docs/known-issues.txt
author "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Tue Jun 24 16:24:09 2008 +0000 (2008-06-24)
changeset 611 eac4dc8da8a9
parent 472 86c2bdaad27c
child 671 7abb354d4570
permissions -rw-r--r--
New patches from Ioannis E. VENETIS to allow building more up-to-date Alpha x-compilers.
Some patches are still missing, though.
See: http://sourceware.org/ml/libc-help/2008-06/msg00061.html

/trunk/patches/glibc/2.5.1/270-glibc-alpha-cfi.patch | 25 25 0 0 ++++++++++
/trunk/patches/glibc/2.5.1/280-glibc-alpha-sigsuspend.patch | 24 24 0 0 ++++++++++
/trunk/patches/glibc/2.5/270-glibc-alpha-cfi.patch | 25 25 0 0 ++++++++++
/trunk/patches/glibc/2.5/280-glibc-alpha-sigsuspend.patch | 24 24 0 0 ++++++++++
/trunk/patches/glibc/linuxthreads-2.3.6/270-glibc-linuxthreads-alpha-cfi.patch | 25 25 0 0 ++++++++++
/trunk/patches/gcc/4.2.0/940-gcc-alpha-signal_h.patch | 17 17 0 0 +++++++
/trunk/patches/gcc/4.2.1/940-gcc-alpha-signal_h.patch | 17 17 0 0 +++++++
/trunk/patches/gcc/4.3.0/940-gcc-alpha-signal_h.patch | 17 17 0 0 +++++++
/trunk/patches/gcc/4.2.2/940-gcc-alpha-signal_h.patch | 17 17 0 0 +++++++
/trunk/patches/gcc/4.3.1/940-gcc-alpha-signal_h.patch | 17 17 0 0 +++++++
/trunk/patches/gcc/4.2.3/940-gcc-alpha-signal_h.patch | 17 17 0 0 +++++++
11 files changed, 225 insertions(+)
yann@469
     1
Glibc does not build: asm/page.h not found
yann@469
     2
yann@469
     3
As of linux-2.6.25, asm/page.h is no longer exported.
yann@538
     4
This means the glibc up to and including 2.6 will not build against headers
yann@538
     5
exported from linux-2.6.25 and up. Starting with glibc-2.6.1, patches are
yann@538
     6
available that solves the issue. Switch to at least 2.6.1 if you can, or try
yann@538
     7
to backport the fix (should be quite easy).
yann@472
     8
yann@472
     9
----
yann@472
    10
dmalloc does not build
yann@472
    11
yann@538
    12
This broken build has been observed for big endian ARM with glibc and NPTL,
yann@472
    13
and with i686 with glibc and NPTL.
yann@472
    14
yann@472
    15
This has not been fully investigated yet. There is a name clash with strdup,
yann@472
    16
where it is a macro somewhere, and dmalloc tries to re-declare a strdup
yann@472
    17
function in dmalloc.h, but the declaration is overwritten with the macro
yann@472
    18
expansion, and thus the build is broken.
yann@472
    19
yann@472
    20
The common denominator seems to be NPTL.