The max heap size was set to 2G.
On Oct 9, 2010, at 21:37, Matt Dew wrote:
> On Sat, Oct 9, 2010 at 7:26 PM, Jeremy Huddleston wrote:
>> As mentioned earlier in the thread, it fails to build without upping the
>> heap...
>>
>> The point is that it should take NO time to build because it should
On Sat, Oct 9, 2010 at 7:26 PM, Jeremy Huddleston wrote:
> As mentioned earlier in the thread, it fails to build without upping the
> heap...
>
> The point is that it should take NO time to build because it shouldn't be
> rebuilding the documentation every time.
True, but when it does build, it
Reviewed-by: Jeremy Huddleston
On Oct 9, 2010, at 16:00, Dan Nicholson wrote:
> Make expects prerequisites to be files with valid timestamps, and
> directories are treated as always being out of date. Thus, any targets
> depending on directories will always be rebuilt.
>
> Instead, the doc rule
On Sat, Oct 9, 2010 at 6:28 PM, Jeremy Huddleston wrote:
>
> On Oct 9, 2010, at 17:25, Matt Dew wrote:
Plus, make install re-gens all of the docs. The dependencies are wrong.
>>
>> While there's been a massive amount of work already done on the docs,
>> there is still a massive amount to do.
On Oct 9, 2010, at 17:25, Matt Dew wrote:
>>> Plus, make install re-gens all of the docs. The dependencies are wrong.
>
> While there's been a massive amount of work already done on the docs,
> there is still a massive amount to do. I'd be shocked if people
> didn't find problems at this point
As mentioned earlier in the thread, it fails to build without upping the heap...
The point is that it should take NO time to build because it shouldn't be
rebuilding the documentation every time.
On Oct 9, 2010, at 16:53, Matt Dew wrote:
> Jeremy,
> Does the build time go down if you up Java's
fop is used to generate the pdf and ps formats of the documentation.
This can significantly slow down the build, especially when creating all
the compose key charts. Since all the docs really only ever need to be
created for the tarball, set the default to 'no', but ensure that
they're built during
The typical default of enabling documentation if the correct tool is
found is usually the right thing to do. However, some packages such as
Xlib have huge amounts of documentation that really only need to be
built once to put into the tarball. Allow packages to specify the
default by passing an arg
Handling the optional documentation parameters with m4 allows the help
string to correctly represent the default.
Signed-off-by: Dan Nicholson
---
xorg-macros.m4.in | 30 --
1 files changed, 12 insertions(+), 18 deletions(-)
diff --git a/xorg-macros.m4.in b/xorg-ma
Here's a few patches that allow fop to be used but defaulted to off in
libX11. The doc macros have been updated to take an optional argument
setting the default that will be used. XORG_WITH_FOP is then passed 'no'
from libX11.
I don't believe this strictly requires a new macro version since
autoco
On Fri, Oct 8, 2010 at 5:31 PM, Jeremy Huddleston wrote:
>
> On Oct 8, 2010, at 16:22, James Cloos wrote:
>
>> Now that the docs are in the modules they document, building from git
>> has become fragile.
That's exactly opposite of the goal. :) With fewer tools required,
it should be less fragi
Jeremy,
Does the build time go down if you up Java's heap for fop?
export FOP_OPTS='-Xmx512m
Matt
On Thu, Oct 7, 2010 at 5:26 PM, Jeremy Huddleston wrote:
> Where did this go? I thought I saw a patch, but it's not pushed, and now I
> can't find it. libX11 takes 2h20m to build/install on my
Make expects prerequisites to be files with valid timestamps, and
directories are treated as always being out of date. Thus, any targets
depending on directories will always be rebuilt.
Instead, the doc rules are changed to always create the target's leading
directory. This should prevent the docu
On Sat, Oct 9, 2010 at 20:47:35 +0400, Mikhail Gusarov wrote:
> This macro was undefined since 7.0. Generated by unifdef -UPATHETICCPP
>
> Signed-off-by: Mikhail Gusarov
Reviewed-by: Julien Cristau
Cheers,
Julien
___
xorg-devel@lists.x.org: X.Org d
This macro was undefined since 7.0. Generated by unifdef -UPATHETICCPP
Signed-off-by: Mikhail Gusarov
---
xrdb.c | 105
1 files changed, 0 insertions(+), 105 deletions(-)
diff --git a/xrdb.c b/xrdb.c
index 21005c0..3c15faa 100644
Takashi Iwai schrieb:
> Signed-off-by: Takashi Iwai
> ---
> include/synaptics-properties.h | 10 ++
> src/keymap.c | 58 ++--
> src/properties.c | 63
>
> src/synaptics.c
On 10/08/2010 07:22 PM, Takashi Iwai wrote:
> Signed-off-by: Takashi Iwai
> ---
> src/eventcomm.c| 56 +++--
> src/synaptics.c| 136
>
> src/synapticsstr.h | 12 +
> src/synproto.h |6 ++
> 4 files cha
Hi Takashi,
> diff --git a/src/eventcomm.c b/src/eventcomm.c
> index 85dfd09..fc5055b 100644
> --- a/src/eventcomm.c
> +++ b/src/eventcomm.c
> @@ -269,6 +269,12 @@ event_query_axis_ranges(InputInfoPtr pInfo)
> }
>
> xf86Msg(X_PROBED, "%s: buttons:%s\n", pInfo->name, buf);
> +
> +
Hi Jeremy,
On Fri, Oct 8, 2010 at 16:25:18 -0700, Jeremy Huddleston wrote:
> I haven't heard any planes crashing or massive electrical grid
> failures since 1.9.1rc1 last wek, and I haven't seen anything
> terribly urgent land on master or in the list, so my plan is to
> release this as 1.9.1 ne
19 matches
Mail list logo