RRSetChanged (immediately above) was immune to screens with no master,
but RRTellChanged was not:
Thread 1 "X" received signal SIGSEGV, Segmentation fault.
RRTellChanged (pScreen=<optimized out>) at ../../randr/randr.c:576
576 mastersp = rrGetScrPriv(master);
(gdb) bt
#0 RRTellChanged (pScreen=<optimized out>) at ../../randr/randr.c:576
#1 0x000055555566f1e9 in RRNoticePropertyChange (value=0x555555bfbf28, property=70, output=0x555555bfef10) at ../../randr/rrproperty.c:153
#2 RRChangeOutputProperty (output=output@entry=0x555555bfef10, property=<optimized out>, type=type@entry=19, format=format@entry=32, mode=<optimized out>, mode@entry=0, len=len@entry=1, value=0x7fffffffe77c, sendevent=1, pending=0)
at ../../randr/rrproperty.c:263
#3 0x000055555566dba5 in RROutputSetNonDesktop (output=output@entry=0x555555bfef10, nonDesktop=nonDesktop@entry=0) at ../../randr/rroutput.c:333
...
Reported-by: Michel Dänzer <michel@daenzer.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
This adds support for RandR CRTC/Output leases through the modesetting
driver, creating a lease using new kernel infrastructure and returning
that to a client through an fd which will have access to only those
resources.
v2: Restore CRTC mode when leases terminate
When a lease terminates for a crtc we have saved data for, go
ahead and restore the saved mode.
v3: Report RR_Rotate_0 rotations for leased crtcs.
Ignore leased CRTCs when selecting screen size.
Stop leasing encoders, the kernel doesn't do that anymore.
Turn off crtc->enabled while leased so that modesetting
ignores them.
Check lease status before calling any driver mode functions
When starting a lease, mark leased CRTCs as disabled and hide
their cursors. Also, check to see if there are other
non-leased CRTCs which are driving leased Outputs and mark
them as disabled as well. Sometimes an application will lease
an idle crtc instead of the one already associated with the
leased output.
When terminating a lease, reset any CRTCs which are driving
outputs that are no longer leased so that they start working
again.
This required splitting the DIX level lease termination code
into two pieces, one to remove the lease from the system
(RRLeaseTerminated) and a new function that frees the lease
data structure (RRLeaseFree).
v4: Report RR_Rotate_0 rotation for leased crtcs.
v5: Terminate all leases on server reset.
Leases hang around after the associated client exits so that
the client doesn't need to occupy an X server client slot and
consume a file descriptor once it has gotten the output
resources necessary.
Any leases still hanging around when the X server resets or
shuts down need to be cleaned up by calling the kernel to
terminate the lease and freeing any DIX structures.
Note that we cannot simply use the existing
drmmode_terminate_lease function on each lease as that wants
to also reset the video mode, and during server shut down that
modesetting: Validate leases on VT enter
The kernel doesn't allow any master ioctls to run when another
VT is active, including simple things like listing the active
leases. To deal with that, we check the list of leases
whenever the X server VT is activated.
xfree86: hide disabled cursors when resetting after lease termination
The lessee may well have played with cursors and left one
active on our screen. Just tell the kernel to turn it off.
v6: Add meson build infrastructure
[Also bumped libdrm requirement - ajax]
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
This lets a DRM client map between X outputs and kernel connectors.
v2:
Change CONNECTOR_ID to enum -- Adam Jackson <ajax@nwnk.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@nwnk.net>
Tracks changes to the non-desktop property so that when non-zero,
outputs will always appear to be disconnected.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@nwnk.net>
Save any value of the kernel non-desktop property in the xf86Output
structure to avoid non-desktop outputs in the default configuration.
[Also bump randrproto requirement to a version that defines
RR_PROPERTY_NON_DESKTOP - ajax]
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@nwnk.net>
At startup, we want to ignore non-desktop monitors unless we don't
find any desktop monitors. Because there are no DIX RandR resources
allocated, let the driver store this information in a new field in the
xf86Output structure and then use that value to help decide whether to
include an output as part of the default configuration.
v2:
Suggested-by: Michel Dänzer <michel@daenzer.net>
Bump XF86_CRTC_VERSION from 7 to 8. This will let out-of-tree
drivers know whether this field is available.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@nwnk.net>
glamor now supports depth 30, so allow use of it.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Antoine Martin <antoine@nagafix.co.uk>
This retains old behavior for depths <= 24, but allows gamma
table and colormap updates to work properly at depth 30.
This needs the xf86Randr12CrtcComputeGamma() fix for depth 30
from a previous commit to work. Otherwise the server will work,
but gamma table updates will silently fail, iow. the server
would always run with a default identity gamma lut.
v2: Simplify as proposed by Michel.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Antoine Martin <antoine@nagafix.co.uk> (v1)
At screen depths > 24 bit, the color palettes passed into
xf86Randr12CrtcComputeGamma() can have a larger number of slots
than the crtc's hardware lut. E.g., at depth 30, 1024 palette
slots vs. 256 hw lut slots. This palette size > crtc gamma size
case is not handled yet and leads to silent failure, so gamma
table updates do not happen.
Add a new subsampling path for this case.
This makes lut updates work again, as tested with the xgamma
utility (uses XF86VidMode extension) and some RandR based
gamma ramp animation.
v2: Better resampling when subsampling the palette, as
proposed by Ville. Now reaches the max index of the
palette and deals with non-power-of-two sizes. Thanks.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Antoine Martin <antoine@nagafix.co.uk> (v1)
Cc: <ville.syrjala@linux.intel.com>
Specifically for xrgb2101010 format.
Tested on KDE Plasma-5 with XRender based composite
acceleration backend. Much smoother and faster.
(v2) Dropped argb2101010, because of depth 32 confusion with
argb8888, as pointed out by Eric. Thanks!
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Antoine Martin <antoine@nagafix.co.uk>
This makes it work properly with OpenGL based desktop
compositing, as tested with EGL and GLX based compositing
under OpenGL-2/3, and also artifact free with XRender
based 2D compositing.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Antoine Martin <antoine@nagafix.co.uk>
Support x-screens of depth 30, so init doesn't fail.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Antoine Martin <antoine@nagafix.co.uk>
Turns out that the kernel DRM ioctl handling returns EINVAL
instead of ENOTTY if one tries to call the new drmCrtcGetSequence()
or drmCrtcQueueSequence() ioctl's introduced in Linux 4.15 on an
older kernel where they are missing. This causes the fallback code
not to fall back to the old drmWaitVblank() ioctl and thereby
failure of vblank stuff.
E.g., on Linux 4.13, glxgears -info runs unthrottled at 10000 fps
instead of 60 fps. Also breakage of OML_sync_control extension.
Check for errno != EINVAL before setting has_queue_sequence = TRUE.
Additionally in case of supported drmCrtcQueueSequence(), set
has_queue_sequence = TRUE on success, or we might get at
least a temporary failure in ms_queue_vblank().
One slight ambiguity is that we can also get EINVAL if
drm_crtc_vblank_get() fails in the kernel, so if that
happened at first invocation of the new api, we'd fall
back to drmWaitVblank() and then fail there, instead of
failing in the new api, but the end result would be the
same.
Fixes: 44d5f2eb8a ("xf86-video-modesetting: Support new vblank kernel API [v2]")
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Cc: Keith Packard <keithp@keithp.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
This plumbs the full width color for solid pictures through to fb, exa,
and glamor. External drivers and acceleration code may wish to make a
similar change for sufficiently new servers.
v2: Don't break ABI (Michel Dänzer)
v2.1: Use the (correct) full color in fb too (Michel Dänzer)
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
These guards were dropped by the commit below, but it turns out they're
needed. Fixes crash on VT switch.
Fixes: d8ec33fe05 ("glx: Use vnd layer for dispatch (v4)")
Bugzilla: https://bugs.freedesktop.org/105233
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
We would throw BadValue here for the GLX_SCREEN attribute. The upper
dispatch layer already checks this, we can ignore it here.
Signed-off-by: Adam Jackson <ajax@redhat.com>
This is mostly for the client library's convenience, if this extension
is listed then it can know the attribute won't be rejected. Note that we
don't honor this attribute for indirect contexts. That's fine, we don't
want to introduce undefined behavior into a potentially privileged
process.
v2: Remember to ignore the attribute (Eric Anholt)
Signed-off-by: Adam Jackson <ajax@redhat.com>
Make a solid-fill picture for this instead of a 1x1 pixmap. In principle
the backend can accelerate this directly, and we also get to preserve
all the bits of the fill color.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
If an implicitly redirected window is unredirected by the reparent
operation, cw will be a stale pointer.
Signed-off-by: Peter Harris <pharris@opentext.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Listing the extensions is useful, despite being annoying for normal
usecases. Print it only when extra (lvl 3) vebose is requested.
v2: Move the logging to InitExtensions(), as requested by Adam.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Add the missing arguments to the function signature.
Fixes: e46820fb89 ("miinitext: introduce LoadExtensionList() to replace
over LoadExtension()")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Set the major/minor version it's currently implemented, not the one
provided by the GLXSERVER_VENDOR_ABI_{MAJOR,MINOR}_VERSION macros.
Those are identical for now, but can change in the future.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
malloc can return NULL, unlike GetVendorDispatchFunc. The latter
provides DispatchBadRequest.
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
When xdg_output support was added to Xwayland, need_rotate parameter was
added to output_get_new_size where true gave you the old pre-xdg_output
behavior and false gave the new behavior. Unfortunately, the two places
where this is called, need_rotate was set backwards. This caused input
get clampped to the wrong dimensions. Also, the logic for deciding
whether or not to flip was wrong because, if need_rotate was false, it
would always flip which is not what you want.
v2 (Daniel Stone):
- Fix output_get_new_size so that it doesn't flip the dimensions when
need_rotate is false.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Olivier Fourdan <ofourdan@redhat.com>
This call was forgotten to be removed in
90996f5909 in which
hostx_destroy_shm_segment() was introduced, which
itself does it.
Signed-off-by: Alexander Volkov <a.volkov@rusbitech.ru>
Reviewed-by: Adam Jackson <ajax@redhat.com>
When xorg_backtrace calls unw_get_proc_name and an error occurs, offset
might not be set for the current frame.
Initialize offset for each frame so that the offset from another frame
cannot be used inadvertently.
Signed-off-by: Jeff Smith <whydoubt@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
/home/ajax/git/xserver/at/../hw/dmx/input/dmxinputinit.c: In function ‘dmxInputInit’:
/home/ajax/git/xserver/at/../hw/dmx/input/dmxinputinit.c:1019:9: warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement]
int found;
Signed-off-by: Adam Jackson <ajax@redhat.com>
/home/ajax/git/xserver/at/../glx/glxdri2.c: In function ‘create_driver_context’:
/home/ajax/git/xserver/at/../glx/glxdri2.c:444:5: warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement]
const __DRIconfig *driConfig = config ? config->driConfig : NULL;
I don't know why autotools has decided not to be C99 but at least it's
trivial to fix.
Signed-off-by: Adam Jackson <ajax@redhat.com>
There's no reason a multithreaded client shouldn't be allowed to
interleave other requests (for other contexts) with a RenderLarge. Move
the check into __glXForceCurrent, and store the state in the context not
the client.
Signed-off-by: Adam Jackson <ajax@redhat.com>
This was only separate because GLX was loadable. The frontend is now
linked statically, so we can use the static extension list directly.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Add a stub for Xnest so it continues to link, but otherwise we support
GLX on every server so there's no need to make every DDX add it.
Signed-off-by: Adam Jackson <ajax@redhat.com>
The big change here is MakeCurrent and context tag tracking. We now
delegate context tags entirely to the vnd layer, and simply store a
pointer to the context state as the tag data. If a context is deleted
while it's current, we allocate a fake ID for the context and move the
context state there, so the tag data still points to a real context. As
a result we can stop trying so hard to detach the client from contexts
at disconnect time and just let resource destruction handle it.
Since vnd handles all the MakeCurrent protocol now, our request handlers
for it can just be return BadImplementation. We also remove a bunch of
LEGAL_NEW_RESOURCE, because now by the time we're called vnd has already
allocated its tracking resource on that XID.
v2: Update to match v2 of the vnd import, and remove more redundant work
like request length checks.
v3: Add/remove the XID map from the vendor private thunk, not the
backend. (Kyle Brenneman)
v4: Fix deletion of ghost contexts (Kyle Brenneman)
Signed-off-by: Adam Jackson <ajax@redhat.com>
This is based on an out-of-tree module written by Kyle:
https://github.com/kbrenneman/libglvnd/tree/server-libglx
I (ajax) did a bunch of cosmetic fixes, ported it off xfree86 API,
added request length checks, and fixed a minor bug or two.
v2: Use separate functions to set/get a context tag's private data, and
call the backend's MakeCurrent when a client disconnects to unbind the
context. (Kyle Brenneman)
Signed-off-by: Adam Jackson <ajax@redhat.com>