Quoting sukadev@xxxxxxxxxx (sukadev@xxxxxxxxxx): > Serge E. Hallyn [serue@xxxxxxxxxx] wrote: > | Quoting sukadev@xxxxxxxxxx (sukadev@xxxxxxxxxx): > | > | + } else > | > | + ERROR("Unexpected item, '%s'\n", item); > | > | + } > | > | + DEBUG("restore_fifo() fd %d, len %d, got 'END FIFO'\n", > | > | + fifofdinfo->fi_fd, fifofdinfo->fi_length); > | > | + return 0; > | > | +} > | > > | > fifofdinfo can be NULL if application does not have any fifos and > | > the above DEBUG() will SIGSEGV. Following updated patch fixes this. > | > Thanks Serge for reporting the bug. > | > | Thanks, Suka. This version works for both pipe and sleep testcases. > | > | Though, interestingly, it segfaults on the -lxc kernel. It succeeds on > | a recent -git from Linus' tree, though. > > I have been testing 2.6.25-mm1. > > I run into the semun_undo underflow bug with -lxc. Is that the same > bug you are running into ? I'm not sure... Not looking into it too deeply. Hopefully Kathy is coming out with a new -lxc in the next few days, and if it still causes me problems then I'll really look into it. -serge _______________________________________________ Containers mailing list Containers@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/containers