** Description changed:

+ [Impact]
+ 
  We identified a problem that is causing slow logging to the console for
  customers.
  
  The following commit resolves this issue as well as other cache relates 
issues:
  325073ae3485 ("video: hyperv_fb: Fix the cache type when mapping the VRAM")
  
  (marcelo.cerri: it's actually commit id
  5f1251a48c17b54939d7477305e39679a565382c in the mainline kernel)
  
  Patch details from it's commit message:
  
  x86 Hyper-V used to essentially always overwrite the effective cache type
  of guest memory accesses to WB. This was problematic in cases where there
  is a physical device assigned to the VM, since that often requires that
  the VM should have control over cache types. Thus, on newer Hyper-V since
  2018, Hyper-V always honors the VM's cache type, but unexpectedly Linux VM
  users start to complain that Linux VM's VRAM becomes very slow, and it
  turns out that Linux VM should not map the VRAM uncacheable by ioremap().
  Fix this slowness issue by using ioremap_cache().
  
  With this change, the VRAM on new Hyper-V is as fast as regular RAM, so
  it's no longer necessary to use the hacks we added to mitigate the
  slowness, i.e. we no longer need to allocate physical memory and use
  it to back up the VRAM in Generation-1 VM, and we also no longer need to
  allocate physical memory to back up the framebuffer in a Generation-2 VM
  and copy the framebuffer to the real VRAM. A further big change will
  address these for v5.11.
  
  Microsoft would like to request this patch in all supported releases.
+ 
+ [Test Case]
+ 
+ The test is very simple. When using the console in a local Hyper-V
+ instance and running a command with output with several lines (ie `ls -l
+ /`) the output delay is very noticeable and it should be instantaneous
+ with the fix.
+ 
+ [Where problems could occur]
+ 
+ The change might cause regressions in the hyperv_fb driver, affecting
+ one of the alternatives users have to debug problems.

-- 
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux-azure in Ubuntu.
https://bugs.launchpad.net/bugs/1908569

Title:
  video: hyperv_fb: Fix the cache type when mapping the VRAM

Status in linux-azure package in Ubuntu:
  New
Status in linux-azure-4.15 package in Ubuntu:
  New
Status in linux-azure source package in Bionic:
  Invalid
Status in linux-azure-4.15 source package in Bionic:
  New
Status in linux-azure source package in Focal:
  New
Status in linux-azure-4.15 source package in Focal:
  Invalid
Status in linux-azure source package in Groovy:
  New
Status in linux-azure-4.15 source package in Groovy:
  Invalid

Bug description:
  [Impact]

  We identified a problem that is causing slow logging to the console
  for customers.

  The following commit resolves this issue as well as other cache relates 
issues:
  325073ae3485 ("video: hyperv_fb: Fix the cache type when mapping the VRAM")

  (marcelo.cerri: it's actually commit id
  5f1251a48c17b54939d7477305e39679a565382c in the mainline kernel)

  Patch details from it's commit message:

  x86 Hyper-V used to essentially always overwrite the effective cache type
  of guest memory accesses to WB. This was problematic in cases where there
  is a physical device assigned to the VM, since that often requires that
  the VM should have control over cache types. Thus, on newer Hyper-V since
  2018, Hyper-V always honors the VM's cache type, but unexpectedly Linux VM
  users start to complain that Linux VM's VRAM becomes very slow, and it
  turns out that Linux VM should not map the VRAM uncacheable by ioremap().
  Fix this slowness issue by using ioremap_cache().

  With this change, the VRAM on new Hyper-V is as fast as regular RAM, so
  it's no longer necessary to use the hacks we added to mitigate the
  slowness, i.e. we no longer need to allocate physical memory and use
  it to back up the VRAM in Generation-1 VM, and we also no longer need to
  allocate physical memory to back up the framebuffer in a Generation-2 VM
  and copy the framebuffer to the real VRAM. A further big change will
  address these for v5.11.

  Microsoft would like to request this patch in all supported releases.

  [Test Case]

  The test is very simple. When using the console in a local Hyper-V
  instance and running a command with output with several lines (ie `ls
  -l /`) the output delay is very noticeable and it should be
  instantaneous with the fix.

  [Where problems could occur]

  The change might cause regressions in the hyperv_fb driver, affecting
  one of the alternatives users have to debug problems.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-azure/+bug/1908569/+subscriptions

-- 
Mailing list: https://launchpad.net/~kernel-packages
Post to     : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help   : https://help.launchpad.net/ListHelp

Reply via email to