Re: [PATCH RFC] hisi_sas_v3: multiqueue support

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

 



On 31/05/2019 14:18, Hannes Reinecke wrote:
On 5/31/19 11:42 AM, John Garry wrote:

-static int hisi_sas_slot_index_alloc(struct hisi_hba *hisi_hba,
-                     struct scsi_cmnd *scsi_cmnd)
+static int hisi_sas_slot_index_alloc(struct hisi_hba *hisi_hba)
 {
     int index;
     void *bitmap = hisi_hba->slot_index_tags;
     unsigned long flags;

-    if (scsi_cmnd)
-        return scsi_cmnd->request->tag;
-
     spin_lock_irqsave(&hisi_hba->lock, flags);
     index = find_next_zero_bit(bitmap, hisi_hba->slot_index_count,
                    hisi_hba->last_slot_index + 1);

Then you switch to hisi_sas_slot_index_alloc() for allocating the
unique
tag via spin_lock & find_next_zero_bit(). Do you think this way is more
efficient than blk-mq's sbitmap?

These are not fast path, as used only for TMF, internal IO, etc.

Having said that, hopefully we can move to scsi_{get,put}_reserved_cmd()
when available, so that the LLDD has to stop managing them.


slot_index_alloc() is only used for commands which do _not_ have a tag
(eg internal commands), or for v2 hardware which has weird allocation
rules.

But this patch has switched to this allocation unconditionally for all
commands:


As Hannes said, v2 had a few bugs which meant that we had to make a
specific version of this function for that hw revision, cf.
slot_index_alloc_quirk_v2_hw(), and it cannot use request queue tag.

But, indeed, we could consider sbitmap for v2 hw. I'm not sure if it
would help, considering the weird rules.

We should be able to get away by shifting all tags by 1 to the left,
and adding '1' to SMP/SAS commands, and '2' to STP commands, no?
Then index '0' would be free, and the allocation rules are satisfied.


The crazy (escalating from weird) rules to workaround the HW bug(s) mean that we need to chop up the command tag range into blocks of 32 even tag indexes per SATA device; this means that SATA device #0 can use 64, 66, 68, 70...126. device #1 can use 128, 130, 132,..., device #2 can use 192, 194,...

I don't know how you can take a rq tag and generate a command tag suitable for a SATA device.

Thanks!

John

I'll see to whip up a patch.

Cheers,

Hannes






[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