re: current/netbsd-10: panic() when MULTIPROCESSOR is disabled

[matthew green]

"Jonathan A. Kollasch" writes:
> On Wed, Apr 24, 2024 at 05:51:32PM +0200, Paolo Pisati wrote:
> > I'm slimming down a GENERIC i386 kernel[1] for an embedded board[2], when i
> > discovered that turning off MULTIPROCESSOR, leads to:
>
> There's a reason it's in std.i386 and std.amd64 and not GENERIC.
> Turning MULTIPROCESSOR off on x86 bare-metal hasn't been expected
> to work for just under 16 years now.  Does it really save that much
> space for you?

it worked for me not too long ago, and not too long before that,
though it normally does take some code patching to compile.

this is the first time i've heard of it not working at all..


.mrg.

re: NetBSD 9.2 (STABLE) kernel hangs without panic or ddb

[matthew green]

i wonder if this is lack of buf_destroy() in rf_FreeDiskQueueData(),
to match the buf_init() in rf_CreateDiskQueueData().

can you try this patch?


.mrg.


Index: rf_diskqueue.c
===
RCS file: /cvsroot/src/sys/dev/raidframe/rf_diskqueue.c,v
retrieving revision 1.55.4.1
diff -p -u -r1.55.4.1 rf_diskqueue.c
--- rf_diskqueue.c  19 Oct 2021 10:55:15 -  1.55.4.1
+++ rf_diskqueue.c  13 Dec 2021 04:45:15 -
@@ -413,6 +413,9 @@ rf_CreateDiskQueueData(RF_IoType_t typ, 
 void
 rf_FreeDiskQueueData(RF_DiskQueueData_t *p)
 {
+
+   buf_destroy(p->bp);
+
pool_put(_pools.bufio, p->bp);
pool_put(_pools.dqd, p);
 }

re: httpd, cgi, ikiwiki, and me.

[matthew green]

> Since upgrading to 9.99.81, httpd doesn't launch a cgi that has
> worked for years, to wit /var/www/cgi-bin/ikiwiki.cgi.
> 
> My rc.conf looks at it always has, at least in regard to httpd:
> 
> httpd=YES httpd_flags="-I "
>httpd_wwwdir="/var/www"
> 

are there any logs or anything more?  can you look in the
(ftp) logs or run ktrace?  there's also a "debug" build
one can do that may help if these don't work.

nothing should have broken.

thanks.


.mrg.

re: My first look at nouveau in xorg

[matthew green]

glad you got it working.


FWIW, in general, modern Xorg server runs with no xorg.conf fine
in almost all cases.  the only reasons i use one are to choose
specific options at start up time that aren't easily available,
but the vast majority of the sections aren't needed.

eg, to get a core dump vs it complaining this is all that you
need in xorg.conf:

---
Section "ServerFlags"
Option "NoTrapSignals" "true"
EndSection
---

HTH,


.mrg.

re: osabi in -current in some difficulty?

[matthew green]

> I've confirmed with upstream that this library is just obsolete. The
> code to support it in the X server was removed 12 years ago.
> 
> I've removed lbxproxy and proxymngr from pkgsrc.

last i saw, pkgsrc was still had the Xprint components,
which have been obsolete a similar or even longer time.
(i removed lbx and Xprint stuff from xsrc almost 10
years ago.)

maybe there is more you can delete :-)


.mrg.

re: bozohttpd minor fixes to man page

[matthew green]

Swift Griggs writes:
> 
> I like NetBSD's httpd. I noticed a couple of minor inconsistencies in the 
> bozohttpd(8) manual page. Where should I report these?

thanks.  directly to me is fine, and is send-pr from netbsd or from
the send-pr form on gnats.netbsd.org.

> * The -v option appears twice in the options summary. It's shown as both a 
>   flag and a switch that takes options. They can't be both right. 

this got fixed late last year by shm@.

> * The -V option is documented in the manual page, but does not appear in 
>   the options summary block at the top of the manual page. It's also 
>   unclear if "slashdir" is an option for -V or if the text refers to the 
>   "slashdir" given as the document root.

OK, i've cleared this one some.

> * -V is also not documented in the usage when you get help directly from 
>   the binary (ie.. just run /usr/libexec/httpd to see what I mean). 

i also fixed this.

thanks!


.mrg.