guessing that's exactly why no entries ever started with whitespace -
they wouldn't actually work. :)
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
> This driver is used for SGI workstations Indy and Indigo2. Manufacturing
> stopped in 1997 and SGI support ended in 2012.
>
> Signed-off-by: Gaetan Nadon
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: h
On Sep 18, 2013 7:50 AM, "Gaetan Nadon" wrote:
>
> This driver has not received server updates such as mibstore removal
> and conditional compiling of xaa architecture. This indicate an implicite
> decisison to remove the driver from active support and future releases.
>
> This driver is used for
On Jul 15, 2013 10:46 PM, "Alan Coopersmith"
wrote:
>
> On 07/14/13 01:33 PM, Thomas Dickey wrote:
>>
>> On Sun, Jul 14, 2013 at 09:45:01PM +0200, Matthieu Herrb wrote:
>>>
>>> Note however that as is luit is currently broken on OpenBSD. I had to
>>> commit a patch to OpenBSD's xenocara to disable
uot;ServerFlags"
> +Option "DontVTSwitch" "on"
> +EndSection
> +
> +Section "InputClass"
> +Identifier "Force input devices to seat"
> +Option "GrabDevice" "on"
> +EndSection
Is that all there is
On Sat, Apr 20, 2013 at 11:28 AM, Gaetan Nadon wrote:
> On 13-04-20 01:08 PM, Alan Coopersmith wrote:
>> Check for groff never got translated from imake to autoconf
>>
>> Signed-off-by: Alan Coopersmith
>> ---
>> configure.ac |5 +
>> defs.h |4
>> 2 files changed, 9 inser
On Sat, Apr 20, 2013 at 10:08 AM, Alan Coopersmith
wrote:
> Check for groff never got translated from imake to autoconf
>
> Signed-off-by: Alan Coopersmith
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archi
pretty good to me, but I
didn't try extremely hard to navigate the previous #ifdef maze.
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
d
> xorg.modules: Add mesa metamodule
> xorg.modules: Add split xcb util repositories so they get tinderboxed
> xorg.modules: Add xf86-video-nested so it gets tinderboxed
Except for a couple questions, these look good to me.
Reviewed-by: Dan Nicholson
_
On Feb 28, 2013 7:10 AM, "Jon TURNEY" wrote:
>
> Add split xcb util repositories and a xcb metamodule
>
> Signed-off-by: Jon TURNEY
> ---
> xorg.modules | 32
> 1 files changed, 32 insertions(+), 0 deletions(-)
>
> diff --git a/xorg.modules b/xorg.modules
> ind
On Feb 28, 2013 7:08 AM, "Jon TURNEY" wrote:
>
> mesa is now sufficently autotooled that:
> - it doesn't need makedepend
> - non-srcdir builds work
> - it doesn't need skip-autogen
> - the check target exists
>
> Signed-off-by: Jon TURNEY
>
> libGL doesn't need makedepend
I haven't checked on th
_cv_header_dlcfn_h -o x$ac_cv_header_dl_h; then
> +if test "x$ac_cv_header_dlcfn_h" = xyes -o "x$ac_cv_header_dl_h" = xyes; then
> HAVE_LOADABLE_MODULES=yes
> else
> HAVE_LOADABLE_MODULES=no
Looks right.
Reviewed-by: Dan Nicholson
_
ata->mtdev);
> proto_data->mtdev = NULL;
> proto_data->num_touches = 0;
> }
>From what I can tell, evdev_query_touch() has the same imbalanced
mtdev_new_open/mtdev_close imbalance. With that fix, too,
Reviewed-by: Dan Nicholson
__
ays?
I think when I was doing the port, blowup seemed to be only a fringe
tool and I figured I'd get back to it later. 3 years later...
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archive
lid or uninitialized input device",/* BadDevice
> */
> "BadEvent, invalid event type",/* BadEvent */
> "BadMode, invalid mode parameter", /* BadMode */
I didn't actually look through the code to see all the usage of these
two variabl
t;id;
> + I->resolution = ifs->resolution;
> + I->minVal = ifs->min_value;
> + I->maxVal = ifs->max_value;
> break;
> }
> case StringFeedbackClass:
> diff --git a/src/XIQueryVersion.c
-status = XkbDDXListComponent(dev, _XkbListSymbols, list, client);
> -if (status == Success)
> -status = XkbDDXListComponent(dev, _XkbListGeometry, list,
client);
> -return status;
> -}
> diff --git a/xkb/xkb.c b/xkb/xkb.c
> index 4440a98..0d3bda3 100644
> --- a/xkb/xkb.c
> +++ b/xkb/xkb.c
> @@ -5621,9 +5621,9 @@ ProcXkbListComponents(ClientPtr client)
> DeviceIntPtr dev;
> xkbListComponentsReply rep;
> unsigned len;
> -int status;
> unsigned char *str;
> -XkbSrvListInfoRec list;
> +uint8_t size;
> +int i;
>
> REQUEST(xkbListComponentsReq);
> REQUEST_AT_LEAST_SIZE(xkbListComponentsReq);
> @@ -5633,40 +5633,27 @@ ProcXkbListComponents(ClientPtr client)
>
> CHK_KBD_DEVICE(dev, stuff->deviceSpec, client, DixGetAttrAccess);
>
> -status = Success;
> str = (unsigned char *) &stuff[1];
> -memset(&list, 0, sizeof(XkbSrvListInfoRec));
> -list.maxRtrn = stuff->maxNames;
> -list.pattern[_XkbListKeycodes] = GetComponentSpec(&str, FALSE,
&status);
> -list.pattern[_XkbListTypes] = GetComponentSpec(&str, FALSE, &status);
> -list.pattern[_XkbListCompat] = GetComponentSpec(&str, FALSE,
&status);
> -list.pattern[_XkbListSymbols] = GetComponentSpec(&str, FALSE,
&status);
> -list.pattern[_XkbListGeometry] = GetComponentSpec(&str, FALSE,
&status);
> -if (status != Success)
> -return status;
> +for (i = 0; i < _XkbListNumComponents; i++) {
> +size = *((uint8_t *)str);
> +str += (size + 1);
> +}
This seems wrong, although I'm not extremely familiar with how the server
writes things on the wire. Overall, I agree with your assessment of the
usefulness of this code. It's far more correct to do this from the XML on
the client side.
Acked-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
On Oct 17, 2012 5:36 PM, "Peter Hutterer" wrote:
>
> On Thu, Oct 18, 2012 at 09:40:11AM +1000, Dave Airlie wrote:
> > On Thu, Oct 18, 2012 at 9:15 AM, Aaron Plattner
wrote:
> > > On 10/17/2012 04:12 PM, Peter Hutterer wrote:
> > >>
> > >> On Wed, Oct 17, 2012 at 03:02:53PM -0700, Aaron Plattner w
LOn Oct 2, 2012 1:09 PM, "Julien Cristau" wrote:
>
> On Sun, Sep 30, 2012 at 16:45:17 -0400, Gaetan Nadon wrote:
>
> > On 12-09-29 04:37 PM, Dan Nicholson wrote:
> > >
> > > > Some distros may prefer maintainer mode. A way to appease everyone
On Sep 29, 2012 11:19 AM, "Chase Douglas" wrote:
>
> On Thu, Sep 27, 2012 at 2:12 PM, Adam Jackson wrote:
> > All this does is make it so editing configure.ac or Makefile.am doesn't
> > rebuild the makefiles. Which is just stupid.
> >
> > Signed-off-by: Adam Jackson
> > ---
> > configure.ac |
ot;? Which is true, if you've said AM_MAINTAINER_MODE in
configure.ac.
It's more correct for downstream in a sense. You don't accidentally get the
autotools pulled in during your tarball build if you patched something.
There are ways to accomplish that (touch) without maintainer
On Mon, Sep 24, 2012 at 8:04 AM, Yaakov (Cygwin/X)
wrote:
> On Mon, 2012-09-24 at 07:15 -0700, Dan Nicholson wrote:
>> On Sep 23, 2012 9:59 PM, "Yaakov (Cygwin/X)" wrote:
>> >
>> > On 2012-09-23 12:35, Matt Turner wrote:
>> >> The proto packa
On Sep 23, 2012 9:59 PM, "Yaakov (Cygwin/X)" <
yselkow...@users.sourceforge.net> wrote:
>
> On 2012-09-23 12:35, Matt Turner wrote:
>>
>> The proto packages install their pc files in $(libdir)/pkgconfig, but
>> this leads to files being installed in /usr/lib32 or /usr/lib64 when
>> there's nothing
On Sep 23, 2012 10:35 AM, "Matt Turner" wrote:
>
> Hi Gaetan,
>
> The proto packages install their pc files in $(libdir)/pkgconfig, but
> this leads to files being installed in /usr/lib32 or /usr/lib64 when
> there's nothing ABI specific about them.
>
> Would it be reasonable to install them to $(
;
> So, add an option to allow explicitly specifying the location of
keysymdef.h,
> and verify that the specified or found path exists.
>
> (original patch by Martin Jansa , revised by
myself)
>
> Signed-off-by: Ross Burton
Reviewed-by: Dan Nicholson
_
if
someone throws in a directory with spaces.
> + AC_MSG_ERROR([$KEYSYMDEFDIR doesn't exist or isn't a directory])
> +fi
> +
> FILES="keysymdef.h XF86keysym.h Sunkeysym.h DECkeysym.h HPkeysym.h"
> for i in $FILES; do
> if test -f "$KEYSYMDEFDIR/$i&qu
On Jul 19, 2012 3:38 PM, "Daniel Stone" wrote:
>
> Hi,
>
> On 19 July 2012 22:53, Keith Packard wrote:
> > Adam Jackson writes:
> >> I think there's a combination of factors that pile on to make it worse
> >> than you'd expect.
> >
> > And, caching is completely free, so it seems sensible to jus
Sorry for the slow reply. This was interesting to read, but I got busy.
On Fri, Jul 13, 2012 at 7:50 AM, Daniel Stone wrote:
> Hi,
>
> On 13 July 2012 13:34, Dan Nicholson wrote:
>> Not that I actually have the time to work on this, but I'd been
>> thinking about xkbc
On Thu, Jul 12, 2012 at 3:16 PM, Daniel Stone wrote:
> Hi,
>
> On 12 July 2012 23:07, Keith Packard wrote:
>> Construct a unique filename based on the display name and the RMLVO
>> values. If that file contains valid contents, use it. Otherwise,
>> compile the keymap to that file and don't unlink
On Fri, Jun 8, 2012 at 3:20 PM, Daniel Stone wrote:
>
> Gaetan is our build system maintainer (sorry Gaetan)
I like how the anointing of build system maintainership requires an apology. :)
--
Dan
___
xorg-devel@lists.x.org: X.Org development
Archives:
quot;);
> + }
> + else {
> AccessXKRGTurnOn((DeviceIntPtr) arg, XkbSlowKeysMask, &cn);
> + LogMessage(X_INFO, "XKB SlowKeys are now enabled. Hold shift to
> disable.\n");
> + }
> +
> return 0;
> }
>
I never knew about that. S
On Tue, Mar 27, 2012 at 2:38 AM, Zhigang Gong
wrote:
> Hi,
>
>
>
> I’m building xorg in a local directory $HOME/xorg, then all libraries will
> go to $HOME/xorg/lib,
>
> And all the configure files will go to $HOME/xorg/share/X11/xorg.conf.d/,
> such as the 10-evdev.conf.
>
>
>
> My question is, w
2012/1/4 Ian Romanick :
> On 01/04/2012 11:05 AM, Michel Dänzer wrote:
>>
>> On Mit, 2012-01-04 at 10:56 -0800, Ian Romanick wrote:
>>>
>>> On 01/04/2012 10:55 AM, Daniel Stone wrote:
Hi,
On 4 January 2012 18:45, Ian Romanick wrote:
>
> Okay, I looked back at your buil
ixes https://bugs.freedesktop.org/show_bug.cgi?id=43320
>
> Signed-off-by: Alan Coopersmith
Looks right, but it would be REALLY great for the build if that stub DDX
existed. Most of my experimenting with building Xorg smarter broke down
because it also needed to build from te
6Init.c:365)
> by 0x425A7A: main (main.c:204)
>
> Signed-off-by: Paulo Zanoni
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
void xf86closeConfigFile(void);
Seems pretty reasonable, but do you know why xf86openConfigDirFiles is
called twice? Oh, I guess it's because we also look at the system
config dir. Now I recall thinking this would leak but not really
caring to fix it at the time. :) I certainly like the idea of passing
around variables when possible and not poking globals.
These functions also get called from hw/xwin/winconfig.c. I'm not sure
how used that code is anymore, but I was trying to keep it updated at
the time. Besides that,
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
0x49AA77: InitOutput (xf86Init.c:365)
> by 0x425A7A: main (main.c:204)
>
> Signed-off-by: Paulo Zanoni
I guess I thought the parser was gonna clean those up.
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archi
On Tue, Nov 1, 2011 at 9:27 PM, Peter Hutterer wrote:
> On Tue, Nov 01, 2011 at 11:12:36AM -0200, przan...@gmail.com wrote:
>> From: Paulo Zanoni
>>
>> It seems appropriate to make the function that frees the list elements
>> also free the list.
>>
>> 80 bytes in 1 blocks are definitely lost in l
On Oct 27, 2011 4:26 AM, "Peter Hutterer" wrote:
>
> On Thu, Oct 27, 2011 at 10:41:06AM +0200, walter harms wrote:
> >
> >
> > Am 27.10.2011 09:55, schrieb Jamey Sharp:
> > > On Thu, Oct 27, 2011 at 09:15:54AM +0200, walter harms wrote:
> > >> Am 27.10.2011 06:21, schrieb Peter Hutterer:
> > >>> +
On Oct 24, 2011 7:00 PM, "Peter Hutterer" wrote:
>
> No functional changes.
>
> Signed-off-by: Peter Hutterer
Nice cleanup.
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.o
On Sep 17, 2011 9:54 AM, "Chase Douglas"
wrote:
>
> On 09/17/2011 06:53 AM, Gaetan Nadon wrote:
> > On Sat, 2011-09-17 at 06:52 +1000, Peter Hutterer wrote:
> >> On Wed, Sep 14, 2011 at 10:33:56PM -0700, Chase Douglas wrote:
> >> > Signed-off-by: Chase Douglas >
> >> > ---
> >> > configure.ac |
On Wed, Aug 10, 2011 at 8:20 PM, Peter Hutterer
wrote:
> Only use one init path for input devices - through NIDR.
>
> This requires that inp_driver and inp_identifier from the
> XF86ConfInputRec are copied over into the options for NIDR to see them.
>
> Signed-off-by: Peter Hutterer
> ---
> hw/x
ough. Maybe another day.
Anyway,
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
On Mon, Aug 15, 2011 at 9:49 PM, Alan Coopersmith
wrote:
> If libfontenc needs it, it will either be linked with it or
> provide -lz in its pkg-config library requirements.
>
> Signed-off-by: Alan Coopersmith
Good catch.
Reviewed-by: D
On Sat, Aug 13, 2011 at 9:54 AM, Alan Coopersmith
wrote:
> I don't think I've ever used proxymngr or known anyone who has, but some
> distros
> still ship lbxproxy, so I'd slightly prefer leaving support for those that do.
>
> What if you leave Makefile.am doing the substitution (i.e. revert that
- expand testing to test key/value changes and insertion-overwrite
> - remove pointless error: label in input_option_new
> - use xf86AddNewOption in NIDR, it strdups for us
Looks good now. I'll try to review the rest of this series today, but
time is scarce. I really like the idea,
* test environment.
> + */
> + sigaddset(&sig_ign, SIGWINCH);
> +#endif
> +
> init_done = 1;
> }
>
Hah, good catch.
Reviewed-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
, "value");
> + assert(list);
> + opt = input_option_find(list, "key");
> + val = input_option_get_value(opt);
> + assert(strcmp(val, "value") == 0);
> +
> + list = input_option_new(list, "2", "v2");
> + opt
On Wed, Jul 20, 2011 at 4:55 PM, Peter Hutterer
wrote:
> On Wed, Jul 20, 2011 at 08:23:15PM +0200, Lennart Poettering wrote:
>> On Wed, 20.07.11 06:08, Dan Nicholson (dbn.li...@gmail.com) wrote:
>>
>> > > 3) We try to avoid setting pretty names in udev rules, s
On Mon, Jul 18, 2011 at 11:56 AM, Lennart Poettering
wrote:
> On Mon, 18.07.11 19:29, Lennart Poettering (lenn...@poettering.net) wrote:
>
>> If this is supposed to help with the wacom driver, then this is bogus
>> too, as the associated rule in /lib/udev/rules.d/70-wacom.rules actually
>> does ch
On Mon, Jul 18, 2011 at 12:17 PM, Lennart Poettering
wrote:
> Don't enumerate/monitor all devices of the system (since that can be
> quite a few), but limit our search to devices from the "input"
> subsystem, as well as the "tty" subsystem (to cover Wacom tablets).
>
> This should make X start up
val.str = NULL;
> HANDLE_LIST (scrn_display_lst,
> xf86parseDisplaySubSection,
> XF86ConfDisplayPtr);
> }
I also see a couple instances of "free (val.st
On Thu, Jun 30, 2011 at 10:01 AM, Gaetan Nadon wrote:
> This reverts commit 4005df66072ceac175ea71427deb16176262f197.
>
> The patch introduces a couple of issues.
> 1) These tools are apps and conceptually do not depend on Xserver,
> so XORG_CFLAGS should not be used. It included pixman header fil
build bugs
unique to specific modules less likely.
Assuming you put your reasons into the commit message,
Acked-by: Dan Nicholson
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
ot;on"
> MatchOS "Linux"
> - MatchDevicePath "/dev/input/mouse"
> + MatchDevicePath "/dev/input/mouse*"
> Option "Ignore" "on"
> EndSection
Seems right. In fact, I only have /dev/input/mouse0 on my system.
Reviewed
On Fri, Jun 24, 2011 at 7:59 AM, Gaetan Nadon wrote:
> All X.Org documentation in all four formats are build by default.
> An expection was introduced by commit 1ec89689fc77 for libX11.
>
> All documentation can be individually disbaled using
> --without-docs/specs/deve-docs.
> The pdf/ps format c
On Thu, Jun 23, 2011 at 4:21 PM, Gaetan Nadon wrote:
>
> Help me chose the lesser of the two evils. I have no problem keeping the "no".
>
> This works, but looks silly:
>
> XORG_WITH_FOP([1.0],[no])
Sorry, but why is that silly? It just says that you need version 1.0
to build but it's disabled by
On Jun 23, 2011 4:21 PM, "Gaetan Nadon" wrote:
>
> On Thu, 2011-06-23 at 20:50 +0200, Julien Cristau wrote:
>>
>> Also, I'm not sure if looking for a version number is a very good way to
>> check for bugs. People backport bug fixes all the time, and introduce
>> new bugs all the time.
>>
>>
> Rig
On Thu, Jun 23, 2011 at 10:30 AM, Gaetan Nadon wrote:
> On Thu, 2011-06-23 at 11:13 -0400, James Cloos wrote:
>
> Do any dists have any fop versions newer than 0.95?
>
> Gentoo and Debian only have 0.95. I haven't searched for other dists'
> package lists.
>
> I doubt so. Debian has it in the exp
On Thu, Jun 23, 2011 at 5:42 AM, Gaetan Nadon wrote:
> This version fixes a bug in fop 0.95 where fop crashes in the
> presence of some characters like the latin capital sharp s.
>
> The XORG_WITH_FOP macro must be at version 1.15 in order
> to be able to specify the minimum version.
>
> Note the
On Wed, Jun 22, 2011 at 11:01 AM, Keith Packard wrote:
> On Tue, 21 Jun 2011 12:16:39 -0700, Dan Nicholson wrote:
>
>> Alright, git://people.freedesktop.org/~dbn/xserver.git no-libxorg has
>> been updated. I think it's ready to pull now.
>
> I'm getting a merge
On Wed, Jun 22, 2011 at 11:31 AM, Gaetan Nadon wrote:
>
> I am wondering if for libX11 we should not take this opportunity to use
> XORG_WITH_FOP([1.0]). This fixes the problem where fop crashes in the nls
> directory. Few people have 1.0 (not in any Debian distro yet), so it would
> be equivalent
On Wed, Jun 22, 2011 at 10:31 AM, Dirk Wallenstein wrote:
> Signed-off-by: Dirk Wallenstein
> ---
> If the URL is wrong, it should not be a problem to change the directory.
Yeesh, I clearly didn't even test adding these modules in 928e1676.
Reviewed-by:
interface consistent with other commands
> -m4_ifval([$1],[AC_MSG_WARN(Checking for MIN-VERSION is not implemented.)])
> -
> AM_CONDITIONAL([HAVE_XSLTPROC], [test "$have_xsltproc" = yes])
> ]) # XORG_WITH_XSLTPROC
Reviewed-by: Dan Nicholson
la_LIBADD = \
> $(top_builddir)/mi/libmi.la \
> $(top_builddir)/os/libos.la \
> @XORG_LIBS@
> +libxservertest_la_DEPENDENCIES = $(libxservertest_la_LIBADD)
> endif
> endif
>
I think that's right.
Reviewed-by: Dan Nicholson
_
lif test "x$use_perl" = x"no" ; then
> + if test "x$PERL" != "x"; then
> + AC_MSG_WARN([ignoring PERL environment variable since --with-perl=no
> was specified])
> + fi
> + have_perl=no
> +else
> + AC_MSG_ERROR([--with-perl expects &#
t order convention. It's too bad those didn't just
get added from the outset. Oh well, it probably won't hurt anything
significant.
If you want to use the version parameter in any other modules you'll
have to wait for the util-macros version bump, right?
Reviewed-by: Dan Nic
On Wed, Jun 22, 2011 at 7:55 AM, Gaetan Nadon wrote:
> Following the same pattern as XMLTO and friends.
> Allows all modules to use the same interface, variables
> and options to check for perl.
> Perl is used in libX11 and xserver.
>
> Signed-off-by: Gaetan Nadon
Seems like a bit of overkill, b
On Mon, Jun 20, 2011 at 3:42 PM, Peter Hutterer
wrote:
> On Mon, Jun 20, 2011 at 12:25:07PM -0700, Dan Nicholson wrote:
>> On Sun, Jun 19, 2011 at 5:25 PM, Peter Hutterer
>> wrote:
>> > On Fri, Jun 17, 2011 at 04:55:56PM -0700, Dan Nicholson wrote:
>> >> He
On Sat, Jun 18, 2011 at 12:07 AM, Oleh Nykyforchyn wrote:
> Well, if people are so accustomed to single arguments and "|" as "or",
> what about:
>
> 1) adding logical '&' to logical '|';
>
> 2) "regex:|pat|" = "regex:/pat/" = "regex:%pat% = (like sed);
>
> 3) single argument, with multiple co
On Sun, Jun 19, 2011 at 5:25 PM, Peter Hutterer
wrote:
> On Fri, Jun 17, 2011 at 04:55:56PM -0700, Dan Nicholson wrote:
>> Here's a few commits that kill off the libxorg "convenience library of
>> convenience libraries". It provides a nice speedup when you're re
mote-tracking branch 'kibi/master' (2011-06-07 12:31:15 -0700)
are available in the git repository at:
git://people.freedesktop.org/~dbn/xserver.git no-libxorg
Dan Nicholson (3):
Don't use empty source files
xfree86: Move sdksyms generation to ddx toplevel
xfree86:
On Fri, Jun 17, 2011 at 12:48 PM, Oleh R. Nykyforchyn
wrote:
> On Fri, 17 Jun 2011 08:44:58 -0700
> Dan Nicholson wrote:
>
>> This is really confusing. I think it says that if the pattern didn't
>> match the attribute and the pattern was preceded by !, then we ke
On Fri, Jun 17, 2011 at 8:44 AM, Gaetan Nadon wrote:
> On Fri, 2011-06-17 at 16:27 +0100, Daniel Stone wrote:
>
> Are you fine with the following patch, which readds the _DEPENDENCIES
> magic to relink Xorg whenever any of the relevant source files have
> changed? If so, I'll integrate that into m
On Fri, Jun 17, 2011 at 8:27 AM, Daniel Stone wrote:
> Hi,
>
> On Tue, Jun 14, 2011 at 11:02:15PM -0700, Dan Nicholson wrote:
>> libxorg.la served to collect all the Xorg convenience libraries into one
>> massive archive to link into Xorg. This made things easy for symbol
On Fri, Jun 17, 2011 at 6:57 AM, Oleh Nykyforchyn wrote:
> xserver: Introduce negated conditions in Match patterns
>
> When processing input devices, it is useful to determine not only which
> attributes are to be accepted, but also which are to be rejected. Hence we
> introduce
> an '!' prefix,
On Fri, Jun 17, 2011 at 6:53 AM, Oleh Nykyforchyn wrote:
> Further polishing, commit message is the same. Comments, please.
>
> xserver: Use lists and constants for matching modes in Match entries
>
> Current implementation of xf86MatchGroup relies on the assumption that
> mode of matching of an i
The symbols in sdksyms.c cover the entire source tree. In order to make
them resolve when libxorg.la goes away, move the objects from libloader
to Xorg. Unfortunately, this means sdksyms needs to get built again for
the test code.
Signed-off-by: Dan Nicholson
---
hw/xfree86/.gitignore
the objects from the sub-libraries and recombining them.
Instead, the archives are linked directly into Xorg. The order of the
libraries had to be tweaked a bit to make symbols resolve correctly with
the lower level code moving later in the link command.
Signed-off-by: Dan Nicholson
---
hw
When an empty _SOURCES variable is declared, automake will recognize that
only linking is needed.
Signed-off-by: Dan Nicholson
---
dix/.gitignore|1 -
dix/Makefile.am |6 ++
hw/xfree86/.gitignore |2 --
hw/xfree86/Makefile.am
On Tue, Jun 14, 2011 at 9:22 PM, Alan Coopersmith
wrote:
> On 06/14/11 08:20 PM, Dan Nicholson wrote:
>> This seems to work. I had to move sdksyms out of libloader and move the
>> order of a couple libraries around, but everything seems to resolve now.
>> It definitely make
The symbols in sdksyms.c cover the entire source tree. In order to make
them resolve when libxorg.la goes away, move the objects from libloader
to Xorg.
Signed-off-by: Dan Nicholson
---
hw/xfree86/.gitignore |3 +
hw/xfree86/Makefile.am| 13 +-
hw/xfree86/loader/.gitignore
the objects from the sub-libraries and recombining them.
Instead, the archives are linked directly into Xorg. The order of the
libraries had to be tweaked a bit to make symbols resolve correctly with
the lower level code moving later in the link command.
Signed-off-by: Dan Nicholson
---
hw
When an empty _SOURCES variable is declared, automake will recognize that
only linking is needed.
Signed-off-by: Dan Nicholson
---
dix/.gitignore|1 -
dix/Makefile.am |6 ++
hw/xfree86/.gitignore |2 --
hw/xfree86/Makefile.am
This seems to work. I had to move sdksyms out of libloader and move the
order of a couple libraries around, but everything seems to resolve now.
It definitely makes a difference.
Dan Nicholson (3):
Don't use empty source files
xfree86: Move sdksyms generation to ddx toplevel
xfree86: R
On Tue, Jun 14, 2011 at 5:55 PM, Daniel Stone wrote:
> Hi,
>
> On Tue, Jun 14, 2011 at 05:42:44PM -0700, Dan Nicholson wrote:
>> One thing I'd note is that I don't think libxorg.la is needed at all.
>> It just collects a bunch of other convenience libraries togeth
On Tue, Jun 14, 2011 at 4:49 PM, Daniel Stone wrote:
> On Tue, Jun 14, 2011 at 04:02:45PM -0700, Dan Nicholson wrote:
>> On Tue, Jun 14, 2011 at 12:31 PM, Alan Coopersmith
>> wrote:
>> > xserver on the other hand is huge and could probably be much more parallel
>> &
On Tue, Jun 14, 2011 at 12:31 PM, Alan Coopersmith
wrote:
> On 06/14/11 11:11 AM, Jamey Sharp wrote:
>> On Tue, Jun 14, 2011 at 03:53:06PM +0200, Diego Elio Pettenò wrote:
>>> Having all of xorg using non-recursive make with parallel install would
>>> indeed be sweet to avoid spending so much time
On Mon, Jun 13, 2011 at 4:44 PM, Gaetan Nadon wrote:
>
> Commit
> http://cgit.freedesktop.org/xorg/driver/xf86-input-synaptics/commit/?id=39afe69ad7d2258d4043044d1283bd6e311e48da
>
> build: collapse all Makefile.am files into a single non-recursive one.
>
> With this change, the whole of the build
On Tue, Jun 14, 2011 at 6:45 AM, Gaetan Nadon wrote:
> On Tue, 2011-06-14 at 10:56 +0200, Diego Elio Pettenò wrote:
>
> Il giorno lun, 13/06/2011 alle 19.43 -0400, Gaetan Nadon ha scritto:
>>
>> Using m4 to check for macro defined by the xserver to ensure it is
>> installed
>> will work as long as
On Tue, Jun 14, 2011 at 1:41 AM, Diego Elio Pettenò wrote:
> Il giorno lun, 13/06/2011 alle 19.43 -0400, Gaetan Nadon ha scritto:
>>
>> This directory is created when the source code is extracted from git.
>> There are 175 other man pages directories in X.Org and none are
>> created from the makef
On Tue, Jun 14, 2011 at 1:56 AM, Diego Elio Pettenò wrote:
> Il giorno lun, 13/06/2011 alle 19.43 -0400, Gaetan Nadon ha scritto:
>>
>> Using m4 to check for macro defined by the xserver to ensure it is
>> installed
>> will work as long as the macro is not removed. The preferred way of
>> checking
dows):
> rm -f $<; \
> $(MAKE) $(AM_MAKEFLAGS) $< || exit 1; \
> fi
> - $(AM_V_GEN)mv -f $(@:.man=.libmansuffix) $@
> + $(AM_V_at)mv -f $(@:.man=.libmansuffix) $@
> endif
AM_V_at silences th
)$(MAKE) $(AM_MAKEFLAGS) $< || exit 1; \
> + rm -f $<; \
> + $(MAKE) $(AM_MAKEFLAGS) $< || exit 1; \
> fi
> $(AM_V_GEN)mv -f $(@:.man=.libmansuffix) $@
> endif
Reviewed-by: Dan Nicholson
___
xorg-devel@lists
sltproc to get the man page.
The rest of the patch looks nice and even contains some more goodness
in the xsl from the version I looked at the other day. The only
further improvement I'd suggest is using more embedded groff to format
the tables. B
On Sat, Jun 4, 2011 at 10:48 PM, Oleh Nykyforchyn wrote:
> Use lists and constants for matching modes in Match entries
You're leaving out some critical information here. Particularly that
each Match entry would take multiple arguments. I'm going to ignore
those details for now since I don't think
On Jun 5, 2011 4:48 AM, "Julien Cristau" wrote:
>
> On Sat, Jun 4, 2011 at 09:25:07 -0400, Gaetan Nadon wrote:
>
> > The DocBook/XML olink databases (this is what those files are) are
> > installed in the location provided by the specs/Makefile variable
> > sgmldbsdir which is set to $(XORG_SGML_
On Fri, Jun 3, 2011 at 1:30 AM, Peter Hutterer wrote:
> On Thu, Jun 02, 2011 at 05:53:14PM +0300, Oleh Nykyforchyn wrote:
>> Please consider a final version of the patch.
>>
>> ---
>>
>> Multiple matching modes selection (incl. regex) support in Match* statements
>>
>>
On Wed, Jun 1, 2011 at 5:34 AM, Mark Kettenis wrote:
>> From: Gaetan Nadon
>> Date: Wed, 01 Jun 2011 07:42:32 -0400
>>
>> > AC_LANG_DEFINES_PROVIDED seems to be only in newer, GPLv3, versions of au=
>> toconf ... is there a way around this?
>>
>> I looked at it yesterday, using AC_LANG_SOURCE sho
On Thu, May 26, 2011 at 10:56 PM, Peter Hutterer
wrote:
> On Wed, May 25, 2011 at 03:08:32PM +0300, Oleh Nykyforchyn wrote:
>> On Wed, 25 May 2011 16:14:42 +1000
>> Peter Hutterer wrote:
>>
>> > I believe | is used frequently in regular expressions, so using it as
>> > decision over regex or simp
1 - 100 of 872 matches
Mail list logo
