It has come to my attention that, on -current PIE architectures
(amd64, mips64*, sparc64),
gcc -static will produce a binary that's "generally" static, but that
will require a working ld.so to run.
This can be a problem if you want to compile static binaries for chroot,
or say, you're trying to
On 4.11.2012. 15:52, Brad Smith wrote:
> On Sat, Nov 03, 2012 at 09:49:02PM +, Christian Weisgerber wrote:
>> Like bge(4), we previously couldn't enable TCP/UDP transmit checksum
>> offload on em(4). We can now.
>>
>> Works fine here on
>> em0 at pci1 dev 1 function 0 "Intel PRO/1000MT (8254
This is somewhat nasty. bsdmake used to have a simplistic approach to
whitespace (kill everything at end of line).
Unfortunately, that's not how other makes work. nor is it how posix
finally standardized it.
Specifically, for variable definitions: make should strip spaces around
the equal sign (a
Can someone, please, point me to the right cvs URL for those changes.
Thanks.
On 3 nov 2012, at 22:41, Christian Weisgerber wrote:
> Henning's epic rewrite of the checksum handling has fixed
> On 3 nov 2012, at 22:41, Christian Weisgerber wrote:
>
> > Henning's epic rewrite of the checksum handling has fixed
On 2012/11/05 18:57, mxb wrote:
> Can someone, please, point me to the right cvs URL for those changes.
> Thanks.
http://anoncvs.estpak.ee/cgi-bin/cgit/openbsd-src/commit/?id=e
On Sun, Nov 04, 2012 at 12:22:31AM +0100, Christian Weisgerber wrote:
> Brad Smith:
>
> > Here is another revision but disabling the UDP checksum offload.
> > There is a bug that results in some UDP packets having a 0 checksum.
>
> Oh, right. I assume you are referencing this FreeBSD commit?
> h
Hope there'll be some numbers over how fast the stack became.
Great work!
On 5 nov 2012, at 19:23, Stuart Henderson wrote:
>> On 3 nov 2012, at 22:41, Christian Weisgerber wrote:
>>
>>> Henning's epic rewrite of the checksum handling has fixed
>
> On 2012/11/05 18:57, mxb wrote:
>> Can some
Howdy-
I'm resending this diff from last week. When windows match more than
one autogroup then the last defined autogroup that matches both name
and class wins. If no autogroup matches both name and class then the
last defined autogroup that matches only class wins.
This allows for different co
When more than one autogroup matches a window the
last defined autogroup that matches both class and
name wins. If no autogroup matches both then the
last defined autogroup that only matches class wins.
Index: cwm/cwmrc.5
===
--- cwm
Previously, quotes were required in autogroup configuration
using the windowname,windowclass syntax because the grammar
does not allow commas in strings. This change removes that
requirement by explicitly matching literal commas. Quotes
are still required around name or class values that contain
Allow users to configure autogroups as:
autogroup X windowclass windowname
The idea is to better reflect the fact that classes match
more generally than names.
Index: cwm/group.c
===
--- cwm.orig/group.c2012-11-02 17:45:39.85172
Howdy-
I sent this to okan@ last week but forgot to CC the list in
case others have an opinion as well. Basically, I think it
should be documented that the values for class and name to
use with autogroup are both set in the WM_CLASS property of
existing windows.
Best,
Kent
Explicitly mention that the values of class and name for use with autogroup
can both be obtained from the WM_CLASS property of existing windows.
Index: cwm/cwmrc.5
===
--- cwm.orig/cwmrc.52012-11-05 10:25:37.777426409 -0600
+++ cw
Howdy-
I originally sent these diffs as part of one big diff last week along
with an unrelated change to autogroup match ordering. I've separated
them out into their own series because they're independent of the match
order change, and I've also split them into three separate steps to
(hopefully)
Make group number the second parameter to match the autogroup
configuration syntax. Split val into class and name, respectively.
No functional change, but this paves the way for some future
changes.
Index: cwm/group.c
===
--- cwm.ori
Hey, dude-
> This breaks how name_to_kbfunc is currently being used;
>
>for (iter = 0; iter < nitems(name_to_kbfunc); iter++) {
> if (strcmp(name_to_kbfunc[iter]->tag, ...
>
> I think you can see where this goes wrong with your patch
> applied (hint: passing NULL to strcmp is not advis
Howdy-
Here is an updated diff that fixes a regression introduced in the first
diff (pointed out by Thomas Pfaff) and also includes the manpage bits.
Again, feedback appreciated. Thanks in advance!
Best,
Kent
Useful for those times you want to use an unbound function, but even
when the function is bound to something you haven't memorized yet it
can be faster than lookup up the keybinding in the manual. Bound to
CM-/ by default.
Index: cwm/calmwm.h
==
On 5.11.2012. 14:38, Hrvoje Popovski wrote:
>
> Hello,
>
> it's working here.
Please ignore "working here" part. It is enabled here.
Will play with combination of 82546GB and 82541GI interfaces.
In the unlikely event that a NULL device is passed to the
priv_pcap_live() function, it will just set the error buffer and
continue, causing a segfault.
The diff below fixes this bug by ensuring that the function returns
NULL in this situation.
Comments/OK?
Lawrence
Index: privsep_pcap.c
=
Hi,
Here is a small rewrite for consistency in vio_attach() error handling path,
also a fix for dmesg message "vlan2: initialized with non-standard mtu
1496 (parent vio0)",
while creating vlan
Index: if_vio.c
===
RCS file: /cvs/src/sy
21 matches
Mail list logo
