AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
13,501 Commits 3 Branches 0 Tags
e56ff3716d7b8d04ee60e42d2ebb6ebf5d0d58db
Commit Graph

13501 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
James Zhu 9509af4b98 libhsakmt: increase default svm.alignment_order 
Since GFX950 can support page table fragment up to 18 without
performance loss. So set GFX950  default svm.alignment_order to 18.

Change-Id: Ibcdb7f041fb07a38e924c471beec261ea227ca1d
Signed-off-by: James Zhu <James.Zhu@amd.com>
 2025-01-28 08:27:19 -05:00
Amber Lin 0b6e457201 kfdtest: Create gfx950 blacklist 
This patch creates the blacklist for gfx950 by copying gfx942 but adding
KFDGWSTest.Semaphore as GWS support is completely removed from gfx950.

Change-Id: I5d7c17e57b8cfd9fae63780ecc9dd55662cfdade
Signed-off-by: Amber Lin <Amber.Lin@amd.com>
 2025-01-28 08:26:44 -05:00
Yiannis Papadopoulos 26bfa0b8f6 rocr/aie: Add dma-buf import support for AIEAgents via the Driver interface 
Change-Id: I70f8d8772dda7c06944d75042cb3034ddd89aff4
 2025-01-27 15:22:46 -05:00
Indic, Vladimir 9f7772d90b Temporarily allow only host-trap sampling (#156) 
[ROCm/rocprofiler-sdk commit: e4d736839d]
 2025-01-27 13:26:11 -06:00
Indic, Vladimir e4d736839d Temporarily allow only host-trap sampling (#156) 2025-01-27 13:26:11 -06:00
Lancelot Six 76052ba028 libhsakmt: gfx950 uses same VGPR block size as gfx940 
Make sure to use allocate the same amount of size for VGPR data in
gfx950 as it is done for gfx940.

Change-Id: I6a0820996389627ccbdfef856e5150c46fac92a1
Signed-off-by: Lancelot SIX <lancelot.six@amd.com>
 2025-01-27 14:06:42 -05:00
Trowbridge, Ian 2fc0e4ecd4 Support for rocJPEG Tracing (#5) 
* Add initial support for rocJPEG API Trace

* Added tests for rocjpeg

* Formatting

[ROCm/rocprofiler-register commit: 366716af06]
 2025-01-27 11:50:12 -06:00
Trowbridge, Ian 366716af06 Support for rocJPEG Tracing (#5) 
* Add initial support for rocJPEG API Trace

* Added tests for rocjpeg

* Formatting
 2025-01-27 11:50:12 -06:00
Shweta Khatri 6361466baa rocr: Use view3dAs2dArray flag, for thick/3D swizzle modes. 
Added HSA_IMAGE_ENABLE_3D_SWIZZLE_DEBUG environment flag to
enable/disable this. Default value is false (view3dAs2dArray = 1)
Enabling this flag will enable support for swizzles that do 3D
interleaving. Note that all features of 3D images are supported
with 2D swizzles,it's just that the access patterns are different
and therefore cache hit-rates may be better or worse, depending
on how it's used. Volumetric algorithms do better with 3D and apps
that tend to access a single slice at a time do better with 2D.

Change-Id: Id8574a6710fe4333a1ee331e5ce9195a81434198
 2025-01-27 09:28:33 -05:00
Tony Gutierrez 8a38f121ea rocr: Add WaitMultiple to core Signal 
Replaces WaitAny with WaitMultiple to more closely align with the
underlying driver API for waiting on multiple events.

WaitMultiple adds a single parameter, wait_on_all, to the WaitAny
interface providing a single function for waiting on multiple
events when we only need AND and OR semantics for the signal
checking logic.

Change-Id: I68a4a45d48151d9d69aef02fd8f7263b9e6c0e75
 2025-01-27 09:21:43 -05:00
Lancelot Six c51aa0d155 libhsakmt: Use the node info to determine LDS size 
The CWSR area size needs to take into account the size of LDS each
active workgroup can have.  The current implementation uses a constant
for that.  This patch refactors this to use the HsaNodeProperties of the
device's the CWSR area is for to figure out the size of LDS.

Change-Id: Ib8585b2b7140ec5c99e7b7d62e67f785697c028a
Signed-off-by: Lancelot Six <Lancelot.Six@amd.com>
Signed-off-by: Amber Lin <Amber.Lin@amd.com>
 2025-01-26 21:46:32 -05:00
Alex Sierra 268054cd28 kfdtest: add support for gfx9.5.0 in shader store 
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Change-Id: I48b98ff631bd1aa1a044b60583ff256e43b17423
 2025-01-26 21:45:07 -05:00
Alex Sierra e94ff8a36c kfdtest: Add gfx 9.5 as FAMILY_AV 
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Change-Id: Ib5696eee1d4f64c9c87d714eae7c80fbbd1e2b23
 2025-01-26 21:43:55 -05:00
David Yat Sin dab8f2fc65 rocr: Add support for gfx950 
<squashed with patch for gfx950 generic targets>

Signed-off-by: Chris Freehill <Chris.Freehill@amd.com>

Change-Id: Ifec6d93cf46c7fbf736c6572882299e279260af6
 2025-01-26 13:04:58 -05:00
Sam Wu c95cfc9bd8 Update documentation requirements (#540) 
[ROCm/rocprofiler-compute commit: f58070aff1]
 2025-01-24 16:34:03 -07:00
Sam Wu f58070aff1 Update documentation requirements (#540) 2025-01-24 16:34:03 -07:00
Galantsev, Dmitrii 99d4d77e20 CMAKE - Move rdc_options into share/rdc/conf/ 
Change-Id: Ib2e792aef180f0f267d86d68c57b852b2cdc8ea6
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
 2025-01-24 12:06:05 -06:00
Ben Vanik 7d64fe49fa rocr: Fix HostQueue to obey the alignment requirement 
Change-Id: I06542e9ff94e826ca0abba0328b301fec50a95ea
 2025-01-24 12:08:11 -05:00
Pryor, Adam 6f358ddc9e SWDEV-508477 Eval Flops Percent (#85) 
SWDEV-508477 - Profiler add FP*_PERCENT

Change-Id: Idb6250fe6b7ba3df6fe7d30861e0fbbda7e9bdce

Signed-off-by: adapryor <Adam.pryor@amd.com>
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
 2025-01-24 10:07:32 -06:00
David Yat Sin 7ea25ebb85 rocr: Add thread priority for AsyncEventHandler 
Set priority to maximum for signal event handler and minimum for
exceptions event handler.

Change-Id: I1b982d3c2e4c880fafc073fe1a542d01692a6fdc
 2025-01-24 10:08:12 -05:00
Elwazir, Ammar b2c71e43cb Fixing collection period rocprofv3 help message (#148) 
Update rocprofv3.py

[ROCm/rocprofiler-sdk commit: 19a912d476]
 2025-01-24 08:39:40 -06:00
Elwazir, Ammar 19a912d476 Fixing collection period rocprofv3 help message (#148) 
Update rocprofv3.py
 2025-01-24 08:39:40 -06:00
Galantsev, Dmitrii e033fd4c55 CMAKE - Rename SMI_*_DIR into AMD_SMI_*_DIR 
Change-Id: I3b8b852e6b68f1448c8ed5d5e6ea4579c470ff53
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
 2025-01-23 20:56:00 -06:00
Ma, Li 9dce427c69 Fix Memory Current Bandwidth (#98) 
Adjust the calculation order to ensure accuracy.


Change-Id: Ica10769fa3dba10c67428d09ffd454fc09ed0da8

Signed-off-by: Li Ma <li.ma@amd.com>
 2025-01-24 10:22:08 +08:00
stali e36d3fae22 fix topology issue 2025-01-24 09:22:42 +08:00
Mallya, Ameya Keshava e5cd8ffa48 Added PSDB trigger for staging 
[ROCm/roctracer commit: 3e1caf69ae]
 2025-01-23 16:54:09 -08:00
Mallya, Ameya Keshava 3e1caf69ae Added PSDB trigger for staging 2025-01-23 16:54:09 -08:00
cfallows-amd a64129b34f Add ubuntu 24.04 detection for roofline bin loading. (#541) 
Also updated docker script for dev usage.

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: 7602341b70]
 2025-01-23 19:53:50 -05:00
cfallows-amd 7602341b70 Add ubuntu 24.04 detection for roofline bin loading. (#541) 
Also updated docker script for dev usage.

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>
 2025-01-23 19:53:50 -05:00
Mallya, Ameya Keshava ae8f4b32a7 Added PSDB trigger for staging 2025-01-23 16:53:25 -08:00
Mallya, Ameya Keshava a3bf6879da Added PSDB trigger for staging 
[ROCm/rocprofiler commit: ae8f4b32a7]
 2025-01-23 16:53:25 -08:00
Galantsev, Dmitrii ef77c0ed92 Fix workflow for rocprof by specifying GPU_TARGETS 
Change-Id: I153f9e73471599fbcf68c73ad0ed9f4db7a742ef
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
 2025-01-23 18:38:34 -06:00
Mallya, Ameya Keshava a0e9b3716a Added PSDB trigger for npi 
[ROCm/roctracer commit: 3ac827eb87]
 2025-01-23 16:13:32 -08:00
Mallya, Ameya Keshava 3ac827eb87 Added PSDB trigger for npi 2025-01-23 16:13:32 -08:00
Mallya, Ameya Keshava 11cd25d1f5 Added PSDB trigger for npi 2025-01-23 16:12:59 -08:00
Mallya, Ameya Keshava 1f6b476eff Added PSDB trigger for npi 
[ROCm/rocprofiler commit: 11cd25d1f5]
 2025-01-23 16:12:59 -08:00
ywang103-amd 8079e2e10b enable rocprofv3 for all old soc architectures (#539) 
* enable rocprofv3 for all old soc archetectures

* remove gfx906 from rocprofv3

[ROCm/rocprofiler-compute commit: 713570908a]
 2025-01-23 15:37:00 -05:00
ywang103-amd 713570908a enable rocprofv3 for all old soc architectures (#539) 
* enable rocprofv3 for all old soc archetectures

* remove gfx906 from rocprofv3
 2025-01-23 15:37:00 -05:00
Ben Vanik 9971e7b004 rocr: Fixing non-portable inline attribute on hsa_flag_* utilities. 
Change-Id: Ie1c53fef407a71b5ec4c6eaf3a3ed00871184408
 2025-01-23 15:09:21 -05:00
Tony Gutierrez 15107afb11 rocr: Generalize driver discovery 
Generalize the driver discovery and move driver-specific
functionality to the concrete driver implementations.
Currently, this process is tightly coupled to the hsakmt
which is GPU and OS specific.

Change-Id: Ie1c53fef407a71b5ec4c6eaf3a3ed00871184409
 2025-01-23 15:09:14 -05:00
Tony Gutierrez 77fa5af618 rocr: Make Open() and Close() virtual in Driver 
Change-Id: Iac054c08383b080ca2b2ec6d65019bf2f083b763
 2025-01-23 15:09:06 -05:00
Tony Gutierrez 8bbc44d51b rocr: Forward declare Driver in the Agent class 
Change-Id: Ib27081bf31446af92602f723f352fb75ec3f378e
 2025-01-23 15:08:59 -05:00
Cole Ramos c5c44a9cba Yearly license update (#538) 
* Add license template designation

Signed-off-by: coleramos425 <colramos@amd.com>

* Update license year on *.py

Signed-off-by: coleramos425 <colramos@amd.com>

* Update license year on *.sh

Signed-off-by: coleramos425 <colramos@amd.com>

* Update license year on misc files

Signed-off-by: coleramos425 <colramos@amd.com>

---------

Signed-off-by: coleramos425 <colramos@amd.com>

[ROCm/rocprofiler-compute commit: 34760ff814]
 2025-01-23 13:09:32 -06:00
Cole Ramos 34760ff814 Yearly license update (#538) 
* Add license template designation

Signed-off-by: coleramos425 <colramos@amd.com>

* Update license year on *.py

Signed-off-by: coleramos425 <colramos@amd.com>

* Update license year on *.sh

Signed-off-by: coleramos425 <colramos@amd.com>

* Update license year on misc files

Signed-off-by: coleramos425 <colramos@amd.com>

---------

Signed-off-by: coleramos425 <colramos@amd.com>
 2025-01-23 13:09:32 -06:00
Shweta Khatri 2d4a578020 Revert "Revert "hsakmt: Only set exec flag when requested"" 
This reverts commit 80da7d5ee4.

Reason for revert: This will put back the change ID - Id1154f08f6ba21c633905fd46b06053994d6f3cc to ROCR repo, which will prevent memory allocations from being automatically granted the 'executable' flag, addressing previously -  incorrect and unsafe behavior in ROCm driver.

Change-Id: I3d45c45859929a80f7791681b411251e099a1901
 2025-01-23 09:08:25 -05:00
Bhardwaj, Gopesh 7c6afe75dd Fixing corrupted CHANGELOG (#152) 
* Fixing corrupted CHANGELOG

* updating changelog

[ROCm/rocprofiler-sdk commit: 2f65f99d2c]
 2025-01-23 00:12:12 -06:00
Bhardwaj, Gopesh 2f65f99d2c Fixing corrupted CHANGELOG (#152) 
* Fixing corrupted CHANGELOG

* updating changelog
 2025-01-23 00:12:12 -06:00
Madsen, Jonathan 3de44de3b5 Upload artifacts only on 'workflow_dispatch' (#147) 
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>

[ROCm/rocprofiler-sdk commit: 34145fa762]
 2025-01-22 19:44:19 -06:00
Madsen, Jonathan 34145fa762 Upload artifacts only on 'workflow_dispatch' (#147) 
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
 2025-01-22 19:44:19 -06:00
Rawat, Swati edb51fc861 update copyright date to 2025 (#102) 
* Update LICENSE

* Update conf.py

* Update copyright year

* [fix] Update copyright year

* Update copyright year "ROCm Developer Tools"

* Add license headers to c++ files

* Add license to *.py

* Update licenses in rocdecode sources

---------

Co-authored-by: srawat <120587655+SwRaw@users.noreply.github.com>
Co-authored-by: Mythreya <mythreya.kuricheti@amd.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>

[ROCm/rocprofiler-sdk commit: 97b7a6315d]
 2025-01-22 19:11:20 -06:00