Commit Graph

1375 Commits

Author SHA1 Message Date
matthieu
52218799b8 Fixes for various integer overflow problems from X.Org:
CVE-2008-2360 - RENDER Extension heap buffer overflow
CVE-2008-2361 - RENDER Extension crash
CVE-2008-2362 - RENDER Extension memory corruption
CVE-2008-1379 - MIT-SHM arbitrary memory read
CVE-2008-1377 - RECORD and Security extensions memory corruption
2008-06-11 15:39:26 +00:00
mbalmer
ef9d4b4b8c Pull in a patch from X.Org-current to let X11 run on AMD Geode LX
CPUs in basic VESA mode:

X86EMU: handle CPUID instruction

Starting X on an AMD Geode LX system such as an Alix board with VGA
connector, the Xorg 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. With this patch it 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.

Tested with & ok matthieu
2008-06-10 22:31:30 +00:00
matthieu
f01cdd7229 - actually zero the memory returned by realloc().
- set the return value for the pthread_once() stub.
2008-06-10 22:01:31 +00:00
oga
eb88ad3f01 Include stdlib.h so we have the prototype for realloc. This means that we don't
assume realloc returns an int. This causes problems...

This took me hours to find.

Ok matthieu@.
2008-06-10 21:24:16 +00:00
phessler
f80be2c38a Remove the special case for amd64 using MTRR. Wasn't compiled in in the
first place, and didn't work anyways.

ok deraadt, oga, matthieu
2008-06-09 22:55:10 +00:00
okan
45b2fad725 Enter -> Return, to be consistant and correct.
found the hard way by johan and discovered by oga.

"do it" oga@
2008-06-05 14:29:01 +00:00
okan
f42ad31e00 prevent trying to exec a null char; could potentially happen with a
canceled or empty searchstr.

ok oga@
2008-06-05 04:01:37 +00:00
okan
9005249c59 actually honor termpath and lockpath if specified in cwmrc.
"now" oga@
2008-06-05 00:07:05 +00:00
okan
2abd40cbdb "Meta is a perfectly well defined concept in X (The keys bound to the
Meta_L or Meta_R keysysm).  no need to redefine it roughly there" - matthieu@
2008-06-03 21:27:47 +00:00
matthieu
c6b3b52580 Revert to 1.4.2.1. The change that pulls Xmd.h in more headers is
causing too much problems in ports. requested by espie@ and deraadt@
2008-06-03 05:13:48 +00:00
todd
35417f08d8 sync 2008-06-01 20:50:40 +00:00
matthieu
cd215b2b5d new upstream releases 2008-06-01 13:05:30 +00:00
matthieu
19a6c8b857 sync 2008-06-01 13:00:45 +00:00
matthieu
80656e86d4 Update to Mesa 7.0.3. tested my oga@ and johan@ 2008-05-31 16:36:47 +00:00
todd
85b293ceec sync 2008-05-31 04:13:27 +00:00
matthieu
8d6a1e5b56 update to inputproto 1.4.3 2008-05-25 11:14:58 +00:00
matthieu
48bc822959 regen. 2008-05-24 20:39:01 +00:00
matthieu
1b5891c504 implement the --disable-ioport configure option to disable the build of
ioport and its symlinks and use that option. Requested by kettenis@
and others.
2008-05-24 20:36:04 +00:00
matthieu
d262dc5de2 Update to printproto 1.0.4 2008-05-24 18:20:21 +00:00
matthieu
1caa452538 sync 2008-05-24 15:46:32 +00:00
matthieu
cae2cf16df Update to libxtrans 1.2 2008-05-24 15:44:21 +00:00
matthieu
d8808c0da2 Update to libxkbfile 1.0.5 2008-05-24 14:21:56 +00:00
matthieu
b9c810fd57 Merge libpciaccess 0.10.2, including kettenis@ OpenBSD port. 2008-05-24 14:14:00 +00:00
matthieu
a8393f0f0c update to libXv 1.0.4 2008-05-24 13:46:47 +00:00
matthieu
b40a014ccf update to libXinerama 1.0.3 2008-05-24 13:31:12 +00:00
matthieu
b6e6f0b4c0 merge libXfont 1.3.2. bump major since some symbols were removed. 2008-05-24 13:29:07 +00:00
matthieu
28676ec1c9 Update to libXext 1.0.4 2008-05-24 09:29:33 +00:00
matthieu
c17f6cf0ff update to libXScrnSaver 1.1.3 2008-05-24 09:17:11 +00:00
matthieu
d8ac6b7b37 update to libFS 1.0.1 2008-05-24 09:02:48 +00:00
okan
ddd8d269ff make sure to take bwdith into account when placing a new window.
ok oga@
2008-05-23 18:57:35 +00:00
oga
6d57649414 Grab the keyboard when we initialise the menu. This stops the keyboard
shortcut code stealing our events in some cases.

"put 'er in" okan@.
2008-05-23 18:48:57 +00:00
matthieu
80fa2403ff More files that got added by mistake. 2008-05-22 06:59:23 +00:00
todd
a81402b204 sync 2008-05-21 21:56:14 +00:00
matthieu
3d060c0d86 oops should not have committed this. Noticed by deraadt@. 2008-05-21 20:27:45 +00:00
matthieu
a939d92966 sync 2008-05-21 20:25:41 +00:00
matthieu
4203da792a Update to xf86-video-intel 2.3.1. Tested by many. 2008-05-21 20:19:51 +00:00
oga
bad8c87b4a Make menu_filter handle mouse movement too. This enables the keyboard
search dialogues to be manipulated with the mouse, too. It also allows
me to shrink the codebase further by killing grab_menu().

One known issue with highlighting the first entry in a search dialogue,
that'll be fixed soonish.

ok okan@, tested by Edd Barrett and todd@.
2008-05-21 14:11:19 +00:00
oga
ca512ce795 Pull out the behaviour in grab_label and search_start into one utility
function menu_filter(). The plan is to eventually merge in grab_menu too.
Shrinks the code a fair bit.

Also, change XMaskEvent for XWindowEvent to prevent getting exposes for other
windows. This is particuarly noticable on slow machines with a LOT of xterms
(todd, you're an odd man).

ok okan@, todd@.
2008-05-20 14:50:51 +00:00
oga
f8b4088fad General cleanup.
ok okan@.
2008-05-19 18:53:09 +00:00
okan
88bce60b1c stop normalizing search input; searching and matching are still
case-insensitive.  since this was the only use of normalizing input,
simplify as well.

allows one to exec with mixed case unmatched commands.

"works for me" oga@
2008-05-19 18:07:53 +00:00
oga
e13d74cfcc Function prototypes should not have parameter names in them. These must
have been missed last time i knfed this.

ok okan.
2008-05-19 17:32:22 +00:00
okan
d07a7f737f client_cyclenext() -> client_cycle() since we now pass an arg.
removes a stray proto as well.

discussed with and ok oga@
2008-05-19 17:24:19 +00:00
oga
796006fd7f allow an autogroup value of 0 to mean no group. This means you can set
automatically "sticky" (in the traditional sense of the word) windows in
autogroup mode.

Based on an initial diff from Andrew Fresh, thanks!

ok okan@.
2008-05-19 17:13:55 +00:00
okan
7f3adeacc7 as done with cycle/rcycle, make prev/next group switching one kbfuncs
and use a flag; adjusted to match and rename to {r,}cycle.

"ok, since i came up with the same thing" oga@
2008-05-19 17:11:19 +00:00
oga
80fd601708 Use the XGrabKeyboard hack in for alt-tabbing as well. This stops the
mru getting the order messed up when gvim/xpdf et all steal key events.

While i'm here, change the logic in client_cyclenext() to use break instead
of goto, it's nicer that way.

Thirdly, instead of two different kbfuncs, just use the one and a flag.

"put your cycle diff in so I can pkg_delete gvim" okan@
2008-05-19 15:17:50 +00:00
okan
0a4376b166 finally implement keyboard binding for group toggling
idea for the
    "slightly-less-abhorrent-hack-but-a-hack-nonetheless-TM" from oga@

grab and ungrab the keyboard to get around some silly X apps that like
stealing events

ok oga@
2008-05-19 12:56:58 +00:00
okan
98aad18f7d send the correct x/y coordinates to XConfigureWindow()
fixes some windows that seem as if they don't fit; noticed by Edd Barrett.

ok oga@
2008-05-18 20:06:36 +00:00
okan
0e709156e6 remove extra calls to client_draw_border()
ok oga@
2008-05-18 20:00:16 +00:00
okan
8d2a0490e4 fix backwards logic in example; found by oga@ 2008-05-18 19:57:43 +00:00
oga
ca54149838 Fix two problems with conf_unbind():
1) it used TAILQ_FOREACH() when it's removing entrys from the list, this
   is bad.
2) We didn't free key, so there was a small memleak too.

also rework conf_bindname's logic slightly to be more simple.

ok okan@
2008-05-18 19:47:19 +00:00