SCSI: Fix NULL pointer dereference in runtime PM
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 1 line · Greg Kroah-Hartman
4.1-stable review patch. If anyone has any objections, please let me know.
It turns out that this patch causes problems with the sr driver. A reversion and alternate solution have already been submitted to James Bottomley (http://marc.info/?l=linux-scsi&m=144185206825609&w=2). We ought to hold off putting it into the -stable kernels.
Alan Stern
Quoted 31 lines
------------------
From: Alan Stern <[email protected]>
commit
49718f0fb8c9af192b33d8af3a2826db04025371upstream.The routines in
scsi_rpm.cassume that if a runtime-PM callback is invoked for a SCSI device, it can only mean that the device's driver has asked the block layer to handle the runtime power management (by callingblk_pm_runtime_init(), which among other things setsq->dev).However, this assumption turns out to be wrong for things like the ses driver. Normally ses devices are not allowed to do runtime PM, but userspace can override this setting. If this happens, the kernel gets a NULL pointer dereference when
blk_post_runtime_resume()tries to use the uninitializedq->devpointer.This patch fixes the problem by calling the block layer's runtime-PM routines only if the device's driver really does have a runtime-PM callback routine. Since ses doesn't define any such callbacks, the crash won't occur.
This fixes Bugzilla #101371.
Signed-off-by Alan Stern <[email protected]>Reported-by Stanisław Pitucha <[email protected]>Reported-by Ilan Cohen <[email protected]>Tested-by Ilan Cohen <[email protected]>Reviewed-by Johannes Thumshirn <[email protected]>Signed-off-by James Bottomley <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>
Greg Kroah-Hartman
maintainerQuoted 8 lines · Alan Stern
Quoted 1 line · Greg Kroah-Hartman
4.1-stable review patch. If anyone has any objections, please let me know.
It turns out that this patch causes problems with the sr driver. A reversion and alternate solution have already been submitted to James Bottomley (http://marc.info/?l=linux-scsi&m=144185206825609&w=2). We ought to hold off putting it into the -stable kernels.
I prefer to take patches like this into a stable release, so as to spur the maintainer on to also get the fix into Linus's tree, which I don't see happening yet. I don't like diverging from what Linus has for obvious reasons.
thanks,
greg k-h