Hi everyone, Thank you all so much for your comments and replies.
@Micha: Understood. I agree that the 800-file diff for the 25.3.3 CVE backport is too intrusive for a clean review. I am withdrawing that proposal. @Fabian: That sounds like the correct path forward. Since you are planning to sync the toolchain (llvm) from Forky, do you have a rough timeline for that or the spirv-* packages? I'd be happy to help test the Mesa 26 build or assist with the package preparation once the dependency stack lands in trixie-backports. I have Intel Arc hardware ready to validate the whole stack. Best regards, stornic56 On Sunday, June 14th, 2026 at 4:41 AM, Fabian Grünbichler <[email protected]> wrote: > On Sun, Jun 14, 2026, at 9:22 AM, Daniel Baumann wrote: > > On 6/13/26 15:00, stornic56 wrote: > >> 4. Mesa 26.0.8 is extra work > > > > not really[0]. > > > > a mesa backport[1] is itself a simple rebuild (no changes needed), > > if the following packages are backported too: > > > > * llvm-toolchain-21 > > => simple rebuild, no changes necessary > > that one is already in backports cause of rustc. I do plan on doing > another sync with Forky. > > > [..] > > > * spirv-llvm-translator-21 > > * spirv-tools > > * spirv-headers > > => simple rebuilds, no changes necessary > > (at least one of them could probably avoided, but not worth the > > effort) > > if that is planned, I'd wait for it and then do the llvm update, since > it would allow reducing the delta ;) > > Fabian > >
