<div dir="ltr">I pushed it. Thanks Nicholas.<div> </div><div>Stéphane</div><div> </div></div><div class="gmail_extra"> <div class="gmail_quote">On Thu, Sep 22, 2016 at 9:40 AM, Roland Scheidegger <<a href="mailto:sroland@vmware.com" target="_blank">sroland@vmware.com</a>> wrote: <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Am 22.09.2016 um 17:40 schrieb Emil Velikov: > Hi Nicholas, > > On 26 August 2016 at 00:31, Nicholas Bishop <<a href="mailto:nbishop@neverware.com">nbishop@neverware.com</a>> wrote: >> From: Nicholas Bishop <<a href="mailto:nicholasbishop@gmail.com">nicholasbishop@gmail.com</a>> >> >> On Intel Pineview M hardware, the i915 gallium driver doesn't output >> the correct gl_FragCoord. It seems to always have an X coord of 0.0 >> and a Y coord of the window's height in pixels, e.g. 600.0f or such. >> >> I believe this is a regression caused in part by this commit: >> afa035031ff9e0c07a2297d864e46c76f7bfff58 >> >> The old behavior used the output at index zero, while the new behavior >> uses actual zeroes. In the case of gl_FragCoord the output at index >> zero happened to be the correct one, so the behavior appeared correct >> although the code already had a bug. >> >> Fixed by checking for I915_SEMANTIC_POS when setting up texCoords. If >> the generic_mapping is I915_SEMANTIC_POS, look for the >> TGSI_SEMANTIC_POSITION instead of a TGSI_SEMANTIC_GENERIC output. >> >> <a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__bugs.freedesktop.org_show-5Fbug.cgi-3Fid-3D97477&d=CwIFaQ&c=Sqcl0Ez6M0X8aeM67LKIiDJAXVeAw-YihVMNtXt-uEs&r=Vjtt0vs_iqoI31UfJxBl7yv9I2FeiaeAYgMTLKRBc_I&m=Mt7V3FvGsInekgNmMd2jOoCYRJDl4If-uvNtbOhGkRQ&s=JlfHHRDgBQvDfacPpT-qFUmY4drEprImNSdf1L6ch3c&e=" rel="noreferrer" target="_blank">https://urldefense.proofpoint.com/v2/url?u=https-3A__bugs.freedesktop.org_show-5Fbug.cgi-3Fid-3D97477&d=CwIFaQ&c=Sqcl0Ez6M0X8aeM67LKIiDJAXVeAw-YihVMNtXt-uEs&r=Vjtt0vs_iqoI31UfJxBl7yv9I2FeiaeAYgMTLKRBc_I&m=Mt7V3FvGsInekgNmMd2jOoCYRJDl4If-uvNtbOhGkRQ&s=JlfHHRDgBQvDfacPpT-qFUmY4drEprImNSdf1L6ch3c&e=</a> >> --- >> src/gallium/drivers/i915/i915_state_derived.c | 7 ++++++- >> 1 file changed, 6 insertions(+), 1 deletion(-) >> >> diff --git a/src/gallium/drivers/i915/i915_state_derived.c b/src/gallium/drivers/i915/i915_state_derived.c >> index 177b854..dbfbc84 100644 >> --- a/src/gallium/drivers/i915/i915_state_derived.c >> +++ b/src/gallium/drivers/i915/i915_state_derived.c >> @@ -145,7 +145,12 @@ static void calculate_vertex_layout(struct i915_context *i915) >> uint hwtc; >> if (texCoords[i]) { >> hwtc = TEXCOORDFMT_4D; >> - src = draw_find_shader_output(i915->draw, TGSI_SEMANTIC_GENERIC, fs->generic_mapping[i]); >> + if (fs->generic_mapping[i] == I915_SEMANTIC_POS) { >> + src = draw_find_shader_output(i915->draw, TGSI_SEMANTIC_POSITION, 0); >> + } >> + else { >> + src = draw_find_shader_output(i915->draw, TGSI_SEMANTIC_GENERIC, fs->generic_mapping[i]); >> + } > > Personally I'm unfamiliar with the i915g driver, so I've Cc'd Roland > (the author of the commit 'breaking' this) and Stéphane (the author of > this driver). > > Gents, any input on the above ? Oh, I wasn't aware I broke this... This looks to me like before my patch there i915g was indeed just relying on draw code redirecting undefined inputs to 0, which near certainly was unspecified behavior. I don't really know i915g, but this looks like it should be correct. Reviewed-by: Roland Scheidegger <<a href="mailto:sroland@vmware.com">sroland@vmware.com</a>> <div class="HOEnZb"><div class="h5"> > > That said, please drop the unneeded brackets and add the following > line in the commit message for v2. > Cc: <a href="mailto:mesa-stable@lists.freedesktop.org">mesa-stable@lists.freedesktop.org</a> > > Thanks, > Emil > _______________________________________________ mesa-dev mailing list <a href="mailto:mesa-dev@lists.freedesktop.org">mesa-dev@lists.freedesktop.org</a> <a href="https://lists.freedesktop.org/mailman/listinfo/mesa-dev" rel="noreferrer" target="_blank">https://lists.freedesktop.org/mailman/listinfo/mesa-dev</a> </div></div></blockquote></div> </div>