Le mar. 28 janv. 2025 à 09:29, Arnd Bergmann <a...@arndb.de> a écrit :
> On Mon, Jan 27, 2025, at 19:31, Jérémy Lal wrote: > > Le lun. 27 janv. 2025 à 17:41, Arnd Bergmann <a...@arndb.de> a écrit : > >> On Mon, Jan 27, 2025, at 15:19, Jérémy Lal wrote: > >> > >> if ((sizeof(long)< sizeof(off_t)) > >> p = dlsym(RTLD_DEFAULT, is_pread ? "preadv64" : > "pwritev64"); > >> else > >> p = dlsym(RTLD_DEFAULT, is_pread ? "preadv" : "pwritev"); > > > > Thank you, it works with that. > > I linked your reply in the upstream issue. > > It appears that the proposed workaround at > > > https://github.com/libuv/libuv/pull/4683/commits/45bf09e2567b6480962d932946608454fab31b87 > > now just always tries to find the preadv64()/pwritev64() symbols, > which is still nonportable because this doesn't work on targets > that use a 32-bit off_t. > > If i386-debian was working before this change, it is now broken. > > The bit I don't understand is why libuv was ever getting built > without largefile support. It probably makes sense to change that > for all architectures regardless of time64 support, but this > is likely an ABI break and requires rebuilding all libuv users > in turn. It builds fine in a i386 chroot on barriere.d.o with https://github.com/libuv/libuv/pull/4683.patch
