rofl0r

for the record, in the case i describe i used mcm to compile a native gcc 6.5.0, whose libgcc.a then exposed the problem while using it inside a chroot with...

i think it would be preferable to merge the patches done by szabolcs nagy, as he's a long-time contributor to both musl and gcc and certainly did a careful forward...

superceded by https://github.com/richfelker/musl-cross-make/pull/124 , this can be closed.

> trying to compile the kernel on the M1 M1 is aarch64, not arm, so you would have to set your target triple accordingly. note that the kernel uses in...

> mcm already knows how to remap them or it wouldn't work at all to get the kernel headers thanks for clarification. i wasn't aware of this and have always...

> Should I ask these questions upstream? maybe not a bad idea. SDL could check if the previous queued event is mousemotion when it gets a new event, and then...

and before that commit we had no audio in CLI. can you debug what's happening, and compare the audio setup code in win32 interface vs posix cli ? my guess...

from commit msg: > - fixed ignoring volume set by the SPU. my naive guess is that the interface sets volume to 0, which was ignored in the past.

> - Fixing a bug I introduced in my interpolation re-write (ADPCM channels weren't restored correctly from savestates) > - Fixing another bug I introduced where using capture-based echo effects...

in case those errors are from gnu tar (looks like it), you can patch the tar binary to default to --no-same-owner using the instructions here https://github.com/sabotage-linux/sabotage/wiki/Running-a-minimal-ubuntu-rootfs-as-regular-user#addendum-making-gnu-tar-less-annoying