New packages: {edk2,ovmf-*}-202102 #29074

Logarithmus · 2021-02-26T17:59:40Z

Continues #17225
Closes #11243, #27229

General

This is a new package and it conforms to the quality requirements

Have the results of the proposed changes been tested?

I use the packages affected by the proposed changes on a regular basis and confirm this PR works for me
I generally don't use the affected packages but briefly tested this PR

Does it build and run successfully?

I built this PR locally for my native architecture, (x86_64-musl)
I built this PR locally for these architectures (if supported. mark crossbuilds):
- x86_64
- x86_64-musl
- i686

ericonr · 2021-02-26T18:12:17Z

Please clean up your commit, you have .rej and .orig files.

Xlaits · 2021-03-24T21:43:52Z

Getting errors attempting to build ovmf-*

=> ca-certificates-20210119_1: running pre-pkg hook: 90-set-timestamps ...
date: invalid date ‘@1616622037\n1616622037’
=> ERROR: ca-certificates-20210119_1: pre-pkg_90-set-timestamps: 'date --date "@$SOURCE_DATE_EPOCH"' exited with 1
=> ERROR:   in hook() at common/hooks/pre-pkg/90-set-timestamps.sh:7
=> ERROR:   in run_func() at common/xbps-src/shutils/common.sh:21
=> ERROR:   in run_pkg_hooks() at common/xbps-src/shutils/common.sh:245
=> ERROR:   in main() at common/xbps-src/libexec/xbps-src-prepkg.sh:47
=> ca-certificates-20210119_1: setting mtimes to 
touch: invalid date format ‘@1616622037\n1616622037’
=> ERROR: ca-certificates-20210119_1: pre-pkg_90-set-timestamps: 'xargs -0 touch -h --date "@$SOURCE_DATE_EPOCH"' exited with 123
=> ERROR:   in hook() at common/hooks/pre-pkg/90-set-timestamps.sh:8
=> ERROR:   in run_func() at common/xbps-src/shutils/common.sh:21
=> ERROR:   in run_pkg_hooks() at common/xbps-src/shutils/common.sh:245
=> ERROR:   in main() at common/xbps-src/libexec/xbps-src-prepkg.sh:47

Logarithmus · 2021-03-26T22:44:31Z

Getting errors attempting to build ovmf-*

=> ca-certificates-20210119_1: running pre-pkg hook: 90-set-timestamps ...
date: invalid date ‘@1616622037\n1616622037’
=> ERROR: ca-certificates-20210119_1: pre-pkg_90-set-timestamps: 'date --date "@$SOURCE_DATE_EPOCH"' exited with 1
=> ERROR:   in hook() at common/hooks/pre-pkg/90-set-timestamps.sh:7
=> ERROR:   in run_func() at common/xbps-src/shutils/common.sh:21
=> ERROR:   in run_pkg_hooks() at common/xbps-src/shutils/common.sh:245
=> ERROR:   in main() at common/xbps-src/libexec/xbps-src-prepkg.sh:47
=> ca-certificates-20210119_1: setting mtimes to 
touch: invalid date format ‘@1616622037\n1616622037’
=> ERROR: ca-certificates-20210119_1: pre-pkg_90-set-timestamps: 'xargs -0 touch -h --date "@$SOURCE_DATE_EPOCH"' exited with 123
=> ERROR:   in hook() at common/hooks/pre-pkg/90-set-timestamps.sh:8
=> ERROR:   in run_func() at common/xbps-src/shutils/common.sh:21
=> ERROR:   in run_pkg_hooks() at common/xbps-src/shutils/common.sh:245
=> ERROR:   in main() at common/xbps-src/libexec/xbps-src-prepkg.sh:47

Please, check now

Logarithmus · 2021-03-27T13:09:28Z

CI passed, tested on x86_64-musl, works fine. Ready for review & merge I think.

Xlaits · 2021-03-29T04:38:11Z

Still having the same issue.

Logarithmus · 2021-03-29T06:25:11Z

Still having the same issue.

Maybe the issue is your date? Check if it's correct. The package have built successfully both on my laptop and CI/CD. Can someone else test it?

Xlaits · 2021-03-29T12:50:38Z

Date and time are correct.

…

On Mon, Mar 29, 2021, 2:25 AM Artur Sinila ***@***.***> wrote: Still having the same issue. Maybe the issue is your date? Check if it's correct. The package have built successfully both on my laptop and CI/CD. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#29074 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHX2YDM64PVNQOCBQ5XEPA3TGAMNPANCNFSM4YI3T56Q> .

noarchwastaken · 2021-04-10T01:44:40Z

The packages built with no problems for me on x86_64-glibc.

noarchwastaken · 2021-04-10T01:57:16Z

I noticed that the VM starts up to a black screen if I have OpenGL enabled for Spice... the VM only starts to display when the OS has booted to a certain point. I'm not sure if it's an upstream issue though.

eoli3n · 2021-05-10T07:27:19Z

up

eoli3n · 2021-05-18T19:26:42Z

The packages built with no problems for me on x86_64-glibc.

Same here, i will test later.

NeonOverflow · 2021-05-31T21:12:50Z

Please merge this!

eoli3n · 2021-05-31T22:21:07Z

ping @ericonr

NeonOverflow · 2021-06-03T00:37:15Z

@ericonr

ericonr · 2021-06-03T00:50:05Z

I only made a lint comment, and haven't claimed this PR. It's a big package that I'm not in a position to review rn.

ahesford · 2021-06-03T01:29:16Z

I really don't see this being merged without massive upstream work. The template is a jumble of manual steps and questionable vendoring (brotli? openssl???). I suspect the only truly valuable part of this package is the half-dozen firmware blobs and associated JSON descriptors, all of which I just manually unpack from an Arch Linux package and be on my way.

ericonr · 2021-06-03T02:16:27Z

I don't want to involve myself, but from https://github.com/tianocore/edk2/blob/master/CryptoPkg/Library/OpensslLib/OpenSSL-HOWTO.txt it would seem at least openssl is used to provide crypto stuff and it would have to be compiled specially for that, which is why they pull it in.

NeonOverflow · 2021-06-04T00:28:30Z

@ahesford The firmware files are primarily used by GPU passthrough users.

ahesford · 2021-06-04T00:37:32Z

I understand the utility of the firmware files; I use them in all of my VMs. My point is that the template requires a lot of manual effort because the EDK build system isn't sane, and these steps are likely to break from one release to the next. Void is a small team and fragile templates that require a lot of manual work become disproportionate time sinks.

In the meantime, the firmware blobs can be manually extracted from packages of other distributions and used in Qemu/KVM installations in Void with minimal effort.

NeonOverflow · 2021-06-04T01:03:53Z

Understandable. I'll probably just do as you said in the meantime.

eoli3n · 2021-06-04T06:17:40Z

In the meantime, the firmware blobs can be manually extracted from packages of other distributions and used in Qemu/KVM installations in Void with minimal effort.

So why not pulling this in the package ?

HadetTheUndying · 2021-06-20T01:03:57Z

In the meantime, the firmware blobs can be manually extracted from packages of other distributions and used in Qemu/KVM installations in Void with minimal effort.

So why not pulling this in the package ?

I would be under the assumption that this is against the package quality requirements since this package would not fall under the nonfree packages, it's just a pain to package.

Matthew-Tate-Scarbrough · 2021-07-26T17:56:45Z

In the meantime, the firmware blobs can be manually extracted from packages of other distributions and used in Qemu/KVM installations in Void with minimal effort.

Would it be a advantageous/within scope of void-packages to just write a script that just steals them from Arch or another popular distro's repos?

RononDex · 2021-08-26T19:52:43Z

Is there any update on the state of the PR? Trying to setup UEFI with qemu

apprehensions · 2022-01-10T15:14:53Z

uh? when?

HadetTheUndying · 2022-01-10T20:19:28Z

uh? when?

The qemu package includes the following:

qemu-6.1.0_3	/usr/share/qemu/edk2-aarch64-code.fd
qemu-6.1.0_3	/usr/share/qemu/edk2-arm-code.fd
qemu-6.1.0_3	/usr/share/qemu/edk2-arm-vars.fd
qemu-6.1.0_3	/usr/share/qemu/edk2-i386-code.fd
qemu-6.1.0_3	/usr/share/qemu/edk2-i386-secure-code.fd
qemu-6.1.0_3	/usr/share/qemu/edk2-i386-vars.fd
qemu-6.1.0_3	/usr/share/qemu/edk2-licenses.txt
qemu-6.1.0_3	/usr/share/qemu/edk2-x86_64-code.fd
qemu-6.1.0_3	/usr/share/qemu/edk2-x86_64-secure-code.fd
qemu-6.1.0_3	/usr/share/qemu/firmware/50-edk2-i386-secure.json
qemu-6.1.0_3	/usr/share/qemu/firmware/50-edk2-x86_64-secure.json
qemu-6.1.0_3	/usr/share/qemu/firmware/60-edk2-aarch64.json
qemu-6.1.0_3	/usr/share/qemu/firmware/60-edk2-arm.json
qemu-6.1.0_3	/usr/share/qemu/firmware/60-edk2-i386.json
qemu-6.1.0_3	/usr/share/qemu/firmware/60-edk2-x86_64.json
```

I have tested the x86_64 firmware with GPU passthrough to a Windows VM and can confirm it works.

apprehensions · 2022-01-11T12:50:21Z

The qemu package includes the following:
...
qemu-6.1.0_3	/usr/share/qemu/edk2-x86_64-code.fd
qemu-6.1.0_3	/usr/share/qemu/edk2-x86_64-secure-code.fd
qemu-6.1.0_3	/usr/share/qemu/firmware/60-edk2-x86_64.json
...
I have tested the x86_64 firmware with GPU passthrough to a Windows VM and can confirm it works.

the fact that it was in qemu this whole time and i never noticed... thanks for telling me this!

ahesford · 2022-01-11T13:14:59Z

Because it seems that qemu already ships viable EFI image and EDK2 is such a nightmare to build, let's close this and revisit if there is a compelling need and we can construct a more maintainable template.

eoli3n · 2022-02-03T09:57:39Z

I can't find a way to run that firmware with vagrant-libvirt : vagrant-libvirt/vagrant-libvirt#1443

ghost · 2023-06-14T09:44:51Z

Has anyone gotten the EFI images shipped with qemu to work for enabling secure boot and creating a Windows 11 VM?

Logarithmus force-pushed the ovmf branch from 77d9e88 to 2c0aac7 Compare February 26, 2021 18:01

Logarithmus force-pushed the ovmf branch from 2c0aac7 to 6e61a06 Compare February 26, 2021 18:13

ahesford added the new-package This PR adds a new package label Feb 27, 2021

Logarithmus force-pushed the ovmf branch from 6e61a06 to b0db0da Compare March 26, 2021 21:44

Logarithmus changed the title ~~New packages: {edk2,ovmf-*}-202011~~ New packages: {edk2,ovmf-*}-202102 Mar 26, 2021

Logarithmus force-pushed the ovmf branch from b0db0da to c006cf3 Compare March 26, 2021 22:41

Logarithmus marked this pull request as ready for review March 26, 2021 22:43

New package: {edk2,ovmf-*}-202102

93287ab

Logarithmus force-pushed the ovmf branch from c006cf3 to 93287ab Compare March 26, 2021 22:46

paper42 mentioned this pull request Aug 18, 2021

qemu/libvirt: virt-aa-helper refuses to create AppArmor VM profiles when using QEMU provided OVMF images from /usr/share/qemu #32562

Closed

ahesford closed this Jan 11, 2022

This was referenced May 9, 2023

PCI passthrough guide void-linux/void-docs#208

Open

Package Request: Tianocore/OVMF #11243

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

New packages: {edk2,ovmf-*}-202102 #29074

New packages: {edk2,ovmf-*}-202102 #29074

Logarithmus commented Feb 26, 2021 •

edited

Loading

ericonr commented Feb 26, 2021

Xlaits commented Mar 24, 2021

Logarithmus commented Mar 26, 2021

Logarithmus commented Mar 27, 2021

Xlaits commented Mar 29, 2021

Logarithmus commented Mar 29, 2021 •

edited

Loading

Xlaits commented Mar 29, 2021 via email

noarchwastaken commented Apr 10, 2021

noarchwastaken commented Apr 10, 2021

eoli3n commented May 10, 2021

eoli3n commented May 18, 2021

NeonOverflow commented May 31, 2021

eoli3n commented May 31, 2021

NeonOverflow commented Jun 3, 2021

ericonr commented Jun 3, 2021

ahesford commented Jun 3, 2021

ericonr commented Jun 3, 2021

NeonOverflow commented Jun 4, 2021

ahesford commented Jun 4, 2021

NeonOverflow commented Jun 4, 2021

eoli3n commented Jun 4, 2021 •

edited

Loading

HadetTheUndying commented Jun 20, 2021

Matthew-Tate-Scarbrough commented Jul 26, 2021

RononDex commented Aug 26, 2021

apprehensions commented Jan 10, 2022

HadetTheUndying commented Jan 10, 2022

apprehensions commented Jan 11, 2022

ahesford commented Jan 11, 2022

eoli3n commented Feb 3, 2022

ghost commented Jun 14, 2023

New packages: {edk2,ovmf-*}-202102 #29074

New packages: {edk2,ovmf-*}-202102 #29074

Conversation

Logarithmus commented Feb 26, 2021 • edited Loading

General

Have the results of the proposed changes been tested?

Does it build and run successfully?

ericonr commented Feb 26, 2021

Xlaits commented Mar 24, 2021

Logarithmus commented Mar 26, 2021

Logarithmus commented Mar 27, 2021

Xlaits commented Mar 29, 2021

Logarithmus commented Mar 29, 2021 • edited Loading

Xlaits commented Mar 29, 2021 via email

noarchwastaken commented Apr 10, 2021

noarchwastaken commented Apr 10, 2021

eoli3n commented May 10, 2021

eoli3n commented May 18, 2021

NeonOverflow commented May 31, 2021

eoli3n commented May 31, 2021

NeonOverflow commented Jun 3, 2021

ericonr commented Jun 3, 2021

ahesford commented Jun 3, 2021

ericonr commented Jun 3, 2021

NeonOverflow commented Jun 4, 2021

ahesford commented Jun 4, 2021

NeonOverflow commented Jun 4, 2021

eoli3n commented Jun 4, 2021 • edited Loading

HadetTheUndying commented Jun 20, 2021

Matthew-Tate-Scarbrough commented Jul 26, 2021

RononDex commented Aug 26, 2021

apprehensions commented Jan 10, 2022

HadetTheUndying commented Jan 10, 2022

apprehensions commented Jan 11, 2022

ahesford commented Jan 11, 2022

eoli3n commented Feb 3, 2022

ghost commented Jun 14, 2023

Logarithmus commented Feb 26, 2021 •

edited

Loading

Logarithmus commented Mar 29, 2021 •

edited

Loading

eoli3n commented Jun 4, 2021 •

edited

Loading