public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: "Gerd Hoffmann" <kraxel@redhat.com>
To: "Lu, Ken" <ken.lu@intel.com>
Cc: Ard Biesheuvel <ardb@kernel.org>,
	"Xu, Min M" <min.m.xu@intel.com>,
	Daniel Kiper <daniel.kiper@oracle.com>,
	"devel@edk2.groups.io" <devel@edk2.groups.io>,
	Ard Biesheuvel <ardb+tianocore@kernel.org>,
	"Aktas, Erdem" <erdemaktas@google.com>,
	James Bottomley <jejb@linux.ibm.com>,
	"Yao, Jiewen" <jiewen.yao@intel.com>
Subject: Re: [edk2-devel] measurement to command-line/initrd for loading kernel via -kernel option
Date: Wed, 21 Sep 2022 09:14:09 +0200	[thread overview]
Message-ID: <20220921071409.5oziya6kcfvkkkp7@sirius.home.kraxel.org> (raw)
In-Reply-To: <DM6PR11MB36749B88FC1F1B8413413861984C9@DM6PR11MB3674.namprd11.prod.outlook.com>

On Tue, Sep 20, 2022 at 02:30:01PM +0000, Lu, Ken wrote:
> > > So there are two types loaders:
> > > 1. QemuKernelLoaderFsDxe  -   this way just put kernel/initrd blob into a FS
> > for any future's usage, may be continue boot or not.
> > > 2. QemuLoadKernelImage,    -    this is consumed by TryRunningQemuKernel()
> > - standard Qemu direct boot path
> > 
> > Nope.  QemuLoadKernelImage loads the linux kernel from the virtual filesystem
> > created by QemuKernelLoaderFsDxe.  And for the initrd it'll just pass
> > 'inittd=initrd' and the stub loads it.
> > 
> > We have two variants:
> >   GenericQemuLoadImageLib - supports efi stub only
> >   X86QemuLoadImageLib     - has fallback code paths for the legacy
> >                             pre-efi-stub boot protocol (guess that
> >                             is the one grub has deprecated for 2.06).
> > 
> > So, yes, with the legacy protocol there is no stub which can measure things, but
> > for the snake of confidential computing we can completely ignore that.  Kernels
> > which are *that* old certainly will not have support for SEV / TDX ...
> > 
> 
> Thanks Hoffman. Hmm.. GenericQemuLoadImageLib sound like is used by ArmVirtQemu.dsc, OvmfXen.dsc, AmdSevX64.dsc,.....
> But X86QemuLoadImageLib is used by OvmfPkgX64.dsc and Intel TDX~~
> 
> Headache.... do you want use GenericQemuLoadImageLib to replace X86 one for OvmfPkgX64.dsc also?

I think most x86 should stick to X86QemuLoadImageLib for backward
compatibility reasons.

The AmdSevX64 and IntelTDX variants can most likely be switched over to
GenericQemuLoadImageLib without breaking stuff.  Linux kernels new
enough to support sev / tdx should not need the legacy support.

> But either in GenericQemuLoadImageLib, it can do measurement for command line and initrd, correct?

Yes, it could.  But why given that the linux kernel efi stub measures anyway?

take care,
  Gerd


  reply	other threads:[~2022-09-21  7:14 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-17  2:52 measurement to command-line/initrd for loading kernel via -kernel option Min Xu
2022-09-18 12:52 ` Ard Biesheuvel
2022-09-19  2:13   ` [edk2-devel] " Min Xu
2022-09-19  6:58     ` Ard Biesheuvel
2022-09-20  0:20       ` Min Xu
2022-09-20 12:29         ` Ard Biesheuvel
2022-09-20 12:55       ` Lu, Ken
2022-09-20 13:03         ` Ard Biesheuvel
2022-09-20 13:24           ` Lu, Ken
2022-09-20 13:43             ` James Bottomley
2022-09-20 14:34               ` Ard Biesheuvel
2022-09-20 14:51             ` Ard Biesheuvel
2022-09-20 15:14               ` Lu, Ken
2022-09-20 13:20         ` Gerd Hoffmann
2022-09-20 13:38           ` Lu, Ken
2022-09-20 14:18             ` Gerd Hoffmann
2022-09-20 14:30               ` Lu, Ken
2022-09-21  7:14                 ` Gerd Hoffmann [this message]
2022-09-21 11:24                   ` Lu, Ken
2022-09-21 12:27                     ` Gerd Hoffmann
2022-09-21 15:41                       ` Ard Biesheuvel
2022-09-23  9:34                         ` Ilias Apalodimas

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-list from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220921071409.5oziya6kcfvkkkp7@sirius.home.kraxel.org \
    --to=devel@edk2.groups.io \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox