[PATCH] nvme: fix possible initialization failure for P4500/P4600 SSDs

Keith Busch kbusch at kernel.org
Fri Dec 10 09:31:45 PST 2021


On Thu, Jun 17, 2021 at 12:05:52PM -0400, Zheng Wu wrote:
> From: Wu Zheng <wu.zheng at intel.com>
> 
> Subnqn isn't supported in P4500/P4600 SSDs.
> It's a fixed unavailable value in each P4500/P4600 SSD.

According to ark.intel.com, P4500/4600 were released nearly 4 years ago.
My understanding was these were a popular product. Why are we hearing
about this issue only now? Is this issue specific to a particular
firmware version, or has Linux really been incompatible with these this
entire time?

> @@ -3326,7 +3326,8 @@ static const struct pci_device_id nvme_id_table[] = {
>  				NVME_QUIRK_DEALLOCATE_ZEROES, },
>  	{ PCI_VDEVICE(INTEL, 0x0a54),	/* Intel P4500/P4600 */
>  		.driver_data = NVME_QUIRK_STRIPE_SIZE |
> -				NVME_QUIRK_DEALLOCATE_ZEROES, },
> +				NVME_QUIRK_DEALLOCATE_ZEROES |
> +				NVME_QUIRK_IGNORE_DEV_SUBNQN, },
>  	{ PCI_VDEVICE(INTEL, 0x0a55),	/* Dell Express Flash P4600 */
>  		.driver_data = NVME_QUIRK_STRIPE_SIZE |
>  				NVME_QUIRK_DEALLOCATE_ZEROES, },

So the generic P4600 is broken, but Dell's is not?



More information about the Linux-nvme mailing list