qemu: 8.2.4 -> 9.0.0 #298097
Merged
qemu: 8.2.4 -> 9.0.0 #298097
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ofborg
bot
added
11.by: package-maintainer
|
Might it be time to set |
|
I don't use or understand macOS, so I don't really have an opinion, except that something needs to be done in the next two weeks. |
|
I'm told the fixes needed to make this work on darwin are currently in staging. Haven't been able to do a staging bootstrap yet to test this. |
alyssais
force-pushed
the
qemu-9.0.0-rc0
branch
from
3706afa to
124808e
Compare
|
staging-next has merged now, so am I understanding correctly that if I rebase this it should build on Darwin without the patches? |
alyssais
force-pushed
the
qemu-9.0.0-rc0
branch
from
124808e to
2793eaf
Compare
|
Still not building on x86_64-darwin. Is there something else that needs to happen? |
|
I no longer have access to an x86 mac (moved on to arm64) and so can't actually run the relevant builds. However, unless #101229 and similar issues happen, it would surprise me to not need these anymore. The patches are basically undoing 10.13+ stuff that is being added to QEMU to allow it to keep building on x86. At this point, 10.13 came out in 2017, and I think we're reaching the point where it's unreasonable to backport software to require an OS's API that has to be 7+ years old. If only because I don't know anyone else who is invested in qemu on darwin x86 on nix, I'm not sure how to test that functionality short of me trial-and-erroring blindly by trying to fix diffs and with ofborg as my guinea pig. |
No, I'm told it missed this staging-next cycle. |
|
Have tried with the new |
|
9.0 (non RC) is out now, if you can and want to pursue this. |
|
I still need somebody to tell me what to do about x86_64-darwin, unless we just mark it broken? |
|
Unless we can find someone with an x86-darwin machine who wants to rebase and test, I think broken is the best bet, but that's just me. |
alyssais
force-pushed
the
qemu-9.0.0-rc0
branch
from
2793eaf to
f2f62fe
Compare
|
@ofborg test cage cagebreak cloud-init-hostname cloud-init mtp os-prober phosh sway systemd-initrd-luks-fido2 tinywl virtualbox |
matdibu
reviewed
Apr 24, 2024
|
Ugh, apparently QEMU is release-critical for x86_64-darwin, so we can't just mark it broken without blocking the channel. @roberth @Enzime seems like you were involved in that — can you help here? |
|
If it's okay to get it removed out of being release critical, that makes the most sense to me. If it's considered release critical, then we'll need to find a volunteer from darwin maintainers who's still on x86 and doesn't mind rebasing the patches (not pinging them yet...) |
|
Following the advice from the docs on using a newer SDK does not seem to work, or at least has other problems: applying this patch onto this PR (after rebasing on nixos-unstable)diff --git a/pkgs/applications/virtualization/qemu/default.nix b/pkgs/applications/virtualization/qemu/default.nix
index a2b01272b1f3..7afca0cbd371 100644
--- a/pkgs/applications/virtualization/qemu/default.nix
+++ b/pkgs/applications/virtualization/qemu/default.nix
@@ -265,7 +265,6 @@ stdenv.mkDerivation (finalAttrs: {
license = licenses.gpl2Plus;
maintainers = with maintainers; [ eelco qyliss ];
platforms = platforms.unix;
- broken = stdenv.hostPlatform.system == "x86_64-darwin";
}
# toolsOnly: Does not have qemu-kvm and there's no main support tool
// lib.optionalAttrs (!toolsOnly) {
diff --git a/pkgs/top-level/all-packages.nix b/pkgs/top-level/all-packages.nix
index eaa23105da40..398179be3fa2 100644
--- a/pkgs/top-level/all-packages.nix
+++ b/pkgs/top-level/all-packages.nix
@@ -34070,8 +34070,8 @@ with pkgs;
eiskaltdcpp = libsForQt5.callPackage ../applications/networking/p2p/eiskaltdcpp { };
- qemu = callPackage ../applications/virtualization/qemu {
- inherit (darwin.apple_sdk.frameworks) CoreServices Cocoa Hypervisor vmnet;
+ qemu = darwin.apple_sdk_11_0.callPackage ../applications/virtualization/qemu {
+ inherit (darwin.apple_sdk_11_0.frameworks) CoreServices Cocoa Hypervisor vmnet;
inherit (darwin.stubs) rez setfile;
inherit (darwin) sigtool;
};
results in these errors when building on x86_64-darwin |
alyssais
added
the
2.status: merge conflict
ofborg
bot
removed
the
2.status: merge conflict
|
@NixOS/darwin-maintainers does anybody care about QEMU on x86_64-darwin? If nobody steps up there's not much we can do apart from remove it from the tested set… |
|
Their is an effort to port nixos test on macos so I think we should repair it. I will try to fix it (not today but tomorrow) but if anyone have a fix earlier it is ok. |
|
I'm also looking into this but won't have time to finish until tomorrow. Feel free to beat me to it. |
|
I'm also working on this. |
wegank
force-pushed
the
qemu-9.0.0-rc0
branch
2 times, most recently
from
beadcc2 to
99dc467
Compare
alyssais
commented
May 8, 2024
|
Thanks a lot @wegank! |
libvirt doesn't build (with sandbox enabled) on x86_64-darwin. It doesn't build on masterOther error are not related. diffoscope build on master but not in this PR. |
So the build failure is unrelated to this PR? |
I have found Meson-based tests to be annoyingly unreliable on Darwin. Try using |
|
Result of 15 packages marked as broken and skipped:
33 packages failed to build:
56 packages built:
|
|
@ofborg build solo5 |
|
@ofborg build linuxKernel.packages.linux_6_8.virtualbox |
alyssais
commented
May 14, 2024
pkgs/by-name/so/solo5/package.nix
Outdated
| @@ -59,6 +59,8 @@ in stdenv.mkDerivation { | |||
| checkPhase = '' | |||
| runHook preCheck | |||
| patchShebangs tests | |||
| substituteInPlace scripts/virtio-run/solo5-virtio-run.sh \ | |||
| --replace-fail "-no-acpi" "-machine acpi=off" | |||
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description of changes
The Darwin patches have stopped applying again. (cc @willcohen)
Things done
nix.conf? (See Nix manual)sandbox = relaxedsandbox = truenix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/)Add a 👍 reaction to pull requests you find important.