On Thu, Mar 29, 2018 at 04:28:54PM +0100, Ben Hutchings wrote:
On Mon, 2018-03-19 at 19:04 +0100, Greg Kroah-Hartman wrote:
4.4-stable review patch. If anyone has any objections, please let me know.
From: Prarit Bhargava prarit@redhat.com
[ Upstream commit fda78d7a0ead144f4b2cdb582dcba47911f4952c ]
[...]
The MSI disabling code was added by d52877c7b1af ("pci/irq: let pci_device_shutdown to call pci_msi_shutdown v2") because a driver left MSI enabled and kdump failed because the kexeced kernel wasn't prepared to receive the MSI interrupts.
Subsequent commits 1851617cd2da ("PCI/MSI: Disable MSI at enumeration even if kernel doesn't support MSI")
This went into 4.2 and hasn't been backported.
To backport that to 3.18.y takes 2 other patches to get it "correct".
and e80e7edc55ba ("PCI/MSI: Initialize MSI capability for all architectures")
This went into 4.5 and hasn't been backported.
Yeah, but that really isn't needed unless 4d9aac397a5d ("powerpc/PCI: Disable MSI/MSI-X interrupts at PCI probe time in OF case") is also there, along with 1851617cd2da.
So I think 3.18-stable and 4.4-stable will need backports of the later changes in order for this to work properly.
I think I'm just going to revert this as no one is complaining about this issue on 4.4.y and 3.18.y kernels. Thanks for the review.
greg k-h