Summary: | [PNV]oglc shad-compiler(advanced.TestTexture1D_VertShader) regressed | ||
---|---|---|---|
Product: | Mesa | Reporter: | fangxun <xunx.fang> |
Component: | Drivers/DRI/i915 | Assignee: | Ian Romanick <idr> |
Status: | CLOSED WONTFIX | QA Contact: | |
Severity: | major | ||
Priority: | medium | CC: | brianp |
Version: | git | ||
Hardware: | All | ||
OS: | Linux (All) | ||
Whiteboard: | |||
i915 platform: | i915 features: | ||
Bug Depends on: | |||
Bug Blocks: | 42993 |
Description
fangxun
2012-02-10 03:28:49 UTC
I was easily able to bisect this to the commit below. This commit actually starts triggering an assertion failure. A later commit (0f3f6cf) removes the assertion, and converts the assertion failure to the reported segfault. commit 68da4b50e9b6aa72a9b155f650952620063e1b94 Author: Brian Paul <brianp@vmware.com> Date: Sun Oct 23 10:44:47 2011 -0600 mesa: add swrast_texture_image::Buffer In the past, swrast_texture_image::Data has been overloaded. It could either point to malloc'd memory storing texture data, or it could point to a current mapping of GPU memory. Now, Buffer always points to malloc'd memory (if we're not using GPU memory) and Data always points to mapped memory. The next step would be to rename Data -> Map. This change also involves adding swrast functions for mapping textures and renderbuffers prior to rendering to setup the Data pointer. Plus, corresponding functions to unmap texures and renderbuffers. This is very much like similar code in the dri drivers. After some tracing and watchpoint setting, I've figure out what's happening, but I'm not sure what the right fix is. Before calling _tnl_run_pipeline, the i915 driver maps all textures used in vertex shaders (via intel_tex_map_images). Later run_vp tries to map the textures again (via _swrast_map_texture). However, this just replaces the actual mappings with NULL pointers. When the pointers are dereferened, the segfault occurs. I believe setting swrast_texture_image::Buffer in intel_tex_map_images would work around the problem, but I have a hard time believing this is the right fix. *** Bug 45885 has been marked as a duplicate of this bug. *** *** Bug 45886 has been marked as a duplicate of this bug. *** Hmm. do we just no longer need to wrap the tnl with our texture mapping now? Please confirm that this is seen on an SNB or newer platform. I think it just impacts PNV. Closing as won't fix. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.