Commit Graph

864 Commits

Author SHA1 Message Date
Eric Anholt
c66d65a645 glamor: Use GL_MESA_tile_raster_order for overlapping blits.
Improves Raspberry Pi 3 x11perf -copywinwin500 from ~480/sec to
~700/sec.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
2017-10-13 17:23:46 -07:00
Olivier Fourdan
9869dcb349 glamor: Avoid overflow between box32 and box16 box
glamor_compute_transform_clipped_regions() uses a temporary box32
internally which is copied back to a box16 to init the regions16,
thus causing a potential overflow.

If an overflow occurs, the given region is invalid and the pixmap
init region will fail.

Simply check that the coordinates won't overflow when copying back to
the box16, avoiding a crash later down the line in glamor.

Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=101894
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Tested-by: Fabrice Bellet <fabrice@bellet.info>
Reviewed-by: Adam Jackson <ajax@redhat.com>
2017-09-13 11:02:52 -04:00
Olivier Fourdan
bd353e9b84 glamor: handle NULL source picture
COMPOSITE_REGION() can pass NULL as a source picture, make sure we
handle that nicely in both glamor_composite_clipped_region() and
glamor_composite_choose_shader().

Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=101894
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
2017-09-13 11:02:48 -04:00
Emil Velikov
e12e02b2e6 glamor: use the lowercase xnf.*alloc API
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Olivier Fourdan <ofourdan@redhat.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-09-06 08:57:06 +10:00
Eric Anholt
27500ee82e glamor: Scissor Render composite operations to the bounds of the drawing.
Unlike the previous two fixes, this one introduces new GL calls and
statechanges of the scissor.  However, given that our Render drawing
already does CPU side transformation and inefficient box upload, this
shouldn't be a limiting factor for Render acceleration.

Surprisingly, it improves x11perf -comppixwin10 -repeat 1 -reps 10000
on i965 by 3.21191% +/- 1.79977% (n=50).

v2: Make the jump to the exit land after scissor disable.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
2017-08-14 12:37:37 -07:00
Eric Anholt
e6ab3b1109 glamor: Scissor CopyArea to the bounds of the drawing.
Like the previous fix to rectangles, this reduces the area drawn on
tiled renderers by letting the CPU-side tile setup know what tiles
might be drawn at all.

Surprisingly, it improves x11perf -copypixwin1 -repeat 1 -reps 10000
on i965 by 2.93185% +/- 1.5561% (n=90).

v2: Drop extra glamor_bounds_union_box() from previous debugging
    (caught by Mark Marshall).

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com> (v1)
2017-08-14 12:36:35 -07:00
Eric Anholt
60cc7e367a glamor: Scissor rectangle drawing to the bounds of the rects.
Scissors provide a critical hint to tiled renderers as to what tiles
need to be load/stored because they could be modified by the
rendering.

The bounds calculation here is limited to when we have a small number
of rects (large enough to cover rounded window corners, but probably
not xeyes) to avoid overhead on desktop GL.

No performance difference on i965 with x11perf -rect1 -repeat 1 -reps
10000 (n=50)

v2: Clamp rectangle bounds addition.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
2017-08-14 12:35:55 -07:00
Emil Velikov
05a4396f08 glamor: update "required EGL extensions" comment
The extensions listed have not been needed in a while. Replace with the
only remaining requirement.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
2017-07-05 10:15:32 -04:00
Emil Velikov
06419a5b17 glamor: remove no longer needed KHR_gl_texture_2D_image requirement
The code that needed it was introduced with commit 7cfd9cc232 ("Add
DRI3 support to glamor") back in 2013. And was nuked a couple of years
ago with commit 51984dddfc ("glamor: Delay making pixmaps shareable
until we need to.")

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
2017-06-29 08:48:25 -07:00
Keith Packard
c86fc56b10 glamor: Clarify variable names in glamor_copy_cpu_fbo
This function creates a temporary pixmap to hold data being moved from
the source to the destination. However, it labeled all of the
variables associated with this as src_, which makes it confusing to
read the code. Rename them tmp_ instead. Also fix the comment
describing the function to note that it copies from CPU to GPU, not
GPU to GPU.

Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
2017-06-12 20:15:49 -07:00
Michel Dänzer
ffda82ed04 glamor: Fix temporary pixmap coordinate offsets
The previous values happened to work in basic cases, but not in general
if the destination is a subwindow or has a border.

Fixes crash with xli, which moves a large subwindow inside a smaller
parent window for scrolling.

No regressions with xterm, x11perf -copyplane or the xscreensaver
phosphor hack.

Bug: https://bugs.debian.org/857983
Reviewed-by: Keith Packard <keithp@keithp.com>
2017-06-13 11:13:22 +09:00
Michel Dänzer
7c88977d33 glamor: Store the actual EGL/GLX context pointer in lastGLContext
Fixes subtle breakage which could sometimes trigger after a server reset
with multiple screens using glamor:

Screen A enters glamor_close_screen last and calls various cleanup
functions, which at some point call glamor_make_current to make sure
screen A's GL context is current. This sets lastGLContext to screen A's
&glamor_priv->ctx. Finally, glamor_close_screen calls
glamor_release_screen_priv, which calls free(glamor_priv).

Later, screen B enters glamor_init, which allocates a new glamor_priv.
With bad luck, this can return the same pointer which was previously
used for screen A's glamor_priv. So when screen B's glamor_init calls
glamor_make_current, lastGLContext == &glamor_priv->ctx, so MakeCurrent
isn't called for screen B's GL context, and the following OpenGL API
calls triggered by glamor_init mess up screen A's GL context.

The observed end result of this was a crash in glamor_get_vbo_space
because glamor_priv->vbo didn't match the GL context, though there might
be other possible outcomes.

Assigning the actual GL context pointer to lastGLContext prevents this
by preventing the false negative test in glamor_make_current.

Reviewed-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
2017-06-05 18:32:15 +09:00
Eric Anholt
87db150d8a glamor: Remove the "delayed fallback" code.
The usage of this died with the old core rendering code.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
2017-06-02 17:46:48 -07:00
Eric Anholt
72ddad7a97 glamor: Drop glamor_set_screen_pixmap().
All that was left here was updating the FBO's size.  However, the FBO
size was always set correctly already through
glamor_set_pixmap_texture() from whoever had attached a new BO to the
pixmap.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
2017-06-02 17:46:46 -07:00
Eric Anholt
49b12cb736 glamor: Stop tracking the screen_fbo.
This means we no longer get "s" for on-screen drawing in glamor_debug,
and there's only "m" (CPU memory) or "f" (Any GPU memory, aka FBOs).
That seems fine to me.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
2017-06-02 17:46:45 -07:00
Eric Anholt
fd0d2523a3 glamor_egl: Stop saving the EGL major/minor version.
We don't use them for anything.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
2017-06-02 17:46:01 -07:00
Eric Anholt
d0675e96f1 glamor_egl: Drop glamor_egl_create_textured_screen_ext().
The function hasn't been doing anything useful since keithp's resource
freeing fixes in 2014.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:33:41 -04:00
Eric Anholt
c36f56808f glamor_egl: Automatically choose a GLES2 context if desktop GL fails.
GLES2 support has been requested multiple times, and we've had this
code laying around trying to implement it.  The GLES2 implementation
is not quite there yet (some pixel transfer failures), but it
shouldn't take much fixing at this point.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:33:38 -04:00
Eric Anholt
f6bf5f1c74 glamor_egl: Remove check for KHR_surfaceless_context_*
Those extensions don't exist.  There's only surfaceless_context.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
2017-05-18 11:33:35 -04:00
Eric Anholt
6e793ae085 glamor_egl: Drop warning about indirect GLX and GLES2.
Indirect GLX uses its own context and doesn't care what glamor is
using.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:33:32 -04:00
Eric Anholt
59de50d6ef glamor_egl: Avoid flink names in glamor_egl_create_textured_pixmap().
Using flink is banned on render nodes, and they needlessly expose our
screen pixmap contents to any authenticated client.  This also
incidentally drops the dependency on EGL_MESA_drm_image.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:33:26 -04:00
Eric Anholt
5d6f17d913 glamor_egl: Drop dead "cpp" field
It's been unused since the initial import.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:33:24 -04:00
Eric Anholt
37618e810b glamor_egl: Drop dead gl_context_depth.
This was replaced in 4afe15d8bf, but not
deleted.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:33:21 -04:00
Eric Anholt
a6a751e5d8 glamor_egl: Drop unnecessary check for KHR_gl_renderbuffer_image.
I couldn't find it being used anywhere in the history of the code.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:33:18 -04:00
Eric Anholt
40a153ee45 glamor_egl: Always require the gbm-based image import path.
This has been associated with dri3 for now, but we need to use it
elsewhere in order to avoid flink.  The extensions have been
implemented for long enough that I couldn't find when it was that we
turned them on.  Oddly, we already required renderbuffer import
support, which is basically as hard to implement as texture import.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
Acked-by: Daniel Stone <daniels@collabora.com>
2017-05-18 11:33:09 -04:00
Eric Anholt
b1b71b7a8b glamor_egl: Drop the has_gem flag.
We're using GBM, so we know we've got GEM.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:33:07 -04:00
Eric Anholt
5102ea38eb glamor_egl: Unifdef GLAMOR_HAS_GBM.
We only build this code with GBM, and supporting non-GBM well would be
invasive.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:33:03 -04:00
Eric Anholt
5064153c79 glamor_egl: Print a useful identifying string on initialization.
The EGL version is not used anywhere in the glamor code, so it's not
interesting.  And when saying that we've started using GL
acceleration, it's nice to know what GL we're actually using.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-05-18 11:32:59 -04:00
Eric Anholt
1549e30372 Add a Meson build system alongside autotools.
This is a work in progress that builds Xvfb, Xephyr, Xwayland, Xnest,
and Xdmx so far.  The outline of Xquartz/Xwin support is in tree, but
hasn't been built yet.  The unit tests are also not done.

The intent is to build this as a complete replacement for the
autotools system, then eventually replace autotools.  meson is faster
to generate the build, faster to run the bulid, shorter to write the
build files in, and less error-prone than autotools.

v2: Fix indentation nits, move version declaration to project(), use
    existing meson_options for version-config.h's vendor name/web.

Signed-off-by: Eric Anholt <eric@anholt.net>
Acked-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
2017-04-26 15:25:27 -07:00
Eric Anholt
c7be7a688a Use #ifdef instead of #if for features to make Meson easier.
We mostly use #ifdef throughout the tree, and this lets the generated
config.h files just be #define TOKEN instead of #define TOKEN 1.

Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-04-26 17:30:12 -04:00
Olivier Fourdan
7bfb87a213 glamor: an FBO is not needed for Xv pixmaps
It appears that on some hardware/diver combo such as nv30/nouveau, using
GL_ALPHA as format for 8-bit depth will cause an incomplete attachment
error (GL_FRAMEBUFFER_INCOMPLETE_ATTACHMENT) when trying to bind the
texture.

As a result, the FBO is NULL and glamor segfaults when trying to access
the FBO width/height in pixmap_priv_get_scale() in glamor_xv_render().

This happens with glamor-xv which uses 8-bit pixmaps, meaning that on
such hardware/driver, trying to play a video using Xv will lead to a
crash of the Xserver. This affects Xwayland, Xephyr, modesetting driver
with glamor accel.

But the use of an FBO is not actually needed for glamox-xv, so by
disabling FBO at pixmap creation, we can avoid the issue entirely.

Fix suggested by Eric Anholt <eric@anholt.net>

Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=100710
Fixes: https://bugzilla.redhat.com/1412814
Reviewed-by: Eric Anholt <eric@anholt.net>
2017-04-21 11:53:13 -07:00
Eric Anholt
266af991c8 glamor: Fix some formatting that confused the unifdef command.
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
2017-03-23 13:17:42 -04:00
Adam Jackson
1ad2306823 glamor: Avoid software fallback for planemasked ZPixmap GetImage
Same trick as in fb: just do a normal GetImage and deal with the
planemask on the CPU if you have to. Since the software fallback hit for
glamor is pretty brutal, this is a much more impressive win for glamor
than it was for fb:

  11100.0  87700.0 (7.901) (copy 0xaaaaaaaa) ShmGetImage 10x10 square
   9840.0  47800.0 (4.858) (copy 0xaaaaaaaa) ShmGetImage 100x100 square
   1550.0   4240.0 (2.735) (copy 0xaaaaaaaa) ShmGetImage 500x500 square
   9450.0  78900.0 (8.349) (0xaaaaaaaa) GetImage 10x10 square
   6910.0  30900.0 (4.472) (0xaaaaaaaa) GetImage 100x100 square
    431.0   2020.0 (4.687) (0xaaaaaaaa) GetImage 500x500 square

Measured with Xephyr -glamor on Skylake GT3e.

Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
2017-03-20 13:48:40 -04:00
Olivier Fourdan
8805a48ed3 glamor: avoid a crash if texture allocation failed
Texture creation in _glamor_create_tex() can fail if a GL_OUT_OF_MEMORY
is raised, in which case the texture returned is zero.

But the texture value is not checked in glamor_create_fbo() and glamor
will abort in glamor_pixmap_ensure_fb() because the fbo->tex is 0:

  Truncated backtrace:
  Thread no. 1 (10 frames)
   #4 glamor_pixmap_ensure_fb at glamor_fbo.c:57
   #5 glamor_create_fbo_from_tex at glamor_fbo.c:112
   #6 glamor_create_fbo at glamor_fbo.c:159
   #7 glamor_create_fbo_array at glamor_fbo.c:210
   #8 glamor_create_pixmap at glamor.c:226
   #9 compNewPixmap at compalloc.c:536
   #10 compAllocPixmap at compalloc.c:605
   #11 compCheckRedirect at compwindow.c:167
   #12 compRealizeWindow at compwindow.c:267
   #13 RealizeTree at window.c:2617

Check the value returned by _glamor_create_tex() in glamor_create_fbo()
and return NULL in the texture is zero.

All callers of glamor_create_fbo() actually check the returned value and
will use a fallback code path if it's NULL.

Please cherry-pick this to active stable branches.

Bugzilla: https://bugzilla.redhat.com/1433305
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
2017-03-17 13:17:31 -07:00
Adam Jackson
0803918e64 fb: Remove 24bpp support (v3)
v2:
- Require power-of-two bpp in ScreenInit
- Eliminate fbCreatePixmapBpp

v3
- Squash in the exa and glamor changes so we can remove pRotatedPixmap
  in the same stroke.

Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
2017-03-17 15:14:42 -04:00
Eric Anholt
fe0b297420 glamor: Fix dashed line rendering.
We were binding the screen pixmap as the dash and sampling its alpha,
which is usually just 1.0 (no dashing at all).

Please cherry-pick this to active stable branches.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
2017-03-16 11:30:58 -07:00
Olivier Fourdan
455051a0f1 glamor: Check glamor_set_destination_drawable() return value
Check the value returned by glamor_set_destination_drawable() and use
the fallback code path where possible.

Bugzilla: https://bugzilla.redhat.com/1417575
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
2017-03-15 15:41:13 -04:00
Olivier Fourdan
04b4bad7c0 glamor: glamor_set_destination_drawable() can fail
The fbo_array of a given glamor pixmap can be NULL in some cases, as
glamor_create_fbo_array() can fail to allocate the FBO array.

If this is the case, glamor_pixmap_fbo_at() will return NULL even though
the box index is valid, and glamor_set_destination_drawable() simply
assumes glamor_pixmap_fbo_at() will return an FBO prior to pass the
value to glamor_set_destination_pixmap_fbo(), which will segfault.

We need a way for glamor_set_destination_drawable() to fail safely and
let the caller know about the failure.

Add a boolean return value to glamor_set_destination_drawable() for that
purpose.

Bugzilla: https://bugzilla.redhat.com/1417575
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
2017-03-15 15:41:12 -04:00
Olivier Fourdan
f40ff18c96 glamor: Check for NULL pixmap in glamor_get_pixmap_texture()
glamor_create_pixmap() would return a NullPixmap if the given size is
larger than the maximum size of a pixmap.

But glamor_get_pixmap_texture() won't check if the given pixmap is
non-null, leading to a segfault if glamor_create_pixmap() failed.

This can be reproduced by passing Xephyr a very large screen width,
e.g.:

 $ Xephyr -glamor -screen 32768x1024 :10

 (EE)
 (EE) Backtrace:
 (EE) 0: Xephyr (OsSigHandler+0x29)
 (EE) 1: /lib64/libpthread.so.0 (__restore_rt+0x0)
 (EE) 2: Xephyr (glamor_get_pixmap_texture+0x30)
 (EE) 3: Xephyr (ephyr_glamor_create_screen_resources+0xc6)
 (EE) 4: Xephyr (ephyrCreateResources+0x98)
 (EE) 5: Xephyr (dix_main+0x275)
 (EE) 6: /lib64/libc.so.6 (__libc_start_main+0xf1)
 (EE) 7: Xephyr (_start+0x2a)
 (EE) 8: ? (?+0x2a) [0x2a]
 (EE)
 (EE) Segmentation fault at address 0x0
 (EE)
 Fatal server error:
 (EE) Caught signal 11 (Segmentation fault). Server aborting
 (EE)
 Aborted (core dumped)

Bugzilla: https://bugzilla.redhat.com/1431633
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
2017-03-15 15:36:44 -04:00
Michel Dänzer
0c1574d988 glamor: Fix typo: "vec2_pos" -> "vec2 pos"
Fixes crash when trying to use dashed lines:

 Failed to compile VS: 0:8(2): error: `vec2_pos' undeclared

Trivial.

Fixes: 	d8161aeb50 ("glamor: Fix missing declaration in dash vertex
                       shader")
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99943
2017-03-09 18:13:31 +09:00
Qiang Yu
9232835bd1 glamor: use drmGetDeviceNameFromFD2 when available
This is for glamor can support fd from DRM render node which is useful
for a render only DDX.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Qiang Yu <Qiang.Yu@amd.com>
2017-03-07 16:00:27 -05:00
Dr.-Ing. Dieter Jurzitza
d8161aeb50 glamor: Fix missing declaration in dash vertex shader
Fixes a GLSL compilation error:

Failed to compile VS: 0:13(43): error: `pos' undeclared
0:13(14): error: operands to arithmetic operators must be numeric
0:13(13): error: operands to arithmetic operators must be numeric

Tested-by: Stefan Dirsch <sndirsch@suse.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
2017-02-23 12:57:26 -05:00
Olivier Fourdan
8646398136 glamor: Two pass won't work on memory pixmaps
When selecting "CA_TWO_PASS" in glamor_composite_clipped_region() when
the hardware does not support "GL_ARB_blend_func_extended", we call
glamor_composite_choose_shader() twice in a row, which in turn calls
glamor_pixmap_ensure_fbo().

On memory pixmaps, the first call will set the FBO and the second one
will fail an assertion in glamor_upload_picture_to_texture() because
the FBO is already set.

Bail out earlier when the mask pixmap is in memory and the hardware
capabilities would require to use two pass, so that the assertion is not
failed and the rendering is correct.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99346
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
2017-01-24 11:27:33 -08:00
Adam Jackson
29a4f3db60 glamor: Squash unused variable warning
Accidentally introduced in 05e19644.

Signed-off-by: Adam Jackson <ajax@redhat.com>
2017-01-02 15:38:45 -05:00
Hans De Goede
05e1964425 glamor: Trust eglGetPlatformDisplayEXT if it exists
If the libEGL we are using has eglGetPlatformDisplayEXT, yet it still
returns NULL, then this very likely means that it does not support the
type (e.g. EGL_PLATFORM_GBM_MESA) passed in, and then returning NULL is
the right thing to do.

This avoids falling back to an eglGetDisplay() implementation which does
not understands the passed in gbm handle, treats it as a pointer to
something else completely, followed by a crash sooner or later.

Specifically this fixes using the nvidia binary driver, with nvidia's
libEGL + the modesetting driver on a secondary GPU crashing inside
glamor_egl_init() sometimes.

Cc: Eric Anholt <eric@anholt.net>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2017-01-02 13:07:25 -05:00
Olivier Fourdan
f43207c1c4 glamor: restore vfunc handlers on init failure
In glamor_init(), if the minimum requirements are not met, glamor may
fail after setting up its own CloseScreen() and DestroyPixmap()
routines, leading to a crash when either of the two routines is called
if glamor failed to complete its initialization, e.g:

  (EE) Backtrace:
  (EE) 0:  Xwayland (OsSigHandler+0x29)
  (EE) 1: /lib64/libpthread.so.0 (__restore_rt+0x0)
  (EE) 2: Xwayland (glamor_sync_close+0x2a)
  (EE) 3: Xwayland (glamor_close_screen+0x52)
  (EE) 4: Xwayland (CursorCloseScreen+0x88)
  (EE) 5: Xwayland (AnimCurCloseScreen+0xa4)
  (EE) 6: Xwayland (present_close_screen+0x42)
  (EE) 7: Xwayland (dix_main+0x4f9)
  (EE) 8: /lib64/libc.so.6 (__libc_start_main+0xf1)
  (EE) 9:  Xwayland (_start+0x2a)

Restore the previous CloseScreen() and DestroyPixmap() vfunc handlers in
case of failure when checking for the minimum requirements, so that if
any of the requirement is not met we don't leave the CloseScreen() and
DestroyPixmap() from glamor handlers in place.

Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1390018

Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
2016-11-30 09:45:52 +01:00
Emil Velikov
7fc96fb02d glamor: don't look for non-existing EGL_KHR_platform_base
The extension does not exist in the registry, thus needs to know they're
using EGL 1.5 in order to determine the eglGetPlatformDisplay function
pointer is valid.

Thus brings us into some lovely circular dependency.

Since mesa won't be able (in the foreseeable future) to export the KHR
flavour of extension (another way one could assume that EGL 1.5 is
available) just drop all the heuristics and use the
EGL_EXT_platform_base extension.

In practise (checked with the Mali driver) any EGL 1.5 driver will
advertise support for EGL_EXT_platform_base.

Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
2016-10-25 13:04:48 -04:00
Adam Jackson
f4a4115547 glamor: Use eglGetPlatformDisplay{,EXT} if we can
eglGetDisplay forces the implementation to guess which kind of display
it's been handed. glvnd does something different from Mesa, and in
general it's impossible for the library to get this right. Add a new
inline that gets the logic right, and works around a quirk in epoxy.

Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
2016-10-05 16:03:13 -04:00
Olivier Fourdan
1c2fcb9548 glamor: Fix pixmap offset for bitplane in glamor_copy_fbo_cpu
Commit cba28d5 - "glamor: Handle bitplane in glamor_copy_fbo_cpu"
introduced a regression as the computed pixmap offset would not match
the actual coordinates and write data elsewhere in memory causing a
segfault in fbBltOne().

Translate the pixmap coordinates so that the data is read and written at
the correct location.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97974
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Reviewed-and-Tested-by: Michel Dänzer <michel.daenzer@amd.com>
2016-10-05 13:57:13 -04:00
Mark Yao
c9b8ce7392 glamor: spans: fixup wrong count on glDrawArrays
In commit 9e9fcf5 (glamor: Add a helper function for the common
GL_QUADS fallback pattern.), the glDrawArrays count change was
accidentally changed to nbox.

Fixes xlogo with MESA_GL_VERSION_OVERRIDE=2.1 and
MESA_GLSL_VERSION_OVERRIDE=120

Signed-off-by: Mark Yao <mark.yao@rock-chips.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
2016-09-30 08:46:03 -07:00