Re: [Mesa-dev] [PATCH 1/3] gallium/hud: fix a problem where objects are free'd while in use.

Hi Emil, On 07.11.2016 18:37, Laurent Carlier wrote: Le lundi 7 novembre 2016, 18:32:07 CET Nicolai Hähnle a écrit : Looks good to me as well, and pushed! Thanks for the respin and sorry it took so long. [...] Maybe cc 13.0 ? It's buggy with 13.0 and it will be a nice fix I didn't think

Re: [Mesa-dev] [PATCH] radv: emit correct last export when Z/stencil export is enabled

Reviewed-by: Bas Nieuwenhuizen On Tue, Nov 8, 2016 at 7:24 AM, Dave Airlie wrote: > From: Dave Airlie > > I was getting a random GPU hang in the renderpass simple tests, > it turns out sometimes radv emitted the wrong thing

[Mesa-dev] [PATCH 1/3] ac/nir: add support for discard_if intrinsic (v2)

From: Dave Airlie We are going to start lowering to this in NIR code, so prepare radv for it. v2: handle conversion to kilp properly (nha) Signed-off-by: Dave Airlie --- src/amd/common/ac_nir_to_llvm.c | 21 + 1 file changed, 21

[Mesa-dev] [PATCH 3/3] radv: enable conditional discard optimisation on radv.

From: Dave Airlie This fixes a bunch of GPU hangs introduced in some CTS tests like dEQP-VK.memory.pipeline_barrier.host_write_uniform_buffer.65536 It works around an issue seen in the LLVM backend, but also makes the radv code work more like the radeonsi stack.

[Mesa-dev] [PATCH 2/3] nir: add conditional discard optimisation (v3)

From: Dave Airlie This is ported from GLSL and converts if (cond) discard; into discard_if(cond); This removes a block, but also is needed by radv to workaround a bug in the LLVM backend. v2: handle if (a) discard_if(b) (nha) cleanup and drop pointless loop (Matt)

[Mesa-dev] [PATCH] radv: emit correct last export when Z/stencil export is enabled

From: Dave Airlie I was getting a random GPU hang in the renderpass simple tests, it turns out sometimes radv emitted the wrong thing "last". This fixes the logic to emit Z/stencil last if they occur, and not mark a color output as last. Also this relies on the Z/STENCIL

[Mesa-dev] [PATCH 1/4] linker: Trivial coding standards fixes

From: Ian Romanick Signed-off-by: Ian Romanick --- src/compiler/glsl/link_uniforms.cpp | 30 ++ 1 file changed, 14 insertions(+), 16 deletions(-) diff --git a/src/compiler/glsl/link_uniforms.cpp

[Mesa-dev] [PATCH 2/4] glsl: Add some comments to methods of ir_variable_refcount_visitor

From: Ian Romanick It was not obvious from the just the .h file what the hash table contained. It was also not obvious that get_variable_entry would create a new entry in the hash table. Signed-off-by: Ian Romanick ---

[Mesa-dev] [PATCH 3/4] linker: Slight code rearrange to prevent duplication in the next commit

From: Ian Romanick Signed-off-by: Ian Romanick --- src/compiler/glsl/link_uniforms.cpp | 13 ++--- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/src/compiler/glsl/link_uniforms.cpp

[Mesa-dev] [PATCH 4/4] linker: Accurately track gl_uniform_block::stageref

From: Ian Romanick As the linked per-stage shaders are processed, mark any block that has a field that is accessed as referenced. When combining all the linked shaders, combine the per-stage stageref masks. This fixes a number of GLES CTS tests including

Re: [Mesa-dev] [PATCH 07/13] mesa: Replace program locks with atomic inc/dec.

There are still some issues with the other patches but is there any reason this one didn't land? On Thu, 2015-08-06 at 17:10 -0700, Matt Turner wrote: > --- >  src/mesa/main/mtypes.h |  1 - >  src/mesa/program/program.c | 15 +++ >  2 files changed, 3 insertions(+), 13 deletions(-)

Re: [Mesa-dev] [PATCH 2/2] Make shader refcounting atomic.

On Sat, 2016-11-05 at 15:37 +0100, Steinar H. Gunderson wrote: > These were racy when using the same shaders (seemingly even from > different > program objects) on multiple theads sharing the same objects, leading > to > issues such as (excerpts from an apitrace dump from a real > application): >

[Mesa-dev] [PATCH] llvmpipe: Fix build after removal of deprecated attribute API

Applies on top of v2 of Tom's gallivm change. Signed-off-by: Aaron Watry CC: Tom Stellard CC: Jan Vesely --- This fixes the build for me. I haven't done more than compile test this and run make check.

[Mesa-dev] [PATCH v4 08/10] anv/batch: Move last_ss_pool_bo_offset to the command buffer

The original reason for putting it in the batch_bo was to allow primaries to share it across secondaries or something like that. However, the relocation lists in secondary command buffers are are always left alone and copied into the primary command buffer's relocation list. This means that the

[Mesa-dev] [PATCH 2/3] anv/device: Return the right error for failed maps

Signed-off-by: Jason Ekstrand Cc: "12.0 13.0" --- src/intel/vulkan/anv_device.c | 9 +++-- src/intel/vulkan/anv_gem.c| 6 ++ 2 files changed, 9 insertions(+), 6 deletions(-) diff --git a/src/intel/vulkan/anv_device.c

[Mesa-dev] [PATCH 3/3] anv/device: Implicitly unmap memory objects in FreeMemory

From the Vulkan spec version 1.0.32 docs for vkFreeMemory: "If a memory object is mapped at the time it is freed, it is implicitly unmapped." Signed-off-by: Jason Ekstrand Cc: "12.0 13.0" --- src/intel/vulkan/anv_device.c | 9

[Mesa-dev] [PATCH 1/3] anv/device: Don't even try to map memory with a size of 0

Signed-off-by: Jason Ekstrand Cc: "12.0 13.0" --- src/intel/vulkan/anv_device.c | 5 + 1 file changed, 5 insertions(+) diff --git a/src/intel/vulkan/anv_device.c b/src/intel/vulkan/anv_device.c index 5393144..8055893 100644 ---

[Mesa-dev] [Bug 98002] Mud rendering bug in Portal 2

https://bugs.freedesktop.org/show_bug.cgi?id=98002 Michel Dänzer changed: What|Removed |Added Resolution|--- |NOTOURBUG

Re: [Mesa-dev] [RFC 03/12] egl: add EGL_ANDROID_native_fence_sync

On Mon, Nov 7, 2016 at 6:29 PM, Rafael Antognolli wrote: > On Mon, Oct 31, 2016 at 08:58:26AM -0700, Rafael Antognolli wrote: >> On Sat, Oct 29, 2016 at 01:15:44PM -0400, Rob Clark wrote: >> > On Fri, Oct 28, 2016 at 7:44 PM, Rafael Antognolli >> >

[Mesa-dev] [PATCH 1/3] swr: fix AND_INVERTED logic op conversion

Signed-off-by: Ilia Mirkin --- src/gallium/drivers/swr/swr_state.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/drivers/swr/swr_state.h b/src/gallium/drivers/swr/swr_state.h index 0e3b49d..8409114 100644 ---

[Mesa-dev] [PATCH 2/3] swr: disable logic op when the rt format is float

Signed-off-by: Ilia Mirkin --- src/gallium/drivers/swr/swr_state.cpp | 5 + 1 file changed, 5 insertions(+) diff --git a/src/gallium/drivers/swr/swr_state.cpp b/src/gallium/drivers/swr/swr_state.cpp index d8a8ee1..acb0452 100644 ---

[Mesa-dev] [PATCH 3/3] swr: [rasterizer jitter] fix logic op to work with unorm/snorm

Most logic op usage is probably going to end up with normalized textures. Scale the floating point values and convert to integer before performing the logic operations. Signed-off-by: Ilia Mirkin --- The gl-1.1-xor-copypixels test still fails. The image stays the same. I'm

Re: [Mesa-dev] [RFC 03/12] egl: add EGL_ANDROID_native_fence_sync

On Mon, Oct 31, 2016 at 08:58:26AM -0700, Rafael Antognolli wrote: > On Sat, Oct 29, 2016 at 01:15:44PM -0400, Rob Clark wrote: > > On Fri, Oct 28, 2016 at 7:44 PM, Rafael Antognolli > > wrote: ... > > Hey, thanks for this. I don't suppose you have a branch

[Mesa-dev] [PATCH v4 02/10] anv: Don't presume to know what address is in a surface relocation

Because our relocation processing happens at EndCommandBuffer time and because RENDER_SURFACE_STATE objects may be shared by batches, we really have no clue whatsoever what address is actually written to the relocation offset in the BO. We need to stop making such claims to the kernel and just

[Mesa-dev] [PATCH 1/3] gallium/scons: OpenSWR Windows support

- Added code to create screen and handle swaps in libgl_gdi.c - Added call to swr SConscript - included llvm 3.9 support for scons (windows swr only support 3.9 and later) - include -DHAVE_SWR to subdirs that need it To buils SWR on windows, use "scons swr libgl-gdi" --- scons/llvm.py

[Mesa-dev] [Bug 98599] xterm menus corrupt since tgsi/scan: handle indirect image indexing correctly

https://bugs.freedesktop.org/show_bug.cgi?id=98599 --- Comment #6 from Andy Furniss --- The patches on mesa-dev fix for me, thanks. -- You are receiving this mail because: You are the assignee for the bug. You are the QA Contact for the

[Mesa-dev] [PATCH 0/3] swr: Support Windows builds

Changes to support windows builds for OpenSWR driver. These are divided into 3 patches: - scons and core mesa-related changes - a fix in macros.h to implement HAS_TRIVIAL_DESTRUCTOR - swr-specific changes The way to build SWR on windows is using scons. Build using the following command line:

[Mesa-dev] [PATCH 2/3] mesa: added msvc HAS_TRIVIAL_DESTRUCTOR implementation

not having it on windows causes a CANARY assertion in src/util/ralloc.c:get_header() Tested only on MSVC 19.00 (DevStudio 14.0), so #ifdef guards reflect that. --- src/util/macros.h | 5 + 1 file changed, 5 insertions(+) diff --git a/src/util/macros.h b/src/util/macros.h index

[Mesa-dev] [PATCH 3/3] swr: Support windows builds

- Added SConscript files - better handling of NOMINMAX for inclusion - Reorder header files in swr_context.cpp to handle NOMINMAX better, since mesa header files include windows.h before we get a chance to #define NOMINMAX - cleaner support for .dll and .so prefix/suffix across OSes - added

[Mesa-dev] [PATCH v4 06/10] anv/batch_chain: Improve write_reloc

The old version wasn't properly handling large addresses where we have to sign-extend to get it into the "canonical form" expected by the hardware. Also, the new version is capable of doing a clflush of the newly written reloc if requested. Signed-off-by: Jason Ekstrand ---

[Mesa-dev] [PATCH v4 10/10] anv: Do relocations in userspace before execbuf ioctl

From: Kristian Høgsberg Kristensen Since our surface state buffer is shared by all batches, the kernel does a full stall and sync with the CPU between batches every time we call execbuf2 because it refuses to do relocations on an active buffer. Doing them in

[Mesa-dev] [PATCH v4 09/10] anv: Move relocation handling from EndCommandBuffer to QueueSubmit

Ever since the early days of the Vulkan driver, we've been setting up the lists of relocations at EndCommandBuffer time. The idea behind this was to move some of the CPU load out of QueueSubmit which the client is required to lock around and into command buffer building which could be done in

[Mesa-dev] [PATCH v4 07/10] anv: Add an anv_execbuf helper struct

This commit adds a little helper struct for storing everything we use to build an execbuf2 call. Since the add_bo function really has nothing to do with a command buffer, it makes sense to break it out a bit. This also reduces some of the churn in the next commit. ---

[Mesa-dev] [PATCH v4 00/10] anv: Rework relocation handling

This is the fourth iteration of my attempt to rework relocation handling and do relocations in userspace. I'm finally getting pretty happy with this and I think I'll probably merge this version if there are no further objections. Jason Ekstrand (9): anv: Add a cmd_buffer_execbuf helper anv:

[Mesa-dev] [PATCH v4 04/10] anv/allocator: Simplify anv_scratch_pool

The previous implementation was being overly clever and using the anv_bo::size field as its mutex. Scratch pool allocations don't happen often, will happen at most a fixed number of times, and never happen in the critical path (they only happen in shader compilation). We can make this much

[Mesa-dev] [PATCH v4 05/10] anv: Initialize anv_bo::offset to -1

Since -1 is an invalid GPU address, this lets us know whether or not we have a valid address for a buffer. We don't get a valid address until the first time that buffer is used in an execbuf2 ioctl. Signed-off-by: Jason Ekstrand --- src/intel/vulkan/anv_private.h | 2 +-

[Mesa-dev] [PATCH v4 03/10] anv: Add a new bo_pool_init helper

This ensures that we're always setting all of the fields in anv_bo Signed-off-by: Jason Ekstrand --- src/intel/vulkan/anv_allocator.c | 9 ++--- src/intel/vulkan/anv_device.c| 10 +++--- src/intel/vulkan/anv_intel.c | 11 +--

[Mesa-dev] [PATCH v4 01/10] anv: Add a cmd_buffer_execbuf helper

This puts the actual execbuf2 call in anv_batch_chain.c along with the other relocation stuff. --- src/intel/vulkan/anv_batch_chain.c | 8 src/intel/vulkan/anv_device.c | 3 +-- src/intel/vulkan/anv_private.h | 2 ++ 3 files changed, 11 insertions(+), 2 deletions(-) diff --git

Re: [Mesa-dev] [PATCH 2/2] i965: Advertise 8 subpixel bits always.

Chris Forbes writes: > Hi Curro, > > Thanks for being thorough about this -- I think there is still one area > where things might be a bit wobbly; if we end up taking a sw fallback, > swrast only does 4 bits. I'm not sure that matters though. > Hmm, I don't think we fall

Re: [Mesa-dev] [PATCH v2] nvc0: only invalidate currently bound tic/tsc

This could be still improved by adding textures/samplers_valid[6] into the context. On 11/07/2016 11:13 PM, Samuel Pitoiset wrote: This is especially useful when switching from compute to 3D. v2: - get rid of one loop with 'x |= (1ULL << y) - 1' instead Signed-off-by: Samuel Pitoiset

[Mesa-dev] [PATCH v2] nvc0: only invalidate currently bound tic/tsc

This is especially useful when switching from compute to 3D. v2: - get rid of one loop with 'x |= (1ULL << y) - 1' instead Signed-off-by: Samuel Pitoiset --- Tested with Elemental on GK208, works fine. src/gallium/drivers/nouveau/nvc0/nvc0_compute.c | 6 +++---

Re: [Mesa-dev] [PATCH 2/2] i965: Advertise 8 subpixel bits always.

Chris Forbes writes: > The mesa default is 4, but we program the hardware for 8 on all > generations. > I happened to come across this inconsistency a couple of weeks ago -- I just double-checked that it doesn't cause any conformance regressions because some of the

Re: [Mesa-dev] [PATCH] gallivm: Fix build after removal of deprecated attribute API v2

On Mon, 2016-11-07 at 21:06 +, Tom Stellard wrote: > v2: > Fix adding parameter attributes with LLVM < 4.0. > --- > src/gallium/auxiliary/draw/draw_llvm.c| 6 +- > src/gallium/auxiliary/gallivm/lp_bld_intr.c | 52 - >

Re: [Mesa-dev] [PATCH 1/2] gallivm: introduce 32x32->64bit lp_build_mul_32_lohi function

Am 07.11.2016 um 22:34 schrieb Jose Fonseca: > On 07/11/16 19:09, Roland Scheidegger wrote: >> Am 06.11.2016 um 16:50 schrieb Jose Fonseca: >>> On 04/11/16 04:14, srol...@vmware.com wrote: From: Roland Scheidegger This is used by shader umul_hi/imul_hi functions

Re: [Mesa-dev] [PATCH 1/2] gallivm: introduce 32x32->64bit lp_build_mul_32_lohi function

On 07/11/16 19:09, Roland Scheidegger wrote: Am 06.11.2016 um 16:50 schrieb Jose Fonseca: On 04/11/16 04:14, srol...@vmware.com wrote: From: Roland Scheidegger This is used by shader umul_hi/imul_hi functions (and soon by draw). It's actually useful separating this out on

Re: [Mesa-dev] [PATCH] swr: allow alphatest without blend or logicop

Reviewed-by: Bruce Cherniak > On Nov 7, 2016, at 1:23 PM, Tim Rowley wrote: > > We need to compile a blend function when alphatest is enabled. > --- > src/gallium/drivers/swr/swr_state.cpp | 3 ++- > 1 file changed, 2 insertions(+), 1

[Mesa-dev] [Bug 98632] Fix build on Hurd without PATH_MAX

https://bugs.freedesktop.org/show_bug.cgi?id=98632 Bug ID: 98632 Summary: Fix build on Hurd without PATH_MAX Product: Mesa Version: 13.0 Hardware: Other OS: other Status: NEW Severity: normal

[Mesa-dev] [PATCH] gallivm: Fix build after removal of deprecated attribute API v2

v2: Fix adding parameter attributes with LLVM < 4.0. --- src/gallium/auxiliary/draw/draw_llvm.c| 6 +- src/gallium/auxiliary/gallivm/lp_bld_intr.c | 52 - src/gallium/auxiliary/gallivm/lp_bld_intr.h | 13 -

[Mesa-dev] [PATCH] anv: Document cmd_buffer_alloc_binding_table

Some of the details of this function are very confusing and have a long history. We should document that history and this seems like the best place to do it. Signed-off-by: Jason Ekstrand --- src/intel/vulkan/anv_batch_chain.c | 71 ++

[Mesa-dev] [PATCH] nir: Avoid an extra NIR op in integer divide lowering.

NIR bools are ~0 for true, so ((unsigned)a >> 31) != 0 -> ((int)a >> 31). --- src/compiler/nir/nir_lower_idiv.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/compiler/nir/nir_lower_idiv.c b/src/compiler/nir/nir_lower_idiv.c index b1e7aeb03c8a..6726b718aaa5 100644 ---

Re: [Mesa-dev] [PATCH 3/7] nvc0: simplify draw parameters upload for vertex shaders

On 11/07/2016 04:36 AM, Ilia Mirkin wrote: On Tue, Oct 25, 2016 at 3:41 PM, Samuel Pitoiset wrote: Signed-off-by: Samuel Pitoiset --- src/gallium/drivers/nouveau/nvc0/nvc0_vbo.c | 14 ++ 1 file changed, 6 insertions(+), 8

Re: [Mesa-dev] [PATCH 2/7] nvc0: only update primitive restart for indexed draws

On 11/07/2016 04:34 AM, Ilia Mirkin wrote: Primitive restart is a thing for non-indexed draws too. There's a method that controls it - NVC0_3D_PRIM_RESTART_WITH_DRAW_ARRAYS. The more recently GL 4.5 behaviour is to *not* do primitive restart for draw arrays, however the older behavior is to do

Re: [Mesa-dev] [PATCH 1/7] nvc0: reduce the number of PUSH_SPACE in draw path

On 11/07/2016 04:32 AM, Ilia Mirkin wrote: On Wed, Oct 26, 2016 at 4:14 AM, Samuel Pitoiset wrote: On 10/25/2016 09:49 PM, Ilia Mirkin wrote: What if instance_count = 1M? (It can happen.) We allocate a giant space in the pushbuf in one shot. Well, anyways

Re: [Mesa-dev] [PATCH 00/25] anv: A major rework of color attachment surface states

On Thu, Oct 27, 2016 at 10:08:33AM +0300, Pohjolainen, Topi wrote: > On Wed, Oct 26, 2016 at 07:11:20PM +0300, Pohjolainen, Topi wrote: > > > > I had a few qeustions in 1 and 3, but regardless patches 1-7 and 9-13 are: > > > > Reviewed-by: Topi Pohjolainen > > 14-17

Re: [Mesa-dev] [PATCH 2/2] i965: Advertise 8 subpixel bits always.

On Sun, Nov 6, 2016 at 10:45 PM, Chris Forbes wrote: > The mesa default is 4, but we program the hardware for 8 on all > generations. > > Signed-off-by: Chris Forbes > --- > src/mesa/drivers/dri/i965/brw_context.c | 1 + > 1 file changed, 1

Re: [Mesa-dev] [PATCH] gallivm: Fix build after removal of deprecated attribute API

On Mon, 2016-11-07 at 18:44 +, Tom Stellard wrote: > --- > > Build tested only so far. > > src/gallium/auxiliary/draw/draw_llvm.c| 6 +- > src/gallium/auxiliary/gallivm/lp_bld_intr.c | 48 +++- > src/gallium/auxiliary/gallivm/lp_bld_intr.h | 13 - >

Re: [Mesa-dev] [PATCH 24/25] Allocate a null state whenever there is depth/stencil

On Sat, Oct 22, 2016 at 10:50:55AM -0700, Jason Ekstrand wrote: > --- > src/intel/vulkan/genX_cmd_buffer.c | 19 ++- > 1 file changed, 10 insertions(+), 9 deletions(-) Reviewed-by: Topi Pohjolainen > > diff --git a/src/intel/vulkan/genX_cmd_buffer.c

Re: [Mesa-dev] [PATCH 22/25] anv/blorp: Use the new clear_attachments entrypoint for attachment clears

On Sat, Oct 22, 2016 at 10:50:53AM -0700, Jason Ekstrand wrote: > This allows us to re-use the surface states emitted from the Vulkan driver > instead of blorp creating its own. > --- > src/intel/vulkan/anv_blorp.c | 93 > +--- > 1 file changed, 52

[Mesa-dev] [PATCH] swr: allow alphatest without blend or logicop

We need to compile a blend function when alphatest is enabled. --- src/gallium/drivers/swr/swr_state.cpp | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/swr/swr_state.cpp b/src/gallium/drivers/swr/swr_state.cpp index 3e02322..424bff2 100644 ---

Re: [Mesa-dev] [PATCH 1/2] gallivm: introduce 32x32->64bit lp_build_mul_32_lohi function

Am 06.11.2016 um 16:50 schrieb Jose Fonseca: > On 04/11/16 04:14, srol...@vmware.com wrote: >> From: Roland Scheidegger >> >> This is used by shader umul_hi/imul_hi functions (and soon by draw). >> It's actually useful separating this out on its own, however the real >> reason

[Mesa-dev] [PATCH] gallivm: Fix build after removal of deprecated attribute API

--- Build tested only so far. src/gallium/auxiliary/draw/draw_llvm.c| 6 +- src/gallium/auxiliary/gallivm/lp_bld_intr.c | 48 +++- src/gallium/auxiliary/gallivm/lp_bld_intr.h | 13 - src/gallium/auxiliary/gallivm/lp_bld_sample_soa.c | 4 +-

Re: [Mesa-dev] [PATCH 3/3] nvc0: refactor textures/samplers validation

On 11/07/2016 04:30 AM, Ilia Mirkin wrote: Patches 1-2 seem OK. I'm a little concerned that this one is changing functionality, since it's removing the "need_flush" thing. It'd be nice if you could get this patch some heavier testing before pushing it out... Yes, it's more fine-grained

Re: [Mesa-dev] [PATCH 1/3] gallium/hud: fix a problem where objects are free'd while in use.

On Mon, Nov 7, 2016 at 12:32 PM, Nicolai Hähnle wrote: > Looks good to me as well, and pushed! Thanks for the respin and sorry it > took so long. You are very welcome. No apology necessary. Thank you for your due diligence and prior feedback. -- Steven Toth - Kernel Labs

Re: [Mesa-dev] [PATCH 1/3] gallium/hud: fix a problem where objects are free'd while in use.

On Mon, Nov 7, 2016 at 12:37 PM, Laurent Carlier wrote: > Le lundi 7 novembre 2016, 18:32:07 CET Nicolai Hähnle a écrit : >> Looks good to me as well, and pushed! Thanks for the respin and sorry it >> took so long. >> >> Cheers, >> Nicolai >> > > Maybe cc 13.0 ? It's buggy

Re: [Mesa-dev] [PATCH 1/3] gallium/hud: fix a problem where objects are free'd while in use.

Le lundi 7 novembre 2016, 18:32:07 CET Nicolai Hähnle a écrit : > Looks good to me as well, and pushed! Thanks for the respin and sorry it > took so long. > > Cheers, > Nicolai > Maybe cc 13.0 ? It's buggy with 13.0 and it will be a nice fix -- Laurent Carlier http://www.archlinux.org

Re: [Mesa-dev] [PATCH 1/3] gallium/hud: fix a problem where objects are free'd while in use.

Looks good to me as well, and pushed! Thanks for the respin and sorry it took so long. Cheers, Nicolai On 24.10.2016 16:10, Steven Toth wrote: Instead of trying to maintain a reference counted list of valid HUD objects, and freeing them accordingly, creating race conditions between

Re: [Mesa-dev] [PATCH 4/8] i965: Store mcs buffer size

On 16-11-07 10:22:46, Lionel Landwerlin wrote: On 07/11/16 10:07, Pohjolainen, Topi wrote: On Thu, Nov 03, 2016 at 10:39:39AM +, Lionel Landwerlin wrote: From: Ben Widawsky libdrm may round up the allocation requested by mesa. As a result, accesses through

[Mesa-dev] [Bug 98555] NWN won't start on 13.0

https://bugs.freedesktop.org/show_bug.cgi?id=98555 Emil Velikov changed: What|Removed |Added Status|NEW |RESOLVED

Re: [Mesa-dev] [PATCH 3/3] gallium/hud: protect against and initialization race

>> A humble ping on this and the two others. >> >> - Steve >> > > Series looks OK to me. > > Reviewed-by: Brian Paul > > Need me to push these for you? > > -Brian > That would be a helpful, yes please. I think they fell through the cracks after the last round of reviews.

Re: [Mesa-dev] [PATCH 1/2] swr: [rasterizer core]: set depth hottile when depth bounds test enabled

Reviewed-by: Tim Rowley > On Nov 1, 2016, at 3:45 PM, Ilia Mirkin > wrote: Signed-off-by: Ilia Mirkin > ---

Re: [Mesa-dev] [PATCH 2/2] swr: add support for EXT_depth_bounds_test

We suspect the remaining failure might be due to not quantizing the depth bounds min/max values. That can be addressed in a future patch. Reviewed-by: Tim Rowley > On Nov 1, 2016, at 3:45 PM, Ilia Mirkin

Re: [Mesa-dev] [PATCH v3] vl/dri3: use external texture as back buffers(v3)

On 11/07/2016 11:31 AM, Nayan Deshmukh wrote: On Mon, Nov 7, 2016 at 8:31 PM, Leo Liu > wrote: On 11/05/2016 02:44 AM, Nayan Deshmukh wrote: Hi Leo, Thanks for the reference patch. There are only a number of

Re: [Mesa-dev] [PATCH v3] vl/dri3: use external texture as back buffers(v3)

On Mon, Nov 7, 2016 at 8:31 PM, Leo Liu wrote: > > > On 11/05/2016 02:44 AM, Nayan Deshmukh wrote: > >> Hi Leo, >> >> Thanks for the reference patch. >> >> There are only a number of output surfaces taking turns as the >> mixer render targets, so we probably can use the

[Mesa-dev] [Bug 98555] NWN won't start on 13.0

https://bugs.freedesktop.org/show_bug.cgi?id=98555 --- Comment #7 from Alan Swanson --- Might be that __nwu_possible return value isn't checked for the opendir override. Will have a look in a day or two but unlikely to be a mesa bug. Unfortunately NWN was

[Mesa-dev] [Bug 98555] NWN won't start on 13.0

https://bugs.freedesktop.org/show_bug.cgi?id=98555 --- Comment #6 from Laurent carlier --- Just tried with my own copy of nwn (resurected from my hard drive) and it starts perfectly (archlinux/mesa-git/llvm-svn) with radeonsi -- You are receiving this mail because: You

Re: [Mesa-dev] [PATCH 3/3] gallium/hud: protect against and initialization race

On 11/07/2016 06:00 AM, Steven Toth wrote: On Mon, Oct 24, 2016 at 10:10 AM, Steven Toth wrote: In the event that multiple threads attempt to install a graph concurrently, protect the shared list. Signed-off-by: Steven Toth A humble ping on this

[Mesa-dev] [Bug 98563] Xorg segfaults with displaylink attached and mesa version >= 13.0

https://bugs.freedesktop.org/show_bug.cgi?id=98563 Lorenzo Bona changed: What|Removed |Added CC|lorenz.b...@gmail.com | -- You are

[Mesa-dev] [Bug 98563] Xorg segfaults with displaylink attached and mesa version >= 13.0

https://bugs.freedesktop.org/show_bug.cgi?id=98563 --- Comment #6 from Emil Velikov --- Andrew, the affected/new codepaths shouldn't do anything that causes such behaviour. Thus I'm inclined that this a separate bug. Can you please check/bisect the offending mesa

Re: [Mesa-dev] [PATCH v3] vl/dri3: use external texture as back buffers(v3)

On 11/05/2016 02:44 AM, Nayan Deshmukh wrote: Hi Leo, Thanks for the reference patch. There are only a number of output surfaces taking turns as the mixer render targets, so we probably can use the same pixmap corresponding to each of output surface texture. The mixer renders to

Re: [Mesa-dev] [PATCH] android: amd/common: add support for libmesa_amd_common

On 5 November 2016 at 21:57, Marek Olšák wrote: > On Nov 5, 2016 9:03 PM, "Mauro Rossi" wrote: >> >> 2016-11-05 18:58 GMT+01:00 Marek Olšák : >> > Hi, >> > >> > I pushed the patch from the bug report. Hopefully it's the same as this >> >

Re: [Mesa-dev] [PATCH] android: amd/common: add support for libmesa_amd_common

On 5 November 2016 at 12:21, Mauro Rossi wrote: > Hi Nicolai, > > please review the attached patch which is necessary fix android build, > as per https://bugs.freedesktop.org/show_bug.cgi?id=98573 > > Tested with nougat-x86 build > Kind regards > > Mauro > > > From

Re: [Mesa-dev] [PATCH 1/3] glsl: use the prefixed name of the lexer generated functions

On 3 November 2016 at 22:42, Timothy Arceri wrote: > On Fri, 2016-11-04 at 08:56 +1100, Timothy Arceri wrote: >> On Thu, 2016-10-27 at 12:41 +0100, Emil Velikov wrote: >> > >> > From: Emil Velikov >> > >> > Flex version 2.6.2 does not

[Mesa-dev] [Bug 98555] NWN won't start on 13.0

https://bugs.freedesktop.org/show_bug.cgi?id=98555 --- Comment #5 from Emil Velikov --- s/good _not_/_not_ good/ of course. Or to put it otherwise - __nwu_possible() [1] should attribute only for files/directories which are known used by NWM. Everything else must

Re: [Mesa-dev] [PATCH 3/3] gallium/hud: protect against and initialization race

On Mon, Oct 24, 2016 at 10:10 AM, Steven Toth wrote: > In the event that multiple threads attempt to install a graph > concurrently, protect the shared list. > > Signed-off-by: Steven Toth A humble ping on this and the two others. - Steve --

Re: [Mesa-dev] [PATCH 2/3] gallium/hud: close a previously opened handle

On Mon, Oct 24, 2016 at 10:10 AM, Steven Toth wrote: > We're missing the closedir() to the matching opendir(). > > Signed-off-by: Steven Toth A humble ping on this and the two others. - Steve -- Steven Toth - Kernel Labs http://www.kernellabs.com

Re: [Mesa-dev] [PATCH 1/3] gallium/hud: fix a problem where objects are free'd while in use.

On Mon, Oct 24, 2016 at 10:10 AM, Steven Toth wrote: > Instead of trying to maintain a reference counted list of valid HUD > objects, and freeing them accordingly, creating race conditions > between unanticipated multiple threads, simply accept they're > allocated once and

Re: [Mesa-dev] [PATCH] glsl: Do not allow scalar types in vector relational functions

2016-11-05 3:23 GMT+08:00 Matt Turner : > On Sun, Oct 30, 2016 at 11:45 PM, Boyan Ding wrote: >> According to OpenGL Shading Language 4.50 spec, Section 8.7 "Vector >> Relational Functions", functions of this type do not operate on scalar >> types, so

Re: [Mesa-dev] [PATCH 8/8] i965/gen9: Allow sampling with hiz when supported

On Thu, Nov 03, 2016 at 10:39:43AM +, Lionel Landwerlin wrote: > From: Jordan Justen > > For gen9+ this will indicate when we should allow hiz based sampling > during rendering. > > Improves performance in : > - Synmark's OglDeferred by 2.2% (n=20) > -

[Mesa-dev] [Bug 98555] NWN won't start on 13.0

https://bugs.freedesktop.org/show_bug.cgi?id=98555 --- Comment #4 from Emil Velikov --- The most likely thing that comes to mind is that one LD_PRELOADs a bunch of libraries, amongst which an older version of libdrm. What's the output of $LD_DEBUG=libs nwn The above

Re: [Mesa-dev] [PATCH 2/2] tgsi/scan: turn a huge if-else-if.. chain into a switch statement

On 05.11.2016 18:38, Marek Olšák wrote: From: Marek Olšák For the series: Reviewed-by: Nicolai Hähnle --- src/gallium/auxiliary/tgsi/tgsi_scan.c | 44 +++--- 1 file changed, 30 insertions(+), 14 deletions(-) diff

[Mesa-dev] [Bug 98555] NWN won't start on 13.0

https://bugs.freedesktop.org/show_bug.cgi?id=98555 --- Comment #3 from Nicolai Hähnle --- amdgpu_dri.so is the name of the amdgpu-pro (closed-source) OpenGL driver. Since you seem to be trying to use Mesa, please make sure to remove the amdgpu-pro installation. -- You are

Re: [Mesa-dev] [PATCH 5/8] i965/miptree: Create a hiz mcs type

On Mon, Nov 07, 2016 at 12:30:14PM +0200, Pohjolainen, Topi wrote: > On Mon, Nov 07, 2016 at 10:24:35AM +, Lionel Landwerlin wrote: > > On 07/11/16 10:18, Pohjolainen, Topi wrote: > > > On Thu, Nov 03, 2016 at 10:39:40AM +, Lionel Landwerlin wrote: > > > > From: Ben Widawsky

[Mesa-dev] [Bug 98002] Mud rendering bug in Portal 2

https://bugs.freedesktop.org/show_bug.cgi?id=98002 --- Comment #13 from almos --- Ok, I managed to download the trace with Konqueror. I haven't tried it with fglrx yet, but I compared the attached screenshot with the YouTube video, and here are my thoughts. Actually, the

[Mesa-dev] [PATCH] main: return error if asking for GL_TEXTURE_BORDER_COLOR in TEXTURE_2D_MULTISAMPLE{_ARRAY} through TexParameterI{i, ui}v()

OpenGL ES 3.2 says in section 8.10. "TEXTURE PARAMETERS", at the end of the section: "An INVALID_ENUM error is generated if target is TEXTURE_2D_- MULTISAMPLE or TEXTURE_2D_MULTISAMPLE_ARRAY , and pname is any sampler state from table 21.12." GL_TEXTURE_BORDER_COLOR is present in that table.

Re: [Mesa-dev] [PATCH 3/3] dri: fixup driver names if needed

2016-11-04 12:59 GMT+01:00 Eric Engestrom : > On Thursday, 2016-11-03 15:25:22 +0100, Christian Gmeiner wrote: >> This makes it possible to 'use' the imx-drm driver. Remeber that it >> is not possible to have sysmbol names in C/C++ with a '-' in it. >> >> Signed-off-by:

Re: [Mesa-dev] [PATCH 5/8] i965/miptree: Create a hiz mcs type

On Mon, Nov 07, 2016 at 10:24:35AM +, Lionel Landwerlin wrote: > On 07/11/16 10:18, Pohjolainen, Topi wrote: > > On Thu, Nov 03, 2016 at 10:39:40AM +, Lionel Landwerlin wrote: > > > From: Ben Widawsky > > > > > > This seems counter to the goal of

Re: [Mesa-dev] [PATCH 5/8] i965/miptree: Create a hiz mcs type

On 07/11/16 10:18, Pohjolainen, Topi wrote: On Thu, Nov 03, 2016 at 10:39:40AM +, Lionel Landwerlin wrote: From: Ben Widawsky This seems counter to the goal of consolidating hiz, mcs, and later ccs buffers. Unfortunately, hiz on gen6 is a thing the code

Re: [Mesa-dev] [PATCH 4/8] i965: Store mcs buffer size

On 07/11/16 10:07, Pohjolainen, Topi wrote: On Thu, Nov 03, 2016 at 10:39:39AM +, Lionel Landwerlin wrote: From: Ben Widawsky libdrm may round up the allocation requested by mesa. As a result, accesses through the gtt may end up accessing memory which does not

Re: [Mesa-dev] [PATCH 5/8] i965/miptree: Create a hiz mcs type

On Thu, Nov 03, 2016 at 10:39:40AM +, Lionel Landwerlin wrote: > From: Ben Widawsky > > This seems counter to the goal of consolidating hiz, mcs, and later ccs > buffers. > Unfortunately, hiz on gen6 is a thing the code supports, and this wart will be > helpful

Re: [Mesa-dev] [PATCH 3/8] i965: Drop the aux mt when not used

On Mon, Nov 07, 2016 at 11:38:09AM +0200, Pohjolainen, Topi wrote: > On Thu, Nov 03, 2016 at 10:39:38AM +, Lionel Landwerlin wrote: > > From: Ben Widawsky > > > > This patch will preserve the BO & offset, and not the miptree for the > > aux_mcs buffer. Eventually

  1   2   >