GOTO Masanori wrote:
Looks good to me. On the C3 box which kicked all this off:I put glibc_2.3.1-11 test packages with this i686/cmov experimental fix at:http://people.debian.org/~gotom/ Could you test them and report me whether it works or not. And please tell me the result of below checks: (1) # ldd `which ssh` (2) # LD_DEBUG=libs ldd `which ssh` I guess the result of (2) about libcrypt.so becomes for example: 01473: find library=libcrypto.so.0.9.6; searching 01473: search cache=/etc/ld.so.cache 01473: lib->hwcap(u)=40000 01473: lib->hwcap(l)=8000 01473: *hwcap=8001bf <- arch dependent 01473: *hwcap(u)=0 01473: *hwcap(l)=8001bf 01473: platform(u)=40000 01473: platform(l)=0 01473: _DL_HWCAP_PLATFORM(u)=f0000 01473: _DL_HWCAP_PLATFORM(l)=0 01473: trying file=/usr/lib/libcrypto.so.0.9.6 <- not cmov dir I concern that my dynamic load fix patch causes minor architecture not to load libraries, because it changes hwcap checking behavior. I already check it's ok on i386, however, -11 may cause another problem, but I would like to know such platform (sid is experimental distro, as you know :-). Yeah, but I think I do right fix.
# ldd `which ssh`
libutil.so.1 => /lib/libutil.so.1 (0x40016000)
libz.so.1 => /lib/libz.so.1 (0x4001a000)
libnsl.so.1 => /lib/libnsl.so.1 (0x40028000)
libcrypto.so.0.9.7 => /usr/lib/libcrypto.so.0.9.7 (0x4003b000)
libc.so.6 => /lib/libc.so.6 (0x4012d000)
libdl.so.2 => /lib/libdl.so.2 (0x4023d000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
# LD_DEBUG=libs ldd `which ssh` 2>&1 >/dev/null | grep -A 11 'find library=libcrypto'
00915: find library=libcrypto.so.0.9.7; searching
00915: search cache=/etc/ld.so.cache
00915: lib->hwcap(u)=80000
00915: lib->hwcap(l)=8000
00915: *hwcap=803135
00915: *hwcap(u)=0
00915: *hwcap(l)=803135
00915: platform(u)=80000
00915: platform(l)=0
00915: _DL_HWCAP_PLATFORM(u)=f0000
00915: _DL_HWCAP_PLATFORM(l)=0
00915: trying file=/usr/lib/libcrypto.so.0.9.7
This is the only 'sid' box I've got so I can't test the other cases.
Cheers,
Scott
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
