On Tue, Jan 8, 2013 at 3:10 AM, Eric Anholt <[email protected]> wrote:
> Kenneth Graunke <[email protected]> writes:
>> diff --git a/src/mesa/drivers/dri/i965/brw_fs_emit.cpp
>> b/src/mesa/drivers/dri/i965/brw_fs_emit.cpp
>> index 63f09fe..0c2b696 100644
>> --- a/src/mesa/drivers/dri/i965/brw_fs_emit.cpp
>> +++ b/src/mesa/drivers/dri/i965/brw_fs_emit.cpp
>> @@ -387,8 +387,11 @@ fs_generator::generate_tex(fs_inst *inst, struct
>> brw_reg dst, struct brw_reg src
>> msg_type = GEN5_SAMPLER_MESSAGE_SAMPLE_RESINFO;
>> break;
>> case SHADER_OPCODE_TXD:
>> - /* There is no sample_d_c message; comparisons are done manually */
>> - msg_type = GEN5_SAMPLER_MESSAGE_SAMPLE_DERIVS;
>> + if (inst->shadow_compare) {
>
> maybe an assert(intel->is_haswell) with a note that otherwise it had
> been lowered, here and in the vs?
>
Following 2 gles3 conformance test cases fail after this commit (899017f):
1. shadow_execution_frag.test
GTF3: brw_fs_emit.cpp:392: Assertion `intel->is_haswell' failed
2. shadow_execution_vert.test
GTF3: brw_vec4_emit.cpp:283: Assertion `intel->is_haswell' failed.
Probably inst->shadow_compare is set incorrectly for Gen6. Both test cases
pass if we set inst->shadow_compare = false for Gen6.
_______________________________________________
mesa-dev mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/mesa-dev