mirror of https://gitlab.freedesktop.org/mesa/mesa
e858da39e5
The intel_perf_query path used for performance queries on GL was
passing a bogus "end" pointer to intel_perf_query_result_accumulate(),
causing it to accumulate garbage values. This was causing the values
of many performance counters to be corrupted.
The "end" pointer was incorrect because the current code was assuming
that different OA reports were located TOTAL_QUERY_DATA_SIZE bytes
apart, which is a hard-coded preprocessor define. However recent
(Gfx12+) hardware generations use a variable query size determined by
the query layout. Use the size derived from it instead, and remove
the stale define.
Fixes:
|
||
|---|---|---|
| .. | ||
| gen_perf.py | ||
| intel_perf.c | ||
| intel_perf.h | ||
| intel_perf_mdapi.c | ||
| intel_perf_mdapi.h | ||
| intel_perf_private.h | ||
| intel_perf_query.c | ||
| intel_perf_query.h | ||
| intel_perf_regs.h | ||
| intel_perf_setup.h | ||
| meson.build | ||
| oa-adl.xml | ||
| oa-bdw.xml | ||
| oa-bxt.xml | ||
| oa-cflgt2.xml | ||
| oa-cflgt3.xml | ||
| oa-chv.xml | ||
| oa-dg1.xml | ||
| oa-ehl.xml | ||
| oa-glk.xml | ||
| oa-hsw.xml | ||
| oa-icl.xml | ||
| oa-kblgt2.xml | ||
| oa-kblgt3.xml | ||
| oa-rkl.xml | ||
| oa-sklgt2.xml | ||
| oa-sklgt3.xml | ||
| oa-sklgt4.xml | ||
| oa-tglgt1.xml | ||
| oa-tglgt2.xml | ||