* Jesse Barnes <jbarnes@xxxxxxxxxxxxxxxx>: > On Sunday, September 21, 2008 11:26 pm Zhao, Yu wrote: > > The get_##name in pci_hotplug_core.c always returns 0, which fails its > > callers to conclude right operations. Change it to return proper values on > > failure. > > > > Signed-off-by: Yu Zhao <yu.zhao@xxxxxxxxx> > > I'd like to get Alex or Kenji-san's ack on this. Also please don't reply to > another thread when posting a completely new message; it messes up my > threading. Well, the changelog comment is a bit wrong. get_##name does _not_ always return 0, it either returns an error or the number of chars that sprintf wrote into the buffer. I think what Yu meant was that his patch gives proper behavior during a race condition between module unload and reading a hotplug file. Previously, if the module unload won the race, the user got a 0. Now, if the module unload wins the race, the user gets -ENODEV. This patch looks good to me. Thanks to Willy for additional confirmation. Acked-by: Alex Chiang <achiang@xxxxxx> /ac -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
