Nvidia needs dma-buf for glamor acceleration to work
when using the modesetting ddx driver.
Signed-off-by: stefan11111 <stefan11111@shitposting.expert>
Port https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/751
to xlibre
Fixes glamor with modesetting on nvidia
This is needed for glamor to work with modesetting
on nvidia, according to the nvidia docs:
https://download.nvidia.com/XFree86/Linux-x86_64/510.39.01/README/gbm.html
From the mr above:
The X server was passing GBM bos directly to
eglCreateImageKHR using the EGL_NATIVE_PIXMAP_KHR
target. Given the EGL GBM platform spec claims it
is invalid to create a EGLSurface from a native
pixmap on the GBM platform, implying there is no
mapping between GBM objects and EGL's concept of
native pixmaps, this seems a bit questionable.
This change modifies the bo import function to
extract all the required data from the bo and then
imports it as a dma-buf instead when the dma-buf +
modifiers path is available.
Signed-off-by: stefan11111 <stefan11111@shitposting.expert>
function prototypes are by default always extern, so no need for
that extra qualifier.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
The consumers always need to include <dix-config.h> at the very top
anyways, so no need to also include it (with extra guards) from
internal headers.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
This symbol is always defined, and the header is always present,
so no need to check for it.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
All in-tree consumers already need to include dix-config.h on the very
top anyways, and there are no out-of-tree consumers (private to fb subsys)
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
All in-tree consumers already need to include dix-config.h on the very
top anyways, and out-of-tree consumers don't have that include at all.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Not supposed to be written to (once the pointer is assigned),
so should be marked const. Also document KdOsInit()
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
These will be used by subsequent commits for generic Kdrive
functions calling back into the OS specific parts
Signed-off-by: stefan11111 <stefan11111@shitposting.expert>
Kdrive X servers used to do the OS-speciffic init part using KdOsInit.
This was changed in modern Xorg because Xephyr is the only kdrive
X server there, so there was no need to keep this generic.
Since we want to eventually add Xfbdev, we need to add this back.
Signed-off-by: stefan11111 <stefan11111@shitposting.expert>
The `types` variable is never used, just assigned NULL - and that NULL
value is passed to free(). This code is really doing nothing, never did so
since introduced in 2009.
Fixes: 8b6a370058
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Not neccessary to have entirely separate workflow for it - that's
just spamming the workflow list.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Some display drivers might cause p->atoms to be null if the display is
disabled. This will cause segfault when checking
if p->atoms[0] != property .
Some display drivers owned by qualcomm is known to cause this bug.
Signed-off-by: fish4terrisa-MSDSM <flyingfish.msdsm@gmail.com>
If ApplyPointerMapping() returns `MappingBusy`, this value needs to be
returned in the `success` reply field. On `Success` the returned value
needs to be `MappingSuccess`.
Closes: https://github.com/X11Libre/xserver/issues/353
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Since Xwayland is long gone, we don't need to install it's build-time
dependencies anymore.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
The assert(iclients) was in a place where the iclients variable could
not have been initialized ever - the first assignment is done *after* that.
Since we already have a BUG_RETURN_VAL() right be before it's actually
dereferenced, there's no need for that assert() at all - so drop it.
See https://github.com/X11Libre/xserver/issues/330
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
This struct (and associated functions) needs to be part of public driver ABI,
so video drivers can get notifications on damage certain operations, but
there hasn't been any documentation on it.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
This symbol is always defined, and the header is always present,
so no need to check for it.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Directly calling abort() doesn't print a stack trace, nor any useful message.
Instead FatalError() should be used in cases, where there's really no other
way than terminating the Xserver. The FatalError function also tries to make
a smooth shutdown (eg. resetting video mode), so console doesn't end up locked.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
The passed name buffer isn't changed by those functions ever
(it would be a bug if they did so), therefore it should be const.
Signed-off-by: stefan11111 <stefan11111@shitposting.expert>
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Add 4th parameter to clone_source() helper for checking out specific commit.
It will clone the given branch (but with full depth) and then checkout
the given commit.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
> ../hw/xquartz/mach-startup/bundle_trampoline.c:53:29: warning: a function declaration without a prototype is deprecated in all versions of C [-Wstrict-prototypes]
> static char *executable_path() {
> ^
> void
> 1 warning generated.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Previously we leaked stack when invalid enum parameters were
specified and caused __glGet*_size functions to return a 0 size.
Further, we read out-of-bounds (and leaked) when the input data was less
than 8 bytes (__glXDispSwap_GetFramebufferAttachmentParameteriv and
__glXDisp_GetRenderbufferParameteriv).
Now we only write a single element in the reply padding, and only when there
is a single element. This is what the Mesa client-side libGL expects, and
restores original GLX server behaviour, matching both pre-public (1996) SGI GLX
and XFree86 4.
The main risk of this change is if we have any error in element count or size;
previously it may not have mattered but now it does.
There are no piglit result changes from this modification using either mesa
libGLX or NVIDIA libGLX.
For performance considerations, an extra conditional and variable-length
memcpy has no meaningful impact on the indirect rendering pipeline cost.
There is still the possiblity to leak if our size checks allow an enum that
the GL implemention does not. Guarding against that requires zero-initializing
all temp storage, which wants re-evaluation of the blind 200-byte buffers
used for many calls and thus is a much bigger change.
Signed-off-by: Nathan Kidd <nkidd@rocketsoftware.com>
Part-of: <https://gitlab.freedesktop.org/xorg/xserver/-/merge_requests/1647>
Some drivers need to call into damage from within their CloseScreen proc,
so damage teardown needs to be done after that, instead of before.
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
