-CURRENT slowdown in last 2 weeks

2001-02-21 Thread Andrea Campi
I am noticing a severe slowdown on my -CURRENT system. It actually started after Feb [3-5] changes in intrupt handling, but I didn't really notice until I run a make world (which I delayed doing because of, well you guess, the libc breakage). When I say severe I mean make buildworld takes x3

Re: latest alpha breakage

2001-02-21 Thread Matthew Jacob
I cvsup every day in periodic/daily- maybe I just missed this one. This has been dead for several days- sorryu I guess I missed your mail about it Umm... yup... sorry for the noise Matthew Jacob writes: === libcrypto ( echo "#ifndef MK1MF_BUILD"; echo " /* auto-generated by

Re: Building procedure of krb5 is broken

2001-02-21 Thread Jun Kuriyama
At 20 Feb 2001 09:44:01 GMT, Makoto MATSUSHITA wrote: It seems that add 'CFLAGS+=-I${KRB5OBJDIR}' to src/kerberos5/lib/libgssapi/Makefile is required to fix this, since krb5_err.h is not in ${KRB5DIR}/lib/krb5. In my environment, buildworld was finished with attached patch. I don't know KRB4

Re: Building procedure of krb5 is broken

2001-02-21 Thread Jacques A. Vidrine
On Wed, Feb 21, 2001 at 10:06:29PM +0900, Jun Kuriyama wrote: At 20 Feb 2001 09:44:01 GMT, Makoto MATSUSHITA wrote: It seems that add 'CFLAGS+=-I${KRB5OBJDIR}' to src/kerberos5/lib/libgssapi/Makefile is required to fix this, since krb5_err.h is not in ${KRB5DIR}/lib/krb5. In my

Re: Building procedure of krb5 is broken

2001-02-21 Thread Assar Westerlund
"Jacques A. Vidrine" [EMAIL PROTECTED] writes: On Wed, Feb 21, 2001 at 10:06:29PM +0900, Jun Kuriyama wrote: So if MAKE_KERBEROS5=YES is set in /etc/make.conf, buildworld is failed because krb.h is not found. Last I looked (before assar's import), MAKE_KERBEROS5=yes required that you

Re: Building procedure of krb5 is broken

2001-02-21 Thread Jun Kuriyama
At 21 Feb 2001 14:19:28 +0100, Assar Westerlund wrote: Which is bogus. I think Jun Kuriyama's patch is correct and I'm trying it now, but buildworld takes a little while to complete :-) Just FYI, I've tested buildworld with that patch in environments below. :-) (1) MAKE_KERBEROS4=YES (2)

Re: Building procedure of krb5 is broken

2001-02-21 Thread Makoto MATSUSHITA
kuriyama In my environment, buildworld was finished with attached kuriyama patch. I don't know KRB4 should be defined in this file or kuriyama not. Sorry I should say when building of krb5 is failed; make buildworld seems OK, but 'make release' will fail during 'release.2' target; there is no

Re: Building procedure of krb5 is broken

2001-02-21 Thread John Hay
Jun Kuriyama [EMAIL PROTECTED] writes: Just FYI, I've tested buildworld with that patch in environments below. :-) (1) MAKE_KERBEROS4=YES (2) MAKE_KERBEROS4=YES and MAKE_KERBEROS5=YES (3) MAKE_KERBEROS5=YES (4) No MAKE_KERBEROS* Thanks for your testing. I've commited the fix.

Re: -CURRENT slowdown in last 2 weeks

2001-02-21 Thread Pierre Y. Dampure
Andrea Campi wrote: I am noticing a severe slowdown on my -CURRENT system. It actually started after Feb [3-5] changes in intrupt handling, but I didn't really notice until I run a make world (which I delayed doing because of, well you guess, the libc breakage). When I say severe I mean make

kernfs

2001-02-21 Thread Daniel C. Sobral
It seems the error for kernfs is activating a couple of other warnings: WARNING: unknown option `KERNFS' removed from /usr/obj/home/src/sys/DCS/opt_dont use.h WARNING: option `FFS' moved from opt_ffs.h to opt_dontuse.h WARNING: unknown option `FFS_ROOT' removed from

Problems syncing disks in recent -current (since 19th)

2001-02-21 Thread Josef Karthauser
Hi Kirk, A number of us have problem reliably syncing disks with softupdates in recent -current from about the 19th. Is it possible that you broke something with your recent commit? Cheers, Joe PGP signature

MP-safe midi(4) and sequencer callout priority

2001-02-21 Thread Seigo Tanimura
The patch at http://people.FreeBSD.org/~tanimura/patches/midimutex.diff.gz makes midi(4) MP-safe. With this patch, my kernel with WITNESS and INVARIANTS plays my favorite midi tunes just fine, except that the callout used by the midi sequencer gets delayed under a heavy load. As the sequencer

RE: -CURRENT slowdown in last 2 weeks

2001-02-21 Thread John Baldwin
On 20-Feb-01 Andrea Campi wrote: I am noticing a severe slowdown on my -CURRENT system. It actually started after Feb [3-5] changes in intrupt handling, but I didn't really notice until I run a make world (which I delayed doing because of, well you guess, the libc breakage). When I say