コミットグラフ

77 コミット

作成者 SHA1 メッセージ 日付
Chun Yang f9017cbdc5 SWDEV-296922 : Incorrect rounding due to integer division in rocprofiler metrics
Changed derived metrics to double from int64.
Fixed standalone test due to int64 to float change
Fixed intercept test due to int64 to float change.

Change-Id: I49631c187406ae9dd94a869b3bb13772012e8cdf
2021-09-23 14:52:35 -07:00
AMD 4df3e0bd9a Add support for gfx90a
Merge gfx90a support from the 'amd-npi' branch.

Change-Id: I9b51711ed4a1d2f1ed42ba9b83cb12136be228b8
2021-06-16 16:35:42 -07:00
Evgeny 780dfa37d4 cleanup after separating for staging and npi branches
Change-Id: Iadd624df21b85f1590e901a8125680743e3281a3
2021-04-08 20:37:47 +00:00
Evgeny 82d7bb2145 SWDEV-265287 : integration spmltgen.py script
Change-Id: Ief3e93225fb6660e72a04e4bd4b379262b73c914
2021-04-08 10:04:39 -04:00
Evgeny e2c9d13e5b SWDEV-274821 SPM initialization fix
Change-Id: I5e27928a60083eff328bab3e79937ce11bce11bd
2021-03-22 09:18:36 +00:00
Evgeny 7e60bf163e SWDEV-255662 : spm kfd mode support
Change-Id: I840c7e92d3d5a59d8e5402c4d8ef86bc123dd07c
2020-12-02 13:02:45 -06:00
Evgeny f2c9980647 fixing sqtt trace for zero size case
Change-Id: I75712485f518725af46a3b419339a212d1e762a0
2020-12-01 18:19:51 -05:00
Evgeny ccc6005c25 fixing c_str() as strdup
Change-Id: Ib5cb68d16ce66fd2ae072168de4c16895f32b57f
2020-10-27 14:45:51 -05:00
Evgeny 0d164ba672 enable contexts wait
Change-Id: Ie2adf04662fddc8051fb5418904c9c659e264d78
2020-09-21 21:06:03 -04:00
Evgeny 8850e46071 kernel objects dumping
Change-Id: I5a16e05b7df438efa903948701b65a9ced99e5f3

initial codeobj event implementation

Change-Id: Ia7fac3c2b9897a004cfe88c4de82ba8c18284196

update - codeobj event implementation

Change-Id: I2b91b6e689875af03f0086f5a0872a97a629fd83

update2 - codeobj event implementation

Change-Id: Icff75f14fd21963e40db95373fa74880957a9e32

fix - codeobj event implementation

Change-Id: I76c33c875cb429fb12a974bb408b217f187b4536

URI buffer fix - codeobj event implementation

Change-Id: I7ce1a758e021455da3fe5b8a6e4ae3ab46e9760e

HSA events exposing

Change-Id: I3664ab4e5111c4ccedaf068dcb19f48055f0ef9b

HSA events data struct normalizing

Change-Id: I365ef0db45e0a9314bd2a1a4d29dd4eb4e91297d
2020-09-11 10:01:54 -05:00
Xianwei Zhang b445610cd1 concurrent: enable/fix the related settings
Concurrent profiling relies on the aqlprofile read_api
and tracker. This patch set those options to enable
the concurrent profiling.

Change-Id: Ib97d4d8facfbc11f2684d83109397cd13f117d5e
2020-08-26 16:04:57 -04:00
Evgeny 80747de208 optimization mechanism fix: correct tracker handler; kernel name query on completion;
Change-Id: I14da152b4ac3c7d8fd1af2f54e9d71f834071622
2020-08-03 23:34:49 -05:00
Evgeny 7364edcc5b kernel name filtering fix - handling [] brakets
Change-Id: I46a62d991a52045694640837393df229cf7a3133
2020-07-29 18:47:31 -05:00
Xianwei Zhang 61c9df4631 pmc: add support of concurrent kernel profiling
The profiling was only enabled in serial mode, i.e., kernels
are serialized in execution, and counters are reset at each
kernel start and read at kernel completion. This patch adds
the concurrent mode, by issuing the process-level start
packet to reset counters, and then reading twice at kernel
start and end time to obtain the counter value difference.
The new concurrent profiling usage needs the integration
with the corresponding augment at aqlprofile side.

Change-Id: I94b4442eadc8c64b8fba51b1e4916fc8b895ad21
2020-07-16 14:39:46 -05:00
Evgeny 2a7f77b290 counters dumping optimization
Change-Id: I8c694e5380e15179453148dd9ab3a3e51b6db861
2020-07-15 09:57:41 -05:00
Evgeny f214238880 cleanup
Change-Id: I518f741d3d3d0bb514936c4db8b727bdc79d9fdc
2020-06-24 12:22:22 -05:00
Evgeny 9f7e936d70 concurrent sqtt support
Change-Id: I91391fafabc93aefa5d244d870ef82b96a59dc52
2020-06-23 20:00:49 -04:00
Evgeny 9950b97567 disabling destruction
Change-Id: I2a7d05a8f597b3bc8bd07bffe7181f9dcace1cbf
2020-05-07 03:34:30 -05:00
Evgeny 3af87a7423 adding pid for kernel results to support multi-process profiling
Change-Id: I283228a4b4145599c5e637dd6faa771b9f4b6345
2020-05-05 05:35:32 -05:00
Evgeny fe70682184 toll destruction fix
Change-Id: If069c820526e21a0a4b80ac516f9669a81f34cab
2020-04-28 03:16:15 -05:00
Evgeny 7be9a42ab3 fixing hsa intercept test
Change-Id: I2671dfc6a9bd3e01a0c926aa3ea367b8c7a0279e
2020-03-28 17:24:16 -05:00
Evgeny 9df9fddcfb PC sampling bringup
Change-Id: I0d041c4c8c3778f2c328cde38432bc72223706a3

pc sampling integration fix

Change-Id: Ia66ff876d2d99ec4d561daf8320b65d75f5cd2fe
2020-03-28 13:07:45 -05:00
Evgeny 299a08e4be update for gfx9 and 3.1
Change-Id: I0c218f8a2dcdad1c8deb44770d8a64ccd95a92fb
2020-03-24 19:07:11 -05:00
Evgeny 7ecf9ef1e3 dynamic kernel lds size
Change-Id: I051b85550d12ca8662127615be7adfb12c5f9585
2020-02-27 17:44:22 -06:00
Evgeny 40730e34e4 adding AgentInfo::lds_block_size
Change-Id: I186893add96dc92570e710ae78b475897ebfe531
2020-02-18 14:00:19 -06:00
Evgeny d2324e780f hsa-interept mode enabling in head
Change-Id: I048a975deaae691868034ee361fea71ffb5bdd83
2020-02-14 00:22:36 -06:00
Evgeny 3d4c4d1cce device id/type for submit callback
Change-Id: I69b31bd28850198c8ca44e55bf8f2ff9540d92fc
2020-02-13 22:00:56 -06:00
Evgeny 4edc6c98cf adding agent to device callback data
Change-Id: I7d631e4faae685ce23704ac65308bf2aae0c157a
2020-02-13 22:00:56 -06:00
Evgeny 304568c1a3 kernel free on executable_destroy
Change-Id: I062e08fd5df94f4d21b443b74494dc6e95201f3d
2020-02-13 22:00:56 -06:00
Evgeny 086c29fff1 adding code object freeze handling, kernel code allocate/device callbacks
Change-Id: I6ec32a0c4e945260cd11f88c1d2a0a7f22d6e10c

documenting comments

Change-Id: I1b7af70d51042c48aef0474d0b6b54275b42c456
2020-02-13 22:00:56 -06:00
Evgeny f95ed067ba adding allocate/device/memcopy/submit callbacks
Change-Id: Ie83ae3cd57cdf5038e83db70891b103439c37d55
2020-02-13 22:00:08 -06:00
Evgeny 389cb50854 trace parameters fix
Change-Id: I5cd8c897b9982f9abc9e36c18a4a488bd65fbd30
2019-12-10 04:33:29 -06:00
Evgeny c8eb0641bf sgpr/vgpr reported sizes in terms of registers; lds size in bytes;
Change-Id: Id527687cc075a4e85ea4d0063ed8f336e34d2454
2019-11-27 15:19:14 -06:00
Evgeny 60043d198b code object tracking and v3 code object support
Change-Id: I081ada13f6364ea4401a97a485dedfa9bf8a45fe
2019-11-19 20:18:09 -06:00
Evgeny dc05345a78 setting of granulated lds, sgpr, vgpr reserved by HW for the kernel
Change-Id: Ib3e60c0b97787f56f769136480a46a8560e00cc8
2019-11-11 18:11:16 -05:00
Evgeny e5b938b5fb unordered metrics expressions config
Change-Id: I998423046abfd7af45f28071911276cce1c4f3da
2019-08-25 05:08:55 -05:00
Evgeny Shcherbakov 4f4e02de7a Merge "build: slightly change files to pass compilation" into amd-master 2019-08-23 11:44:44 -04:00
Evgeny aecc7abc76 Mem32Bwrites -> MemWrites32B renaming
Change-Id: I86f570b97178caa3d3d5fcbc5ecc716216dad504
2019-08-22 22:34:36 -05:00
Evgeny dff3256dd5 mi100 metrics fix
Change-Id: I862578a2b7827a88cc6926d636d5eb9a8320a244
2019-08-22 22:30:48 -05:00
Xianwei Zhang ee917344ff build: slightly change files to pass compilation
Compiled using Clang and encountered errors on ambiguous pointer
type, implicit conversion NULL to bool, and unused function, etc.
As a fix, this patch slightly updates the related codes.

Change-Id: I4a73d66293313dda36749036f5c73aaf2a445231
2019-08-22 23:29:16 -04:00
Evgeny 82ecf9e7a4 overwriting on inheritance
Change-Id: I6faf79874fc14e898c75972aa2f372e228105f7b
2019-08-08 01:54:29 -05:00
Evgeny 09eed4f35b gfx908 events update
Change-Id: Ieb7c72f66b7e5e5dfc29d97650a29722a6276a31
2019-08-07 19:44:31 -05:00
Evgeny 855a801ae7 adding gfx908 namespace
Change-Id: Ifd45385777be5066c8bfef23ee24bbabadf4dffa
2019-07-27 03:06:10 -05:00
Xianwei Zhang ca9a714b77 xml: add extra events for gfx906 GPUs
Compard to gfx900, e.g., Vega 10, gfx906 adds extra counter events.
A typical difference is on TCC-EA that gfx906 (e.g., Vega 20) has 2
EAs per TCC, while only one single EA/TCC on gfx900. As such,
additional counters must be profiled to get correct results. This
patch adds one extra events to specifically handle gfx906.

Change-Id: Id6c9d37548a102c80bbfddcfa11e77d20f17431a
2019-07-16 12:56:45 -05:00
Xianwei Zhang 8b445d2c00 test: add tests to validate cache/mem blocks
To validate cache and memory blocks profiling, this patch prepares
tests to profile dedicated kernels using specified counters, to
compare the profiled results against expected ones, and further
show the test is a fail or pass.  Tests here are focusing on cache
hit/miss, memory fetch/write size.

Change-Id: Icbc8096a6e15256dec66297597a57c7665a533b8
2019-07-16 12:39:38 -05:00
Evgeny 46fd670818 adding xml inheritance
Change-Id: Iafbfb8a6fb1936cf1315c1f5215879e99187da77
2019-07-16 09:20:53 -07:00
Evgeny 988283533f adding SPM test
Change-Id: I55dde52e668206e238ea860dd11e0ed5501a582a
2019-06-04 15:34:21 -05:00
Evgeny 6d6eb2b2d0 adding SPM trace support
Change-Id: I07c0d3ec5c4ec51ea00fea539c930c6ed27b1163
2019-06-04 05:07:19 -05:00
Evgeny 8c15ff4bf8 extending xontext pool API
Change-Id: I56809354a9ea991489479117160f3d4f451f82b3
2019-03-29 11:12:40 -05:00
Evgeny f834529669 improving rocprof tracing usage
Change-Id: I4db513c93406a7038b8b681d12a0dc4298975f01
2019-03-12 18:19:17 -05:00