Dynamically added outputs should have their properties
properly updated as well. Otherwise we're left with an output
with many of its propeties not exposed.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Michel Dänzer <michel@daenzer.net>
Closes: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1143
(cherry picked from commit 0c5179c280)
A lot of that code is the same as in xf86-amdgpu and xf86-nouveau drivers. By removing that functions from
ms namespace we can move that code to common implementation.
Signed-off-by: Łukasz Spintzyk <lukasz.spintzyk@synaptics.com>
(cherry picked from commit 5be3b80b8d)
This mode for displays running on evdi/udl as side effect of failed glamor_egl_init
reverse_prime_offload_mode was initialized to FALSE
After Mesa upgrade to 21.0.0 GL_RENDERER is not llvmpipe that results in successful glamor_egl_init
and reverse_prime_offload_mode enabled.
This commit is explicitly disabling reverse_prime_offload_mode for evdi and udl drivers
Signed-off-by: Łukasz Spintzyk <lukasz.spintzyk@synaptics.com>
(cherry picked from commit 7e7c147105)
There was a time when setting a mode on a CRTC would not depend on the
associated connector's state. If a mode had been set successfully once,
it would mean it would work later on.
This changed with the introduction of new connectors type that now
require a link training sequence (DP, HDMI 2.0), and that means that
some events may have happened while the X server was not master that
would then prevent the mode from successfully be restored to its
previous state.
This patch relaxes the requirement that all modes should be restored on
EnterVT, or the entire X-Server would go down by allowing modesets to
fail (with some warnings). If a modeset fails, the CRTC will be
disabled, and a RandR event will be sent for the desktop environment to
fix the situation as well as possible.
Additional patches might be needed to make sure that the user would
never be left with all screens black in some scenarios.
v2 (Martin Peres):
- whitespace fixes
- remove the uevent handling (it is done in a previous patch)
- improve the commit message
- reduce the size of the patch by not changing lines needlessly
- return FALSE if one modeset fails in ignore mode
- add comments/todos to explain why we do things
- disable the CRTCs that failed the modeset
Signed-off-by: Kishore Kadiyala <kishore.kadiyala@intel.com>
Signed-off-by: Martin Peres <martin.peres@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Tested-by: Kishore Kadiyala <kishore.kadiyala@intel.com>
Closes: #1010
(cherry picked from commit efb3abddd4)
Normally, we would receive a uevent coming from Linux's DRM subsystem,
which would trigger the check for disappearing/appearing resources.
However, this event is not received when X is not master (another VT
is selected), and so the userspace / desktop environment would not be
notified about the changes that happened while X wasn't master.
To fix the issue, this patch forces a refresh on EnterVT by splitting
the kms-checking code from the uevent handling into its own (exported)
function called drmmode_update_kms_state. This function is then called
from both the uevent-handling function, and on EnterVT right before
restoring the modes.
Signed-off-by: Martin Peres <martin.peres@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Kishore Kadiyala <kishore.kadiyala@intel.com>
Tested-by: Kishore Kadiyala <kishore.kadiyala@intel.com>
(cherry picked from commit 293cf660c9)
EDID1.4 replaced GTF Bit with Continuous or Non-Continuous Frequency Display.
Check the "Display Range Limits Descriptor" for GTF support.
If panel doesn't support GTF, then add gtf modes.
Otherwise X will only show the modes in "Detailed Timing Descriptor".
V2: Coding style changes.
V3: Coding style changes, remove unused variate.
V4: remove unused variate.
BugLink: https://gitlab.freedesktop.org/drm/intel/issues/313
Signed-off-by: Aaron Ma <aaron.ma@canonical.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
(cherry picked from commit 6a79a737e2)
Since the introduction of "modesetting: Remove unnecessary fb addition from
drmmode_xf86crtc_resize" the fb_id isn't initialited at
drmmode_xf86crtc_resize.
Rotate operation of XRandR uses rotate_bo. So in this case the fb_id
associated to the front_bo is not initialized at drmmode_set_mode_major.
So fd_id remains 0.
As every call to drmmode_xf86crtc_resize allocates a new front_bo we should
destroy unconditionally the old_front_bo if operation success. So we free
the allocated GBM handles.
This avoids crashing xserver with a OOM in the RPI4 1Gb at 4k resolution
after 3 series xrandr rotations from normal to left and vice versa reported at
https://github.com/raspberrypi/firmware/issues/1345
Signed-off-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Closes: https://gitlab.freedesktop.org/xorg/xserver/-/issues/1024
Fixes: 8774532121 "modesetting: Remove unnecessary fb addition from
drmmode_xf86crtc_resize"
(cherry picked from commit 73480f172a)
The miPointerSpriteFunc swcursor code expects there to only be a single
framebuffer and when the cursor moves it will undo the damage of the
previous draw, potentially overwriting what ever is there in a new
framebuffer installed after a flip.
This leads to all kind of artifacts, so we need to disable pageflipping
when a swcursor is used.
The code for this has shamelessly been copied from the xf86-video-amdgpu
code.
Fixes: https://gitlab.freedesktop.org/xorg/xserver/issues/828
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
(cherry picked from commit 0aaac8d783)
Signed-off-by: Łukasz Spintzyk <lukasz.spintzyk@displaylink.com>
The atomic driver has issues with modesetting when stealing
connectors from a different crtc, a black screen when doing rotation
on a different crtc, and in general is just a mapping of the legacy
helpers to atomic. This is already done in the kernel, so just
fallback to legacy by default until this is fixed.
Please backport to 1.20, as we don't want to enable it for everyone
there. It breaks for existing users.
The fixes to make the xserver more atomic have been pending on the
mailing list for ages.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110375
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110030
References: https://gitlab.freedesktop.org/xorg/xserver/merge_requests/36/commits
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
(cherry picked from commit f0d78b47ac)
I introduced this error with the MST hotplug code, but it can trigger
on zaphod setups, and is perfectly fine. There is no support for
MST/hotplug on zaphod setups currently, so we can just skip over
the dynamic connector handling here. However we shouldn't skip
over the lease handling so move it into the codepath.
Fixes: 9257b1252d ("modesetting: add dynamic connector hotplug support (MST) (v3)")
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
(cherry picked from commit 1cfdd1a965)
For the miClearDrawable prototype. Apparently it doesn't get pulled in
for some build configurations, breaking the build.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
(cherry picked from commit a24a786fc8)
Calling rrGetScrPriv when RandR isn't initialized causes an assertion
failure that aborts the server:
Xorg: ../include/privates.h:121: dixGetPrivateAddr: Assertion `key->initialized' failed.
Thread 1 "Xorg" received signal SIGABRT, Aborted.
0x00007ffff78a8f25 in raise () from /usr/lib/libc.so.6
(gdb) bt
#0 0x00007ffff78a8f25 in raise () from /usr/lib/libc.so.6
#1 0x00007ffff7892897 in abort () from /usr/lib/libc.so.6
#2 0x00007ffff7892767 in __assert_fail_base.cold () from /usr/lib/libc.so.6
#3 0x00007ffff78a1526 in __assert_fail () from /usr/lib/libc.so.6
#4 0x00007ffff7fb57c1 in dixGetPrivateAddr (privates=0x555555ab1b60, key=0x555555855720 <rrPrivKeyRec>) at ../include/privates.h:121
#5 0x00007ffff7fb5822 in dixGetPrivate (privates=0x555555ab1b60, key=0x555555855720 <rrPrivKeyRec>) at ../include/privates.h:136
#6 0x00007ffff7fb586a in dixLookupPrivate (privates=0x555555ab1b60, key=0x555555855720 <rrPrivKeyRec>) at ../include/privates.h:166
#7 0x00007ffff7fb8445 in CreateScreenResources (pScreen=0x555555ab1790) at ../hw/xfree86/drivers/modesetting/driver.c:1335
#8 0x000055555576c5e4 in xf86CrtcCreateScreenResources (screen=0x555555ab1790) at ../hw/xfree86/modes/xf86Crtc.c:744
#9 0x00005555555d8bb6 in dix_main (argc=4, argv=0x7fffffffead8, envp=0x7fffffffeb00) at ../dix/main.c:214
#10 0x00005555557a4f0b in main (argc=4, argv=0x7fffffffead8, envp=0x7fffffffeb00) at ../dix/stubmain.c:34
This can happen, for example, if the server is configured with Xinerama
and there is more than one X screen:
Section "ServerLayout"
Identifier "crash"
Screen 0 "modesetting"
Screen 1 "dummy" RightOf "modesetting"
Option "Xinerama"
EndSection
Section "Device"
Identifier "modesetting"
Driver "modesetting"
EndSection
Section "Screen"
Identifier "modesetting"
Device "modesetting"
EndSection
Section "Device"
Identifier "dummy"
Driver "dummy"
EndSection
Section "Screen"
Identifier "dummy"
Device "dummy"
EndSection
The problem does not reproduce if there is only one X screen because of
this code in xf86RandR12Init:
#ifdef PANORAMIX
/* XXX disable RandR when using Xinerama */
if (!noPanoramiXExtension) {
if (xf86NumScreens == 1)
noPanoramiXExtension = TRUE;
else
return TRUE;
}
#endif
Fix the problem by checking dixPrivateKeyRegistered(rrPrivKey) before
calling rrGetScrPriv. This is similar to what the xf86-video-amdgpu
driver does:
fd66f5c0be/src/amdgpu_kms.c (L388)
Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
(cherry picked from commit 4226c6d032)
Fixes random garbage being visible intermittently.
Reviewed-by: Adam Jackson <ajax@redhat.com>
(Cherry picked from commit 9ba13bac9d)
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
This makes sure any pending drawing to a new scanout buffer will be
visible from the start.
This makes the finish call in drmmode_copy_fb superfluous, so remove it.
Reviewed-by: Adam Jackson <ajax@redhat.com>
(Cherry picked from commit c66c548eab)
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
We now ask Glamor to use EGL_MESA_query_driver to obtain the DRI driver
name; if successful, we use that as the DRI driver name. Following the
existing dri2.c logic, we also use the same name for the VDPAU driver,
except for i965 (and now iris), where we switch to the "va_gl" fallback.
This allows us to bypass the PCI ID lists in xserver and centralize the
driver selection mechanism inside Mesa. The hope is that we no longer
have to update these lists for any future hardware.
(backported from commit 8d4be7f6c4)
Acked-by: Michel Dänzer <mdaenzer@redhat.com>
ms_present_get_crtc() returns an RRCrtcPtr, but derives it from a xf86CrtcPtr
found via ms_dri2_crtc_covering_drawable()=>ms_covering_crtc(). As a result, it
depends on all associated DIX ScreenRecs having an xf86CrtcConfigPtr DDX
private.
Some DIX ScreenRecs don't have an xf86CrtcConfigPtr DDX private, but do have an
rrScrPrivPtr DDX private. Given that we can derive all of the information we
need from RandR, we can support these screens by avoiding the use of xf86Crtc.
This change implements an RandR-based path for ms_present_get_crtc(), allowing
drawables to successfully fall back to syncing to the primary output, even if
the slave doesn't have an xf86CrtcConfigPtr DDX private.
Without this change, if a slave doesn't have an xf86CrtcConfigPtr DDX private,
drawables will fall back to 1 FPS if they overlap an output on that slave.
Signed-off-by: Alex Goins <agoins@nvidia.com>
(cherry picked from commit 562c7888be)
DIX ScreenRecs don't necessarily have an xf86CrtcConfigPtr DDX private.
ms_covering_crtc() assumes that they do, which can result in a segfault.
Update ms_covering_crtc() to check the XF86_CRTC_CONFIG_PTR() returned pointer
before dereferencing it. This will still mean that ms_covering_crtc() can't fall
back to the primary output when a drawable overlaps a slave output (going to the
1 FPS default instead), but it won't segfault.
Signed-off-by: Alex Goins <agoins@nvidia.com>
(cherry picked from commit 797e7a0ceb)
ms_covering_crtc() uses RRFirstOutput() to determine a primary output to fall
back to if a drawable is overlapping a slave output.
If the primary output is a slave output, RRFirstOutput() will return a slave
output even if passed a master ScreenPtr. ms_covering_crtc() dereferences the
output's devPrivate, which is invalid for non-modesetting outputs, and can
crash.
Changing RRFirstOutput() could have unintended side effects for other callers,
so this change replaces the call to RRFirstOutput() with ms_first_output().
ms_first_output() ignores the primary output if it doesn't match the given
ScreenPtr, choosing the first connected output instead.
Signed-off-by: Alex Goins <agoins@nvidia.com>
(cherry picked from commit 3ef9029ace)
The autoconf build for the modesetting driver still relied on
xorg-macros.m4 for string replacements and did not include the
top-level manpages.am. As a result, no substitutions took place after
commit 2e497bf887.
This should be a candidate for the 1.20 branch.
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
(cherry picked from commit de0d39f825)
Noticed when porting this logic to xf86-video-nouveau, and valgrind
complained about conditional jump based on uninitialized data.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.com>
(cherry picked from commit 48b1af2718)
Believe it or not, somehow we've never done this in legacy mode! We
currently simply change the DPMS property on the CRTC's output's
respective DRM connector, but this means that we're just setting the
CRTC as inactive-not disabled. From the perspective of the kernel, this
means that any shared resources used by the CRTC are still in use.
This can cause problems for drivers that are not yet fully atomic,
despite using the atomic helpers internally. For instance: if CRTC-1 and
CRTC-2 are still enabled and use shared resources within the kernel (an
MST topology, for example), and then userspace tries to go enable CRTC-3
on the same topology this might suddenly fail if CRTC-3 needs the shared
resources CRTC-1 and CRTC-2 are using. While I don't know of any
situations in the mainline kernel that actually trigger this, future
plans for reworking the atomic check of MST drivers are absolutely
going to make this into a real issue (they already are in my WIP
branches for the kernel).
So: actually do the right thing here and disable CRTCs when they're not
going to be used anymore, even in legacy mode.
Signed-off-by: Lyude Paul <lyude@redhat.com>
(cherry picked from commit 7a44e8d400)
The destination is always either on the stack or in the middle of some
struct.
Signed-off-by: Adam Jackson <ajax@redhat.com>
(cherry picked from commit 43a0f9a5db)
This saves us having to make sure we clean it up.
Pointed out by coverity.
Signed-off-by: Dave Airlie <airlied@redhat.com>
(cherry picked from commit b6c29a881e)
The X will be crashed on the system with other DDX driver,
such as amdgpu.
show the log like:
randr: falling back to unsynchronized pixmap sharing
(EE)
(EE) Backtrace:
(EE) 0: /usr/lib/xorg/Xorg (xorg_backtrace+0x4e)
(EE) 1: /usr/lib/xorg/Xorg (0x55cb0151a000+0x1b5ce9)
(EE) 2: /lib/x86_64-linux-gnu/libpthread.so.0 (0x7f1587a1d000+0x11390)
(EE)
(EE) Segmentation fault at address 0x0
(EE)
The issue is that modesetting as the master, and amdgpu as the slave.
Thus, when the master attempts to access pSlavePixPriv in ms_dirty_update(),
problems result due to the fact that it's accessing AMD's 'ppriv' using the
modesetting structure definition.
Apart from fixing crash issue, the patch fix other issue in master interface
in which driver should refer to master pixmap.
Signed-off-by: Jim Qu <Jim.Qu@amd.com>
Reviewed-by: Alex Goins <agoins@nvidia.com>
(cherry picked from commit f79e536851)
Instead of having every video driver loop over any pending leases to
free them during CloseScreen, do this up in the DIX layer by
terminating leases when a leased CRTC or Output is destroyed and
(just to make sure), also terminating leases in RRCloseScreen. The
latter should "never" get invoked as any lease should be associated
with a resource which was destroyed.
This is required as by the time the driver's CloseScreen function is
invoked, we've already freed all of the DIX randr structures and no
longer have any way to reference the leases
Signed-off-by: Keith Packard <keithp@keithp.com>
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=106960
Cc: Thomas Hellstrom <thellstrom@vmware.com>
(cherry picked from commit 1ef7aed3e2)
The recent rewrite of modesetting driver broke the 24bpp support.
As typically found on cirrus KMS, it leads to a blank screen, spewing
the error like:
failed to add fb -22
(EE) modeset(0): failed to set mode: Invalid argument
The culript is that the wrong bpp value of the front buffer is passed
to drmModeAddFB(). Fix it by replacing with the back buffer bpp,
drmmode->kbpp.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Tested-by: Stefan Dirsch <sndirsch@suse.de>
Reviewed-by: Adam Jackson <ajax@redhat.com>
(cherry picked from commit d625e16918)
When setting DefaultDepth to 16 in the Screen section, the current
code requests a 32 bpp framebuffer, however the X-Server seems to
assumes 16 bpp.
Fixes commit 21217d0216 ("modesetting: Implement 32->24 bpp
conversion in shadow update")
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Stefan Agner <stefan@agner.ch>
(cherry picked from commit 1c7f34e99f)
If we're using atomic modesetting, then we're also using universal
planes, and so the lease we create needs to include the plane.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
(cherry picked from commit d83efc47b7)
We don't want universal_planes unless we're using atomic APIs for
modesetting, and the kernel already enables universal_planes
automatically when atomic is enabled.
If we enable universal_planes when we're not using atomic, then we
won't have selected a plane for each crtc, and this will break lease
creation which requires planes for each output when universal_planes
is enabled.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
(cherry picked from commit 4a11f66e46)
This lets an application open a suitable DRM device and pass the file
descriptor to the mode setting driver through an X server command line
option, '-masterfd'.
There's a companion application, xlease, which creates a DRM master by
leasing an output from another X server. That is available at
git clone git://people.freedesktop.org/~keithp/xlease
v2:
Always print usage, but note that it can't be used if
setuid/gid
Suggested-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
(cherry picked from commit 38ff29ec8e)
This really sucked to find out :(
Signed-off-by: Lyude Paul <lyude@redhat.com>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
(cherry picked from commit c41d4ff48f)
So, this did actually work on older kernels at one point in time,
however it seems that this working was a result of some of the Linux
kernel's atomic modesetting helpers not preserving the CRTC's enabled
state in the right spots. This was fixed in:
846c7dfc1193 ("drm/atomic: Try to preserve the crtc enabled state in drm_atomic_remove_fb, v2")
As a result, atomic commits which simply disassociate a DRM connector
with it's CRTC while leaving the CRTC in an enabled state aren't enough
to disable the CRTC, and result in the atomic commit failing. This
currently can cause issues with MST hotplugging where X will end up
failing to disable the MST outputs after they've left the system. A
simple reproducer:
- Start up Xorg
- Connect an MST hub with displays connected to it
- Remove the hub
- Now there should be CRTCs stuck on the orphaned MST connectors, and X
won't be able to reclaim them.
Signed-off-by: Lyude Paul <lyude@redhat.com>
Cc: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
(cherry picked from commit c12f1bd4b7)
drmmode_shadow_allocate() still uses drmModeAddFB() which may fail if
the format is not as expected, preventing from using a rotated output.
Change it to use the new function drmmode_bo_import() which takes care
of calling the drmModeAddFB2() API.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106715
Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Tested-by: Tomas Pelka <tpelka@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
(cherry picked from commit a85e94a50c)
We don't want DRM file descriptors to leak to child processes.
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
(cherry picked from commit 315c63c41d)
Modifiers support needs gbm as a dependency. Without setting the dependency
included headers are not found reliably and the build might fail if the
headers are not placed in the default system include paths.
Signed-off-by: Roman Gilg <subdiff@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
The old 32-Bit wraparound handling didn't actually work, due to some
integer casting bug, and the mapping was ill equipped to deal with input
from the new true 64-bit GetCrtcSequence/QueueCrtcSequence api's
introduced in Linux 4.15.
For 32-Bit truncated input from pageflip events and old vblank events
and old drmWaitVblank ioctl, implement new wraparound handling, which
also allows to deal with wraparound in the other direction, e.g., if a
32-Bit truncated sequence value is passed in, whose true 64-Bit
in-kernel hw value is within 2^30 counts of the previous processed
value, but whose 32-bit truncated sequence value happens to lie just
above or below a 2^32 boundary, iow. one of the two values 'sequence'
vs. 'msc_prev' lies above a 2^32 border, the other one below it.
The method is directly translated from Mesa's proven implementation of
the INTEL_swap_events extension, where a true underlying 64-Bit wide
swapbuffers count (SBC) needs to get reconstructed from a 32-Bit LSB
truncated SBC transported over the X11 protocol wire. Same conditions
apply, ie. successive true 64-Bit SBC values are close to each other,
but don't always get received in strictly monotonically increasing
order. See Mesa commit cc5ddd584d17abd422ae4d8e83805969485740d9 ("glx:
Handle out-of-sequence swap completion events correctly. (v2)") for
explanation.
Additionally add a separate path for true 64-bit msc input originating
from Linux 4.15+ drmCrtcGetSequence/QueueSequence ioctl's and
corresponding 64-bit vblank events. True 64-bit msc's don't need
remapping and must be passed through.
As a reliability bonus, they are also used here to update the tracking
values msc_prev and ms_high with perfect 64-Bit ground truth as baseline
for mapping msc from pageflip completion events, because pageflip events
are always 32-bit wide, even when the new kernel api's are used. Because
each pageflip(-event) is always preceeded close in time (and vblank
count) by a drmCrtcQueueSequence queued event or drmCrtcGetSequence
query as part of DRI2 or DRI3+Present swap scheduling, we can be certain
that each pageflip event will get its truncated 32-bit msc remapped
reliably to the true 64-bit msc of flip completion whenever the sequence
api is available, ie. on Linux 4.15 or later.
Note: In principle at least the 32-bit mapping path could also be
backported to earlier server branches, as this seems to be broken for at
least server 1.16 to 1.19.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Cc: Keith Packard <keithp@keithp.com>
Cc: Michel Dänzer <michel.daenzer@amd.com>
The function is ported from intel-ddx uxa backend around 2013, where its
stated purpose was to apply a vblank_offset to msc values to correct for
problems with those kernel provided msc values. Some (somewhat magic and
puzzling to myself) heuristic tried to guess if provided values were
unreasonable and tried to adapt the corrective vblank_offset to account
for that.
Except: It wasn't applied to kernel provided msc values, but the values
delivered by clients via DRI2 or Present, so valid client targetmsc
values, e.g., requesting a vblank event > 1000 vblanks in the future,
triggered the offset correction in arbitrarily wrong ways, leading to
wrong msc values being returned and thereby vblank events queued to the
kernel for the wrong time. This causes glXSwapBuffersMscOML and
glXWaitForMscOML to swap / return immediately whenever a swap/wait in >
1000 vblanks is requested.
The original code was also written to only deal with 32 bit mscs, but
server 1.20 modesetting ddx can now use new Linux 4.15+ kernel vblank
api to process true 64 bit msc's, which may confuse the heuristic even
more due to 32 bit integer truncation/wrapping.
This code caused various problems in the intel-ddx in the past since
year 2013, and was removed there in 2015 by Chris Wilson in commit
42ebe2ef9646be5c4586868cf332b4cd79bb4618:
" uxa: Remove the filtering of bogus Present MSC values
If the intention was to filter the return values from the kernel, the
filtering would have been applied to the kernel values and not to the
incoming values from Present. This filtering introduces crazy integer
promotion and truncation bugs all because Present feeds garbage into its
vblank requests.
"
Indeed, i found a Mesa bug yesterday which can cause Mesa's
PresentPixmap request to spuriously feed garbage targetMSC's into the
driver under some conditions. However, while other video drivers seem to
cope relatively well with that, modesetting ddx causes KDE-5's
plasmashell to lock up badly quite frequently, and my suspicion is that
the code removed in this commit is one major source of the extra
fragility.
Also my own tests fail for any swap scheduled more than 1000 vblanks
into the future, which is not uncommon for some scientific applications.
Iow. modesetting's swap scheduling seems to be more robust without this
function afaics.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Keith Packard <keithp@keithp.com>
Tested-by: Mike Lothian <mike@fireburn.co.uk>
GBM objects were never destroyed after looking for format and
modifier compatibility when deciding whether flipping or copying
a presented pixmap.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106106
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Use the DRM_CAP_ADDFB2_MODIFIERS query to make sure the kms
driver supports modifiers in the addfb2 ioctl, and fall back
to addfb ioctl without modifiers if modifiers are unsupported.
E.g., as of Linux 4.17, nouveau-kms so far does not suppport
modifiers and gets angry if drmModeAddFB2WithModifiers() is
called (-> failure to set a video mode -> blank screen), but
Mesa's nvc0+ gallium driver causes gbm_bo_get_modifier() to
return a valid modifier by translating the default tiling of
bo's created via gbm_bo_create() into a modifier other than
DRM_FORMAT_MOD_INVALID (see Mesa's nvc0_miptree_get_modifier()).
Testing for != DRM_FORMAT_MOD_INVALID is apparently not
sufficient for safe use of drmModeAddFB2WithModifiers.
Bonus: Handle potential failure of populate_format_modifiers().
The required DRM_CAP is defined since libdrm v2.4.65, and we
require v2.4.89+ for the server, so we can use it unconditionally.
Tested on intel-kms, radeon-kms, nouveau-kms. Fixes failure on
NVidia Pascal.
Fixes: 2f807c2324 ("modesetting: Add support for multi-plane pixmaps when page-flipping")
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Cc: Daniel Stone <daniels@collabora.com>
Cc: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
ms_queue_vblank() returns false on failure.
Reported-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Frank Binns <frank.binns@imgtec.com>
Tested-by: Mike Lothian <mike@fireburn.co.uk>
We would fail to get the FB ID if it wasn't already imported, since we
were checking to see if the pointer was NULL (it never was) rather than
if the content of the pointer was 0.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reported-by: Olivier Fourdan <ofourdan@redhat.com>
Tested-by: Olivier Fourdan <ofourdan@redhat.com>
Reviewed-by: Olivier Fourdan <ofourdan@redhat.com>
drmmode_crtc_set_mode has a loop nested inside another loop, where both
of them were using 'i' as the loop iterator. Rename it to avoid an
infinite loop.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Reported-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-and-Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Non-atomic kms drivers like radeon-kms (or nouveau-kms with
default setting of "atomic ioctl disabled") don't export
any formats, so num_formats == 0.
Some atomic drivers (nouveau-kms with boot param nouveau.atomic=1,
or intel-kms on, e.g., Linux 4.13) expose num_formats == 0, or
don't expose any modifiers, so num_modifiers == 0.
Let the drmmode_is_format_supported() check pass in these cases
to allow page flipping, as it works just fine.
Tested on NV-96 for nouveau, HD-5770 for radeon, Intel Ivybridge
with Linux 4.13 and drm-next to fix page flipping.
Fixes: 9d147305b4 ("modesetting: Check if buffer format is supported when flipping")
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
We need to make sure that the atomic commit are consistent
or else the kernel will reject it. For example, when moving
a CRTC from one output to another one, the first output CRTC_ID
property needs to be reset. Also if the second output was using
another CRTC beforehands, it needs to be disabled to avoid an
inconsistent state.
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Tested-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Ville Syrjälä
Łukasz Spintzyk
Kishore Kadiyala
Martin Peres
Aaron Ma
Jose Maria Casanova Crespo
Hans de Goede
Maarten Lankhorst
Dave Airlie
Michel Dänzer
Aaron Plattner
Kenneth Graunke
Alex Goins
Sven Joachim
Ilia Mirkin
Lyude Paul
Adam Jackson
Jim Qu
Keith Packard
Takashi Iwai
Stefan Agner
Olivier Fourdan
Michel Dänzer
Roman Gilg
Mario Kleiner
Louis-Francis Ratté-Boulianne
Matt Turner
Daniel Stone