frederik
/
xserver
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
4,982 Commits 293 Branches 13 Tags 61 MiB
Commit Graph

1611 Commits

Author SHA1 Message Date
Adam Jackson 3b73d62791 xf86DDCMonitorSet: Honor the DisplaySize from the config file. 
We honor sync ranges and pixel clock settings from the config here, no
reason to ignore DisplaySize.
 2008-03-03 15:43:22 -05:00
David Nusinow f7ab2d3821 Add missing file from previous commit. 2008-03-02 18:36:25 -05:00
David Nusinow b5ce0e1d0b Bug #13860: Ensure that the DRI mode is in octal format. 2008-03-02 17:12:02 -05:00
Matthieu Herrb c934366424 use UTILS_SYS_LIBS to pass SYS_LIBS to utils/ioports correctly 2008-03-02 19:32:01 +01:00
George Sapountzis 3d64290547 clean some "unused" warnings 2008-03-01 19:49:24 +02:00
Maarten Maathuis 8af2c39bcc Fix big mistake in commit fd41f46ac6. 
- When a mode is deleted, the name pointer is also free()'ed.
- This leaves other modes with an invalid pointer.
 2008-03-01 16:54:01 +01:00
Eamon Walsh ef60632e20 dix: Modify callers of property and selection API to use new interfaces. 2008-02-29 18:01:37 -05:00
Eamon Walsh 34bf308a9e dix: Refactoring of selection code to allow for polyinstantiation. 
Introduces dixLookupSelection() API.
Removes NumCurrentSelections from API.
 2008-02-29 18:01:37 -05:00
Eamon Walsh d5715f7bea dix: Refactoring of property code to allow for polyinstantiation. 
Introduces dixLookupProperty() API.
 2008-02-29 18:01:37 -05:00
Adam Jackson b7eb92774a EDID 1.4: Print additional CVT support data in the log. 2008-02-29 16:01:15 -05:00
Adam Jackson 31014d88af EDID 1.4: Decode additional CVT support information. 
Table 3.28: Display Range Limits & CVT Support Definition
 2008-02-29 16:01:15 -05:00
Adam Jackson d1c48955f8 Fix CVT abuse in DDCModesFromStandardTiming. 
CVT is enough different from GTF that it should not be used on monitors
that aren't expecting it.  This brings us closer to what the spec says
the correct behaviour is.
 2008-02-29 16:01:15 -05:00
Adam Jackson 26c2e95fa5 Nuke a duplicate SYMFUNC(xf86CVTMode) 2008-02-29 16:01:15 -05:00
Adam Jackson ca5625b911 Add xf86GTFMode(). 
This should probably be shared like xf86CVTMode().
 2008-02-29 16:01:15 -05:00
Adam Jackson 4cb4817c10 Remove the duplicate copy of xf86cvt.c 2008-02-29 16:01:15 -05:00
Adam Jackson e65e51a99b Constify the built-in mode tables. 2008-02-29 16:01:14 -05:00
Adam Jackson 6828d8fc2b Clean up DisplayModeRec handling in many places. 
Use xf86DuplicateMode() instead of rolling our own, and change
malloc+memset to calloc.
 2008-02-29 16:01:14 -05:00
Adam Jackson 85617b56e5 Remove some #if 0 guarding code duplicated in xf86Modes.c 2008-02-29 16:01:14 -05:00
Adam Jackson fd41f46ac6 Allow xf86DuplicateMode() to work correctly on read-only modes. 
Before this it was meaningless to try to mark DisplayModeRec tables
const, since the mode name would be emitted as a pointer to an
anonymous string constant, and therefore would have to be fixed up by
ld.so and so couldn't live in .rodata.  With this change the standard
mode lists can live in .rodata, and modes duplicated from them will
have their names filled in on the fly.
 2008-02-29 16:01:14 -05:00
Adam Jackson 8ac19d16a0 Add several comments documenting our EDID failures. 2008-02-29 16:01:14 -05:00
Kristian Høgsberg 13bfa5937d GLX: Adjust to changes in DRI driver interface. 2008-02-29 15:11:13 -05:00
Tiago Vignatti aebd9dc252 Oops, there's one more parenthesis. 2008-02-28 01:22:31 -03:00
Dave Airlie 08afc70513 quirks: another LPL panel with the cm/mm wrong 2008-02-28 10:45:41 +10:00
Dave Airlie d12b7b6632 xf86Crtc: add higher level modesetting entry point. 
For kernel modesetting work we need a bigger stick to beat the modesetting path
 2008-02-27 07:09:14 +10:00
Tiago Vignatti c46f7b62d2 Fine, we don't need pciInit() anymore. Nuke, nuke, nuke... 2008-02-25 18:14:08 -03:00
Tiago Vignatti a9050d5424 Jesus, pciInit() was called 32 times in my machine without need! 
xf86scanpci() will always call pciInit() before any other function, so
there's no need to guarantee it after.
 2008-02-25 18:10:18 -03:00
Tiago Vignatti 81272f7ec9 Some doc clean up: 
clean up legacy things in os-support/bus/Pci.c.

    Put InitOutput() comment to its right place.
 2008-02-25 17:39:33 -03:00
Tiago Vignatti 7c582dd6de remove lnxPciInit() unused function. 2008-02-25 17:31:37 -03:00
Tiago Vignatti 9727db88d5 No more "-scanpci" given that we already have it in libpciaccess. 
(Yeah, lets nuke dead code!)
 2008-02-25 17:08:07 -03:00
Tiago Vignatti f19f7b8e53 Clean some garbage caused when pciaccess was merged: 
FindPCIVideoInfo() function isn't need anymore.

    xf86scanpci() is being called only once so we don't need permanent
    (static) variables there.

    restorePciState() is not used for now (until we find why multiple
    cards aren't working).
 2008-02-25 17:07:07 -03:00
Adam Jackson 4b50e71bf1 Bug #13736: Fix %bx in VBEGetPixelClock to match spec. 
Reported by Yannick Henault.
 2008-02-24 20:25:13 -05:00
Adam Jackson e6a4cde16d Use the client-side XKB headers for the config utilities 2008-02-22 18:36:29 -05:00
Adam Jackson 347db49ebe s/via/openchrome/ in the autoconfig logic. 
Xorg's via driver is effectively dead anyway.
 2008-02-22 16:05:33 -05:00
Adam Jackson ee21aba6be Fix Xinerama's consolidated visual handling. 
Formerly the code claimed it could only handle up to 256 visuals, which
was true.  Also true, but not explicitly stated, was that it could only
handle visuals with VID < 256.  If you have enough screens, and subsystems
that add lots of visuals, you can easily run off the end.  (Made worse
because we allocate visual IDs from the same pool as XIDs.)  If your app
then chooses a visual > 256, then the Xinerama code would throw BadMatch
on CreateColormap and your app wouldn't start.

With this change, PanoramiXVisualTable is gone.  Other subsystems that
were using it as a translation table between each screen's visuals now
use a PanoramiXTranslateVisual() helper.
 2008-02-22 15:19:54 -05:00
Peter Hutterer 67a78e84a8 Revert "xfree86: plug a memory leak in xf86LoadModules." 
This reverts commit 3abce3ea2b and
6cbaf15e61.

The memory returned to xf86LoadModule was allocated in doLoadModule, which
calls the respective module's PreInit. As it turns out, input and output
drivers store a pointer to the module elswhere, so freeing it in
xf86LoadModule is a bad idea.

For further reference: hw/xfree86/common/xf86Helper.c
    Input drivers: xf86InputDriverList[blah]->module = module;
    Output drivers: xf86DriverList[blah]->module = module;

Unloading the module would not look pretty then.
 2008-02-19 21:31:50 +10:30
Peter Hutterer 6cbaf15e61 xfree86: guard against NULL-pointer dereferences in xf86LoadModules. 
Should have been part of 3abce3ea2b, but I
forgot to git-update-index before I committed.

Thanks to Bill Crawford for pointing it out.
 2008-02-19 12:05:13 +10:30
Arjan van de Ven 3abce3ea2b xfree86: plug a memory leak in xf86LoadModules. 
LoadModule() returns the only reference to a fresh piece of memory (a
ModuleDescPtr). Sadly, xf86LoadModules dropped the return value on the floor
leaking memory for each module it loaded.

Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
 2008-02-18 18:14:40 +10:30
Arjan van de Ven 6dc71f6b2c xfree86: plug memory leak in InitPathList. 
All the failure paths were very diligent in freeing the "fullpath" temporary
string, but the success case was not. All the content only got strdup()d, so
it's not live memory anymore.

Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
 2008-02-18 18:14:33 +10:30
Arjan van de Ven 1bec6ad897 xfree86: plug memory leak in xf86LogInit() 
xf86LogInit allocates a piece of memory, stores it in lf. LogInit() will then
effectively strdup it, but lf is never freed again.

Signed-off-by: Peter Hutterer <peter@cs.unisa.edu.au>
 2008-02-18 18:14:20 +10:30
Daniel Stone 68bd7ac193 XKB: Move headers into the server tree 
We need to start breaking the XKB API to enforce sanity, so drag whichever
headers we need to do so into the server tree, as the client API is set in
stone, being part of Xlib.
 2008-02-17 22:52:07 +02:00
Maarten Maathuis a48cc88ea2 Fix rotation for multi-monitor situation. 
- The (x,y)-coordinates of the crtc were not being passed as xFixed values, which made it an obscure bug to find.
- Fix bug #13787.
 2008-02-17 18:47:28 +01:00
Kristian Høgsberg 0ffb6a3ad0 GLX: Implement support for TTM BO based TFP when available. 2008-02-14 22:20:56 -05:00
Kristian Høgsberg 879515b139 Add GLX provider for DRI2. 2008-02-14 19:53:49 -05:00
Kristian Høgsberg b71dc54352 Add DRI2 module. 2008-02-14 19:53:49 -05:00
Adam Jackson 3e0353c785 Bah, macros are hard. 2008-02-15 06:42:48 +11:00
Maarten Maathuis c38feeb149 Fix typo in commit b8b7cdf6df. 2008-02-14 18:52:47 +01:00
Adam Jackson b8b7cdf6df Silence useless debug spew from XFree86-Misc extension. 2008-02-15 01:31:46 +11:00
Adam Jackson ba69904148 Simplify dlloader a bit more. 2008-02-14 08:52:43 +11:00
Adam Jackson f750ce53ac Remove some braindamage from ModuleDesc. 
Yay dead code elimination.
 2008-02-14 08:22:57 +11:00
Adam Jackson e91ff09568 Friends don't let friends call xf86AddModuleInfo. 
That code only existed to leak memory.  It can go now, thanks.
 2008-02-14 08:21:44 +11:00
Adam Jackson 0bdd20a045 Eradicate the VTInit code. 
"This option should rarely be used."  Never sounds like a better idea.
 2008-02-14 07:52:46 +11:00
Adam Jackson 24089b0624 Don't bother warning people about the keyboard driver rename, just do it. 2008-02-14 07:52:33 +11:00
Adam Jackson 2aaf6ac134 Don't even attempt to parse -bpp in xfree86 DDX option parsing. 
This hasn't been valid since 1999.  Get with the times, man.
 2008-02-14 07:52:26 +11:00
Adam Jackson f3b3b37ec6 Use strerror instead of errno values in user strings. 2008-02-14 07:52:02 +11:00
Bart Trojanowski 66f8001b61 X86EMU: handle CPUID instruction 
After trying to switch from X to VT (or just quit) the video-amd driver
attempts to issue INT 10/0 to go to mode 3 (VGA).  The emulator, running
the BIOS code, would then spit out:

        c000:0282: A2 ILLEGAL EXTENDED X86 OPCODE!

The opcode was 0F A2, or CPUID; it was not implemented in the emulator.
This simple patch, against 1.3.0.0, handles the CPUID instruction in one of
two ways:
 1) if ran on __i386__ or __x86_64__ then it calls the CPUID instruction
     directly.
 2) if ran elsewhere it returns a canned 486dx4 set of values for
     function 1.

This fix allows the video-amd driver to switch back to console mode,
with the GSW BIOS.

Thanks to Symbio Technologies for funding my work, and ThinCan for
providing hardware :)

Signed-off-by: Bart Trojanowski <bart@jukie.net>
Acked-by: Eric Anholt <eric@anholt.net>
 2008-02-08 10:48:40 -08:00
liuhong d3c36fe721 validate mode clock for probed modes 
Some modes claimed in monitor EDID data may not be supported by
the monitor. So also validating the max clock for probed modes.
 2008-02-05 12:45:27 -08:00
liuhong 4b5b6e7baa fix max clock unit 
max clock from EDID data is in MHz, while we need KHz to validate modes.
 2008-02-05 12:45:16 -08:00
Peter Hutterer 10617dc0fb xfree86: stick two more checks in for num_valuators < MAX_VALUATORS 2008-02-01 14:24:04 +10:30
Matthieu Herrb c68f063be6 xfree86: don't call xalloc from signal handlers when posting events. 
Reviewed-by: Peter Hutterer <peter@cs.unisa.edu.au>
 2008-02-01 14:14:17 +10:30
Coleman Kane df325be394 Bug 13101: xorg-server has a typo in hw/xfree86/os-support/bsd/i386_video.c 
X.Org Bug #13101 <https://bugs.freedesktop.org/show_bug.cgi?id=13101>
Patch #12360 <https://bugs.freedesktop.org/attachment.cgi?id=12360>
 2008-01-29 09:48:00 -08:00
Julien Goodwin 94f412cb7e Loader: Fix verbosity confusion 
'Loading foo' is verbosity 3, whereas 'already built-in' is verbosity 0.
This means that gdm's log would just be full of bare 'module already
built-in' messages.
 2008-01-29 14:46:15 +11:00
Julien Goodwin 442838fcb3 xorg.conf.man: Fix monitor/output confusion in monitor positioning 
On the Intel driver at least, LeftOf/RightOf/Above/Below in xorg.conf refers
to output names, not monitor names.  Fix nomenclature in xorg.conf.man.
 2008-01-29 14:46:15 +11:00
Peter Hutterer 5b8641a5fd xfree86: fix AlwaysCore handling. (Bug #14256) 
Assume AlwaysCore being set by default, just like the other options.

X.Org Bug 14256 <http://bugs.freedesktop.org/show_bug.cgi?id=14256>
 2008-01-29 10:01:42 +10:30
Eamon Walsh e915a26397 xselinux: Move the extension to extmod instead of being built-in. 2008-01-25 19:22:19 -05:00
Eamon Walsh f6a78ee143 XACE: Remove the extension code entirely, XACE is completely static now. 2008-01-25 18:04:10 -05:00
Hong Liu 734e115871 Bug #12439: add a quirk to use +hsync +vsync for the probed detailed mode. 
Samsung 205BW quirk is somehow reworked.
 2008-01-23 21:24:09 +08:00
Adam Jackson cc22b05ea0 There is no such thing as /dev/cpu/mtrr. 2008-01-22 18:57:11 -05:00
Matthieu Herrb bbde5b62a1 Fix for CVE-2007-5760 - XFree86 Misc extension out of bounds array index 2008-01-17 15:26:41 +01:00
Dave Airlie 1f83f40525 xf86Cursors: fix memset for non-square cursors 2008-01-15 10:21:17 +10:00
Peter Harris 0883e838e2 xf86misc.c: Avoid use of swapped values 2008-01-09 18:43:55 -08:00
Alan Coopersmith 59a63d72a1 Fix reference to old versions of XFree86 to not say "Xorg 4.2 or earlier" 2008-01-09 18:42:58 -08:00
David Nusinow 30375cd6d1 Don't log DPMS enabling as being from xorg.conf if it's not 2008-01-07 20:57:30 -05:00
David Nusinow 260505e3c5 Log enabling of DPMS even when it's not set in xorg.conf 2008-01-06 16:37:13 -05:00
Julien Cristau 7e76221659 Fix the name of the XFree86-Misc extension in the xorg.conf manpage. 2008-01-06 18:23:09 +01:00
Julien Cristau 59df687835 Document the AllowEmptyInput, AutoAddDevices and AutoEnableDevices flags 
Add documentation for the new AllowEmptyInput, AutoAddDevices and
AutoEnableDevices server flags in the xorg.conf manpage.
 2008-01-06 17:02:03 +01:00
Alex Deucher 20eb26f9d1 Fix potential crasher in xf86CrtcRotate() 
xf86CrtcRotate() is called by randr 1.2 drivers via xf86CrtcSetMode() or xf86SetDesiredModes()
during ScreenInit() at which point pScrn->pScreen is not set. If a user specifies a rotation
in their config file pScrn->pScreen is dereferenced and boom.
 2008-01-03 22:53:36 -05:00
Alan Coopersmith ccf6636d2c Kill xf86getpagesize even harder (dummylib & ioport) 2008-01-02 19:28:33 -08:00
Alan Coopersmith f6666dcc3b Add dixAllocatePrivate stub to dummylib for utils 
Normally not necessary, except when building non-optimized/debug causes
the inline functions from private.h to appear in os-support/libxorgos.la
 2008-01-02 19:19:55 -08:00
Alan Coopersmith 73f4229960 Fix names/types of new vuidMouse{Get,Set}ScreenPrivates macros 2008-01-02 19:17:54 -08:00
Adam Jackson 85365ddf16 EDID 1.4: If given a native pixel format, use it when inferring virtual. 2007-12-28 17:38:52 -05:00
Adam Jackson 312b30cb03 EDID 1.4: First detailed mode is always preferred. 
... so act like it in the modelist generator, not just the parser.
 2007-12-28 17:38:51 -05:00
Adam Jackson bae459cfc4 Don't carp on EDID 1.4 blocks anymore. 
Also whine more loudly when we get something other than 1.x.
 2007-12-28 16:53:17 -05:00
Adam Jackson 70b2d6cfeb Check the gamma value, not its address. 2007-12-28 16:53:17 -05:00
Adam Jackson 592d814ee0 EDID 1.4: Additional semantics for display feature bits. 
First mode is _always_ preferred in 1.4; the bit that used to mean this
now means that the preferred mode is also the native pixel format.  The
old "is GTF" bit now means "is continuous-frequency" instead.

Section 3.6.4, Table 3.14: Feature Support, Notes 4 and 5.
 2007-12-28 16:53:17 -05:00
Adam Jackson 322d0103ae EDID 1.4: Alternate color encodings for digital inputs. 
Section 3.6.4, Table 3.14: Feature support.
 2007-12-28 16:53:17 -05:00
Adam Jackson f1f43caf7e EDID 1.4: Allow for gamma definition in extension blocks. 
Section 3.6.3, Table 3.13: Display Transfer Characteristics (Gamma)
 2007-12-28 16:53:16 -05:00
Adam Jackson 861ee38817 EDID 1.4: Additional aspect ratio semantics for screen size fields. 
Section 3.6.2, Table 3.12: Horizontal and Vertical Screen Size or
Aspect Ratio.
 2007-12-28 16:53:16 -05:00
Adam Jackson 14b5c8a447 EDID 1.4: Extended support for digital interfaces. 
Section 3.6.1, Table 3.11: Video Input Definition.
 2007-12-28 16:53:16 -05:00
Adam Jackson a948216dcc EDID 1.4: Decode CVT 3-byte codes and add them to the mode pool. 
Section 3.10.3.8: CVT 3 Byte Code Descriptor Definition.
 2007-12-28 16:53:16 -05:00
Adam Jackson f6df66cc89 EDID 1.4: Trivial support for new detailed sections. 
Nothing actually decoded yet, but at least we print what they are.

New in EDID 1.4:
- Color Management Data (0xF9), Section 3.10.3.7
- CVT 3 Byte Code Descriptor (0xF8), Section 3.10.3.8
- Established Timings III Descriptor (0xF7), section 3.10.3.9
- Manufacturer-specified data tag (0x00 - 0x0F), section 3.10.3.12
 2007-12-28 16:53:16 -05:00
Adam Jackson bac3ecde39 EDID 1.4: Allow for sync range offsets. 
Table 3.26: Display Range Limits & Timing Descriptor Block Definition
 2007-12-28 16:53:16 -05:00
Eamon Walsh 643c52be32 xselinux: Remove "X" prefix on remaining functions and strings. 
Should be evident from the context.
 2007-12-28 13:27:28 -05:00
Adam Jackson 938da5ee38 Add HDMI and DisplayPort connector types. 2007-12-28 12:23:44 -05:00
David S. Miller ae869fc766 [SBUS]: Fix build, use getpagesize() instead of xf86getpagesize(). 
xf86getpagesize() was removed, but this one call site was
not fixed up.

Signed-off-by: David S. Miller <davem@davemloft.net>
 2007-12-27 07:34:09 +01:00
Dave Airlie 7ef7727b80 entity sharing: make !shareable entity non-fatal. 
Just because the entity isn't shareable, we should bring down the server.
Just ignore the extra screen and keep going.
 2007-12-20 10:11:26 +10:00
Adam Jackson 7721d3e921 Reference cvt and gtf in the xorg.conf man page. 2007-12-18 19:14:26 -05:00
Samuel Thibault 58c2898b62 xfree86: permit access to io port 0xffff on the hurd 2007-12-16 01:21:45 +01:00
Eamon Walsh a2df51f8e9 Bump video driver ABI and extension ABI for devPrivates rework. 2007-12-14 00:53:54 -05:00
Eamon Walsh 86b2e59bfb Merge branch 'master' into XACE-SELINUX 
Conflicts:

	configure.ac

This is the last pull before merging to master.
 2007-12-14 00:32:24 -05:00
Eamon Walsh c8feb73f58 Merge branch 'master' into XACE-SELINUX 
Conflicts:

	Xext/EVI.c
	Xext/bigreq.c
	Xext/cup.c
	Xext/dpms.c
	Xext/fontcache.c
	Xext/mitmisc.c
	Xext/xcmisc.c
	Xext/xf86bigfont.c
	Xext/xtest.c
	configure.ac
	dbe/dbe.c
	hw/darwin/darwin.h
	hw/darwin/darwinEvents.c
	hw/darwin/iokit/xfIOKit.h
	hw/darwin/iokit/xfIOKitCursor.c
	hw/darwin/quartz/fullscreen/fullscreen.c
	hw/darwin/quartz/fullscreen/quartzCursor.c
	hw/darwin/quartz/quartz.c
	hw/darwin/quartz/quartzCommon.h
	hw/darwin/quartz/quartzCursor.c
	hw/darwin/quartz/xpr/dri.c
	hw/darwin/quartz/xpr/dristruct.h
	hw/darwin/quartz/xpr/xprCursor.c
	hw/darwin/quartz/xpr/xprFrame.c
	hw/xfree86/modes/xf86RandR12.c
	include/cursor.h
	miext/rootless/rootlessCommon.h
	miext/rootless/rootlessScreen.c
	miext/rootless/rootlessWindow.c
	render/picturestr.h

Trying to pick up the pieces from the darwin churn here...
 2007-12-13 18:38:25 -05:00