Re: [edk2-devel] Adding Bhyve support into upstream EDK2

["Laszlo Ersek" <lersek@redhat.com>]

On 03/09/20 23:54, Laszlo Ersek wrote:
> On 03/09/20 07:08, Sean via Groups.Io wrote:

>> From someone who has developed and maintained platforms with edk2 for
>> the last decade +, I can tell you that having OVMF in edk2 hasn't kept
>> the tree free from regressions.
> 
> Of course. Changes to core packages (even spec changes) are always
> motivated by platform goals. And regressions are unavoidable, as much as
> we try to prevent them (even *spec* regressions). I hope you're not
> claiming that OVMF has been a major source of regressions in the core
> packages.
> 
> And, you have the advantage of *plausibly* claiming that OVMF-oriented
> changes have caused regressions in the core -- plausibly because those
> contributions, and the dependent OVMF changes, are upstream. Whereas I
> can't classify any other regressions as associated by Microsoft platform
> needs, simply because those platform needs have hardly ever been public.
> (One of my never-ending crusades has been "better commit messages",
> i.e., with use case spelled out.)

I think I misread your point.

IIUC, you weren't blaming OVMF for regressions, instead, you were
questioning the usefulness of OVMF in catching regressions in the core.

Putting aside your rhetorical expression "keep the tree free from
regressions" -- because how could any single platform keep the whole
tree free of regressions? --, OVMF *has* caught regressions. In fact,
running OVMF on QEMU/KVM has been one of the best ways for rooting out
multiprocessing / synchronization bugs in UefiCpuPkg, due to the VCPU
jitter that KVM tends to introduce. This is why I asked (years ago) to
be CC'd on all UefiCpuPkg patches, so I could regression-test them
before they are pushed. (Hence my "R:" role in UefiCpuPkg.)

$ git log \
  --oneline \
  --reverse \
  --grep='Regression-tested-by: Laszlo Ersek <lersek@redhat.com>' \
  master -- \
  ':!OvmfPkg' \
  ':!ArmVirtPkg' \
  ':!ArmPlatformPkg/ArmVirtualizationPkg' \
| wc -l

(83 patches)


This is also (part of) why I wrote:

https://github.com/tianocore/tianocore.github.io/wiki/Testing-SMM-with-QEMU,-KVM-and-libvirt


Two semi-recent commits that fixed regressions (exposed by OVMF) have been:

- 4d05a4b709ce ("MdeModulePkg/BdsDxe: Fix calling
PlatformBootManagerWaitCallback on 0", 2019-10-15)

- a52355624440 ("UefiCpuPkg/PiSmmCpuDxeSmm: fix 2M->4K page splitting
regression for PDEs", 2020-01-17)

Thanks
Laszlo