Mesa (staging/21.3): intel/perf: fix perf equation subslice mask generation for gfx12+

GitLab Mirror gitlab-mirror at kemper.freedesktop.org
Sun Nov 7 21:49:51 UTC 2021


Module: Mesa
Branch: staging/21.3
Commit: 059f30e8cdc51c680a428f6bdee17c84304d2ae0
URL:    http://cgit.freedesktop.org/mesa/mesa/commit/?id=059f30e8cdc51c680a428f6bdee17c84304d2ae0

Author: Lionel Landwerlin <lionel.g.landwerlin at intel.com>
Date:   Sat Apr  3 13:34:11 2021 +0300

intel/perf: fix perf equation subslice mask generation for gfx12+

v2: Fix comment change (Marcin)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin at intel.com>
Cc: <mesa-stable at lists.freedesktop.org>
Reviewed-by: Francisco Jerez <currojerez at riseup.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10015>
(cherry picked from commit 67619d8153249e2cc3c36ac9a28aec9ded0ca1ba)

---

 .pick_status.json           | 2 +-
 src/intel/perf/intel_perf.c | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/.pick_status.json b/.pick_status.json
index 982d34499ba..d967218ebab 100644
--- a/.pick_status.json
+++ b/.pick_status.json
@@ -76,7 +76,7 @@
         "description": "intel/perf: fix perf equation subslice mask generation for gfx12+",
         "nominated": true,
         "nomination_type": 0,
-        "resolution": 0,
+        "resolution": 1,
         "main_sha": null,
         "because_sha": null
     },
diff --git a/src/intel/perf/intel_perf.c b/src/intel/perf/intel_perf.c
index 044bbe2c007..a013e025ab3 100644
--- a/src/intel/perf/intel_perf.c
+++ b/src/intel/perf/intel_perf.c
@@ -364,15 +364,15 @@ compute_topology_builtins(struct intel_perf_config *perf,
    perf->sys_vars.eu_threads_count = devinfo->num_thread_per_eu;
 
    /* The subslice mask builtin contains bits for all slices. Prior to Gfx11
-    * it had groups of 3bits for each slice, on Gfx11 it's 8bits for each
-    * slice.
+    * it had groups of 3bits for each slice, on Gfx11 and above it's 8bits for
+    * each slice.
     *
     * Ideally equations would be updated to have a slice/subslice query
     * function/operator.
     */
    perf->sys_vars.subslice_mask = 0;
 
-   int bits_per_subslice = devinfo->ver == 11 ? 8 : 3;
+   int bits_per_subslice = devinfo->ver >= 11 ? 8 : 3;
 
    for (int s = 0; s < util_last_bit(devinfo->slice_masks); s++) {
       for (int ss = 0; ss < (devinfo->subslice_slice_stride * 8); ss++) {



More information about the mesa-commit mailing list