similar to: [PATCH] nv50/ir: saturate FRC result to avoid completely bogus values

On Tue, Nov 18, 2014 at 8:54 AM, Roland Scheidegger <sroland at vmware.com> wrote: > Am 18.11.2014 um 05:03 schrieb Ilia Mirkin: >> For values above integer accuracy in floats, val - floor(val) might >> actually produce a value greater than 1. For such large floats, it's >> reasonable to be imprecise, but it's unreasonable for FRC to return a >> value that

On 18/11/14 14:34, Roland Scheidegger wrote: > Am 18.11.2014 um 15:05 schrieb Ilia Mirkin: >> On Tue, Nov 18, 2014 at 8:54 AM, Roland Scheidegger <sroland at vmware.com> wrote: >>> Am 18.11.2014 um 05:03 schrieb Ilia Mirkin: >>>> For values above integer accuracy in floats, val - floor(val) might >>>> actually produce a value greater than 1. For such

Am 18.11.2014 um 05:03 schrieb Ilia Mirkin: > For values above integer accuracy in floats, val - floor(val) might > actually produce a value greater than 1. For such large floats, it's > reasonable to be imprecise, but it's unreasonable for FRC to return a > value that is not between 0 and 1. > > Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu> > --- >

Am 18.11.2014 um 15:05 schrieb Ilia Mirkin: > On Tue, Nov 18, 2014 at 8:54 AM, Roland Scheidegger <sroland at vmware.com> wrote: >> Am 18.11.2014 um 05:03 schrieb Ilia Mirkin: >>> For values above integer accuracy in floats, val - floor(val) might >>> actually produce a value greater than 1. For such large floats, it's >>> reasonable to be imprecise,

Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu> --- .../drivers/nouveau/codegen/nv50_ir_emit_nvc0.cpp | 66 +++++++++++++++++++++- 1 file changed, 63 insertions(+), 3 deletions(-) diff --git a/src/gallium/drivers/nouveau/codegen/nv50_ir_emit_nvc0.cpp b/src/gallium/drivers/nouveau/codegen/nv50_ir_emit_nvc0.cpp index dfb093c..e38a3b8 100644 ---

On 21/05/14 00:39, Ilia Mirkin wrote: > From: Christoph Bumiller <christoph.bumiller at speed.at> > > Reviewed-by: Ilia Mirkin <imirkin at alum.mit.edu> > Cc: "10.2" <mesa-stable at lists.freedesktop.org> > --- > src/gallium/drivers/nouveau/codegen/nv50_ir_from_tgsi.cpp | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff

On 21/05/14 19:53, Ilia Mirkin wrote: > On Wed, May 21, 2014 at 2:51 PM, Emil Velikov <emil.l.velikov at gmail.com> wrote: >> On 21/05/14 00:39, Ilia Mirkin wrote: >>> From: Christoph Bumiller <christoph.bumiller at speed.at> >>> >>> Reviewed-by: Ilia Mirkin <imirkin at alum.mit.edu> >>> Cc: "10.2" <mesa-stable at

Retrieving the high 32 bits of a signed multiply is rather annoying. It appears that the simplest way to do this is to compute the absolute value of the arguments, and perform a u32 x u32 -> u64 operation. If the arguments' signs differ, then negate the result. Since there is no u64 support in the cvt instruction, we have the perform the 2's complement negation "by hand".

I went through the gallium-nine tree and picked out nouveau patches that are general bug-fixes. The first bunch I'd like to also get into 10.2. I've reviewed all of them and they make sense to me, but sending them out for public review as well in case there are any objections. Unless I hear objections, I'd like to push this by Friday. Christoph Bumiller (11): nv50,nvc0: always pull

Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu> --- Untested beyond compiling a few shaders to see if they look like they might work. nvdisasm agrees with envydis's decoding of these things. Will definitely get ahold of a G200 to run tests on before pushing this. .../drivers/nouveau/codegen/nv50_ir_emit_nv50.cpp | 94 ++++++++++++++++++---

Signed-off-by: Tobias Klausmann <tobias.johannes.klausmann at mni.thm.de> --- .../drivers/nouveau/codegen/nv50_ir_driver.h | 1 + .../drivers/nouveau/codegen/nv50_ir_from_tgsi.cpp | 27 ++++++++++++++++++++-- 2 files changed, 26 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/nouveau/codegen/nv50_ir_driver.h b/src/gallium/drivers/nouveau/codegen/nv50_ir_driver.h

previously, if we had something like: gl_ViewportIndex = idx; for(int i = 0; i < gl_in.length(); i++) { gl_Position = gl_in[i].gl_Position; EmitVertex(); } EndPrimitive(); we failed to set the right ViewportIndex. To resolve this, save the ViewportIndex and store it to the right register on each emit. This fixes the remaining piglit tests in ARB_viewport_array for nvc0. Note: Not

previously, if we had something like: gl_ViewportIndex = idx; for(int i = 0; i < gl_in.length(); i++) { gl_Position = gl_in[i].gl_Position; EmitVertex(); } EndPrimitive(); we failed to set the right ViewportIndex. To resolve this, save the ViewportIndex and store it to the right register on each emit. This fixes the remaining piglit tests in ARB_viewport_array for nvc0. Note: Not

Does this give correct results for special floats (0, infs)? We tried to improve (for single floats) x86 rcp in llvmpipe with newton-raphson, but unfortunately not being able to give correct results for these two cases (without even more additional code) meant it got all disabled in the end (you can still see that code in the driver) since the problems are at least as bad as those due to bad

This is parallel to the pre-SM50 change which does this. Adjusts the shuffles / quadops to make the values correct relative to lane 0, and then splat the results to all lanes for the final move into the target register. Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu> --- Entirely untested beyond compilation. Should check bin/tex-miplevel-selection textureGrad Cube

From: Christoph Bumiller <christoph.bumiller at speed.at> Reviewed-by: Ilia Mirkin <imirkin at alum.mit.edu> Cc: "10.2" <mesa-stable at lists.freedesktop.org> --- src/gallium/drivers/nouveau/codegen/nv50_ir_from_tgsi.cpp | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/nouveau/codegen/nv50_ir_from_tgsi.cpp

On Wed, May 21, 2014 at 3:01 PM, Emil Velikov <emil.l.velikov at gmail.com> wrote: > On 21/05/14 19:53, Ilia Mirkin wrote: >> On Wed, May 21, 2014 at 2:51 PM, Emil Velikov <emil.l.velikov at gmail.com> wrote: >>> On 21/05/14 00:39, Ilia Mirkin wrote: >>>> From: Christoph Bumiller <christoph.bumiller at speed.at> >>>> >>>>

On Wed, May 21, 2014 at 2:51 PM, Emil Velikov <emil.l.velikov at gmail.com> wrote: > On 21/05/14 00:39, Ilia Mirkin wrote: >> From: Christoph Bumiller <christoph.bumiller at speed.at> >> >> Reviewed-by: Ilia Mirkin <imirkin at alum.mit.edu> >> Cc: "10.2" <mesa-stable at lists.freedesktop.org> >> --- >>

OK, so there's a bunch of stuff in here. The geometry stuff is based on the work started by Bryan Cain and Christoph Bumiller. Patches 01-12: Add support for geometry shaders and fix related issues Patches 13-14: Make it possible for fb clears to operate on texture attachments with an explicit layer set (as is allowed in gl 3.2). Patches 15-17: Make ARB_texture_multisample work

Maybe I didn't describe the question very well. I want to match these patterns: (and (mul node:$val0, node:$val1)) (and (add node:$val0, node:$val1)) (and (sub node:$val0, node:$val1)) (and (udiv node:$val0, node:$val1)) into "four" different machines instructions, respectively: MUL_AND ADD_AND SUB_AND UDIV_AND That is, this machine is capable of doing an