Skip to contentSkip to conversation

cvs glibc bug?

DISCUSSION Benjamin Menküc 3 participants 4 messages May 16, 2003 – May 20, 2003 MIPS

lore
Share & tools
lore
#0 lore raw

Hi,

is this a bug?

[benmen@linuxpc1 mipsel-glibc] LC_ALL=C LD_LIBRARY_PATH="" CFLAGS="-g -finline-limit=10000 -O2" CC="mipsel-linux-gcc" AS="mipsel-linux-as" ../glibc-cvs/configure --build=i686-linux --host=mipsel-linux --enable-add-ons --prefix=/home/benmen/mipsel --with-headers=/home/benmen/mips/kernel/mips-2.4.20/include --disable-profile

....

[benmen@linuxpc1 mipsel-glibc] LC_ALL=C mak

...

mipsel-linux-gcc -mabi=32 ../sysdeps/unix/sysv/linux/sigtimedwait.c -c -std=gnu99 -O2 -Wall -Winline -Wstrict-prototypes -Wwrite-strings
-finline-limit=10000 -g -I../include -I.
-I/home/benmen/mips/mipsel-glibc/signal -I.. -I../libio -I/home/benmen/mips/mipsel-glibc -I../sysdeps/mips/elf -I../linuxthreads/sysdeps/unix/sysv/linux/mips -I../linuxthreads/sysdeps/unix/sysv/linux -I../linuxthreads/sysdeps/pthread -I../sysdeps/pthread -I../linuxthreads/sysdeps/unix/sysv -I../linuxthreads/sysdeps/unix -I../linuxthreads/sysdeps/mips -I../sysdeps/unix/sysv/linux/mips/mips32 -I../sysdeps/unix/sysv/linux/mips -I../sysdeps/unix/sysv/linux -I../sysdeps/gnu -I../sysdeps/unix/common -I../sysdeps/unix/mman -I../sysdeps/unix/inet -I../sysdeps/unix/sysv -I../sysdeps/unix/mips/mips32 -I../sysdeps/unix/mips -I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/mips/mips32 -I../sysdeps/mips -I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754/dbl-64 -I../sysdeps/wordsize-32 -I../sysdeps/mips/fpu -I../sysdeps/ieee754 -I../sysdeps/generic/elf -I../sysdeps/generic -nostdinc -isystem /home/benmen/mipsel/lib/gcc-lib/mipsel-linux/3.2.3/include -isystem /home/benmen/mips/kernel/mips-2.4.20/include -D_LIBC_REENTRANT -include
../include/libc-symbols.h -DPIC -o
/home/benmen/mips/mipsel-glibc/signal/sigtimedwait.o -MD -MP -MF /home/benmen/mips/mipsel-glibc/signal/sigtimedwait.o.dt ../sysdeps/unix/sysv/linux/sigtimedwait.c: In function `do_sigtimedwait':
../sysdeps/unix/sysv/linux/sigtimedwait.c:44: `SI_TKILL' undeclared (first use in this function)
../sysdeps/unix/sysv/linux/sigtimedwait.c:44: (Each undeclared identifier is reported only once
../sysdeps/unix/sysv/linux/sigtimedwait.c:44: for each function it appears in.)
make[2]: *** [/home/benmen/mips/mipsel-glibc/signal/sigtimedwait.o] Error 1 make[2]: Leaving directory `/home/benmen/mips/glibc-cvs/signal' make[1]: *** [signal/subdir_lib] Error 2
make[1]: Leaving directory `/home/benmen/mips/glibc-cvs'
make: *** [all] Error 2

regards,

Ben

#1 lore raw
Quoted 3 lines · Benjamin Menküc

Hi,

is this a bug?

You could argue so.

[snip]

Quoted 3 lines

../sysdeps/unix/sysv/linux/sigtimedwait.c: In function `do_sigtimedwait':
../sysdeps/unix/sysv/linux/sigtimedwait.c:44: `SI_TKILL' undeclared (first use in this function)

SI_TKILL is new in glibc and not yet ported to mips.

Thiemo

#2 lore raw

can I disable SI_TKILL stuff somehow so that glibc compiles for mips?

regards,

Ben

#3 lore raw
In reply to Thiemo Seufer
Quoted 1 line · Thiemo Seufer

SI_TKILL is new in glibc and not yet ported to mips.

Well, the kernel's include/asm-mips/siginfo.h defines SI_KILL as -6 while glibc defines SI_ASYNCNL as -6. Is there a reason for this or is this just a typo, all other architectures define the later as -60. If it's a typo i'd like to keep this in sync with other architectures:

Index: sysdeps/unix/sysv/linux/mips/bits/siginfo.h
===================================================================
RCS file: /cvs/glibc/libc/sysdeps/unix/sysv/linux/mips/bits/siginfo.h,v retrieving revision 1.9

unix/sysv/linux/mips/bits/siginfo.h +3 -1
diff -u -p -u -r1.9 siginfo.h
@@ -119,8 +119,10 @@ typedef struct siginfo
    signals.  */
 enum
 {
-  SI_ASYNCNL = -6,		/* Sent by asynch name lookup completion.  */
+  SI_ASYNCNL = -60,		/* Sent by asynch name lookup completion.  */
 # define SI_ASYNCNL	SI_ASYNCNL
+  SI_TKILL = -6,		/* Sent by tkill. */
+# define SI_TKILL	SI_TKILL
   SI_SIGIO,			/* Sent by queued SIGIO. */
 # define SI_SIGIO	SI_SIGIO
   SI_MESGQ,			/* Sent by real time mesq state change.  */

Regards,
-- Guido