Re: [Mesa-dev] [PATCH] radeonsi/gfx9: add a temporary workaround for a tessellation driver bug
On Fri, Aug 25, 2017 at 8:42 PM, Marek Olšákwrote: > On Tue, Aug 22, 2017 at 2:15 PM, Nicolai Hähnle wrote: >> On 22.08.2017 14:10, Nicolai Hähnle wrote: >>> >>> On 22.08.2017 13:00, Marek Olšák wrote: On Tue, Aug 22, 2017 at 9:37 AM, Nicolai Hähnle wrote: > > On 18.08.2017 19:06, Marek Olšák wrote: >> >> >> Ping. >> >> On Wed, Aug 16, 2017 at 12:57 AM, Marek Olšák wrote: >>> >>> >>> From: Marek Olšák >>> >>> The workaround will do for now. The root cause is still unknown. >>> >>> This fixes new piglit: 16in-1out > > > > I don't see this test. Did you already send it out? "[PATCH] arb_tessellation_shader: new tests for a radeonsi bug" on the piglit ML. >>> >>> >>> Curious, I can't reproduce the problem on my Raven. >> >> >> VGT_LS_HS_CONFIG.NUM_PATCHES is 16, so there should definitely be more than >> one wave per thread-group. > > The test is insufficient to reproduce the issue, but you'll see it > when you run the test with ST_DEBUG=wf with and without the fix. I just pushed a commit into piglit that adjusts the test, so that random geometry results in a failure. Marek ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] radeonsi/gfx9: add a temporary workaround for a tessellation driver bug
On Tue, Aug 22, 2017 at 2:15 PM, Nicolai Hähnlewrote: > On 22.08.2017 14:10, Nicolai Hähnle wrote: >> >> On 22.08.2017 13:00, Marek Olšák wrote: >>> >>> On Tue, Aug 22, 2017 at 9:37 AM, Nicolai Hähnle >>> wrote: On 18.08.2017 19:06, Marek Olšák wrote: > > > Ping. > > On Wed, Aug 16, 2017 at 12:57 AM, Marek Olšák wrote: >> >> >> From: Marek Olšák >> >> The workaround will do for now. The root cause is still unknown. >> >> This fixes new piglit: 16in-1out I don't see this test. Did you already send it out? >>> >>> >>> "[PATCH] arb_tessellation_shader: new tests for a radeonsi bug" on the >>> piglit ML. >> >> >> Curious, I can't reproduce the problem on my Raven. > > > VGT_LS_HS_CONFIG.NUM_PATCHES is 16, so there should definitely be more than > one wave per thread-group. The test is insufficient to reproduce the issue, but you'll see it when you run the test with ST_DEBUG=wf with and without the fix. Marek ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] radeonsi/gfx9: add a temporary workaround for a tessellation driver bug
On 22.08.2017 14:10, Nicolai Hähnle wrote: On 22.08.2017 13:00, Marek Olšák wrote: On Tue, Aug 22, 2017 at 9:37 AM, Nicolai Hähnlewrote: On 18.08.2017 19:06, Marek Olšák wrote: Ping. On Wed, Aug 16, 2017 at 12:57 AM, Marek Olšák wrote: From: Marek Olšák The workaround will do for now. The root cause is still unknown. This fixes new piglit: 16in-1out I don't see this test. Did you already send it out? "[PATCH] arb_tessellation_shader: new tests for a radeonsi bug" on the piglit ML. Curious, I can't reproduce the problem on my Raven. VGT_LS_HS_CONFIG.NUM_PATCHES is 16, so there should definitely be more than one wave per thread-group. >> Marek -- Lerne, wie die Welt wirklich ist, Aber vergiss niemals, wie sie sein sollte. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] radeonsi/gfx9: add a temporary workaround for a tessellation driver bug
On 22.08.2017 13:00, Marek Olšák wrote: On Tue, Aug 22, 2017 at 9:37 AM, Nicolai Hähnlewrote: On 18.08.2017 19:06, Marek Olšák wrote: Ping. On Wed, Aug 16, 2017 at 12:57 AM, Marek Olšák wrote: From: Marek Olšák The workaround will do for now. The root cause is still unknown. This fixes new piglit: 16in-1out I don't see this test. Did you already send it out? "[PATCH] arb_tessellation_shader: new tests for a radeonsi bug" on the piglit ML. Curious, I can't reproduce the problem on my Raven. Marek -- Lerne, wie die Welt wirklich ist, Aber vergiss niemals, wie sie sein sollte. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] radeonsi/gfx9: add a temporary workaround for a tessellation driver bug
On Tue, Aug 22, 2017 at 9:37 AM, Nicolai Hähnlewrote: > On 18.08.2017 19:06, Marek Olšák wrote: >> >> Ping. >> >> On Wed, Aug 16, 2017 at 12:57 AM, Marek Olšák wrote: >>> >>> From: Marek Olšák >>> >>> The workaround will do for now. The root cause is still unknown. >>> >>> This fixes new piglit: 16in-1out > > > I don't see this test. Did you already send it out? "[PATCH] arb_tessellation_shader: new tests for a radeonsi bug" on the piglit ML. Marek ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] radeonsi/gfx9: add a temporary workaround for a tessellation driver bug
On 18.08.2017 19:06, Marek Olšák wrote: Ping. On Wed, Aug 16, 2017 at 12:57 AM, Marek Olšákwrote: From: Marek Olšák The workaround will do for now. The root cause is still unknown. This fixes new piglit: 16in-1out I don't see this test. Did you already send it out? Cc: 17.1 17.2 Reviewed-by: Nicolai Hähnle --- src/gallium/drivers/radeonsi/si_state_draw.c | 6 +- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeonsi/si_state_draw.c b/src/gallium/drivers/radeonsi/si_state_draw.c index f17f570..a26e38d 100644 --- a/src/gallium/drivers/radeonsi/si_state_draw.c +++ b/src/gallium/drivers/radeonsi/si_state_draw.c @@ -187,21 +187,25 @@ static void si_emit_derived_tess_state(struct si_context *sctx, /* Make sure the output data fits in the offchip buffer */ *num_patches = MIN2(*num_patches, (sctx->screen->tess_offchip_block_dw_size * 4) / output_patch_size); /* Not necessary for correctness, but improves performance. The * specific value is taken from the proprietary driver. */ *num_patches = MIN2(*num_patches, 40); - if (sctx->b.chip_class == SI) { + if (sctx->b.chip_class == SI || + /* TODO: fix GFX9 where a threadgroup contains more than 1 wave and +* LS vertices per patch > HS vertices per patch. Piglit: 16in-1out */ + (sctx->b.chip_class == GFX9 && +num_tcs_input_cp > num_tcs_output_cp)) { /* SI bug workaround, related to power management. Limit LS-HS * threadgroups to only one wave. */ unsigned one_wave = 64 / MAX2(num_tcs_input_cp, num_tcs_output_cp); *num_patches = MIN2(*num_patches, one_wave); } /* The VGT HS block increments the patch ID unconditionally * within a single threadgroup. This results in incorrect * patch IDs when instanced draws are used. -- 2.7.4 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev -- Lerne, wie die Welt wirklich ist, Aber vergiss niemals, wie sie sein sollte. ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Re: [Mesa-dev] [PATCH] radeonsi/gfx9: add a temporary workaround for a tessellation driver bug
Ping. On Wed, Aug 16, 2017 at 12:57 AM, Marek Olšákwrote: > From: Marek Olšák > > The workaround will do for now. The root cause is still unknown. > > This fixes new piglit: 16in-1out > > Cc: 17.1 17.2 > --- > src/gallium/drivers/radeonsi/si_state_draw.c | 6 +- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff --git a/src/gallium/drivers/radeonsi/si_state_draw.c > b/src/gallium/drivers/radeonsi/si_state_draw.c > index f17f570..a26e38d 100644 > --- a/src/gallium/drivers/radeonsi/si_state_draw.c > +++ b/src/gallium/drivers/radeonsi/si_state_draw.c > @@ -187,21 +187,25 @@ static void si_emit_derived_tess_state(struct > si_context *sctx, > /* Make sure the output data fits in the offchip buffer */ > *num_patches = MIN2(*num_patches, > (sctx->screen->tess_offchip_block_dw_size * 4) / > output_patch_size); > > /* Not necessary for correctness, but improves performance. The > * specific value is taken from the proprietary driver. > */ > *num_patches = MIN2(*num_patches, 40); > > - if (sctx->b.chip_class == SI) { > + if (sctx->b.chip_class == SI || > + /* TODO: fix GFX9 where a threadgroup contains more than 1 wave > and > +* LS vertices per patch > HS vertices per patch. Piglit: > 16in-1out */ > + (sctx->b.chip_class == GFX9 && > +num_tcs_input_cp > num_tcs_output_cp)) { > /* SI bug workaround, related to power management. Limit LS-HS > * threadgroups to only one wave. > */ > unsigned one_wave = 64 / MAX2(num_tcs_input_cp, > num_tcs_output_cp); > *num_patches = MIN2(*num_patches, one_wave); > } > > /* The VGT HS block increments the patch ID unconditionally > * within a single threadgroup. This results in incorrect > * patch IDs when instanced draws are used. > -- > 2.7.4 > ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
[Mesa-dev] [PATCH] radeonsi/gfx9: add a temporary workaround for a tessellation driver bug
From: Marek OlšákThe workaround will do for now. The root cause is still unknown. This fixes new piglit: 16in-1out Cc: 17.1 17.2 --- src/gallium/drivers/radeonsi/si_state_draw.c | 6 +- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeonsi/si_state_draw.c b/src/gallium/drivers/radeonsi/si_state_draw.c index f17f570..a26e38d 100644 --- a/src/gallium/drivers/radeonsi/si_state_draw.c +++ b/src/gallium/drivers/radeonsi/si_state_draw.c @@ -187,21 +187,25 @@ static void si_emit_derived_tess_state(struct si_context *sctx, /* Make sure the output data fits in the offchip buffer */ *num_patches = MIN2(*num_patches, (sctx->screen->tess_offchip_block_dw_size * 4) / output_patch_size); /* Not necessary for correctness, but improves performance. The * specific value is taken from the proprietary driver. */ *num_patches = MIN2(*num_patches, 40); - if (sctx->b.chip_class == SI) { + if (sctx->b.chip_class == SI || + /* TODO: fix GFX9 where a threadgroup contains more than 1 wave and +* LS vertices per patch > HS vertices per patch. Piglit: 16in-1out */ + (sctx->b.chip_class == GFX9 && +num_tcs_input_cp > num_tcs_output_cp)) { /* SI bug workaround, related to power management. Limit LS-HS * threadgroups to only one wave. */ unsigned one_wave = 64 / MAX2(num_tcs_input_cp, num_tcs_output_cp); *num_patches = MIN2(*num_patches, one_wave); } /* The VGT HS block increments the patch ID unconditionally * within a single threadgroup. This results in incorrect * patch IDs when instanced draws are used. -- 2.7.4 ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev