Re: [PATCH v2 13/20] scsi: ufs: Fix a deadlock in the error handler

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, 2021-11-30 at 11:32 -0800, Bart Van Assche wrote:
> On 11/30/21 12:54 AM, Bean Huo wrote:
> > We are looking for alternative
> > methods: for example, to fix this problem from the SCSI layer;
> > Add a new dedicated hardware device management queue on the UFS
> > device
> > side.
> 
> Hi Bean,
> 
> I don't think that there are any alternatives. If all host controller
> tags
> are in use, it is not possible to implement a mechanism in software
> to
> create an additional tag. Additionally, stealing a tag is not
> possible since
> no new request can be submitted to a UFS controller if all tags are
> in use.
> 
> Thanks,
> 
> Bart.

Bart, 

How about adding a hardware-specific UFS device management queue to the
UFS device? Compared with NVMe and eMMC CMDQ, they both have dedicated
tags for device management commands. NVMe is queue 0, eMMC CMDQ is CMDQ
slot 31.

I'm thinking about the benefits of using this device manage queue. If
the benefits are significant, we can submit the Jedec proposal for the
UFS equipment management queue.

Kind regards,
Bean




[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [SCSI Target Devel]     [Linux SCSI Target Infrastructure]     [Kernel Newbies]     [IDE]     [Security]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux ATA RAID]     [Linux IIO]     [Samba]     [Device Mapper]

  Powered by Linux