* modutils-2.3.21: modprobe looping @ 2000-11-28 19:22 Kurt Garloff 2000-11-28 20:33 ` Keith Owens 0 siblings, 1 reply; 5+ messages in thread From: Kurt Garloff @ 2000-11-28 19:22 UTC (permalink / raw) To: Keith Owens; +Cc: Linux kernel list [-- Attachment #1.1: Type: text/plain, Size: 927 bytes --] Hi Keith, thanks for your modutils-2.3.21! During testing I found a problem: modprobe pppoe was recursing endlessly in build_stack(). Find attached the modules.dep that caused this: There is a circular dependency of pppoe on pppox on pppoe on .... modprobe has code to detect this in build_stack(), but it seems to not work. The old dep is thrown away and the new one is taken. And checked for dependencies again :-( Of course, one could also think of depmod making sure that such circular dependencies do not occur, but I guess, it's more robust if depmod has a working test in any case. Test system was a PPC machine running 2.4.0-test9. Do you have a fix for this? Regards, -- Kurt Garloff <garloff@suse.de> Eindhoven, NL GPG key: See mail header, key servers Linux kernel development SuSE GmbH, Nuernberg, FRG SCSI, Security [-- Attachment #1.2: modules.dep --] [-- Type: text/plain, Size: 857 bytes --] /lib/modules/2.4.0-test9/kernel/drivers/net/ppp_async.o: /lib/modules/2.4.0-test9/kernel/drivers/net/ppp_generic.o /lib/modules/2.4.0-test9/kernel/drivers/net/ppp_deflate.o: /lib/modules/2.4.0-test9/kernel/drivers/net/ppp_generic.o /lib/modules/2.4.0-test9/kernel/drivers/net/ppp_generic.o: /lib/modules/2.4.0-test9/kernel/drivers/net/ppp_synctty.o: /lib/modules/2.4.0-test9/kernel/drivers/net/ppp_generic.o /lib/modules/2.4.0-test9/kernel/drivers/net/pppoe.o: /lib/modules/2.4.0-test9/kernel/drivers/net/pppox.o \ /lib/modules/2.4.0-test9/kernel/drivers/net/ppp_generic.o /lib/modules/2.4.0-test9/kernel/drivers/net/pppox.o: /lib/modules/2.4.0-test9/kernel/drivers/net/pppoe.o \ /lib/modules/2.4.0-test9/kernel/drivers/net/ppp_generic.o /lib/modules/2.4.0-test9/kernel/drivers/net/skfp/skfp.o: /lib/modules/2.4.0-test9/kernel/drivers/net/slip.o: [-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --] ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: modutils-2.3.21: modprobe looping 2000-11-28 19:22 modutils-2.3.21: modprobe looping Kurt Garloff @ 2000-11-28 20:33 ` Keith Owens 2000-11-28 20:36 ` Rod Stewart 2000-11-28 21:24 ` Kurt Garloff 0 siblings, 2 replies; 5+ messages in thread From: Keith Owens @ 2000-11-28 20:33 UTC (permalink / raw) To: Kurt Garloff; +Cc: Linux kernel list On Tue, 28 Nov 2000 20:22:59 +0100, Kurt Garloff <garloff@suse.de> wrote: >Find attached the modules.dep that caused this: There is a circular >dependency of pppoe on pppox on pppoe on .... The kernel code is broken. Circular dependencies make no sense, the pppoe maintainer agrees and I thought that bug was fixed. >modprobe has code to detect this in build_stack(), but it seems to not work. >The old dep is thrown away and the new one is taken. And checked for >dependencies again :-( >Of course, one could also think of depmod making sure that such circular >dependencies do not occur, but I guess, it's more robust if depmod has a >working test in any case. Circular dependencies are not supported, nor are they correrctly detected. The existing code to walk the inter module relationships, including dependencies, alias, probe, probeall, before and after statements is a mess. It just grew over the years with special cases being added and is not robust. In modutils 2.5 the entire code will be discarded and replaced with a standard graph walking algorithm with loop detection and back tracking instead of special case code. That might change some modutils behaviour in rare cases and I do not want to change its behaviour just before kernel 2.4 is released. I have a list of changes that might break backwards compatibility waiting for modutils 2.5, this is just one of them. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/ ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: modutils-2.3.21: modprobe looping 2000-11-28 20:33 ` Keith Owens @ 2000-11-28 20:36 ` Rod Stewart 2000-11-28 21:24 ` Kurt Garloff 1 sibling, 0 replies; 5+ messages in thread From: Rod Stewart @ 2000-11-28 20:36 UTC (permalink / raw) To: Keith Owens; +Cc: Kurt Garloff, Linux kernel list On Wed, 29 Nov 2000, Keith Owens wrote: > On Tue, 28 Nov 2000 20:22:59 +0100, > Kurt Garloff <garloff@suse.de> wrote: > >Find attached the modules.dep that caused this: There is a circular > >dependency of pppoe on pppox on pppoe on .... > > The kernel code is broken. Circular dependencies make no sense, the > pppoe maintainer agrees and I thought that bug was fixed. It is fixed in test10/11. -Rms - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/ ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: modutils-2.3.21: modprobe looping 2000-11-28 20:33 ` Keith Owens 2000-11-28 20:36 ` Rod Stewart @ 2000-11-28 21:24 ` Kurt Garloff 2000-11-29 0:48 ` Kurt Garloff 1 sibling, 1 reply; 5+ messages in thread From: Kurt Garloff @ 2000-11-28 21:24 UTC (permalink / raw) To: Keith Owens; +Cc: Linux kernel list [-- Attachment #1: Type: text/plain, Size: 3250 bytes --] On Wed, Nov 29, 2000 at 07:33:53AM +1100, Keith Owens wrote: > On Tue, 28 Nov 2000 20:22:59 +0100, > Kurt Garloff <garloff@suse.de> wrote: > >Find attached the modules.dep that caused this: There is a circular > >dependency of pppoe on pppox on pppoe on .... > > The kernel code is broken. Circular dependencies make no sense, the > pppoe maintainer agrees and I thought that bug was fixed. No question that circular dependencies are a bug in the kernel/modules. I just thought, that it would be nice if modprobe would not run crazy and end up in an endless loop, possible risking the machine to run out of mem with all the nice side effects this has on Linux. This is especially bad, because modprobe is called invisibly by kmod. Here's what happened on the test machine (2.4.0-test9 on Dual Power3): Swapping like crazy; getting unresponsive; ... Here's the last output top was able to produce: 10:12pm up 5:56, 3 users, load average: 1.57, 0.41, 0.16 65 processes: 63 sleeping, 2 running, 0 zombie, 0 stopped CPU states: 36.9% user, 19.0% system, 0.0% nice, 44.0% idle Mem: 509320K av, 508272K used, 1048K free, 28172K shrd, 76K buff Swap: 131064K av, 107388K used, 23676K free 7048K cached PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND 23543 root 15 0 574M 472M 328 R 0 85.8 95.0 0:55 modprobe 2 root 11 0 0 0 0 DW 0 26.3 0.0 0:06 kswapd 23544 root 14 0 568 428 328 R 0 2.1 0.0 0:01 top [...] (10 mins later) I'm afraid I can't wait until the machine recovers, if it ever will. [2.2 with AA VM would have started to kill processes (which is bad) till it would have finally killed modprobe (which is good); but I'm afraid that 2.4 will just not survive at all :-( ] > Circular dependencies are not supported, nor are they correrctly > detected. The existing code to walk the inter module relationships, > including dependencies, alias, probe, probeall, before and after > statements is a mess. It just grew over the years with special cases > being added and is not robust. This is a pity, because I think the current behaviour is not acceptable, as it can kill the machine by just being invoked by kmod. I will try to make sense out of the code and make sure that modprobe will not go crazy, by either detecting loops (if I can do that in a way wihtout breaking things) or by limiting the recursion depth. I'll send you a patch. > In modutils 2.5 the entire code will be discarded and replaced with a > standard graph walking algorithm with loop detection and back tracking > instead of special case code. That might change some modutils > behaviour in rare cases and I do not want to change its behaviour just > before kernel 2.4 is released. I have a list of changes that might > break backwards compatibility waiting for modutils 2.5, this is just > one of them. Looking forward to it. Regards, -- Kurt Garloff <garloff@suse.de> Eindhoven, NL GPG key: See mail header, key servers Linux kernel development SuSE GmbH, Nuernberg, FRG SCSI, Security [-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --] ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: modutils-2.3.21: modprobe looping 2000-11-28 21:24 ` Kurt Garloff @ 2000-11-29 0:48 ` Kurt Garloff 0 siblings, 0 replies; 5+ messages in thread From: Kurt Garloff @ 2000-11-29 0:48 UTC (permalink / raw) To: Keith Owens, Linux kernel list [-- Attachment #1.1: Type: text/plain, Size: 1049 bytes --] On Tue, Nov 28, 2000 at 10:24:48PM +0100, Kurt Garloff wrote: > This is a pity, because I think the current behaviour is not acceptable, > as it can kill the machine by just being invoked by kmod. > I will try to make sense out of the code and make sure that modprobe > will not go crazy, by either detecting loops (if I can do that in a way > wihtout breaking things) or by limiting the recursion depth. I'll send > you a patch. Find it attached. As the dependency generation looked indeed rather cumbersome to me, I didn't really touch it. I just did implement the recursion limit to prevent the modprobe process grabbing all the memory of the system ... root@cantaloupe:~ > modprobe -k pppoe modprobe: Too deep recursion in module dependencies! modprobe: Circular dependency? pppox pppoe Aborted Regards, -- Kurt Garloff <garloff@suse.de> Eindhoven, NL GPG key: See mail header, key servers Linux kernel development SuSE GmbH, Nuernberg, FRG SCSI, Security [-- Attachment #1.2: modutils-2.3.21-limitrecursion.diff --] [-- Type: text/plain, Size: 1911 bytes --] diff -uNr modutils-2.3.21/insmod/modprobe.c modutils-2.3.21.SuSE/insmod/modprobe.c --- modutils-2.3.21/insmod/modprobe.c Tue Nov 21 10:10:54 2000 +++ modutils-2.3.21.SuSE/insmod/modprobe.c Wed Nov 29 00:39:54 2000 @@ -363,6 +363,9 @@ return desc; } +/* KG, 2000-11-29: Limit recursion in build_stack() */ +#define BSTACKRECMAX 128 +static int bstackrecurs = 0; /* * Create a linked list of the stack of modules needed in the kernel * in order to satisfy a request for "name" @@ -395,6 +398,15 @@ if (rev && strcmp(rev->kname, desc->kname) == 0) return 0; + if (++bstackrecurs > BSTACKRECMAX) { + fprintf (stderr, "modprobe: Too deep recursion in module dependencies!\n"); + fprintf (stderr, "modprobe: Circular dependency?"); + for (lp = *stack; lp; lp = lp->next) + fprintf (stderr, " %s", lp->item.d->kname); + fprintf (stderr, "\n"); + abort (); + } + /* * Get the modules that this one should pull in * according to modules.conf ("above") @@ -414,7 +426,7 @@ rev ? rev : desc, MODE_ABOVE, 1); if (r < 0) - return r; + { bstackrecurs--; return r; } } } } @@ -424,7 +436,7 @@ if (strcmp(desc->kname, lp->item.d->kname) == 0) { /* Avoid infinite loops ("above") */ if (rev && rev == lp->item.d) - return 0; + { bstackrecurs--; return 0; } /* else */ /* Remove the entry, it is in the wrong place */ if (prev) @@ -452,7 +464,7 @@ dir, rev, MODE_BELOW, 1)) < 0) - return r; + { bstackrecurs--; return r; } } } } @@ -465,10 +477,10 @@ for (nod = desc->nod->dep; nod; nod = nod->next) { if ((r = build_stack(stack, nod->key, nod->str, dir, rev, MODE_NORMAL, 0)) < 0) - return r; + { bstackrecurs--; return r; } } - return 0; + bstackrecurs--; return 0; } /* For debugging */ [-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --] ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2000-11-29 1:21 UTC | newest] Thread overview: 5+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2000-11-28 19:22 modutils-2.3.21: modprobe looping Kurt Garloff 2000-11-28 20:33 ` Keith Owens 2000-11-28 20:36 ` Rod Stewart 2000-11-28 21:24 ` Kurt Garloff 2000-11-29 0:48 ` Kurt Garloff
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.