diff options
| author | Joshua Bakita <jbakita@cs.unc.edu> | 2024-02-14 15:36:25 -0500 |
|---|---|---|
| committer | Joshua Bakita <jbakita@cs.unc.edu> | 2024-02-14 15:36:25 -0500 |
| commit | b5281f5fc01fc925898c9323edab41b817df8661 (patch) | |
| tree | 6a14c270c7a3bb061bd167d87430d3dd9f3696ec /libsmctrl_test_stream_mask.c | |
| parent | 973b919cfe6d05fdb3b82f538b1afbc3233a7008 (diff) | |
Add test that higher-granularity masks override lower-granularity ones
Stream-level masks should always override globally-set masks.
Next-kernel masks should always override both stream-level masks
and globally-set masks.
Tests reveal an issue with the next-kernel mask not overriding the
stream mask on CUDA 11.0+. CUDA appears to apply the per-stream
mask to the QMD/TMD after `launchCallback()` is triggered, making
it impossible to override as currently implemented.
Diffstat (limited to 'libsmctrl_test_stream_mask.c')
| -rw-r--r-- | libsmctrl_test_stream_mask.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/libsmctrl_test_stream_mask.c b/libsmctrl_test_stream_mask.c index 063f934..d8895df 100644 --- a/libsmctrl_test_stream_mask.c +++ b/libsmctrl_test_stream_mask.c | |||
| @@ -1,5 +1,5 @@ | |||
| 1 | // Copyright 2023 Joshua Bakita | 1 | // Copyright 2024 Joshua Bakita |
| 2 | // Test libsmctrl_set_global_mask(). | 2 | // Test libsmctrl_set_stream_mask_ext(). |
| 3 | // All types of partitioning use the same test, so this file is trival. | 3 | // All types of partitioning use the same test, so this file is trival. |
| 4 | 4 | ||
| 5 | #include "libsmctrl_test_mask_shared.h" | 5 | #include "libsmctrl_test_mask_shared.h" |