first KSE/3 commit will be:
After discussing it with Peter we agreed that we will do the milestone 3 commi in several stages. First will be a diff-reduction commit, in which we make very little functional differnce to current but add and use various macros and structure renamings that are needed for KSE/3 but which ar enot part of the real MEAT of the work this will reduce the size of the KSE/3 diff considerably and allow us to focus on the real differences. We expect to try do this commit at the end of this week. The meat of the commit is actually small but will need more testing and should follow in a week or two after that. thanks for your patience.. Julian -- ++ __ _ __ | __--_|\ Julian Elischer | \ U \/ / hard at work in | / \ [EMAIL PROTECTED] +--x USA\ a very strange | ( OZ)\___ ___ | country ! +- X_.---._/presently in San Francisco \_/ \\ v To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
buildword break on STABLE
Hi I am trying to build -CURRENT world on 4.4-RELEASE machine: walder#/ext/current/src 129_ make buildworld -- Rebuilding the temporary build tree -- rm -rf /usr/obj/ext/current/src/i386 mkdir -p /usr/obj/ext/current/src/i386/usr/bin mkdir -p /usr/obj/ext/current/src/i386/usr/lib/compat/aout mkdir -p /usr/obj/ext/current/src/i386/usr/games mkdir -p /usr/obj/ext/current/src/i386/usr/libdata/ldscripts ... skip a bit ... cc -O -pipe -c /ext/current/src/usr.bin/xinstall/xinstall.c cc -O -pipe -c /ext/current/src/usr.bin/xinstall/../../lib/libc/gen/strtofflags.c /ext/current/src/usr.bin/xinstall/../../lib/libc/gen/strtofflags.c:38: syntax error before string constant /ext/current/src/usr.bin/xinstall/../../lib/libc/gen/strtofflags.c:38: warning: data definition has no type or storage class *** Error code 1 Stop in /ext/current/src/usr.bin/xinstall. *** Error code 1 Stop in /ext/current/src. *** Error code 1 Stop in /ext/current/src. *** Error code 1 Stop in /ext/current/src. walder#/ext/current/src 134_ fgrep -n FBSDID \ /ext/current/src/usr.bin/xinstall/../../lib/libc/gen/strtofflags.c 38:__FBSDID($FreeBSD: src/lib/libc/gen/strtofflags.c,v 1.19 2002/02/01 00:57:29 obrien Exp $); walder#/ext/current/src 135_ Any suggestions ? ps: This was woring for me before well. -- SW Soft, Moscow Vladimir B. Grebenschikov, [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: /dev/ugen?
CHOI Junho wrote: KS == KT Sin [EMAIL PROTECTED] writes: KS If you have usbd running, just hot attach the USB device (serial, etc). KS You are probably using devfs on your -CURRENT. The correct /dev/* entries KS will be created automagically once the hardware device is probed and attached KS successfully. KS kt I have attached USB serial device(PDA cradle), and usbd is running. What should I do more? As long as I know, devfs will create new entry automatically but in this case I have no idea... You probably need put manufacturer ID and device ID of your device into src/sys/dev/usb/usbdevs. -Maxim KS On Mon, Feb 04, 2002 at 02:47:41AM +0900, CHOI Junho wrote: Hi, How can I see/make /dev/ugen* devices? I wish to use it but I can't find a way to see/use this. I want to use coldsync(USB serial on Palm). My machine has USB, of course; FreeBSD gradius.wdb.co.kr 5.0-CURRENT FreeBSD 5.0-CURRENT #16: Thu Dec 27 18:31:20 KST 2001 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GRADIUS i386 port 0xfca0-0xfcbf irq 9 at device 7.2 on pci0 on uhci0 usb0: USB revision 1.0 uhub0: Intel UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub0: 2 ports with 2 removable, self powered uhub1: Philips Semiconductors hub, class 9/0, rev 1.10/1.10, addr 2 uhub1: 3 ports with 3 removable, self powered -- Any opinions in this posting are my own and not those of my employers +++ To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message -- +++ Any opinions in this posting are my own and not those of my employers +++ CHOI Junho [sleeping now]http://www.kr.FreeBSD.org/~cjh [while sleeping] cjh @ kr.FreeBSD.ORG cjh @ FreeBSD.ORG cjh @ wdb.co.kr Korea FreeBSD Users Group www.kr.FreeBSD.org Web Data Bankwww.wdb.co.kr To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Junior Annoying Hacker Task
Thus spake Miguel Mendez [EMAIL PROTECTED]: On Sat, 02 Feb 2002 03:29:50 -0800 Terry Lambert [EMAIL PROTECTED] wrote: Hi Terry et al, Let me know the form you want the hierarchy to take, so you can stick it into the GTK hierarchy thingy; I'll be happy to crank out some quick yacc and lex code to do the parsing of the file into that format for you. A structure definition, with links, and how you want it linked, would be ideal. 8-). I've been thinking about it and I'd like to provide two frontends, a GTK one and a ncurses one, what do you think? This is getting pretty ambitious. I'm not about to say that a procedural interface to rc.conf wouldn't be nice, or that it couldn't be done well, but when it comes down to it, what problem are we trying to solve? Certainly rc.conf supports a lot of options, and a hierarchical arrangement would simplify things. However, most people only need to deal with a small subset of the options, and organization doesn't seem to be a big issue. Of course there's also the possibility that someone could leave out a quotation mark and end up with a non-working system, but that's rare and solvable with simpler tools, as has been discussed. Besides, both Microsoft and the linuxconf people have done far worse than rc.conf---especially Microsoft. Know what to do when a bad update eats a VXD that isn't on the Windows CD? :P [I'm not subscribed to -current, so please CC me any replies if that's where this thread has drifted.] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: AMD AGP Bug
On Fri, 1 Feb 2002, Peter Wemm wrote: - AMD write cache allocation due to speculative writes being cancelled and then written back later vs no cache snooping on AGP regions. I'm somewhat perplexed about this issue, there's lots of conflicting info going around, a good deal of it which does not make much sense [to me :-)]. I really dont see what PSE has to do with this for several reasons.. if the page/ region is cacheable, why does a 4MB vs 4K page make any difference? cacheable vs no-cache-snooping is a recipe for disaster.. why would 4K pages on a non-coherent region be safer? Or is the problem that write allocation happens on uncacheable/non-write-back regions in 4MB pages? Or something else? Speculative writes can only happen to pages in the TLB (so you don't get speculative TLB misses and replacements), not having a large amount of 4M pages around in the TLB means that addresses covered by these can't possibly be involved in speculative writes. I personaly suspect the reason the cache line flushes of speculatively written to cache lines derive from the AMD-s use of MOESI coherency and mapping that to actual bits. Another minor side effect is that you get direct cache-to-cahce transfers in SMP systems for shared data. Sander Cheers, -Peter -- Peter Wemm - [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] All of this is for nothing if we don't go to the stars - JMS/B5 To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
panic: softdep_setup_inomapdep: found inode
IdlePTD at phsyical address 0x00487000
initial pcb at physical address 0x003702c0
panicstr: bremfree: bp 0xd3bf08d0 not locked
panic messages:
---
panic: softdep_setup_inomapdep: found inode
syncing disks... panic: bremfree: bp 0xd3bf08d0 not locked
Uptime: 1h49m20s
pccbb0: pccbb_power: CARD_VCC_0V and CARD_VPP_0V [44]
pccbb1: pccbb_power: CARD_VCC_0V and CARD_VPP_0V [44]
/dev/vmmon: Module vmmon: unloaded
dumping to dev ad0s2b, offset 2000128
dump ata0: resetting devices .. done
1023 1022 [...]
---
#0 dumpsys () at /usr/src/sys/kern/kern_shutdown.c:492
492 if (!dodump)
(kgdb) t
[Current thread is 0 (process 0)]
(kgdb) where
#0 dumpsys () at /usr/src/sys/kern/kern_shutdown.c:492
#1 0xc01bf5ff in boot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:335
#2 0xc01bfa39 in panic (fmt=0xc02fbec2 bremfree: bp %p not locked)
at /usr/src/sys/kern/kern_shutdown.c:634
#3 0xc01eedb1 in bremfree (bp=0xd3bf08d0) at /usr/src/sys/kern/vfs_bio.c:545
#4 0xc01f0383 in vfs_bio_awrite (bp=0xd3bf08d0)
at /usr/src/sys/kern/vfs_bio.c:1534
#5 0xc01a23d8 in spec_fsync (ap=0xe7c5e88c)
at /usr/src/sys/fs/specfs/spec_vnops.c:404
#6 0xc01a1fc1 in spec_vnoperate (ap=0xe7c5e88c)
at /usr/src/sys/fs/specfs/spec_vnops.c:119
#7 0xc02813bd in ffs_sync (mp=0xc755d600, waitfor=2, cred=0xc2069b00,
td=0xc03ab804) at vnode_if.h:441
#8 0xc01fb461 in sync (td=0xc03ab804, uap=0x0)
at /usr/src/sys/kern/vfs_syscalls.c:657
#9 0xc01bf248 in boot (howto=256) at /usr/src/sys/kern/kern_shutdown.c:244
#10 0xc01bfa39 in panic (fmt=0xc030dca0 softdep_setup_inomapdep: found inode)
at /usr/src/sys/kern/kern_shutdown.c:634
#11 0xc027a62a in softdep_setup_inomapdep (bp=0xd3c7bce4, ip=0xc8202200,
newinum=503542) at /usr/src/sys/ufs/ffs/ffs_softdep.c:1182
#12 0xc0274014 in ffs_nodealloccg (ip=0xc8202200, cg=46, ipref=503460,
mode=33152) at /usr/src/sys/ufs/ffs/ffs_alloc.c:1391
#13 0xc0273083 in ffs_hashalloc (ip=0xc8202200, cg=46, pref=503460,
---Type return to continue, or q return to quit---
size=33152, allocator=0xc0273e14 ffs_nodealloccg)
at /usr/src/sys/ufs/ffs/ffs_alloc.c:871
#14 0xc0272b10 in ffs_valloc (pvp=0xe78ce120, mode=33152, cred=0xc8084c00,
vpp=0xe7c5e9dc) at /usr/src/sys/ufs/ffs/ffs_alloc.c:611
#15 0xc028a252 in ufs_makeinode (mode=33152, dvp=0xe78ce120, vpp=0xe7c5ec40,
cnp=0xe7c5ec54) at /usr/src/sys/ufs/ufs/ufs_vnops.c:2264
#16 0xc0287ae8 in ufs_create (ap=0xe7c5eb38)
at /usr/src/sys/ufs/ufs/ufs_vnops.c:199
#17 0xc028a5b9 in ufs_vnoperate (ap=0xe7c5eb38)
at /usr/src/sys/ufs/ufs/ufs_vnops.c:2658
#18 0xc0200740 in vn_open_cred (ndp=0xe7c5ec2c, flagp=0xe7c5ebf8, cmode=384,
cred=0xc8084c00) at vnode_if.h:90
#19 0xc02005de in vn_open (ndp=0xe7c5ec2c, flagp=0xe7c5ebf8, cmode=384)
at /usr/src/sys/kern/vfs_vnops.c:86
#20 0xc01fbcbe in open (td=0xe7924404, uap=0xe7c5ed20)
at /usr/src/sys/kern/vfs_syscalls.c:1117
#21 0xc02c1678 in syscall (frame={tf_fs = -1070923729, tf_es = 47, tf_ds = 47,
tf_edi = 0, tf_esi = 134901314, tf_ebp = -1077940792,
tf_isp = -406459020, tf_ebx = 33256, tf_edx = 0, tf_ecx = 135024096,
tf_eax = 5, tf_trapno = 22, tf_err = 2, tf_eip = 134557391, tf_cs = 31,
tf_eflags = 647, tf_esp = -1077940852, tf_ss = 47})
at /usr/src/sys/i386/i386/trap.c:1032
#22 0xc02b58bd in syscall_with_err_pushed ()
---Type return to continue, or q return to quit---
#23 0x804a952 in ?? ()
#24 0x80485d9 in ?? ()
#25 0x8048135 in ?? ()
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: AMD AGP Bug
Narvi wrote: Speculative writes can only happen to pages in the TLB (so you don't get speculative TLB misses and replacements), not having a large amount of 4M pages around in the TLB means that addresses covered by these can't possibly be involved in speculative writes. I personaly suspect the reason the cache line flushes of speculatively written to cache lines derive from the AMD-s use of MOESI coherency and mapping that to actual bits. Another minor side effect is that you get direct cache-to-cahce transfers in SMP systems for shared data. I think that the problem is more related to the fact that there are 16 TLB entries for 4K data pages, 16 for 4K code pages, and another 8 for 4M pages. Peter is right, in other words, because there is a problem with the interaction of 4K and 4M pages. I've seen this myself, as I've previously reported, as well as seeing other problems (and knowing how to work around them, after weeks of investigation into characterizing them). Note that I was lucky, in that I had modified the FreeBSD kernel to use certain types of mappings in a certain way; I think it would be very difficult, or impossible, for anyone else to duplicate the problem in order to better characterize it beyond DISABLE_PSE, except the chip vendors themselves, if they started from first principles with a simulation. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Panics in ffs_clusteracct with todays -current
On Sun, 3 Feb 2002, John Polstra wrote: The kernel from today's current (CVSupped 3 Feb 2002 around 17:40 PST) can't stay up for more than a few minutes without getting a page-not-present panic at line 1815 of ufs/ffs/ffs_alloc.c revision 1.86. It is in this code: /* * Find the size of the cluster going backward. */ start = blkno - 1; end = start - fs-fs_contigsumsize; if (end 0) end = -1; mapp = freemapp[start / NBBY]; map = *mapp--; ^ BANG! This was broken by a recent change to the type of NBBY. `start' is apparently negative (it would be for blkno == 0). Small negative values of `start' used to give an index of 0 in freemapp (not even small negative, since integer division bogusly rounds negative values towards 0). They now give an index of about 0x1fff on 32-bit machines. I also got panics in vm. The vm code apparently trapped while trying to map the preposterous address generated by the above. This patch just backs out the change to NBBY. %%% Index: types.h === RCS file: /home/ncvs/src/sys/sys/types.h,v retrieving revision 1.50 diff -u -2 -r1.50 types.h --- types.h 3 Feb 2002 11:36:59 - 1.50 +++ types.h 4 Feb 2002 13:08:05 - @@ -172,5 +172,5 @@ #ifndef _POSIX_SOURCE -#defineNBBY8U /* number of bits in a byte */ +#defineNBBY8 /* number of bits in a byte */ /* %%% Bruce To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
pam_ssh world breakage (was: Re: cvs commit: src/lib/libpam Makefile.inc)
On Sun, Feb 03, 2002 at 07:51:52AM -0800, Mark Murray wrote:
markm 2002/02/03 07:51:52 PST
Modified files:
lib/libpam Makefile.inc
Log:
Turn on fascist warning mode.
Revision ChangesPath
1.4 +3 -1 src/lib/libpam/Makefile.inc
This delta breaks buildworld. gcc(1) has a known bug-feature
of hiding some errors in standard system headers, making them
invisible without -I.
During buildworld, ${WORLDTMP}/usr/include is populated with
headers, and -I${WORLDTMP}/usr/include is added to CFLAGS,
and compile now can't survive the WARNS=4.
Try this to see the breakage:
cd /usr/src/lib/libpam/modules/pam_ssh
make cleandir
make DESTDIR=/
Please back it out.
Cheers,
--
Ruslan Ermilov Sysadmin and DBA,
[EMAIL PROTECTED] Sunbay Software AG,
[EMAIL PROTECTED] FreeBSD committer,
+380.652.512.251Simferopol, Ukraine
http://www.FreeBSD.org The Power To Serve
http://www.oracle.com Enabling The Information Age
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: Junior Annoying Hacker Task
Andrew Cowan wrote: However, I have previously thought that a system that used xml files to store application configs (that would then be used to generate valid conf files) would be useful. I was on the verge of doing so the other day. Basically, I wanted to have standard configuration data describing the network, services and service configuration stored in XML and use XSLT to produce which-ever config files you need. You then introduce some inheritance and allow configuration to be overridden for particular hosts, subnets, networks, and/or platforms, and you have a powerful site-wide configuration management tool. You get all the usual benefits from using XML as well; data source, processing and output independence, so the configuration data could be use to automatically generate HTML for helpdesk pages, could be sourced from existing a variety of new or pre-existing data repositories (LDAP, CVS, file system, etc.) and could be processed and edited using a number of standard tools. If only I had not run out of time write it, I'd be happily using this today. Mike. -- Mike Gratton [EMAIL PROTECTED] Every motive escalate. Blatant self-promotion: http://web.vee.net/ To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: AMD AGP Bug
On Mon, 4 Feb 2002, Terry Lambert wrote: Narvi wrote: Speculative writes can only happen to pages in the TLB (so you don't get speculative TLB misses and replacements), not having a large amount of 4M pages around in the TLB means that addresses covered by these can't possibly be involved in speculative writes. I personaly suspect the reason the cache line flushes of speculatively written to cache lines derive from the AMD-s use of MOESI coherency and mapping that to actual bits. Another minor side effect is that you get direct cache-to-cahce transfers in SMP systems for shared data. I think that the problem is more related to the fact that there are 16 TLB entries for 4K data pages, 16 for 4K code pages, and another 8 for 4M pages. Peter is right, in other words, because there is a problem with the interaction of 4K and 4M pages. I've seen this myself, as I've previously reported, as well as seeing other problems (and knowing how to work around them, after weeks of investigation into characterizing them). I wasn't aware that I was contradicting Peter 8-) Yes, it is possible that there are issues (some of which may in turn be bugs) in the 4KB and 4MB page handling. I'm not sure that 4K vs. 4M handling is done in any special way when it comes to speculative vs. normal writes. Speculative writes don't cross page boundaries (which means a different thing depending on which TLB you found the page from) - meaning that having a change of objects inside a mapped 4M page is asking for trouble. Writing back the cache lines that may have been written to is imho reasonable behaviour. Note that I was lucky, in that I had modified the FreeBSD kernel to use certain types of mappings in a certain way; I think it would be very difficult, or impossible, for anyone else to duplicate the problem in order to better characterize it beyond DISABLE_PSE, except the chip vendors themselves, if they started from first principles with a simulation. It may even well be possible to get different results with aligned vs. misaligned reads and writes, or a proper mix thereof. It may be possible to build a model to track down the what is really going on part, but its not clear its worth the trouble over just picking what (not) to do. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Verba Volant
We have been requested to insert the following email address, [EMAIL PROTECTED], in the Verba Volant Newsletter database. Through this daily service you will receive a quotation, selected from amongst the most celebrated philosophers, writers and poets of all time and translated into many languages and dialects by volunteers worldwide. If you would like to confirm your subscription to Verba Volant, please click on the following link: http://www.logos.net/owa-l/press.subscribe?lang=en[EMAIL PROTECTED] If you do not wish to click on the link, your subscription will be cancelled. Thank you for your time. Verba Volant Il nous a été demandé d'ajouter l'adresse électronique [EMAIL PROTECTED] dans la liste des destinataires de Verba Volant, un service qui tous les jours vous adressera une citation sélectionnée parmi les uvres des meilleurs philosophes, écrivains, poètes de tous les temps et traduite en de très nombreuses langues grâce à des volontaires du monde entier. Pour confirmer l'inscription à Verba Volant, veuillez vous connecter au lien suivant: http://www.logos.net/owa-l/press.subscribe?lang=fr[EMAIL PROTECTED] Si vous préférez ne pas cliquer sur le lien, vous ne recevrez rien. Merci dans tous les cas de nous avoir accordé quelques secondes. Verba Volant Se nos ha solicitado insertar la dirección de correo electrónico [EMAIL PROTECTED] en el listado de envíos de Verba Volant, un servicio que diariamente le enviará citas elegidas entre los mejores filosofos, escritores, poetas, etc., traducidas a varios idiomas y dialectos. Dichas citas están traducidas por voluntarios que se conectan a nuestra web desde todo el mundo. Si quiere confirmar la suscripción a Verba Volant, le rogamos entre en: http://www.logos.net/owa-l/press.subscribe?lang=es[EMAIL PROTECTED] Si no entra en la dirección señalada no recibirá las citas. Muchas gracias por el tiempo que nos ha dedicado. Verba Volant Ci è stato chiesto di inserire l'indirizzo di posta elettronica [EMAIL PROTECTED] nellelenco dei destinatari di Verba Volant, un servizio che ogni giorno ti invierà una citazione scelta tra quelle dei migliori filosofi, scrittori, poeti di tutti i tempi e tradotta in moltissime lingue e dialetti grazie alla collaborazione di volontari da tutto il mondo. Se desideri confermare l'iscrizione, ti preghiamo di collegarti al seguente link: http://www.logos.net/owa-l/press.subscribe?lang=it[EMAIL PROTECTED] Nel caso preferissi non cliccare sul link, non riceverai nulla. Grazie comunque per i secondi che ci hai dedicato. Cordiali saluti. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Panics in ffs_clusteracct with todays -current
If this fixes a problem, then please go ahead and commit it. :-) M On Sun, 3 Feb 2002, John Polstra wrote: The kernel from today's current (CVSupped 3 Feb 2002 around 17:40 PST) can't stay up for more than a few minutes without getting a page-not-present panic at line 1815 of ufs/ffs/ffs_alloc.c revision 1.86. It is in this code: /* * Find the size of the cluster going backward. */ start = blkno - 1; end = start - fs-fs_contigsumsize; if (end 0) end = -1; mapp = freemapp[start / NBBY]; map = *mapp--; ^ BANG! This was broken by a recent change to the type of NBBY. `start' is apparently negative (it would be for blkno == 0). Small negative values of `start' used to give an index of 0 in freemapp (not even small negative, since integer division bogusly rounds negative values towards 0). They now give an index of about 0x1fff on 32-bit machines. I also got panics in vm. The vm code apparently trapped while trying to map the preposterous address generated by the above. This patch just backs out the change to NBBY. %%% Index: types.h === RCS file: /home/ncvs/src/sys/sys/types.h,v retrieving revision 1.50 diff -u -2 -r1.50 types.h --- types.h 3 Feb 2002 11:36:59 - 1.50 +++ types.h 4 Feb 2002 13:08:05 - @@ -172,5 +172,5 @@ #ifndef _POSIX_SOURCE -#define NBBY8U /* number of bits in a byte */ +#define NBBY8 /* number of bits in a byte */ /* %%% Bruce -- o Mark Murray \_ FreeBSD Services Limited O.\_Warning: this .sig is umop ap!sdn To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: buildword break on STABLE
In message: [EMAIL PROTECTED] Vladimir B. Grebenschikov [EMAIL PROTECTED] writes: : I am trying to build -CURRENT world on 4.4-RELEASE machine: ... : Any suggestions ? I can tell you that 4.5-release will work. Warner To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
RE: Junior Annoying Hacker Task
Hi, I don't know whether the suggested approach is really a good one or not, but as far as implementing some registry-like features in FreeBSD, we have developped something that proves to be useful. The idea was to extend the sysctl mechanism to make it dynamic from the user-land point of view. This was implemented as a kernel module (a simple one too) and a modified sysctl command line. With this mod we can add/remove nodes in the sysctl tree and use it to store information that is needed by many applications. The implementation we have is certainly rough, but it gets the job done and represents a very easy and convenient tool for: - sharing non sensitive bits of information between processes - storing configuration/status data in an easy to use manner - benefit from the tree like presentation of sysctl (building a gui around this should be easy) - don't loose the 'vi' compatibility (sysctl values can be loaded at boot time from a text file) Note that the security model around these 'user-land dynamic sysctl entries' can be improved significantly, but as a proof of concept the mod is useful. Some may argue that storing userland data in the kernel space is Not A Nice Thing(tm) but it certainly makes things a lot easier. Patrick. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Panics in ffs_clusteracct with todays -current
In article [EMAIL PROTECTED],
Alfred Perlstein [EMAIL PROTECTED] wrote:
* Bruce Evans [EMAIL PROTECTED] [020204 05:26] wrote:
This was broken by a recent change to the type of NBBY.
`start' is apparently negative (it would be for blkno == 0). Small
negative values of `start' used to give an index of 0 in freemapp (not
even small negative, since integer division bogusly rounds negative
values towards 0). They now give an index of about 0x1fff on
32-bit machines.
I also got panics in vm. The vm code apparently trapped while trying
to map the preposterous address generated by the above.
This patch just backs out the change to NBBY.
Wouldn't this make more sense:
Index: ffs/ffs_alloc.c
===
RCS file: /home/ncvs/src/sys/ufs/ffs/ffs_alloc.c,v
retrieving revision 1.86
diff -u -r1.86 ffs_alloc.c
--- ffs/ffs_alloc.c 2 Feb 2002 01:42:44 - 1.86
+++ ffs/ffs_alloc.c 4 Feb 2002 16:08:34 -
@@ -1790,7 +1790,7 @@
end = start + fs-fs_contigsumsize;
if (end = cgp-cg_nclusterblks)
end = cgp-cg_nclusterblks;
- mapp = freemapp[start / NBBY];
+ mapp = freemapp[start 0 ? 0 : start / NBBY];
map = *mapp++;
bit = 1 (start % NBBY);
for (i = start; i end; i++) {
I think your change makes sense here, but I also think the change
to NBBY should be backed out. In general it is very dangerous to
change constants to explicitly unsigned. There's no reason why NBBY
shouldn't be used in signed contexts, but making it unsigned promotes
all of the other ints in containing expressions to unsigned as well
(on the i386). NBBY is used in lots of code, including 3rd party
applications. It has a long tradition, and now isn't the time to
change its type. There's no reason for it to be explicitly unsigned.
The constant 8 is every bit as positive as the constant 8U. :-)
John
--
John Polstra
John D. Polstra Co., Inc.Seattle, Washington USA
Disappointment is a good sign of basic intelligence. -- Chögyam Trungpa
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: Latest binutils import breaks Alpha cross compiles
On Tue, Jan 29, 2002 at 12:02:18PM -0800, David O'Brien wrote:
On Tue, Jan 29, 2002 at 07:08:21PM +0200, Ruslan Ermilov wrote:
David!
After the latest binutils import, attempts to cross-compile
Alpha fail at the cross-tools stage of buildworld as shown
below. Please also note sed(1) complaints about nonexistent
ldscripts/ files. I suspect that normal alpha worlds may
be broken as well, but I can't tell for sure.
I am sure it probably is. There is a lack of developers using -current
on Alpha's that progress on i386, IA64, x86-64, and sparc64 cannot be
held back. I will fix the Alpha problems when I get a chance, but it may
be a week.
The fix was as simple as this:
Index: Makefile
===
RCS file: /home/ncvs/src/gnu/usr.bin/binutils/libbfd/Makefile,v
retrieving revision 1.16
diff -u -r1.16 Makefile
--- Makefile2002/01/27 22:47:22 1.16
+++ Makefile2002/02/04 16:40:01
@@ -5,7 +5,7 @@
.PATH: ${SRCDIR}/bfd ${SRCDIR}/opcodes
LIB= bfd
-SRCS+= archive.c archures.c bfd.c binary.c cache.c \
+SRCS+= archive.c archive64.c archures.c bfd.c binary.c cache.c \
coffgen.c corefile.c elf.c elf-eh-frame.c elf-strtab.c format.c \
hash.c ihex.c init.c libbfd.c linker.c merge.c opncls.c reloc.c \
section.c srec.c stab-syms.c stabs.c syms.c targets.c tekhex.c \
Verified with both i386 and alpha.
Cheers,
--
Ruslan Ermilov Sysadmin and DBA,
[EMAIL PROTECTED] Sunbay Software AG,
[EMAIL PROTECTED] FreeBSD committer,
+380.652.512.251Simferopol, Ukraine
http://www.FreeBSD.org The Power To Serve
http://www.oracle.com Enabling The Information Age
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: Panics in ffs_clusteracct with todays -current
The kernel from today's current (CVSupped 3 Feb 2002 around 17:40 PST) can't stay up for more than a few minutes without getting a page-not-present panic at line 1815 of ufs/ffs/ffs_alloc.c revision 1.86. It is in this code: /* * Find the size of the cluster going backward. */ start = blkno - 1; end = start - fs-fs_contigsumsize; if (end 0) end = -1; mapp = freemapp[start / NBBY]; map = *mapp--; ^ BANG! I just reverted the upgrade of NBBY (to unsigned). M -- o Mark Murray \_ FreeBSD Services Limited O.\_Warning: this .sig is umop ap!sdn To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: make buildworld error in libpam/modules/pam_ssh
I'm experiencing the following error while bulding world: Fix committed. M -- o Mark Murray \_ FreeBSD Services Limited O.\_Warning: this .sig is umop ap!sdn To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Latest binutils import breaks Alpha cross compiles
On Mon, Feb 04, 2002 at 06:43:23PM +0200, Ruslan Ermilov wrote: The fix was as simple as this: Thanks!! Committed. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: pam_ssh world breakage (was: Re: cvs commit: src/lib/libpam Makefile.inc)
On Mon, Feb 04, 2002 at 03:30:32PM +0200, Ruslan Ermilov wrote: This delta breaks buildworld. gcc(1) has a known bug-feature of hiding some errors in standard system headers, making them invisible without -I. ... and compile now can't survive the WARNS=4. Not to mention there is ZERO way this code will pass WARNS=4 for GCC 3. Please Committers, do not try to WARNS code right now -- there just is no use. It will only get in the way later. Well, of course feel free to make the code changes, but PLEASE do not commit any stronger WARNS levels to Makefiles. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Panics in ffs_clusteracct with todays -current
* Bruce Evans [EMAIL PROTECTED] [020204 05:26] wrote:
On Sun, 3 Feb 2002, John Polstra wrote:
The kernel from today's current (CVSupped 3 Feb 2002 around 17:40
PST) can't stay up for more than a few minutes without getting a
page-not-present panic at line 1815 of ufs/ffs/ffs_alloc.c revision
1.86. It is in this code:
/*
* Find the size of the cluster going backward.
*/
start = blkno - 1;
end = start - fs-fs_contigsumsize;
if (end 0)
end = -1;
mapp = freemapp[start / NBBY];
map = *mapp--;
^
BANG!
This was broken by a recent change to the type of NBBY.
`start' is apparently negative (it would be for blkno == 0). Small
negative values of `start' used to give an index of 0 in freemapp (not
even small negative, since integer division bogusly rounds negative
values towards 0). They now give an index of about 0x1fff on
32-bit machines.
I also got panics in vm. The vm code apparently trapped while trying
to map the preposterous address generated by the above.
This patch just backs out the change to NBBY.
Wouldn't this make more sense:
Index: ffs/ffs_alloc.c
===
RCS file: /home/ncvs/src/sys/ufs/ffs/ffs_alloc.c,v
retrieving revision 1.86
diff -u -r1.86 ffs_alloc.c
--- ffs/ffs_alloc.c 2 Feb 2002 01:42:44 - 1.86
+++ ffs/ffs_alloc.c 4 Feb 2002 16:08:34 -
@@ -1790,7 +1790,7 @@
end = start + fs-fs_contigsumsize;
if (end = cgp-cg_nclusterblks)
end = cgp-cg_nclusterblks;
- mapp = freemapp[start / NBBY];
+ mapp = freemapp[start 0 ? 0 : start / NBBY];
map = *mapp++;
bit = 1 (start % NBBY);
for (i = start; i end; i++) {
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: Panics in ffs_clusteracct with todays -current
On Mon, 4 Feb 2002, Alfred Perlstein wrote:
* Bruce Evans [EMAIL PROTECTED] [020204 05:26] wrote:
This patch just backs out the change to NBBY.
Wouldn't this make more sense:
Index: ffs/ffs_alloc.c
===
RCS file: /home/ncvs/src/sys/ufs/ffs/ffs_alloc.c,v
retrieving revision 1.86
diff -u -r1.86 ffs_alloc.c
--- ffs/ffs_alloc.c 2 Feb 2002 01:42:44 - 1.86
+++ ffs/ffs_alloc.c 4 Feb 2002 16:08:34 -
@@ -1790,7 +1790,7 @@
end = start + fs-fs_contigsumsize;
if (end = cgp-cg_nclusterblks)
end = cgp-cg_nclusterblks;
- mapp = freemapp[start / NBBY];
+ mapp = freemapp[start 0 ? 0 : start / NBBY];
map = *mapp++;
bit = 1 (start % NBBY);
for (i = start; i end; i++) {
Except it doesn't fix the potentially millions of other lines that are
affected by the fix, and leaves at best confusing code: mapp, map and
bit are unitialized to bogus values. This seems to be harmless because
they are not used (the loop is iterated 0 times if start 0).
Bruce
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: Panics in ffs_clusteracct with todays -current
On Mon, Feb 04, 2002 at 03:56:45PM +, Mark Murray wrote: If this fixes a problem, then please go ahead and commit it. :-) Many reasons to not rush this: Maybe because we are in the middle of a discussion. Maybe to educate others why the change was bad so others will not attempt the same change later. Maybe the person that made the change does not agree with the analysis and wants to fix things a different way. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
buildworld is still broken in libpam/libpam
Hi! Alpha buildworld is still broken due to the WARNS=4 change: : cc -O -pipe -nostdinc :-I/CURRENT/usr/src/lib/libpam/libpam/../../../contrib/libpam/libpam/include :-I/CURRENT/usr/src/lib/libpam/libpam/../../../contrib/libpam/libpam_misc/include :-I/CURRENT/usr/src/lib/libpam/libpam/../../../contrib/libpam/libpamc/include -I. :-I/CURRENT/usr/src/lib/libpam/libpam -DDEFAULT_MODULE_PATH=\/usr/lib/\ :-DLIBPAM_VERSION_MAJOR=0 -DLIBPAM_VERSION_MINOR=75 :-I/usr/obj/alpha/CURRENT/usr/src/i386/usr/include -Werror -Wall -W :-Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Wreturn-type -Wcast-qual :-Wwrite-strings -Wswitch -Wshadow -Wcast-align -Wno-uninitialized -c :/CURRENT/usr/src/lib/libpam/libpam/pam_get_pass.c -o pam_get_pass.o : cc1: warnings being treated as errors : /CURRENT/usr/src/lib/libpam/libpam/pam_get_pass.c: In function `pam_conv_pass': : /CURRENT/usr/src/lib/libpam/libpam/pam_get_pass.c:60: warning: type mismatch in :implicit declaration for built-in function `memset' : /CURRENT/usr/src/lib/libpam/libpam/pam_get_pass.c:60: warning: type mismatch in :implicit declaration for built-in function `strlen' : *** Error code 1 : : Stop in /CURRENT/usr/src/lib/libpam/libpam. : *** Error code 1 The patch is attached. Cheers, -- Ruslan Ermilov Sysadmin and DBA, [EMAIL PROTECTED] Sunbay Software AG, [EMAIL PROTECTED] FreeBSD committer, +380.652.512.251Simferopol, Ukraine http://www.FreeBSD.org The Power To Serve http://www.oracle.com Enabling The Information Age Index: pam_get_pass.c === RCS file: /home/ncvs/src/lib/libpam/libpam/pam_get_pass.c,v retrieving revision 1.3 diff -u -p -r1.3 pam_get_pass.c --- pam_get_pass.c 2001/09/30 22:10:58 1.3 +++ pam_get_pass.c 2002/02/04 17:31:06 @@ -28,6 +28,7 @@ __FBSDID($FreeBSD: src/lib/libpam/libpam/pam_get_pass.c,v 1.3 2001/09/30 22:10:58 dillon Exp $); #include stdlib.h +#include string.h #include security/pam_modules.h #include pam_mod_misc.h
Re: Latest binutils import breaks Alpha cross compiles
On Mon, Feb 04, 2002 at 09:06:56AM -0800, David O'Brien wrote: On Mon, Feb 04, 2002 at 06:43:23PM +0200, Ruslan Ermilov wrote: The fix was as simple as this: Thanks!! Committed. Great! /me fires up the DS10.. -- | / o / /_ _ [EMAIL PROTECTED] |/|/ / / /( (_) Bulte Arnhem, the Netherlands To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: pam_ssh world breakage (was: Re: cvs commit: src/lib/libpam Makefile.inc)
David O'Brien [EMAIL PROTECTED] writes: Not to mention there is ZERO way this code will pass WARNS=4 for GCC 3. Please Committers, do not try to WARNS code right now -- there just is no use. It will only get in the way later. Well, of course feel free to make the code changes, but PLEASE do not commit any stronger WARNS levels to Makefiles. Alternatively, developers working on WARNS could use a newer GCC from the ports collection. Best regards, Mike Barcroft To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: buildworld is still broken in libpam/libpam
The patch is attached. Thanks! Committed. M -- o Mark Murray \_ FreeBSD Services Limited O.\_Warning: this .sig is umop ap!sdn To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
HEADS UP! ATA subsystem has changed..
In order to make some real progress on the ATA RAID support code I've also made some rather trivial but many changes all over the ATA subsystem, so if behavior changes on ordinary system I'd like to know... In short - you have been warned ... -Søren To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
buildworld fails in pam on Alpha
Now that current on Alpha builds a lot better than a few days back it fails in: cc -nostdinc -O -pipe -mcpu=ev4 -mcpu=ev4 -I/usr/src/lib/libpam/libpam/../../../contrib/libpam/libpam/include -I/usr/src/lib/libpam/libpam/../../../contrib/libpam/libpam_misc/include -I/usr/src/lib/libpam/libpam/../../../contrib/libpam/libpamc/include -I. -I/usr/src/lib/libpam/libpam -DDEFAULT_MODULE_PATH=\/usr/lib/\ -DLIBPAM_VERSION_MAJOR=0 -DLIBPAM_VERSION_MINOR=75 -I/usr/obj/usr/src/alpha/usr/include -Werror -Wall -W -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Wreturn-type -Wcast-qual -Wwrite-strings -Wswitch -Wshadow -Wcast-align -Wno-uninitialized -c /usr/src/lib/libpam/libpam/pam_get_pass.c -o pam_get_pass.o cc1: warnings being treated as errors /usr/src/lib/libpam/libpam/pam_get_pass.c: In function `pam_conv_pass': /usr/src/lib/libpam/libpam/pam_get_pass.c:60: warning: type mismatch in implicit declaration for built-in function `memset' /usr/src/lib/libpam/libpam/pam_get_pass.c:60: warning: type mismatch in implicit declaration for built-in function `strlen' *** Error code 1 Stop in /usr/src/lib/libpam/libpam. *** Error code 1 -- | / o / /_ _ [EMAIL PROTECTED] |/|/ / / /( (_) Bulte Arnhem, the Netherlands To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: PAM and current
M. Warner Losh [EMAIL PROTECTED] writes: I just upgraded my laptop to -current to start working on some lingering newcard + acpi problems (like the panic on boot due to interrupt issues). So I notice a new message on my console when I log in via xdm: Feb 3 16:18:44 hammer : PAM unable to dlopen(/usr/lib/pam_opie.so) Feb 3 16:18:44 hammer : PAM [dlerror: /usr/lib/libopie.so.2: Undefined symbol __xuname] Feb 3 16:18:44 hammer : PAM adding faulty module: /usr/lib/pam_opie.so Feb 3 16:18:44 hammer : PAM unable to dlopen(/usr/lib/pam_opieaccess.so) Feb 3 16:18:44 hammer : PAM [dlerror: /usr/lib/libopie.so.2: Undefined symbol __xuname] Feb 3 16:18:44 hammer : PAM adding faulty module: /usr/lib/pam_opieaccess.so Your xdm is linked against -STABLE libc, but the PAM modules expect a -CURRENT libc. Since PAM modules aren't versioned, there's no way to keep -STABLE modules around for -STABLE apps; you have to rebuild all PAM apps when upgrading from -STABLE to -CURRENT. DES -- Dag-Erling Smorgrav - [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: buildword break on STABLE
On Mon, 2002-02-04 at 19:00, M. Warner Losh wrote: In message: [EMAIL PROTECTED] Vladimir B. Grebenschikov [EMAIL PROTECTED] writes: : I am trying to build -CURRENT world on 4.4-RELEASE machine: ... : Any suggestions ? I can tell you that 4.5-release will work. What happens ? Only about 2 weeks ago all was fine. Warner -- SW Soft, Moscow Vladimir B. Grebenschikov, [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Junior Annoying Hacker Task
On Tue, Feb 05, 2002 at 01:30:53AM +1030, Mike Gratton wrote: However, I have previously thought that a system that used xml files to store application configs (that would then be used to generate valid conf files) would be useful. I was on the verge of doing so the other day. Basically, I wanted to have standard configuration data describing the network, services and service configuration stored in XML and use XSLT to produce which-ever config files you need. You then introduce some inheritance and allow configuration to be overridden for particular hosts, subnets, networks, and/or platforms, and you have a powerful site-wide configuration management tool. Great minds and all that: http://people.freebsd.org/~nik/xml-servers/ I'm going to try and do a 5 minute 'Work in Progress' on this at BSDCon next week if anyone's interested. Feel free to take that as a starting point. N -- FreeBSD: The Power to Serve http://www.freebsd.org/ (__) FreeBSD Documentation Projecthttp://www.freebsd.org/docproj/\\\'',) \/ \ ^ --- 15B8 3FFC DDB4 34B0 AA5F 94B7 93A8 0764 2C37 E375 --- .\._/_) msg34120/pgp0.pgp Description: PGP signature
Re: AMD AGP Bug
Narvi wrote: I wasn't aware that I was contradicting Peter 8-) Sorry; looked like it to me... 8-). It may even well be possible to get different results with aligned vs. misaligned reads and writes, or a proper mix thereof. It may be possible to build a model to track down the what is really going on part, but its not clear its worth the trouble over just picking what (not) to do. Unless you were a chip vendor telling people to disable PSE in all rour systems that ran Windows or Linux, and the resulting systems were losing 4-14% of their total performance as a result? ;^). -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: Panics in ffs_clusteracct with todays -current
David O'Brien wrote: If this fixes a problem, then please go ahead and commit it. :-) Many reasons to not rush this: Maybe because we are in the middle of a discussion. Maybe to educate others why the change was bad so others will not attempt the same change later. Maybe the person that made the change does not agree with the analysis and wants to fix things a different way. I have to agree with David; there had to be some reason involved in the change in the first place. I can think of a couple of reasons, haviong to do with sign extended math, which might prompt the change. It may be that Alfred's fix is the best one, despite the risks Bruce has identified. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
buildworld broken
Hi All, I've been battling with building current for a few days now, I've noticed that some of the errors I've been seeing have been fixed, but today in desperation I deleted my /usr/src tree and cvs'd it up again from the cvsup.au.freebsd.org server. Todays error: === libpam/modules/pam_ssh rm -f .depend mkdep -f .depend -a snip /usr/src/lib/libpam/modules/pam_ssh/pam_ssh.c /usr/src/lib/libpam/modules/pam_ssh/pam_ssh.c:48: ssh.h: No such file or directory No such file or dir for these also: openssl/dsa.h openssl/evp.h key.h authfd.h authfile.h log.h Thanks, Anthony
RE: Junior Annoying Hacker Task
On Mon, 2002-02-04 at 11:03, Patrick Bihan-Faou wrote: Some may argue that storing userland data in the kernel space is Not A Nice Thing(tm) but it certainly makes things a lot easier. Why am I reminded of when terminal type information was bodged into the stty settings as a 2-character code? -- brandon s. allbery [os/2][linux][solaris][japh] [EMAIL PROTECTED] system administrator [WAY too many hats][EMAIL PROTECTED] electrical and computer engineeringKF8NH carnegie mellon university [better check the oblivious first -ke6sls] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: buildworld broken
From: [EMAIL PROTECTED] Date: Tue, 5 Feb 2002 11:45:32 +1100 Hi All, I've been battling with building current for a few days now, I've noticed that some of the errors I've been seeing have been fixed, but today in desperation I deleted my /usr/src tree and cvs'd it up again from the cvsup.au.freebsd.org server. What source collection(s) are you using? I suggest src-all. Cheers, david (links to my resume at http://www.catwhisker.org/~david) -- David H. Wolfskill [EMAIL PROTECTED] I believe it would be irresponsible (and thus, unethical) for me to advise, recommend, or support the use of any product that is or depends on any Microsoft product for any purpose other than personal amusement. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: buildworld fails in pam on Alpha
Wilko Bulte [EMAIL PROTECTED] wrote: Now that current on Alpha builds a lot better than a few days back it fails in: /usr/src/lib/libpam/libpam/pam_get_pass.c:60: warning: type mismatch in implicit declaration for built-in function `memset' /usr/src/lib/libpam/libpam/pam_get_pass.c:60: warning: type mismatch in implicit declaration for built-in function `strlen' That's the characteristic signature of a missing #include string.h. -- Christian naddy Weisgerber [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
RE: Making bootable recovery CD using cdboot/loader fails
On 02-Feb-02 Michael Reifenberger wrote: Hi, a CD created using the following commands fails during loading /boot/kernel/kernel (just freezes up) under a recent current: cd /usr/tmp mkdir root dump -0a -f - / | ( cd root; restore -r -f - ) dd if=/dev/zero of=root/boot/boot.catalog count=5 mkisofs -b boot/cdboot -no-emul-boot -c boot/boot.catalog \ -r -J -V LiveCD \ -o root.iso root burncd -e -f /dev/acd0c blank data root.iso fixate Using a loader binary from 4.5-mini.iso works insofar that it works with a 4.5 /kernel but can't execute a -current kernel (loader version too low...) Any clues how to debug this problem? Hmm, I've never done a manual boot.catalog, always just ran mkisofs as so: mkisofs -b boot/cdboot -no-emul-boot -r -J -o foo.iso /some/dir and let mkisofs create its own boot.catalog. If you do that instead of creating one yourself does it work any better? If not, then you can revert the revisions that turned off all the debugging output, hook up a serial console on COM1, and prepare to be spammed with lots of debug output. :) -- John Baldwin [EMAIL PROTECTED]http://www.FreeBSD.org/~jhb/ Power Users Use the Power to Serve! - http://www.FreeBSD.org/ To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: buildworld fails in pam on Alpha
Now that current on Alpha builds a lot better than a few days back it fails in: cc -nostdinc -O -pipe -mcpu=ev4 -mcpu=ev4 -I/usr/src/lib/libpam/libpam/../../../contrib/libpam/libpam/include Already fixed. M -- o Mark Murray \_ FreeBSD Services Limited O.\_Warning: this .sig is umop ap!sdn To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: buildworld broken
Hi All, I've been battling with building current for a few days now, I've noticed that some of the errors I've been seeing have been fixed, but today in desperation I deleted my /usr/src tree and cvs'd it up again from the cvsup.au.freebsd.org server. Please turn off HTML in your mailer. Todays error: === libpam/modules/pam_ssh Already fixed. :-) M -- o Mark Murray \_ FreeBSD Services Limited O.\_Warning: this .sig is umop ap!sdn To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: pam_ssh world breakage (was: Re: cvs commit: src/lib/libpam Makefile.inc)
On Mon, Feb 04, 2002 at 09:17:17AM -0800, David O'Brien wrote: On Mon, Feb 04, 2002 at 03:30:32PM +0200, Ruslan Ermilov wrote: This delta breaks buildworld. gcc(1) has a known bug-feature of hiding some errors in standard system headers, making them invisible without -I. ... and compile now can't survive the WARNS=4. Not to mention there is ZERO way this code will pass WARNS=4 for GCC 3. Please Committers, do not try to WARNS code right now -- there just is no use. It will only get in the way later. If a new compiler generates more warnings then we should temporarily disable the warnings when the compiler is imported by setting WARNS=0 in bsd.sys.mk. All of the Makefile.inc's which set a default WARNS value for a subdirectory use WARNS?= for this reason (so they can be globally overridden until someone goes through and fixes the new ones). If you use the argument that one shouldn't set WARNS because a new compiler will cause the tree to break, then there's no point having it at all since that condition will always be true. Kris msg34130/pgp0.pgp Description: PGP signature
usb mouse won't work
FreeBSD does not want to work with my USB mouse. I believe it to be a problem with the motherboard, a Tyan S1834. The mouse does work under BeOS and Windows so I know it's not toast. Also, the mouse works with FreeBSD on my Tyan S1832 motherboard so that's a good sign too. Here's my kernel output: Feb 4 20:51:54 max syslogd: kernel boot file is /boot/kernel/kernel Feb 4 20:51:54 max kernel: Copyright (c) 1992-2002 The FreeBSD Project. Feb 4 20:51:54 max kernel: Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 Feb 4 20:51:54 max kernel: The Regents of the University of California. All rights reserved. Feb 4 20:51:54 max kernel: FreeBSD 5.0-CURRENT #1: Mon Feb 4 20:32:50 CST 2002 Feb 4 20:51:54 max kernel: [EMAIL PROTECTED]:/usr/obj/usr/src/sys/MAXKERNEL Feb 4 20:51:54 max kernel: Preloaded elf kernel /boot/kernel/kernel at 0xc040. Feb 4 20:51:54 max kernel: Preloaded elf module /boot/kernel/acpi.ko at 0xc04000a8. Feb 4 20:51:54 max kernel: Timecounter i8254 frequency 1193182 Hz Feb 4 20:51:54 max kernel: CPU: Pentium III/Pentium III Xeon/Celeron (933.03-MHz 686-class CPU) Feb 4 20:51:54 max kernel: Origin = GenuineIntel Id = 0x686 Stepping = 6 Feb 4 20:51:54 max kernel: Features=0x383fbffFPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,MMX,FXSR,S SE Feb 4 20:51:54 max kernel: real memory = 402587648 (393152K bytes) Feb 4 20:51:54 max kernel: avail memory = 387457024 (378376K bytes) Feb 4 20:51:54 max kernel: Programming 24 pins in IOAPIC #0 Feb 4 20:51:54 max kernel: IOAPIC #0 intpin 2 - irq 0 Feb 4 20:51:54 max kernel: IOAPIC #0 intpin 16 - irq 12 Feb 4 20:51:54 max kernel: IOAPIC #0 intpin 17 - irq 10 Feb 4 20:51:54 max kernel: IOAPIC #0 intpin 19 - irq 11 Feb 4 20:51:54 max kernel: FreeBSD/SMP: Multiprocessor System Detected: 2 CPUs Feb 4 20:51:54 max kernel: cpu0 (BSP): apic id: 0, version: 0x00040011, at 0xfee0 Feb 4 20:51:54 max kernel: cpu1 (AP): apic id: 1, version: 0x00040011, at 0xfee0 Feb 4 20:51:54 max kernel: io0 (APIC): apic id: 2, version: 0x00170011, at 0xfec0 Feb 4 20:51:54 max kernel: Pentium Pro MTRR support enabled Feb 4 20:51:54 max kernel: Using $PIR table, 8 entries at 0xc00fdb50 Feb 4 20:51:54 max kernel: npx0: math processor on motherboard Feb 4 20:51:55 max kernel: npx0: INT 16 interface Feb 4 20:51:55 max kernel: acpi0: VIA694 AWRDACPI on motherboard Feb 4 20:51:55 max kernel: acpi0: power button is handled as a fixed feature programming model. Feb 4 20:51:55 max kernel: Timecounter ACPI frequency 3579545 Hz Feb 4 20:51:55 max kernel: ACPI-0294: *** Error: Method execution failed, AE_AML_UNINITIALIZED_LOCAL Feb 4 20:51:55 max kernel: can't fetch resources for \_SB_.PCI0.FDC0 - AE_AML_UNINITIALIZED_LOCAL Feb 4 20:51:55 max kernel: acpi_timer0: 24-bit timer at 3.579545MHz port 0x4008-0x400b on acpi0 Feb 4 20:51:55 max kernel: acpi_cpu0: CPU on acpi0 Feb 4 20:51:55 max kernel: acpi_cpu1: CPU on acpi0 Feb 4 20:51:55 max kernel: acpi_button0: Power Button on acpi0 Feb 4 20:51:55 max kernel: acpi_pcib0: Host-PCI bridge port 0x5000-0x500f,0x4080-0x40ff,0x4000-0x407f,0xcf8-0xcff on acpi0 Feb 4 20:51:55 max kernel: pci0: PCI bus on acpi_pcib0 Feb 4 20:51:55 max kernel: agp0: VIA 82C691 (Apollo Pro) host to PCI bridge mem 0xe000-0xe3ff at device 0.0 on pci0 Feb 4 20:51:55 max kernel: pcib1: PCI-PCI bridge at device 1.0 on pci0 Feb 4 20:51:55 max kernel: pci1: PCI bus on pcib1 Feb 4 20:51:55 max kernel: pci1: display, VGA at device 0.0 (no driver attached) Feb 4 20:51:55 max kernel: isab0: PCI-ISA bridge at device 7.0 on pci0 Feb 4 20:51:55 max kernel: isa0: ISA bus on isab0 Feb 4 20:51:55 max kernel: atapci0: VIA 82C596 ATA66 controller port 0xd000-0xd00f at device 7.1 on pci0 Feb 4 20:51:55 max kernel: ata0: at 0x1f0 irq 14 on atapci0 Feb 4 20:51:55 max kernel: ata1: at 0x170 irq 15 on atapci0 Feb 4 20:51:55 max kernel: uhci0: VIA 83C572 USB controller port 0xd400-0xd41f irq 11 at device 7.2 on pci0 Feb 4 20:51:55 max kernel: usb0: VIA 83C572 USB controller on uhci0 Feb 4 20:51:55 max kernel: usb0: USB revision 1.0 Feb 4 20:51:55 max kernel: uhub0: VIA UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 Feb 4 20:51:55 max kernel: uhub0: 2 ports with 2 removable, self powered Feb 4 20:51:55 max kernel: ums0: Logitech USB-PS/2 Mouse, rev 1.00/1.20, addr 2, iclass 3/1 Feb 4 20:51:55 max kernel: ums0: 3 buttons Feb 4 20:51:55 max kernel: ums_attach: sc=0xc1eb0400 Feb 4 20:51:55 max kernel: ums_attach: X 8/8 Feb 4 20:51:55 max kernel: ums_attach: Y 16/8 Feb 4 20:51:55 max kernel: ums_attach: B1 0/1 Feb 4 20:51:55 max kernel: ums_attach: B2 1/1 Feb 4 20:51:55 max kernel: ums_attach: B3 2/1 Feb 4 20:51:55 max kernel: ums_attach: size=3, id=0 Feb 4 20:51:55 max kernel: uscanner0: Hewlett Packard Hewlett Packard ScanJet 5300C, rev 1.00/0.01, addr 3 Feb 4 20:51:55 max kernel: ahc0: Adaptec 2940 Ultra2
RE: buildworld broken
Hi David, I've cvsupd again using src-all and it seems to be working again now. Thanks, Anthony -Original Message- From: David Wolfskill [mailto:[EMAIL PROTECTED]] Sent: Tuesday, 5 February 2002 12:48 PM To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: buildworld broken From: [EMAIL PROTECTED] Date: Tue, 5 Feb 2002 11:45:32 +1100 Hi All, I've been battling with building current for a few days now, I've noticed that some of the errors I've been seeing have been fixed, but today in desperation I deleted my /usr/src tree and cvs'd it up again from the cvsup.au.freebsd.org server. What source collection(s) are you using? I suggest src-all. Cheers, david (links to my resume at http://www.catwhisker.org/~david) -- David H. Wolfskill[EMAIL PROTECTED] I believe it would be irresponsible (and thus, unethical) for me to advise, recommend, or support the use of any product that is or depends on any Microsoft product for any purpose other than personal amusement. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
On 20-Jan-02 Robert Watson wrote: On Sat, 19 Jan 2002, Alp Atici wrote: Is gcc 3.x going to be the default compiler starting from FBSD 5.x series? Is the development on current branch compiled using gcc 3.0 (or up)? Is 5.x series going to be based on a preemptible kernel? Can't answer the gcc question, but yes, John Baldwin currently has support for preemption in his SMPng development tree. The kernel is already somewhat preemptive. The kernel in 5.0 will certainly be preemptible, as making a kernel SMP safe makes it laregly preemptible (i.e, safe for preemption) as well. Making the kernel fully preemptive (i.e., we can switch tasks on any setrunqueue() if the conditions favor that) is actually a fairly esay thing to do, I'm just not sure how well it works right now. :) I just recently fixed some bugs in the alpha pmap code that should help out with getting our kernel closer to that goal. -- John Baldwin [EMAIL PROTECTED]http://www.FreeBSD.org/~jhb/ Power Users Use the Power to Serve! - http://www.FreeBSD.org/ To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
