AuroraMiddleware/glibc
1
0
Fork 0
mirror of https://sourceware.org/git/glibc.git synced 2024-11-26 15:00:06 +00:00
Code Issues Packages Projects Releases Wiki Activity
3079f652d7
glibc/sysdeps/x86_64/fpu/e_ilogbl.S

37 lines
613 B
ArmAsm
Raw Normal View History

Update. 2003-11-17 Jakub Jelinek <jakub@redhat.com> * posix/regcomp.c (optimize_utf8): Optimize multi-byte chars as well. * posix/bug-regex20.c (tests): Add new tests. Multi-byte char followed by dup operator is expected to be optimized. * posix/regexec.c (check_node_accept_bytes): Move nrules and j variables to the block where they are only used, initialize nrules only immediately before using it. 2003-11-15 Andreas Jaeger <aj@suse.de> * sysdeps/x86_64/fpu/s_scalbnl.S: New file. * sysdeps/x86_64/fpu/s_truncl.S: New file. * sysdeps/x86_64/fpu/s_nearbyintl.S: New file. * sysdeps/x86_64/fpu/s_floorl.S: New file. * sysdeps/x86_64/fpu/s_ilogbl.S: New file. * sysdeps/x86_64/fpu/e_remainderl.S: New file. * math/libm-test.inc (floor_test): Test also ±0.25. (ceil_test): Test -0.25.
2003-11-18 07:25:02 +00:00
/*
* Public domain.
*/
#include <machine/asm.h>
Fix ilogb exception and errno (bug 6794) [BZ #6794] Following Joseph comments about bug 6794, here is a proposed fix. It turned out to be a large fix mainly because I had to move some file along to follow libm files/names conventions. Basically I have added wrappers (w_ilogb.c, w_ilogbf.c, w_ilogbl.c) that now calls the symbol '__ieee754_ilogb'. The wrappers checks for '__ieee754_ilogb' output and set the errno and raise exceptions as expected. The '__ieee754_ilogb' is implemented in sysdeps. I have moved the 's_ilogb[f|l]' files to e_ilogb[f|l] and renamed the '__ilogb[f|l]' to '__ieee754_ilogb[f|l]'. I also found out a bug in i386 and x86-64 assembly coded ilogb implementation where it raises a FE_DIVBYZERO when argument is '0.0'. I corrected this issue as well. Finally I added the errno and FE_INVALID tests for 0.0, NaN and +-InF argument. Tested on i386, x86-64, ppc32 and ppc64.
2012-04-11 19:30:13 +00:00
ENTRY(__ieee754_ilogbl)
Update. 2003-11-17 Jakub Jelinek <jakub@redhat.com> * posix/regcomp.c (optimize_utf8): Optimize multi-byte chars as well. * posix/bug-regex20.c (tests): Add new tests. Multi-byte char followed by dup operator is expected to be optimized. * posix/regexec.c (check_node_accept_bytes): Move nrules and j variables to the block where they are only used, initialize nrules only immediately before using it. 2003-11-15 Andreas Jaeger <aj@suse.de> * sysdeps/x86_64/fpu/s_scalbnl.S: New file. * sysdeps/x86_64/fpu/s_truncl.S: New file. * sysdeps/x86_64/fpu/s_nearbyintl.S: New file. * sysdeps/x86_64/fpu/s_floorl.S: New file. * sysdeps/x86_64/fpu/s_ilogbl.S: New file. * sysdeps/x86_64/fpu/e_remainderl.S: New file. * math/libm-test.inc (floor_test): Test also ±0.25. (ceil_test): Test -0.25.
2003-11-18 07:25:02 +00:00
fldt 8(%rsp)
/* I added the following ugly construct because ilogb(+-Inf) is
required to return INT_MAX in ISO C99.
-- jakub@redhat.com. */
fxam /* Is NaN or +-Inf? */
fstsw %ax
movb $0x45, %dh
andb %ah, %dh
cmpb $0x05, %dh
je 1f /* Is +-Inf, jump. */
Fix ilogb exception and errno (bug 6794) [BZ #6794] Following Joseph comments about bug 6794, here is a proposed fix. It turned out to be a large fix mainly because I had to move some file along to follow libm files/names conventions. Basically I have added wrappers (w_ilogb.c, w_ilogbf.c, w_ilogbl.c) that now calls the symbol '__ieee754_ilogb'. The wrappers checks for '__ieee754_ilogb' output and set the errno and raise exceptions as expected. The '__ieee754_ilogb' is implemented in sysdeps. I have moved the 's_ilogb[f|l]' files to e_ilogb[f|l] and renamed the '__ilogb[f|l]' to '__ieee754_ilogb[f|l]'. I also found out a bug in i386 and x86-64 assembly coded ilogb implementation where it raises a FE_DIVBYZERO when argument is '0.0'. I corrected this issue as well. Finally I added the errno and FE_INVALID tests for 0.0, NaN and +-InF argument. Tested on i386, x86-64, ppc32 and ppc64.
2012-04-11 19:30:13 +00:00
cmpb $0x40, %dh
je 2f /* Is +-Inf, jump. */
Update. 2003-11-17 Jakub Jelinek <jakub@redhat.com> * posix/regcomp.c (optimize_utf8): Optimize multi-byte chars as well. * posix/bug-regex20.c (tests): Add new tests. Multi-byte char followed by dup operator is expected to be optimized. * posix/regexec.c (check_node_accept_bytes): Move nrules and j variables to the block where they are only used, initialize nrules only immediately before using it. 2003-11-15 Andreas Jaeger <aj@suse.de> * sysdeps/x86_64/fpu/s_scalbnl.S: New file. * sysdeps/x86_64/fpu/s_truncl.S: New file. * sysdeps/x86_64/fpu/s_nearbyintl.S: New file. * sysdeps/x86_64/fpu/s_floorl.S: New file. * sysdeps/x86_64/fpu/s_ilogbl.S: New file. * sysdeps/x86_64/fpu/e_remainderl.S: New file. * math/libm-test.inc (floor_test): Test also ±0.25. (ceil_test): Test -0.25.
2003-11-18 07:25:02 +00:00
fxtract
fstp %st
fistpl -4(%rsp)
fwait
movl -4(%rsp),%eax
ret
1: fstp %st
movl $0x7fffffff, %eax
ret
Fix ilogb exception and errno (bug 6794) [BZ #6794] Following Joseph comments about bug 6794, here is a proposed fix. It turned out to be a large fix mainly because I had to move some file along to follow libm files/names conventions. Basically I have added wrappers (w_ilogb.c, w_ilogbf.c, w_ilogbl.c) that now calls the symbol '__ieee754_ilogb'. The wrappers checks for '__ieee754_ilogb' output and set the errno and raise exceptions as expected. The '__ieee754_ilogb' is implemented in sysdeps. I have moved the 's_ilogb[f|l]' files to e_ilogb[f|l] and renamed the '__ilogb[f|l]' to '__ieee754_ilogb[f|l]'. I also found out a bug in i386 and x86-64 assembly coded ilogb implementation where it raises a FE_DIVBYZERO when argument is '0.0'. I corrected this issue as well. Finally I added the errno and FE_INVALID tests for 0.0, NaN and +-InF argument. Tested on i386, x86-64, ppc32 and ppc64.
2012-04-11 19:30:13 +00:00
2: fstp %st
movl $0x80000000, %eax /* FP_ILOGB0 */
ret
END (__ieee754_ilogbl)
Reference in New Issue Copy Permalink