On 09/18/2009 10:58 PM, David Cantrell wrote: > On Fri, 18 Sep 2009, Chris Lumens wrote: > >>> Avoid SIGSEGV in addOption() in loader/modules.c. Just because found >>> does not equal 0 does not mean we found the module requested. >> >> Can you explain what's going on in this patch and function? Got a bug >> number? > It's actually related to the first patch where I reduce the > mlLoadModuleSet() > call on s390x. When mlLoadModuleSet() was called with floppy, pcspkr, > etc on > s390x, the console was getting a backtrace dumped and it fell in to > addOption(). > > The assumption in addOption() is that the module is always found, which Not that I would understand all the details of the existing code, but readModuleOpts() seems to check with isValidModule() before calling addOption() while doLoadModule() just calls addOption() unconditionally. I guess your fix in addOption() would make it safe in all cases. > isn't > the case. We can prevent that by only calling mlLoadModuleSet() with > modules > that exist on the platform, but in order to prevent these backtraces in the > future, I figured it was worth fixing addOption() too. +1 Steffen Linux on System z Development IBM Deutschland Research & Development GmbH Vorsitzender des Aufsichtsrats: Martin Jetter Geschäftsführung: Erich Baier Sitz der Gesellschaft: Böblingen Registergericht: Amtsgericht Stuttgart, HRB 243294 _______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
