| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| |
|
|
|
|
|
| |
The git logic is handled in the live ebuild directly, and the only use
of this eclass is in non-live ebuilds. In fact, the two newest non-live
ebuilds do not make use of this eclass at all, and as far as I can see
nowhere in the history of this file does it inherit git-2 at all.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The error is found by wm_ attempting to build a cross-compiler
CHOST=powerpc-unknown-linux-gnu CTARGET=mips64-unknown-linux-gnu
Boostrap failed on glibc-headers (./configure phase) as:
configure:2623: powerpc-unknown-linux-gnu-gcc -mabi=n32 -c -O1 -pipe -U_FORTIFY_SOURCE conftest.c >&5
powerpc-unknown-linux-gnu-gcc: error: unrecognized command line option '-mabi=n32'; did you mean '-mabi=d32'?
Note how target's ABI is mexed into host's compiler flags.
The error above happens on every host=non-multilib target=multilib combination.
I've reproduced it on:
CHOST=i686-pc-linux-gnu CTARGET=mips64-unknown-linux-gnu
The fix is not to inject target's CFLAGS ABI into CC for
USE=headers-only bootstrap step as we don't have cross-compiler yet.
Tested by successfully bootstrapping cross-compiler on
CHOST=i686-pc-linux-gnu CTARGET=mips64-unknown-linux-gnu
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
|
| |
|
|
| |
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
|
| |
|
|
| |
glibc aborts if rpath is set.
|
| |
|
|
| |
alt_build_headers already has EPREFIX.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
USE=headers-only
CROSSCOMPILE_OPTS is a USE_EXPAND of a single item: headers-only.
Convert it to a global USE flag instead.
Mechanical ebuild rename done as:
$ sed -e 's@crosscompile_opts_headers-only@headers-only@g' \
-i $(git grep -l headers-only)
'headers-only' flag is used by crossdev to bootstrap stage1 compiler
before libc is available.
crossdev switched to USE=headers-only in =sys-devel/crossdev-20171230.
Bug: https://bugs.gentoo.org/642712
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We explicitly disable IFUNC support on the following targets:
alpha/hppa/ia64/mips/m68k/nios2/riscv/sh
to workaround weak IFUNC detection on binutils/glibc side.
Otherwise at least on ia64 glibc generates IFUNC entries against
compat librt.so.1 symbols (to redirect them back to libc.so.6)
but linker does not produce correct relocations. As a result all
IFUNC-backed functions don't work.
Reported-by: Émeric Maschino
Bug: https://sourceware.org/PR22634
Closes: https://bugs.gentoo.org/641216
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Otherwise dlopen() crashes (found on 'top' from 'procps' package).
Original backtrace from qemu-mipsn32 (fails the same way on native host):
```
Program received signal SIGSEGV, Segmentation fault.
0x408cb908 in _dlerror_run (operate=operate@entry=0x408cadf0 <dlopen_doit>, args=args@entry=0x407feb28) at dlerror.c:163
163 result->errcode = _dl_catch_error (&result->objname, &result->errstring,
(gdb) bt
#0 0x408cb908 in _dlerror_run (operate=operate@entry=0x408cadf0 <dlopen_doit>, args=args@entry=0x407feb28) at dlerror.c:163
#1 0x408caf4c in __dlopen (file=file@entry=0x10012d58 "libnuma.so", mode=mode@entry=1) at dlopen.c:87
#2 0x1000306c in before (me=0x407ff382 "/usr/mips64-unknown-linux-gnu/usr/bin/top") at top/top.c:3308
#3 0x10001a10 in main (dont_care_argc=<optimized out>, argv=0x407ff1a4) at top/top.c:5721
```
Reported-by: Matt Turner
Bug: https://bugs.gentoo.org/640130
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
|
| |
|
|
|
| |
Credit: Martin Mokrejš
Bug: 608234
|
| |
|
|
|
|
|
|
|
|
| |
Currently gcc on powerpc32 generates invalid code for
__builtin_return_address(0) calls. Normally programs
don't do that but malloc() hooks in glibc do:
Bug: https://gcc.gnu.org/PR81996
Bug: https://bugs.gentoo.org/629054
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
|
| | |
|
| |
|
|
|
|
|
| |
-freorder-blocks is enabled by default on -O and upper.
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
Signed-off-by: Matthias Maier <tamiko@gentoo.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The failure happens when live glibc-9999 ebuild is installed:
* QA Notice: Missing gen_usr_ldscript for libm-2.26.90.so
* ERROR: sys-libs/glibc-9999::gentoo failed:
* add those ldscripts
The problem here is how upstream glibc version is detected:
dosym ../../$(get_libdir)/libm-${PV}.so $(alt_usrlibdir)/libm-${PV}.so
Change to use 'version.h' to pick upstream version.
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
Signed-off-by: Matthias Maier <tamiko@gentoo.org>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
For gcc-6 and newer the old logic in the toolchain-glibc eclass:
if use hardened && gcc-specs-pie ; then
append-cppflags -DPIC
else
filter-flags -fPIE
fi
is obsolete. Simply disable the check.
|
| | |
|
| |
|
|
| |
Signed-off-by: Matthias Maier <tamiko@gentoo.org>
|
| |
|
|
|
|
|
|
|
|
| |
-fstack-protector-all (bug #609048).
configure accepts --enable-stack-protector=... option which results
in build system passing appropriate -fstack-protector... option
when possible.
Signed-off-by: Matthias Maier <tamiko@gentoo.org>
|
| | |
|
| |
|
|
|
|
| |
Use local variable to limit the scope of ABI variable override. This is
a builtin bash solution that is reliable and simple, unlike the complex
evar_* logic that unnecessarily pollutes global variables.
|
| |
|
Michał Górny
Marty E. Plummer
Sergei Trofimovich
Benda Xu
Andreas K. Hüttel
Matthias Maier
Arfrever Frehtes Taifersar Arahesis
GitWeb