Non-Adreno: find every driver pair, Venus before the bridge, fix the bridge's begin info

From the logs testers sent since c9ebb247:
- The app read the drivers repo's newest 15 releases only; the component releases (gamescope,
  Venus) pushed every PanVK pair and RADV-Xclipse past them, so Auto found no pair on any Mali
  and switched the Mali bridge on. The non-Adreno repo is now read in one page of 100.
- Auto's stand-in for a missing pair: Mali (Venus) where the apk carries it (Steam runs on it
  from the G57 to the G1-Ultra), the bridge otherwise, nothing on a driver below Vulkan 1.2. It
  no longer clears the Linux driver, does nothing when the list could not be read, records that
  it chose the route, and turns it off again once a pair is installed. A bridge that earlier
  builds left on under Auto is taken as Auto's.
- Mali bridge: vkBeginCommandBuffer sent its begin info without the count the server decodes, so
  Mali got a zeroed begin info with stale flags and crashed at 0x10 (Pixel 8 Pro G715, Lenovo
  G615, Xiaomi G925, Tab S11).
- Mali-G1 series (MT6993, arch v14) known by its chip; its label says Venus, not the bridge.
- A session Android killed for memory is said so on the next start (the S24 FE "crashes" were
  low-memory kills, not driver faults).
- The in-app updater never installs an official (Adreno-only) build over this test build.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
The412BannerandClaude Opus 5.5 committed 2026-10-19 09:07:44 +00:00
1 parent 6bb7ad74bf
commit 8e8c27be95
11 files changed
+153 -22

No files matched your search

+11 -2
View File
@@ -170,8 +170,17 @@ VKAPI_ATTR VkResult VKAPI_CALL vkb_ep_vkBeginCommandBuffer(VkCommandBuffer comma
vkb_enc *e = vkb_cmd_record_begin(commandBuffer, VKB_CMD_vkBeginCommandBuffer);
vkb_enc_u64(e, cb->obj.remote);
tls_encoding = cb;
vkb_enc_u8(e, pBeginInfo != NULL);
if (pBeginInfo) vkb_enc_VkCommandBufferBeginInfo(e, pBeginInfo);
/* The server decodes a present pointer as a count and that many structs (vkb_sv_vkBeginCommandBuffer,
* as vkb_wire_vkBeginCommandBuffer sends it). Without the count it read 0 and handed Mali a zeroed
* 16-byte begin info, so flags and pInheritanceInfo came from stale memory: the driver crashed in
* vkBeginCommandBuffer at 0x10 whenever those flags had RENDER_PASS_CONTINUE set. */
if (pBeginInfo) {
vkb_enc_u8(e, 1);
vkb_enc_u32(e, 1);
vkb_enc_VkCommandBufferBeginInfo(e, pBeginInfo);
} else {
vkb_enc_u8(e, 0);
}
tls_encoding = NULL;
vkb_cmd_record_end(commandBuffer);
return cb->broken ? VK_ERROR_OUT_OF_HOST_MEMORY : VK_SUCCESS;