On Fri, May 18, 2007 at 06:01:03PM +0200, Aurelien Jarno wrote: > Matthias Klose a écrit : > > Peter Green writes: > >> package:gcc-4.1 > >> version:4.1.2-7 > >> severity:serious > >> > >> from the relavent buildd logs: > > > > Note, that the severity is not RC for 68k; I do not intend to fix > > that. Same for arm, and arm porters don't seem to care that much, so > > maybe we should drop arm as a release architecture? > > > > > You can find a preliminary patch here: > > http://temp.aurel32.net/libffi-arm-closure.diff > > With it gcc-4.1 builds well, I am restarting a full build with the testsuite >
The patch is correct, at least the testsuite shows the exact same results, though not very good. Please find attached a patch against the current debian package. Bye, Aurelien -- .''`. Aurelien Jarno | GPG: 1024D/F1BCDB73 : :' : Debian developer | Electrical Engineer `. `' [EMAIL PROTECTED] | [EMAIL PROTECTED] `- people.debian.org/~aurel32 | www.aurel32.net
diff -u gcc-4.1-4.1.2/debian/patches/libjava-backport-updates2.dpatch gcc-4.1-4.1.2/debian/patches/libjava-backport-updates2.dpatch --- gcc-4.1-4.1.2/debian/patches/libjava-backport-updates2.dpatch +++ gcc-4.1-4.1.2/debian/patches/libjava-backport-updates2.dpatch @@ -32564,0 +32565,27 @@ +Index: libffi/src/arm/ffi.c +=================================================================== +--- libffi/src/arm/ffi.c 2007-05-18 17:49:21.000000000 +0200 ++++ libffi/src/arm/ffi.c 2007-05-18 17:51:59.000000000 +0200 +@@ -269,16 +269,17 @@ + /* the cif must already be prep'ed */ + + ffi_status +-ffi_prep_closure (ffi_closure* closure, +- ffi_cif* cif, +- void (*fun)(ffi_cif*,void*,void**,void*), +- void *user_data) ++ffi_prep_closure_loc (ffi_closure* closure, ++ ffi_cif* cif, ++ void (*fun)(ffi_cif*,void*,void**,void*), ++ void *user_data, ++ void *codeloc) + { + FFI_ASSERT (cif->abi == FFI_SYSV); + + FFI_INIT_TRAMPOLINE (&closure->tramp[0], \ + &ffi_closure_SYSV, \ +- (void*)closure); ++ codeloc); + + closure->cif = cif; + closure->user_data = user_data;