Am 18.07.2013 13:37, schrieb v...@picaros.org:
http://bugs.freedesktop.org/show_bug.cgi?id=65210
Commit 37d956e3ac9513b74078882dff489f9b0a7a5a28 presets
config-compat_output = -1 to signal an unset compat_output.
Since compat_output is used to index config-output[] during initial
The current code allows for compat_output = num_output when
num_output decreases to zero. Thus hw/xfree86/modes/xf86Crtc.h
xf86CompatOutput() may read beyond the end of the output array via
config-output[config-compat_output].
Move the compat_output update into its own if() block. Then it is
xserver/hw/modes: compat_output bad dereference.
The commit 37d956e3ac9513b74078882dff489f9b0a7a5a28 into
xserver/hw/modes/xf86Crtc.c xf86CrtcConfigInit() initializes
config-compat_output = -1;
So an unset compat_output is now invalid, a good property since the
previous initial
The commit 37d956e3ac9513b74078882dff489f9b0a7a5a28 into
xserver/hw/modes/xf86Crtc.c xf86CrtcConfigInit() modified the
initialization of config-compat_output from default 0 to -1.
This change exposes a bug. During initial configuration a monitor
detection is attempted and if an EDID block is
On 11/21/2012 04:12 AM, v...@picaros.org wrote:
Add support for the Option ZoomModes in a monitor section:
Section Monitor
Identifier a21inch
Option PreferredMode 1600x1200
Option ZoomModes 1600x1200 1280x1024 1280x1024 640x480
EndSection
ZoomModes seems like an
Add support for the Option ZoomModes in a monitor section:
Section Monitor
Identifier a21inch
Option PreferredMode 1600x1200
Option ZoomModes 1600x1200 1280x1024 1280x1024 640x480
EndSection
The option's effect is to search for and mark once each named mode in
the output modes list. So
Just an observation: setting the word none in italic hints the
reader that this is a symbolic name or parameter. However, it is to
be typed in literally. So perhaps
diff --git a/man/Xserver.man b/man/Xserver.man
index 1a36b09..023e009 100644
--- a/man/Xserver.man
+++ b/man/Xserver.man
@@
On Sat, Sep 17, 2011 at 06:58:18PM +0200, v...@picaros.org wrote:
This patch fixes a potential buffer overrun in xf86addComment(). The
overrun occurs if the comment string to add doesn't start with a
leading '#'.
out of interest - how do you trigger this? Presumably xf86addComment
Anyone willing to review this triviality ? The code is included below.
This patch fixes a memory leak in xf86CrtcCreate(). It frees the
crtc-gamma_red buffer when a quit on error occurs.
xf86CrtcPtr
xf86CrtcCreate (ScrnInfoPtr scrn,
const xf86CrtcFuncsRec *funcs)
This patch fixes a potential buffer overrun in xf86addComment(). The
overrun occurs if the comment string to add doesn't start with a
leading '#'.
The bug can be seen from the code assuming
curlen 0 eol_seen == 0 iscomment == 0 :
char *
xf86addComment(char *cur, char *add)
...
diff --git a/hw/xfree86/man/xorg.conf.man b/hw/xfree86/man/xorg.conf.man
index 6774dbc..6aae650 100644
--- a/hw/xfree86/man/xorg.conf.man
+++ b/hw/xfree86/man/xorg.conf.man
@@ -948,6 +948,17 @@ is equivalent to
This option controls the startup behavior only, a device
may be
Am 31.08.2011 07:06, schrieb v...@picaros.org:
[PATCH xserver/hw/xfree86/parser/scan.c] potential buffer overflow
The patch below fixes a potential buffer overflow in xf86addComment().
This occurs if curlen 0 eol_seen == 0 iscomment == 0 , as
follows from the code:
char
[PATCH xserver/hw/xfree86/parser/scan.c] potential buffer overflow
The patch below fixes a potential buffer overflow in xf86addComment().
This occurs if curlen 0 eol_seen == 0 iscomment == 0 , as
follows from the code:
char *xf86addComment(char *cur, char *add)
...
len =
This tiny patch fixes a memory leak in xf86CrtcCreate().
Signed-off-by: Servaas Vandenberghe
diff --git a/hw/xfree86/modes/xf86Crtc.c b/hw/xfree86/modes/xf86Crtc.c
index c2814d4..74c8d66 100644
--- a/hw/xfree86/modes/xf86Crtc.c
+++ b/hw/xfree86/modes/xf86Crtc.c
@@ -133,6 +133,7 @@ xf86CrtcCreate
The xorg.conf manual uses the following convention in most of its
sections:
bold = text to be copied literally to the config file,
italic = a symbolic name to be substituted by a true value.
Some configuration keywords seem to have been changed into generic
options. Prepending Option to the
This patch adds printing of the DisplayMode type bits to
xf86PrintModeline(). It helps to trace the modeline origin and to
understand the initial configured modeline.
This is rather useful if you're trying to figure out where a mode
might have come from.
Example output:
Modeline
This patch adds support for the
Option ZoomModes 1600x1200 1280x1024 1280x1024 640x480
in a monitor section. The code tries to match each mode name and
sets the M_T_USERDEF mode type bit if found. A mode will not be
rematched so specifying a name twice selects the next, usually lower
The patch here below is unchanged, except for the letter bitmap
description added in a comment block.
This patch adds printing of the DisplayMode type bits. It helps to
trace the modeline origin and to understand the initial configured
modeline.
This is rather useful if you're trying to
Assuming this is the only patch to be applied, I'd like some
explanation in the code as to why some of the letters are capitalized.
As a suggestion :
/* U and P are capitalized because USERPREF and PREFERRED have
priority when sorting. */
Perhaps a description would be best. The text
The patch below should apply to xserver from git.
Signed-off-by: Servaas Vandenberghe
diff --git a/hw/xfree86/modes/xf86Modes.c b/hw/xfree86/modes/xf86Modes.c
index 51eb4c9..eff13f1 100644
--- a/hw/xfree86/modes/xf86Modes.c
+++ b/hw/xfree86/modes/xf86Modes.c
@@ -283,10 +283,29 @@ add(char **p,
The patch below adds printing of the DisplayMode type bits. It helps
to trace the modeline origin and to understand the initial configured
modeline.
type_bit flag origin
M_T_USERPREF U set by the Option PreferredMode
M_T_DRIVER e driver: EDID, flat panel native
Once upon a time xdm was created. This program manages X terminals by
forking 1 or 2 processes per display:
1. a local server 'X :0' if required, and
2. the session manager greeter '-:0'.
The dm.c WaitForChild() loop interpretes the session exit code and
uses a ++startTries=startAttempts
22 matches
Mail list logo