* David Miller <davem@xxxxxxxxxxxxx> [161221 14:13]: > From: "Liam R. Howlett" <Liam.Howlett@xxxxxxxxxx> > Date: Wed, 21 Dec 2016 13:43:38 -0500 > > > Check if the PCI memory starting address is disabled and return -EINVAL. > > This will have minimal impact to performance as it is only during the > > set up of the mmap and will avoid consequences of a failed load/store. > > > > Signed-off-by: Liam R. Howlett <Liam.Howlett@xxxxxxxxxx> > > Some X server or similar userland application may be depending upon > the current behavior. > > I'd rather not check, and let the user take the SIGBUS if it tries > to access it. > > Furthermore, the user can legitimately map a disabled BAR then enable > it with PCI config accesses. > > So I'm not applying this, sorry. I appreciate you taking the time to tell me how this patch is not the right way to fix this issue. I will rework my solution to this issue to behave in a more appropriate manner. I should have mentioned that this patch isn't to remap a SIGBUS to an -EINVAL, it's to stop the system from a panic due to a non-recoverable error occurring when trying to read an mmap'ed resource that is not enabled. Thank you, Liam -- To unsubscribe from this list: send the line "unsubscribe sparclinux" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
