[pve-devel] [PATCH pve-kernel] fix #5448: cherry-pick SCSI VPD fix

[Fabian Grünbichler]

ping, it's in linux-next in the meantime :)

On May 23, 2024 11:38 am, Fabian Grünbichler wrote:
> Signed-off-by: Fabian Grünbichler <f.gruenbichler at proxmox.com>
> ---
> 6.8/master only, patch straight from the linked LKML thread
> 
> affected user confirmed it fixes the issue, patch is very small and
> straight-forward.
> 
>  ...-devices-which-return-an-unusually-l.patch | 51 +++++++++++++++++++
>  1 file changed, 51 insertions(+)
>  create mode 100644 patches/kernel/0016-scsi-core-Handle-devices-which-return-an-unusually-l.patch
> 
> diff --git a/patches/kernel/0016-scsi-core-Handle-devices-which-return-an-unusually-l.patch b/patches/kernel/0016-scsi-core-Handle-devices-which-return-an-unusually-l.patch
> new file mode 100644
> index 0000000..513d79b
> --- /dev/null
> +++ b/patches/kernel/0016-scsi-core-Handle-devices-which-return-an-unusually-l.patch
> @@ -0,0 +1,51 @@
> +From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
> +From: "Martin K. Petersen" <martin.petersen at oracle.com>
> +Date: Mon, 20 May 2024 22:30:40 -0400
> +Subject: [PATCH] scsi: core: Handle devices which return an unusually large
> + VPD page count
> +MIME-Version: 1.0
> +Content-Type: text/plain; charset=UTF-8
> +Content-Transfer-Encoding: 8bit
> +
> +Peter Schneider reported that a system would no longer boot after
> +updating to 6.8.4.  Peter bisected the issue and identified commit
> +b5fc07a5fb56 ("scsi: core: Consult supported VPD page list prior to
> +fetching page") as being the culprit.
> +
> +Turns out the enclosure device in Peter's system reports a byteswapped
> +page length for VPD page 0. It reports "02 00" as page length instead
> +of "00 02". This causes us to attempt to access 516 bytes (page length
> ++ header) of information despite only 2 pages being present.
> +
> +Limit the page search scope to the size of our VPD buffer to guard
> +against devices returning a larger page count than requested.
> +
> +Cc: stable at vger.kernel.org
> +Reported-by: Peter Schneider <pschneider1968 at googlemail.com>
> +Tested-by: Peter Schneider <pschneider1968 at googlemail.com>
> +Fixes: b5fc07a5fb56 ("scsi: core: Consult supported VPD page list prior to fetching page")
> +Link: https://lore.kernel.org/all/eec6ebbf-061b-4a7b-96dc-ea748aa4d035@googlemail.com/
> +Signed-off-by: Martin K. Petersen <martin.petersen at oracle.com>
> +Signed-off-by: Fabian Grünbichler <f.gruenbichler at proxmox.com>
> +---
> + drivers/scsi/scsi.c | 7 +++++++
> + 1 file changed, 7 insertions(+)
> +
> +diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
> +index 8cad9792a562..b3ff3a633a1e 100644
> +--- a/drivers/scsi/scsi.c
> ++++ b/drivers/scsi/scsi.c
> +@@ -350,6 +350,13 @@ static int scsi_get_vpd_size(struct scsi_device *sdev, u8 page)
> + 		if (result < SCSI_VPD_HEADER_SIZE)
> + 			return 0;
> + 
> ++		if (result > sizeof(vpd)) {
> ++			dev_warn_once(&sdev->sdev_gendev,
> ++				      "%s: long VPD page 0 length: %d bytes\n",
> ++				      __func__, result);
> ++			result = sizeof(vpd);
> ++		}
> ++
> + 		result -= SCSI_VPD_HEADER_SIZE;
> + 		if (!memchr(&vpd[SCSI_VPD_HEADER_SIZE], page, result))
> + 			return 0;
> -- 
> 2.39.2
> 
> 
> 
> _______________________________________________
> pve-devel mailing list
> pve-devel at lists.proxmox.com
> https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
>