Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
82,532 Commits 606 Branches 0 Tags 3.1 GiB
cca727ab8c
Commit Graph

729 Commits

Author SHA1 Message Date
is
a3a6f324e6 Use dbeq loops, code submitted by Hiroshi Horimoto in PR 3223, integrated 
and tested by Allen Briggs.
Needs 1/3 less time on 68030, 45% less time on 68040.
XXX On 68060, 1/2 less time for the padding part, but 40% more time on the
copy part. I've stared at the 68060 User Manual, but can't understand why;
must be some wierd (branch) cache interaction.
 2000-11-27 22:35:30 +00:00
eeh
5d73d71676 Fix modf() (I hope. Can't be any worse than the old version.) 2000-11-01 23:32:41 +00:00
jeffs
4912e50a59 Make memcpy and memmove use bcopy assembly for mips. Tweak forward copy 
code for backward copy to make overlap copy faster.  This happens in
some SW like sleepycat db.  Use temp labels in order to not confused
the profiling tools.
 2000-10-10 21:51:54 +00:00
kleink
a253819635 Whitespace botch in previous, noticed by Leo Weppelman and Thorsten Frueauf. 2000-10-02 14:52:39 +00:00
kleink
bea4776498 _SYSCALL() argument order botch in previous. 2000-10-02 14:51:57 +00:00
kleink
82fd76bb06 Need an internal name for pipe(2). 2000-09-28 08:38:53 +00:00
msaitoh
f9b799b651 check LIBC_SCCS 2000-09-13 22:32:25 +00:00
thorpej
9c5afcb2f0 Move the Alpha _mcount() into a .S file so that it can be 
pre-processed, like other Alpha assembly files.
 2000-09-05 16:28:30 +00:00
eeh
d3312c2719 Fix layout of fields in jmp_buf. [N.B. we could make JMP_BUFLEN 9 if we want.] 2000-08-28 00:45:59 +00:00
joda
991b448f39 add newline at end of file, to please (the quite buggy) mkid 2000-08-22 12:24:17 +00:00
ad
136f4e3b32 %edx is free; use it instead of %ebx. 2000-08-07 14:46:52 +00:00
matt
09f029058b Fix some ELF/a.out lossage 2000-08-07 03:18:04 +00:00
tsubai
7272e1b36f s/minbrk/_C_LABEL(__minbrk)/. 2000-08-04 09:19:39 +00:00
tsubai
511556bf86 Change comment character from '#' to '!'. (binutils-2.10 doesn't like '#') 2000-08-04 08:47:03 +00:00
eeh
9c2bf87977 Need to reserve a full 64-bit stackframe, not a 32-bit one. 2000-07-26 00:53:50 +00:00
mycroft
01a39929fc Store the whole 64-bit pointer... 2000-07-25 20:15:40 +00:00
mycroft
fb41c8728c Use cpp comment style. 2000-07-25 15:14:45 +00:00
mycroft
c54917029b Make __dtoul() actually DTRT. The calling convention was totally wrong, and 
the `big' value should be 2^63, not 2^31.
Also, nuke __fixunsdfsi(), since this is never output by the sparc64 compiler.
 2000-07-25 04:26:12 +00:00
mycroft
4272204d8d Offset the %i2 stack slot from %sp, not %fp. Otherwise we may modify the 
caller's registers!
 2000-07-24 15:36:49 +00:00
mycroft
fc590aafb4 There's no good reason not to assume a 64-bit stack here. 2000-07-24 15:24:55 +00:00
mycroft
eded30220a Don't bother saving %f16-%f31, since we don't use them. 2000-07-24 07:30:54 +00:00
mycroft
b3e75eb4ed _errno -> __errno 2000-07-24 00:11:10 +00:00
mycroft
e740400671 Just use a store/load pair to get the pattern into the FP reg. This eliminates 
a bunch of cruft and avoids using a v9a instruction.
In addition, eliminate 8 of the fmovda's, which we are not using the result of
anyway.
Net result is that this should be faster in all cases.
 2000-07-23 20:34:07 +00:00
eeh
32ceb997a8 Fix stack BIAS issues. 2000-07-23 07:12:22 +00:00
eeh
e507cc60e6 Update some more assembly code for PIC issues. 2000-07-21 03:14:15 +00:00
eeh
067b7521b5 Fix PIC bugs. 2000-07-18 22:38:11 +00:00
bjh21
3b9cd873cb Revert previous change -- turned out I'd got my build environment in a twist. 2000-07-15 15:07:34 +00:00
bjh21
985033a7a6 Add fixunssfsi and fixunsdfsi. These are provided by libgcc2.c in GCC 2.95, 
but not in EGCS.

This version of fixunssfsi is almost exactly the arm32 version.  fixunsdfsi is
mostly based on fixdfsi, but with simplifications for 64-bit integers.

Only minimal testing has been applied so far.
 2000-07-15 13:26:50 +00:00
eeh
1b72cecb45 Block align pointer *AFTER* fixing up possible stack bias. 2000-07-08 16:02:15 +00:00
itohy
af83a97717 Add special treatment of shmat(2). 
On m68k ELF calling standard (__SVR4_ABI__), if a function
returns a pointer, the return value should go to a0 as well as d0.
 2000-07-07 08:20:50 +00:00
kleink
de9e80b40f Missed in previous: replace the previously used _brk label with an ordinary 
local label; problem noticed by Matthias Drochner.
 2000-07-04 07:37:33 +00:00
matt
7661e75ad4 Fix a major botch in 1.4 (which I did not do) which broke _brk and malloc. 2000-07-01 05:21:12 +00:00
matt
c3530f0c12 Fix some bugs and clean up for ELF & a.out support 2000-06-30 23:46:36 +00:00
mrg
6b9ed41192 sync with libkern version; <vm/vm.h> -> <uvm/uvm_extern.h> 2000-06-29 06:30:02 +00:00
matt
ae4364e1d3 Make these work with either a.out or ELF. 2000-06-28 19:20:16 +00:00
kleink
b8a4d7adf9 Need an internal name for fork(). 2000-06-26 06:32:54 +00:00
kleink
2234f31b26 Add a _SYSCALL(entry,syscall) macro like the one found on other 
architectures.
 2000-06-26 06:31:37 +00:00
kleink
15bc7ed349 Thinko in previous: we do always use an internal name. 2000-06-26 06:25:36 +00:00
kleink
03f996bd0c Add internal names for brk() and sbrk(). 2000-06-23 15:40:36 +00:00
kleink
4d8e5be65e Delete the leftover _brk entry; as discussed with Ragge. 2000-06-23 07:18:09 +00:00
cgd
db755e7c76 sweep of my licenses (userland files w/o only my copyright) for 
consistency.  (no functional changes)
 2000-06-14 17:24:02 +00:00
cgd
d220ca5ba3 fix up NetBSD RCS Ids to match the standard, and the leading comment as 
to match as well.  No functional changes.
 2000-06-14 06:48:47 +00:00
msaitoh
b2e2c7ec87 movstrSI*() for gcc -Os 2000-06-09 04:47:31 +00:00
tsubai
bba37c9174 Add softfloat glue for sh3. 2000-06-06 17:28:37 +00:00
bjh21
4472dbe5e3 Initial commit of SoftFloat 2a import. This should provide a sensible 
mostly-MI floating-point implementation for use by gcc -msoft-float.
It's currently only used by arm26, but should be usable by other ports
without too much hacking, assuming doubles and u_int64_ts are passed and
returned the same way, and FP formats are IEEEish.
 2000-06-06 08:14:49 +00:00
castor
5510a708c4 Make __cerror be LEAF_NOPROFILE. I relies upon having v0 and v1 be unchanged 
from the caller.   From Ethan Solomita <ethan@geocast.com>
 2000-05-29 15:07:41 +00:00
bjh21
6c97e2bd78 Initial commit of arm26 port 2000-05-09 21:55:44 +00:00
erh
b6ab3db38e Fix unaligned accesses more correctly. 2000-04-25 05:55:58 +00:00
erh
9396b0cf82 Fix unaligned accesses. 2000-04-25 05:54:33 +00:00
tsubai
dc6a892a44 Add missing files. 2000-04-20 13:48:14 +00:00
mark
a319e58c04 Fix alignment of __nanf and __infinity. Fixes PR9925. 2000-04-18 22:07:17 +00:00
mycroft
fedba1d813 Clean up. 2000-04-03 04:20:44 +00:00
mycroft
72e604e414 Clean up. 
* _REENTRANT is always defined.
* The vfork() stubs can use cerror.
 2000-04-03 04:13:36 +00:00
mycroft
7f7027bb59 Repair syscall stubs that were not updated when the cerror interface was 
modified.
 2000-04-03 03:13:07 +00:00
msaitoh
9c587bcf2f fill delay slots (from Klaus Klein) 2000-03-27 16:27:16 +00:00
mycroft
78a4ddbd60 Do a little rearrangement here. No functional change, but it's a little 
clearer what's going on.
The assembler still whines about the lack of `.cprestore' when profiling,
but I'm not going there right now.
 2000-03-23 04:58:59 +00:00
soren
89c5a767f8 Fix doubled 'the's. 2000-03-13 22:55:05 +00:00
kleink
e4d7c2e329 Add ISO C99 long long integer general utility interfaces; partially addresses 
PR standards/9482.
 2000-03-06 18:32:22 +00:00
shin
e68fefb855 remove unused files (MIPS SOFTFLOAT). 2000-03-05 05:45:46 +00:00
shin
ac9896415f remove MIPS SOFTFLOAT. 2000-03-05 05:41:03 +00:00
mycroft
19b51e87a9 Remove pointless #ifdef. 2000-02-25 17:40:30 +00:00
kleink
7b61131dba Missed in end -> _end cleanup. 2000-02-24 15:37:34 +00:00
kleink
a379081935 For consistency with all other ports, clear errno in the prologue to the 
syscall here as well - while the manual page advises applications to clear
errno themselves their actual behaviour may be questionable, especially in
light of the other ports' implementations.
 2000-02-23 20:16:57 +00:00
mycroft
9ff0891c83 Fix bogons. 2000-02-22 03:28:04 +00:00
mycroft
5256a370bf Do the minimum work to make this compile with -Werror. 2000-02-22 03:14:20 +00:00
mycroft
32d50d3388 Do the minimum work to make this compile with -Werror. 2000-02-22 01:18:28 +00:00
mycroft
d018facbbf A better fix for the stub-in-other-module problem: 
Load the address of cerror through the GOT, so we can do a PIC_EPILOGUE and
leave the stack clean.
 2000-02-19 21:04:58 +00:00
kleink
b94df6841c (Re-)use the assembly version for labs() as well. 2000-02-06 13:14:51 +00:00
kleink
b9d73b3129 Implement *__errno() redirection, with help from Matt Thomas. 2000-02-05 21:02:51 +00:00
kleink
c54d3e28e2 Go via *__errno() here, too. 2000-02-05 19:53:54 +00:00
kleink
c3ed25e892 Implement *__errno() indirection. 2000-02-05 19:20:53 +00:00
kleink
fb2f23e243 Oops, need to save r2 and r3 as well. 2000-01-27 15:11:17 +00:00
kleink
9d7ea6523c Support *__errno() indirection. 2000-01-27 14:58:48 +00:00
kleink
1a3fa11c95 Support *__errno() indirection. 2000-01-24 21:57:39 +00:00
mycroft
5888ab27bd Make cerror() work correctly for stubs in other shlibs -- e.g. libposix. 2000-01-23 23:38:32 +00:00
mycroft
c26054b799 Delint. 2000-01-22 22:40:58 +00:00
kleink
02867b49c9 Nuke the assembly version, which has been unused for > 2 years and is actually 
identical to GCC's output from processing abs.c.
 2000-01-20 18:28:33 +00:00
kleink
6910891781 Add NAN constant glue for sh3. 2000-01-17 16:27:53 +00:00
kleink
34c7685f30 Adjust for little endian environments (confirmed by msaitoh). 2000-01-17 16:21:36 +00:00
kleink
5d03e70658 (Re-)use the assembly version for labs(), too. 2000-01-16 00:32:32 +00:00
msaitoh
df208e8ae8 sync with libkern 2000-01-14 22:55:46 +00:00
msaitoh
95d875fb90 libc for SH3. 
TODO: fplib
 2000-01-05 14:07:29 +00:00
eeh
0ba088f0a3 New bzero() using block store insns. 1999-12-30 15:31:39 +00:00
eeh
d1b815fe32 Sample code to implement ffs() using the popc insn (if it's ever implemented). 1999-12-30 15:30:26 +00:00
shin
509e885e3b fix name space pollution by adding prefix '_mips_'. 
closes PR 8770.
 1999-12-26 00:22:31 +00:00
kleink
11e6c54cfc C99: Define a NAN macro in <math.h> which evaulates to a constant expression of 
a single-precision quiet NaN; only to be defined on platforms that do support
this value.
 1999-12-23 10:15:05 +00:00
matthias
87bb3a7b2b make this work again with -mrtd. 1999-11-16 18:16:53 +00:00
is
eaf45b75c1 Fix spelling of contributors name 1999-10-27 09:10:27 +00:00
thorpej
3b435a7396 Add register prefixes. 1999-10-25 23:48:07 +00:00
itohy
1d0a433095 ELF PIC support. 1999-10-18 03:24:28 +00:00
itohy
29215efa45 ELF PIC support. 
Small optimization.
 1999-10-18 03:22:14 +00:00
shin
54de12eae6 remove extra lw (SOFTFLOAT) 1999-10-11 05:25:35 +00:00
shin
f761ddb744 adjust #endif position (SOFTFLOAT) 1999-10-11 05:21:06 +00:00
takemura
cdabc2eb91 Hpcmips needs completely FPU instruction free user land. 
XXX, this modification is kinda ad-hoc.
 1999-09-29 13:19:13 +00:00
kleink
30ae96c534 Oops, missed these in the mcount -> __mcount renaming. 1999-09-28 08:27:08 +00:00
is
e203553a11 Finish support for auto-generated libkern divsi3.S. 1999-09-21 09:43:38 +00:00
is
897dbf6c67 Use an identical source file for the libc and the libkern version of 
divsi3.S, conditionally compiling the divide by zero case. This way,
after a change to the one, we can just copy it over to update the other
without creating ... problems. By Chris G.  Demetriou.
 1999-09-20 09:22:47 +00:00
mycroft
959acd6d47 Looks like I forgot to add this copy of isnan.c. 1999-09-17 21:32:53 +00:00
takemura
2f85eb6142 Import NetBSD/hpcmips source files from hpcmips repository. 
file list:
        sys/arch/hpcmips/
        distrib/hpcmips/
        etc/etc.hpcmips/
        lib/libc/arch/mips/fplib/
        lib/libc/arch/mips/fplib/Makefile.inc
        lib/libc/arch/mips/fplib/environment.h
        lib/libc/arch/mips/fplib/fplib_glue.c
        lib/libc/arch/mips/fplib/fplib_libc.c
        lib/libc/arch/mips/fplib/hpcmips-gcc.h
        lib/libc/arch/mips/fplib/softfloat-macros.h
        lib/libc/arch/mips/fplib/softfloat-specialize.h
        lib/libc/arch/mips/fplib/softfloat.c
        lib/libc/arch/mips/fplib/softfloat.h
        lib/libc/arch/mips/gen/ieee.h
        lib/libc/arch/mips/gen/sf_fabs.c
        lib/libc/arch/mips/gen/sf_flt_rounds.c
        lib/libc/arch/mips/gen/sf_fpgetmask.c
        lib/libc/arch/mips/gen/sf_fpgetround.c
        lib/libc/arch/mips/gen/sf_fpgetsticky.c
        lib/libc/arch/mips/gen/sf_fpsetmask.c
        lib/libc/arch/mips/gen/sf_fpsetround.c
        lib/libc/arch/mips/gen/sf_fpsetsticky.c
        lib/libc/arch/mips/gen/sf_isinf.c
        lib/libc/arch/mips/gen/sf_ldexp.c
        lib/libc/arch/mips/gen/sf_modf.c
 1999-09-16 12:13:17 +00:00
is
bc086d24af In the division core: if the highest bit of the dividend is set, 
we have to be careful when shifting the divisor. Test this.

This fixes PR 7760 by Richard Earnshaw.
 1999-09-16 09:15:05 +00:00
drochner
ec2a7d4249 SYS.h was nuked from libkern 1999-09-10 15:41:51 +00:00
chs
f70d067294 use RODATA() instead of ENTRY() for __ffstab so that it works when profiling. 1999-09-06 19:00:24 +00:00
mycroft
b3b07a4c6e Another microoptimization. Take advantage of the addition already having set 
the flags for the <=0 comparison.
 1999-08-30 18:28:24 +00:00
mycroft
0560606383 Reuse the trick of loading 0.0 (rather than 1.0) and then modifying the 
exponent.  It's a bit faster on some machines.
 1999-08-30 18:01:40 +00:00
mycroft
5579a03f91 Nuke unused variable. 1999-08-30 16:48:07 +00:00
augustss
75d2dfe657 Remove more unused mulexp variables. 1999-08-30 15:11:32 +00:00
augustss
6a4d7f0bc4 Make it compile again by removing an unused variable. 1999-08-30 12:14:09 +00:00
mycroft
9cba4741ee Use C versions of isinf() and isnan(). The assembler code is basically 
identical, but was endian-dependent.
 1999-08-29 23:01:40 +00:00
mycroft
6b221be206 ldexp(denormal, exp>1023) would generate the wrong result in all non-overflow 
cases.  Totally rework this code to fix this bug *and* make it faster.
 1999-08-29 22:50:25 +00:00
mycroft
5f34ffabca ldexp(denormal, exp>1023) would generate the wrong result in all non-overflow 
cases.  Totally rework this code to fix this bug *and* make it faster.
 1999-08-29 22:45:32 +00:00
mycroft
c3140da584 Oops; committed wrong version here. 1999-08-29 19:42:54 +00:00
mycroft
f784c98496 Reimplement the FSCALE version as C-with-assembler. (Make the compiler deal 
with ABI crap.)
 1999-08-29 19:41:53 +00:00
mycroft
0ae41d0984 Use the C version of ldexp(). The assembler version was broken in several 
ways (ldexp(0.0, 5000) returned +Inf, didn't do normalization/denormalization,
etc.).
 1999-08-29 19:08:44 +00:00
mycroft
02a48bde49 Don't ever return -0.0. 1999-08-29 19:04:02 +00:00
mycroft
0d708431b7 Minor style change. 1999-08-29 18:40:50 +00:00
mycroft
ea4b733810 Trivial KNF. 1999-08-29 18:39:35 +00:00
mycroft
6850b97bb3 Sync with SPARC version. 1999-08-29 18:35:51 +00:00
mycroft
ed01c49a3f Use the C version. The assembler version was just transliterated (and buggy) 
anyway.
 1999-08-29 18:32:36 +00:00
mycroft
099dd01a1d Make the C version of frexp() consitent across platforms, and handle the 
Inf/NaN case.
 1999-08-29 18:30:14 +00:00
mycroft
291053e1e2 Use ieee.h. 1999-08-29 18:23:26 +00:00
mycroft
c8190acc41 Separate isnan() into a separate module, and make isinf()/isnan() use ieee.h. 1999-08-29 18:11:28 +00:00
mycroft
17030e8106 Separate isnan() into a separate module, and make isinf()/isnan() use ieee.h. 1999-08-29 18:08:27 +00:00
thorpej
30dca7f7db Add __msgctl13(), __semctl13(), and __shmctl13() (symbols are __RENAME()'d 
in their respective header files).  Add PSEUDO stubs for the old versions
of these system calls.

Note that __semctl13() doesn't require the extra stub that the old
semctl()/__semctl() did, as the SUSv2 version takes variable arguments,
and thus does not require the extra indirection.
 1999-08-25 05:09:27 +00:00
kleink
74511f97b1 Use _ALIGN_TEXT from <machine/asm.h>. 1999-08-23 08:45:09 +00:00
danw
687108b42f add the ieeefp functions 1999-07-07 01:55:07 +00:00
danw
e8559a61eb mark an asm statement volatile so it doesn't get optimized out 1999-07-07 01:53:38 +00:00
pk
0ffda58ae9 A new implementation of _setjmp/_longjmp that uses the T_FLUSHWIN trap. 
This means the stack frame and program counter can restored directly
from the jmp_buf, eliminating the need to backtrack through frames
which does not work in all cases.
 1999-06-27 11:10:18 +00:00
tron
d7678457d6 Fix definition of infinity. Patch supplied by Richard Earnshaw in 
PR lib/7829.
 1999-06-21 22:31:32 +00:00
matthias
71dd06e171 Go for correctness instead of speed as suggested by Ian Dall. This fixes 
pr#port-pc532/7710.
 1999-06-06 20:20:46 +00:00
matthias
bb36354c10 Fix spurious page faults by double word aligning the source and not the 
destination. Thank's to Ian Dall for this fix.
 1999-06-06 19:49:11 +00:00
kleink
15fb0e4315 Shameless hack to reference end if using a.out and _end if using ELF. 1999-06-03 20:14:58 +00:00
kleink
5b4ea02521 _C_FUNC() -> _C_LABEL() 1999-05-06 14:01:20 +00:00
kleink
c66e4d7122 * Prefix softfloat symbols which require external linkage with _arm32_ to 
move them out of user namespace; declare those not requiring external linkage
static.
* In due course, disable those functions that require neither external nor
internal linkage.
Addresses PR port-arm32/7517.
 1999-05-06 13:56:28 +00:00
tsubai
cb9a32fdf9 * Add getcachelinesize() prototype. 
* Make this compile with -D_STANDALONE.
 1999-05-05 12:35:54 +00:00
kleink
5b2a4f0824 Call cerror through the PLT. 1999-05-02 23:19:49 +00:00
kleink
18d5ba6151 We can shave off a register-register move when using the SVR4 ABI: __errno() 
will return its pointer result in a0.
 1999-05-02 20:52:02 +00:00
kleink
4e0adc06d5 When using the SVR4 ABI, return pointer results in a0, too. 1999-05-02 20:34:04 +00:00
kleink
1bd013aa27 Oops, typo. 1999-05-02 20:18:17 +00:00
kleink
d658cafb30 When using the SVR4 ABI, just return floating point values in fp0. 1999-05-02 20:17:20 +00:00
kleink
bcd9993e4e When using the SVR4 ABI, return the pointer in a0, too. 1999-05-02 20:09:29 +00:00
ross
c5a82dd848 * New MI man page for fp{set,get}{round,sticky,mask}(3). 
* Implement fpgetsticky() for alpha.
* Direct fpsetsticky() and fp{get,set}mask() into alpha kernel via sysarch(2).
* Define new sysarch(2) stub for above and install and distribute sysarch.h
for alpha.  (The fpcr IS user mode r/w, but for reasons beyond the scope
of a commit message kernel calls are needed.) And much kernel Magick is
required before these do anything, but this way programs compiled under
1.4 will DTRT on future snapshots and releases.
 1999-04-30 00:58:30 +00:00
kleink
b74f892f6d Declare cerror using _ENTRY() to get, among other things, a .type statement. 1999-04-29 15:37:21 +00:00
ws
5423093850 Modify syncicache on PowerPC from an inline to a real function. 
Support different cache line sizes with the same object code in userland.
While here, move the function to implementation name space.
 1999-04-17 21:16:45 +00:00
ragge
765d11306b Set infinity to the highest D-float possible. 
XXX - infinity does not exist on vax...
 1999-04-17 14:53:32 +00:00
eeh
713a1d89ff Add _dtoul() which gcc generates a lot. Maybe it should go in another file. 1999-04-04 21:01:39 +00:00
kleink
dccf2a15e4 While in transition, conditionally rename _ASM_LABEL(cerror) and 
_ASM_LABEL(curbrk) to _C_LABEL(__cerror) and _C_LABEL(__curbrk), respectively,
depending on the object format chosen, to avoid user namespace clashes.
 1999-03-31 18:14:27 +00:00
kleink
5af7e7a902 Change references to end' to _end' as not to invade (or being invaded by) 
the user's name space.
 1999-03-31 11:53:07 +00:00
jonathan
cd6244f1cc Assembly-coded, word-at-a-time unrolled memset(). 1999-03-15 06:37:43 +00:00
mycroft
2b504929fa Just some minor cleanup. 1999-03-10 08:15:43 +00:00
mycroft
f1c544b6a8 Use ieee.h structs and macros. 1999-03-10 08:14:44 +00:00