Re: CVS commit: src/sys/net80211

2013-03-29 Thread Christos Zoulas 
In article ,
NONAKA Kimihiro   wrote:
>Hi,
>
>2013/3/30 Christos Zoulas :
>
>> Module Name:src
>> Committed By:   christos
>> Date:   Sat Mar 30 01:05:49 UTC 2013
>>
>> Modified Files:
>> src/sys/net80211: ieee80211.h ieee80211_proto.h
>>
>> Log Message:
>> EDCA and QOS additions from OpenBSD
>
>You have to add the following lines. It seems same as WME_AC_ *.
>
>-
> /*
>+ * EDCA Access Categories.
>+ */
>+enum ieee80211_edca_ac {
>+ EDCA_AC_BK  = 1, /* Background */
>+ EDCA_AC_BE  = 0, /* Best Effort */
>+ EDCA_AC_VI  = 2, /* Video */
>+ EDCA_AC_VO  = 3 /* Voice */
>+};
>+#define EDCA_NUM_AC 4
>+
>-
>
>WME_AC_ *
>- sys/net80211/ieee80211.h
>#define WME_AC_BE 0 /* best effort */
>#define WME_AC_BK 1 /* background */
>#define WME_AC_VI 2 /* video */
>#define WME_AC_VO 3 /* voice */
>-

They are there? I should gc the old one I guess.

christos

Re: CVS commit: src/sys/net80211

2013-03-29 Thread NONAKA Kimihiro 
Hi,

2013/3/30 Christos Zoulas :

> Module Name:src
> Committed By:   christos
> Date:   Sat Mar 30 01:05:49 UTC 2013
>
> Modified Files:
> src/sys/net80211: ieee80211.h ieee80211_proto.h
>
> Log Message:
> EDCA and QOS additions from OpenBSD

You have to add the following lines. It seems same as WME_AC_ *.

-
 /*
+ * EDCA Access Categories.
+ */
+enum ieee80211_edca_ac {
+ EDCA_AC_BK  = 1, /* Background */
+ EDCA_AC_BE  = 0, /* Best Effort */
+ EDCA_AC_VI  = 2, /* Video */
+ EDCA_AC_VO  = 3 /* Voice */
+};
+#define EDCA_NUM_AC 4
+
-

WME_AC_ *
- sys/net80211/ieee80211.h
#define WME_AC_BE 0 /* best effort */
#define WME_AC_BK 1 /* background */
#define WME_AC_VI 2 /* video */
#define WME_AC_VO 3 /* voice */
-

Regards,
--
NONAKA Kimihiro

Re: CVS commit: src/usr.bin/tftp

2013-03-29 Thread Christos Zoulas 
In article <20130329185322.aab9717...@cvs.netbsd.org>,
Radoslaw Kujawa  wrote:
>-=-=-=-=-=-
>
>Module Name:   src
>Committed By:  rkujawa
>Date:  Fri Mar 29 18:53:21 UTC 2013
>
>Modified Files:
>   src/usr.bin/tftp: Makefile
>
>Log Message:
>Work around "variable might be clobbered by longjmp" gcc warning when
>compiling with CPUFLAGS=-march=armv7. Should be fixed properly one
>day...

Have you tried volatile? Can you provide more details? No matter what
this should be documented in HACKS if indeed it is a compiler bug, or
just make the variable volatile if it is not.

christos

Re: CVS commit: src/lib/libpthread

2013-03-29 Thread Christos Zoulas 
In article <20130329032804.23b1514a...@mail.netbsd.org>,
YAMAMOTO Takashi  wrote:
>
>"mono" escaping its scope?

Yes, I already added it to lwp_park to avoid the hack but not using
the syscall yet until we decide what we need to pass to lwp_park
to avoid priority inversion and fix the races in pthread_cond_*.

christos

Re: CVS commit: [agc-symver] src/lib/libisns

2013-03-29 Thread Christos Zoulas 
In article <20130329032650.gb24...@nef.pbox.org>,
Alistair Crooks   wrote:
>
>That's what FreeBSD used (although, admittedly, they also had
>Symbol.map and Versions.def, which we do without).

ok.

>And maybe, in the spirit of "shlib_version", "version_map", or
>"version_info" or some other name is better?

let's vote!

>Anyway, let me get them all added with the same name on the branch,
>and I can change them all in one fell swoop before merging the branch.

Sounds good.

christos