[Mesa-dev] [PATCH] compiler/spirv: reject invalid shader code properly

After bebe3d626e5, b->fail_jump is prepared after vtn_create_builder which can longjmp(3) to it through its vtx_assert()s. This corrupts the stack and creates confusing core dumps, so we need to avoid it. While there, I decided to print the offending values for debugability. ---

Re: [Mesa-dev] [PATCH] compiler/spirv: reject invalid shader code properly

And pushed. On Fri, Jun 1, 2018 at 8:04 AM, Jason Ekstrand wrote: > Reviewed-by: Jason Ekstrand > > On Fri, Jun 1, 2018 at 5:27 AM, Martin Pelikán wrote: > >> After bebe3d626e5, b->fail_jump is prepared after vtn_create_builder >> which can longjmp(3) to it through its vtx_assert()s. This

Re: [Mesa-dev] [PATCH] compiler/spirv: reject invalid shader code properly

Reviewed-by: Jason Ekstrand On Fri, Jun 1, 2018 at 5:27 AM, Martin Pelikán wrote: > After bebe3d626e5, b->fail_jump is prepared after vtn_create_builder > which can longjmp(3) to it through its vtx_assert()s. This corrupts > the stack and creates confusing core dumps, so we need to avoid it.

Re: [Mesa-dev] [PATCH] compiler/spirv: reject invalid shader code properly

On Sun, May 13, 2018 at 9:01 AM, Martin Pelikán wrote: > After bebe3d626e5, b->fail_jump is prepared after vtn_create_builder > which can longjmp(3) to it through its vtx_assert()s. This corrupts > the stack and creates confusing core dumps, so we need to avoid it. > > While there, I decided to

[Mesa-dev] [PATCH] compiler/spirv: reject invalid shader code properly

After bebe3d626e5, b->fail_jump is prepared after vtn_create_builder which can longjmp(3) to it through its vtx_assert()s. This corrupts the stack and creates confusing core dumps, so we need to avoid it. While there, I decided to print the offending values for debugability. ---

Re: [Mesa-dev] [PATCH] compiler/spirv: reject invalid shader code properly

Ugh... I know this worked correctly when I did it. I think the setjmp must not have quite survived over of the refactors. I need to take a serious look at out on Monday. On May 13, 2018 09:02:24 Martin Pelikán wrote: After bebe3d626e5, b->fail_jump is prepared after