Wykres commitów

64512 Commity

Autor SHA1 Wiadomość Data
Ajay 17813d3e3c SWDEV-414828 - set AMD_DK_ROOT variable to env DK_ROOT
this ensures python to be picked from DK

Change-Id: Idd006b7a2bb819c6aa747136aa55cfc28737a5a9


[ROCm/clr commit: 0d3a5c4c81]
2023-08-17 12:21:16 -04:00
Ajay 0d3a5c4c81 SWDEV-414828 - set AMD_DK_ROOT variable to env DK_ROOT
this ensures python to be picked from DK

Change-Id: Idd006b7a2bb819c6aa747136aa55cfc28737a5a9
2023-08-17 12:21:16 -04:00
Cole Ramos b9b72711cc Update CHANGES for v1.0.9
Signed-off-by: Cole Ramos <colramos@amd.com>

[ROCm/rocprofiler-compute commit: 7244c01760]
2023-08-17 11:04:43 -05:00
Cole Ramos 7244c01760 Update CHANGES for v1.0.9
Signed-off-by: Cole Ramos <colramos@amd.com>
2023-08-17 11:04:43 -05:00
Cole Ramos 3f6ce7b4f7 Change README to v1.0.9
Signed-off-by: Cole Ramos <colramos@amd.com>

[ROCm/rocprofiler-compute commit: 9a15f37960]
2023-08-17 10:39:05 -05:00
Cole Ramos 9a15f37960 Change README to v1.0.9
Signed-off-by: Cole Ramos <colramos@amd.com>
2023-08-17 10:39:05 -05:00
coleramos425 8e30256bc1 Documentation enhancements
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: 064debf356]
2023-08-17 09:50:38 -05:00
coleramos425 064debf356 Documentation enhancements
Signed-off-by: coleramos425 <colramos@amd.com>
2023-08-17 09:50:38 -05:00
Ammar ELWazir 2b78928d60 SWDEV-415259: Replacing HCC with AMD
__HIP_PLATFORM_HCC__ with __HIP_PLATFORM_AMD__

Change-Id: I2e79cff5a0216125b4ce2aa90104d4a1696bad08


[ROCm/roctracer commit: 67b3717da9]
2023-08-17 09:23:45 -04:00
Ammar ELWazir 67b3717da9 SWDEV-415259: Replacing HCC with AMD
__HIP_PLATFORM_HCC__ with __HIP_PLATFORM_AMD__

Change-Id: I2e79cff5a0216125b4ce2aa90104d4a1696bad08
2023-08-17 09:23:45 -04:00
Jonathan Kim 8f21793a3e Fix SDMA ganging circular deadlock in oversubscription
When oversubscribing SDMA gangs, a circular deadlock can occur since
gang enqueue is staggered with respect to SDMA engine leader based
on source to destination.
As a result, an enqueued leader may be waiting on a gang item that is
waiting on another enqueued leader or gang item and so on.

To prevent this, first lock the submission to ensure dma status query
and submissions are atomic.  Once this is in place, be more stringent
with ganging in that all SDMA engines must be available in order to gang.

Finally, re-enable SDMA ganging by default.

Change-Id: I4511e3487db9d26475b5aece4897f10168cc5322
2023-08-17 08:49:09 -04:00
Jonathan Kim 704d9c5e19 Fix SDMA ganging circular deadlock in oversubscription
When oversubscribing SDMA gangs, a circular deadlock can occur since
gang enqueue is staggered with respect to SDMA engine leader based
on source to destination.
As a result, an enqueued leader may be waiting on a gang item that is
waiting on another enqueued leader or gang item and so on.

To prevent this, first lock the submission to ensure dma status query
and submissions are atomic.  Once this is in place, be more stringent
with ganging in that all SDMA engines must be available in order to gang.

Finally, re-enable SDMA ganging by default.

Change-Id: I4511e3487db9d26475b5aece4897f10168cc5322


[ROCm/ROCR-Runtime commit: 8f21793a3e]
2023-08-17 08:49:09 -04:00
Jonathan Kim 4c74e47e91 Update D2D SDMA ganging for non-SPX modes
xGMI for compute partitioning in non-SPX modes does not have
a reported bandwith.
Fix it to at most 2 since each partition is either bounded
by the number of xGMI links or the number of available
SDMA contexts.

Change-Id: I09094bd7548d9eee6f039b0efe849838e5de166e
2023-08-17 07:25:08 -04:00
Jonathan Kim 58d5f7354f Update D2D SDMA ganging for non-SPX modes
xGMI for compute partitioning in non-SPX modes does not have
a reported bandwith.
Fix it to at most 2 since each partition is either bounded
by the number of xGMI links or the number of available
SDMA contexts.

Change-Id: I09094bd7548d9eee6f039b0efe849838e5de166e


[ROCm/ROCR-Runtime commit: 4c74e47e91]
2023-08-17 07:25:08 -04:00
Jonathan Kim 30982ff6aa Bump the number of SDMA engines for gfx940
GFX940 can support up to 16 SDMA engines so bump it.

Change-Id: I41a95e66383036735712e317a57b239d84fcb78d
2023-08-17 07:25:08 -04:00
Jonathan Kim 2994cfa875 Bump the number of SDMA engines for gfx940
GFX940 can support up to 16 SDMA engines so bump it.

Change-Id: I41a95e66383036735712e317a57b239d84fcb78d


[ROCm/ROCR-Runtime commit: 30982ff6aa]
2023-08-17 07:25:08 -04:00
Jonathan Kim f8664e88e0 Break when finding ganged agent
There's no need to keep looking in the list once we find a ganged agent.

Change-Id: Ia0b9b484c88221a7966a814456942c19b1741978
2023-08-17 07:25:08 -04:00
Jonathan Kim 64e0037743 Break when finding ganged agent
There's no need to keep looking in the list once we find a ganged agent.

Change-Id: Ia0b9b484c88221a7966a814456942c19b1741978


[ROCm/ROCR-Runtime commit: f8664e88e0]
2023-08-17 07:25:08 -04:00
Hao Zhou 8edd88817a Merge amd-staging into amd-master 20230817
Signed-off-by: Hao Zhou <Hao.Zhou@amd.com>
Change-Id: Idb8b837f8b65e3e820719f39a6f2c0c4ebc5b7f9


[ROCm/rocm_smi_lib commit: 3921278c4e]
2023-08-17 13:27:13 +08:00
Hao Zhou 3921278c4e Merge amd-staging into amd-master 20230817
Signed-off-by: Hao Zhou <Hao.Zhou@amd.com>
Change-Id: Idb8b837f8b65e3e820719f39a6f2c0c4ebc5b7f9
2023-08-17 13:27:13 +08:00
Jaydeep Patel 86bac29b0a SWDEV-412393 - Force alloc memory to avoid another hsa image creation.
Change-Id: Ia3cd99eb736231e6dfe013ebae6c41fd4cc657bc


[ROCm/clr commit: 289535e805]
2023-08-17 05:18:43 +00:00
Jaydeep Patel 289535e805 SWDEV-412393 - Force alloc memory to avoid another hsa image creation.
Change-Id: Ia3cd99eb736231e6dfe013ebae6c41fd4cc657bc
2023-08-17 05:18:43 +00:00
Cole Ramos aabfac4363 Merge pull request #162 from AMDResearch/dev
Prep v1.0.9

[ROCm/rocprofiler-compute commit: 9aeac83961]
2023-08-16 14:29:40 -07:00
Cole Ramos 9aeac83961 Merge pull request #162 from AMDResearch/dev
Prep v1.0.9
2023-08-16 14:29:40 -07:00
Cole Ramos 0d03a08084 Merge branch 'main' into dev
[ROCm/rocprofiler-compute commit: eec0323945]
2023-08-16 14:29:29 -07:00
Cole Ramos eec0323945 Merge branch 'main' into dev 2023-08-16 14:29:29 -07:00
coleramos425 a6b78ee34d Renaming csv_converter to make more succinct
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: 3ae0198f8e]
2023-08-16 16:16:46 -05:00
coleramos425 3ae0198f8e Renaming csv_converter to make more succinct
Signed-off-by: coleramos425 <colramos@amd.com>
2023-08-16 16:16:46 -05:00
coleramos425 34946c95a7 Remove debug logging
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: f27142160c]
2023-08-16 15:34:55 -05:00
coleramos425 f27142160c Remove debug logging
Signed-off-by: coleramos425 <colramos@amd.com>
2023-08-16 15:34:55 -05:00
coleramos425 7e3d92e66d Prevent users from ipblock filtering in standalone GUI
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: 5d74b142b6]
2023-08-16 14:16:50 -05:00
coleramos425 5d74b142b6 Prevent users from ipblock filtering in standalone GUI
Signed-off-by: coleramos425 <colramos@amd.com>
2023-08-16 14:16:50 -05:00
coleramos425 d1406ba55a Add a -d option to analyze mode for consistency with profile
Signed-off-by: coleramos425 <colramos@amd.com>


[ROCm/rocprofiler-compute commit: 74f816ed47]
2023-08-16 13:48:31 -05:00
coleramos425 74f816ed47 Add a -d option to analyze mode for consistency with profile
Signed-off-by: coleramos425 <colramos@amd.com>
2023-08-16 13:48:31 -05:00
Juan Manuel MARTINEZ CAAMAÑO 10e2958197 SWDEV-410182: Do not force wave64 for geekbench
Change-Id: I505af6b0ca70e54a963abe9b097b91eea94d3268


[ROCm/clr commit: d6dc82b220]
2023-08-16 10:22:52 -04:00
Juan Manuel MARTINEZ CAAMAÑO d6dc82b220 SWDEV-410182: Do not force wave64 for geekbench
Change-Id: I505af6b0ca70e54a963abe9b097b91eea94d3268
2023-08-16 10:22:52 -04:00
David Yat Sin a20a0a5bac Temporarily disable SDMA ganging by default
SDMA ganging is causing some regressions with some applications hanging.
Temporarily disabling SDMA ganging by default until issue is fixed.

Change-Id: I65e172923a53a967df27b30d969ad5d215c4fa09
2023-08-15 23:17:34 +00:00
David Yat Sin 38bec00960 Temporarily disable SDMA ganging by default
SDMA ganging is causing some regressions with some applications hanging.
Temporarily disabling SDMA ganging by default until issue is fixed.

Change-Id: I65e172923a53a967df27b30d969ad5d215c4fa09


[ROCm/ROCR-Runtime commit: a20a0a5bac]
2023-08-15 23:17:34 +00:00
Jenkins 8136e20d1c SWDEV-299127 - Merge 'develop' into 'amd-staging'
Change-Id: Id1921f5867ee9f2cb98748a192aa85b7e0fd5ca8


[ROCm/hip commit: 72d3f7e7a4]
2023-08-15 23:09:54 +00:00
Jenkins 72d3f7e7a4 SWDEV-299127 - Merge 'develop' into 'amd-staging'
Change-Id: Id1921f5867ee9f2cb98748a192aa85b7e0fd5ca8
2023-08-15 23:09:54 +00:00
Cole Ramos 7e0fb0fc16 Merge pull request #160 from JoseSantosAMD/pct_of_peak
Pct of peak

[ROCm/rocprofiler-compute commit: abd424beea]
2023-08-15 13:56:38 -07:00
Cole Ramos abd424beea Merge pull request #160 from JoseSantosAMD/pct_of_peak
Pct of peak
2023-08-15 13:56:38 -07:00
Cole Ramos 384bd90214 Merge pull request #158 from JoseSantosAMD/instr_mix_min_max_avg
Instr mix min max avg

[ROCm/rocprofiler-compute commit: 6f6bca1383]
2023-08-15 13:55:22 -07:00
Cole Ramos 6f6bca1383 Merge pull request #158 from JoseSantosAMD/instr_mix_min_max_avg
Instr mix min max avg
2023-08-15 13:55:22 -07:00
Giovanni LB 157eacd2bb DCGPUBU-44: Added arbitrary envvars to file/dir names. Squashed some fixes.
- Added arbitrary env-vars
- Fixed some UBs with atomic variables
- Fixed loading of file plugin
- ATT vs. kernel-trace off by one

Change-Id: I69c75f66f722e4085b5279f41afd05813628846d
2023-08-15 17:53:57 -03:00
Giovanni LB c869df7272 DCGPUBU-44: Added arbitrary envvars to file/dir names. Squashed some fixes.
- Added arbitrary env-vars
- Fixed some UBs with atomic variables
- Fixed loading of file plugin
- ATT vs. kernel-trace off by one

Change-Id: I69c75f66f722e4085b5279f41afd05813628846d


[ROCm/rocprofiler commit: 157eacd2bb]
2023-08-15 17:53:57 -03:00
JoseSantosAMD 32492ec598 add simple_bar styling
Signed-off-by: JoseSantosAMD <Jose.Santos@amd.com>


[ROCm/rocprofiler-compute commit: 6d39955708]
2023-08-15 15:14:20 -05:00
JoseSantosAMD 6d39955708 add simple_bar styling
Signed-off-by: JoseSantosAMD <Jose.Santos@amd.com>
2023-08-15 15:14:20 -05:00
Cole Ramos e703d1a71b Merge pull request #157 from JoseSantosAMD/kernel_name_mappings
Kernel name demangle

[ROCm/rocprofiler-compute commit: b79b4d3512]
2023-08-15 13:07:22 -07:00
Cole Ramos b79b4d3512 Merge pull request #157 from JoseSantosAMD/kernel_name_mappings
Kernel name demangle
2023-08-15 13:07:22 -07:00