Keith Packard [Sat, 9 Nov 2013 20:36:47 +0000 (12:36 -0800)]
present: Signal destroyed flip window with vblank->window == NULL
This eliminates dereferencing freed window pointers when there is a
flip for that window in progress. The flip will complete, and then
immediately get undone (as we can't stop an in-progress flip).
Remove the vblank->window_destroyed field as we can signal this with
vblank->window == NULL instead.
Change check to vblank->window == NULL in:
present_flip_notify
Add check for vblank->window == NULL in:
present_vblank_notify
present_execute
present_flip_notify was also using vblank->window->drawable.pScreen,
so stop doing that and use vblank->screen instead.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Sat, 9 Nov 2013 20:33:02 +0000 (12:33 -0800)]
present: Ignore event_id 0 from driver vblank notify
We use event_id 0 to mean 'no such event'; if a driver sends us that
event_id, make sure we don't accidentally match it.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Thu, 7 Nov 2013 22:17:12 +0000 (14:17 -0800)]
present: Round fake MSC computations intead of truncating
If the timer fired too early, we'd sometimes mis-compute the MSC for
fake vblanks. Rounding the computation to the nearest MSC fixes this nicely.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Peter Hutterer [Mon, 18 Nov 2013 22:07:09 +0000 (08:07 +1000)]
configure: allow for --enable-libunwind and --disable-libunwind
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Thu, 14 Nov 2013 08:02:04 +0000 (17:02 +0900)]
Merge remote-tracking branch 'whot/for-keith'
Adam Jackson [Tue, 29 Oct 2013 16:09:27 +0000 (12:09 -0400)]
mieq: Bump default queue size to 512
Based on some bugzilla scraping I did around November 2012. Of xserver
bugs in Red Hat bugzilla with an EQ size message in the log, the
distribution looked like:
String | Matches
-------------------------------------
Increasing EQ size to 512 | 460
Increasing EQ size to 1024 | 52
Increasing EQ size to 2048 | 6
Increasing EQ size to 4096 | 0
Most of the "512" ones appear to be mostly harmless, some relatively
expensive path in either rendering or resource destruction simply taking
too long due to external pressures like paging or CPU contention. So
let's raise the initial queue size, both to reduce the number of
spurious abrt reports and to drop fewer events in all but the most
pathological cases.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
Dave Airlie [Tue, 29 Oct 2013 16:09:26 +0000 (12:09 -0400)]
xfree86: return NULL for compat output if no outputs.
With outputless GPUs showing up we crash here if there are not outputs
try and recover with a bit of grace.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Dan Horák [Tue, 29 Oct 2013 16:09:25 +0000 (12:09 -0400)]
test: build the touch test only when building Xorg
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Dan Horák <dan@danny.cz>
Signed-off-by: Keith Packard <keithp@keithp.com>
Peter Hutterer [Wed, 13 Nov 2013 07:14:11 +0000 (17:14 +1000)]
include: export key_is_down and friends
VNC needs key_is_down to check if a key is processed as down before it
simulates various key releases. Make it available, because I seriously can't
be bothered thinking about how to rewrite VNC to not need that.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Acked-by: Keith Packard <keithp@keithp.com>
Alan Coopersmith [Tue, 12 Nov 2013 22:46:04 +0000 (14:46 -0800)]
Stop including inline assembly .il file for Solaris Studio builds
Since all the inb/outb/etc. use in the X server itself (except for
xf86SlowBcopy) has been replaced by calls to libpciaccess, we no
longer need to pass inline assembly files to replace the gcc inline
assembly from hw/xfree86/common/compiler.h when building Xorg itself.
The .il files are still generated and installed in the SDK for the
benefit of drivers who may use them.
Binary diff of before and after showed that xf86SlowBcopy was the
only function changed across the Xorg binary and all modules built
in the Xserver build, it just calls the outb() function now instead
of having the outb instructions inlined, making it a slightly slower
bcopy.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Peter Hutterer [Fri, 4 Oct 2013 00:55:52 +0000 (10:55 +1000)]
kdrive: fix cursor jumps on CursorOffScreen behavior
This patch fixes cursor jumps when there is a grab on the Xephyr window and
the pointer moves outside the window.
So on two side-by-side 640x480 screens, a coordinate of 0/481
triggers KdCursorOffscreen.
If the delta between two screens is 0, they share the same offset for
that dimension. When searching for the new screen, the loop always rules out
the current screen. So we get to the second screen, trigger the conditions
where dy <= 0 and decide that this new screen is the correct one. The result
is that whenever KdCursorOffScreen is called, the pointer jumps to the other
screen.
Change to check for dy < 0 etc. so that the cursor stays on the same screen if
there is no other screen at the target location.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Peter Hutterer [Tue, 29 Oct 2013 04:24:41 +0000 (14:24 +1000)]
kdrive: modify ephyr events to use POINTER_DESKTOP and scale them to that
A multi-head Xephyr instance has the pointer stuck on one screen
because of bad coordinate calculation. The coordinates passed to
GetPointerEvents are per-screen, so the cursor gets stuck on the left-most
screen by default.
Adjust and mark the events as POINTER_DESKTOP, so the DIX
can adjust them accordingly.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Keith Packard [Thu, 14 Nov 2013 03:21:54 +0000 (13:21 +1000)]
kdrive: handle WxH as valid geometry spec
If a screen size was specified as WxH, the loop returned early and kdOrigin
was never advanced. Thus, screen->origin was always 0 (or whatever was given
at the -origin commandline flag).
If a screen size was given with a bit depth (WxHxD), kdOrigin would always
advance by the current screen, offsetting the next screen.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Peter Hutterer [Tue, 29 Oct 2013 04:33:56 +0000 (14:33 +1000)]
ephyr: xcb_connect returns an error, not NULL
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Keith Packard [Thu, 14 Nov 2013 01:26:20 +0000 (10:26 +0900)]
Update to version 1.14.99.902
1.15 RC2
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Tue, 12 Nov 2013 22:01:45 +0000 (07:01 +0900)]
xext: Fix shmint.h to not use headers outside of sdk_HEADERS
shmint.h is part of sdk_HEADERS, and so can't use anything not
included in sdk_HEADERS.
busfault.h includes dix-config.h which is not. Leave the use of
struct busfault in shmint.h and move the include of busfault.h to
shm.c.
protocol-versions.h is not part of sdk_HEADERS, so instead of using
that, just use XTRANS_SEND_FDS to choose whether to expose the fd
passing requests directly.
Reported-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Keith Packard <keithp@keithp.com>
Tested-by: Knut Petersen <Knut_Petersen@t-online.de>
Reviewed-by: Julien Cristau <jcristau@debian.org>
v2: also avoid using protocol-versions.h
Jeremy Huddleston Sequoia [Fri, 8 Nov 2013 22:55:33 +0000 (14:55 -0800)]
xfree86: Fix build without libpciaccess
Regression fix from commit
04ab07ca19236d6c9a947e065fb69b0dd0d16639
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Reviewed-by: Connor Behan <connor.behan@gmail.com>
Alan Coopersmith [Fri, 8 Nov 2013 05:08:56 +0000 (21:08 -0800)]
Enable XTRANS_SEND_FDS on Solaris too.
Requires passing through the __EXTENSIONS__ and _XOPEN_SOURCE defines
in order to expose the msg_control members in struct msghdr.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Alan Coopersmith [Fri, 8 Nov 2013 05:05:33 +0000 (21:05 -0800)]
Avoid conflicts with Solaris <sys/regset.h> defines that clash with our names
When building on Solaris with _XOPEN_SOURCE set to a recent XPG release,
<stdlib.h> and other core headers start including <sys/regset.h>, which
has a bunch of unfortunately named macros such as "CS", "ES", etc. for
x86 & x64 registers which clash with existing variable & struct member
names in Xorg - so #undef these so they don't interfere with our use.
(Yes, have filed a bug against the system headers for exposing these,
but this solves the problem for building on existing releases.)
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Mon, 11 Nov 2013 23:48:41 +0000 (15:48 -0800)]
Proper spelling of MAP_ANONYMOUS is MAP_ANON.
The former doesn't exist on BSD and the latter is available everywhere
AFAIK (checked Solaris and Linux).
You also might want to wrap that line ;).
Reported-by: Mark Kettenis <mark.kettenis@xs4all.nl>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Thu, 7 Nov 2013 23:18:28 +0000 (15:18 -0800)]
Get rid of the rest of the FD passing code when XTRANS_SEND_FDS isn't set
req_fds and SetReqFds in include/dixstruct.h
ReadFdFromClient, WriteFdToClient and the FD flushing in os/io.c
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Mon, 11 Nov 2013 23:26:12 +0000 (15:26 -0800)]
Merge remote-tracking branch 'jeremyhu/master'
Adam Jackson [Thu, 7 Nov 2013 17:01:41 +0000 (12:01 -0500)]
present: Don't try to initialize when building without present support
There's a --disable-present, so it'd be nice if it worked.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Thu, 7 Nov 2013 11:19:42 +0000 (03:19 -0800)]
present: Change debug output a bit to help diagnose missing vblank signals
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Thu, 7 Nov 2013 07:56:26 +0000 (23:56 -0800)]
Present: Check for window/fence destroyed when idling pixmaps
A client destroying objects in the middle of an unflip can end up
having the screen flip window or fence set to NULL in the unflip
notify path. Check for these and don't try to use those objects.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Sun, 3 Nov 2013 21:12:40 +0000 (13:12 -0800)]
Trap SIGBUS to handle truncated shared memory segments
If a client passes a section of memory via file descriptor and then
subsequently truncates that file, the underlying pages will be freed
and the addresses invalidated. Subsequent accesses to the page will
fail with a SIGBUS error.
Trap that SIGBUS, figure out which segment was causing the error and
then allocate new pages to fill in for that region. Mark the offending
shared segment as invalid and free the resource ID so that the client
will be able to tell when subsequently attempting to use the segment.
Signed-off-by: Keith Packard <keithp@keithp.com>
v2: Use MAP_FIXED to simplify the recovery logic (Mark Kettenis)
v3: Also catch errors in ShmCreateSegment
Conflicts:
include/dix-config.h.in
include/xorg-config.h.in
Keith Packard [Thu, 7 Nov 2013 07:22:51 +0000 (23:22 -0800)]
Require libXtrans version 1.3.2
This has the FD passing support included
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Sun, 3 Nov 2013 18:08:15 +0000 (10:08 -0800)]
Xext: Enable MIT-SHM FD-passing request definitions only when possible
Check to see if xtrans FD passing is available and use that to
advertise the appropriate version of the SHM extension
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Keith Packard [Thu, 7 Nov 2013 07:06:28 +0000 (23:06 -0800)]
Require xextproto version 7.2.99.901
This includes the MIT-SHM FD passing requests
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Keith Packard [Thu, 7 Nov 2013 22:20:26 +0000 (14:20 -0800)]
Remove dix-config.h config variables from xorg-config.h
xorg-config.h includes dix-config, so there's no need to duplicate.
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Thu, 7 Nov 2013 07:05:46 +0000 (23:05 -0800)]
Make XTrans FD passing support optional. Define only on Linux
Until other operating systems have a libXtrans port for FD passing,
disable this on non-Linux systems.
Note that this define affects how libXtrans gets built into the X
server, which is why it need only define the symbol
Signed-off-by: Keith Packard <keithp@keithp.com>
Jeremy Huddleston Sequoia [Sun, 10 Nov 2013 19:40:27 +0000 (11:40 -0800)]
XQuartz: Don't sleep forever on exit if usleep() returns an error
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Jeremy Huddleston Sequoia [Sun, 10 Nov 2013 19:26:49 +0000 (11:26 -0800)]
XQuartz: Update Info.plist copyright years
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Jeremy Huddleston Sequoia [Mon, 4 Nov 2013 20:13:45 +0000 (12:13 -0800)]
configure.ac: Move GLX section after DRI
The GLX section of configura.ac checks the state of DRI2, so it
needs to be after DRI2=auto is resolved.
Also reset libgl requirement to 7.1.0 in non-dri2 case.
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Andreas Schwab [Thu, 15 Aug 2013 10:03:42 +0000 (12:03 +0200)]
ARM64: Add support for aarch64
Signed-off-by: Andreas Schwab <schwab@suse.de>
Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Wed, 6 Nov 2013 02:39:11 +0000 (18:39 -0800)]
Use $GL_LIBS instead of -lGL for linking
-lGL presumes that the GL library is in the system path, while
$GL_LIBS is auto-detected.
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Wed, 6 Nov 2013 02:37:12 +0000 (18:37 -0800)]
Link with xshmfence, reference miSyncShmScreenInit in sdksyms
This gets the server to link with xshmfence again, and also ensures
that the miSyncShm code is linked into the server with the reference
from sdksyms.
Signed-off-by: Keith Packard <keithp@keithp.com>
Julien Cristau [Tue, 5 Nov 2013 15:08:21 +0000 (07:08 -0800)]
os: Actually use the computed clockid in GetTimeInMicros
The selection of which clock to use for this function was not actually
getting used when fetching the final clock value.
Reported-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Tue, 5 Nov 2013 03:01:26 +0000 (19:01 -0800)]
hw/xfree86: Link libdri3 only when DRI3 is defined
Don't attempt to link non-existant libraries...
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Sun, 3 Nov 2013 17:56:02 +0000 (09:56 -0800)]
Disable DRI3 and sync fence FD functions if xshmfence isn't available
Make sure the server can build when the xshmfence library isn't present
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Fri, 1 Nov 2013 08:01:58 +0000 (01:01 -0700)]
Set version to 1.14.99.901 (1.15 RC1)
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Fri, 1 Nov 2013 08:43:59 +0000 (01:43 -0700)]
Add dri3 and present directorys to distribution
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Fri, 1 Nov 2013 07:34:18 +0000 (00:34 -0700)]
Merge remote-tracking branch 'whot/for-keith'
Egbert Eich [Wed, 14 Aug 2013 16:18:56 +0000 (18:18 +0200)]
config/APM: Add option to disable building of APM support on Linux
APM support in the Xserver was used to restore the console mode
prior to a power management event. This was to ensure the mode
upon suspend/resume was one that the system firmware or kernel
could deal with.
APM support is now largely obsolete, KMS drivers don't require a
mode restoration anyhow. Therefore it should be possible to disable
this feature.
(small modification by keithp - move test for XF86PM flag after check
for APM, then move XF86PM flag to xorg-config.h.in)
Signed-off-by: Egbert Eich <eich@freedesktop.org>
Tested-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Peter Hutterer [Wed, 30 Oct 2013 05:40:58 +0000 (15:40 +1000)]
sync: fix corner-case in triggering idle alarms
ProcessInputEvent() resets the device idle times. If idle time was higher than
the lower bracket, this should trigger an event in the idle time wakeup
handler.
If processing is slow, the idle time may advance past the lower bracket
between the reset and the time the BlockHandler is called. In that case, we'd
never schedule a wakeup to handle the event, causing us to randomly miss
events.
Ran tests with a neg transition trigger on 5ms with 200 repeats of the test
and it succeeded. Anything below that gets a bit tricky to make sure the
server sees the same idle time as the client usleeps for.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Egbert Eich [Wed, 14 Aug 2013 16:18:39 +0000 (18:18 +0200)]
config/ACPI: Add option to disable building of ACPI support on Linux
ACPI support in the Xserver was used to restore the console mode
prior to a power management event. This was to ensure the mode
upon suspend/resume was one that the system firmware or kernel
could deal with.
The feature depended on acpid to be running. Most of this functionality
is now take over by systemd, KMS drivers don't require a mode restoration
anyhow. Therefore it should be possible to disable this feature under
some circumstances.
Tested-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Egbert Eich [Mon, 12 Aug 2013 16:21:33 +0000 (18:21 +0200)]
DDX/Randr: Avoid server crash when xrandr SetConfig is called while switched away
A call to Xrandr SetScreenConfig (for randr 1.1) causes the Xserver to
crash when xf86SetViewport() which does not check if the hardware is
accessible.
Wrap accesses to xf86SetViewport() with if (vtSema) { ... } to avoid that.
Signed-off-by: Egbert Eich <eich@freedesktop.org>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Egbert Eich [Mon, 12 Aug 2013 16:20:36 +0000 (18:20 +0200)]
DDX/Events: Distinguish between Input- and GeneralHandlers in xf86VTSwitch()
When enabling/disabling input handlers in xf86VTSwitch() we treat Input-
and GeneralHandlers equally. The result is that after a VT switch the
masks for EnabledDevices and AllSockets are equal and the distiction
between both types is lost.
Signed-off-by: Egbert Eich <eich@freedesktop.org>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
Egbert Eich [Sat, 5 Oct 2013 06:02:28 +0000 (08:02 +0200)]
DDX/modes: Add a sanity check when using screen sizes from EDID
EDID sometimes lies about screen sizes. Since the screen size is used
by clients to determine the DPI a wrong ration will lead to terrible
looking fonts.
Add a sanity check for the h/v ratio cutting off at 2.4. This would
still accept the cinemascope aspect ratio as valid.
Also add message suggesting to add a quirk table entry.
Signed-off-by: Egbert Eich <eich@freedesktop.org>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Egbert Eich [Wed, 14 Aug 2013 09:50:26 +0000 (11:50 +0200)]
DDX/DPMS: Call dixSaveScreens() also when screen is turned on
DMPS calls dixSaveScreens() when turned off but not when turned
on. In most cases this is irrelevant as DPMS is done when a
key is hit in which case dixSaveScreens() will be called to
unblank anyhow. This isn't the case if we use xset (or the
DPMS extension directly) to unblank.
Check screenIsSaved to make sure the state needs to be changed
before calling dixSaveScreens().
Signed-off-by: Egbert Eich <eich@freedesktop.org>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Søren Sandmann Pedersen [Mon, 21 Oct 2013 21:11:56 +0000 (17:11 -0400)]
ephyr: Ensure stride of private framebuffer is multiple of 4
The fb layer of X can't deal with strides that are not a multiple of
4, so when Xephyr allocates its own framebuffer it should make sure to
align it.
This fixes crashes and rendering corruption when Xephyr runs in a
depth that is different from the host X server and its screen size is
not a multiple of 4 / depth. (This is particularly easy to trigger if
you use the -resizeable option).
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Soren Sandmann <ssp@redhat.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Søren Sandmann Pedersen [Mon, 21 Oct 2013 20:58:54 +0000 (16:58 -0400)]
ephyr: hostx_screen_init(): Fix bits_per_pixel and bytes_per_line
When the depth of the Xephyr server matches that of the host X server,
Xephyr simply uses the buffer associated with the XImage as its
framebuffer. In this case, it is correct to get the bits_per_pixel and
bytes_per_line values returned from hostx_screen_init() from the XImage.
However, when the depth doesn't match the host, Xephyr uses a private
framebuffer that is periodically copied to the XImage. In this case,
the returned values of bits_per_pixel and bytes_per_line should be
those of the private framebuffer, not those of the XImage.
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Soren Sandmann <ssp@redhat.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Søren Sandmann Pedersen [Tue, 8 Oct 2013 21:45:40 +0000 (17:45 -0400)]
xf86AddBusDeviceToConfigure(): Store device in DevToConfig[i].pVideo
After
fc3ab84d the pVideo field in DevToConfig[i] is no longer
initialized, so it's always NULL. This causes the duplicate finding
algorithm in the beginning of the function to not work anymore as it
is based on this field.
The symptom of this bug is that X -configure reports
Number of created screens does not match number of detected devices.
Configuration failed.
Server terminated with error (2). Closing log file.
rather than producing a working config file.
This patch fixes that bug by initializing the field before calling
xf86PciConfigureNewDev().
Cc: tvignatti@gmail.com
Signed-off-by: Soren Sandmann <ssp@redhat.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Fri, 1 Nov 2013 01:18:19 +0000 (18:18 -0700)]
Merge remote-tracking branch 'jeremyhu/master'
Jeremy Huddleston Sequoia [Thu, 31 Oct 2013 15:57:56 +0000 (08:57 -0700)]
miext/damage: Partial revert of "Only wrap into the GC ops chain if there's a listener (v3)"
Fixes regression from:
4dc2a76740d921c824a4d8193f39dd373475f02a
http://lists.x.org/archives/xorg-devel/2013-May/036241.html
https://trac.macports.org/ticket/38993
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Fri, 1 Nov 2013 00:05:48 +0000 (17:05 -0700)]
Merge remote-tracking branch 'keithp/dri3'
Adam Jackson [Thu, 31 Oct 2013 22:39:22 +0000 (18:39 -0400)]
glx: Lie about GLX_Y_INVERTED_EXT
Well, that was lame. The problem with reporting y inversion honestly is
that libGL asks the driver _its_ opinion of Y inversion, which it just
fabricates from whole cloth. So then when libGL goes to compare the
driver's idea of fbconfigs with that of the server - a fairly dumb idea
to begin with - nothing matches, and direct rendering fails, and
sadness.
So until the DRI drivers are fixed we should just continue to lie about
Y inversion. GLX_DONT_CARE is what libGL would make up for that
attribute if we hadn't sent it, so just send that instead.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Tested-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Thu, 11 Jul 2013 23:11:57 +0000 (16:11 -0700)]
present: Add Present extension
Provides both a software implementation using timers and driver hooks
to base everything on vblank intervals.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Wed, 10 Apr 2013 02:59:39 +0000 (19:59 -0700)]
dri3: Add DRI3 extension
Adds DRM compatible fences using futexes.
Uses FD passing to get pixmaps from DRM applications.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Thu, 17 Jan 2013 21:46:55 +0000 (13:46 -0800)]
Add support for MIT-SHM AttachFd request
This passes a file descriptor from the client to the server, which is
then mmap'd
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Thu, 17 Jan 2013 21:43:02 +0000 (13:43 -0800)]
Add interfaces to get FDs from clients over the socket
This adds two interfaces:
void SetReqFds(ClientPtr client, int req_fds)
Marks the number of file descriptors expected for this
request. Call this before any request processing so that
any un-retrieved file descriptors will be closed
automatically.
int ReadFdFromClient(ClientPtr client)
Reads the next queued file descriptor from the connection. If
this request is not expecting any more file descriptors, or
if there are no more file descriptors available from the
connection, then this will return -1.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Thu, 31 Oct 2013 20:01:46 +0000 (13:01 -0700)]
misync: Don't smash custom screen sync functions
There was a check to avoid smashing custom functions, but the sense
was backwards causing it to always smash them, and also not set them otherwise.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Thu, 11 Jul 2013 23:10:34 +0000 (16:10 -0700)]
os: Add GetTimeInMicros
64-bit higher resolution current time value.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Thu, 11 Jul 2013 23:09:34 +0000 (16:09 -0700)]
Add swapll to byte swap 64-bit datatypes
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Thu, 11 Jul 2013 23:08:41 +0000 (16:08 -0700)]
Add a RegionDuplicate function
This allocates a new region structure and copies a source region into
it in a single API rather than forcing the caller to do both steps themselves.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Mon, 14 Jan 2013 22:24:36 +0000 (14:24 -0800)]
Clean up a couple of warnings in os/
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Keith Packard [Sat, 19 Jan 2013 05:43:40 +0000 (21:43 -0800)]
Xext: Move MIT-SHM 'ShmDesc' to shmint.h
This data structure is required to use shared memory objects in any
extension. That includes the Xv extension, which (before this patch)
duplicated the definition of this structure in its own code.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Peter Hutterer [Tue, 22 Oct 2013 04:24:52 +0000 (14:24 +1000)]
os: use a constant for backtrace array size
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
Peter Hutterer [Tue, 29 Oct 2013 05:24:11 +0000 (15:24 +1000)]
configure: remove a comment
94ed0ba1b5043ad9fc33b42756af447d5ab15bbd moved backtracing into the DIX, so
this comment is outdated. since no-one noticed and it's easier to just grep
than update file references, remove the comment.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
Jeremy Huddleston Sequoia [Thu, 31 Oct 2013 15:50:57 +0000 (08:50 -0700)]
XQuartz: Bump bundle version to 2.7.6
Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Adam Jackson [Thu, 31 Oct 2013 13:23:52 +0000 (09:23 -0400)]
dri: Fix GLX_Y_INVERTED_EXT fbconfig attribute
We're Y-inverted from GL's coordinates, so this is correct. gnome-shell
doesn't seem to check this - somewhat reasonable since the only server
that answered the other way around was Xglx - but kwin does, and
upside-down hilarity ensues.
Tested-by: maelcum on #xorg-devel
Signed-off-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Adam Jackson [Tue, 29 Oct 2013 16:58:14 +0000 (12:58 -0400)]
xfree86: Bump video/input/extension ABIs for 1.15
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Wed, 30 Oct 2013 15:54:24 +0000 (08:54 -0700)]
kdrive/ephyr: Don't discard one-time driver structure at server reset
KdScreenInfo is constructed at server startup time, and not
re-generated at server reset time. Freeing the 'driver' element at
reset time means this information is lost, and the server crashes
pretty quickly afterwards.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Michal Srb [Wed, 30 Oct 2013 11:33:51 +0000 (13:33 +0200)]
randr: deliver Output and Crtc events of attached output providers.
Consider all attached output providers when looking for changed outputs and
crtcs.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Michal Srb <msrb@suse.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Michal Srb [Wed, 30 Oct 2013 11:33:50 +0000 (13:33 +0200)]
randr: send RRResourceChangeNotify event
Send RRResourceChangeNotify event when provider, output or crtc was created or
destroyed. I.e. when the list of resources returned by RRGetScreenResources and
RRGetProviders changes.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Michal Srb <msrb@suse.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Michal Srb [Wed, 30 Oct 2013 11:33:49 +0000 (13:33 +0200)]
randr: send RRProviderChangeNotify event
Send RRProviderChangeNotify event when a provider becomes output source or
offload sink.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Michal Srb <msrb@suse.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Wed, 30 Oct 2013 04:00:13 +0000 (21:00 -0700)]
Merge remote-tracking branch 'whot/for-keith'
Gaetan Nadon [Mon, 28 Oct 2013 22:32:36 +0000 (18:32 -0400)]
config: replace deprecated use of AC_OUTPUT with AC_CONFIG_FILES
Fix Automake warning: AC_OUTPUT should be used without arguments.
www.gnu.org/software/autoconf/manual/autoconf.html#Configuration-Files
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Peter Hutterer [Wed, 23 Oct 2013 05:16:25 +0000 (15:16 +1000)]
dmx: queue button events with a flags of 0
Setting POINTER_SCREEN with a unset valuator mask causes a jump to 0/0. Set
the flags to 0 so we don't generate any motion on a button event.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Connor Behan [Fri, 18 Oct 2013 01:26:28 +0000 (18:26 -0700)]
xfree86: Find primary entity when bus types are nominally different
As of server 1.13, systems with DRM and Udev will have BUS_PLATFORM as
their primary bus type. However, drivers not implementing a
platformProbe function will still create entities of type BUS_PCI. We
need to account for this when checking for the primary entity.
Signed-off-by: Connor Behan <connor.behan@gmail.com>
Acked-by: Tormod Volden <debian.tormod@gmail.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Tue, 29 Oct 2013 16:37:30 +0000 (09:37 -0700)]
Merge remote-tracking branch 'idr/glx-float-fbconfig'
Adam Jackson [Wed, 10 Jul 2013 14:00:46 +0000 (10:00 -0400)]
glx: convert to direct GL dispatch (v2)
We now expect to be linked against something that provides the GL API,
instead of manually grubbing about in the DRI driver's dispatch table.
Since the GLX we expose calls GL functions that are meant to be looked
up dynamically, also add a way to thunk through to GetProcAddress.
This includes a refresh of the generated sources, which requires a
correspondingly new Mesa.
The GetProcAddress stubs are at the moment merely enough to make this
link against Mesa 9.2, but should really be provided for everything not
in the OpenGL 1.2 ABI.
v2: Explicitly hide the GetProcAddress stubs so we can't conflict with
libGL symbols; fix leading tab/space issues [anholt]
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Adam Jackson [Fri, 4 Oct 2013 16:58:19 +0000 (12:58 -0400)]
glx: Remove DRI1 AIGLX (v2)
Mesa doesn't ship DRI1 drivers as of 8.0, which is about 18 months and
three releases ago. The main reason to have wanted DRI1 AIGLX was to
get a GLX compositor working, but DRI1's (lack of) memory management API
meant that the cost of a GLX compositor was breaking direct GLX apps,
which isn't a great tradeoff.
Of the DRI1 drivers Mesa has dropped, I believe only mga stands to lose
some functionality here, since it and only it has support for
NV_texture_rectangle. Since that's required for every extant GLX
compositor I know of, I conclude that anybody with a savage, say, would
probably not notice AIGLX going away, since they wouldn't be running a
GLX compositor in the first place.
In the future we'd like to use GL in the server in a more natural way,
as just another EGL client, including in the GLX implementation itself.
Since there's no EGL implemented for DRI1 drivers, this would already
doom AIGLX on DRI1 (short of entirely forking the GLX implementation,
which I'm not enthusiastic about).
v2: Remove DRI1 from AIGLX conditionals in configure.ac [anholt]
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Adam Jackson [Wed, 2 Oct 2013 21:38:05 +0000 (17:38 -0400)]
glx: Simplify glXDestroyContext
We can just free the resource unconditionally here. ContextGone (which
FreeResourceByType will call) already does:
cx->idExists = GL_FALSE;
if (!cx->currentClient) {
__glXFreeContext(cx);
}
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Adam Jackson [Mon, 5 Aug 2013 19:46:11 +0000 (15:46 -0400)]
glx: Fix a corner case in DrawableGone
Losing the drawable does not change our notion of current client. Since
the GL under us doesn't understand having a current context without
current drawables (sigh), we do still need to loseCurrent so that we
re-bind the context on the next request.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=65030
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Adam Jackson [Sat, 3 Aug 2013 13:47:55 +0000 (09:47 -0400)]
glx: Fix memory leak in context garbage collection (v2)
I broke this, back in:
commit
a48dadc98a28c969741979b70b7a639f24f4cbbd
Author: Adam Jackson <ajax@redhat.com>
Date: Mon Mar 21 11:59:29 2011 -0400
glx: Reimplement context tags
In that, I changed the glx client state to not explicitly track the list
of current contexts for the client (since that was what we were deriving
tags from). The bug was that I removed the code for same from
glxClientCallback without noticing that it had the side effect of
effectively de-currenting those contexts, so that ContextGone could free
them. So, if you had a client exit with a context still current, the
context's memory would leak. Not a huge deal for direct clients, but
viciously bad for indirect, since the swrast context state at the bottom
of Mesa is like 15M.
Fix this by promoting Bool isCurrent to ClientPtr currentClient, so that
we have a back-pointer to chase when walking the list of contexts when
ClientStateGone happens.
v2: Explicitly call __glXFreeContext on the ClientStateGone path. Our
current context might be one we got from EXT_import_context and whose
creating client has since died. Without the explicit call, the creating
client's FreeClientResources would not free the context because it's
still current, and the using client's FreeClientResources would not free
the context because it's not an XID it created. This matches the logic
from a48dadc.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Adam Jackson [Tue, 8 Oct 2013 19:02:44 +0000 (15:02 -0400)]
glx: Only take a Pixmap reference if creating the GLXPixmap succeeded
Reviewed-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Peter Hutterer [Mon, 28 Oct 2013 02:10:10 +0000 (12:10 +1000)]
Xi: fix logic error when calculating emulated motion events
gcc -Wlogical-op
exevents.c: In function 'DeliverEmulatedMotionEvent':
exevents.c:1480:13: warning: logical 'or' of collectively exhaustive
tests is always true [-Wlogical-op]
The relevant snippet of exevents.c:
1479 if (ti->listeners[0].type != LISTENER_POINTER_REGULAR ||
1480 ti->listeners[0].type != LISTENER_POINTER_GRAB)
1481 return;
This condition was always true, causing dropped motion events.
Reported-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Daniel Czarnowski [Mon, 21 Oct 2013 17:48:38 +0000 (19:48 +0200)]
glx: Enable GLX_ARB_fbconfig_float for DRI2 drivers
v2 (idr): Maintain sort order of extension list. Fixed rebase error
(comment header for initializeExtensions was in front of
__glXDRIinvalidateBuffers). Re-write commit message.
Signed-off-by: Daniel Czarnowski <daniel.czarnowski@intel.com>
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Daniel Czarnowski [Wed, 23 Oct 2013 10:53:59 +0000 (12:53 +0200)]
glx: Handle float config types in glxConvertConfigs
Replaces old use of floatMode attribute with new, extended range of
values in __DRI_ATTRIB_RENDER_TYPE. Also adds new conditions, where the
float modes support requires it. Enables support for not only float
configs, but packed float configs as well.
v2 (idr): Whitespace and formatting fixes. Refactor render type
vs. pbuffer checking to a separate function that includes a quote from
the spec. Re-write commit message. Fix compiler warnings:
glxdricommon.c: In function 'glxConvertConfigs':
glxdricommon.c:212:35: warning: pointer targets in passing argument 3 of 'core->getConfigAttrib' differ in signedness [-Wpointer-sign]
glxdricommon.c:212:35: note: expected 'unsigned int *' but argument is of type 'int *'
glxdricommon.c:230:35: warning: pointer targets in passing argument 3 of 'core->getConfigAttrib' differ in signedness [-Wpointer-sign]
glxdricommon.c:230:35: note: expected 'unsigned int *' but argument is of type 'int *'
Signed-off-by: Daniel Czarnowski <daniel.czarnowski@intel.com>
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Daniel Czarnowski [Mon, 21 Oct 2013 17:57:53 +0000 (19:57 +0200)]
glx: Correctly set render type enum
The type specified is bits, but the context wants an enum.
v2 (idr): Fix various whitespace and formatting problems. Rename
convFBconfRenderTypeBits2CtxRenderType to
renderTypeBitsToRenderTypeEnum. Re-write commit message.
Signed-off-by: Daniel Czarnowski <daniel.czarnowski@intel.com>
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Daniel Czarnowski [Mon, 21 Oct 2013 17:45:35 +0000 (19:45 +0200)]
glx: Allow float renderType in glXCreateContextAttribsARB
This enables GLX_RGBA_FLOAT_TYPE_ARB and
GLX_RGBA_UNSIGNED_FLOAT_TYPE_EXT as valid renderType parameters to
glXCreateContextAttribsARB.
v2 (idr) : Use C-style comments and fix word wrapping. Re-write commit
message.
Signed-off-by: Daniel Czarnowski <daniel.czarnowski@intel.com>
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Eric Anholt [Tue, 22 Oct 2013 21:22:04 +0000 (14:22 -0700)]
glx: Add support for the new DRI loader entrypoint.
This is going to be exposed (and not the old entrypoint) for some DRI
drivers once the megadrivers series lands, and the plan is to
eventually transition all drivers to that. Hopefully this is
unobtrusive enough to merge to stable X servers so that they can be
compatible with new Mesa versions.
v2: typo fix in the comment
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Strake [Fri, 18 Oct 2013 04:20:20 +0000 (23:20 -0500)]
unbreak KdEnableInput
pi = ki = 0 here, so not follow them.
Signed-off-by: Strake <strake888@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Keith Packard [Sat, 19 Oct 2013 00:31:01 +0000 (17:31 -0700)]
Update to version 1.14.99.3
Signed-off-by: Keith Packard <keithp@keithp.com>
Keith Packard [Sat, 19 Oct 2013 00:22:52 +0000 (17:22 -0700)]
Merge commit '
2523a445a09a75a8baf642608d099a5e12d5437f'
This is merging the commits along Peter's for-keith branch, without
also merging in the spurious merge at the top of that branch.
Peter Hutterer [Fri, 18 Oct 2013 06:50:12 +0000 (16:50 +1000)]
Merge branch 'xsync-transition-bug-59644' into for-keith
Peter Hutterer [Thu, 17 Oct 2013 02:02:27 +0000 (12:02 +1000)]
sync: split updating and triggering a counter up
Now that the brackets are always the nearest change points (regardless of
transition) we need to update the counters whenever we check for any updates.
Otherwise we end up with a situation where counter->value is out of date and
an alarm doesn't trigger because we're still using the value from last time
something actually triggered.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Peter Hutterer [Wed, 16 Oct 2013 06:31:15 +0000 (16:31 +1000)]
sync: always set the brackets (#59644)
The current code sets bracket_greater to the first trigger after the current
value, and bracket_less to the last trigger before the current value.
For example, the idle timer with three negative and three positive transitions
would set this:
nt1 nt2 nt3
|--------|------|--|------- idle --|---|--|-----> t
pt1 pt2 pt3
bracket_less == nt2
bracket_greater == pt2
This is an optimization so we can skip code paths in the block/wakeup handlers
if the current value doesn't meet any of the trigger requirements. Those
handlers largely do a
if (bracket_less is less than current value &&
bracket_greater is greater than current value)
return, nothing to do
However, unless the bracket values are updated at the correct time, the
following may happen:
nt
|--------------|---------- idle ------|--------> t
pt
In this case, neither bracket is set, we're past the pos transition and not
yet at the neg transition. idle may now go past nt, but the brackets are not
updated. If idle is then reset to 0, no alarm is triggered for nt. Likewise,
idle may now go past pt and no alarm is triggered.
Changing an alarm or triggering an alarm will re-calculate the brackets, so
this bug is somewhat random. If any other client triggers an alarm when the
brackets are wrongly NULL, the recalculation will set them this bug may not
appear.
This patch changes the behavior, so that the brackets are always the nearest
positive or negative transitions to the current counter value. In the example
above, nt will trigger a wakeup and a re-calculation of the brackets, so that
going past it in the negative direction will then cause the proper alarm
triggers.
Or, in Keith's words:
Timer currently past a positive trigger
No bracket values, because no trigger in range
Timer moves backwards before the positive trigger
Brackets not reset, even though there is now a trigger in range
Timer moves forward past the positive trigger
Trigger doesn't fire because brackets not set
Setting the LT bracket in this case will cause everything to get
re-evaluated when the sync value moves backwards before the trigger
value.
X.Org Bug 59644 <http://bugs.freedesktop.org/show_bug.cgi?id=59644>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Peter Hutterer [Wed, 16 Oct 2013 00:08:46 +0000 (10:08 +1000)]
sync: if the idle time was reset, force alarms to trigger (#70476)
The time between the idle reset and the IdleTimeWakeupHandler to be called is
indeterminate. Clients with an PositiveTransition or NegativeTransition alarm
on a low threshold may miss an alarm.
Work around this by keeping a reset flag for each device. When the
WakeupHandler triggers and the reset flag is set, we force a re-calculation of
everything and pretend the current idle time is zero. Immediately after is the
next calculation with the real idle time.
Relatively reproducible test case: Set up a XSyncNegativeTransition alarm for
a threshold of 1 ms. May trigger, may not.
X.Org Bug 70476 <http://bugs.freedesktop.org/show_bug.cgi?id=70476>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>