On Thu, 2025-08-21 at 14:17 -0500, Rob Landley wrote: > On 8/20/25 01:40, John Paul Adrian Glaubitz wrote: > > This can be worked around by just disabling io_uring support on sh4: > > > > --- libuv1-1.50.0.orig/src/unix/linux.c > > +++ libuv1-1.50.0/src/unix/linux.c > > @@ -470,6 +470,8 @@ static int uv__use_io_uring(uint32_t fla > > #elif defined(__powerpc64__) || defined(__ppc64__) > > /* See https://github.com/libuv/libuv/issues/4283. */ > > return 0; /* Random SIGSEGV in signal handler. */ > > +#elif defined(__sh__) > > + return 0; > > #else > > /* Ternary: unknown=0, yes=1, no=-1 */ > > static _Atomic int use_io_uring; > > Is this actually a different bug than powerpc is seeing? Because that > bug report _also_ says they're seeing crashes in cmake. Possibly this > should just add a third || defined() to the previous line...
If you look at the code in question, you can see that there are separate lines for each architectures which is okay since libuv is crashing for different reasons. Although I'm surprised this doesn't seem to be a problem on powerpc32. Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer `. `' Physicist `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
