Re: FW: Filesystem gets a huge performance boost
On Thu, Apr 19, 2001 at 03:46:39PM +0300, Maxim Sobolev wrote:
What do you think about attached patch?
-Maxim
Index: Makefile
===
RCS file: /home/ncvs/src/etc/Makefile,v
retrieving revision 1.248
diff -d -u -r1.248 Makefile
--- Makefile 2001/03/29 14:02:59 1.248
+++ Makefile 2001/04/19 12:44:09
@@ -15,8 +15,8 @@
printcap profile protocols \
rc rc.atm rc.devfs rc.diskless1 rc.diskless2 rc.firewall rc.firewall6 \
rc.isdn rc.network rc.network6 rc.pccard rc.serial rc.shutdown \
- rc.syscons rc.sysctl remote rpc security services shells syslog.conf \
- usbd.conf \
+ rc.syscons rc.sysctl remote rpc security services shells sysctl.conf \
+ syslog.conf usbd.conf \
etc.${MACHINE_ARCH}/disktab \
etc.${MACHINE_ARCH}/rc.${MACHINE_ARCH} \
etc.${MACHINE_ARCH}/ttys \
Index: sysctl.conf
===
RCS file: /home/ncvs/src/etc/sysctl.conf,v
retrieving revision 1.1
diff -d -u -r1.1 sysctl.conf
--- sysctl.conf 2000/07/27 22:53:42 1.1
+++ sysctl.conf 2001/04/19 12:44:09
@@ -3,3 +3,6 @@
# This file is read when going to multi-user and its contents piped thru
# ``sysctl -w'' to adjust kernel values. ``man 5 sysctl.conf'' for details.
#
+
+vfs.vmiodirenable=0 # Set to 1 to enable a new directory allocation policy
+ # (codenamed "dirpref")
Oops, the comment is wrong. Dirpref is totally another thing which has
nothing to do with that sysctl.
--
Alex Kapranoff, Voice: +7(0832)791845
We've lived 2608 hours in the brand new millenium...
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
Re: ACL support in 5.0-CURRENT (was: cvs commit: src/sys/conf files options src/sys/ufs/ufs acl.h ufs_acl.c ufs_vnops.c (fwd))
On Tue, Mar 27, 2001 at 02:14:38PM -0500, Robert Watson wrote: After the EA backing files are initialized, you can manually start the EA's on the file system, or if UFS_EXTATTR_AUTOSTART is enabled, simply remount the file systems, or reboot the system. To manually start and enable the attributes for the root file system, do the following: extattrctl start / extattrctl enable / posix1e.acl_access /.attribute/system/posix1e.acl_access extattrctl enable / posix1e.acl_default /.attribute/system/posix1e.acl_default When I did this several days ago, I had to actually issue: extattrctl enable / system posix1e.acl_access /.attribute/system/posix1e.acl_access extattrctl enable / system posix1e.acl_default /.attribute/system/posix1e.acl_default ^^ Note namespace argument. Your directions in RAEDME probably got out of sync with `extattrctl' implementaion which now requires attrnamesspace argument. Manpage is accurate though. -- Alex Kapranoff, Voice: +7(0832)791845 We've lived 12 weeks in the brand new millenium... To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: pcm driver and DEVFS
On Fri, Feb 02, 2001 at 04:11:29PM +0900, Yoshihiro Koya wrote: Hello, I did make world a couple days ago. The system was built from cvsup'd source on Jan 30: -- elf make world started on Tue Jan 30 06:23:38 JST 2001 -- The system uses DEVFS. But I have some issue around sound drivers. I usually use mpg123(Version 0.59r (1999/Jun/15)) or x11amp(version 0.8). Before using DEVFS, I was able to adjust sound volume in the sophisticated manner. But, after installing DEVFS, I wasnt adjust sound volume. It might be difficult to run x11amp with DEVFS. On the other hand, mpg123 works. But, its sound is too loud. Added to this, before install DEVFS, I found /dev/dsp1 or /dev/dsp0 in /dev. But I only found the different kind of files: % ls /dev [skip] The files /dev/dsp1.0 and /dev/dsp1.1 are new to me. Of course, I tried to do % x11amp -e /dev/dsp1.0 % x11amp -e /dev/dsp1.1 % x11amp -e /dev/dspW1.0 % x11amp -e /dev/dspW1.1 But in vain. Does some have solution or suggestion? Yep. I have these in my /etc/rc.devfs: = ln -fs /dev/audio1.0 /dev/audio ln -fs /dev/dsp1.0 /dev/dsp ln -fs /dev/mixer1 /dev/mixer = This produces almost exactly same environment both with DEVFS and without. -- Alex Kapranoff, Voice: +7(0832)791845 We've lived 32 days in the brand new millenium... To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: cc1 gets segmentation faults
Huh, I removed '-O' switch from ${CFLAGS} and managed to rebuild
and reinstall gcc. This new (in fact the same) gcc now works fine both
with '-O' and without it. Looks like a pilot error. Strange, anyway.
Sorry to bother you all.
On Sun, Jan 21, 2001 at 02:50:48PM +0300, Alex Kapranoff wrote:
I have -CURRENT on a UP P166 box.
I've just managed to build a new kernel and world (cvsuped yesterday)
and installed them both. Now I get:
cc: Internal compiler error: program cc1 got fatal signal 11
*** Error code 1
Stop in /usr/home/alex/work/own.
whenever I want to compile something not so trivial as `hello world'.
So, am I stuck in this situation? I cannot build neither kernel nor
gcc now. Are there any ways out? Unfortunately, I don't backup system
binaries, libs or headers.
How can I debug this? BTW, other big programs such as INN, apache +
mod_php4 work like a charm.
--
Alex Kapranoff, Voice: +7(0832)791845
We've lived 2 weeks in the brand new millenium...
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
Lots of page faults
I have a fairly recent CURRENT: FreeBSD kapran.bitmcnit.bryansk.su 5.0-CURRENT FreeBSD 5.0-CURRENT #0: Sat Dec 30 12:41:53 MSK 2000 [EMAIL PROTECTED]:/usr/src/sys/compile/KAPRAN i386 And what I noticed is a massive slowdown when running make(1). It seems that I am going to finish a new `buildkernel' in a millennium or two. Additional symptoms include very high system CPU state percentage and a lot of page faults. vmstat(8) shows something like (while building world): procs memory pagedisks faults cpu r b w avm fre flt re pi po fr sr ad0 ac0 in sy cs us sy id 2 0 0 24216 8444 235 1 1 0 215 23 0 0 810 805 213 15 43 42 2 0 0 23776 8508 356 0 0 0 356 0 16 0 296 879 216 18 82 1 2 0 0 24252 8232 382 0 0 0 313 0 12 0 299 987 240 21 79 0 2 0 0 23664 8528 328 0 0 0 407 0 11 0 289 711 216 9 91 0 2 0 0 23816 8424 314 0 0 0 304 0 7 0 285 890 209 21 75 5 2 0 0 23792 8440 362 0 0 0 338 0 12 0 292 979 222 24 76 0 2 0 0 24284 8144 372 0 0 0 285 0 12 0 291 981 227 24 74 2 4 0 0 22724 8964 140 0 0 0 401 0 71 0 363 463 340 2 62 36 2 1 0 22392 8472 429 0 0 0 279 0 11 0 297 834 215 17 82 2 4 0 0 22628 8312 334 0 0 0 297 0 2 0 286 869 213 15 85 0 From time to time I get a core dump from `cc1' or `cpp' or even `as' too. Is my RAM rotting or what? -- Alex Kapranoff, Voice: +7(0832)791845 We've lived 2 weeks in the brand new millenium... To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: Current stalls...(now also panic)
On Fri, Dec 29, 2000 at 10:57:10PM +0100, Szilveszter Adam wrote: Hi! Maybe this is related, maybe not... I upgraded to the latest CURRENT available this morning and now I also see occasional (albeit short) hangs sometimes, although the machine seems to be responsive otherwise. But, not only that, I also got a cool panic while trying to do some stuff (like compiling the docs) and pressing Ctrl-Z in another tty. No X, no nothing. I was dropped into the debugger, but since I am not a ddb artist, I tried to avoid to type the whole trace by hand and in the process managed to reboot the box... oh well. Next time I will know. But, the panic message was this: panic: blockable mtx_enter() of lockmgr interlock when not legal @ ../../kern/kern_lock.c:247 Maybe this rings a bell with someone. The error was appearently caught by WITNESS, which I also have enabled in my kernel (albeit without MUTEX_DEBUG, because *that* really made it impossible to do anything sensible on the machine...) I see the same panic. I managed to collect some info in kern/23935. http://www.freebsd.org/cgi/query-pr.cgi?pr=23935 I can reliably reproduce it. -- Alex Kapranoff, Voice: +7(0832)791845 36 hours before the brand new millenium... To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: possible fix for gdb hanging the kernel
On Fri, Dec 29, 2000 at 04:19:38PM -0800, Alfred Perlstein wrote:
I'd appreciate it if those who are having issues with gdb were to
try this patch and let me know if it fixes things.
Sorry, but gdb keeps hanging my box after this patch.
In fact, the behaviour didn't changed.
Index: sys_process.c
===
RCS file: /home/ncvs/src/sys/kern/sys_process.c,v
retrieving revision 1.57
diff -u -u -r1.57 sys_process.c
--- sys_process.c 2000/12/28 08:34:21 1.57
+++ sys_process.c 2000/12/30 00:24:38
@@ -381,8 +381,8 @@
if (p-p_stat == SSTOP) {
p-p_xstat = uap-data;
setrunnable(p);
- psignal(p, SIGCONT);
mtx_exit(sched_lock, MTX_SPIN);
+ psignal(p, SIGCONT);
} else {
mtx_exit(sched_lock, MTX_SPIN);
if (uap-data) {
--
-Alfred Perlstein - [[EMAIL PROTECTED]|[EMAIL PROTECTED]]
"I have the heart of a child; I keep it in a jar on my desk."
--
Alex Kapranoff, Voice: +7(0832)791845
1 days before the brand new millenium...
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
buildworld fails on -CURRENT after recent libcrypt commits
I cvsuped 4 or 5 hours ago. While building cross-tools: cc -O -pipe -march=pentium -I/usr/src/gnu/usr.bin/perl/miniperl/../../../../cont rib/perl5 -I/usr/obj/usr/src/i386/usr/src/gnu/usr.bin/perl/miniperl -DPERL_EXTER NAL_GLOB -I/usr/obj/usr/src/i386/usr/include -L/usr/obj/usr/src/i386/usr/src/ gnu/usr.bin/perl/miniperl/../libperl -static -o miniperl miniperlmain.o opmini.o /usr/obj/usr/src/i386/usr/src/gnu/usr.bin/perl/miniperl/../libperl/libperl.a - lm -lcrypt -lutil /usr/lib/libcrypt.a(crypt-md5.o): In function `crypt_md5': crypt-md5.o(.text+0xc0): undefined reference to `__MD5Init' crypt-md5.o(.text+0xe1): undefined reference to `__MD5Update' crypt-md5.o(.text+0x107): undefined reference to `__MD5Update' crypt-md5.o(.text+0x11d): undefined reference to `__MD5Update' crypt-md5.o(.text+0x12f): undefined reference to `__MD5Init' crypt-md5.o(.text+0x150): undefined reference to `__MD5Update' crypt-md5.o(.text+0x169): undefined reference to `__MD5Update' crypt-md5.o(.text+0x18a): undefined reference to `__MD5Update' crypt-md5.o(.text+0x19a): undefined reference to `__MD5Final' crypt-md5.o(.text+0x1d6): undefined reference to `__MD5Update' crypt-md5.o(.text+0x237): undefined reference to `__MD5Update' crypt-md5.o(.text+0x291): undefined reference to `__MD5Final' crypt-md5.o(.text+0x2b1): undefined reference to `__MD5Init' crypt-md5.o(.text+0x2ea): undefined reference to `__MD5Update' crypt-md5.o(.text+0x31c): undefined reference to `__MD5Update' crypt-md5.o(.text+0x362): undefined reference to `__MD5Update' crypt-md5.o(.text+0x399): undefined reference to `__MD5Update' crypt-md5.o(.text+0x3ac): undefined reference to `__MD5Final' *** Error code 1 1 error *** Error code 2 Worked around with % cd /usr/src/lib/libcrypt; make install before buildworld To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
Re: randomdev entropy gathering is really weak
On Mon, Jul 17, 2000 at 05:08:35PM +0200, Leif Neland wrote: On Mon, 17 Jul 2000, Steve O'Hara-Smith wrote: On 17-Jul-00 Poul-Henning Kamp wrote: NTP is the perfect way to gather entropy at bootup! Only if in reach of an NTP server ? If you can't reach a NTP server, you are not connected to the internet. In that case you don't need to worry so much about security... Flawed logic. That's cryptography. It's about information protection. And you of course know that about 80% of computer crimes are commited by local cow orkers in a LAN environment behind a twenty five firewalls, proxies and the like. -- Alex Kapranoff, 2:50/383.20@fidonet, Voice: +7(0832)791845. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
