samples/powerpc-e500v2-linux-gnuspe/reported.by
author "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Wed Oct 22 20:50:10 2008 +0000 (2008-10-22)
changeset 962 2ba4393e0441
parent 930 a98f0615133f
child 1076 faaf799d6b5e
permissions -rw-r--r--
Make Super-H finally compile a complete (C-only) toolchain:
- new, un-tested patches
- an sh4 sample to work on, and to try to reproduce later.

/trunk/patches/glibc/2.7/250-sh-chop-linux-version.patch | 49 49 0 0 ++
/trunk/patches/glibc/2.7/240-sh-lowlevellock-asm.patch | 56 56 0 0 +++
/trunk/patches/glibc/2.7/270-sh-fix-procfs.patch | 11 11 0 0 +
/trunk/patches/glibc/2.7/280-sh-fix-kernel-heders-location.patch | 23 23 0 0 +
/trunk/patches/glibc/2.7/260-sh-syscall-error-path.patch | 26 26 0 0 +
/trunk/samples/sh4-unknown-linux-gnu/crosstool.config | 311 311 0 0 ++++++++++++++
/trunk/samples/sh4-unknown-linux-gnu/reported.by | 3 3 0 0 +
7 files changed, 479 insertions(+)
yann@930
     1
reporter_name="Nate CASE"
yann@930
     2
reporter_url="http://sourceware.org/ml/crossgcc/2008-10/msg00016.html"
yann@935
     3
reporter_comment="This is a sample config file for Freescale e500v2 processors (e.g.,
yann@935
     4
MPC8548, MPC8572).  It uses eglibc (for e500/SPE patches) and a recent
yann@935
     5
gcc (4.3.1, for e500v2 DPFP support) and will generate appropriate
yann@935
     6
dual-precision floating point instructions by default.
yann@935
     7
yann@935
     8
Note: If building a Linux kernel with this toolchain, you will want to
yann@935
     9
make sure -mno-spe AND -mspe=no are passed to gcc to prevent SPE
yann@935
    10
ABI/instructions from getting into the kernel (which is currently
yann@935
    11
unsupported).  At this time, the kernel build system only passes
yann@935
    12
-mno-spe by default (this should be fixed soon hopefully).
yann@935
    13
yann@935
    14
A binutils snapshot is used to fix a bug present in 2.18 preventing
yann@935
    15
U-Boot from being compiled (CodeSourcery issue #2297: internal.h
yann@935
    16
(ELF_IS_SECTION_IN_SEGMENT): Adjust to cope with segments at the end of
yann@935
    17
memory)."