Skip to contentSkip to conversation

back to glibc-2.3.3 & gcc-3.4.1 pb

parisc-linux John David Anglin 2 participants 3 messages Sep 6, 2004 1 message not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw
Quoted 3 lines
Quoted 3 lines

../sysdeps/generic/unwind-dw2.c:1021: warning:
'fallback_frame_state_for' defined but not used rtld.c:730: warning: implicit declaration of function

This problem in serious and needs to be fixed. The function fallback_frame_state_for is needed. It is only called from an asm and gcc doesn't know about calls from asm's.

Dave

Signature · 6 lines
J. David Anglin                                  [email protected]
National Research Council of Canada              (613) 990-0752 (FAX: 952-6602)
_______________________________________________
parisc-linux mailing list
[email protected]
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux
#2 lore raw
Quoted 7 lines · John David Anglin
Quoted 3 lines
Quoted 3 lines

../sysdeps/generic/unwind-dw2.c:1021: warning:
'fallback_frame_state_for' defined but not used rtld.c:730: warning: implicit declaration of function

This problem in serious and needs to be fixed. The function fallback_frame_state_for is needed. It is only called from an asm and gcc doesn't know about calls from asm's.

It will only become important the day we need dwarf2 compat. That function is used as a fallback if 'frame_state_for' doesn't exist, and if libgcc_s.so.1 doesn't include '__frame_state_for', in which case we use a fallback onto the dwarf2 compat code in glibc.

In an sjlj environment fallback_frame_state_for just becomes abort if the above tests fail, and hence there is an extra fallback_frame_state_for that doesn't get used in the glibc build with an sjlj compiler.

See libc/sysdeps/generic/framestate.c for the fixes I made to this file.

c.

_______________________________________________
parisc-linux mailing list
[email protected]
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux

#3 lore raw
In reply to a message not in the archive
Quoted 11 lines
Quoted 11 lines

the weird in this last warning is that it's a very small fnct:
[...]
static inline void
__set_cr27(struct _pthread_descr_struct * cr27)

{
     asm(
                "ble    0xe0(%%sr2, %%r0)\n\t"
        "copy    %0, %%r26"
     : : "r" (cr27) : "r26" );
}

[...]

There might be a bug in gcc's size estimate for asm's. See the subsection "Size of an @code{asm}" in the gcc manual. This could be checked with a small test program that tries to inline the above.

Quoted 2 lines

244 static inline pthread_descr thread_self (void) __attribute__
((const,always_inline));

This is correct.

Dave

Signature · 6 lines
J. David Anglin                                  [email protected]
National Research Council of Canada              (613) 990-0752 (FAX: 952-6602)
_______________________________________________
parisc-linux mailing list
[email protected]
http://lists.parisc-linux.org/mailman/listinfo/parisc-linux