> Date: Wed, 13 May 2026 10:04:04 +0100 > From: Christopher Bazley <[email protected]>
> Hello Hans-Peter, > > I'm pleased to report that my patch to add > -Wzero-init-padding-bits=(standard|unions|all) was merged to GCC trunk > yesterday as > 2502a459db595905f6406b7994a27729c1e99a1d. > > I'll leave it up to the maintainers to decide whether it should now be > backported to those releases that contain 0547dbb725b (namely, > releases/gcc-15 and releases/gcc-16). > > I'm not sure of the policy on backporting patches that are not bugfixes. Good news but also followed by somewhat bad news. Even though the patch is now NAK:ed for branches, I've checked that the committed patch appears to cherry-pick cleanly to the gcc-16 branch, but I haven't built and tested. For the gcc-15 branch, it needs some trivial editing to invoke.texi. I'm sure no-one would object if you post a rebased and tested version that works for the gcc-15 branch, helping distros now pondering whether to include it. While the change to not initiate union padding bits might be mostly harmless, I know that the uncertainty has caused some fear, and I still hope for it to be back-ported: it warns about changed behaviour in gcc-15 (and IMHO it honestly doesn't look intrusive). > > Thanks, > -- > Christopher Bazley > Staff Software Engineer, GNU Tools Team. > Arm Ltd, 110 Fulbourn Road, Cambridge, CB1 9NJ, UK. > http://www.arm.com/ > Thank you for your work and thanks to maintainers that reviewed and considered it. brgds, H-P
