If the linux vm86 backend changes look somewhat horrifying to you,
that's because you have taste.
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
The only driver even pretending to check the result is mach64 anyway.
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
This API sucks. Fortunately it's mostly unused at this point. geode,
sis, and xgi need minor patches to use the corresponding pciaccess code,
neomagic will (more explicitly) lose its non-PCI support, and newport
will need to be ported to /dev/mem or the platform bus or something.
This should also make it pretty clear that alpha's sparse memory support
was basically not a thing anymore, very few tears shed.
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Instead of sending every little rect. Lets x11perf run to completion,
makes 'while true; do gtkperf -a; done' take longer to crash.
This is effectively a resend of the same logic against the old
xfree86+xwayland branch:
http://lists.x.org/archives/xorg-devel/2013-October/038453.html
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
This implements simple throttling that keeps us to one attach per
frame. There isn't really an active performance benefit, since the
buffers will be redrawn only once per frame anyway, but it does cut down
on the chatty network traffic. Since the Wayland sockets might fill
up as well, the cut down on the volume of data we send out also provides
us with a big stability benefit.
Namely, mutter is a lot more stable running gtkperf, a fairly intensive
X11 application, after this change.
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
If something quickly maps and unmaps a window, then we'll immediately
create and destroy the Wayland surface that cooresponds to that
window. If our mouse pointer is over the window when the surface is
created, we'll receive a enter on the window.
Since resource creation and destruction is not synchronous, that
means that the compositor will queue up an event for a resource that's
eventually destroyed. On the client-side, when we receive this message,
we note that the resource isn't allocated, and get a NULL surface in our
enter handler. We immediately try to dereference this, and then crash.
This was caused by running gtkperf while moving the window a lot.
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Jasper St. Pierre <jstpierre@mecheye.net>
The piglit test glx_make_current triggers this assertion, by making the context
current on a different drawable before issuing a glXSwapBuffers()
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
It happens whenever a GLX client uses GL on a window before it's been mapped, so
don't log it like an error.
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
fbConfigToPixelFormatIndex()'s drawableTypeOverride parameter is a drawable
type bitmask, not a drawable type enum value
WGL_SWAP_COPY_ARB is a value of the WGL_SWAP_METHOD_ARB attribute for
wglChoosePixelFormatARB(), not an attribute itself
also remove duplicate error reporting for wglChoosePixelFormat() and fix a
comment
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Update to align with rewrite of wrapper generation script in commit
583a1146233f16d861706926706e5feec3baffba
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Don't override the server supported GL extensions string.
The string reported to the client is the intersection of client, server and GL
implementation extensions.
Overriding the server supported GL extensions string like this causes extensions
which are supported by the client and implementation, but not by the server, to
be erroneously reported, so don't change it.
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Add an idempotency guard to glwindows.h
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Seen during shutdown when using '-fullscreen' and '-depth 8'
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
Choose the display name used to connect to internal clients and exported into
environment of processes started from the traymenu so that it uses a transport
we know is working
This should mean the server can start correctly with -multiwindow and/or
-clipboard and any two of -nolisten inet6, -nolisten inet and -nolisten unix
(the server will correctly refuse to start if all 3 are used, as it must be
listening on at least one socket)
v2:
Place prototype for winGetDisplayName() in windisplay.h, and include it where
needed.
v3:
Include xwin-config.h, so that _XSERVER64 is defined, just in case anything
relies on that.
v4:
Replace grovelling around in the server's list of listeners with new Xtrans
TransIsListening() interface, added in Xtrans 1.3.3
See also [1]
[1] https://sourceware.org/bugzilla/show_bug.cgi?id=10725
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Yaakov Selkowitz <yselkowitz@users.sourceforge.net>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
- Rename XWIN_GLX_LINK_FLAGS -> XWIN_GLX_SYS_LIBS for consistency
- Rename MULTIWINDOW_LIBS -> MULTIWINDOW_SYS_LIBS for consistency
- Don't link with XWin with $(MAIN_LIB), it provides it's own main()
- Put one library per line for more intelligible diffs when one is added
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
This fixes a problem when using Khronos registry data since the change of
2013-08-16 removed glBlend(Color|Equation) from GL1.2 and added them to
GL_ARB_imaging.
If shim generation considers all features, no shims are generated for
glBlend(Color|Equation) as they are first emitted for GL 1.4 (which we ignore as
shims are only generated for GL version <=1.2), then emission for GL_ARB_imaging
is skipped as they have already been emitted.
Also improve feature name matching so it is exact, not on an initial substring,
so 'GL_ARB_texture_compression_bptc' and 'GL_ARB_texture_compression_rgtc'
aren't matched by 'GL_ARB_texture_compression'.
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Reviewed-by: Colin Harrison <colin.harrison@virgin.net>
On platforms that don't support PCI or have no GPU attached to the PCI
bus, there can still be a primary device on a non-PCI bus.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Tested-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
When neither of the various bus implementations was able to find a
primary bus and device, fallback to using the platform bus as primary
bus and the first platform device as primary device.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Tested-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
GCC 4.2 doesn't accept 2 typedef declarations of the same type, so
remove the extra one from xf86Xinput.h and have xf86Xinput.h #include
xf86.h to make sure everyone using just that file gets the typedef.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
Nobody was using it.
v2: Merge the hunk that was accidentally in the previous commit into
this one.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
v2: Fix accidentally squashed-in change for dropping client from the
arguments, which should have been in the next commit.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com> (v2)
Privates are initially cleared to zero by dixInitPrivates().
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
It's been unused since mach64 was deleted, and now there's a helper in
core XV.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Color key overlay implementations want to reuse this code, and XF86's
had bugs (to be fixed in the next commit).
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
XV was going against convention by having the core infrastructure
allocate the private on behalf of the DDX. I was interested in this
because I was trying to make multiple pieces of DDX be able to
allocate adaptors, and that wasn't going to work if DDX-specific code
was hung off of a single global screen private.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
The core was passing pointers to pxvs's nAdaptors and pAdaptors, and
the two hardware implementations were copying pxvs's nAdaptors and
pAdaptors into those pointers.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Since any DDX XV screen cleanup would need this same code for freeing
the tree of pointers for xv adaptors, move it to the dix.
v2: Unconditionalize the pPorts freeing, to match the block above it.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com> (v1)
As far as I can see, nothing has ever used this flag except possibly
the i.mx6 xorg ddx debug during bringup.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
As far as I can see (looking at trees on my disk, plus googling for
the term), nothing has ever used this flag
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Don't try to destroy rotation_damage in the xf86RotateCloseScreen; it
will have been destroyed when the screen pixmap was destroyed.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
When systemd isn't being used, systemd_logind_release_fd is defined
as an empty macro, leaving the arguments unused. Fix the compiler
warnings by simply removing the local variables and referencing the
structure within the macro call.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
A few files in the server are including xorg-server.h, which is only
for use by Xorg server drivers. This fixes those errors and then adds
a check to make sure it doesn't happen again.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
i810, mga, savage, and tdfx do reference these slots, but only to set
them to NULL, so while this does have API impact it's not actually used.
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Cargo-culted from DRI1, not actually used for anything.
Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
I ported these to pciaccess in:
commit 858fbbb40d
Author: Adam Jackson <ajax@redhat.com>
Date: Fri Sep 16 13:33:04 2011 -0400
pci: Port xf86MapLegacyIO to pciaccess
As of yet there are still no drivers using them, and there's not a lot
of value in having the wrappers when they just trivially call pciaccess
anyway. Nuke 'em.
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
