Kristian Høgsberg [Sat, 9 May 2015 05:32:37 +0000 (22:32 -0700)]
Add vulkan driver for BDW
Kenneth Graunke [Sat, 9 May 2015 02:36:19 +0000 (19:36 -0700)]
nir: Fix aggressive typos in nir_from_ssa.c.
s/agressive/aggressive/g
Trivial.
Jason Ekstrand [Fri, 8 May 2015 16:48:33 +0000 (09:48 -0700)]
nir/search: Save/restore the variables_seen bitmask when matching
Shader-db results on Broadwell:
total instructions in shared programs:
7152330 ->
7137006 (-0.21%)
instructions in affected programs:
1330548 ->
1315224 (-1.15%)
helped: 5797
HURT: 76
GAINED: 0
LOST: 8
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Fri, 8 May 2015 16:42:05 +0000 (09:42 -0700)]
nir/search: Assert that variable id's are in range
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Fri, 8 May 2015 15:33:01 +0000 (08:33 -0700)]
nir/search: handle explicitly sized sources in match_value
Previously, this case was being handled in match_expression prior to
calling match_value. However, there is really no good reason for this
given that match_value has all of the information it needs. Also, they
weren't being handled properly in the commutative case and putting it in
match_value gives us that for free.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Fri, 24 Apr 2015 17:16:27 +0000 (10:16 -0700)]
nir/nir: Use a linked list instead of a hash set for use/def sets
This commit switches us from the current setup of using hash sets for
use/def sets to using linked lists. Doing so should save us quite a bit of
memory because we aren't carrying around 3 hash sets per register and 2 per
SSA value. It should also save us CPU time because adding/removing things
from use/def sets is 4 pointer manipulations instead of a hash lookup.
Running shader-db 50 times with USE_NIR=0, NIR, and NIR + use/def lists:
GLSL IR Only: 586.4 +/- 1.653833
NIR with hash sets: 675.4 +/- 2.502108
NIR + use/def lists: 641.2 +/- 1.557043
I also ran a memory usage experiment with Ken's patch to delete GLSL IR and
keep NIR. This patch cuts an aditional 42.9 MiB of ralloc'd memory over
and above what we gained by deleting the GLSL IR on the same dota trace.
On the code complexity side of things, some things are now much easier and
others are a bit harder. One of the operations we perform constantly in
optimization passes is to replace one source with another. Due to the fact
that an instruction can use the same SSA value multiple times, we had to
iterate through the sources of the instruction and determine if the use we
were replacing was the only one before removing it from the set of uses.
With this patch, uses are per-source not per-instruction so we can just
remove it safely. On the other hand, trying to iterate over all of the
instructions that use a given value is more difficult. Fortunately, the
two places we do that are the ffma peephole where it doesn't matter and GCM
where we already gracefully handle duplicates visits to an instruction.
Another aspect here is that using linked lists in this way can be tricky to
get right. With sets, things were quite forgiving and the worst that
happened if you didn't properly remove a use was that it would get caught
in the validator. With linked lists, it can lead to linked list corruption
which can be harder to track. However, we do just as much validation of
the linked lists as we did of the sets so the validator should still catch
these problems. While working on this series, the vast majority of the
bugs I had to fix were caught by assertions. I don't think the lists are
going to be that much worse than the sets.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Tue, 28 Apr 2015 03:40:11 +0000 (20:40 -0700)]
util/list: Add a list validation function
Acked-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Jason Ekstrand [Tue, 28 Apr 2015 03:39:37 +0000 (20:39 -0700)]
util/list: Add list_empty and list_length functions
v2: Don't use C99 when iterating over the list
Acked-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Jason Ekstrand [Tue, 28 Apr 2015 01:56:02 +0000 (18:56 -0700)]
util/list: Add C99-based iterator macros
v2: Use LIST_ENTRY instead of container_of in iterators
Acked-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Jason Ekstrand [Tue, 28 Apr 2015 00:41:27 +0000 (17:41 -0700)]
util: Move gallium's linked list to util
The linked list in gallium is pretty much the kernel list and we would like
to have a C-based linked list for all of mesa. Let's not duplicate and
just steal the gallium one.
Acked-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Jason Ekstrand [Mon, 27 Apr 2015 23:58:29 +0000 (16:58 -0700)]
gallium/double_list: s/INLINE/inline and remove the p_compiler include
Acked-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Rob Clark <robclark@freedesktop.org>
Jason Ekstrand [Tue, 21 Apr 2015 19:12:26 +0000 (12:12 -0700)]
nir: Use nir_instr_rewrite_src in copy propagation
We were rolling our own rewrite_src variant in copy-propagation. Let's
stop doing that and use the ones in core NIR.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Fri, 24 Apr 2015 17:34:30 +0000 (10:34 -0700)]
nir: Add a function for rewriting the condition of an if statement
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Wed, 22 Apr 2015 01:00:21 +0000 (18:00 -0700)]
nir: Add and use initializer #defines for nir_src and nir_dest
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Tue, 21 Apr 2015 18:16:04 +0000 (11:16 -0700)]
nir: Modernize the out-of-SSA pass
The out-of-SSA pass was one of the first passes written when getting SSA
up-and-going (for obvious reasons). As such, it came before a lot of the
nifty SSA-based helpers were introduced. This commit modernizes it so that
we're no longer doing nearly as much manual banging on use/def sets.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Jason Ekstrand [Fri, 24 Apr 2015 00:10:42 +0000 (17:10 -0700)]
nir/validate: Validate SSA def parent instructions
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Ilia Mirkin [Fri, 8 May 2015 22:54:08 +0000 (18:54 -0400)]
nv50/ir: only propagate saturate up if some actual folding took place
The former logic would copy the saturate up to any mul with an immediate
if there was a subsequent mul with a saturate. However we only want to
do that if we collapsed 2 muls by multiplying their immediates (or were
able to put the immediate in as a post-multiplier).
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
Ian Romanick [Thu, 7 May 2015 00:04:15 +0000 (17:04 -0700)]
nir: Delete all traces of nir_op_flog
Nothing produces it, and nothing can consume it.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Acked-by: Jason Ekstrand <jason.ekstrand@intel.com>
Ian Romanick [Thu, 7 May 2015 00:01:37 +0000 (17:01 -0700)]
nir: Don't produce nir_op_flog from GLSL IR
All paths that produce GLSL IR for NIR lower ir_unop_log. All paths
that consume NIR will explode if they geta nir_op_flog.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Acked-by: Jason Ekstrand <jason.ekstrand@intel.com>
Ian Romanick [Thu, 7 May 2015 00:00:18 +0000 (17:00 -0700)]
nir: Delete all traces of nir_op_fexp
Nothing produces it, and nothing can consume it.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Acked-by: Jason Ekstrand <jason.ekstrand@intel.com>
Ian Romanick [Wed, 6 May 2015 23:57:22 +0000 (16:57 -0700)]
nir: Don't produce nir_op_fexp from GLSL IR
All paths that produce GLSL IR for NIR lower ir_unop_exp. All paths
that consume NIR will explode if they geta nir_op_fexp.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Acked-by: Jason Ekstrand <jason.ekstrand@intel.com>
Ian Romanick [Wed, 6 May 2015 23:54:06 +0000 (16:54 -0700)]
prog_to_nir: OPCODE_EXP is not nir_op_fexp
It's a weird thing that provides some values related to 2**x. It's also
already handled by a case in the switch.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Acked-by: Jason Ekstrand <jason.ekstrand@intel.com>
Neil Roberts [Fri, 8 May 2015 15:13:52 +0000 (16:13 +0100)]
i965/fs: Improve a comment about stripping trailing zeroes
Originally I wrote that removing the first parameter doesn't work but
I didn't know why. I now found a mention of this in the PRM so it's
probably worthing adding it to the comment.
Fredrik Höglund [Wed, 18 Mar 2015 19:50:06 +0000 (20:50 +0100)]
docs: Update the ARB_direct_state_access status
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 18:07:52 +0000 (19:07 +0100)]
mesa: Implement GetVertexArrayIndexed[64]iv
v2: Fix the name of the entry point in the error messages.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 18:04:16 +0000 (19:04 +0100)]
mesa: Add support for querying GL_VERTEX_ATTRIB_ARRAY_LONG
This parameter was added in OpenGL 4.3 and GL_ARB_direct_state_access.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:58:36 +0000 (18:58 +0100)]
mesa: Add a vao parameter to get_vertex_array_attrib
This is needed to implement glGetVertexArrayIndexediv and
glGetVertexArrayIndexed64iv.
v2: Make the vao parameter const.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:56:31 +0000 (18:56 +0100)]
mesa: Implement GetVertexArrayiv
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:52:36 +0000 (18:52 +0100)]
mesa: Implement VertexArrayBindingDivisor
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:50:49 +0000 (18:50 +0100)]
mesa: Add a vao parameter to vertex_binding_divisor
This is needed to implement VertexArrayBindingDivisor.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:49:06 +0000 (18:49 +0100)]
mesa: Implement VertexArrayAttribBinding
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:46:42 +0000 (18:46 +0100)]
mesa: Add a vao parameter to vertex_attrib_binding
This is needed to implement VertexArrayAttribBinding.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:44:00 +0000 (18:44 +0100)]
mesa: Implement VertexArrayAttrib[I|L]Format
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:41:09 +0000 (18:41 +0100)]
mesa: Add a vao parameter to update_array_format
This is needed to implement VertexArrayAttrib*Format.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:39:50 +0000 (18:39 +0100)]
mesa: Refactor VertexAttrib[I|L]Format
The only difference between these functions is the legal types and
sizes, so consolidate the code into a single vertex_attrib_format()
function and call it from all three entry points.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:37:27 +0000 (18:37 +0100)]
mesa: Implement VertexArrayVertexBuffers
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:35:10 +0000 (18:35 +0100)]
mesa: Implement VertexArrayVertexBuffer
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:33:14 +0000 (18:33 +0100)]
mesa: Add a vao parameter to bind_vertex_buffer
This is needed to implement VertexArrayVertexBuffer and
VertexArrayVertexBuffers.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:30:12 +0000 (18:30 +0100)]
mesa: Implement VertexArrayElementBuffer
v2: Add a doxygen comment.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:27:58 +0000 (18:27 +0100)]
mesa: Implement EnableVertexArrayAttrib
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:27:18 +0000 (18:27 +0100)]
mesa: Implement DisableVertexArrayAttrib
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:25:45 +0000 (18:25 +0100)]
mesa: Keep track of the last looked-up VAO
This saves the cost of repeated hash table lookups when the same
vertex array object is referenced in a sequence of calls such as:
glVertexArrayAttribFormat(vao, ...);
glVertexArrayAttribBinding(vao, ...);
glEnableVertexArrayAttrib(vao, ...);
...
Note that VAO's are container objects that are not shared between
contexts.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:24:36 +0000 (18:24 +0100)]
mesa: Add _mesa_lookup_vao_err
This is a convenience function that generates GL_INVALID_OPERATION
when the array object doesn't exist.
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Fredrik Höglund [Mon, 2 Mar 2015 17:22:50 +0000 (18:22 +0100)]
mesa: Implement CreateVertexArrays
v2: Update the documentation for gen_vertex_arrays().
Reviewed-by: Laura Ekstrand <laura@jlekstrand.net>
Neil Roberts [Thu, 7 May 2015 13:20:17 +0000 (14:20 +0100)]
i965/skl: In opt_sampler_eot always set destination register to null
opt_sampler_eot enables a direct write to framebuffer from a sample.
In order to do this the sample message needs to have a message header
so if there wasn't one already then the function adds one. In addition
the function sets the destination register to null because it's no
longer used. However it was only doing this in cases where it was
adding a message header. This patch just moves setting the destination
so that it happens even if there's a messge header. In practice this
doesn't seem to make any difference but it's a bit cleaner.
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Neil Roberts [Thu, 7 May 2015 17:55:55 +0000 (18:55 +0100)]
i965/fs: Set the header_size on LOAD_PAYLOAD in opt_sampler_eot
Commit
94ee908448 added a header size parameter to the function to
create the LOAD_PAYLOAD instruction. However this broke
opt_sampler_eot which manually constructs the instruction and so
wasn't setting the header_size. This ends up making the parameters for
the send message all have the wrong location and it all falls apart.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Martin Peres [Thu, 7 May 2015 13:57:48 +0000 (16:57 +0300)]
docs: document the LIBGL_DRI3_DISABLE environment variable
Suggested-by: Axel Davy <axel.davy@ens.fr>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Martin Peres <martin.peres@linux.intel.com>
Dave Airlie [Fri, 24 Apr 2015 02:50:21 +0000 (12:50 +1000)]
docs: update ARB_vertex_attrib_64bit status
Add to GL3.txt and release notes.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Fri, 20 Feb 2015 01:42:19 +0000 (11:42 +1000)]
st/mesa: add double input support including lowering (v3.1)
This takes a different approach to previously, we cannot index into the
inputMapping with anything but the mesa attribute index, so we can't use
the just add one to index trick, we need more info to add one to it
after we've mapped the input.
(Fixed copy propgation and cleaned up a little)
v2: drop float64 format check, just attr->Doubles.
merge enable patch.
v3: cleanup code a bit.
v3.1: minor review fixups (comment, newline) (Ilia)
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Fri, 20 Feb 2015 01:41:01 +0000 (11:41 +1000)]
mesa/vbo: add support for 64-bit vertex attributes. (v1)
This adds support in the vbo and array code to handle
double vertex attributes.
v0.2: merge code to handle doubles in vbo layer.
v1: don't use v0, merge api_array elt code.
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Thu, 30 Apr 2015 00:42:06 +0000 (10:42 +1000)]
glsl: check total count of multi-slot double vertex attribs
The spec is vague all over the place about this, but this seems
to be the intent, we can probably make this optional later if
someone makes hw that cares and writes a driver.
Basically we need to double count some of the d types but
only for totalling not for slot number assignment.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Wed, 8 Apr 2015 04:38:19 +0000 (14:38 +1000)]
glsl: track which program inputs are doubles
instead of doing the attempts at dual slot handling here,
let the backend do it.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Fri, 20 Feb 2015 01:38:12 +0000 (11:38 +1000)]
glsl: add ARB_vertex_attrib_64bit support. (v2)
Just more boilerplate stuff.
v2:
bad fallthrough on versioning,
this is my ugly but self contained solution (Ian)
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Fri, 20 Feb 2015 01:32:55 +0000 (11:32 +1000)]
mesa: add ARB_vertex_attrib_64bit to extensions. (v2)
Just add the boilerplate bits.
v2: add to version.c
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Fri, 20 Feb 2015 01:26:33 +0000 (11:26 +1000)]
mapi: add GL_ARB_vertex_attrib_64bit support
This just adds the glapi bits.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Fri, 20 Feb 2015 03:58:18 +0000 (13:58 +1000)]
st/glsl_to_tgsi: fix ir_assignment hack doing bad things for doubles
This hack for fixing gl_FragDepth apparantly caused a GLSL shader
outputting a single double to try and output a dvec4, but we hadn't
assigned outputs for the secondary bit.
This avoids going into the hack code for scalar doubles.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Topi Pohjolainen [Thu, 7 May 2015 13:07:15 +0000 (16:07 +0300)]
i965/wm/gen6: Add option for disabling statistics collection
Normally this is always needed but for internal blits and clears
we need to be able to disable it.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Topi Pohjolainen [Wed, 6 May 2015 17:31:49 +0000 (20:31 +0300)]
i965/wm/gen6: Refactor state setup
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Anuj Phogat [Wed, 6 May 2015 22:11:42 +0000 (15:11 -0700)]
i965: Remove unused variables
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Anuj Phogat [Wed, 15 Apr 2015 18:46:53 +0000 (11:46 -0700)]
i965: Change the order of conditions tested in if
Reduces the number of conditions tested in if to one in case of
non-integer formats. Makes no functional changes.
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Matt Turner [Wed, 6 May 2015 05:54:59 +0000 (22:54 -0700)]
nir: Allow feq/fne/ieq/ine to be optimized with inot.
instructions in affected programs: 380 -> 376 (-1.05%)
helped: 2
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Glenn Kennard <glenn.kennard@gmail.com>
Matt Turner [Wed, 6 May 2015 03:25:07 +0000 (20:25 -0700)]
nir: Recognize (a < c || b < c) as min(a, b) < c.
... and (a >= c) || (b >= c) as max(a, b) >= c.
Similar to commit
97e6c1b9.
total instructions in shared programs:
6182276 ->
6182180 (-0.00%)
instructions in affected programs: 6400 -> 6304 (-1.50%)
helped: 68
HURT: 4
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Glenn Kennard <glenn.kennard@gmail.com>
Matt Turner [Wed, 6 May 2015 03:20:30 +0000 (20:20 -0700)]
nir: Recognize trivial min/max.
No changes, but does prevent some regressions in the next commit.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Glenn Kennard <glenn.kennard@gmail.com>
Matt Turner [Wed, 6 May 2015 02:56:27 +0000 (19:56 -0700)]
nir: Recognize i2b(b2i(x)) as x.
Helps the same set of programs as the previous commit.
instructions in affected programs: 4490 -> 4346 (-3.21%)
helped: 8
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Glenn Kennard <glenn.kennard@gmail.com>
Matt Turner [Wed, 6 May 2015 02:50:58 +0000 (19:50 -0700)]
nir: Recognize imul(b2i(a), b2i(b)) as a logical AND.
Four shaders in Unreal 4's Sun Temple are helped, and gain SIMD16
because we avoid an integer multiplication.
instructions in affected programs: 2353 -> 2245 (-4.59%)
helped: 4
GAINED: 4
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Glenn Kennard <glenn.kennard@gmail.com>
Chad Versace [Wed, 6 May 2015 02:05:32 +0000 (19:05 -0700)]
i965/sync: Implement DRI2_Fence extension
This enables EGL_KHR_fence_sync and EGL_KHR_wait_sync.
Below is the difference in piglit results, before and after this patch.
No regressions and several tests improve from 'skip' to 'pass'. Out of
EGL_KHR_fence_sync tests, two of the multithreaded tests skip; all other
tests pass.
cmdline: piglit run -p gbm -t sync tests/quick.py
mesa: master@
1ac7db0
piglit:
4069bec
hw: Ivybridge
| before after
------+-------------
pass | 32 46
fail | 0 0
crash | 0 0
skip | 35 21
total | 67 67
v2:
- Set fence->signalled = true in brw_fence_has_completed() too.
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chad Versace [Wed, 6 May 2015 02:05:32 +0000 (19:05 -0700)]
i965/sync: Replace prefix 'intel_sync' -> 'intel_gl_sync'
I'm about to implement DRI2_Fenc in intel_syncobj.c. To prevent
madness, we need to prefix functions for GL_ARB_sync with 'gl' and
functions for DRI2_Fence with 'dri'. Otherwise, the file will become
a jumble of similiarly named functions.
For example:
old-name: intel_client_wait_sync()
new-name: intel_gl_client_wait_sync()
soon-to-come: intel_dri_client_wait_sync()
I wrote this renaming commit separately from the commit that implements
DRI2_Fence because I wanted the latter diff to be reviewable.
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chad Versace [Wed, 6 May 2015 02:05:31 +0000 (19:05 -0700)]
i915/sync: Return early when calloc fails
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chad Versace [Wed, 6 May 2015 02:05:30 +0000 (19:05 -0700)]
i965/sync: Return NULL when calloc fails
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chad Versace [Wed, 6 May 2015 02:05:29 +0000 (19:05 -0700)]
i915/sync: Don't crash when deleting sync object
Don't pass NULL to drm_intel_bo_unreference(). It doesn't like that.
Bug found by code inspection.
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chad Versace [Wed, 6 May 2015 02:05:28 +0000 (19:05 -0700)]
i965/sync: Don't crash when deleting sync object
Don't pass NULL to drm_intel_bo_unreference(). It doesn't like that.
Bug found by code inspection.
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chad Versace [Thu, 7 May 2015 15:09:07 +0000 (08:09 -0700)]
egl/dri2: Fix codestyle in a comment
Pointed out by Kenneth Graunke. Trivial fix.
Martin Peres [Wed, 6 May 2015 10:31:30 +0000 (13:31 +0300)]
glx: report which DRI version is used when in verbose debug mode
This should make it more obvious in bug reports while also removing
any sort of guesswork for developers.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Martin Peres <martin.peres@linux.intel.com>
Vinson Lee [Wed, 6 May 2015 19:39:09 +0000 (12:39 -0700)]
glapi: Add positional argument specifier.
Fix build error introduced with commit
1c5a57a "glapi/es3.1: Add support
for GLES versions > 3.0" with Python < 2.7.
File "src/mapi/glapi/gen/gl_genexec.py", line 230, in <module>
printer.Print(api)
File "src/mapi/glapi/gen/gl_XML.py", line 120, in Print
self.printBody(api)
File "src/mapi/glapi/gen/gl_genexec.py", line 187, in printBody
condition_parts.append('(ctx->API == API_OPENGLES2 && ctx->Version >= {})'.format(int(f.api_map['es2'] * 10)))
ValueError: zero length field name in format
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Ilia Mirkin [Thu, 7 May 2015 00:48:40 +0000 (20:48 -0400)]
nv50/ir: add SHL to the list of U32 opcodes
Having the wrong inferred type prevents a number of optimizations,
including constant propagation (since float immediates work differently
than integer immediates).
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Ian Romanick [Tue, 21 Apr 2015 04:43:32 +0000 (13:43 +0900)]
i965: Sort extension enable lists
Sort by GEN, then sort by extension name.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Vinson Lee [Wed, 6 May 2015 17:09:38 +0000 (10:09 -0700)]
r600g: Fix Clang return-type build error.
Fix Clang return-type error introduced with commit
96f164f6f047833091eb98a73aa80c31dc94f962 "gallium: make
pipe_context::begin_query return a boolean".
CC r600_query.lo
r600_query.c:443:3: error: non-void function 'r600_begin_query' should return a value [-Wreturn-type]
return;
^
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Kenneth Graunke [Tue, 10 Mar 2015 11:18:06 +0000 (04:18 -0700)]
i965/fs: Allow copy propagation on ATTR file registers.
This especially helps with NIR because we currently emit MOVs at the top
of the shader to copy from various ATTR registers to a giant VGRF array
of all inputs. (This could potentially be done better, but since
there's only ever one write to each register, it should be trivial to
copy propagate away...)
With NIR - only vertex shaders:
total instructions in shared programs:
3129373 ->
2889581 (-7.66%)
instructions in affected programs:
3119717 ->
2879925 (-7.69%)
helped: 20833
Without NIR - only vertex shaders:
total instructions in shared programs:
2745901 ->
2724483 (-0.78%)
instructions in affected programs: 693426 -> 672008 (-3.09%)
helped: 3516
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Jason Ekstrand [Thu, 2 Apr 2015 01:15:42 +0000 (18:15 -0700)]
i965/fs_inst: Get rid of the effective_width field
The effective_width field was an ill-concieved hack to get around issues in
the LOAD_PAYLOAD instruction. Now that the LOAD_PAYLOAD instruction is far
more sane, this field can die.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Wed, 25 Mar 2015 00:00:04 +0000 (17:00 -0700)]
i965/fs: Rework the fs_visitor LOAD_PAYLOAD instruction
The newly reworked instruction is far more straightforward than the
original. Before, the LOAD_PAYLOAD instruction was lowered by a the
complicated and broken-by-design pile of heuristics to try and guess
force_writemask_all, exec_size, and a number of other factors on the
sources.
Instead, we use the header_size on the instruction to denote which sources
are "header sources". Header sources are required to be a single physical
hardware register that is copied verbatim. The registers that follow are
considered the actual payload registers and have a width that correspond's
to the LOAD_PAYLOAD's exec_size and are treated as being per-channel. This
gives us a fairly straightforward lowering:
1) All header sources are copied directly using force_writemask_all and,
since they are guaranteed to be a single register, there are no
force_sechalf issues.
2) All non-header sources are copied using the exact same force_sechalf
and force_writemask_all modifiers as the LOAD_PAYLOAD operation itself.
3) In order to accommodate older gens that need interleaved colors,
lower_load_payload detects when the destination is a COMPR4 register
and automatically interleaves the non-header sources. The
lower_load_payload pass does the right thing here regardless of whether
or not the hardware actually supports COMPR4.
This patch commit itself is made up of a bunch of smaller changes squashed
together. Individual change descriptions follow:
i965/fs: Rework fs_visitor::LOAD_PAYLOAD
We rework LOAD_PAYLOAD to verify that all of the sources that count as
headers are, indeed, exactly one register and that all of the non-header
sources match the destination width. We then take the exec_size for
LOAD_PAYLOAD directly from the destination width.
i965/fs: Make destinations of load_payload have the appropreate width
i965/fs: Rework fs_visitor::lower_load_payload
v2: Don't allow the saturate flag on LOAD_PAYLOAD instructions
i965/fs_cse: Support the new-style LOAD_PAYLOAD
i965/fs_inst::is_copy_payload: Support the new-style LOAD_PAYLOAD
i965/fs: Simplify setup_color_payload
Previously, setup_color_payload was a a big helper function that did a
lot of gen-specific special casing for setting up the color sources of
the LOAD_PAYLOAD instruction. Now that lower_load_payload is much more
sane, most of that complexity isn't needed anymore. Instead, we can do
a simple fixup pass for color clamps and then just stash sources
directly in the LOAD_PAYLOAD. We can trust lower_load_payload to do the
right thing with respect to COMPR4.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Tue, 24 Mar 2015 17:37:33 +0000 (10:37 -0700)]
i965/fs: Make LOAD_PAYLOAD take a header size
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Tue, 24 Mar 2015 20:55:10 +0000 (13:55 -0700)]
i965/fs: Make emit_single_fb_write take an explicit exec_size
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Wed, 1 Apr 2015 22:38:23 +0000 (15:38 -0700)]
i965/fs_inst: Add an is_copy_payload helper
This commit adds a new is_copy_payload helper to fs_inst that takes the
place of the similarly named functions in cse and register coalesce. The
two is_copy_payload functions in CSE and register coalesce were subtly
different and potentially subtly broken. The new version unifies the two
and should be more correct.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Tue, 24 Mar 2015 17:17:32 +0000 (10:17 -0700)]
i965: Change header_present to header_size in backend_instruction
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Tue, 24 Mar 2015 22:06:24 +0000 (15:06 -0700)]
i965/fs_cse: Factor out code to create copy instructions
v2: Get rid of the block parameter and make src a const reference
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Tue, 5 May 2015 22:57:11 +0000 (15:57 -0700)]
i965/fs: Make half(fs_reg, unsigned) handle register files more explicitly
Previously, we had a special case for uniforms and immediates and then a
bunch of asserts for various other pessimal things. This commit changes it
so that it explicitly does something on each register file. Some of them
are disallowed and others are treated properly.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Francisco Jerez [Wed, 22 Apr 2015 12:01:24 +0000 (15:01 +0300)]
i965/fs: Fix passing an immediate to half().
Immediates are generally uniform, they yield the same value to both
halves of any instruction.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Jeremy Huddleston Sequoia [Fri, 2 Jan 2015 03:48:40 +0000 (19:48 -0800)]
swrast: Build fix for darwin
Fixes regression from commit
64b1dc44495890cbc2c7c5509cb830264020998c
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90147
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
CC: Emil Velikov <emil.l.velikov@gmail.com>
CC: jon.turney@dronecode.org.uk
CC: ionic@macports.org
Chad Versace [Wed, 6 May 2015 02:05:20 +0000 (19:05 -0700)]
egl/dri2: Check return value of __DRI2fence::create_fence()
If it returns NULL, then return early with an error.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Roland Scheidegger [Wed, 6 May 2015 13:56:17 +0000 (15:56 +0200)]
draw: (trivial) fix out-of-bounds vector initialization
Was off-by-one. llvm says inserting an element with an index higher than the
number of elements yields undefined results. Previously such inserts were
ignored but as of llvm revision 235854 the vector gets replaced with undef,
causing failures.
This fixes piglit gl-3.2-layered-rendering-gl-layer, as mentioned in
https://llvm.org/bugs/show_bug.cgi?id=23424.
Reviewed-by: Brian Paul <brianp@vmware.com>
Cc: mesa-stable@lists.freedesktop.org
Martin Peres [Mon, 4 May 2015 13:59:54 +0000 (16:59 +0300)]
main/queryobj: add GL_QUERY_TARGET support to GetQueryObjectiv()
This was missing from my patchset to support the query-related entry
points of Direct State Access.
Reported-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Martin Peres <martin.peres@linux.intel.com>
Chia-I Wu [Wed, 6 May 2015 08:32:28 +0000 (16:32 +0800)]
ilo: silence a compiler warning
Silence
ilo_query.c:120:7: warning: 'return' with no value, in function returning non-void
since commit
96f164f6.
Tapani Pälli [Wed, 6 May 2015 06:36:15 +0000 (09:36 +0300)]
mesa: support compute stage in _mesa_program_resource_prop
Increases pass rate of ES31-CTS.*program_interface_query* tests
when run with MESA_EXTENSION_OVERRIDE='GL_ARB_compute_shader'. Many
of the negative tests that happen to use compute stage in queries
start passing.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Martin Peres <martin.peres@linux.intel.com>
Tapani Pälli [Thu, 30 Apr 2015 06:27:00 +0000 (09:27 +0300)]
glsl: mark special built-in inputs referenced by vertex stage
Refactoring done on active attribute queries did not take in to
account special built-in inputs for the vertex stage. This commit
sets them referenced by vertex stage so that they get enumerated
properly.
Fixes Piglit test 'get-active-attrib-returns-all-inputs' failure.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90243
Acked-by: Jose Fonseca <jfonseca@vmware.com>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Reviewed-By: Martin Peres <martin.peres@linux.intel.com>
Chris Forbes [Wed, 6 May 2015 07:05:17 +0000 (19:05 +1200)]
relnotes: Note support for viewport arrays on i965/Gen6.
Signed-off-by: Chris Forbes <chrisf@ijw.co.nz>
Chris Forbes [Wed, 6 May 2015 05:45:11 +0000 (17:45 +1200)]
i965/gen6: Enable ARB_viewport_array and AMD_vertex_shader_viewport_index
Signed-off-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chris Forbes [Wed, 6 May 2015 05:35:37 +0000 (17:35 +1200)]
i965/gen6: Upload all the SF viewports
Signed-off-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chris Forbes [Wed, 6 May 2015 05:34:27 +0000 (17:34 +1200)]
i965/gen6: Upload all the clip viewports
Signed-off-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Chris Forbes [Wed, 6 May 2015 02:43:34 +0000 (14:43 +1200)]
i965/gen6: setup limits for ARB_viewport_array
Signed-off-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Brian Paul [Tue, 5 May 2015 21:49:56 +0000 (15:49 -0600)]
st/mesa: fix pipe_query_result result initializer
Fixes MSVC build error.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>