The screen already has a pointer to the (base) winsys object.
With the latter of which implemented/sub-classed as either drm or sw
based one, depending on the target.
Signed-off-by: Emil Velikov
---
src/gallium/drivers/virgl/virgl_public.h | 1 -
src/gallium/drivers/virgl/virgl_screen.c | 1
Signed-off-by: Emil Velikov
---
src/gallium/winsys/virgl/drm/virgl_drm_winsys.h | 2 --
1 file changed, 2 deletions(-)
diff --git a/src/gallium/winsys/virgl/drm/virgl_drm_winsys.h
b/src/gallium/winsys/virgl/drm/virgl_drm_winsys.h
index eac1d3e..a547654 100644
--- a/src/gallium/winsys/virgl/drm
The only two remaining cases of (struct virgl_resource *) require a
closer look. Either the error checking is missing or the arguments
provided feel wrong.
Signed-off-by: Emil Velikov
---
src/gallium/drivers/virgl/virgl_buffer.c| 2 +-
src/gallium/drivers/virgl/virgl_context.c | 28
Signed-off-by: Emil Velikov
---
src/gallium/winsys/virgl/drm/virgl_drm_winsys.c | 10 +-
src/gallium/winsys/virgl/drm/virgl_drm_winsys.h | 14 ++
src/gallium/winsys/virgl/vtest/virgl_vtest_winsys.c | 10 +-
src/gallium/winsys/virgl/vtest/virgl_vtest_winsys.h
The uppercase versions are wrappers which must be matched.
Signed-off-by: Emil Velikov
---
src/gallium/winsys/virgl/drm/virgl_drm_winsys.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/winsys/virgl/drm/virgl_drm_winsys.c
b/src/gallium/winsys/virgl/drm
Signed-off-by: Emil Velikov
---
src/gallium/winsys/virgl/drm/virgl_drm_winsys.c | 12 +++-
1 file changed, 3 insertions(+), 9 deletions(-)
diff --git a/src/gallium/winsys/virgl/drm/virgl_drm_winsys.c
b/src/gallium/winsys/virgl/drm/virgl_drm_winsys.c
index d9b4d58..0616de3 100644
--- a
Signed-off-by: Emil Velikov
---
src/gallium/winsys/virgl/drm/virgl_drm_winsys.c | 5 ++---
src/gallium/winsys/virgl/vtest/virgl_vtest_winsys.c | 3 +--
2 files changed, 3 insertions(+), 5 deletions(-)
diff --git a/src/gallium/winsys/virgl/drm/virgl_drm_winsys.c
b/src/gallium/winsys/virgl
Signed-off-by: Emil Velikov
---
src/gallium/drivers/virgl/virgl_context.c | 2 +-
src/gallium/drivers/virgl/virgl_encode.c | 4 ++--
src/gallium/drivers/virgl/virgl_encode.h | 5 +
src/gallium/drivers/virgl/virgl_resource.c | 4 ++--
src/gallium/drivers/virgl/virgl_screen.h | 2 +-
5
Signed-off-by: Emil Velikov
---
src/gallium/drivers/virgl/virgl_buffer.c| 6 +-
src/gallium/drivers/virgl/virgl_context.c | 98 ++---
src/gallium/drivers/virgl/virgl_context.h | 17 +
src/gallium/drivers/virgl/virgl_encode.c| 2 +-
src/gallium/drivers
Signed-off-by: Emil Velikov
---
src/gallium/drivers/virgl/virgl_query.c | 15 ++-
1 file changed, 10 insertions(+), 5 deletions(-)
diff --git a/src/gallium/drivers/virgl/virgl_query.c
b/src/gallium/drivers/virgl/virgl_query.c
index ea50f2f..b06635b 100644
--- a/src/gallium/drivers
Hi Emmanuel,
On 29 October 2015 at 15:22, Emmanuel Gil Peyrot
wrote:
> This was causing compilation issues when one of its providers wasn’t
> already included before gbm.h.
Cc: "11.0"
Reviewed-by: Emil Velikov
I'll push this later on today. For future patches please in
On 29 October 2015 at 17:15, Grazvydas Ignotas wrote:
> Hi,
>
> On Thu, Oct 29, 2015 at 6:21 PM, Emil Velikov
> wrote:
>> On 29 October 2015 at 15:22, Emmanuel Gil Peyrot
>> wrote:
>>> This was causing compilation issues when one of its providers wasn’t
>>
it so might as well not bother
with it for now.
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
for (i = 0; i < num_surfaces; i++) {
> + if (surfaces[i] != VA_INVALID_ID)
Should have caught this one sooner - this looks odd for couple of reasons.
- vlVaDestroySurfaces removes a total of i surfaces. Thus we don't
need a loop here.
- handle_table_add() returns 0 on error and
VA_INVALID_ID;
> + img->width = surf->buffer->width;
> + img->height = surf->buffer->height;
> + img->num_palette_entries = 0;
> + img->entry_bytes = 0;
> + w = align(surf->buffer->width, 2);
> + h = align(surf->buffer->
t; + }
>
>
> It would be nice to have this snip-set in something like a vl_winsys_drm.c
> if possible, but that's not a hard requirement and can happen when we
> actually need this elsewhere as well.
>
I doubt we'll need it elsewhere any time soon, so whenever one
On 20 October 2015 at 17:40, Arnaud Vrac wrote:
> On Tue, Oct 20, 2015 at 6:35 PM, Emil Velikov
> wrote:
>>
>> On 20 October 2015 at 17:06, Julien Isorce
>> wrote:
>> >
>> >
>> > On 19 October 2015 at 17:16, Emil Velikov
>> > w
On 29 October 2015 at 23:04, Arnaud Vrac wrote:
> On Thu, Oct 29, 2015 at 11:28 PM, Emil Velikov
> wrote:
>>
>> On 20 October 2015 at 17:40, Arnaud Vrac wrote:
>> > On Tue, Oct 20, 2015 at 6:35 PM, Emil Velikov
>> > wrote:
>> >>
>> >
rry about that, I'll be more careful about that in the future.
>
Fwiw some of us tend to add the revision based on which the r-b was
given, when the revision history is present in the commit message.
Something like:
v2: Changed foo and bar.
Reviewed-by: Slim Shady (v1)
Then again, you might want to follow your colleague's suggestion.
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 30 October 2015 at 09:59, Julien Isorce wrote:
>
>
> -Original Message-
> From: Emil Velikov [mailto:emil.l.veli...@gmail.com]
> Sent: 29 October 2015 19:04
> To: Julien Isorce
> Cc: ML mesa-dev
> Subject: Re: [Mesa-dev] [PATCH v4 3/9] st/va: impleme
On 29 October 2015 at 21:50, Dave Airlie wrote:
> On 29 October 2015 at 21:12, Emil Velikov wrote:
>> On 29 October 2015 at 11:12, Emil Velikov wrote:
>>> Hi all,
>>>
>>> A slightly longer series (that builds on top of the previous sent a
>>> minu
On 19 October 2015 at 18:41, Emil Velikov wrote:
> On 19 October 2015 at 17:07, Brian Paul wrote:
>>
>> I'm not too familiar with this code or these changes but I'm wondering how
>> much of chance there is of this breaking any driver/target builds.
>>
if there's
> anything useful in there that could use the same codepaths.
>
Bth, I'm not sure if one can share much with the dri backed loaders
(libgl/libegl/libgbm) - on the drm/gbm side from we (vl) don't require
anything special.
In theory the idea is great, in practise I t
upon before
NIR got merged ? Perhaps it's worth porting them over ?
Thanks
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 2 November 2015 at 19:41, Jason Ekstrand wrote:
> On Mon, Nov 2, 2015 at 9:33 AM, Connor Abbott wrote:
>> On Mon, Nov 2, 2015 at 8:35 AM, Emil Velikov
>> wrote:
>>> Hi all,
>>>
>>> From a quick look, it seems that NIR copies (almost ?) all the sta
e https://gcc.gnu.org/onlinedocs/gcc/Unnamed-Fields.html
FYI on old versions of GCC where this extension is not enabled by
default, we flip in on. We're using -std=gnu99, although perhaps we
should use -fms-extensions - not a big deal either way.
Btw I had an almost identical patch locally :
On 3 November 2015 at 00:29, Matt Turner wrote:
> Put fields that are meaningless with an immediate in the same storage
> with the immediate.
There is something funky here. Should it be
"Put fields that are meaningless _without_ an immediate in the same storage
_as_ the immediate
RF:
> @@ -48,7 +49,7 @@ static uint32_t brw_file_from_reg(fs_reg *reg)
> case UNIFORM:
>unreachable("not reached");
> }
> - return 0;
> + return GRF;
Although unreachable (and gcc being silly) GRF looks wrong. Use
BRW_ARCHITECTURE_REGISTER_FILE perha
reg.writemask = inst->dst.writemask;
>
> -inst->dst.file = HW_REG;
> -inst->dst.fixed_hw_reg = reg;
> + inst->dst = reg;
Same concern, as in fs_visitor::assign_curb_setup. Past the struct
brw_reg, dst_reg will be uninitialized. Or is the
dst_reg::ds
OPERATORS macro and fs_reg::fs_reg()
kick in ? If not things look quite fragile and perhaps we should wire
them up.
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
r_file)reg.file;
You're not adding the cast in the rest of the ctors (be that fs_reg,
src_reg or dst_reg) so might as well drop this one ?
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
reg(struct brw_reg reg) :
> backend_reg(reg)
> {
> - this->file = (enum register_file)reg.file;
Should we fold the remaining this->file = foo into the backend_reg() ctors ?
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop
n as a replacement for the
> vector float src_reg/fs_reg constructors.
Yes please. If we have any floats we can easily convert them with the
existing brw_float_to_vf().
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
;
> + unsigned vf;
> + memcpy(&vf, imm, sizeof(vf));
> + vec4_instruction *mov = MOV(inst->dst, brw_imm_vf(vf));
You can drop the temp variable + memcpy call and use brw_imm_vf4(imm[x],)
-Emil
___
mesa-dev mailing
move reladdr to backend_reg. Although
that'll be obviously follow up patches.
Fwiw I really like the whole series and barring a few small comments I
would love to see it land.
Big thanks for doing this.
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
eventually be dropped. We've gone through this several times in
> the past.
>
Out of curiosity: doesn't the existing scons build allow you to do all
sort of builds or there is something special that can be done with the
project files ?
Thanks
Emil
__
or now ?
Additionally dropping the GALLIUM_HUD_VISIBLE will eliminate some of
the issues pointed out so far. One can easily toggle off as the
application is up.
Regards,
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 3 November 2015 at 17:27, Matt Turner wrote:
> On Tue, Nov 3, 2015 at 2:38 AM, Emil Velikov wrote:
>> On 3 November 2015 at 00:29, Matt Turner wrote:
>>> Put fields that are meaningless with an immediate in the same storage
>>> with the immediate.
>> There i
On 3 November 2015 at 18:02, Matt Turner wrote:
> On Tue, Nov 3, 2015 at 8:05 AM, Emil Velikov wrote:
>> On 3 November 2015 at 00:29, Matt Turner wrote:
>>> The test (file == BAD_FILE) works on registers for which the constructor
>>> has not run because BAD_FILE is
On 3 November 2015 at 17:55, Matt Turner wrote:
> On Tue, Nov 3, 2015 at 8:04 AM, Emil Velikov wrote:
>> On 3 November 2015 at 00:29, Matt Turner wrote:
>>
>> Please add a bit of commit message - "Mostly unused as of last commit.
>> Fold the remaining cases (GR
On 3 November 2015 at 18:10, Matt Turner wrote:
> On Tue, Nov 3, 2015 at 8:07 AM, Emil Velikov wrote:
>> On 3 November 2015 at 00:29, Matt Turner wrote:
>>
>>> index 6eeafd5..3d2b051 100644
>>> --- a/src/mesa/drivers/dri/i965/brw_fs.cpp
>>>
On 3 November 2015 at 18:20, Matt Turner wrote:
> On Tue, Nov 3, 2015 at 8:09 AM, Emil Velikov wrote:
>> On 3 November 2015 at 00:29, Matt Turner wrote:
>>
>>> @@ -387,7 +342,9 @@ vec4_visitor::opt_vector_float()
>>>
>>>remaining_chann
can you please provide a fix against 11.0 if we still want this
in the stable branch.
Thanks
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
From: Emil Velikov
Signed-off-by: Emil Velikov
---
src/mesa/drivers/dri/i965/brw_vec4.cpp | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/mesa/drivers/dri/i965/brw_vec4.cpp
b/src/mesa/drivers/dri/i965/brw_vec4.cpp
index e2f8a0a..6975a19 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4
efer we can make them non-inline.
The lot can be found in branch bye-bye-memset over at
https://github.com/evelikov/Mesa
Cheers,
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
From: Emil Velikov
By using a default value for brw_reg_type the compiler will reuse
the existing constructor.
Signed-off-by: Emil Velikov
---
src/mesa/drivers/dri/i965/brw_fs.cpp | 9 -
src/mesa/drivers/dri/i965/brw_ir_fs.h | 4 ++--
2 files changed, 2 insertions(+), 11 deletions
From: Emil Velikov
Remove unneeded this->file assignment in dst_reg(const src_reg &f) ctor
Signed-off-by: Emil Velikov
---
src/mesa/drivers/dri/i965/brw_vec4.cpp | 1 -
1 file changed, 1 deletion(-)
diff --git a/src/mesa/drivers/dri/i965/brw_vec4.cpp
b/src/mesa/drivers/dri/i965/brw_v
From: Emil Velikov
By providing default values for type and writemask, the compiler will
'create' the constructor for us. The values chosen are the exact same
ones, used in the default ctor - dst_reg(), as such the compiler should
be smart enough to optimise the duplicated (identical)
From: Emil Velikov
As of last commit one can easily the whole brw_reg as two unsigned
values. Lets use that and avoid the nastylooking memcmp + casts.
XXX: reladdr looks a bit strange. we tend to do dst_reg<>src_reg and in
some cases this might cause someheadaches. should we zero it
From: Emil Velikov
Rather than doing memset(), implicitly use the backend_reg ctor(s) which
will initialise its variables. For the fs_reg ones, set them manually.
Signed-off-by: Emil Velikov
---
src/mesa/drivers/dri/i965/brw_fs.cpp | 18 +++---
src/mesa/drivers/dri/i965
From: Emil Velikov
Just fold the brw_type_for_base_type() and use a delegated
constructor.
Note this is a c++11 feature and the compiler will warn us if it's
set/defaults to an earlier version of the standard.
XXX: Should we just toggle c++11 on ? We seem to be using some of its
function
From: Emil Velikov
Signed-off-by: Emil Velikov
---
src/mesa/drivers/dri/i965/brw_ir_vec4.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_ir_vec4.h
b/src/mesa/drivers/dri/i965/brw_ir_vec4.h
index b70dca5..2c7293d 100644
--- a/src/mesa
From: Emil Velikov
Just default the size to 1, and let the compiler do its magic.
XXX: Not 100% sure this is correct.
Signed-off-by: Emil Velikov
---
src/mesa/drivers/dri/i965/brw_ir_vec4.h| 3 +--
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 19 +++
2 files
From: Emil Velikov
This commits partially reverts "i965: Initialize registers' file to
BAD_FILE."
No longer needed as of commit "i965/fs: properly construct fs_reg" which
removes the memset(...0...), and correctly sets BAD_FILE.
Signed-off-by: Emil Velikov
---
From: Emil Velikov
Step one towards having a normal constructors for the *_reg primitives,
and removing a handful of somewhat ugly code.
Signed-off-by: Emil Velikov
---
src/mesa/drivers/dri/i965/brw_fs.cpp | 2 +-
src/mesa/drivers/dri/i965/brw_shader.h | 11 +--
src/mesa/drivers
From: Emil Velikov
Will allow us an easy once-off get/set of the attribs. Thus we can drop
all the memset/memcmp + magic casting.
XXX: alternative name(s) are welcome
Signed-off-by: Emil Velikov
---
src/mesa/drivers/dri/i965/brw_reg.h | 21 +
1 file changed, 13 insertions
From: Emil Velikov
Analogous to the previous (fs) commit - drop the explicit memset and
set the non-inherited variables (reladdr) manually.
Signed-off-by: Emil Velikov
---
src/mesa/drivers/dri/i965/brw_ir_vec4.h| 13 +++
src/mesa/drivers/dri/i965/brw_vec4.cpp | 48
On 3 November 2015 at 18:02, Matt Turner wrote:
> On Tue, Nov 3, 2015 at 8:05 AM, Emil Velikov wrote:
>> On 3 November 2015 at 00:29, Matt Turner wrote:
>>> The test (file == BAD_FILE) works on registers for which the constructor
>>> has not run because BAD_FILE is
On 3 November 2015 at 16:04, Emil Velikov wrote:
>> @@ -1596,8 +1586,7 @@ vec4_visitor::lower_attributes_to_hw_regs(const int
>> *attribute_map,
>> reg.type = inst->dst.type;
>> reg.writemask = inst->dst.writemask;
>>
>> -
> --- a/src/gallium/state_trackers/va/va_private.h
> +++ b/src/gallium/state_trackers/va/va_private.h
> @@ -49,6 +49,7 @@
> #define VL_VA_PSCREEN(ctx) (VL_VA_DRIVER(ctx)->vscreen->pscreen)
>
> #define VL_VA_MAX_IMAGE_FORMATS 9
> +#define VL_VA_MAX_SURFACE_ATTRIB
On 5 November 2015 at 08:24, Julien Isorce wrote:
> Some lines were using 4 indentation spaces instead of 3.
>
> The switch in vlVaAcquireBufferHandle actually had wrong brackets
> surrounding case+default.
>
Please don't mix whitespace (trivial) patches
On 5 November 2015 at 16:18, Emil Velikov wrote:
> From: Emil Velikov
>
> This commits partially reverts "i965: Initialize registers' file to
> BAD_FILE."
>
> No longer needed as of commit "i965/fs: properly construct fs_reg" which
> removes th
debug_state_int(ctx, cap, state);
There is another similar hunk in _mesa_IsEnabled. With that the patch is
Reviewed-by: Emil Velikov
I will send a "append KHR to function names when in ES context" patch
and with that we can enable the extension.
Thanks for bringing this up !
Emil
__
As defined in the spec
when implemented in an OpenGL ES context, all entry points defined
by this extension must have a "KHR" suffix.
Signed-off-by: Emil Velikov
---
The final piece afaict for KHR_debug + ES contexts.
Mildly related - the enum tests seems somewhat busted/
As defined in the spec
when implemented in an OpenGL ES context, all entry points defined
by this extension must have a "KHR" suffix.
Signed-off-by: Emil Velikov
---
Actually build tested this time.
src/mesa/main/errors.c | 40 ---
On 5 November 2015 at 19:15, Matt Turner wrote:
> On Thu, Nov 5, 2015 at 8:18 AM, Emil Velikov wrote:
>> From: Emil Velikov
>>
>> This commits partially reverts "i965: Initialize registers' file to
>> BAD_FILE."
>>
>> No longer needed a
On 5 November 2015 at 19:04, Matt Turner wrote:
> On Thu, Nov 5, 2015 at 8:17 AM, Emil Velikov wrote:
>> From: Emil Velikov
>>
>> Just fold the brw_type_for_base_type() and use a delegated
>> constructor.
>>
>> Note this is a c++11 feature and the compil
On 5 November 2015 at 16:17, Emil Velikov wrote:
> From: Emil Velikov
>
> As of last commit one can easily the whole brw_reg as two unsigned
The about above should read
"... one can easily _access_ the whole..."
-Emil
___
mesa-de
On 5 November 2015 at 22:06, Matt Turner wrote:
> On Thu, Nov 5, 2015 at 12:24 PM, Emil Velikov
> wrote:
>> On 5 November 2015 at 19:15, Matt Turner wrote:
>>> On Thu, Nov 5, 2015 at 8:18 AM, Emil Velikov
>>> wrote:
>>>> From: Emil Velikov
>
> +#include "pipe-loader/pipe_loader.h"
Please add the following include.
#include "state_tracker/drm_driver.h"
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
se {
omx_display = XOpenDisplay()
vl_screen_create()
}
> }
>
> if (!omx_screen) {
> - omx_screen = vl_screen_create(omx_display, 0);
> - if (!omx_screen) {
> - pipe_mutex_unlock(omx_lock);
> - return NULL;
> - }
> + if (omx_render
#3 and #4 after addressing remarks + split.
>
I've pushed the first two patches. Thanks again for working on these !
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
; i < VL_VA_MPEG4_BYTES_FOR_LOOKUP ; i++) {
> + if (memcmp (p, start_code, sizeof(start_code)) == 0) {
> +found = true;
Just use startcode_available directly ?
> +break;
> + }
> + p += 1;
> + extraSize += 1;
> + }
> + if (found) {
> + startcode_available = true;
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
Seems that we've been forgetting to update this test, leading to
decreased usefulness. Thus lets add a test which walks over the mesa
generated enums and checks the string (if non "0x" prefixed) against the
local table.
Signed-off-by: Emil Velikov
---
I'm not 100% sure on
c4: Replace src_reg(imm) constructors with brw_imm_*().
Keep the memcpy(), although we really should address the uninitialised
imm[] warning/bug. In a separate commit of course and optionally nuke
the memcpy.
With these and a jenkins spin to catch any subtle bugs the series is
Reviewed-by: Emil Vel
oord_conventions",
> o(ARB_fragment_coord_conventions), GL, 2009 },
Please add gl_extensions::ARB_enhanced_layouts and use it in the above
two tables. Otherwise the extension override won't work. We can always
nuke it at the end, if it turns out that this extension can be enabled
everywhere.
Thanks
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
nally I would split the validate_layout_qualifiers() introduction
and the CompileStatus/InfoLog movement into separate patches.
Feel free to ignore, if others don't share my sentiment
-Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
> - this->layout.stream);
> + int qual_stream;
> + if (process_qualifier_constant(state, &loc, "stream", this->layout.stream,
> + &qual_stream)) {
> + validate_stream_qualifier(&loc, state, qual_stream);
Didn
On 6 November 2015 at 03:19, Liu, Leo wrote:
>>-Original Message-
>>From: Emil Velikov [mailto:emil.l.veli...@gmail.com]
>>Sent: Thursday, November 05, 2015 6:30 PM
>>To: Liu, Leo
>>Cc: ML mesa-dev
>>Subject: Re: [Mesa-dev] [PATCH 4/4] st/omx: add head
Hello Leo,
On 6 November 2015 at 16:00, Liu, Leo wrote:
> Hi Emil,
>
>>Shorten the variable name - OMX_RENDER_NODE ? debug_get_option is only
>>available on debug builds so better use getenv.
>>
>
> The debug_get_*_option() functions are used all around mesa, a
ing -count
> 1023.
I haven't been following on the VBO patches that you've sent recently.
Do you think that any of them are mesa-stable material ? If so can you
please forward the sha's the the ML (or just list them here).
Thanks
Emil
_
lay)
> + XCloseDisplay(omx_display);
Missing close(fd) ?
> + pipe_mutex_unlock(omx_lock);
> + return NULL;
> }
>
> void omx_put_screen(void)
> {
> pipe_mutex_lock(omx_lock);
> if ((--omx_usecount) == 0) {
> - vl_screen_destroy(omx_screen);
> - XCloseDisplay(
creen*
> +vl_drm_screen_create(int fd)
> +{
> + struct vl_screen *vscreen;
> +
> + vscreen = CALLOC_STRUCT(vl_screen);
> + if (!vscreen)
> + return NULL;
> +
> +#if GALLIUM_STATIC_TARGETS
> + vscreen->pscreen = dd_create_screen(fd);
> +#else
> + if (pipe_loader_drm_probe_fd(&vscreen->dev, fd)) {
Add a dup() in the above. So that it reads
if (pipe_loader_drm_probe_fd(&vscreen->dev, dup(fd))) {
Thanks
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
ing the 11.0 series - thank you for the mesa-stable tag.
Fire^WCommit at will
Regards,
Emil
P.S. Is it me or does writing commit messages feels like pulling
teeth, sometimes ?
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freede
src/glsl/nir
LOCAL_GENERATED_SOURCES += $(addprefix $(intermediates)/, \
$(LIBGLCPP_GENERATED_FILES) \
Formatting might be broken (thanks gmail), although the gist is there.
Can you give it a try (note the order is important)
Thanks
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
ds
>>
>>if (pipe_loader_drm_probe_fd(&vscreen->dev, dup(fd))) {
>>
>
> The dup is handled is in st/va/context.c before calling vl_drm_screen_create
>
Please, move it here ? There little point in getting more of these
ugl
On 6 November 2015 at 19:26, Marek Olšák wrote:
> On Fri, Nov 6, 2015 at 8:24 PM, Marek Olšák wrote:
>> On Thu, Oct 15, 2015 at 2:28 PM, Emil Velikov
>> wrote:
>>> On 3 October 2015 at 12:19, Emil Velikov wrote:
>>>> On 3 October 2015 at 02:12, Marek Olšá
On 6 November 2015 at 17:50, Brian Paul wrote:
> On 11/06/2015 09:40 AM, Emil Velikov wrote:
>>
>> Hi Brian,
>>
>> On 31 October 2015 at 13:38, Brian Paul wrote:
>>>
>>> Very long line loops which spanned 3 or more vertex buffers were not
>>
04th 2015 - Release candidate 3
December 11th 2015 - Release candidate 4/Mesa 11.1.0
We have roughly _two_ weeks to get any more new features in.
Thanks
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/lis
u have any questions or comments that you would like to share
before the release, please go ahead.
Cheers,
Emil
Mesa stable queue
-
Nominated (13)
==
Ben Widawsky (1):
i965/skl/gt4: Fix URB programming restriction.
Boyan Ding (1):
i915: Ad
the whole piglit run :(
Are you seeing a similar thing or there is something funny with my setup ?
I've removed the patch from the queue for now, and if we cannot
resolve the regressions I will have to drop it from 11.0.
Thanks
Emil
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 7 November 2015 at 19:18, Kenneth Graunke wrote:
> On Saturday, November 07, 2015 04:37:05 PM Emil Velikov wrote:
>> Jason, Kenneth,
>>
>> I had to pick a few extra commits as prerequirements to the "nir:
>> Properly invalidate metadata in nir_foo" patc
Hi Mark,
On 7 November 2015 at 23:26, Mark Janes wrote:
> Hi Emil,
>
> I get this regression testing the new branch:
>
> piglit.spec.oes_compressed_paletted_texture.basic api
>
> /tmp/build_root/m64/lib/piglit/bin/oes_compressed_paletted_texture-api -auto
> -f
[Dropping Jason, Ken, adding Roland to CC list]
Hi Oded,
On 8 November 2015 at 09:22, Oded Gabbay wrote:
> Hi Emil,
>
> I would like to propose the following patch to be added to 11.0.5:
> 39b4dfe6ab1003863778a25c091c080e098833ec llvmpipe: use simple coeffs
> calc for 128bit
On 8 November 2015 at 11:41, Christian König wrote:
> On 06.11.2015 20:28, Ilia Mirkin wrote:
>>
>> On Fri, Nov 6, 2015 at 2:15 PM, Christian König
>> wrote:
>>>
>>> On 06.11.2015 20:10, Ilia Mirkin wrote:
>>>>
>>>> On Fri, Nov 6,
On 7 November 2015 at 21:57, Dave Airlie wrote:
> On 8 November 2015 at 02:47, Emil Velikov wrote:
>> Hi Dave,
>>
>> On 9 October 2015 at 01:38, Dave Airlie wrote:
>>> From: Dave Airlie
>>>
>>> So I've known this was broken before, cogl h
Hi Tim,
On 6 November 2015 at 21:09, Timothy Arceri wrote:
> On Fri, 2015-11-06 at 13:02 +0000, Emil Velikov wrote:
>> Hi Tim,
>>
>> A few comments below
>>
>> On 5 November 2015 at 11:17, Timothy Arceri wrote:
>> > From: Timothy Arceri
>> >
&
On 6 November 2015 at 21:13, Timothy Arceri wrote:
> On Fri, 2015-11-06 at 13:16 +0000, Emil Velikov wrote:
>> On 5 November 2015 at 11:17, Timothy Arceri wrote:
>> > From: Timothy Arceri
>> >
>> > This is in preperation for compile-time constant support.
>
even mentioned a few times a way
that we can share those and minimise these issues), but I believe
Chih-Wei was not really a fan of them. If he's ok with it I'll push
your original patch.
Regards,
Emil
P.S. Typos - for each one we fix, another we introduce another one or more :-P
___
***
> + *
> + * Copyright 2015 Advanced Micro Devices, Inc.
> + * All Rights Reserved.
> + *
Copying someone else's code does _not_ give you the right to remove
ignore their copyright. This piece here should include Julien and
myself.
Pl
401 - 500 of 8586 matches
Mail list logo
