OpenHarmony/third_party_mesa3d
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

intel/perf: fix roll over PERF_CNT counter accumulation

Browse Source 
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 5ba6d9941b ("intel/perf: add mdapi writes for register perf counters")
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9163>
This commit is contained in:
Lionel Landwerlin
2021-02-19 22:10:33 +02:00
committed by Marge Bot
parent 018393d851
commit 8b44e45347
1 changed files with 6 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/intel/perf/gen_perf.c
View File

@@ -1144,8 +1144,12 @@ gen_perf_query_result_read_perfcnts(struct gen_perf_query_result *result,
const uint64_t *end)
{
for (uint32_t i = 0; i < 2; i++) {
result->accumulator[query->perfcnt_offset + i] =
(end[i] & PERF_CNT_VALUE_MASK) - (start[i] & PERF_CNT_VALUE_MASK);
uint64_t v0 = start[i] & PERF_CNT_VALUE_MASK;
uint64_t v1 = end[i] & PERF_CNT_VALUE_MASK;
result->accumulator[query->perfcnt_offset + i] = v0 > v1 ?
(PERF_CNT_VALUE_MASK + 1 + v1 - v0) :
(v1 - v0);
}
}