Arch Linux User Repository

• Leaving eclairvoyant's comment pinned because it contains some useful information.
• Patch policy is different, but similar.
  • Primary purpose of this package are the ACS override and i915 VGA arbiter patches.
  • Upstream Arch patches will be applied to maintain parity with the standard kernel.
  • No other patches will be added.
• Package does have options that can be changed, like building with clang or disabling Arch patches.
  • Defaults will match Arch packages, except when incompatible with this package's primary patches.
  • Options are subject to change. Read PKGBUILD for details.
• Avoid commenting and flagging at the same time for the same issue.
  • Flag for common issues with standard solutions, like new releases, key changes, etc.
  • Comment for issues requiring explanation or debugging.
    • Use a pastebin for blocks of text more than a few lines.

This package exists for the specific purpose of adding ported patches based on those originally created by Alex Williamson for:

• Bypassing the default IOMMU grouping by overriding PCIe ACS support, and

• Enabling VGA arbitration support on Intel iGPUs.

Arbitrary patches will not be added.

Refer to the wiki on PCI passthrough and this blog post on IOMMU groups for risks/caveats before using this package.

Regular AUR etiquette applies as well (knowledge of makepkg and searching the wiki/Arch forums is expected, and AUR helpers or Arch-based distros that are not Arch Linux are unsupported).

Tested with 5.0.9 by changing the git tag in the PKGBUILD, working fine, including testing with QEMU 4.0. Noteworthy though, that I had to disable building the doc package (removed htmldocs from make), since there was a build error with sphinx.

By the way, I think it might be a good idea to change the default kernel config in this package to enable voluntary preemption (CONFIG_PREEMPT_VOLUNTARY=y), since this fixes a bug with QEMU/OMVF boot times when configured with a lot of RAM/cores (see https://forums.unraid.net/topic/72172-unraid-os-version-653-available/). Personally using this configuration, have not noticed any side-effects or performance losses on the host so far.

Just wanted to say that In regards to jmandawg and markzz that I have been updating by just changing the versions for about a week now (starting with 5.0-1) with no unexpected behaviour or side effects. Although I should state I haven't done any deep testing outside of just running my normal vms etc.

jmandawg: You can certainly try just changing the PKGBUILD with no new patches, but I would bet on the fact that the current patches will not work. The issue is, it's on my list of things I have to get done, but that list has grown in the last few months to the point where I'm trying to catch up. No ETA at this moment.

EDIT: I take it all back, it looks like the patches did apply, so I'm running a build and will quickly test it tomorrow.

Thanks for the great work, 5.0.x coming anytime soon? Or is it safe to change the _srcver and rebuild?

The latest version (4.19.8-arch1-1-vfio) has build error due to patch:

drivers/pci/quirks.c:4617:28: error: ‘pci_acs_overrides’ undeclared here (not in a function); did you mean ‘pcie_acs_overrides’? { PCI_ANY_ID, PCI_ANY_ID, pci_acs_overrides }, ^~~~~~~~~~~~~~~~~ pcie_acs_overrides

I have pushed something that the patches apply, but I haven't tested the package itself.

whight: The vga arbiter patch does not apply in 4.19.x.

Applying: i915: Add module option to support VGA arbiter on HD devices (4.17)
error: patch failed: drivers/gpu/drm/i915/i915_params.c:133
error: drivers/gpu/drm/i915/i915_params.c: patch does not apply
error: patch failed: drivers/gpu/drm/i915/i915_params.h:58
error: drivers/gpu/drm/i915/i915_params.h: patch does not apply
error: patch failed: drivers/gpu/drm/i915/intel_display.c:15402
error: drivers/gpu/drm/i915/intel_display.c: patch does not apply

You shouldn't have any troubles using the patch in your sources for 4.19.8. Built and tested it myself with linux from the core repo. If you'd like, I can help keep this maintained.

archqemu: I try to keep up to date with regards to whatever version is in [core]. The problem is, on version releases (e.g. 4.14 -> 4.15), I often have to rewrite the patches (this isn't normally a problem on patch releases like 4.16.7 -> 4.16.8). Because I'm both a full time student and work full time as a software engineer, I am sometimes unable to set aside enough time to update the patches and I get to it when I can. The semester is almost over at the time of this comment, so I should be able to set aside time for sure then, if not sooner.