Public bug reported:

SRU Justification

[Impact]

* Microsoft requested patch to improve efficiency of SCSI I/O on Hyper-V
by issuing IO on same CPU as return CPU.

[Fix]

* clean cherry-pick of linux-next patch:
    * b69ffeaa0ae4: "scsi: storvsc: Prefer returning channel with the same CPU 
as on the I/O issuing CPU"

[Test Plan]

* Compile tested
* Microsoft to test

[Regression potential]

* Patch limited to SCSI module. Possible regressions could result in
unexpected performance desegregation in non-Hyper-V cases.

[Other info]

* SF#00420316

** Affects: linux-azure (Ubuntu)
     Importance: Undecided
         Status: New

** Affects: linux-azure (Ubuntu Noble)
     Importance: Undecided
         Status: New

** Affects: linux-azure (Ubuntu Plucky)
     Importance: Undecided
         Status: New

** Affects: linux-azure (Ubuntu Questing)
     Importance: Undecided
         Status: New

** Also affects: linux-azure (Ubuntu Plucky)
   Importance: Undecided
       Status: New

** Also affects: linux-azure (Ubuntu Questing)
   Importance: Undecided
       Status: New

** Also affects: linux-azure (Ubuntu Noble)
   Importance: Undecided
       Status: New

** Description changed:

  SRU Justification
  
  [Impact]
  
  * Microsoft requested patch to improve efficiency of SCSI I/O on Hyper-V
  by issuing IO on same CPU as return CPU.
  
  [Fix]
  
- * clean cherry-pick of upstream patch:
-     * b69ffeaa0ae4: "scsi: storvsc: Prefer returning channel with the same 
CPU as on the I/O issuing CPU"
+ * clean cherry-pick of linux-next patch:
+     * b69ffeaa0ae4: "scsi: storvsc: Prefer returning channel with the same 
CPU as on the I/O issuing CPU"
  
  [Test Plan]
  
  * Compile tested
  * Microsoft to test
  
  [Regression potential]
  
  * Patch limited to SCSI module. Possible regressions result in
  unexpected performance desegregation in non-Hyper-V cases.
  
  [Other info]
  
  * SF#00420316

** Description changed:

  SRU Justification
  
  [Impact]
  
  * Microsoft requested patch to improve efficiency of SCSI I/O on Hyper-V
  by issuing IO on same CPU as return CPU.
  
  [Fix]
  
  * clean cherry-pick of linux-next patch:
      * b69ffeaa0ae4: "scsi: storvsc: Prefer returning channel with the same 
CPU as on the I/O issuing CPU"
  
  [Test Plan]
  
  * Compile tested
  * Microsoft to test
  
  [Regression potential]
  
- * Patch limited to SCSI module. Possible regressions result in
+ * Patch limited to SCSI module. Possible regressions could result in
  unexpected performance desegregation in non-Hyper-V cases.
  
  [Other info]
  
  * SF#00420316

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2128842

Title:
  [STORVSC] [Backport] Backport for Linux storage driver patch for
  fixing performance issues

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-azure/+bug/2128842/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to