There were two memory leaks in the function: one was the lack of free
for "enabled", the other was the full lack of releasing anything when
configuration was too small. The first issue was fixed by adding the
missing free, the other was addressed by replacing the duplicate
memory releasing sequences with one that is gotoed into.
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit d3adf2d935)
v2: Slightly more obvious sizing math.
==14882== Invalid write of size 2
==14882== at 0x6750267: VBEGetVBEInfo (vbe.c:400)
==14882== by 0x6142064: ??? (in /usr/lib64/xorg/modules/drivers/vesa_drv.so)
==14882== by 0x471895: InitOutput (xf86Init.c:519)
==14882== by 0x422778: main (main.c:205)
==14882== Address 0x4f32fa8 is 72 bytes inside a block of size 73 alloc'd
==14882== at 0x4A0640D: malloc (vg_replace_malloc.c:236)
==14882== by 0x675024B: VBEGetVBEInfo (vbe.c:398)
==14882== by 0x6142064: ??? (in /usr/lib64/xorg/modules/drivers/vesa_drv.so)
==14882== by 0x471895: InitOutput (xf86Init.c:519)
==14882== by 0x422778: main (main.c:205)
Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
(cherry picked from commit d8caa78200)
Reviewed-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Julien Cristau <jcristau@debian.org>
(cherry picked from commit 3f0d3f4d97)
The EDID processing regards physical dimensions of 0mm x 0mm as
invalid. Previously the old values for height and width would be
preserved if none of the physical dimension specifications in the new
EDID were considered valid.
This will come up in particular if first a monitor is connected to an
output, and then a projector is connected. Since projectors generally
report physical dimensions of 0mm x 0mm, this would result in the
projector claiming to have the physical dimensions of the monitor.
Signed-off-by: Evan Broder <ebroder@mokafive.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit 12b0f7df2c)
Inferring modes from sync ranges is only valid if the monitor says it's
valid. If the monitor says it's valid, then we'll have already added
those modes during EDID block parse. If it doesn't, then we should
believe it.
If there's no EDID for an output, but sync ranges from the config, we'll
still add default modes as normal.
Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
(cherry picked from commit dc498b433f)
We now support using RandR to set the resolution of the primary display (and
place a shielding window on other displays) in multi-monitor configurations.
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
(cherry picked from commit 8cf3348e90)
This will prevent native windows from resizing as we change resolutions.
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
(cherry picked from commit 13578b852b)
If there's a GPU copy and a non-zero devKind was passed in, set the GPU copy
pitch to that instead of to a possibly bogus value derived from the new width.
This is e.g. used by the radeon driver's drmmode_xf86crtc_resize hook, fixes
https://bugs.freedesktop.org/show_bug.cgi?id=33929 .
On the other hand, the system memory copy doesn't need the pitch to be aligned
beyond the PixmapBytePad of the width.
Signed-off-by: Michel Dänzer <daenzer@vmware.com>
Acked-by: Cyril Brulebois <kibi@debian.org>
Tested-by: Cyril Brulebois <kibi@debian.org>
Reported-by: Thierry Vignaud <thierry.vignaud@gmail.com>
Tested-by: Thierry Vignaud <thierry.vignaud@gmail.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit 31704510f4)
RecordFlushReplyBuffer can call itself recursively through
WriteClient->CallCallbacks->_CallCallbacks->RecordFlushAllContexts
when the recording client's buffer cannot be completely emptied in one
WriteClient. When a such a recursion occurs, it will not be broken out
of which results in segmentation fault when the stack is exhausted.
This patch adds a counter (a flag, really) that guards against this
situation, to break out of the recursion.
One alternative to this change would be to change _CallCallbacks to
check the corresponding counter before the callback loop, but that
might affect existing behavior, which may be relied upon.
Reviewed-by: Rami Ylimäki <rami.ylimaki@vincit.fi>
Signed-off-by: Erkki Seppälä <erkki.seppala@vincit.fi>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit 0801afbd7c)
M_DRAWABLE_PIXMAP is the lookup mask to dixLookupDrawable, and _not_ the
type value in the drawable itself.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Eamon Walsh <ewalsh@tycho.nsa.gov>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit ac0a00a840)
We get an XP_EVENT_DISPLAY_CHANGED event when our display configuration is
changed. If this change was caused by hotplugging a monitor or Mac Display
Preferences changes by the user, we need to call RRScreenSizeNotify in order
to ensure new connections get the correct screen size.
http://xquartz.macosforge.org/trac/ticket/460
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
(cherry picked from commit 418bb57a39)
Button events may be sent with no valuators (e.g. to simply indicate
ButtonPress or ButtonRelease without any coordinates); when this happens
the server would read uninitialized memory.
==9999== Conditional jump or move depends on uninitialised value(s)
==9999== at 0x48E87E8: pixman_f_transform_point (in /usr/lib/libpixman-1.so.0.18.2)
==9999== Uninitialised value was created by a stack allocation
==9999== at 0x37524: GetPointerEvents (getevents.c:1074)
==9999==
==9999== Conditional jump or move depends on uninitialised value(s)
==9999== at 0x496D074: lround (s_lround.c:40)
==9999== by 0x3773B: GetPointerEvents (getevents.c:1048)
==9999== by 0x683BB: xf86PostButtonEventP (xf86Xinput.c:1162)
==9999== by 0x6853B: xf86PostButtonEvent (xf86Xinput.c:1126)
==9999== by 0x5779037: process_state (multitouch.c:321) (xf86-input-mtev)
==9999== by 0x577908F: read_input (multitouch.c:331)) (xf86-input-mtev)
==9999== by 0x66B4F: xf86SigioReadInput (xf86Events.c:298)
==9999== by 0x112697: xf86SIGIO (sigio.c:118)
==9999== by 0x4A12B2F: ??? (sigrestorer.S:51)
==9999== Uninitialised value was created by a stack allocation
==9999== at 0x37524: GetPointerEvents (getevents.c:1074)
Signed-off-by: Oliver McFadden <oliver.mcfadden@nokia.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
(cherry picked from commit bf48082f30)
RemoveBlockAndWakeupHandlers requires caller to pass same block data
parameter as for RegisterBlockAndWakeupHandlers.
Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
(cherry picked from commit 3e1455505a)
cc2c73ddcb4370a7c3ad439cda4da825156c26c9's three-cent titanium tax
doesn't go too far enough. Fix the rest of the call and jmp
instructions to handle the data prefix correctly.
Reference: Intel 64 and IA-32 Architectures Software Developer's Manual
Volume 2A: Instruction Set Reference, A-M
http://www.intel.com/Assets/PDF/manual/253666.pdf
Reviewed-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Adam Jackson <ajax@redhat.com>
(cherry picked from commit bb18f27715)
Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
(cherry picked from commit 617b7d2211)
In animate cursor block handler code assumes GetTimeInMillis returns
always nonzero value. This isn't true when time wraps around.
To prevent any problems in case GetTimeInMillis would return zero use
activeDevice variable to track if we have received time.
Signed-off-by: Pauli Nieminen <ext-pauli.nieminen@nokia.com>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
(cherry picked from commit aa8cea953d)
Assume that a mode can be used in either landscape or portrait
orientation. I suppose the correct thing to do would be to
collect all the supported rotations from the CRTCs that can be used
with a specific output, but that information doesn't seem to be
readily available when these checks are done. So just assume that
either orientation is fine.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
(cherry picked from commit 2e781457d4)
If the drawable size doesn't match the pixmap size page flipping should
not be allowed.
If the window is larger than the pixmap, page flipping might need to
reposition the CRTC somewhere in the middle of the pixmap. I didn't
spot any code that would handle that at least in the intel driver.
Also the root pixmap could then move to some negative screen
coordinates. Not sure if all bits of code could handle that. Perhaps
when composite is enabled screen_x/y would make it work, but without
composite there's no way that it would work AFAICS.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
(cherry picked from commit 0ce25fd790)
On some systems, using CLOCK_MONOTONIC forces a readback of HPET or some
similarly expensive timer. CLOCK_MONOTONIC_COARSE can alleviate this,
at the cost of negligibly-reduced resolution, so prefer that where we
can.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Julien Cristau <jcristau@debian.org>
Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
(cherry picked from commit 44adb31bfe)
Return a error if the screen is configured to an invalid size.
Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit d1107918d4)
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=32803 .
Signed-off-by: Michel Dänzer <daenzer@vmware.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit 6358a60065)
Most extensions have a version defined
in the protocol headers, and also in the
server's protocol-versions.h. The latter
defines which version the server advertises
support for. Sync wasn't included in
protocol-versions.h, and was advertising
support for whatever was in the protocol
headers the server was built against.
Signed-off-by: James Jones <jajones@nvidia.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit 27593eea7e)
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=24703 .
Signed-off-by: Michel Dänzer <daenzer@vmware.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit e06fa80400)
Like some other LPL panels, this one reports the vertical size in cm rather
than mm.
Patch taken from Launchpad bug #380009 <https://launchpad.net/bugs/380009>
X.Org Bug 28414 <https://bugs.freedesktop.org/show_bug.cgi?id=28414>
Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit 4b88c7be8d)
Ensure that if we're called exactly on the threshold of a
NegativeTransition trigger that we reshedule to pick up
an idle time over the threshold.
Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit a2e67a6412)
The {Positive,Negative}Transition triggers only fire when the counter
goes from strictly {below,above} the threshold. If
SyncComputeBracketValues gets called exactly at this threshold we may update
the bracket values so that the counter is not updated past the threshold.
Signed-off-by: Christopher James Halse Rogers <christopher.halse.rogers@canonical.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
(cherry picked from commit b55bf24858)
Byte padding and conversion is interesting for the rage of 0-8 bytes, and
then interesting towards the end of the valid range (INT_MAX - 7 and INT_MAX
- 3).
Note: this changes the upper range for pad_to_int32() and bytes_to_int32()
from the previous (INT_MAX - 4) to (INT_MAX - 3).
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Julien Cristau <jcristau@debian.org>
(cherry picked from commit d435e1ecb8)
We calculate the expected bytes for each value, let's use it.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Julien Cristau <jcristau@debian.org>
(cherry picked from commit f49ee9074a)
It is currently assumed that an event button delieved to a master device
corresponds to the slave button states. However, the event button is a
logical (mapped) slave button and slave button states correspond to
physical (unmapped) slave buttons. This leads to incorrect update of the
master button state and incorrect events devlivered to clients. Fix the
situation by taking the slave button map into account when querying a
slave button state.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=24887
Signed-off-by: Eoghan Sherry <ejsherry@gmail.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
(cherry picked from commit 36b614dedf)
