This batch is the straightforward set - others are more complex and
need more analysis to determine right size to pass.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
memType is a uint64_t on powerpc. Using memType only really makes
sense for *physical* addresses, which can be 64-bit for 32-bit
systems running on 64-bit hardware.
However, unmapVidMem() only deals with *virtual* addresses, which
are guaranteed to fit into an uintptr_t.
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
Also remove odd definition MAP_FAILED.
Signed-off-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Bug introduced by 9dca441670
xfree86: add a hook to replace the new console handler.
console_handler was not being set, making the server eat up CPU spinning
in WaitForSomething selecting consoleFd over and over again, every time
trying to unregister drain_console without success due to
console_handler being NULL.
Let's just fix the unregistration in xf86SetConsoleHandler() and use that.
But wait, there could be a catch: If some driver replaced the handler using
xf86SetConsoleHandler(), the unregistration in xf86CloseConsole will unregister
that one. I don't understand Xorg well enough to know whether this poses a
problem (could mess up driver deinit somehow or something like that). As it is,
xf86SetConsoleHandler() doesn't offer any way to prevent this (i.e. check which
handler is currently registered).
I had been using it for two days on my machine that previously hit 100% CPU
several times a day. That has now gone away without any new problems appearing.
Signed-off-by: Tomas Trnka <tomastrnka@gmx.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
People that don't want VGA arbiter active can go to the library and enable the
stubs there.
Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com>
Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
Reviewed-by: Jamey Sharp <jamey@minilop.net>
pciaccess handles this now.
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
Tested-by: Jeremy Huddleston <jeremyhu@apple.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Jamey Sharp <jamey@minilop.net>
Stop duplicating in each os-support variant before it gets replicated
even further.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Julien Cristau <jcristau@debian.org>
Reviewed-by: Alexandr Shadchin <Alexandr.Shadchin@gmail.com>
When the smart scheduler is enabled, the VT ioctls (particularly
VT_WAITACTIVE) can be interrupted by the smart scheduler's SIGALRMs.
Previously, this caused the server to immediately continue on to
ScreenInit, almost certainly causing a crash or failure because the X
server that owned the VT hadn't finished cleaning up. As of commit
7ee965a300, it causes a FatalError
instead.
Retrying the ioctl as long as it fails with errno == EINTR fixes the
problem and allows server regenerations to trigger VT switches that
actually succeed.
Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Cyril Brulebois <kibi@debian.org>
Signed-off-by: Keith Packard <keithp@keithp.com>
Cast it to a char *, mimicking the return immediately below it.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
lnx_agp.c: In function ‘xf86DeallocateGARTMemory’:
lnx_agp.c:267: warning: cast to pointer from integer of different size
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
This makes more things fatal than were fatal before, but that's correct;
if you need the VT, then failing to get it on regeneration means things
are about to go very very badly.
Reviewed-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Adam Jackson <ajax@redhat.com>
If the vt gets a vhangup from under us, then the tty will appear ready
in select(), but trying to tcflush() it will return -EIO, so we'll spin
around at 100% CPU for no reason. Notice this condition and unregister
the handler if it happens.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
Reviewed-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Disable timer/keyboard trapping on GNU/Hurd for now
Trapping disabled for now, as some VBIOSes (mga-g450 notably) use these
ports, and the int10 wrapper is not emulating them.
It's effectively what happens in the Linux variant too, as iopl() is used there,
making the ioperm() meaningless.
Signed-off-by: Olaf Buddenhagen <antrik@users.sf.net>
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
This hook is only necessary for the keyboard driver to remove the race
condition between drain_console() and the driver's ReadInput (Bug 29969).
The idea is that a driver that needs to handle events from the console
calls xf86ReplaceConsoleHandler() with it's own ReadInput (or NULL) and thus
removes the drain_console call. It's the driver's responsibility to restore
the previous behaviour when the driver is unloaded.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
CC: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
The drain_console() function will race with new keyboard events being added
by the hardware causing the server to lose keyboard events if the console fd
is used for input.
Only use the drain_console() when AllowEmptyInput is off which is the best
indicator we have for whether the keyboard driver will be used. This patch
will only fix the bug when hotplugging is disabled.
What we really need is a way to figure out either whether we're _not_ using
the keyboard driver (not predictable) or a way for the keyboard driver to
disable drain_console().
X.Org Bug 29969 <http://bugs.freedesktop.org/show_bug.cgi?id=29969>
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
(cherry picked from commit 71972c2534)
The drain_console() function will race with new keyboard events being added
by the hardware causing the server to lose keyboard events if the console fd
is used for input.
Only use the drain_console() when AllowEmptyInput is off which is the best
indicator we have for whether the keyboard driver will be used. This patch
will only fix the bug when hotplugging is disabled.
What we really need is a way to figure out either whether we're _not_ using
the keyboard driver (not predictable) or a way for the keyboard driver to
disable drain_console().
X.Org Bug 29969 <http://bugs.freedesktop.org/show_bug.cgi?id=29969>
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
This patch was generated by the following Perl code:
perl -i -pe 's/([^_])return\s*\(\s*([^(]+?)\s*\)s*;(\s+(\n))?/$1return $2;$4/g;'
Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
Reviewed-by: Jamey Sharp <jamey@minilop.net>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Keith Packard <keithp@keithp.com>
The only remaining X-functions used in server are XNF*, the rest is converted to
plain alloc/calloc/realloc/free/strdup.
X* functions are still exported from server and x* macros are still defined in
header file, so both ABI and API are not affected by this change.
Signed-off-by: Mikhail Gusarov <dottedmag@dottedmag.net>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
Let's let glibc do the right thing for dense/sparse selection.
The _alpha_iobase code has been unused since the switch to libpciaccess. It
really should have been killed by fba700f1f6a8976.
Signed-off-by: Matt Turner <mattst88@gmail.com>
Tested-by: Michael Cree <mcree@orcon.net.nz>
Signed-off-by: Keith Packard <keithp@keithp.com>
I couldn't find any version of the X xserver that ever used lnx_font.c
so let's delete it. I tried contacting its author, Egbert, multiple
times on IRC and email [*] but never got any response. It also hasn't
been seriously touched since January 2005.
[*] http://lists.x.org/archives/xorg-devel/2009-October/002855.html
Signed-off-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
* mattst88/master:
[alpha] assume we have __NR_pciconfig_iobase
[alpha] don't return from void functions
Fix undefined symbols on alpha
Fix breakage on alpha caused by c7680befe5
Revert "alpha: kill xf86SlowBCopyToBus and xf86SlowBCopyFromBus"
The code path if we didn't have support has been broken since before we
switched to git.
The pciconfig_iobase syscall has been supported since 2000.
Signed-off-by: Matt Turner <mattst88@gmail.com>
When testing if an fd is valid, the required construct is >= 0, not > 0.
[Daniel: Fixed up the Linux MTRR case as well.]
Signed-off-by: Martin Ettl <ettl.martin@gmx.de>
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
Move misplaced } to get the flow of
if (!ShareVTs) {
VT_ACTIVATE
VT_WAITACTIVE
}
X.Org Bug 11477 <http://bugs.freedesktop.org/show_bug.cgi?id=11477>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Acked-by: Daniel Stone <daniel@fooishbar.org>
Leave consoleFd open over the course of the server, even though any use
of it in this context is likely to be disastrous.
Signed-off-by: Michael Witrant <mike@lepton.fr>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
In non-setuid root installations, we shouldn't try to adjust VT/tty
ownership. It will fail, and shouldn't be necessary anyway (since
startup scripts or PAM should be handling perms for us in that case).
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This changes the ABI, but since the video ABI is at 6 already
it should be fine.
driver changes are in the pipeline after this.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Otherwise APM events get treated as input events, which messes up idle
time accounting and screensavers and such. Not, we hope, that anyone
is using APM anymore.
This was all a glorified no-op. We rely on pciaccess to create device
maps anyway, so we should have no reason to care about what the host
address is.
Acked-by: Ian Romanick <ian.d.romanick at intel.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
The previous check for AEI on left us with the possibility that AEI is forced
off in the config, but devices are added through evdev nonetheless. A keyboard
added this way can CTRL+C the server. Even when we use kbd, we can set the
mode to RAW, so it's safer alround to to so.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Restoring it unconditionally means we restore to whatever tty_mode has as
default value (i.e. 0). K_RAW happens to be 0x00, so we always restore to raw
mode if allowEmptyInput is off.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Julien Cristau <jcristau@debian.org>
Save in a few special cases, _X_EXPORT should not be used in C source
files. Instead, it should be used in headers, and the proper C source
include that header. Some special cases are symbols that need to be
shared between modules, but not expected to be used by external drivers,
and symbols that are accessible via LoaderSymbol/dlopen.
This patch also adds conditionally some new sdk header files, depending
on extensions enabled. These files were added to match pattern for
other extensions/modules, that is, have the headers "deciding" symbol
visibility in the sdk. These headers are:
o Xext/panoramiXsrv.h, Xext/panoramiX.h
o fbpict.h (unconditionally)
o vidmodeproc.h
o mioverlay.h (unconditionally, used only by xaa)
o xfixes.h (unconditionally, symbols required by dri2)
LoaderSymbol and similar functions now don't have different prototypes,
in loaderProcs.h and xf86Module.h, so that both headers can be included,
without the need of defining IN_LOADER.
xf86NewInputDevice() device prototype readded to xf86Xinput.h, but
not exported (and with a comment about it).
The warnings corrected were only the ones that should correct
real problems. The most common one is 64 bit integers as
"printf %l" arguments.
Note that there is a patch related to this at:
http://bugs.freedesktop.org/show_bug.cgi?id=18204
Alan Coopersmith
Jeremy Huddleston
Alexandr Shadchin
Keith Packard
Tomáš Trnka
Tiago Vignatti
Adam Jackson
Aaron Plattner
Daniel Stone
Peter Hutterer
Samuel Thibault
Macpaul Lin
Thomas Hellstrom
Mikhail Gusarov
Matt Turner
Martin Ettl
Michael Cree
Michael Witrant
Jesse Barnes
Dave Airlie
Julien Cristau
Paulo Cesar Pereira de Andrade