نمودار کامیت
22 کامیت‌ها
مولف SHA1 پیام تاریخ
Vladimir Indic 56e33c7543 More comprehensive IOCTL PC sampling checks (mi200 and mi300) (#1045)
* More comprehensive IOCTL PC sampling checks (mi200 and mi300)

* PC sampling tests: formatting

[ROCm/rocprofiler-sdk commit: 8bf2ce622c]
2024-09-06 09:45:05 -05:00
Vladimir Indic 3ed3489670 PC sampling: online partial PC sampling decoding (#1004)
* PC sampling: online partial PC sampling decoding

PC sampling service decodes a PC sample partially
by replacing the PC with an id of the loaded code object instance
containing PC and the offset of the PC within that code object instance.

* PC sampling: marker records removed

* PC sampling parser: minor doc update in mock

* PC sampling: introducing rocprofiler_pc_t

* NULL value of the code object id introduced.

* Clarifying documenation related to PC offset.

* PC offset documentation improvement

* PC sampling parser benchmark: Reducing the number of samples to recreate half of performance.

[ROCm/rocprofiler-sdk commit: 93e82663d9]
2024-09-05 11:35:46 -05:00
Vladimir Indic 3cd5bdf100 PC sampling: chiplet id + integration test fix (#983)
* PCS: show chiplet; cover loading/unloading in integration test

* Use (code_object_id, pc_addr) pair as instruction id.

[ROCm/rocprofiler-sdk commit: 0f89f0449d]
2024-07-22 16:00:59 +05:30
Vladimir Indic 1c98e94488 PC sampling: integration test with instruction decoding (#929)
* PC sampling: integration test with instruction decoding

* PC sampling: verifying internal and external CIDs

The PC sampling integration test has been extended
to verify internal and external correlation IDs.

* tmp solution of using Instructions as keys

* wrapper for HIP call

* PCS integration test: ld_addr as instruction id

For the sake of the integration test, use  as the
instruction identifier. To support code object unloading
and relocations, use  as the identifier
(the change in the decoder is required).

* PCS integration test: removing shared_ptr

Completely removing usage of shared pointers.

* PCS integration test: removing decoder

When a code object has been unloaded, ensure all PC samples
corresponding to that object are decoded, prior to removing
the decoder.

* PCS integration test: fixing build flags and imports

* PCS integration test: fixing labels

* PCS integration test: cmake flags fix

* PC sampling cmake labels renamed

* PCS integration test refactoring

* PCS integration test: minimize usage of raw pointers

* PCS integration test: at least one sample should be delivered.

* PC sampling lables: pc-sampling

[ROCm/rocprofiler-sdk commit: 498b1f2bd7]
2024-07-04 11:19:27 +05:30
Vladimir Indic 2e10d7e2fa PC sampling IOCTL versioning refactored (#945)
The following changes are introduced:
- Use functions instead of macros.
- Verify the error code when querying KFD IOCTL version.
- Skip tests and samples if KFD IOCTL < 1.16 or PC Sampling IOCTL < 0.1.

[ROCm/rocprofiler-sdk commit: 79e81c9b39]
2024-06-21 11:23:00 +02:00
Vladimir Indic 439598f169 PC Sampling IOCTL version check introduced (#944)
[ROCm/rocprofiler-sdk commit: b6833710dc]
2024-06-20 08:20:46 +05:30
Vladimir Indic 6119a60191 Disable counter collection if PC sampling is enabled (#924)
[ROCm/rocprofiler-sdk commit: c66aacd537]
2024-06-13 18:26:05 +02:00
Vladimir Indic ba1e5d1ff3 The NULL value of an internal correlation ID defined (#901)
[ROCm/rocprofiler-sdk commit: b4f7154716]
2024-06-10 16:12:01 +02:00
Vladimir Indic b1c3ec2212 Disable PC sampling service if counter collection service is configured (#899)
[ROCm/rocprofiler-sdk commit: 211ee219c4]
2024-06-10 15:13:49 +02:00
Vladimir Indic cb69616d46 PC sampling client: using raw pointers (#902)
* PC sampling client: using raw pointers to prevent premature destruction of buffers

* PCS client: freeing buffer_ids

[ROCm/rocprofiler-sdk commit: b0c41827c3]
2024-06-04 13:16:26 -05:00
Vladimir Indic 592d9edf6e PC sampling: CID manager unit test (#898)
[ROCm/rocprofiler-sdk commit: 1c5bf63dbc]
2024-06-04 13:16:26 -05:00
Vladimir IndicوJonathan R. Madsen 86b6063aad PC Sampling - blit kernels handling (#878)
* correlation id 0 for blit kernels' samples

* rocprofiler_pc_sampling_snapshot_v1_t docs update (for future use)

* ROCPROFILER_CORRELATION_ID_VALUE_NONE introduced

* Simplifying ROCPROFILER_CORRELATION_ID_VALUE_NONE

* Apply suggestions from code review

Co-authored-by: Vladimir Indic <139573562+vlaindic@users.noreply.github.com>

---------

Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>

[ROCm/rocprofiler-sdk commit: c45bf712d1]
2024-05-29 12:55:47 +02:00
Vladimir Indic 176d812103 (IMPORTANT: AFAR V ) Hide PC sampling if HSA does not support it. (#881)
* Hide PC sampling if HSA does not support it.

* missing hsa.hpp include

[ROCm/rocprofiler-sdk commit: 7201aa321e]
2024-05-29 03:02:14 +02:00
Vladimir Indic bbfe00e759 Updating kfd_ioctl.h 1.17 (#872)
The KFD minor version has been bumped to 17.

[ROCm/rocprofiler-sdk commit: 05890c2231]
2024-05-28 17:26:18 +02:00
Vladimir Indic c28c399b84 avoiding early destruction of code objects list (#847)
[ROCm/rocprofiler-sdk commit: 358c599c3f]
2024-05-08 11:00:08 +02:00
Vladimir IndicوJonathan R. Madsen d0016ef728 Restructure code object source code (#826)
* public codeobj info

* Restructure code object source code file layout

* Update get_unloaded_code_objects + add iterate_loaded_code_objects

* Remove get_unloaded_code_objects from visible internal API

- iterate_loaded_code_objects + functor which filters on the hsa_executable_t effectively reproduces this behavior

* Whitespace removal

---------

Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>

[ROCm/rocprofiler-sdk commit: 733aa8e438]
2024-04-25 14:03:04 -05:00
843398eb3c Introducing PcSamplingExtTable (#735)
* pcs: updating the PCS table

* Fixing Clang Tidy errors

* pcs: reverting old table version

* testint wrong table size

* new size

* testing step

* reverting old steps

* hsa_amd_queue_get_info introduced

* pcs: testing table version

* formatting

* removing redundand declarations

* removing unnecessary files

* Apply suggestions from code review

* Apply suggestions from code review

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Update source/lib/rocprofiler-sdk/hsa/pc_sampling.cpp

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Enable function pointer offset check in hsa::pc_sampling::copy_table

- add offset() to HSA_API_META_DEFINITION
- check if offset() >= size of struct

* Support build without PC sampling API table

* ids for ROCr's PC sampling public functions

---------

Co-authored-by: Ammar ELWazir <aelwazir@hpe6u-21.amd.com>
Co-authored-by: Jonathan R. Madsen <jrmadsen@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>

[ROCm/rocprofiler-sdk commit: deabd869b5]
2024-04-22 20:07:28 -05:00
Vladimir Indic 98b62efd12 pcs: API updated (#770)
[ROCm/rocprofiler-sdk commit: 200da0f274]
2024-04-16 22:51:16 -05:00
Vladimir Indicوgithub-actions[bot] 141783bb83 PCS parser is aware of external correlation IDs (#639)
* PCS parser is aware of external correlation IDs

* source formatting (clang-format v11) (#640)

Co-authored-by: vlaindic <139573562+vlaindic@users.noreply.github.com>

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

[ROCm/rocprofiler-sdk commit: 78939e705a]
2024-03-15 14:04:06 -05:00
43d5712dcb AmdExtTable updated (#292)
* AmdExtTable updated

* hsa_amd_agent_set_async_scratch_limit introduced

* source formatting (clang-format v11) (#294)

Co-authored-by: vlaindic <vlaindic@users.noreply.github.com>

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: vlaindic <vlaindic@users.noreply.github.com>

[ROCm/rocprofiler-sdk commit: 0666f6a197]
2023-12-12 10:36:38 -06:00
2271ed7730 pcs parser: includes fixed, GLOB removed (#286)
* pcs parser: includes fixed, GLOB removed

* source formatting (clang-format v11) (#287)

Co-authored-by: vlaindic <vlaindic@users.noreply.github.com>

* cmake formatting (cmake-format) (#288)

Co-authored-by: vlaindic <vlaindic@users.noreply.github.com>

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: vlaindic <vlaindic@users.noreply.github.com>

[ROCm/rocprofiler-sdk commit: 200be03bb4]
2023-12-08 17:36:32 +01:00
697c751c62 PC sampling examples: single user, multiple agents (#31)
* PC sampling examples: single user, multiple agents

The example assumes that a single user activates PC sampling service
on multiple agents. Preferably, the user chooses stochastic sampling
over host-trap sampling.

* cmake formatting (cmake-format) (#33)

Co-authored-by: vlaindic <vlaindic@users.noreply.github.com>

* source formatting (clang-format v11) (#32)

Co-authored-by: vlaindic <vlaindic@users.noreply.github.com>

* pc sampling samples: anynomus namespace and string_view

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: vlaindic <vlaindic@users.noreply.github.com>

[ROCm/rocprofiler-sdk commit: 76717efa64]
2023-08-30 17:08:48 -05:00