Signed-off-by: Connor Abbott
---
After this, `git status' reports a clean working directory again.
.gitignore | 1 +
1 file changed, 1 insertion(+)
diff --git a/.gitignore b/.gitignore
index 00e3ce3..21aa35c 100644
--- a/.gitignore
+++ b/.gitignore
@@ -45,3 +45,4 @@ manifest.txt
Of course, these commit messages should be swapped... d'oh! Fixed locally.
On Fri, Apr 17, 2015 at 12:59 PM, Connor Abbott wrote:
> Signed-off-by: Connor Abbott
> ---
> src/mesa/main/.gitignore | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/src/mesa/main/.gi
On Fri, Apr 17, 2015 at 1:17 PM, Matt Turner wrote:
> On Fri, Apr 17, 2015 at 9:59 AM, Connor Abbott wrote:
>> Signed-off-by: Connor Abbott
>> ---
>> src/mesa/main/.gitignore | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/src/mesa/mai
I think it would be better to call this "nir_vector" rather than
"nir_array," since this is essentially a poor man's version of
std::vector<>. Is this ok with you?
On Fri, Apr 10, 2015 at 8:48 PM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir_array.h | 96
>
On Fri, Apr 10, 2015 at 8:48 PM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir_lower_vars_to_ssa.c | 28 ++--
> 1 file changed, 14 insertions(+), 14 deletions(-)
>
> diff --git a/src/glsl/nir/nir_lower_vars_to_ssa.c
> b/src/glsl/nir/nir_lower_vars_to_ssa.c
> index 8b7261
On Wed, Apr 22, 2015 at 6:06 PM, Jason Ekstrand wrote:
> On Wed, Apr 22, 2015 at 3:03 PM, Connor Abbott wrote:
>> I think it would be better to call this "nir_vector" rather than
>> "nir_array," since this is essentially a poor man's version of
>
Except for the one minor comment on patch 6, the series (including
patches 13 and 14) is
Reviewed-by: Connor Abbott
On Fri, Apr 10, 2015 at 8:48 PM, Jason Ekstrand wrote:
> Right now, most of the code to handle indirect dereferences of variables is
> dead. However, I have a patch
On Mon, Apr 27, 2015 at 1:36 PM, Eric Anholt wrote:
> Jason Ekstrand writes:
>
>> This commit adds a C-based linked list implementation for NIR. Unlike
>> exec_list in glsl/list.h, there is no C++ API. Also, this list is based on
>> wl_list (from the Wayland project) which is, in turn, based on
On Mon, Feb 15, 2016 at 4:38 PM, Matt Turner wrote:
> On Mon, Feb 15, 2016 at 12:50 PM, Ilia Mirkin wrote:
>> In a few places your indentation is off -- please look at the 'git
>> diff' output, it should be pretty obvious. You used 2 spaces instead
>> of 3 (in just a handful of places).
>
> If yo
AFAIK buffers and images can be backed by coherent or non-coherent
memory types. Found by inspection, only compile tested.
Signed-off-by: Connor Abbott
---
src/vulkan/anv_device.c | 14 --
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/src/vulkan/anv_device.c b/src
The first thing I would do is to try running it under valgrind and see
if there are any errors. Make sure to have the valgrind development
headers installed when building mesa, as anvil has a lot of special
valgrind hooks builtin that it will only use if it finds the header
during the build.
On Tu
I believe this is correct, and won't cause any issues with phi node
placement, etc. before dead_cf comes and gets rid any code in the
unreachable block(s).
Reviewed-by: Connor Abbott
BTW, I'd ask that you hold off on pushing this until I at least get a
chance to look at the phi bui
+ nir_intrinsic_instr_create(b->shader, nir_intrinsic_load_var);
> + load->num_components = orig_instr->num_components;
> + load->variables[0] =
> + nir_deref_as_var(nir_copy_deref(load, &deref->deref));
> + nir_ssa_dest_init(&load-&
Reviewed-by: Connor Abbott
On Sat, Feb 13, 2016 at 9:14 PM, Jason Ekstrand wrote:
> This can happen if a function ends in a return instruction and you remove
> the return.
> ---
> src/compiler/nir/nir_control_flow.c | 7 +--
> 1 file changed, 5 insertions(+), 2 deletions(-)
On Sat, Feb 13, 2016 at 9:14 PM, Jason Ekstrand wrote:
> ---
> src/compiler/nir/nir.c | 63
> ++
> src/compiler/nir/nir.h | 2 ++
> 2 files changed, 65 insertions(+)
>
> diff --git a/src/compiler/nir/nir.c b/src/compiler/nir/nir.c
> index cd78475.
Reviewed-by: Connor Abbott
On Sat, Feb 13, 2016 at 9:14 PM, Jason Ekstrand wrote:
> ---
> src/compiler/nir/nir_control_flow.c | 9 +
> 1 file changed, 9 insertions(+)
>
> diff --git a/src/compiler/nir/nir_control_flow.c
> b/src/compiler/nir/nir_control_flow.c
> i
On Fri, Feb 26, 2016 at 10:39 AM, Iago Toral wrote:
> On Fri, 2016-02-26 at 16:28 +0100, Roland Scheidegger wrote:
>> Am 26.02.2016 um 11:25 schrieb Iago Toral:
>> >
>> > On Thu, 2016-02-25 at 18:20 -0500, Ilia Mirkin wrote:
>> >> On Thu, Feb 25, 2016 at 6:16 PM, Francisco Jerez
>> >> wrote:
>>
On Tue, Mar 8, 2016 at 1:34 PM, Jason Ekstrand wrote:
>
>
> On Wed, Feb 24, 2016 at 10:45 PM, Connor Abbott wrote:
>>
>> On Sat, Feb 13, 2016 at 9:14 PM, Jason Ekstrand
>> wrote:
>> > ---
>> > src/compiler/nir/nir.c | 63
>> > ++
On Thu, Mar 10, 2016 at 9:30 AM, tournier.elie wrote:
> First, thank you all for your answers.
>
> So if I summarize what was said, we need
> Ian:
> - add
> - negate
> - absolute value
> - multiply
> - reciprocal
> - convert to single precision
> - convert from single precision
> Roland:
>
On Mon, Mar 7, 2016 at 3:45 AM, Samuel Iglesias Gonsálvez
wrote:
> From: Jason Ekstrand
>
> v2: Fix size/type mask to properly handle 8-bit types.
>
> Signed-off-by: Juan A. Suarez Romero
> ---
> src/compiler/nir/nir.h | 17 -
> 1 file changed, 16 insertions(+), 1 deletion(-)
>
On Fri, Mar 11, 2016 at 2:33 AM, Samuel Iglesias Gonsálvez
wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA256
>
>
>
> On 11/03/16 01:08, Jason Ekstrand wrote:
>> On Thu, Mar 10, 2016 at 4:00 PM, Connor Abbott
>> wrote:
>>
>>> On Mon, Mar
On Mon, Mar 14, 2016 at 2:10 PM, Jason Ekstrand wrote:
>
>
> On Mon, Mar 7, 2016 at 12:46 AM, Samuel Iglesias Gonsálvez
> wrote:
>>
>> From: Connor Abbott
>>
>> Some opcodes need explicit bitsizes, and sometimes we need to use the
>> double version w
On Mon, Mar 14, 2016 at 7:48 PM, Jason Ekstrand wrote:
>
>
> On Mon, Mar 7, 2016 at 12:46 AM, Samuel Iglesias Gonsálvez
> wrote:
>>
>> From: Connor Abbott
>>
>> v2: Use the bit-size information from the opcode information if defined
>> (Iago)
>>
&
On Tue, Mar 15, 2016 at 5:53 AM, Iago Toral wrote:
> On Mon, 2016-03-14 at 16:48 -0700, Jason Ekstrand wrote:
>>
>>
>> On Mon, Mar 7, 2016 at 12:46 AM, Samuel Iglesias Gonsálvez
>> wrote:
>> From: Connor Abbott
>>
>> v2: Use the bit
On Tue, Mar 15, 2016 at 10:43 AM, Connor Abbott wrote:
> On Tue, Mar 15, 2016 at 5:53 AM, Iago Toral wrote:
>> On Mon, 2016-03-14 at 16:48 -0700, Jason Ekstrand wrote:
>>>
>>>
>>> On Mon, Mar 7, 2016 at 12:46 AM, Samuel Iglesias Gonsálvez
>>
So overall, I think that there needs to be some explanation of the
design choices in the implementation. The API documentation is great,
but digging into the implementation might be a little daunting without
knowing e.g. why NEEDS_PHI is a thing. From what I gather, there are
three potential states
On Mon, Mar 21, 2016 at 6:51 PM, Connor Abbott wrote:
> So overall, I think that there needs to be some explanation of the
> design choices in the implementation. The API documentation is great,
> but digging into the implementation might be a little daunting without
> knowing e.g. w
On Mon, Mar 21, 2016 at 7:37 PM, Jason Ekstrand wrote:
>
>
> On Mon, Mar 21, 2016 at 4:21 PM, Connor Abbott wrote:
>>
>> On Mon, Mar 21, 2016 at 6:51 PM, Connor Abbott
>> wrote:
>> > So overall, I think that there needs to be some explanation of the
>&
On Wed, May 25, 2016 at 1:53 PM, Jason Ekstrand wrote:
> Now that we have the better nir_foreach_block macro, there's no reason to
> use the archaic block version for everything.
> ---
> src/compiler/nir/nir_inline_functions.c | 53
> -
> 1 file changed, 25 insert
On Wed, May 25, 2016 at 1:53 PM, Jason Ekstrand wrote:
> ---
> src/compiler/nir/nir_inline_functions.c | 26 --
> 1 file changed, 20 insertions(+), 6 deletions(-)
>
> diff --git a/src/compiler/nir/nir_inline_functions.c
> b/src/compiler/nir/nir_inline_functions.c
> index
Not sure how this wasn't already caught by valgrind, but it fixes an
issue with the vectorizer.
Signed-off-by: Connor Abbott
---
src/util/set.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/util/set.c b/src/util/set.c
index f01f869..331ff58 100644
--- a/src
ite a much better pass
using the magic of SSA. Shader-db results are in the last patch.
Series is also available as the 'nir-vectorize' branch of my FDO repository.
Connor Abbott (5):
util/set: don't compare against deleted entries
util/hash_table: don't compare deleted entri
Signed-off-by: Connor Abbott
---
src/glsl/nir/nir_array.h | 21 +
1 file changed, 21 insertions(+)
diff --git a/src/glsl/nir/nir_array.h b/src/glsl/nir/nir_array.h
index 1db4e8c..d704119 100644
--- a/src/glsl/nir/nir_array.h
+++ b/src/glsl/nir/nir_array.h
@@ -84,13 +84,34
ctorize.c
new file mode 100644
index 000..2a34a42
--- /dev/null
+++ b/src/glsl/nir/nir_opt_vectorize.c
@@ -0,0 +1,447 @@
+/*
+ * Copyright © 2015 Connor Abbott
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files
16
NoDDChk 1Q };
mov(8) g117<1>.zwD g14<4,4,1>.zzzwD{ align16
NoDDChk 1Q };
So we eliminated two add's, but then had to introduce four mov's to
transpose the result. I don't think there's much we can do about this at
the NIR level, u
On Mon, Nov 16, 2015 at 6:56 PM, Timothy Arceri wrote:
> On Sat, 2015-11-14 at 21:59 -0500, Connor Abbott wrote:
>> Not sure how this wasn't already caught by valgrind, but it fixes an
>> issue with the vectorizer.
>
> Can you give a more detailed description of the
On Mon, Nov 16, 2015 at 11:00 AM, Jason Ekstrand wrote:
> On Sat, Nov 14, 2015 at 6:59 PM, Connor Abbott wrote:
>> Shader-db results on bdw with INTEL_DEBUG=vec4:
>>
>> total instructions in shared programs: 1634044 -> 1612936 (-1.29%)
>> instructions in affec
I think Ken already pushed a similar patch so we can drop this.
On Thu, Nov 19, 2015 at 5:05 AM, Iago Toral Quiroga wrote:
> From: Connor Abbott
>
> Reviewed-by: Iago Toral Quiroga
> ---
> src/mesa/drivers/dri/i965/brw_fs.cpp | 3 +++
> 1 file changed, 3 insertions(+)
&g
On Thu, Nov 19, 2015 at 6:40 AM, Matt Turner wrote:
> On Thu, Nov 19, 2015 at 2:05 AM, Iago Toral Quiroga wrote:
>> From: Connor Abbott
>>
>> It appears that not only math instructions, but also MOV_BYTES or
>> any instruction that uses Align1 mode cannot be in th
On Thu, Nov 19, 2015 at 10:31 AM, Connor Abbott wrote:
> On Thu, Nov 19, 2015 at 6:40 AM, Matt Turner wrote:
>> On Thu, Nov 19, 2015 at 2:05 AM, Iago Toral Quiroga
>> wrote:
>>> From: Connor Abbott
>>>
>>> It appears that not only math instructions,
into registers with, say, 8
32-bit things per SIMD8 register. So to get, say, the second uniform,
we need to say "get the second channel of this register and splat it
to all the channels of the source," which we can do with a source of
stride 0 and offset 1.
>
> No piglit regressions on I
On Thu, Nov 19, 2015 at 1:54 PM, Matt Turner wrote:
> On Thu, Nov 19, 2015 at 7:31 AM, Connor Abbott wrote:
>> On Thu, Nov 19, 2015 at 6:40 AM, Matt Turner wrote:
>>> On Thu, Nov 19, 2015 at 2:05 AM, Iago Toral Quiroga
>>> wrote:
>>>> From: Connor Abb
On Thu, Nov 19, 2015 at 2:07 PM, Connor Abbott wrote:
> On Thu, Nov 19, 2015 at 1:54 PM, Matt Turner wrote:
>> On Thu, Nov 19, 2015 at 7:31 AM, Connor Abbott wrote:
>>> On Thu, Nov 19, 2015 at 6:40 AM, Matt Turner wrote:
>>>> On Thu, Nov 19, 2015 at 2:05 AM, I
On Thu, Nov 19, 2015 at 3:11 PM, Matt Turner wrote:
> On Thu, Nov 19, 2015 at 11:35 AM, Connor Abbott wrote:
>> On Thu, Nov 19, 2015 at 2:07 PM, Connor Abbott wrote:
>>> On Thu, Nov 19, 2015 at 1:54 PM, Matt Turner wrote:
>>>> On Thu, Nov 19, 2015 at 7:31 AM, Conn
On Fri, Nov 20, 2015 at 2:46 AM, Iago Toral wrote:
> On Thu, 2015-11-19 at 10:14 -0500, Connor Abbott wrote:
>> I think Ken already pushed a similar patch so we can drop this.
>
> I don't see that in master, but sure, we can hold this back if he is
> planning to push the s
On Wed, Nov 25, 2015 at 4:48 AM, Iago Toral wrote:
> On Tue, 2015-11-17 at 21:54 -0800, Jordan Justen wrote:
>> Signed-off-by: Jordan Justen
>> ---
>> src/glsl/nir/glsl_to_nir.cpp | 29 +
>> src/glsl/nir/nir_intrinsics.h | 1 +
>> 2 files changed, 30 insertions(+)
>
On Wed, Dec 2, 2015 at 3:26 AM, Kenneth Graunke wrote:
> Normally, get_nir_src() won't return an immediate value - it moves it to
> a temporary VGRF. There are consumers of get_nir_src() that rely on
> this, and are unprepared to handle immediate values.
>
> This patch introduces new helpers whic
On Thu, Dec 17, 2015 at 11:44 AM, Matt Turner wrote:
> On Wed, Dec 9, 2015 at 4:15 AM, Iago Toral Quiroga wrote:
>> Right now we rely on the code at the bottom of brw_set_dest to set the
>> correct execution size for anything that does not operate on a full SIMD
>> register (dst.width < BRW_EXE
We haven't allowed NIR in core gallium before, since core gallium has
to be built with some old version of MSVC that doesn't support many
C99 features that we really wanted to use. The only reason that
-Werror exists is for compatibility with old MSVC, and if you want to
use NIR with something that
Why don't you just introduce a reference-counted wrapper around a
nir_shader * and pass that around instead? This seems like a
gallium-specific problem, so modifying other things to work around it
doesn't seem so appealing.
On Sat, Dec 19, 2015 at 8:55 PM, Rob Clark wrote:
> From: Rob Clark
>
>
sa/st doesn't need to be built with old
MSVC, just gallium.
>
> BR,
> -R
>
> On Sat, Dec 19, 2015 at 8:58 PM, Connor Abbott wrote:
>> We haven't allowed NIR in core gallium before, since core gallium has
>> to be built with some old version of MSVC that doesn&
On Sat, Dec 19, 2015 at 9:52 PM, Rob Clark wrote:
> On Sat, Dec 19, 2015 at 9:30 PM, Connor Abbott wrote:
>> On Sat, Dec 19, 2015 at 9:18 PM, Rob Clark wrote:
>>> Note that this is *only* about the header files.. not the src files.
>>> I'm not proposing to make
On Sat, Dec 19, 2015 at 10:19 PM, Rob Clark wrote:
> On Sat, Dec 19, 2015 at 10:05 PM, Connor Abbott wrote:
>> On Sat, Dec 19, 2015 at 9:52 PM, Rob Clark wrote:
>>> On Sat, Dec 19, 2015 at 9:30 PM, Connor Abbott wrote:
>>>> On Sat, Dec 19, 2015 at 9:18 PM, Rob
On Sat, Dec 19, 2015 at 11:08 PM, Rob Clark wrote:
> On Sat, Dec 19, 2015 at 10:30 PM, Connor Abbott wrote:
>> On Sat, Dec 19, 2015 at 10:19 PM, Rob Clark wrote:
>>> On Sat, Dec 19, 2015 at 10:05 PM, Connor Abbott wrote:
>>>> On Sat, Dec 19, 2015 at 9:52 PM, Rob C
On Sun, Dec 20, 2015 at 12:02 AM, Rob Clark wrote:
> On Sat, Dec 19, 2015 at 11:46 PM, Connor Abbott wrote:
>> On Sat, Dec 19, 2015 at 11:08 PM, Rob Clark wrote:
>>> On Sat, Dec 19, 2015 at 10:30 PM, Connor Abbott wrote:
>>>> On Sat, Dec 19, 2015 at 10:19 PM,
NIR has never been built with MSVC2008, so we shouldn't add
MSVC_2008_COMPAT_CFLAGS to anything that uses it. This allows us to get
rid of the pragma in tgsi_to_nir.c.
Build tested with freedreno.
Signed-off-by: Connor Abbott
---
src/gallium/auxiliary/Makefile.am
On Sun, Dec 20, 2015 at 10:04 PM, Rob Clark wrote:
> On Sun, Dec 20, 2015 at 9:12 PM, Jason Ekstrand wrote:
>>
>> On Dec 19, 2015 5:55 PM, "Rob Clark" wrote:
>>>
>>> From: Rob Clark
>>>
>>> Jason,
>>>
>>> How much do you hate this idea? Seems like an easy alternative to
>>> using ralloc ctx's
On Sun, Dec 20, 2015 at 10:43 AM, Rob Clark wrote:
> On Sun, Dec 20, 2015 at 12:25 AM, Connor Abbott wrote:
>>> But, this would still be an issue for tgsi_to_nir, which I guess
>>> wouldn't make sense in glsl/nir. Although turns out we solve that
>>>
On Mon, Dec 21, 2015 at 6:39 AM, Marek Olšák wrote:
> On Mon, Dec 21, 2015 at 6:48 AM, Jason Ekstrand wrote:
>>
>> On Dec 20, 2015 7:43 PM, "Rob Clark" wrote:
>>>
>>> On Sun, Dec 20, 2015 at 10:29 PM, Connor Abbott
>>> wrote:
>>> &g
On Mon, Dec 21, 2015 at 11:45 AM, Marek Olšák wrote:
> On Mon, Dec 21, 2015 at 4:38 PM, Connor Abbott wrote:
>> On Mon, Dec 21, 2015 at 6:39 AM, Marek Olšák wrote:
>>> On Mon, Dec 21, 2015 at 6:48 AM, Jason Ekstrand
>>> wrote:
>>>>
>>&
On Tue, Dec 22, 2015 at 9:02 PM, Rob Clark wrote:
> On Mon, Dec 21, 2015 at 1:48 PM, Jason Ekstrand wrote:
>>
>> I think two different concepts of ownership are getting conflated here:
>> Right/responsibility to delete and right to modify.
>>
>> The way I understand it, gallium, as it stands, giv
On Tue, Dec 22, 2015 at 9:55 PM, Rob Clark wrote:
> On Tue, Dec 22, 2015 at 9:47 PM, Connor Abbott wrote:
>> On Tue, Dec 22, 2015 at 9:02 PM, Rob Clark wrote:
>>> On Mon, Dec 21, 2015 at 1:48 PM, Jason Ekstrand
>>> wrote:
>>>>
>>>> I th
On Thu, Dec 24, 2015 at 8:34 PM, Kenneth Graunke wrote:
> This morning, I woke up and somehow "knew" what was causing my HS GPU hangs
> on Gen7/7.5. It turns out I was (completely) wrong, but through some
> miraculous series of illogical leaps, I arrived at a solution anyway.
>
> I don't honestly
On Sat, Dec 26, 2015 at 2:09 PM, Jason Ekstrand wrote:
> This commit adds a NIR pass for lowering away returns in functions. If the
> return is in a loop, it is lowered to a break. If it is not in a loop,
> it's lowered away by moving/deleting code as needed.
> ---
>
> Unfortunately, the loop-ha
On Sat, Dec 26, 2015 at 5:57 PM, Jason Ekstrand wrote:
>
> On Dec 26, 2015 2:22 PM, "Connor Abbott" wrote:
>>
>> On Sat, Dec 26, 2015 at 2:09 PM, Jason Ekstrand
>> wrote:
>> > This commit adds a NIR pass for lowering away returns in functions.
Reviewed-by: Connor Abbott
On Sat, Dec 26, 2015 at 2:09 PM, Jason Ekstrand wrote:
> When Connor originally drafted NIR, he copied the same function+overload
> system that GLSL IR had with a few names changed. However, this
> double-indirection is not really needed and has only
On Sat, Dec 26, 2015 at 7:08 PM, Jason Ekstrand wrote:
>
> On Dec 26, 2015 3:04 PM, "Connor Abbott" wrote:
>>
>> On Sat, Dec 26, 2015 at 5:57 PM, Jason Ekstrand
>> wrote:
>> >
>> > On Dec 26, 2015 2:22 PM, "Connor Abbott" wrot
On Sat, Dec 26, 2015 at 9:55 PM, Jason Ekstrand wrote:
>
>
> On Sat, Dec 26, 2015 at 5:15 PM, Jason Ekstrand
> wrote:
>>
>>
>>
>> On Sat, Dec 26, 2015 at 4:59 PM, Connor Abbott
>> wrote:
>>>
>>> On Sat, Dec 26, 2015 at 7:08 PM, Ja
On Sat, Dec 26, 2015 at 2:09 PM, Jason Ekstrand wrote:
> This commit adds a new NIR pass that lowers all function calls away by
> inlining the functions.
> ---
>
> There are still two things missing here:
>
> 1) We really shouldn't do the inline recursively. We should keep a
> hash-table of
On Sun, Dec 27, 2015 at 4:35 PM, Jason Ekstrand wrote:
> On Sun, Dec 27, 2015 at 10:15 AM, Connor Abbott wrote:
>>
>> On Sat, Dec 26, 2015 at 2:09 PM, Jason Ekstrand
>> wrote:
>> > This commit adds a new NIR pass that lowers all function calls away
On Mon, Dec 28, 2015 at 10:13 AM, Rob Clark wrote:
> On Tue, Dec 22, 2015 at 10:11 PM, Connor Abbott wrote:
>> On Tue, Dec 22, 2015 at 9:55 PM, Rob Clark wrote:
>>> On Tue, Dec 22, 2015 at 9:47 PM, Connor Abbott wrote:
>>>> On Tue, Dec 22, 2015 at 9:02 PM, Rob C
On Mon, Aug 4, 2014 at 9:46 AM, Eric Anholt wrote:
> Connor Abbott writes:
>
>> On Fri, Aug 1, 2014 at 12:25 PM, Eric Anholt wrote:
>>> Connor Abbott writes:
>>>
>>>> On Fri, Aug 1, 2014 at 11:51 AM, Eric Anholt wrote:
>>>>> Connor Abb
On Mon, Aug 4, 2014 at 10:17 AM, Eric Anholt wrote:
> Connor Abbott writes:
>
>> On Mon, Aug 4, 2014 at 9:46 AM, Eric Anholt wrote:
>>> Connor Abbott writes:
>>>
>>>> On Fri, Aug 1, 2014 at 12:25 PM, Eric Anholt wrote:
>>>>> Connor Abb
On Mon, Aug 4, 2014 at 10:42 AM, Connor Abbott wrote:
> On Mon, Aug 4, 2014 at 10:17 AM, Eric Anholt wrote:
>> Connor Abbott writes:
>>
>>> On Mon, Aug 4, 2014 at 9:46 AM, Eric Anholt wrote:
>>>> Connor Abbott writes:
>>>>
>>&
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 2 +-
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 11 ++-
3 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/src/mesa/drivers/dri/i965
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 4 +---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 8
2 files changed, 5 insertions(+), 7 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_fs_fp.cpp
b/src/mesa/drivers/dri/i965/brw_fs_fp.cpp
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 5 ++---
2 files changed, 3 insertions(+), 4 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_fs.h
b/src/mesa/drivers/dri/i965/brw_fs.h
index d3c2e5f
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 3 ++-
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 10 +-
3 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/src/mesa/drivers/dri/i965
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 4 +++-
src/mesa/drivers/dri/i965/brw_shader.cpp | 13 +
src/mesa/drivers/dri/i965/brw_shader.h | 3 ++-
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 4 +++-
4 files changed, 13
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 5 +++--
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 5 +++--
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 30 +---
3 files changed, 24 insertions(+), 16 deletions(-)
diff --git a/src
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 12 -
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 6 ++---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 37 +++-
3 files changed, 29 insertions(+), 26 deletions(-)
diff --git a
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 15 ---
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 9 ++--
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 67 +++-
3 files changed, 49 insertions(+), 42 deletions(-)
diff --git a/src
also get to clean up the Mesa IR frontend which was
previously creating a fake ir_texture * to hack around the mess.
No piglit changes on my Ivy Bridge.
Connor Abbott (13):
i965/fs: don't use ir->shadow_comparitor in emit_texture_*
i965: make brw_texture_offset() more gene
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 2 +-
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 4 ++--
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 16
3 files changed, 11 insertions(+), 11 deletions(-)
diff --git a/src/mesa/drivers/dri
We already have the type from the original destination.
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
b/src/mesa/drivers/dri/i965
At this point, the only thing it's used for is the opcode.
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 27 --
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 6 +--
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 55 +++---
es all the information needed to make a texture instruction and
handles all the setup, and all we have to do to emit a texture
instruction while converting from GLSL IR, Mesa IR, or any new backend
is to extract the information emit_texture() needs and then call it.
Signed-off-by: Connor Abbott
---
Now that we've made all the texture emit code mostly independent of GLSL
IR, this isn't necessary any more.
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 19 +--
1 file changed, 5 insertions(+), 14 deletions(-)
diff --git a/src/mesa/driver
also get to clean up the Mesa IR frontend which was
previously creating a fake ir_texture * to hack around the mess.
No piglit changes on my Ivy Bridge.
v2: fix build failure in the middle of the series.
Connor Abbott (13):
i965/fs: don't use ir->shadow_comparitor in emit_texture_*
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 4 +---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 8
2 files changed, 5 insertions(+), 7 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_fs_fp.cpp
b/src/mesa/drivers/dri/i965/brw_fs_fp.cpp
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 5 +++--
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 5 +++--
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 30 +---
3 files changed, 24 insertions(+), 16 deletions(-)
diff --git a/src
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 15 ---
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 9 ++--
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 67 +++-
3 files changed, 49 insertions(+), 42 deletions(-)
diff --git a/src
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 4 +++-
src/mesa/drivers/dri/i965/brw_shader.cpp | 13 +
src/mesa/drivers/dri/i965/brw_shader.h | 3 ++-
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 4 +++-
4 files changed, 13
es all the information needed to make a texture instruction and
handles all the setup, and all we have to do to emit a texture
instruction while converting from GLSL IR, Mesa IR, or any new backend
is to extract the information emit_texture() needs and then call it.
Signed-off-by: Connor Abbott
---
Now that we've made all the texture emit code mostly independent of GLSL
IR, this isn't necessary any more.
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 19 +--
1 file changed, 5 insertions(+), 14 deletions(-)
diff --git a/src/mesa/driver
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 12 -
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 6 ++---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 37 +++-
3 files changed, 29 insertions(+), 26 deletions(-)
diff --git a
We already have the type from the original destination.
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_fs_visitor.cpp
b/src/mesa/drivers/dri/i965
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 2 +-
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 11 ++-
3 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/src/mesa/drivers/dri/i965
v2: fix build failure
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 3 ++-
src/mesa/drivers/dri/i965/brw_fs_fp.cpp | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 12 ++--
3 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/src
v2: fix build failure
Signed-off-by: Connor Abbott
---
src/mesa/drivers/dri/i965/brw_fs.h | 2 +-
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 7 +++
2 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_fs.h
b/src/mesa/drivers/dri/i965
801 - 900 of 1264 matches
Mail list logo
