diff options
| author | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-10-07 13:12:06 -0400 |
|---|---|---|
| committer | Arthur Hanson Rasmusson <arthur@vgpu.io> | 2026-10-07 13:12:06 -0400 |
| commit | bcaab155a1318373b6c3a4c83fbfe2d0dbf1f44e (patch) | |
| tree | 6054797e3c7ed5d86f3f4dc77eae6ad45c518889 /kernel/gsp_phys_gpu.h | |
| parent | 3dd883f43c325f942b25be1accad82c7b431520c (diff) | |
| download | open-iov-bcaab155a1318373b6c3a4c83fbfe2d0dbf1f44e.tar.gz open-iov-bcaab155a1318373b6c3a4c83fbfe2d0dbf1f44e.zip | |
vgsp: m91 - gsp_phys_rmcall_gr_ptr_guest: keep QMD pointer pairs at the guest VAs
The m90/m91 trap decoded the SM I-fetch failure precisely: GPC0/GPC1 TPC0
SM0 WARP_ESR ERROR=0x20 ADDR_ERROR_TYPE=0xb = MMU_NACK, PC=0x109880 (the
rewritten window-0 shadow program VA). The FE reads the same small-PT line
fine (the forwarded inline QMD in the qmdwin page is processed to CWD
ACTIVE), so the SM walk is not using the channel/ladder VAS the shadow
lives in - it appears to use the guest subctx page directory (instance
SC_PAGE_DIRECTORY_BASE 0x10956b000), which maps the guest VAs the real
client's captured QMD points at, and not our window.
Add gr_ptr_guest: when set, the forwarded launch leaves the QMD pointer
pairs (param w32/33, program w48/49, prefetch w8/w31) exactly as the guest
driver wrote them (as the real driver's captured QMD carries), and only
relocates the QMD page itself (qmdwin, a proven FE read). Shadows are still
copied for logging but unused.
Diffstat (limited to 'kernel/gsp_phys_gpu.h')
| -rw-r--r-- | kernel/gsp_phys_gpu.h | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/kernel/gsp_phys_gpu.h b/kernel/gsp_phys_gpu.h index 0b85758..62cc16d 100644 --- a/kernel/gsp_phys_gpu.h +++ b/kernel/gsp_phys_gpu.h @@ -337,6 +337,9 @@ int gsp_phys_rmcall_faultbuf_sys_value(void); * (proven small-PT line) instead of the window-1 scratch range (default 0). */ int gsp_phys_rmcall_gr_shadow_win0_value(void); +/* m91: keep the forwarded QMD pointer pairs at the guest VAs (default 0). */ +int gsp_phys_rmcall_gr_ptr_guest_value(void); + /* m72: append the vendor local-memory context block to the GR prelude so * SKEDCHECK05 LOCAL_MEMORY_TOTAL_SIZE passes (default 0); base VA and * per-TPC size are overridable for A/B runs. */ |