[PATCH 01/12] glamor: Don't try to do rendering with unsupported formats.

dercheck -t blend -o src -f a8r8g8b8,x2r10g10b10 Signed-off-by: Eric Anholt --- glamor/glamor_render.c | 38 ++ 1 file changed, 38 insertions(+) diff --git a/glamor/glamor_render.c b/glamor/glamor_render.c index 488d1a7..3048cd8 100644 --- a/glamor/glamor_r

[PATCH 05/12] glamor: Move glamor_render.c pict handling to glamor_picture.c

These functions aren't used by anything else, and are specific to the temporary-upload-as-a-weird-format path of glamor_render.c, called through glamor_upload_picture_to_texture(). Signed-off-by: Eric Anholt --- glamor/glamor_picture.c

[PATCH 07/12] glamor: Drop tracking of the last picture attached to pixmaps.

Signed-off-by: Eric Anholt --- glamor/glamor.c | 11 --- glamor/glamor_picture.c | 48 glamor/glamor_priv.h| 15 --- glamor/glamor_render.c | 3 --- glamor/glamor_utils.h | 10 +- glamor/glamor_window.c

[PATCH 02/12] glamor: Ignore picture formats when choosing color for core rendering.

s would be wrong). Fixes all failures in "rendercheck -t fill" and, as a result, the remaining failures in "rendercheck -t blend -o src -f a8r8g8b8,x2r10g10b10" Signed-off-by: Eric Anholt --- glamor/glamor_utils.h | 11 +-- 1 file changed, 1 insertion(+), 10 del

[PATCH 10/12] glamor: Drop another dead function.

This hasn't been used since the format swap/revert stuff for pictures was added back in 2012. Signed-off-by: Eric Anholt --- glamor/glamor_utils.h | 7 --- 1 file changed, 7 deletions(-) diff --git a/glamor/glamor_utils.h b/glamor/glamor_utils.h index 59e6b64..e22bbc1 100644 --- a/g

[PATCH 04/12] glamor: Drop dead glamor_upload_bits_to_pixmap_texture() proto.

Signed-off-by: Eric Anholt --- glamor/glamor_priv.h | 9 - 1 file changed, 9 deletions(-) diff --git a/glamor/glamor_priv.h b/glamor/glamor_priv.h index 1614b6f..508433a 100644 --- a/glamor/glamor_priv.h +++ b/glamor/glamor_priv.h @@ -741,15 +741,6 @@ Bool

[PATCH 03/12] glamor: Use the new upload/download interface for XV uploads.

We don't need any of its weird handling of picture formats, since our XV pixmaps don't have any pictures attached. Signed-off-by: Eric Anholt --- glamor/glamor_pixmap.c | 2 +- glamor/glamor_priv.h | 4 glamor/glamor_xv.c | 38 -- 3 fil

[PATCH 12/12] glamor: Drop unused box translation/bounds code.

These are dead since the glamor_copy.c replacement. Signed-off-by: Eric Anholt --- glamor/glamor_utils.h | 39 --- 1 file changed, 39 deletions(-) diff --git a/glamor/glamor_utils.h b/glamor/glamor_utils.h index 8a96c61..e648af2 100644 --- a/glamor

[PATCH 11/12] glamor: Move cache_format to glamor_fbo.c, where it's used.

Signed-off-by: Eric Anholt --- glamor/glamor_fbo.c | 15 +++ glamor/glamor_utils.h | 15 --- 2 files changed, 15 insertions(+), 15 deletions(-) diff --git a/glamor/glamor_fbo.c b/glamor/glamor_fbo.c index 1eee304..262033f 100644 --- a/glamor/glamor_fbo.c +++ b/glamor

[PATCH 06/12] glamor: Use the actual picture's format when uploading memory pixmaps.

ORY pixmap as both the source and mask using different formats, where we smash the source's format to a new value so that the mask can use the same uploaded bits. Dropping most of that should be safe, since it will be uploaded as the source first, so the smashed format will still be used. Sig

glamor: rendercheck fixes and picture handling simplification

This series came out of the process of trying to figure out if I could safely store a8 pixmaps in an a8r8g8b8 FBO, to improve glamor performance on VC4 (which doesn't support 8-bit FBOs). It turns out the answer is "not quite", but the Render code had a lot of bugs obscuring that fact. __

Re: [PATCH 4/4] glamor: Drop a bunch of glamor_priv == NULL checks.

Keith Packard writes: > Eric Anholt writes: > >> Now that it's always non-null when the pixmap is non-null, we don't >> need so much of this. glamor_get_pixmap_private() itself still >> accepts a NULL pixmap and returns NULL, because of glamor

Re: [PATCH] glamor_egl: Properly free resources on init-error and exit

e fd > dup()-ed by eglInitialize() to stay open, call eglTerminate() from the new > glamor_egl_cleanup() to fix this. > > Signed-off-by: Hans de Goede Reviewed-by: Eric Anholt I'll stick it in my next pull request, unless keithp grabs it first. signatur

Re: [PATCH 4/4] glamor: Drop a bunch of glamor_priv == NULL checks.

Michel Dänzer writes: > On 08.07.2015 09:26, Eric Anholt wrote: >> >> diff --git a/glamor/glamor_utils.h b/glamor/glamor_utils.h >> index 0a7de82..a923b7a 100644 >> --- a/glamor/glamor_utils.h >> +++ b/glamor/glamor_utils.h >> @@ -756,8 +756,8 @@ glamor

glamor: Always allocate pixmap privates

The commit message in #3 has the meat of the series, but for further explanation, note that this reduces the glamor library by 8588 bytes (4.8%) on my x86-64 build, and I've only got 94 client-allocated pixmaps on my desktop at the moment according to xrestop. _

[PATCH 4/4] glamor: Drop a bunch of glamor_priv == NULL checks.

Now that it's always non-null when the pixmap is non-null, we don't need so much of this. glamor_get_pixmap_private() itself still accepts a NULL pixmap and returns NULL, because of glamor_render.c Signed-off-by: Eric Anholt --- glamor/glamor.c | 17 ++-

[PATCH 2/4] glamor: Drop dead glamor_is_large_picture().

It died as of keithp's new glyphs code. Signed-off-by: Eric Anholt --- glamor/glamor_utils.h | 12 1 file changed, 12 deletions(-) diff --git a/glamor/glamor_utils.h b/glamor/glamor_utils.h index cef62c5..0a7de82 100644 --- a/glamor/glamor_utils.h +++ b/glamor/glamor_ut

[PATCH 3/4] glamor: Ask the server to always allocate our private.

start out zero-filled, which matches the callocs we were doing when making our own privates, and in the case of an fb pixmap that has a priv where it didn't before, the type ends up being GLAMOR_MEMORY as we would want. Signed-off-by: Eric Anholt --- glamor/glamor.c

[PATCH 1/4] glamor: Reuse the glamor_is_memory helper.

Signed-off-by: Eric Anholt --- glamor/glamor_composite_glyphs.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/glamor/glamor_composite_glyphs.c b/glamor/glamor_composite_glyphs.c index cd88524..389c8f4 100644 --- a/glamor/glamor_composite_glyphs.c +++ b/glamor

[git pull v2] glamor: gles2 text improvements + regression fix + cleanups

Dave Airlie (1): xserver: fix build with glamor disabled. Eric Anholt (23): glamor: Fix text rendering on GLES2. glamor: Drop CloseScreen-time GL resource cleanup code. glamor: Use the normal GL_QUADS drawing helper in the render code. glamor: Use GL_EXT_map_buffer_ran

Re: [PATCH 1/7] glamor: Propagate the fact that pbo_valid is never set.

Keith Packard writes: > Eric Anholt writes: > >> The code to set it was deleted in keithp's big rewrite. > > Looks like you can now remove the 'pbo' parameter to > upload_sub_pixmap_to_texture as it's always 0 now? Sure, but I've got another big

Re: [PATCH 1/2] glamor: Fix bad rendering of glyphs after an a1 glyph upload.

Keith Packard writes: > Eric Anholt writes: > >> Fixes regressions since my "don't make an FBO for the glyph atlas" >> change. The a1 upload was a fallback, as I expected. However, >> fallback reads use glReadPixels() because there's no >> gl

[git pull] glamor: gles2 text improvements + regression fix + cleanups

up to b6a814630e48c76c8682432bf9d74b271840da9e: xserver: fix build with glamor disabled. (2015-07-02 10:07:02 -0700) Dave Airlie (1): xserver: fix build with glamor disabled. Eric Anholt (24): glamor: Fix text rendering on

Re: [git pull] glamor code cleanups

Eric Anholt writes: > As I was getting ready to send out another cleanups series, I found this > old one laying around, reviewed by Ken back in April. It merges cleanly > with the performance series I sent yesterday, and xtests successfully, > too. > > The following ch

[PATCH] glamor: Use GL_ARB_debug_output to log GL errors.

This should help people debugging when glamor does something stupid on their driver. Signed-off-by: Eric Anholt --- glamor/glamor.c | 41 + 1 file changed, 41 insertions(+) diff --git a/glamor/glamor.c b/glamor/glamor.c index 295f415..317f451 100644

[PATCH] xorg.modules: Drop external glamor.

It's been replaced by in-server glamor for a while. Signed-off-by: Eric Anholt --- build.sh | 1 - xorg.modules | 17 - 2 files changed, 18 deletions(-) diff --git a/build.sh b/build.sh index 6eac65d..a87263a 100755 --- a/build.sh +++ b/build.sh @@ -1174,7 +1

[PATCH 2/2] glamor: Add an assert to catch the previous bug.

Signed-off-by: Eric Anholt --- glamor/glamor_transfer.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/glamor/glamor_transfer.c b/glamor/glamor_transfer.c index aa5e861..155d7e0 100644 --- a/glamor/glamor_transfer.c +++ b/glamor/glamor_transfer.c @@ -186,6 +186,8 @@ glamor_download_boxes

[PATCH 1/2] glamor: Fix bad rendering of glyphs after an a1 glyph upload.

exsubimaging back and forth. Signed-off-by: Eric Anholt --- glamor/glamor_composite_glyphs.c | 67 ++-- 1 file changed, 37 insertions(+), 30 deletions(-) diff --git a/glamor/glamor_composite_glyphs.c b/glamor/glamor_composite_glyphs.c in

[PATCH 7/7] glamor: Use ARRAY_SIZE in a couple more places for consistency.

Signed-off-by: Eric Anholt --- glamor/glamor_program.c | 6 ++ glamor/glamor_render.c | 5 + 2 files changed, 3 insertions(+), 8 deletions(-) diff --git a/glamor/glamor_program.c b/glamor/glamor_program.c index 5619216..416c54a 100644 --- a/glamor/glamor_program.c +++ b/glamor

[PATCH 4/7] glamor: Drop dead glamor_es2_pixmap_read_prepare().

It's been unused since I killed glamor_download_pixmap_to_cpu(). Signed-off-by: Eric Anholt --- glamor/glamor_pixmap.c | 70 -- glamor/glamor_priv.h | 6 - 2 files changed, 76 deletions(-) diff --git a/glamor/glamor_pixmap.c b/g

[PATCH 2/7] glamor: Fix up some weird formatting in _glamor_create_fbo_array().

Signed-off-by: Eric Anholt --- glamor/glamor_fbo.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/glamor/glamor_fbo.c b/glamor/glamor_fbo.c index cab2ff9..acb5d0d 100644 --- a/glamor/glamor_fbo.c +++ b/glamor/glamor_fbo.c @@ -424,8 +424,8 @@ glamor_create_fbo_array

[PATCH 5/7] glamor: Make a bunch of single-file glamor functions static.

Signed-off-by: Eric Anholt --- glamor/glamor_largepixmap.c | 10 +++--- glamor/glamor_priv.h| 22 -- glamor/glamor_render.c | 4 ++-- 3 files changed, 9 insertions(+), 27 deletions(-) diff --git a/glamor/glamor_largepixmap.c b/glamor/glamor_largepixmap.c

[PATCH 6/7] glamor: Mark a bunch of single-file data static.

This gives the compiler a chance to optimize when the data is never changed -- for example, with pict_format_combine_tab, the compiler ends up inlining the 24 bytes of data into just 10 more bytes of code. Signed-off-by: Eric Anholt --- glamor/glamor_core.c| 2 +- glamor/glamor_program.c

[PATCH 1/7] glamor: Drop a redundant check.

Above, we've already checked for ->fbo && ->fbo->fb and returned. Signed-off-by: Eric Anholt --- glamor/glamor_pixmap.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/glamor/glamor_pixmap.c b/glamor/glamor_pixmap.c index f2bf223..0e515

[PATCH 3/7] glamor: Restore the hook to glamor_composite_rectangles().

It was apparently accidentally dropped in keithp's removal of _nf functions in 90d326fcc687e6d6d4b308f6272ededcf8145a17. Signed-off-by: Eric Anholt --- glamor/glamor.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/glamor/glamor.c b/glamor/glamor.c index 71ca4f4..29

[git pull] glamor code cleanups

pixmap extents calculations. (2015-06-30 15:32:13 -0700) Eric Anholt (7): glamor: Propagate the fact that pbo_valid is never set. glamor: Remove dead gl_tex flag in the fbo struct. glamor: Drop the dead "PBO&q

Re: glamor regression caused by 9c679d06 (Re: xserver: Branch 'master' - 5 commits)

Michel Dänzer writes: > On 30.06.2015 13:02, Keith Packard wrote: >> >> commit 9c679d06055cc62aa9209318705e87dc33fba4c8 >> Author: Eric Anholt >> Date: Sun May 31 16:07:01 2015 -0700 >> >> glamor: Skip actual FBO setup in our glyph atlas. >&

[git pull] GLES2/VC4 text performance series

git://people.freedesktop.org/~anholt/xserver glamor-next for you to fetch changes up to 0e3f1252dacdc3194a99a2d090b5c13f070f8799: glamor: Avoid using GL_QUADS on VC4. (2015-06-29 21:43:36 -0700) Eric Anholt (7): glamor: Fix text renderi

Re: [PATCH 07/16] modesetting: reverse prime support (v1.1)

This allows modesetting to support this scenario while still > supporting the USB offload one. > > v1.1: > fix comment + ret = bits (Eric) Looks nicer, thanks! This and patch 11-15 are: Reviewed-by: Eric Anholt Patch 16 is: Acked-by: Eric Anholt (I don't think I'll get aroun

Re: [PATCH 5/7] glamor: Provide a fallback path for using an index buffer to do quads.

Dave Airlie writes: > On 20 June 2015 at 12:09, Eric Anholt wrote: >> Improves x11perf -aa10text performance by 1377.59% +/- 23.8198% (n=93) >> on Intel with GLES2. >> > > Do we destroy glamor_priv->ib anywhere? > > Other than that, this seems sane, it m

[PULL] glamor VC4 crash fixes

es up to 790311cec30ac3d35e580b9f1266236f558033d4: glamor: Don't try to disable attrib divisors without the extension. (2015-06-19 19:11:05 -0700) ---- Eric Anholt (4): glamor: Actually allow glyphs of dimension 65 to 128 in

[PATCH] modesetting: Fix a compiler warning about CARD32 vs uint32_t.

Signed-off-by: Eric Anholt --- hw/xfree86/drivers/modesetting/dri2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/xfree86/drivers/modesetting/dri2.c b/hw/xfree86/drivers/modesetting/dri2.c index 63cb065..8dc12b7 100644 --- a/hw/xfree86/drivers/modesetting/dri2.c +++ b

[PATCH 4/7] glamor: Use GL_EXT_map_buffer_range if present.

We were only looking for the desktop GL version of the extension, so GLES2 missed out. Signed-off-by: Eric Anholt --- glamor/glamor.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/glamor/glamor.c b/glamor/glamor.c index 0378388..935fb74 100644 --- a/glamor/glamor.c +++ b

[PATCH 2/7] glamor: Drop CloseScreen-time GL resource cleanup code.

These will all be freed when the context is freed. Signed-off-by: Eric Anholt --- glamor/glamor.c | 5 - glamor/glamor_core.c | 11 --- glamor/glamor_gradient.c | 21 - glamor/glamor_priv.h | 3 --- glamor/glamor_render.c | 21

[PATCH 1/7] glamor: Fix text rendering on GLES2.

The GL_QUADS helper takes a number of quads, not a number of vertices. --- glamor/glamor_composite_glyphs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/glamor/glamor_composite_glyphs.c b/glamor/glamor_composite_glyphs.c index 1f0d75e..e92f1d9 100644 --- a/glamor/glamor_comp

[PATCH 6/7] glamor: Use proper Bools for some boolean values.

Signed-off-by: Eric Anholt --- glamor/glamor_priv.h | 18 +- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/glamor/glamor_priv.h b/glamor/glamor_priv.h index 17f8253..681895f 100644 --- a/glamor/glamor_priv.h +++ b/glamor/glamor_priv.h @@ -194,16 +194,16 @@ typedef

[PATCH 5/7] glamor: Provide a fallback path for using an index buffer to do quads.

Improves x11perf -aa10text performance by 1377.59% +/- 23.8198% (n=93) on Intel with GLES2. Signed-off-by: Eric Anholt --- glamor/glamor.c | 62 +++ glamor/glamor_priv.h | 11 + glamor/glamor_utils.h | 4 +--- 3 files changed, 74

[PATCH 7/7] glamor: Avoid using GL_QUADS on VC4.

Improves text rendering from about 284k glyphs per second to 320k glyphs per second. There's no GL extension for probing this, because of the philosophy of "Don't expose whether things are really in hardware or not." Signed-off-by: Eric Anholt --- glamor/glamor.c |

[PATCH 3/7] Use the normal GL_QUADS drawing helper in the render code.

We use this for all of our other performance-sensitive rendering, too. Signed-off-by: Eric Anholt --- glamor/glamor.c| 1 - glamor/glamor_priv.h | 7 +++ glamor/glamor_render.c | 50 -- 3 files changed, 7 insertions(+), 51

Re: [PATCH 5/6] cursor: add hw cursor support for prime

or(ScreenPtr pScreen, int x, int y) > +static void > +xf86ScreenMoveCursor(ScreenPtr pScreen, int x, int y) > { > xf86CursorScreenPtr ScreenPriv = > (xf86CursorScreenPtr) dixLookupPrivate(&pScreen->devPrivates, > @@ -202,6 +270,24 @@ xf86MoveCursor(ScreenPtr pScreen, int x, int y) >

Re: [PATCH 4/6] modesetting: reverse prime support

Dave Airlie writes: > This adds support for reverse prime to the modesetting driver. > > Reverse prime is where we have two GPUs in the display chain, > but the second GPU can't scanout from the shared pixmap, so needs > an extra copy to the on screen pixmap. > > This allows modesetting to suppor

Re: [PATCH 3/6] glamor: egl: add function to back a pixmap with a dma-buf.

CARD16 stride, CARD8 depth, CARD8 bpp) > { > #ifdef GLAMOR_HAS_GBM > -ScrnInfoPtr scrn = xf86ScreenToScrn(screen); > +ScrnInfoPtr scrn = xf86ScreenToScrn(pixmap->drawable.pScreen); I slightly prefer having a local "ScreenPtr screen = pixmap->drawable.pScreen"

Re: [PATCH 2/6] modesetting: add output master support

know the slave stuff really, but it seems reasonable. Acked-by: Eric Anholt signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel

Re: [PATCH 1/6] glamor: add support for allocating linear buffers (v1.1)

ed back into the FBO cache. Other than that, Reviewed-by: Eric Anholt signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel

[PATCH 2/4] glamor: Clean up some declarations in glyph rendering.

Signed-off-by: Eric Anholt --- glamor/glamor_composite_glyphs.c | 11 --- 1 file changed, 4 insertions(+), 7 deletions(-) diff --git a/glamor/glamor_composite_glyphs.c b/glamor/glamor_composite_glyphs.c index 39ed854..47bf647 100644 --- a/glamor/glamor_composite_glyphs.c +++ b/glamor

[PATCH 4/4] glamor: Don't try to disable attrib divisors without the extension.

Fixes epoxy assertion failures on vc4. Signed-off-by: Eric Anholt --- glamor/glamor_composite_glyphs.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/glamor/glamor_composite_glyphs.c b/glamor/glamor_composite_glyphs.c index c30cbed..1f0d75e 100644 --- a/glamor

[PATCH 1/4] glamor: Actually allow glyphs of dimension 65 to 128 in the cache.

The cache was trying to allow glyph_max_dim in, but since we were putting over 64x64 into HW memory, it would end up in the single-glyph-per-render bail_one path. Signed-off-by: Eric Anholt --- glamor/glamor.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/glamor

composite_glyphs on vc4

After this series, I'm at 143000 glyphs/sec, up from 2640 glyphs/sec with the old always-fallbacks path. ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel

[PATCH 3/4] glamor: Skip actual FBO setup in our glyph atlas.

re general software fallback for a1 (since the glyph is also in system memory). Signed-off-by: Eric Anholt --- glamor/glamor_composite_glyphs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/glamor/glamor_composite_glyphs.c b/glamor/glamor_composite_glyphs.c index 47bf647..c30c

Re: [PATCH] glamor: don't do render ops with matching source/dest (v3)

evel, and check it earlier. > (I assume the was for this case.) > v3: add same code to largepixmap paths (Keith) Reviewed-by: Eric Anholt signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives: http:

Re: [PATCH 1/7] render: Inline common FindGlyph case

Keith Packard writes: > Provide the common path for FindGlyph as an inline function, falling > back to a function for the slow case. From the end of your current branch, reverting this commit gives me this on aa10text: x glamor-new-glyphs + revert +--

Re: [PATCH 7/7] glamor: Replace CompositeGlyphs code

PictFormatPtr glyph_format, > +INT16 x_src, > +INT16 y_src, int nlist, GlyphListPtr list, > + GlyphPtr *glyphs) > +{ > +int this_time; > +GLshort *v = NULL; > +DrawablePtr drawable = ds

Re: [PATCH 6/7] glamor: Adapt glamor_program API to handle render acceleration

Summary: Tiny nitpicks, one functional concern. Keith Packard writes: > This extends the existing API to support options needed for render > accleratoin, including an additional fragment, 'combine', (which acceleration > provides a place to perform the source IN mask operation before the > fina

Re: [PATCH 4/7] glamor: Remove destination drawable argument from glamor_set_texture

Keith Packard writes: > This argument wasn't used at all. > > Signed-off-by: Keith Packard Reviewed-by: Eric Anholt signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org

Re: [PATCH 3/7] glamor: Pass depth to glamor_pm_is_solid and glamor_set_planemask

p. The gc keeps a copy of the depth of the drawable? Is it always the same? If so, Reviewed-by: Eric Anholt signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info:

Re: [PULL] glamor-next

Markus Wick writes: > Am 2015-03-26 03:33, schrieb Eric Anholt: >> The only reason _nf is useful is if you need to implement faster >> software fallbacks -- otherwise, you should wrap above glamor, since >> _nf >> would have been equivalent to calling down. > &g

Re: [PULL] glamor-next

Michel Dänzer writes: > On 26.03.2015 04:14, Eric Anholt wrote: >> Michel Dänzer writes: >>> On 25.03.2015 05:21, Eric Anholt wrote: >>>> >>>> Keith Packard (13): >>> [...] >>>> glamor: Eliminate GLAMOR_USE_SCREEN and GL

Re: [PULL] glamor-next

Michel Dänzer writes: > On 26.03.2015 04:14, Eric Anholt wrote: >> Michel Dänzer writes: >> >>> On 25.03.2015 05:21, Eric Anholt wrote: >>>> >>>> Keith Packard (13): >>> [...] >>>> glamor: Eliminate GLAMOR_USE_SCREEN

Re: [PULL] glamor-next

Michel Dänzer writes: > On 25.03.2015 05:21, Eric Anholt wrote: >> >> Keith Packard (13): > [...] >> glamor: Eliminate GLAMOR_USE_SCREEN and GLAMOR_USE_PICTURE_SCREEN >> glamor: Remove _nf rendering functions >> glamor: Remove ddx fallba

RE: [PATCH] glamor: add fallback if NULL EGLConfig is not supported

"Zhou, Jammy" writes: > Hi Eric, > > Thanks for your response. > > I think some binary blobs don't support EGL_MESA_configless > extension. We can argue that they should add this support, but it will > be good if we can also have some fallback there. You wrote the patch, so presumably you were w

glamor cleanups

Here are some documentation fixes and dead code removal from glamor, as well as a tiny patch from trying to figure out how some of the large pixmap code works. This can be found on the glamor-cleanups branch of my tree. ___ xorg-devel@lists.x.org: X.Org

[PATCH 1/7] glamor: Propagate the fact that pbo_valid is never set.

The code to set it was deleted in keithp's big rewrite. Signed-off-by: Eric Anholt --- glamor/glamor_pixmap.c | 18 ++ glamor/glamor_priv.h | 2 -- 2 files changed, 2 insertions(+), 18 deletions(-) diff --git a/glamor/glamor_pixmap.c b/glamor/glamor_pixmap.c index 89

[PATCH 3/7] glamor: Drop the dead "PBO" field in an FBO cache struct.

PBOs are only used at the whole pixmap level. Signed-off-by: Eric Anholt --- glamor/glamor_fbo.c | 2 -- glamor/glamor_priv.h | 2 -- 2 files changed, 4 deletions(-) diff --git a/glamor/glamor_fbo.c b/glamor/glamor_fbo.c index aedc804..5d59913 100644 --- a/glamor/glamor_fbo.c +++ b/glamor

[PATCH 4/7] glamor: Move doxygen for private structs into the structs.

This should hopefully keep the comments more up to date with the structure comments. While I'm here, I've reworded a few of them to be more accurate, and dropped a bunch of stale comments. Signed-off-by: Eric Anholt --- glamor/glamor_pr

[PATCH 7/7] glamor: Simplify some temp pixmap extents calculations.

Signed-off-by: Eric Anholt --- glamor/glamor_largepixmap.c | 16 +++- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/glamor/glamor_largepixmap.c b/glamor/glamor_largepixmap.c index 391f376..4c2534d 100644 --- a/glamor/glamor_largepixmap.c +++ b/glamor

[PATCH 6/7] glamor: Drop a dead flag to glamor_create_fbo_array().

Signed-off-by: Eric Anholt --- glamor/glamor_fbo.c | 35 +++ 1 file changed, 11 insertions(+), 24 deletions(-) diff --git a/glamor/glamor_fbo.c b/glamor/glamor_fbo.c index 5d59913..ffe8ef9 100644 --- a/glamor/glamor_fbo.c +++ b/glamor/glamor_fbo.c @@ -355,11

[PATCH 5/7] glamor: Drop dead GLAMOR_FBO_DOWNLOADED flag.

Signed-off-by: Eric Anholt --- glamor/glamor_priv.h | 5 - glamor/glamor_utils.h | 1 - 2 files changed, 6 deletions(-) diff --git a/glamor/glamor_priv.h b/glamor/glamor_priv.h index 0960055..b9bd1ba 100644 --- a/glamor/glamor_priv.h +++ b/glamor/glamor_priv.h @@ -334,11 +334,6 @@ enum

[PATCH 2/7] glamor: Remove dead gl_tex flag in the fbo struct.

This used to be used in the old copy_area path. --- glamor/glamor_fbo.c | 6 -- glamor/glamor_priv.h | 2 -- 2 files changed, 8 deletions(-) diff --git a/glamor/glamor_fbo.c b/glamor/glamor_fbo.c index ea0e801..aedc804 100644 --- a/glamor/glamor_fbo.c +++ b/glamor/glamor_fbo.c @@ -476,12 +47

[PULL] glamor-next

glamor: Fix up indentation Eric Anholt (3): glamor: Just set the logic op to what we want at the start of all rendering. glamor: Don't optimize out scissor updates in CopyArea. glamor: Add a helper function for the common GL_QUADS fallback pattern. Jon TURNEY (2):

Re: [PATCH] ephyr: Avoid a segfault with 'DISPLAY= Xephy -glamor'

Daniel Martin writes: > On 9 February 2015 at 15:09, Jon TURNEY wrote: >> On 09/02/2015 14:03, Daniel Martin wrote: >>> >>> On 9 February 2015 at 14:45, Jon TURNEY >>> wrote: ephyr_glamor_connect() returns NULL if we failed, but applying xcb_connection_has_error() to NULL is not

Re: [PATCH] glamor: Fix up indentation

Adam Jackson writes: > Signed-off-by: Adam Jackson Reviewed and applied to -next. signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailma

Re: [PATCH] glamor: add fallback if NULL EGLConfig is not supported

Jammy Zhou writes: > Signed-off-by: Kai Guo > Signed-off-by: Jammy Zhou What driver does this apply to? Is there a good reason for that driver not supporting EGL_MESA_configless context? It's a very nice extension, and trivial to implement. signature.asc Description: PGP signature

Re: [PATCH] glamor: Perform texture2D() separately from swizzle.

Matt Turner writes: > The texture2D() happens in each branch, so we may as well do it as early > as possible and hide some of its latency in the branching instructions. > Moving it outside the (uniform) control flow reduces the number of > instructions in the fs_source shader from 64 to 46 and in

Re: [PATCH] configure.ac: remove remaining TLS references

Emil Velikov writes: > No longer used with the removal of the GL dispatch (glapi) from libglx a > few releases ago. Reviewed-by: Eric Anholt signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives

Re: [PATCH] glamor: Fix build when configured --enable-glamor --disable-xshmfence

Jon TURNEY writes: > On 05/02/2015 19:10, Eric Anholt wrote: >> Jon TURNEY writes: >> >>> Signed-off-by: Jon TURNEY >>> --- >>> glamor/glamor_sync.c | 2 ++ >>> 1 file changed, 2 insertions(+) > [...] >> >> Reviewed and ap

Re: [PATCH] glamor: Fix build when configured --enable-glamor --disable-xshmfence

Jon TURNEY writes: > Signed-off-by: Jon TURNEY > --- > glamor/glamor_sync.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/glamor/glamor_sync.c b/glamor/glamor_sync.c > index 5e158c3..fbc47d4 100644 > --- a/glamor/glamor_sync.c > +++ b/glamor/glamor_sync.c > @@ -94,8 +94,10 @@ glamo

Re: [PATCH v3] glamor: check max native ALU instructions

Olivier Fourdan writes: > When using glamor (either in Xephyr or Xwayland) on hardware with too > low instructions limit, glamor fallbacks to sw due to large shaders. > > This makes glamor unbearably slow on such hardware. > > Check reported value for GL_MAX_PROGRAM_NATIVE_ALU_INSTRUCTIONS_ARB >

Re: [PATCH 1/3] glamor: Just set the logic op to what we want at the start of all rendering.

Keith Packard writes: > Eric Anholt writes: > >> By dropping the unconditional logic op disable at the end of >> rendering, this fixes GL errors being thrown in GLES2 contexts (which >> don't have logic ops). On desktop, this also means a little less >> ove

Re: [PATCH 2/3] glamor: Don't optimize out scissor updates in CopyArea.

Keith Packard writes: > Eric Anholt writes: > >> This possibly is a minor hit for immediate mode renderers (no >> difference on copypixin100 on my hsw, n=12), but it gives important >> information about drawing bounds to a deferred renderer (3.1x >> improvement in

[PATCH 2/3] glamor: Don't optimize out scissor updates in CopyArea.

This possibly is a minor hit for immediate mode renderers (no difference on copypixin100 on my hsw, n=12), but it gives important information about drawing bounds to a deferred renderer (3.1x improvement in copypixwin100 on vc4). Signed-off-by: Eric Anholt --- glamor/glamor_copy.c | 17

[PATCH 1/3] glamor: Just set the logic op to what we want at the start of all rendering.

eap as the caller caching it would be. v2: Don't forget to set the logic op in Xephyr's drawing. Signed-off-by: Eric Anholt --- This is 2/3 a resend of patches that got swamped by a discussion that wasn't related to the patches themselves. The only feedback there that I see missin

[PATCH 3/3] glamor: Add a helper function for the common GL_QUADS fallback pattern.

We should do better than this with an index buffer, but for now at least make it so that we don't have to copy the same code to new places. Signed-off-by: Eric Anholt --- glamor/glamor_copy.c | 8 +--- glamor/glamor_rects.c | 9 + glamor/glamor_spans.c | 9 + g

Re: [PATCH] glamor: Don't unbind the GL_ARRAY_BUFFER in glamor_put_vbo_space().

nder xf86-video-modesetting > (which doesn't do page flipping yet), glamor_get_vbo_space() accounted > for 4.24% of the CPU instructions in each glamor_copy_fbo_fbo_draw() > operation. With this patch, it became only 0.87%. > > Signed-off-by: Kenneth Graunke > Cc: Eric Anholt >

Re: [PATCH 1/3] glamor: Just set the logic op to what we want at the start of all rendering.

Kenneth Graunke writes: > On Tuesday 30 December 2014 14:54:27 Eric Anholt wrote: >> By dropping the unconditional logic op disable at the end of >> rendering, this fixes GL errors being thrown in GLES2 contexts (which >> don't have logic ops). On desktop, thi

Re: [PATCH] xwayland: Add dependency on glamor libs

Olivier Fourdan writes: > So that Xwayland gets re-linked each time glamor is modified. > > Signed-off-by: Olivier Fourdan Reviewed and merged for -next. signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives

Re: [PATCH v2] ephyr: Fail if glamor is requested but not usable

Olivier Fourdan writes: > Signed-off-by: Olivier Fourdan Reviewed and picked it up for my -next branch. signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: htt

Re: [PATCH v2] glamor: check max native ALU instructions

Olivier Fourdan writes: > When using glamor (either in Xephyr or Xwayland) on hardware with too > low instructions limit, glamor fallbacks to sw due to large shaders. > > This makes glamor unbearably slow on such hardware. > > Check reported value for GL_MAX_PROGRAM_NATIVE_ALU_INSTRUCTIONS_ARB >

Re: [PATCH] xwayland: Set glamor filter to nearest

us Wick Reviewed-by: Eric Anholt signature.asc Description: PGP signature ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel

Re: [PATCH 6/7] glamor: Re-enable a8 textures for GLES2

Maarten Lankhorst writes: > Hey, > > Op 13-01-15 om 03:48 schreef Eric Anholt: >> Maarten Lankhorst writes: >> >>> This will probably break PVR, but fixes other implementations. >> GL ES 2.0.25 spec says: "Formats not listed in table 4.5, including >

Re: [PATCH 7/7] XXX glamor: Only use supported tex formats

Maarten Lankhorst writes: > Some GLES2 implementations don't support GL_UNSIGNED__REV and only > guarantee less fancy things to work. > > I'm only certain it fixes the GLES2 errors, but is this a sane way to handle > things? Or am I just shutting up the errors? I always find the GL format ar

<    1   2   3   4   5   6   7   8   9   10   >