Revīziju grafs

60 Revīzijas

Autors SHA1 Ziņojums Datums
Sean Keely 960fcd7a41 Remove legacy build files.
Change-Id: I540846b46714b6b07039e8b84929588f156652e0


[ROCm/ROCR-Runtime commit: a5e492253e]
2020-06-19 22:33:36 -04:00
Ramesh Errabolu ab6b820dbf Update License header and Cleanup IP references
Change-Id: I0a6636e1d8457045d034d05383cfb5d4e7680fee


[ROCm/ROCR-Runtime commit: 42b38daa22]
2020-06-19 22:33:36 -04:00
Ramesh Errabolu c9f453a8a8 Build ROCr core and image libraries as one shared object
Change-Id: I3a16c1227e7db2e386ab33886965596fa0fb0c87


[ROCm/ROCR-Runtime commit: 0ca0691ca7]
2020-06-19 22:33:36 -04:00
Sean Keely 890e666011 Select a non-empty pool for image kernarg.
Zero size pools have no numa bindings.  Selecting a pool with numa
bindings should prevent thrashing due to numa balancing daemon.

Change-Id: Ib0082cb9af66e24e07a2adbb83c1045145d51403


[ROCm/ROCR-Runtime commit: 32bb10086d]
2020-05-28 23:34:42 -04:00
Sean Keely 699a60f2db Remove gfx1030 references.
Options seem to be gfx1030 and gfx1030.

Change-Id: I3e33e3a950d1b950e10163612b892d9c382628fa


[ROCm/ROCR-Runtime commit: 2ffc9ecbb3]
2020-05-15 16:53:11 -05:00
Sean Keely 31268c6f4d Allow linear swizzle mode with tiled image requests.
Mesa address lib faults if the only acceptable swizzle modes are
forbidden.  The old address lib simply ignored the forbidden list
in this case.  Mesa addrlib will not select linear unless there
is no other option so allowing linear mode for tiled images will
still use tiled modes when possible.

Change-Id: I1aa44d072db902c968484dbff67b482af03b45d9


[ROCm/ROCR-Runtime commit: c60364e1e0]
2020-05-11 23:34:44 -04:00
Matt Arsenault b5f6574895 Do not use llvm-dis to pick the triple
There are a couple problems with this. First, llvm-dis is an unstable
llvm development tool and 3rd party users should generally not rely on
it. The text format is unstable, and the regex here isn't even
explicitly looking for the target triple field, so it could
accidentally find something else. Second, picking the target to
compile based on the library you are linking is a fundamentally
backwards decision. The target you're compiling for changes the
library you would want to link. The device libraries are only ever
compiled with amdgcn-amd-amdhsa. If we had a second triple, this
should be explicitly building for any it cares about.

Change-Id: I3bae8398f60f78df61ab2177aa9e83f47ec6dea4


[ROCm/ROCR-Runtime commit: 96d4140609]
2020-05-06 13:28:39 -04:00
Sean Keely 3b0c3e83a3 Update addrlib with latest Mesa source.
Change-Id: Idd8cdaac9ad370397d62f6a32687ca7bc7d7462b


[ROCm/ROCR-Runtime commit: 3da81968cb]
2020-05-01 20:33:09 -04:00
Sean Keely 2269234579 Remove dead code from image_manager_xx.cpp
Image swizzle mode will be set by the preferred surface info
function.

Change-Id: I41e639be53cafbb4db6cf15c159aa2bd457ec5be


[ROCm/ROCR-Runtime commit: 1440da3e15]
2020-05-01 20:32:45 -04:00
Sean Keely 1fc7f2dec7 Move Images code to hsa-runtime folder
Change-Id: I53c1845d985ac3e9708d952865009c0021f3bb4f


[ROCm/ROCR-Runtime commit: 7e3db20826]
2020-04-30 19:35:57 -05:00