]> git.ipfire.org Git - thirdparty/linux.git/commit
Drivers: hv: Always do Hyper-V panic notification in hv_kmsg_dump()
authorMichael Kelley <mhklinux@outlook.com>
Wed, 31 Dec 2025 20:14:47 +0000 (12:14 -0800)
committerWei Liu <wei.liu@kernel.org>
Thu, 15 Jan 2026 07:01:33 +0000 (07:01 +0000)
commit49f49d47af67f8a7b221db1d758fc634242dc91a
tree629789fc87529ab7ccc2485cde59e8142867fb4d
parentfc558188f251f5df6dd11ac409d668733d9baac7
Drivers: hv: Always do Hyper-V panic notification in hv_kmsg_dump()

hv_kmsg_dump() currently skips the panic notification entirely if it
doesn't get any message bytes to pass to Hyper-V due to an error from
kmsg_dump_get_buffer(). Skipping the notification is undesirable because
it leaves the Hyper-V host uncertain about the state of a panic'ed guest.

Fix this by always doing the panic notification, even if bytes_written
is zero. Also ensure that bytes_written is initialized, which fixes a
kernel test robot warning. The warning is actually bogus because
kmsg_dump_get_buffer() happens to set bytes_written even if it fails, and
in the kernel test robot's CONFIG_PRINTK not set case, hv_kmsg_dump() is
never called. But do the initialization for robustness and to quiet the
static checker.

Fixes: 9c318a1d9b50 ("Drivers: hv: move panic report code from vmbus to hv early init code")
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/all/202512172103.OcUspn1Z-lkp@intel.com/
Signed-off-by: Michael Kelley <mhklinux@outlook.com>
Reviewed-by: Roman Kisel <vdso@mailbox.org>
Signed-off-by: Wei Liu <wei.liu@kernel.org>
drivers/hv/hv_common.c