marfrit/marfrit-packages
0
0
Fork 1
Code Issues 2 Pull Requests Actions Packages Projects Releases Wiki Activity

mesa-panvk-bifrost: r3 -> r4 — iter17 XFB primitive decomposition #70

Merged
marfrit merged 1 commits from claude-noether/marfrit-packages:noether/mesa-panvk-bifrost-r4-iter17-xfb-decomp into main 2026-05-21 12:18:23 +00:00
Conversation 0 Commits 1 Files Changed 2
+646 -2
marfrit commented 2026-05-21 12:18:05 +00:00
Owner
Copy Link
Copy Source

iter17 closes the 162 winding_* CTS failures from iter15's baseline by
replacing the upstream pan_nir_lower_xfb call with a panvk-specific NIR
pass (panvk_per_arch(nir_lower_xfb)) that handles per-primitive
decomposition for non-LIST topologies (LINE_STRIP, TRIANGLE_STRIP,
TRIANGLE_FAN, and the four _WITH_ADJACENCY variants).

Topology + per-instance output vertex count are threaded as new sysvals
(vs.xfb_topology + vs.xfb_output_count) so the NIR pass can dispatch
per-topology at runtime without compiling 7+ shader variants.

dEQP-VK.transform_feedback.simple.* result (133596 cases total):
iter15 baseline -> iter17
Pass: 796 958 (+162)
Fail: 243 81 (-162; resume_* by-design only)
NotSupported: 132551 132551
Fatal-skip: 6 6
Pass rate of runnable: 76.2% -> 91.7% (+15.5pp)

100% of the iter15 winding-fail cluster closed. The remaining 81 fails
are all resume_* (pause/resume XFB, by design — we advertise
transformFeedbackDraw=false).

Second-model review (janet) produced 3 findings; Findings 1+2 were
already fixed in the in-tree applied state (stale applied_state/ snapshot
read by reviewer), Finding 3 (degenerate N underflow on N<2) addressed
by gating non-LIST emission on output_count > 0 predicate.

Co-Authored-By: Claude Opus 4.7 noreply@anthropic.com

iter17 closes the 162 winding_* CTS failures from iter15's baseline by replacing the upstream pan_nir_lower_xfb call with a panvk-specific NIR pass (panvk_per_arch(nir_lower_xfb)) that handles per-primitive decomposition for non-LIST topologies (LINE_STRIP, TRIANGLE_STRIP, TRIANGLE_FAN, and the four _WITH_ADJACENCY variants). Topology + per-instance output vertex count are threaded as new sysvals (vs.xfb_topology + vs.xfb_output_count) so the NIR pass can dispatch per-topology at runtime without compiling 7+ shader variants. dEQP-VK.transform_feedback.simple.* result (133596 cases total): iter15 baseline -> iter17 Pass: 796 958 (+162) Fail: 243 81 (-162; resume_* by-design only) NotSupported: 132551 132551 Fatal-skip: 6 6 Pass rate of runnable: 76.2% -> 91.7% (+15.5pp) 100% of the iter15 winding-fail cluster closed. The remaining 81 fails are all resume_* (pause/resume XFB, by design — we advertise transformFeedbackDraw=false). Second-model review (janet) produced 3 findings; Findings 1+2 were already fixed in the in-tree applied state (stale applied_state/ snapshot read by reviewer), Finding 3 (degenerate N underflow on N<2) addressed by gating non-LIST emission on `output_count > 0` predicate. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
marfrit added 1 commit 2026-05-21 12:18:05 +00:00
mesa-panvk-bifrost: r3 -> r4 — iter17 XFB primitive decomposition 83e8eca56d 
iter17 closes the 162 winding_* CTS failures from iter15's baseline by
replacing the upstream pan_nir_lower_xfb call with a panvk-specific NIR
pass (panvk_per_arch(nir_lower_xfb)) that handles per-primitive
decomposition for non-LIST topologies (LINE_STRIP, TRIANGLE_STRIP,
TRIANGLE_FAN, and the four _WITH_ADJACENCY variants).

Topology + per-instance output vertex count are threaded as new sysvals
(vs.xfb_topology + vs.xfb_output_count) so the NIR pass can dispatch
per-topology at runtime without compiling 7+ shader variants.

dEQP-VK.transform_feedback.simple.* result (133596 cases total):
                  iter15 baseline  ->  iter17
  Pass:             796               958   (+162)
  Fail:             243               81    (-162; resume_* by-design only)
  NotSupported:     132551            132551
  Fatal-skip:       6                 6
  Pass rate of runnable: 76.2% -> 91.7% (+15.5pp)

100% of the iter15 winding-fail cluster closed. The remaining 81 fails
are all resume_* (pause/resume XFB, by design — we advertise
transformFeedbackDraw=false).

Second-model review (janet) produced 3 findings; Findings 1+2 were
already fixed in the in-tree applied state (stale applied_state/ snapshot
read by reviewer), Finding 3 (degenerate N underflow on N<2) addressed
by gating non-LIST emission on `output_count > 0` predicate.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
marfrit merged commit a9f1b833b9 into main 2026-05-21 12:18:23 +00:00
marfrit deleted branch noether/mesa-panvk-bifrost-r4-iter17-xfb-decomp 2026-05-21 12:18:23 +00:00
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
No items
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
marfrit (Markus Fritsche)
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: marfrit/marfrit-packages#70
Delete Branch "claude-noether/marfrit-packages:noether/mesa-panvk-bifrost-r4-iter17-xfb-decomp"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?