional system IDs on some >> systems have been seen. >> >> This patch set introduces an API that allows the allocation of trace IDs >> in a dynamic manner. >> >> Architecturally reserved IDs are never allocated, and the system is >> limited to allocating only valid IDs. >> >> Each of the current trace sources ETM3.x, ETM4.x and STM is updated to >> use >> the new API. >> >> For the ETMx.x devices IDs are allocated on certain events >> a) When using sysfs, an ID will be allocated on hardware enable, or a >> read of >> sysfs TRCTRACEID register and freed when the sysfs reset is written. >> >> b) When using perf, ID is allocated on during setup AUX event, and >> freed on >> event free. IDs are communicated using the AUX_OUTPUT_HW_ID packet. >> The ID allocator is notified when perf sessions start and stop >> so CPU based IDs are kept constant throughout any perf session. >> >> >> Note: This patchset breaks some backward compatibility for perf record >> and >> perf report. >> >> The version of the AUXTRACE_INFO has been updated to reflect the fact >> that >> the trace source IDs are generated differently. This will >> mean older versions of perf report cannot decode the newer file. >> >> Appies to coresight/next >> ... >> Mike Leach (15): >> coresight: trace-id: Add API to dynamically assign Trace ID values >> coresight: Remove obsolete Trace ID unniqueness checks >> coresight: perf: traceid: Add perf ID allocation and notifiers >> coresight: stm: Update STM driver to use Trace ID API >> coresight: etm4x: Update ETM4 driver to use Trace ID API >> coresight: etm3x: Update ETM3 driver to use Trace ID API >> coresight: etmX.X: stm: Remove trace_id() callback >> coresight: trace id: Remove legacy get trace ID function. >> perf: cs-etm: Move mapping of Trace ID and cpu into helper function >> perf: cs-etm: Update record event to use new Trace ID protocol >> kernel: events: Export perf_report_aux_output_id() >> perf: cs-etm: Handle PERF_RECORD_AUX_OUTPUT_HW_ID packet >> coresight: events: PERF_RECORD_AUX_OUTPUT_HW_ID used for Trace ID >> coresight: trace-id: Add debug & test macros to Trace ID allocation >> coresight: etm3x: docs: Alter sysfs documentation for trace id updates > > I have pushed the kernel patches on this series to coresight tree next[0]. > > I will be sending them out to Greg for v6.3. Please could you queue the > perf tool changes in the series ? i.e., Patches 9,10 and 12. > > > [0] > https://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux.git/log/?h=next > Please could you pick up the perf tool changes above ? Suzuki
Hi Arnaldo On 24/01/2023 11:36, Suzuki K Poulose wrote: > Hi Arnaldo > > Gentle ping. > > On 19/01/2023 12:00, Suzuki K Poulose wrote: >> Hi Arnaldo, >> >> On 16/01/2023 12:49, Mike Leach wrote: >>> The current method for allocating trace source ID values to sources is >>> to use a fixed algorithm for CPU based sources of (cpu_num * 2 + 0x10). >>> The STM is allocated ID 0x1. >>> >>> This fixed algorithm is used in both the CoreSight driver code, and by >>> perf when writing the trace metadata in the AUXTRACE_INFO record. >>> >>> The method needs replacing as currently:- >>> 1. It is inefficient in using available IDs. >>> 2. Does not scale to larger systems with many cores and the algorithm >>> has no limits so will generate invalid trace IDs for cpu number > 44. >>> >>> Additionally requirements to allocate additional system IDs on some >>> systems have been seen. >>> >>> This patch set introduces an API that allows the allocation of trace >>> IDs >>> in a dynamic manner. >>> >>> Architecturally reserved IDs are never allocated, and the system is >>> limited to allocating only valid IDs. >>> >>> Each of the current trace sources ETM3.x, ETM4.x and STM is updated >>> to use >>> the new API. >>> >>> For the ETMx.x devices IDs are allocated on certain events >>> a) When using sysfs, an ID will be allocated on hardware enable, or a >>> read of >>> sysfs TRCTRACEID register and freed when the sysfs reset is written. >>> >>> b) When using perf, ID is allocated on during setup AUX event, and >>> freed on >>> event free. IDs are communicated using the AUX_OUTPUT_HW_ID packet. >>> The ID allocator is notified when perf sessions start and stop >>> so CPU based IDs are kept constant throughout any perf session. >>> >>> >>> Note: This patchset breaks some backward compatibility for perf >>> record and >>> perf report. >>> >>> The version of the AUXTRACE_INFO has been updated to reflect the fact >>> that >>> the trace source IDs are generated differently. This will >>> mean older versions of perf report cannot decode the newer file. >>> >>> Appies to coresight/next >>> > > ... > >>> Mike Leach (15): >>> coresight: trace-id: Add API to dynamically assign Trace ID values >>> coresight: Remove obsolete Trace ID unniqueness checks >>> coresight: perf: traceid: Add perf ID allocation and notifiers >>> coresight: stm: Update STM driver to use Trace ID API >>> coresight: etm4x: Update ETM4 driver to use Trace ID API >>> coresight: etm3x: Update ETM3 driver to use Trace ID API >>> coresight: etmX.X: stm: Remove trace_id() callback >>> coresight: trace id: Remove legacy get trace ID function. >>> perf: cs-etm: Move mapping of Trace ID and cpu into helper function >>> perf: cs-etm: Update record event to use new Trace ID protocol >>> kernel: events: Export perf_report_aux_output_id() >>> perf: cs-etm: Handle PERF_RECORD_AUX_OUTPUT_HW_ID packet >>> coresight: events: PERF_RECORD_AUX_OUTPUT_HW_ID used for Trace ID >>> coresight: trace-id: Add debug & test macros to Trace ID allocation >>> coresight: etm3x: docs: Alter sysfs documentation for trace id >>> updates >> >> I have pushed the kernel patches on this series to coresight tree >> next[0]. >> >> I will be sending them out to Greg for v6.3. Please could you queue the >> perf tool changes in the series ? i.e., Patches 9,10 and 12. >> >> >> [0] >> https://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux.git/log/?h=next >> > > Please could you pick up the perf tool changes above ? Please could you confirm if you are able to queue the perf tool changes ? Or would you like me route it via coresight tree ? Kind regards Suzuki
Arnaldo Gentle ping On 31/01/2023 11:49, Suzuki K Poulose wrote: > Hi Arnaldo > > On 24/01/2023 11:36, Suzuki K Poulose wrote: >> Hi Arnaldo >> >> Gentle ping. >> >> On 19/01/2023 12:00, Suzuki K Poulose wrote: >>> Hi Arnaldo, >>> >>> On 16/01/2023 12:49, Mike Leach wrote: >>>> The current method for allocating trace source ID values to sources is >>>> to use a fixed algorithm for CPU based sources of (cpu_num * 2 + 0x10). >>>> The STM is allocated ID 0x1. >>>> >>>> This fixed algorithm is used in both the CoreSight driver code, and by >>>> perf when writing the trace metadata in the AUXTRACE_INFO record. >>>> >>>> The method needs replacing as currently:- >>>> 1. It is inefficient in using available IDs. >>>> 2. Does not scale to larger systems with many cores and the algorithm >>>> has no limits so will generate invalid trace IDs for cpu number > 44. >>>> >>>> Additionally requirements to allocate additional system IDs on some >>>> systems have been seen. >>>> >>>> This patch set introduces an API that allows the allocation of >>>> trace IDs >>>> in a dynamic manner. >>>> >>>> Architecturally reserved IDs are never allocated, and the system is >>>> limited to allocating only valid IDs. >>>> >>>> Each of the current trace sources ETM3.x, ETM4.x and STM is updated >>>> to use >>>> the new API. >>>> >>>> For the ETMx.x devices IDs are allocated on certain events >>>> a) When using sysfs, an ID will be allocated on hardware enable, or >>>> a read of >>>> sysfs TRCTRACEID register and freed when the sysfs reset is written. >>>> >>>> b) When using perf, ID is allocated on during setup AUX event, and >>>> freed on >>>> event free. IDs are communicated using the AUX_OUTPUT_HW_ID packet. >>>> The ID allocator is notified when perf sessions start and stop >>>> so CPU based IDs are kept constant throughout any perf session. >>>> >>>> >>>> Note: This patchset breaks some backward compatibility for perf >>>> record and >>>> perf report. >>>> >>>> The version of the AUXTRACE_INFO has been updated to reflect the >>>> fact that >>>> the trace source IDs are generated differently. This will >>>> mean older versions of perf report cannot decode the newer file. >>>> >>>> Appies to coresight/next >>>> >> >> ... >> >>>> Mike Leach (15): >>>> coresight: trace-id: Add API to dynamically assign Trace ID values >>>> coresight: Remove obsolete Trace ID unniqueness checks >>>> coresight: perf: traceid: Add perf ID allocation and notifiers >>>> coresight: stm: Update STM driver to use Trace ID API >>>> coresight: etm4x: Update ETM4 driver to use Trace ID API >>>> coresight: etm3x: Update ETM3 driver to use Trace ID API >>>> coresight: etmX.X: stm: Remove trace_id() callback >>>> coresight: trace id: Remove legacy get trace ID function. >>>> perf: cs-etm: Move mapping of Trace ID and cpu into helper function >>>> perf: cs-etm: Update record event to use new Trace ID protocol >>>> kernel: events: Export perf_report_aux_output_id() >>>> perf: cs-etm: Handle PERF_RECORD_AUX_OUTPUT_HW_ID packet >>>> coresight: events: PERF_RECORD_AUX_OUTPUT_HW_ID used for Trace ID >>>> coresight: trace-id: Add debug & test macros to Trace ID allocation >>>> coresight: etm3x: docs: Alter sysfs documentation for trace id >>>> updates >>> >>> I have pushed the kernel patches on this series to coresight tree >>> next[0]. >>> >>> I will be sending them out to Greg for v6.3. Please could you queue the >>> perf tool changes in the series ? i.e., Patches 9,10 and 12. >>> >>> >>> [0] >>> https://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux.git/log/?h=next >>> >> >> Please could you pick up the perf tool changes above ? > > Please could you confirm if you are able to queue the perf tool changes > ? Or would you like me route it via coresight tree ? > > Kind regards > Suzuki > Suzuki
Mike, Please could you send all the perf tool related patches as a seperate series, with all the tags to Arnaldo directly at -rc1. Please add a cover letter explaining that the kernel changes are available upstream. Kind regards Suzuki On 20/02/2023 16:24, Suzuki K Poulose wrote: > Arnaldo > > Gentle ping > > > On 31/01/2023 11:49, Suzuki K Poulose wrote: >> Hi Arnaldo >> >> On 24/01/2023 11:36, Suzuki K Poulose wrote: >>> Hi Arnaldo >>> >>> Gentle ping. >>> >>> On 19/01/2023 12:00, Suzuki K Poulose wrote: >>>> Hi Arnaldo, >>>> >>>> On 16/01/2023 12:49, Mike Leach wrote: >>>>> The current method for allocating trace source ID values to sources is >>>>> to use a fixed algorithm for CPU based sources of (cpu_num * 2 + >>>>> 0x10). >>>>> The STM is allocated ID 0x1. >>>>> >>>>> This fixed algorithm is used in both the CoreSight driver code, and by >>>>> perf when writing the trace metadata in the AUXTRACE_INFO record. >>>>> >>>>> The method needs replacing as currently:- >>>>> 1. It is inefficient in using available IDs. >>>>> 2. Does not scale to larger systems with many cores and the algorithm >>>>> has no limits so will generate invalid trace IDs for cpu number > 44. >>>>> >>>>> Additionally requirements to allocate additional system IDs on some >>>>> systems have been seen. >>>>> >>>>> This patch set introduces an API that allows the allocation of >>>>> trace IDs >>>>> in a dynamic manner. >>>>> >>>>> Architecturally reserved IDs are never allocated, and the system is >>>>> limited to allocating only valid IDs. >>>>> >>>>> Each of the current trace sources ETM3.x, ETM4.x and STM is updated >>>>> to use >>>>> the new API. >>>>> >>>>> For the ETMx.x devices IDs are allocated on certain events >>>>> a) When using sysfs, an ID will be allocated on hardware enable, or >>>>> a read of >>>>> sysfs TRCTRACEID register and freed when the sysfs reset is written. >>>>> >>>>> b) When using perf, ID is allocated on during setup AUX event, and >>>>> freed on >>>>> event free. IDs are communicated using the AUX_OUTPUT_HW_ID packet. >>>>> The ID allocator is notified when perf sessions start and stop >>>>> so CPU based IDs are kept constant throughout any perf session. >>>>> >>>>> >>>>> Note: This patchset breaks some backward compatibility for perf >>>>> record and >>>>> perf report. >>>>> >>>>> The version of the AUXTRACE_INFO has been updated to reflect the >>>>> fact that >>>>> the trace source IDs are generated differently. This will >>>>> mean older versions of perf report cannot decode the newer file. >>>>> >>>>> Appies to coresight/next >>>>> >>> >>> ... >>> >>>>> Mike Leach (15): >>>>> coresight: trace-id: Add API to dynamically assign Trace ID values >>>>> coresight: Remove obsolete Trace ID unniqueness checks >>>>> coresight: perf: traceid: Add perf ID allocation and notifiers >>>>> coresight: stm: Update STM driver to use Trace ID API >>>>> coresight: etm4x: Update ETM4 driver to use Trace ID API >>>>> coresight: etm3x: Update ETM3 driver to use Trace ID API >>>>> coresight: etmX.X: stm: Remove trace_id() callback >>>>> coresight: trace id: Remove legacy get trace ID function. >>>>> perf: cs-etm: Move mapping of Trace ID and cpu into helper function >>>>> perf: cs-etm: Update record event to use new Trace ID protocol >>>>> kernel: events: Export perf_report_aux_output_id() >>>>> perf: cs-etm: Handle PERF_RECORD_AUX_OUTPUT_HW_ID packet >>>>> coresight: events: PERF_RECORD_AUX_OUTPUT_HW_ID used for Trace ID >>>>> coresight: trace-id: Add debug & test macros to Trace ID allocation >>>>> coresight: etm3x: docs: Alter sysfs documentation for trace id >>>>> updates >>>> >>>> I have pushed the kernel patches on this series to coresight tree >>>> next[0]. >>>> >>>> I will be sending them out to Greg for v6.3. Please could you queue the >>>> perf tool changes in the series ? i.e., Patches 9,10 and 12. >>>> >>>> >>>> [0] >>>> https://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux.git/log/?h=next >>>> >>> >>> Please could you pick up the perf tool changes above ? >> >> Please could you confirm if you are able to queue the perf tool >> changes ? Or would you like me route it via coresight tree ? >> >> Kind regards >> Suzuki >> > > > Suzuki >
Hi Mike, On 27-02-2023 03:56 pm, Suzuki K Poulose wrote: > Mike, > > Please could you send all the perf tool related patches as a seperate > series, with all the tags to Arnaldo directly at -rc1. > > Please add a cover letter explaining that the kernel changes are > available upstream. Looks like, Perf tool patches corresponding to dynamic trace id are not yet merged in to perf-tools-next. Without these patches, the mainline perf tool breaks for the coresight for higher cores. > > Kind regards > Suzuki > > On 20/02/2023 16:24, Suzuki K Poulose wrote: >> Arnaldo >> >> Gentle ping >> >> >> On 31/01/2023 11:49, Suzuki K Poulose wrote: >>> Hi Arnaldo >>> >>> On 24/01/2023 11:36, Suzuki K Poulose wrote: >>>> Hi Arnaldo >>>> >>>> Gentle ping. >>>> >>>> On 19/01/2023 12:00, Suzuki K Poulose wrote: >>>>> Hi Arnaldo, >>>>> >>>>> On 16/01/2023 12:49, Mike Leach wrote: >>>>>> The current method for allocating trace source ID values to >>>>>> sources is >>>>>> to use a fixed algorithm for CPU based sources of (cpu_num * 2 + >>>>>> 0x10). >>>>>> The STM is allocated ID 0x1. >>>>>> >>>>>> This fixed algorithm is used in both the CoreSight driver code, >>>>>> and by >>>>>> perf when writing the trace metadata in the AUXTRACE_INFO record. >>>>>> >>>>>> The method needs replacing as currently:- >>>>>> 1. It is inefficient in using available IDs. >>>>>> 2. Does not scale to larger systems with many cores and the algorithm >>>>>> has no limits so will generate invalid trace IDs for cpu number > 44. >>>>>> >>>>>> Additionally requirements to allocate additional system IDs on some >>>>>> systems have been seen. >>>>>> >>>>>> This patch set introduces an API that allows the allocation of >>>>>> trace IDs >>>>>> in a dynamic manner. >>>>>> >>>>>> Architecturally reserved IDs are never allocated, and the system is >>>>>> limited to allocating only valid IDs. >>>>>> >>>>>> Each of the current trace sources ETM3.x, ETM4.x and STM is >>>>>> updated to use >>>>>> the new API. >>>>>> >>>>>> For the ETMx.x devices IDs are allocated on certain events >>>>>> a) When using sysfs, an ID will be allocated on hardware enable, >>>>>> or a read of >>>>>> sysfs TRCTRACEID register and freed when the sysfs reset is written. >>>>>> >>>>>> b) When using perf, ID is allocated on during setup AUX event, and >>>>>> freed on >>>>>> event free. IDs are communicated using the AUX_OUTPUT_HW_ID packet. >>>>>> The ID allocator is notified when perf sessions start and stop >>>>>> so CPU based IDs are kept constant throughout any perf session. >>>>>> >>>>>> >>>>>> Note: This patchset breaks some backward compatibility for perf >>>>>> record and >>>>>> perf report. >>>>>> >>>>>> The version of the AUXTRACE_INFO has been updated to reflect the >>>>>> fact that >>>>>> the trace source IDs are generated differently. This will >>>>>> mean older versions of perf report cannot decode the newer file. >>>>>> >>>>>> Appies to coresight/next >>>>>> >>>> >>>> ... >>>> >>>>>> Mike Leach (15): >>>>>> coresight: trace-id: Add API to dynamically assign Trace ID values >>>>>> coresight: Remove obsolete Trace ID unniqueness checks >>>>>> coresight: perf: traceid: Add perf ID allocation and notifiers >>>>>> coresight: stm: Update STM driver to use Trace ID API >>>>>> coresight: etm4x: Update ETM4 driver to use Trace ID API >>>>>> coresight: etm3x: Update ETM3 driver to use Trace ID API >>>>>> coresight: etmX.X: stm: Remove trace_id() callback >>>>>> coresight: trace id: Remove legacy get trace ID function. >>>>>> perf: cs-etm: Move mapping of Trace ID and cpu into helper >>>>>> function >>>>>> perf: cs-etm: Update record event to use new Trace ID protocol >>>>>> kernel: events: Export perf_report_aux_output_id() >>>>>> perf: cs-etm: Handle PERF_RECORD_AUX_OUTPUT_HW_ID packet >>>>>> coresight: events: PERF_RECORD_AUX_OUTPUT_HW_ID used for Trace ID >>>>>> coresight: trace-id: Add debug & test macros to Trace ID >>>>>> allocation >>>>>> coresight: etm3x: docs: Alter sysfs documentation for trace id >>>>>> updates >>>>> >>>>> I have pushed the kernel patches on this series to coresight tree >>>>> next[0]. >>>>> >>>>> I will be sending them out to Greg for v6.3. Please could you queue >>>>> the >>>>> perf tool changes in the series ? i.e., Patches 9,10 and 12. >>>>> >>>>> >>>>> [0] >>>>> https://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux.git/log/?h=next >>>>> >>>> >>>> Please could you pick up the perf tool changes above ? >>> >>> Please could you confirm if you are able to queue the perf tool >>> changes ? Or would you like me route it via coresight tree ? >>> >>> Kind regards >>> Suzuki >>> >> >> >> Suzuki >> > Thanks, Ganapat
Em Wed, Mar 29, 2023 at 09:44:47AM +0530, Ganapatrao Kulkarni escreveu: > Hi Mike, > > On 27-02-2023 03:56 pm, Suzuki K Poulose wrote: > > Mike, > > > > Please could you send all the perf tool related patches as a seperate > > series, with all the tags to Arnaldo directly at -rc1. > > > > Please add a cover letter explaining that the kernel changes are > > available upstream. > > Looks like, Perf tool patches corresponding to dynamic trace id are not yet > merged in to perf-tools-next. > > Without these patches, the mainline perf tool breaks for the coresight for > higher cores. I'm pushing what I have to tmp.perf-tools-next while this + other changes goes thru my container set build tests. - Arnaldo > > > > Kind regards > > Suzuki > > > > On 20/02/2023 16:24, Suzuki K Poulose wrote: > > > Arnaldo > > > > > > Gentle ping > > > > > > > > > On 31/01/2023 11:49, Suzuki K Poulose wrote: > > > > Hi Arnaldo > > > > > > > > On 24/01/2023 11:36, Suzuki K Poulose wrote: > > > > > Hi Arnaldo > > > > > > > > > > Gentle ping. > > > > > > > > > > On 19/01/2023 12:00, Suzuki K Poulose wrote: > > > > > > Hi Arnaldo, > > > > > > > > > > > > On 16/01/2023 12:49, Mike Leach wrote: > > > > > > > The current method for allocating trace source ID > > > > > > > values to sources is > > > > > > > to use a fixed algorithm for CPU based sources of > > > > > > > (cpu_num * 2 + 0x10). > > > > > > > The STM is allocated ID 0x1. > > > > > > > > > > > > > > This fixed algorithm is used in both the CoreSight > > > > > > > driver code, and by > > > > > > > perf when writing the trace metadata in the AUXTRACE_INFO record. > > > > > > > > > > > > > > The method needs replacing as currently:- > > > > > > > 1. It is inefficient in using available IDs. > > > > > > > 2. Does not scale to larger systems with many cores and the algorithm > > > > > > > has no limits so will generate invalid trace IDs for cpu number > 44. > > > > > > > > > > > > > > Additionally requirements to allocate additional system IDs on some > > > > > > > systems have been seen. > > > > > > > > > > > > > > This patch set introduces an API that allows the > > > > > > > allocation of trace IDs > > > > > > > in a dynamic manner. > > > > > > > > > > > > > > Architecturally reserved IDs are never allocated, and the system is > > > > > > > limited to allocating only valid IDs. > > > > > > > > > > > > > > Each of the current trace sources ETM3.x, ETM4.x and > > > > > > > STM is updated to use > > > > > > > the new API. > > > > > > > > > > > > > > For the ETMx.x devices IDs are allocated on certain events > > > > > > > a) When using sysfs, an ID will be allocated on > > > > > > > hardware enable, or a read of > > > > > > > sysfs TRCTRACEID register and freed when the sysfs reset is written. > > > > > > > > > > > > > > b) When using perf, ID is allocated on during setup > > > > > > > AUX event, and freed on > > > > > > > event free. IDs are communicated using the AUX_OUTPUT_HW_ID packet. > > > > > > > The ID allocator is notified when perf sessions start and stop > > > > > > > so CPU based IDs are kept constant throughout any perf session. > > > > > > > > > > > > > > > > > > > > > Note: This patchset breaks some backward > > > > > > > compatibility for perf record and > > > > > > > perf report. > > > > > > > > > > > > > > The version of the AUXTRACE_INFO has been updated to > > > > > > > reflect the fact that > > > > > > > the trace source IDs are generated differently. This will > > > > > > > mean older versions of perf report cannot decode the newer file. > > > > > > > > > > > > > > Appies to coresight/next > > > > > > > > > > > > > > > > > ... > > > > > > > > > > > > Mike Leach (15): > > > > > > > coresight: trace-id: Add API to dynamically assign Trace ID values > > > > > > > coresight: Remove obsolete Trace ID unniqueness checks > > > > > > > coresight: perf: traceid: Add perf ID allocation and notifiers > > > > > > > coresight: stm: Update STM driver to use Trace ID API > > > > > > > coresight: etm4x: Update ETM4 driver to use Trace ID API > > > > > > > coresight: etm3x: Update ETM3 driver to use Trace ID API > > > > > > > coresight: etmX.X: stm: Remove trace_id() callback > > > > > > > coresight: trace id: Remove legacy get trace ID function. > > > > > > > perf: cs-etm: Move mapping of Trace ID and cpu > > > > > > > into helper function > > > > > > > perf: cs-etm: Update record event to use new Trace ID protocol > > > > > > > kernel: events: Export perf_report_aux_output_id() > > > > > > > perf: cs-etm: Handle PERF_RECORD_AUX_OUTPUT_HW_ID packet > > > > > > > coresight: events: PERF_RECORD_AUX_OUTPUT_HW_ID used for Trace ID > > > > > > > coresight: trace-id: Add debug & test macros to > > > > > > > Trace ID allocation > > > > > > > coresight: etm3x: docs: Alter sysfs documentation > > > > > > > for trace id updates > > > > > > > > > > > > I have pushed the kernel patches on this series to > > > > > > coresight tree next[0]. > > > > > > > > > > > > I will be sending them out to Greg for v6.3. Please > > > > > > could you queue the > > > > > > perf tool changes in the series ? i.e., Patches 9,10 and 12. > > > > > > > > > > > > > > > > > > [0] https://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux.git/log/?h=next > > > > > > > > > > > > > > > > Please could you pick up the perf tool changes above ? > > > > > > > > Please could you confirm if you are able to queue the perf tool > > > > changes ? Or would you like me route it via coresight tree ? > > > > > > > > Kind regards > > > > Suzuki > > > > > > > > > > > > > Suzuki > > > > > > > Thanks, > Ganapat >
On 29-03-2023 06:49 pm, Arnaldo Carvalho de Melo wrote: > Em Wed, Mar 29, 2023 at 09:44:47AM +0530, Ganapatrao Kulkarni escreveu: >> Hi Mike, >> >> On 27-02-2023 03:56 pm, Suzuki K Poulose wrote: >>> Mike, >>> >>> Please could you send all the perf tool related patches as a seperate >>> series, with all the tags to Arnaldo directly at -rc1. >>> >>> Please add a cover letter explaining that the kernel changes are >>> available upstream. >> >> Looks like, Perf tool patches corresponding to dynamic trace id are not yet >> merged in to perf-tools-next. >> >> Without these patches, the mainline perf tool breaks for the coresight for >> higher cores. > > I'm pushing what I have to tmp.perf-tools-next while this + other > changes goes thru my container set build tests. Thanks Arnaldo! > > - Arnaldo > >>> >>> Kind regards >>> Suzuki >>> >>> On 20/02/2023 16:24, Suzuki K Poulose wrote: >>>> Arnaldo >>>> >>>> Gentle ping >>>> >>>> >>>> On 31/01/2023 11:49, Suzuki K Poulose wrote: >>>>> Hi Arnaldo >>>>> >>>>> On 24/01/2023 11:36, Suzuki K Poulose wrote: >>>>>> Hi Arnaldo >>>>>> >>>>>> Gentle ping. >>>>>> >>>>>> On 19/01/2023 12:00, Suzuki K Poulose wrote: >>>>>>> Hi Arnaldo, >>>>>>> >>>>>>> On 16/01/2023 12:49, Mike Leach wrote: >>>>>>>> The current method for allocating trace source ID >>>>>>>> values to sources is >>>>>>>> to use a fixed algorithm for CPU based sources of >>>>>>>> (cpu_num * 2 + 0x10). >>>>>>>> The STM is allocated ID 0x1. >>>>>>>> >>>>>>>> This fixed algorithm is used in both the CoreSight >>>>>>>> driver code, and by >>>>>>>> perf when writing the trace metadata in the AUXTRACE_INFO record. >>>>>>>> >>>>>>>> The method needs replacing as currently:- >>>>>>>> 1. It is inefficient in using available IDs. >>>>>>>> 2. Does not scale to larger systems with many cores and the algorithm >>>>>>>> has no limits so will generate invalid trace IDs for cpu number > 44. >>>>>>>> >>>>>>>> Additionally requirements to allocate additional system IDs on some >>>>>>>> systems have been seen. >>>>>>>> >>>>>>>> This patch set introduces an API that allows the >>>>>>>> allocation of trace IDs >>>>>>>> in a dynamic manner. >>>>>>>> >>>>>>>> Architecturally reserved IDs are never allocated, and the system is >>>>>>>> limited to allocating only valid IDs. >>>>>>>> >>>>>>>> Each of the current trace sources ETM3.x, ETM4.x and >>>>>>>> STM is updated to use >>>>>>>> the new API. >>>>>>>> >>>>>>>> For the ETMx.x devices IDs are allocated on certain events >>>>>>>> a) When using sysfs, an ID will be allocated on >>>>>>>> hardware enable, or a read of >>>>>>>> sysfs TRCTRACEID register and freed when the sysfs reset is written. >>>>>>>> >>>>>>>> b) When using perf, ID is allocated on during setup >>>>>>>> AUX event, and freed on >>>>>>>> event free. IDs are communicated using the AUX_OUTPUT_HW_ID packet. >>>>>>>> The ID allocator is notified when perf sessions start and stop >>>>>>>> so CPU based IDs are kept constant throughout any perf session. >>>>>>>> >>>>>>>> >>>>>>>> Note: This patchset breaks some backward >>>>>>>> compatibility for perf record and >>>>>>>> perf report. >>>>>>>> >>>>>>>> The version of the AUXTRACE_INFO has been updated to >>>>>>>> reflect the fact that >>>>>>>> the trace source IDs are generated differently. This will >>>>>>>> mean older versions of perf report cannot decode the newer file. >>>>>>>> >>>>>>>> Appies to coresight/next >>>>>>>> >>>>>> >>>>>> ... >>>>>> >>>>>>>> Mike Leach (15): >>>>>>>> coresight: trace-id: Add API to dynamically assign Trace ID values >>>>>>>> coresight: Remove obsolete Trace ID unniqueness checks >>>>>>>> coresight: perf: traceid: Add perf ID allocation and notifiers >>>>>>>> coresight: stm: Update STM driver to use Trace ID API >>>>>>>> coresight: etm4x: Update ETM4 driver to use Trace ID API >>>>>>>> coresight: etm3x: Update ETM3 driver to use Trace ID API >>>>>>>> coresight: etmX.X: stm: Remove trace_id() callback >>>>>>>> coresight: trace id: Remove legacy get trace ID function. >>>>>>>> perf: cs-etm: Move mapping of Trace ID and cpu >>>>>>>> into helper function >>>>>>>> perf: cs-etm: Update record event to use new Trace ID protocol >>>>>>>> kernel: events: Export perf_report_aux_output_id() >>>>>>>> perf: cs-etm: Handle PERF_RECORD_AUX_OUTPUT_HW_ID packet >>>>>>>> coresight: events: PERF_RECORD_AUX_OUTPUT_HW_ID used for Trace ID >>>>>>>> coresight: trace-id: Add debug & test macros to >>>>>>>> Trace ID allocation >>>>>>>> coresight: etm3x: docs: Alter sysfs documentation >>>>>>>> for trace id updates >>>>>>> >>>>>>> I have pushed the kernel patches on this series to >>>>>>> coresight tree next[0]. >>>>>>> >>>>>>> I will be sending them out to Greg for v6.3. Please >>>>>>> could you queue the >>>>>>> perf tool changes in the series ? i.e., Patches 9,10 and 12. >>>>>>> >>>>>>> >>>>>>> [0] https://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux.git/log/?h=next >>>>>>> >>>>>> >>>>>> Please could you pick up the perf tool changes above ? >>>>> >>>>> Please could you confirm if you are able to queue the perf tool >>>>> changes ? Or would you like me route it via coresight tree ? >>>>> >>>>> Kind regards >>>>> Suzuki >>>>> >>>> >>>> >>>> Suzuki >>>> >>> >> >> Thanks, >> Ganapat >> > Thanks, Ganapat