AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • 9c7ee12822 Signal Fix: The signals in a stream are re-used 1. Before, the signal pool is increased depending on the usage 2. After, a static number of signals are allocated to the pool Only these are used by hip in a stream 3. If the signals required are more than the pool size, the stream has to wait to make sure all the signals are available 4. Once they are available, the stream can use them 5. Removed HIP_NUM_SIGNALS_PER_STREAM because of redundancy with HIP_STREAM_SIGNALS 6. Increased signal count from 2 to 32. Future Work: Dynamically increase the pool size depending on the number of streams allocated by the application. And, null stream should have more signals Aditya Atluri 2016-07-28 23:01:35 -05:00
  • 2f53a50718 Signal Fix: The signals in a stream are re-used 1. Before, the signal pool is increased depending on the usage 2. After, a static number of signals are allocated to the pool Only these are used by hip in a stream 3. If the signals required are more than the pool size, the stream has to wait to make sure all the signals are available 4. Once they are available, the stream can use them 5. Removed HIP_NUM_SIGNALS_PER_STREAM because of redundancy with HIP_STREAM_SIGNALS 6. Increased signal count from 2 to 32. Future Work: Dynamically increase the pool size depending on the number of streams allocated by the application. And, null stream should have more signals Aditya Atluri 2016-07-28 23:01:35 -05:00
  • 9062ebcf3a Signal Fix: The signals in a stream are re-used 1. Before, the signal pool is increased depending on the usage 2. After, a static number of signals are allocated to the pool Only these are used by hip in a stream 3. If the signals required are more than the pool size, the stream has to wait to make sure all the signals are available 4. Once they are available, the stream can use them 5. Removed HIP_NUM_SIGNALS_PER_STREAM because of redundancy with HIP_STREAM_SIGNALS 6. Increased signal count from 2 to 32. Future Work: Dynamically increase the pool size depending on the number of streams allocated by the application. And, null stream should have more signals Aditya Atluri 2016-07-28 23:01:35 -05:00
  • 381f67e067 Remove dead enqueueBarrier function. Ben Sander 2016-07-28 22:48:22 -05:00
  • f7ab82cb39 Remove dead enqueueBarrier function. Ben Sander 2016-07-28 22:48:22 -05:00
  • 3044711042 Remove dead enqueueBarrier function. Ben Sander 2016-07-28 22:48:22 -05:00
  • 666c227c7d Remove dead enqueueBarrier function. Ben Sander 2016-07-28 22:48:22 -05:00
  • 8bb14fb26f P4 to Git Change 1296696 by gandryey@gera-w8 on 2016/07/28 16:20:00 foreman 2016-07-28 16:28:15 -04:00
  • fa73da1714 P4 to Git Change 1296696 by gandryey@gera-w8 on 2016/07/28 16:20:00 foreman 2016-07-28 16:28:15 -04:00
  • a30f29af72 P4 to Git Change 1296360 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/28 03:00:07 foreman 2016-07-28 03:09:56 -04:00
  • 4bef3bbb4a P4 to Git Change 1296360 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/28 03:00:07 foreman 2016-07-28 03:09:56 -04:00
  • 337e8b1313 Remove unneeded deb build script Sylvain Jeaugey 2016-07-27 17:58:00 -07:00
  • b3a9e1333d Remove unneeded deb build script Sylvain Jeaugey 2016-07-27 17:58:00 -07:00
  • a853d10446 Cleanup sync code. Ben Sander 2016-07-27 18:31:11 -05:00
  • ef61aae878 Cleanup sync code. Ben Sander 2016-07-27 18:31:11 -05:00
  • 91368967bb Cleanup sync code. Ben Sander 2016-07-27 18:31:11 -05:00
  • 02dd7a7399 Cleanup sync code. Ben Sander 2016-07-27 18:31:11 -05:00
  • fd3b0ef0e5 Shorten the device list in PerfCounter Amber Lin 2016-07-27 16:38:04 -04:00
  • 6c4d19a9d2 Shorten the device list in PerfCounter Amber Lin 2016-07-27 16:38:04 -04:00
  • 058e927c9b Fix API string message for hipDeviceGetAttribute Ben Sander 2016-07-27 16:18:14 -05:00
  • f5118ce3cd Fix API string message for hipDeviceGetAttribute Ben Sander 2016-07-27 16:18:14 -05:00
  • 07c9fb8b38 Fix API string message for hipDeviceGetAttribute Ben Sander 2016-07-27 16:18:14 -05:00
  • 4cf0f86e99 Fix API string message for hipDeviceGetAttribute Ben Sander 2016-07-27 16:18:14 -05:00
  • b81d34bcdf Refactor Trap Handler Code Ramesh Errabolu 2016-07-26 18:00:36 -05:00
  • 570301ffd0 Refactor Trap Handler Code Ramesh Errabolu 2016-07-26 18:00:36 -05:00
  • 96fe085832 Signal Fix: Added signal limit to allocSignal 1. Did not change the logic in allocSignal 2. Added guard to wait on signal limit Aditya Atluri 2016-07-27 13:48:49 -05:00
  • 1b2a24d0b8 Signal Fix: Added signal limit to allocSignal 1. Did not change the logic in allocSignal 2. Added guard to wait on signal limit Aditya Atluri 2016-07-27 13:48:49 -05:00
  • c2b4a3936c Signal Fix: Added signal limit to allocSignal 1. Did not change the logic in allocSignal 2. Added guard to wait on signal limit Aditya Atluri 2016-07-27 13:48:49 -05:00
  • 1859c6e515 Signal Fix: Added signal limit to allocSignal 1. Did not change the logic in allocSignal 2. Added guard to wait on signal limit Aditya Atluri 2016-07-27 13:48:49 -05:00
  • 489d9ed801 P4 to Git Change 1295979 by gandryey@gera-w8 on 2016/07/27 11:00:55 foreman 2016-07-27 12:10:55 -04:00
  • 9dd15fc110 P4 to Git Change 1295979 by gandryey@gera-w8 on 2016/07/27 11:00:55 foreman 2016-07-27 12:10:55 -04:00
  • 5fa1029861 Update release notes for 0.92.00 release Maneesh Gupta 2016-07-27 20:30:04 +05:30
  • de7c9769a4 Update release notes for 0.92.00 release Maneesh Gupta 2016-07-27 20:30:04 +05:30
  • 1acf8919ee Update release notes for 0.92.00 release Maneesh Gupta 2016-07-27 20:30:04 +05:30
  • 059a8d51ba Update release notes for 0.92.00 release Maneesh Gupta 2016-07-27 20:30:04 +05:30
  • 622ce5acc5 P4 to Git Change 1295865 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/27 03:00:09 foreman 2016-07-27 03:22:05 -04:00
  • 934647d3c0 P4 to Git Change 1295865 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/27 03:00:09 foreman 2016-07-27 03:22:05 -04:00
  • 664e843fd9 P4 to Git Change 1295719 by cpaquot@hog-ocl on 2016/07/26 18:50:05 foreman 2016-07-26 19:03:01 -04:00
  • e911ffa47e P4 to Git Change 1295719 by cpaquot@hog-ocl on 2016/07/26 18:50:05 foreman 2016-07-26 19:03:01 -04:00
  • b99410bd07 Signal Fix: Moved kernel count to critical stream 1. Added environment variable HIP_NUM_KERNELS_INFLIGHT 2. Moved kernelcount variable inside stream critical section Aditya Atluri 2016-07-26 17:09:27 -05:00
  • 7be196de48 Signal Fix: Moved kernel count to critical stream 1. Added environment variable HIP_NUM_KERNELS_INFLIGHT 2. Moved kernelcount variable inside stream critical section Aditya Atluri 2016-07-26 17:09:27 -05:00
  • 6ff74b7780 Signal Fix: Moved kernel count to critical stream 1. Added environment variable HIP_NUM_KERNELS_INFLIGHT 2. Moved kernelcount variable inside stream critical section Aditya Atluri 2016-07-26 17:09:27 -05:00
  • 0a31b47e2e Signal Fix: Moved kernel count to critical stream 1. Added environment variable HIP_NUM_KERNELS_INFLIGHT 2. Moved kernelcount variable inside stream critical section Aditya Atluri 2016-07-26 17:09:27 -05:00
  • d8ea207f61 Signal Fix: Changed global signal count to per stream signal count 1. The number of kernels that can use signals are increased to 128 2. The kernel count is now specific to the stream Aditya Atluri 2016-07-26 14:03:51 -05:00
  • 2e754d27dc Signal Fix: Changed global signal count to per stream signal count 1. The number of kernels that can use signals are increased to 128 2. The kernel count is now specific to the stream Aditya Atluri 2016-07-26 14:03:51 -05:00
  • eccab29360 Signal Fix: Changed global signal count to per stream signal count 1. The number of kernels that can use signals are increased to 128 2. The kernel count is now specific to the stream Aditya Atluri 2016-07-26 14:03:51 -05:00
  • 53d7629a85 Signal Fix: Changed global signal count to per stream signal count 1. The number of kernels that can use signals are increased to 128 2. The kernel count is now specific to the stream Aditya Atluri 2016-07-26 14:03:51 -05:00
  • bb363f9d63 removed redundant signal destroy Aditya Atluri 2016-07-26 13:35:35 -05:00
  • 524127b4a4 removed redundant signal destroy Aditya Atluri 2016-07-26 13:35:35 -05:00
  • 06abacce98 removed redundant signal destroy Aditya Atluri 2016-07-26 13:35:35 -05:00
  • fa7933eb91 removed redundant signal destroy Aditya Atluri 2016-07-26 13:35:35 -05:00
  • 0ed0ad30da Doc update for FAQ and future RELEASE notes Ben Sander 2016-07-26 11:51:54 -05:00
  • 77be00e8d3 Doc update for FAQ and future RELEASE notes Ben Sander 2016-07-26 11:51:54 -05:00
  • 3027a1f269 Doc update for FAQ and future RELEASE notes Ben Sander 2016-07-26 11:51:54 -05:00
  • 99bc0f03e1 Doc update for FAQ and future RELEASE notes Ben Sander 2016-07-26 11:51:54 -05:00
  • 542948ad75 Doc update for README.md - add more intro text, example Ben Sander 2016-07-25 17:45:31 -05:00
  • 75d1f8c297 Doc update for README.md - add more intro text, example Ben Sander 2016-07-25 17:45:31 -05:00
  • 46cb4cdc7b Doc update for README.md - add more intro text, example Ben Sander 2016-07-25 17:45:31 -05:00
  • 6ea9d2eace Doc update for README.md - add more intro text, example Ben Sander 2016-07-25 17:45:31 -05:00
  • 4094170fd7 Make HCC ignore register keyword Ben Sander 2016-07-25 10:15:02 -05:00
  • b9e48d6066 Make HCC ignore register keyword Ben Sander 2016-07-25 10:15:02 -05:00
  • b909cbb2a3 Make HCC ignore register keyword Ben Sander 2016-07-25 10:15:02 -05:00
  • 0a74e00ac8 Make HCC ignore register keyword Ben Sander 2016-07-25 10:15:02 -05:00
  • e207805f81 Added re-fix for memcpy kernel sync 1. The patch uses HIP signal pools to sync between copy and kernel commands 2. The hsa_signal_create is removed 3. Left the redundant enqueueBarrier method just in case Aditya Atluri 2016-07-26 09:22:59 -05:00
  • 0232e6bbb4 Added re-fix for memcpy kernel sync 1. The patch uses HIP signal pools to sync between copy and kernel commands 2. The hsa_signal_create is removed 3. Left the redundant enqueueBarrier method just in case Aditya Atluri 2016-07-26 09:22:59 -05:00
  • 69aed2c6ea Added re-fix for memcpy kernel sync 1. The patch uses HIP signal pools to sync between copy and kernel commands 2. The hsa_signal_create is removed 3. Left the redundant enqueueBarrier method just in case Aditya Atluri 2016-07-26 09:22:59 -05:00
  • 4bdf26a82e Added re-fix for memcpy kernel sync 1. The patch uses HIP signal pools to sync between copy and kernel commands 2. The hsa_signal_create is removed 3. Left the redundant enqueueBarrier method just in case Aditya Atluri 2016-07-26 09:22:59 -05:00
  • 582482aa65 P4 to Git Change 1295296 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/26 03:00:13 foreman 2016-07-26 03:09:08 -04:00
  • 0b1059a98d P4 to Git Change 1295296 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/26 03:00:13 foreman 2016-07-26 03:09:08 -04:00
  • 5d954a49e2 Merge branch 'amd-develop' into amd-master Maneesh Gupta 2016-07-26 11:41:34 +05:30
  • 9909773247 Merge branch 'amd-develop' into amd-master Maneesh Gupta 2016-07-26 11:41:34 +05:30
  • d20574ad1f Merge branch 'amd-develop' into amd-master Maneesh Gupta 2016-07-26 11:41:34 +05:30
  • 75f804fbf7 Merge branch 'amd-develop' into amd-master Maneesh Gupta 2016-07-26 11:41:34 +05:30
  • 088aa4a518 P4 to Git Change 1295087 by asalmanp@asalmanp-opencl-stg on 2016/07/25 16:08:34 foreman 2016-07-25 16:20:18 -04:00
  • 8f5943e292 P4 to Git Change 1295087 by asalmanp@asalmanp-opencl-stg on 2016/07/25 16:08:34 foreman 2016-07-25 16:20:18 -04:00
  • 884cf38d01 P4 to Git Change 1295077 by gandryey@gera-w8 on 2016/07/25 15:54:39 foreman 2016-07-25 16:08:09 -04:00
  • 594dcef8a4 P4 to Git Change 1295077 by gandryey@gera-w8 on 2016/07/25 15:54:39 foreman 2016-07-25 16:08:09 -04:00
  • 43d2570611 Merge remote-tracking branch 'github/master' into public Sylvain Jeaugey 2016-07-25 10:53:01 -07:00
  • 428ec5b2a3 Merge remote-tracking branch 'github/master' into public Sylvain Jeaugey 2016-07-25 10:53:01 -07:00
  • 159408c5d5 Fixed redundant contexts in multi-process apps Nathan Luehr 2016-07-22 17:29:13 -07:00
  • 55c42ad681 Fixed redundant contexts in multi-process apps Nathan Luehr 2016-07-22 17:29:13 -07:00
  • cc9a901498 Documentation updates Maneesh Gupta 2016-07-25 14:53:15 +05:30
  • b29ed98f9a Documentation updates Maneesh Gupta 2016-07-25 14:53:15 +05:30
  • adfaa47840 Documentation updates Maneesh Gupta 2016-07-25 14:53:15 +05:30
  • 90a3e4e1da Documentation updates Maneesh Gupta 2016-07-25 14:53:15 +05:30
  • 7a70966cb1 D2H and H2D unpinned memory transfer support Rahul Garg 2016-07-09 19:29:55 +05:30
  • d11d65d401 D2H and H2D unpinned memory transfer support Rahul Garg 2016-07-09 19:29:55 +05:30
  • d506d58919 D2H and H2D unpinned memory transfer support Rahul Garg 2016-07-09 19:29:55 +05:30
  • 42a3ed544c D2H and H2D unpinned memory transfer support Rahul Garg 2016-07-09 19:29:55 +05:30
  • 32fbc0aa47 P4 to Git Change 1294784 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/25 03:00:08 foreman 2016-07-25 03:10:14 -04:00
  • cc7cddf745 P4 to Git Change 1294784 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/25 03:00:08 foreman 2016-07-25 03:10:14 -04:00
  • 13f7baa74b P4 to Git Change 1294684 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/24 03:00:09 foreman 2016-07-24 03:13:25 -04:00
  • 11f04dd081 P4 to Git Change 1294684 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/24 03:00:09 foreman 2016-07-24 03:13:25 -04:00
  • 776ea7a1f0 Partial fix async after kernel launch signal issue Aditya Atluri 2016-07-23 14:54:20 -05:00
  • 1704006bed Partial fix async after kernel launch signal issue Aditya Atluri 2016-07-23 14:54:20 -05:00
  • 6f95ad0a8d Partial fix async after kernel launch signal issue Aditya Atluri 2016-07-23 14:54:20 -05:00
  • c756bb3398 Partial fix async after kernel launch signal issue Aditya Atluri 2016-07-23 14:54:20 -05:00
  • d22df116fa P4 to Git Change 1294664 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/23 03:00:09 foreman 2016-07-23 03:13:11 -04:00
  • 94a62b0da4 P4 to Git Change 1294664 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/07/23 03:00:09 foreman 2016-07-23 03:13:11 -04:00
  • 2c945dba23 Add empty stubs for threadfence family routines, changes include: - stubs and documentation in include/hcc_details/hip_runtime.h - stubs with "no-op" in src/hip_memory.cpp - document update in hip_kernel_language.md, add suggestions to disable L1 and L2 caches when using the threadfence routines. pensun 2016-07-22 10:40:58 -05:00