I know I had another solution to this problem, but... meh. I don't
particularly care what solution gets chosen.
Reviewed-by: Connor Abbott
On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand wrote:
> ---
> src/mesa/drivers/dri/i965/brw_fs_nir.cpp | 32
> +---
On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand wrote:
> ---
> src/mesa/drivers/dri/i965/brw_fs_nir.cpp | 17 ++---
> 1 file changed, 14 insertions(+), 3 deletions(-)
>
> diff --git a/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
> b/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
> index 22f4c0
As a future cleanup, would it be worth getting rid of fddx and fddy
entirely and just generating the correct variant based on the
glHint/drirc configuration?
On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir_opcodes.h | 4
> 1 file changed, 4 insertions(+)
>
>
To be fully correct here, we should be plumbing through
ctx->Const.NativeIntegers and setting it to 1.0f if that's false, but
other things are already broken for hw without native integers so that
can probably wait. Still, might be worth adding a TODO. In any case,
Reviewed-by: Connor Abb
On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand wrote:
> ---
> src/glsl/Makefile.sources| 1 +
> src/glsl/nir/nir.c | 19 +++-
> src/glsl/nir/nir.h | 21 --
> src/glsl/nir/nir_dominance.c | 6 ++---
> src/glsl/nir/nir_metadata.c | 52
> +
Patches 23-26, 28, 30-35, 37-38, 40, (41 gets killed later so I didn't
review it), 42-44 are
Reviewed-by: Connor Abbott
I'm going to bed now, but I'll try to do some more later.
On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand wrote:
> NIR (pronounced "ner&
On Wed, Dec 17, 2014 at 6:52 AM, Jason Ekstrand wrote:
>
>
> On Tue, Dec 16, 2014 at 10:31 PM, Connor Abbott wrote:
>>
>> As a future cleanup, would it be worth getting rid of fddx and fddy
>> entirely and just generating the correct variant based on the
>> glHin
On Wed, Dec 17, 2014 at 7:04 AM, Jason Ekstrand wrote:
>
>
> On Tue, Dec 16, 2014 at 10:58 PM, Connor Abbott wrote:
>>
>> On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand
>> wrote:
>> > ---
>> > src/glsl/Makefile.sources|
Sorry to ask, but... this one is so trivial and touches code that no
earlier patch touches, why don't we just squash it into the commit
that adds nir.c?
On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/src/
One thing I'm a little worried about is that passes might forget to
require the right metadata, and they'll just happen to work since the
pass before also requires the same metadata and preserves it. I think
a good thing to do to combat this is to have a debug mode that dirties
*all* the metadata i
On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand wrote:
> From: Connor Abbott
>
> ---
> src/glsl/Makefile.sources | 1 +
> src/glsl/nir/nir.h| 3 +
> src/glsl/nir/nir_opt_copy_propagate.c | 313
> ++
&g
On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand wrote:
> Since we don't actually have an "if" instruction, this is a very common
> pattern when iterating over instructions. This adds a helper function for
> it to make things a little less painful.
> ---
> src/glsl/nir/nir.c | 17 +++
This patch needs to get renamed, probably something like "nir: set
reg_alloc and ssa_alloc when indexing registers and SSA values"
On Tue, Dec 16, 2014 at 1:05 AM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir.c | 3 +++
> src/glsl/nir/nir.h | 4 ++--
> 2 files changed, 5 insertions(+), 2 delet
I'm sure you're already aware, but there are two things we could do to
speed this up:
1. Pre-compute def/kill sets for each block similar to what i965 does.
2. Use a worklist + an array of flags for "this block is in the
worklist" rather than walking all the basic blocks in reverse to find
the few
On Wed, Dec 17, 2014 at 5:11 PM, Jason Ekstrand wrote:
> On Wed, Dec 17, 2014 at 1:52 PM, Connor Abbott wrote:
>>
>> I'm sure you're already aware, but there are two things we could do to
>> speed this up:
>>
>> 1. Pre-compute def/kill sets for each
On Wed, Dec 17, 2014 at 5:20 PM, Jason Ekstrand wrote:
>
>
> On Wed, Dec 17, 2014 at 1:10 PM, Jason Ekstrand
> wrote:
>>
>>
>>
>> On Wed, Dec 17, 2014 at 12:07 PM, Connor Abbott
>> wrote:
>>>
>>> On Tue, Dec 16, 2014 at 1
On Tue, Dec 16, 2014 at 1:05 AM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir.c | 45 -
> src/glsl/nir/nir.h | 23 +++
> src/glsl/nir/nir_print.c | 21 +
> 3 files changed, 88 insertions(+), 1 deleti
Whew! Other than a few minor things below,
Reviewed-by: Connor Abbott
I tried to understand it all as much as I could, but it is rather
tricky... but I can't suggest anything to make it easier to
understand, after all the paper itself is rather tricky and your
comments help a lot. If anyon
On Wed, Dec 17, 2014 at 5:59 PM, Jason Ekstrand wrote:
>
>
> On Wed, Dec 17, 2014 at 11:51 AM, Connor Abbott wrote:
>>
>> One thing I'm a little worried about is that passes might forget to
>> require the right metadata, and they'll just happen to work since
On Wed, Dec 17, 2014 at 6:02 PM, Jason Ekstrand wrote:
>
>
> On Wed, Dec 17, 2014 at 2:26 PM, Connor Abbott wrote:
>>
>> On Wed, Dec 17, 2014 at 5:20 PM, Jason Ekstrand
>> wrote:
>> >
>> >
>> > On Wed, Dec 17, 2014 at 1:10 PM, Jason Ekstra
I think we're missing a few things:
* Phi node sources
* Parallel copy entries
Whether you care enough to validate those is up to you. Otherwise
Reviewed-by: Connor Abbott
On Tue, Dec 16, 2014 at 1:05 AM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir_validate.c | 13 +++
On Wed, Dec 17, 2014 at 8:00 PM, Jason Ekstrand wrote:
>
>
> On Wed, Dec 17, 2014 at 4:52 PM, Connor Abbott wrote:
>>
>> I think we're missing a few things:
>>
>> * Phi node sources
>
>
> Added
>
>>
>> * Parallel copy entries
>
Would it be possible to drop this patch since this all gets deleted later?
On Tue, Dec 16, 2014 at 1:05 AM, Jason Ekstrand wrote:
> ---
> src/glsl/Makefile.sources| 1 +
> src/glsl/nir/nir.h | 1 +
> src/glsl/nir/nir_opcodes.h | 1 +
> sr
We shouldn't ever have global registers until we start dealing with
subroutines, so you don't need to handle them here.
On Tue, Dec 16, 2014 at 1:05 AM, Jason Ekstrand wrote:
> ---
> src/mesa/drivers/dri/i965/brw_fs.h | 4 +--
> src/mesa/drivers/dri/i965/brw_fs_nir.cpp | 46
> +++
On Tue, Dec 16, 2014 at 1:05 AM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir_from_ssa.c | 40 -
> src/mesa/drivers/dri/i965/brw_fs_nir.cpp | 50
> +---
> 2 files changed, 65 insertions(+), 25 deletions(-)
>
> diff --git a/src/gl
Again, this looks pretty self-contained and pretty easy to squash.
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/glsl/nir/nir.h b/src/glsl/nir/nir.h
> index b04a137..fc16cb5 100644
>
Can we move this right after patch 53 (nir: Automatically handle SSA
uses when an instruction is inserted), since they're doing similar
things?
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir.c | 9 -
> 1 file changed, 4 insertions(+), 5 deletions(-)
>
> d
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand wrote:
> ---
> src/glsl/Makefile.sources | 1 +
> src/glsl/nir/nir.h | 2 +
> src/glsl/nir/nir_split_var_copies.c | 225
>
> 3 files changed, 228 insertions(+)
> create mode 10064
On Wed, Dec 17, 2014 at 8:48 PM, Jason Ekstrand wrote:
>
>
> On Wed, Dec 17, 2014 at 2:23 PM, Connor Abbott wrote:
>>
>> On Wed, Dec 17, 2014 at 5:11 PM, Jason Ekstrand
>> wrote:
>> > On Wed, Dec 17, 2014 at 1:52 PM, Connor Abbott
>> > wrote:
&
On Wed, Dec 17, 2014 at 8:41 PM, Jason Ekstrand wrote:
>
>
> On Wed, Dec 17, 2014 at 5:30 PM, Connor Abbott wrote:
>>
>> On Tue, Dec 16, 2014 at 1:05 AM, Jason Ekstrand
>> wrote:
>> > ---
>> > src/glsl/nir/nir_from_ssa.c | 40 +++
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand wrote:
> This pass analizes all of the load/store operations and, when a variable is
> never aliased (potentially used by an indirect operation), it is lowered
> directly to an SSA value. This pass translates to SSA directly and does
> not require a
On Wed, Dec 17, 2014 at 9:38 PM, Jason Ekstrand wrote:
>
>
> On Wed, Dec 17, 2014 at 6:01 PM, Connor Abbott wrote:
>>
>> > +
>> > +static nir_deref *
>> > +get_deref_tail(nir_deref *deref)
>> > +{
>> > + while (deref->child !
On Wed, Dec 17, 2014 at 10:13 PM, Connor Abbott wrote:
> On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand wrote:
>> This pass analizes all of the load/store operations and, when a variable is
>> never aliased (potentially used by an indirect operation), it is lowered
>> dir
On Wed, Dec 17, 2014 at 1:59 AM, Connor Abbott wrote:
> Patches 23-26, 28, 30-35, 37-38, 40, (41 gets killed later so I didn't
> review it), 42-44 are
>
> Reviewed-by: Connor Abbott
>
> I'm going to bed now, but I'll try to do some more later.
Patches 47-48,
On Wed, Dec 17, 2014 at 10:50 PM, Jason Ekstrand wrote:
>
>
> On Wed, Dec 17, 2014 at 7:13 PM, Connor Abbott wrote:
>>
>> On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand
>> wrote:
>> > This pass analizes all of the load/store operations and, when a
On Thu, Dec 18, 2014 at 2:01 AM, Eric Anholt wrote:
> Jason Ekstrand writes:
>
>> From: Connor Abbott
>>
>> This is similar to ir_validate.cpp.
>>
>> v2: Jason Ekstrand :
>>whitespace fixes
>
> I have again not reviewed the control
On Thu, Dec 18, 2014 at 1:49 PM, Eric Anholt wrote:
> Connor Abbott writes:
>
>> On Thu, Dec 18, 2014 at 2:01 AM, Eric Anholt wrote:
>>> Jason Ekstrand writes:
>>>
>>>> From: Connor Abbott
>>>>
>>>> This is similar to ir_val
On Thu, Dec 18, 2014 at 5:01 PM, Jason Ekstrand wrote:
>
>
> On Wed, Dec 17, 2014 at 10:22 PM, Eric Anholt wrote:
>>
>> Jason Ekstrand writes:
>>
>> > From: Connor Abbott
>> >
>> > This is similar to ir_print_visitor.cpp.
>>
>>
On Thu, Dec 18, 2014 at 8:15 PM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir.h | 43 +--
> 1 file changed, 33 insertions(+), 10 deletions(-)
>
> diff --git a/src/glsl/nir/nir.h b/src/glsl/nir/nir.h
> index 19781c1..aa128ad 100644
> --- a/src/glsl/nir/nir
Reviewed-by: Connor Abbott
On Thu, Dec 18, 2014 at 9:44 PM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir.h | 53 +++--
> 1 file changed, 43 insertions(+), 10 deletions(-)
>
> diff --git a/src/glsl/nir/nir.h b/src/glsl/nir/nir.h
Ok, I'm going to try reviewing this again. I'm pasting the latest version
of the file from review/nir-v1 and replying to that so that I won't get
confused between all the various changes and reorganizing things. Here we
go!
> /*
> * Copyright © 2014 Intel Corporation
> *
> * Permission is hereb
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand
wrote:
> ---
> src/glsl/Makefile.sources | 1 +
> src/glsl/nir/nir.h| 2 +
> src/glsl/nir/nir_lower_global_vars_to_local.c | 107
> ++
> 3 files changed, 110 insertions(+)
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand
wrote:
> ---
> src/glsl/Makefile.sources | 1 +
> src/glsl/nir/nir.h | 2 +
> src/glsl/nir/nir_lower_locals_to_regs.c | 313
>
> 3 files changed, 316 insertions(+)
> create mo
Oh, and I forgot... can we rename this to lower_local_to_regs_scalar or at
least add a note that this won't work for vec4 backends yet due to the
different indexing?
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand
wrote:
> ---
> src/glsl/Makefile.sources | 1 +
> src/glsl/nir/n
I'd also like to rename or at least note that this is a scalar-only thing
for now... otherwise,
Reviewed-by: Connor Abbott
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand
wrote:
> ---
> src/glsl/Makefile.sources | 1 +
> src/glsl/nir/nir.h | 2 +
> src/glsl/n
Except for the minor stale comment and assuming you checked that we don't
call nir_create_local_reg() anymore,
Reviewed-by: Connor Abbott
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand
wrote:
> With this commit, the GLSL IR -> NIR pass generates NIR in more-or-less SSA
> form
I'm not so sure how I feel about checking that outputs are write-only...
eventually we'll want to do lower_input_reads in NIR itself, at which point
we'll need to remove that part from the validator. At the same time, for
now this is somewhat useful. I'm just not sure if it's worth it (making
sure
Reviewed-by: Connor Abbott
Nice job getting this variable lowering stuff all done!
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand
wrote:
> This commit switches us over to the new variable lowering code which is
> capable of properly handling lowering indirects as we go.
> ---
&g
Reviewed-by: Connor Abbott
Nice to see that this idea worked out well!
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand
wrote:
> We used to have the number of components built into the intrinsic. This
> meant that all of our load/store intrinsics had vec1, vec2, vec3, and vec4
>
Reviewed-by: Connor Abbott
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand
wrote:
> ---
> src/glsl/Makefile.sources |1 -
> src/glsl/nir/nir_lower_variables_scalar.c | 1249
> -
> 2 files changed, 1250 deletions(-)
> delete mo
This is a general question for the interpolation support:
Why are we using the variable-based intrinsics directly, instead of
lowering it to something index-based in the lower_io pass just like we do
for normal inputs?
On Tue, Dec 16, 2014 at 1:12 AM, Jason Ekstrand
wrote:
> ---
> src/mesa/dri
On Tue, Dec 16, 2014 at 1:12 AM, Jason Ekstrand
wrote:
> ---
> src/glsl/nir/nir_intrinsics.h | 32 +++-
> src/glsl/nir/nir_lower_io.c | 16 ++--
> 2 files changed, 21 insertions(+), 27 deletions(-)
>
> diff --git a/src/glsl/nir/nir_intrinsics.h b/src/gls
Hi,
Was it your intention to not support non-per-component things like dot
product at all? I've made a few inline comments about how to do it,
and it doesn't seem like it's that hard.
On Tue, Dec 16, 2014 at 1:12 AM, Jason Ekstrand wrote:
>
> This framework provides a simple way to do simple sea
On Tue, Dec 16, 2014 at 1:12 AM, Jason Ekstrand wrote:
> This commit builds on the nir_search.h infastructure by adds a bit of
adding
> python code that makes it stupid easy to write an algebraic transformation
> pass. The nir_algebraic.py file contains four python classes that
> correspond dir
On Tue, Dec 16, 2014 at 1:12 AM, Jason Ekstrand wrote:
> This pass uses the previously built algebraic transformations framework and
> should act as an example for anyone else wanting to make an algebraic
> transformation pass for NIR.
> ---
> src/glsl/Makefile.am | 10 -
>
On Tue, Dec 16, 2014 at 1:12 AM, Jason Ekstrand wrote:
> ---
> src/glsl/Makefile.sources| 1 +
> src/glsl/nir/nir.h | 1 +
> src/glsl/nir/nir_opt_constant_folding.c | 283
> +++
> src/mesa/drivers/dri/i965/brw_fs_nir.cpp |
>>
>> I was mislead by the "leaf" name the first time I reviewed this; having a
>> comment explaining what it does helps, but I still think that it's a pretty
>> misleading name. "Leaf," at least to me, implies that it's a leaf of the
>> dereference tree, which in this case isn't true unless I'm mi
I created nir_tex_src_sampler_index for exactly this purpose, which
fits in with the "stick all the sources in an array so we can easily
iterate over them" philosophy. If you decide to keep with this
solution, though, at least remove that.
On Tue, Dec 16, 2014 at 1:13 AM, Jason Ekstrand wrote:
>
I don't really feel qualified to review this, since I didn't write any
of the original code (I just copied-n-pasted it from what i965 was
doing...). I'm not sure who would be able to, maybe Chris Forbes since
he did the indirect sampler work?
On Tue, Dec 16, 2014 at 1:13 AM, Jason Ekstrand wrote:
Again, not my area of expertise here.
On Tue, Dec 16, 2014 at 1:13 AM, Jason Ekstrand wrote:
> ---
> src/mesa/drivers/dri/i965/brw_fs_nir.cpp | 24 ++--
> 1 file changed, 22 insertions(+), 2 deletions(-)
>
> diff --git a/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
> b/src/mesa/
The commit message should probably say something like "nir: Make
nir_ssa_undef_instr_create initialize the destination". That we now
need to pass in the number of components is just a side effect.
On Tue, Dec 16, 2014 at 1:13 AM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir.c |
On Wed, Dec 17, 2014 at 8:04 PM, Jason Ekstrand wrote:
> Previously, we were doing a lazy creation of the parallel copy
> instructions. This is confusing, hard to get right, and involves some
> extra state tracking of the copies. This commit adds an extra walk over
> the basic blocks to add the
Can you make the fields lowercase to match the algebraic properties
enum? Or did you make that uppercase? Lowercase feels better to me
since it's an enum, but I don't really care too much.
On Fri, Dec 19, 2014 at 8:02 PM, Jason Ekstrand wrote:
> This should be much better for debugging as GDB wil
Patches 100, 102-104, 106-109, 114-117, 121-146 are:
Reviewed-by: Connor Abbott
Whew! I think I've looked through everything.
On Tue, Dec 16, 2014 at 1:04 AM, Jason Ekstrand
wrote:
> NIR (pronounced "ner") is a new IR (internal representation) for the Mesa
> shader c
On Tue, Jan 6, 2015 at 6:36 PM, Jason Ekstrand wrote:
>
>
> On Mon, Jan 5, 2015 at 10:45 PM, Connor Abbott wrote:
>>
>> I created nir_tex_src_sampler_index for exactly this purpose, which
>> fits in with the "stick all the sources in an array so we can easily
>
On Tue, Jan 6, 2015 at 6:27 PM, Jason Ekstrand wrote:
>
>
> On Mon, Jan 5, 2015 at 11:26 PM, Connor Abbott wrote:
>>
>> Can you make the fields lowercase to match the algebraic properties
>> enum? Or did you make that uppercase? Lowercase feels better to me
>>
On Sat, Jan 3, 2015 at 2:18 PM, Thomas Helland
wrote:
> Also handle undefined behaviour for sqrt(x) where x < 0
> and rsq(x) where x <= 0.
>
> This gives us some reduction in instruction count on three
> Dungeon Defenders shaders as they are doing: max(exp(x), 0)
So initially when you said that D
;
> + break;
> + default:
> + assert("Invalid deref chain");
> return false;
> - break;
> - }
> - case nir_deref_type_struct:
> - if (nir_deref_as_struct(a)->index != nir_deref_as_struct(b)->index)
> + }
> +
> + assert((a->child == NULL) == (b->child == NULL));
> + if((a->child == NULL) != (b->child == NULL))
> return false;
Hmm, if you're already doing the assert I don't think the if is really
necessary?
> - break;
> - default:
> - unreachable("Invalid dreference type");
> }
>
> - assert((a->child == NULL) == (b->child == NULL));
> - if (a->child)
> - return derefs_equal(a->child, b->child);
> - else
> - return true;
> + return true;
> }
>
> static int
> --
> 2.2.0
>
> ___
> mesa-dev mailing list
> mesa-dev@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Other than the a few minor things,
Reviewed-by: Connor Abbott
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
With Matt's suggested commit message,
Reviewed-by: Connor Abbott
On Tue, Jan 6, 2015 at 7:34 PM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir.h | 4
> src/glsl/nir/nir_validate.c | 7 +++
> 2 files changed, 11 insertions(+)
>
> diff --git a/src/gl
Whoops... I was looking over the actual file (instead of the patch),
and there's a comment above (at least in lower_variables) about magic
numbers that needs to be deleted or rewritten.
On Fri, Jan 9, 2015 at 1:30 PM, Connor Abbott wrote:
> As you mentioned, you should add somethin
On Sun, Jan 4, 2015 at 4:01 PM, Connor Abbott wrote:
[...]
>> static bool
>> deref_may_be_aliased(nir_deref_var *deref,
>> struct lower_variables_state *state)
>> {
>>nir_deref_var var_deref = *deref;
>>var_deref.deref.child =
On Tue, Jan 6, 2015 at 7:34 PM, Jason Ekstrand wrote:
> Additional description was added to a variety of places. Also, we no
> longer use the term "leaf" to describe fully-qualified direct derefs.
> Instead, we simply use the term "direct" or spell it out completely.
> ---
> src/glsl/nir/nir_low
Reviewed-by: Connor Abbott
On Wed, Jan 7, 2015 at 9:03 PM, Jason Ekstrand wrote:
> In particular, we rename nir_tex_src_sampler_index to _sampler_offset and
> add a sampler_array_size field to nir_tex_instr. This way we can pass the
> size of sampler arrays through to backends e
BTW, now that we're adding an extra possible source, do we need to
expand the size of the source array?
Anyways,
Reviewed-by: Connor Abbott
On Wed, Jan 7, 2015 at 9:03 PM, Jason Ekstrand wrote:
> This helps a lot with things like lowering passes that may need to add
> sources.
&
On Fri, Jan 9, 2015 at 5:41 PM, Jason Ekstrand wrote:
>
>
> On Fri, Jan 9, 2015 at 11:34 AM, Connor Abbott wrote:
>>
>> BTW, now that we're adding an extra possible source, do we need to
>> expand the size of the source array?
>
>
> I don't know that
+ case nir_intrinsic_copy_var:
+ unreachable("There should be no copies whatsoever at this
point");
+ break;
>>>
>>>
>>> Are you sure about this? My impression is that lower_variables will lower
>>> copies involving things that aren't indirectly referenced,
Ok, with the assert removed,
Reviewed-by: Connor Abbott
On Fri, Jan 9, 2015 at 11:16 PM, Jason Ekstrand wrote:
>
>
> On Fri, Jan 9, 2015 at 8:04 PM, Jason Ekstrand wrote:
>>
>> This solves a number of problems. First is the ability to change the
>> numbe
On Fri, Jan 9, 2015 at 11:04 PM, Jason Ekstrand wrote:
> This solves a number of problems. First is the ability to change the
> number of sources that a texture instruction has. Second, it solves the
> delema that may occur if a texture instruction has more than 4 sources.
> ---
> src/glsl/nir/
On Tue, Jan 6, 2015 at 1:04 AM, Jason Ekstrand wrote:
>
>
> On Sun, Jan 4, 2015 at 9:15 PM, Connor Abbott wrote:
>>
>> This is a general question for the interpolation support:
>>
>> Why are we using the variable-based intrinsics directly, instead of
>>
o standard because of D3D. It
seems worth preserving this behavior, and more generally being careful
about NaN's when doing range analysis (unless the hardware is known to
never return NaN's), to keep games from potentially breaking on edge
cases like these.
>
> On Fri, Jan 9, 2015
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand wrote:
> ---
> src/glsl/Makefile.sources | 1 +
> src/glsl/nir/nir.h | 2 +
> src/glsl/nir/nir_split_var_copies.c | 225
>
> 3 files changed, 228 insertions(+)
> create mode 10064
On Tue, Jan 6, 2015 at 5:21 PM, Jason Ekstrand wrote:
>
>
> On Mon, Jan 5, 2015 at 10:00 PM, Jason Ekstrand
> wrote:
>>
>>
>>
>> On Mon, Jan 5, 2015 at 9:12 PM, Connor Abbott wrote:
>>>
>>> Hi,
>>>
>>> Was it your inten
Reviewed-by: Connor Abbott
On Tue, Jan 6, 2015 at 7:34 PM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir_lower_variables.c | 111
> +
> 1 file changed, 50 insertions(+), 61 deletions(-)
>
> diff --git a/src/glsl/nir/nir_lower_variables.c
actly how that
> all gets hooked up for other gallium drivers beyond vc4 is outside the
> scope of this series.
>
> I have pushed a branch to my personal freedesktop.org account. For certain
> types of review, it may be easier to look at the end result rather than the
> patc
On Wed, Jan 14, 2015 at 3:36 PM, Jason Ekstrand wrote:
>
>
> On Fri, Jan 9, 2015 at 11:31 AM, Connor Abbott wrote:
>>
>> On Tue, Jan 6, 2015 at 7:34 PM, Jason Ekstrand
>> wrote:
>> > Additional description was added to a variety of places. Also, we n
On Fri, Jan 9, 2015 at 8:27 PM, Jason Ekstrand wrote:
>
>
> On Fri, Jan 9, 2015 at 4:38 PM, Connor Abbott wrote:
>>
>> >>>> + case nir_intrinsic_copy_var:
>> >>>> + unreachable("There should be no copies wha
Reviewed-by: Connor Abbott
On Wed, Jan 14, 2015 at 3:43 PM, Jason Ekstrand wrote:
> The original name wasn't particularly descriptive. This one indicates that
> it actually gives you SSA values as opposed to the old pass which lowered
> variables to registers.
>
With this squashed in, patch 90 is
Reviewed-by: Connor Abbott
On Wed, Jan 14, 2015 at 2:57 PM, Jason Ekstrand wrote:
> ---
> src/glsl/nir/nir_split_var_copies.c | 96
> ++---
> 1 file changed, 78 insertions(+), 18 deletions(-)
>
> diff --
Reviewed-by: Connor Abbott
On Wed, Jan 14, 2015 at 5:02 PM, Jason Ekstrand wrote:
> This refactor allows you to more easily get the deref node associated with
> a given variable. We then use that new functionality in the
> deref_may_be_aliased function instead of creating a 1-elem
On Wed, Jan 14, 2015 at 5:05 PM, Jason Ekstrand wrote:
>
>
> On Wed, Jan 14, 2015 at 1:05 PM, Connor Abbott wrote:
>>
>> On Wed, Jan 14, 2015 at 3:36 PM, Jason Ekstrand
>> wrote:
>> >
>> >
>> > On Fri, Jan 9, 2015 at 11:31 AM, Connor Abbot
On Wed, Jan 14, 2015 at 5:15 PM, Jason Ekstrand wrote:
>
>
> On Sun, Jan 11, 2015 at 7:08 PM, Connor Abbott wrote:
>>
>> On Tue, Jan 6, 2015 at 5:21 PM, Jason Ekstrand
>> wrote:
>> >
>> >
>> > On Mon, Jan 5, 2015 at 10:00 PM, Jason Ekstr
This, patch 157, and patch 158 are
Reviewed-by: Connor Abbott
On Wed, Jan 14, 2015 at 6:28 PM, Jason Ekstrand wrote:
> ---
> src/glsl/Makefile.sources | 1 +
> src/glsl/nir/nir.h | 3 +
> src/glsl/nir/nir_lower_var_co
Together with the patches to lower all wildcard copies,
Reviewed-by: Connor Abbott
On Tue, Dec 16, 2014 at 1:11 AM, Jason Ekstrand wrote:
> ---
> src/glsl/Makefile.sources | 1 +
> src/glsl/nir/nir.h | 2 +
> src/glsl/nir/nir_lower_locals_to_
On Wed, Jan 14, 2015 at 6:53 PM, Jason Ekstrand wrote:
>
>
> On Sun, Jan 11, 2015 at 7:52 PM, Connor Abbott wrote:
>>
>> Patches without my Reviewed-by on them on your branch are:
>>
>> "i965/fs_nir: Add support for sample_pos and sample_id"
>>
&g
Reviewed-by: Connor Abbott
Just curious, how did this come about?
On Wed, Jan 14, 2015 at 10:15 PM, Jason Ekstrand wrote:
> Without the break, it was possible that an instruction would match multiple
> expressions. If this happened, you could end up trying to replace it
> multiple
Series is
Reviewed-by: Connor Abbott
As you know, I have a branch that generalizes this and adds a worklist
for SSA definitions as well. But I think we won't want the SSA-def
worklist for DCE, since just like with phi-node placement we only ever
push something onto the worklist once, and w
On Thu, Jan 15, 2015 at 5:28 PM, Brian Paul wrote:
> Warning seen with gcc 4.8.2
> ---
> src/glsl/nir/nir.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/glsl/nir/nir.c b/src/glsl/nir/nir.c
> index f112db8..6d6f910 100644
> --- a/src/glsl/nir/nir.c
> +++ b/src/glsl/
Reviewed-by: Connor Abbott
On Thu, Jan 15, 2015 at 6:16 PM, Brian Paul wrote:
> v2: use proper argument
> ---
> src/glsl/nir/nir.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/glsl/nir/nir.c b/src/glsl/nir/nir.c
> index f112db8..81dec1c 10
Btw, apparently nir_validate.c has the same problem on line 391.
On Thu, Jan 15, 2015 at 6:31 PM, Connor Abbott wrote:
> Reviewed-by: Connor Abbott
>
> On Thu, Jan 15, 2015 at 6:16 PM, Brian Paul wrote:
>> v2: use proper argument
>> ---
>> src/glsl/nir/nir.c |
Reviewed-by: Connor Abbott
On Thu, Jan 15, 2015 at 7:39 PM, Brian Paul wrote:
> Silences a compiler warning.
> ---
> src/glsl/nir/nir_validate.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/glsl/nir/nir_validate.c b/src/glsl/nir/nir_validate
401 - 500 of 1264 matches
Mail list logo
