similar to: question about KLIBCSHAREDFLAGS

On 11/23/18 8:09 AM, eno wrote: > Dear experts, > > I'm new in klibc and currently studying the source codes. > I'm looking at the codes in usr/klibc/arch and can't figure out why > does, e.g. in x86_64, the text segment offset differ from system v > x86_64 abi (which is 4 Mb)? How does this value was calculated? > > Thanks in advance! > Nikita. It is set

This reverts commit 9d8d648e604026b32cad00a84ed6c29cbd157641, which broke signal handing on some architectures. On m68k and parisc, signal return depends on a trampoline that the kernel writes on the stack. On alpha, s390, and sparc (32-bit), we can avoid this by providing our own function as sa_restorer, but we currently don't. Signed-off-by: Ben Hutchings <ben at decadent.org.uk>

When using clang+llvm and lld, I get the following error: ld.lld: error: -Ttext-segment is not supported. Use --image-base if you intend to set the base address Let's test whether linker is lld and use --image-base instead of -Ttext-segment. See llvm's rationale for not supporting -Ttext-segment: https://github.com/llvm/llvm-project/commit/4dc2fb123dcfe9a97ad6f3a1135053b74efd0bc9 Note

From: Stanislav Fomichev <sdf at google.com> Clang's lld prefers the use of "-image-base" instead of "-Ttext-segment". Signed-off-by: Stanislav Fomichev <sdf at google.com> Signed-off-by: Bill Wendling <morbo at google.com> --- Makefile | 3 +++ usr/klibc/arch/i386/MCONFIG | 2 +- usr/klibc/arch/mips64/MCONFIG | 2 +-

On Sat, 2020-07-25 at 15:16 -0700, Bill Wendling wrote: > On Sat, Jul 25, 2020 at 2:56 PM Ben Hutchings <ben at decadent.org.uk> wrote: > > On Sat, 2020-07-25 at 14:36 -0700, Bill Wendling wrote: > > [...] > > > I upstreamed a few patches to klibc to fix a few linker errors. Some > > > of these might help: > > > > I applied v3 of your patch

This is a series of patches for clang compatibility: - Using flags needed flags and removing unsupported flags. - Adding support for clang's LLD linker. - Removing a variety of warnings. Bill Wendling (3): [klibc] Kbuild: use "libc.a" with clang [klibc] Kbuild: Add "-fcommon" for clang builds [klibc] Clean up clang warnings Michael Davidson (1): [klibc] Kbuild:

- Renamed IMAGE_BASE to LD_IMAGE_BASE_OPT. - Moved "-fcommon" to KLIBCREQFLAGS in scripts/Kbuild.klibc. - Remove "dash" warning fixes which have been upstreamed. - Conditionalize the inclusion of compiler flags by using the proper compiler name or "cc-option". - Added "-Werror" to "cc-option" to catch flags that cause warnings. - Retain

Commit-ID: 47e429ec6c758b98a00e69df541448942c6f3ba3 Gitweb: http://git.kernel.org/?p=libs/klibc/klibc.git;a=commit;h=47e429ec6c758b98a00e69df541448942c6f3ba3 Author: Ben Hutchings <ben at decadent.org.uk> AuthorDate: Sat, 19 Jan 2019 22:47:18 +0000 Committer: Ben Hutchings <ben at decadent.org.uk> CommitDate: Sat, 19 Jan 2019 22:51:00 +0000 [klibc] Build and install shared

while this doesn't fix the problem that shared doesn't work there, it at least ensures that klibc.so and (e.g.) sh.shared do not try to occupy the same virtual addresses mips64 is also missing KLIBCSHAREDFLAGS and needs treatment Signed-off-by: Thorsten Glaser <tg at mirbsd.org> --- usr/klibc/arch/ia64/MCONFIG | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git

This series fixes link/load issues I found on i386 with Debian's toolchain (gcc 8 with PIE as default, binutils 2.31). Ben. Ben Hutchings (2): [klibc] i386: Use -Ttext-segment to avoid address collision [klibc] Disable PIE scripts/Kbuild.klibc | 3 ++- usr/klibc/arch/i386/MCONFIG | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) -------------- next part -------------- A

Hi, I noticed that klibc started crashing on 64-bit MIPS and in my quest to fix the bug I got a bit carried away and fixed a few other things as well. Here are various miscellaneous MIPS patches, although the first patch is the important one. Thanks, James *** BLURB HERE *** James Cowgill (5): mips64: compile with -mno-abicalls mips: use -Ttext-segment when linking shared library

We currently use KLIBCKERNELSRC and KLIBCKERNELOBJ variables which used to be required to point to a kernel source and object directories. However the kernel headers were reorganised starting in Linux 3.7 such that that doesn't work properly. This series removes KLIBCKERNELOBJ and changes the use and documentation of KLIBCKERNELSRC to be consistent. Ben. Ben Hutchings (4): [klibc]

Hello! I have encountered a problem when building shared klibc with recent binutils versions - apparently the breakage happened since binutils 2.17.50.0.2 due to the following change: http://sources.redhat.com/ml/binutils/2006-05/msg00279.html http://sourceware.org/ml/binutils/2006-05/msg00466.html --- a/binutils/bfd/elf64-x86-64.c +++ b/binutils/bfd/elf64-x86-64.c @@ -3630,7 +3630,8 @@ static

In the v3.6 kernel, many of the headers were split out into a uapi variant. klibc should be using those variants to compile. There is also a generated subdir, to seperate out generated headers. The old kernel header location needs to be included as well, because linux/compiler.h is available only in the base include directory. Signed-off-by: Robin H. Johnson <robbat2 at gentoo.org> ---

Ugh, I was upgrading the AD server running on a CentOS 6.8 which uses named as its back-end. I have been running it for years with no problems. Today after upgrading bind named will not start. I get this error: Sep 28 23:32:25 nikita named[6369]: ---------------------------------------------------- Sep 28 23:32:25 nikita named[6369]: BIND 9 is maintained by Internet Systems Consortium, Sep 28

This fixes a long standing issue where it was not possible to do "make usr/klibc/arch/x86_64/longjmp.o" in the kernel. The principle is that all .o files to be part of klibc are listed with klib-y. For each directory a klib.list file is made that specify all .o file and the final AR then adds all .o files to create libc.a. This patch introduce the infrastructure and converts x86_64 to

The following patchset implements 3 seperate series of changes. External Building ================= Patches 1 through 8 enable to use of klibc's build system while leaving the src tree pristine (and potentially read only). Specifically: - srctree=<Sources for klibc> - objtree=<Ouput directory for klibc> - KLIBCKERNELSRC=<Kernel sources> - KLIBCKERNELOBJ=<Kernel

? KLIBCCC usr/dash/eval.o usr/dash/eval.c: In function 'evaltree': usr/dash/eval.c:277:19: warning: logical not is only applied to the left hand side of comparison [-Wlogical-not-parentheses] ?? if (!exitstatus == isor) ?????????????????? ^ Not sure what the fix should be: ?? if (!(exitstatus == isor)) Or ?? if ((!exitstatus) == isor) Any idea ? Christophe

On Sun, 2 Oct 2016 22:01:32 -0600 "Paul R. Ganci via samba" <samba at lists.samba.org> wrote: > > > On 10/02/2016 07:57 PM, Paul R. Ganci via samba wrote: > > > > > > On 10/02/2016 06:15 PM, Paul R. Ganci via samba wrote: > >> On 09/11/2016 10:38 AM, Paul R. Ganci via samba wrote: > >> > >>> On 09/11/2016 01:23 AM, Rowland

I have updated the klibc patchset based on feedback received. In particular, the patchset has been reorganized so as not to break git-bisect. Additionally, this updates the patch base to 2.6.17-git12 (d38b69689c349f35502b92e20dafb30c62d49d63) and klibc 1.4.8; the main difference on the klibc side is removal of obsolete code. This is also available as a git tree at: