46fd488dd6
Cast to unsigned int before doing shift operations, to appease lint.
2012-03-20 10:51:23 +00:00
a1a4d3d82c
Sprinkle some /* ARGSUSED */ before the dummy functions to appease lint.
2012-03-20 10:40:48 +00:00
5d44f1d51d
Convert to C89 definitions.
...
Cleanup up type usage.
2012-03-19 22:23:10 +00:00
1af1a7c82f
fix lint
2012-03-18 16:26:34 +00:00
0abf123a71
delint
2012-03-17 21:35:06 +00:00
e9be25776b
Avoid lint warnings
2012-03-17 20:48:58 +00:00
90310d68b0
Rename __zero to __sf128_zero to avoid clashes with the strange new
...
inline function in cdefs.h
2012-03-14 17:03:10 +00:00
938140d192
Use the _end symbol rather than the end symbol. Prevents version info
...
problems in elflink.c for the heimdal libraries.
2012-03-14 14:18:10 +00:00
fc3d31db9e
PowerPC doesn't have CERROR since it never had a.out.
...
Use _C_LABEL(__cerror) instead.
2012-02-27 15:47:11 +00:00
501876fe73
Fix pasto.
2012-02-27 14:48:34 +00:00
f46fb1259f
Make __cerror protected to restore ABI leak in older libc versions,
...
where e.g. librt referenced the copy from libc. This still allows libc
to use PC-relative jumps.
2012-02-27 12:26:21 +00:00
62335767ce
Add workaround for ARM cores that don't properly implement the MRC instruction
...
used for reading the thread pointer. This is the corresponding change which
was made to <arm/mcontext.h>
2012-02-14 17:32:43 +00:00
2b9ee4cb40
let one bit more through to SSE, to make FP_X_IMP work
2012-01-19 16:37:18 +00:00
d331b836f8
define _UC_TLSBASE and use it to pass the TLS pointer to setcontext().
...
since there is no available space in ucontext_t on vax, pass the
TLS pointer on the stack referenced by the ucontext_t instead.
suggested by joerg.
2012-01-07 16:47:42 +00:00
6133e96c53
Add TLS support for m68k.
2011-11-22 15:25:28 +00:00
452fee2663
gcc 4.5 on vax creates PLT stubs with an entry mask of 0xffc
...
rather than 0 like gcc 4.1 did, so the sneaky assembly functions
that "ret" without really returning now clobber their registers.
adjust these functions to avoid this problem.
2011-11-21 16:17:48 +00:00
d7796dcc07
Exploit hidden __cerror
2011-11-18 20:43:01 +00:00
4e5324e222
Mark __cerror as hidden and exploit that to simplify assembler.
2011-11-18 20:21:41 +00:00
4219a5c627
Explicitly mark __cerror as hidden
2011-11-18 20:17:46 +00:00
10dc456d4b
Hook up TLS support on ARM. Tested by martin@, kernel support from matt@
2011-11-18 16:10:02 +00:00
0b7c4e1b46
Everything else uses PIC branches to __cerror except those two...
2011-10-29 16:03:33 +00:00
157b9e4ee9
Sync with other ports: in PIC mode make __cerror hidden; it is not supposed
...
to be used outside libc.
2011-10-29 16:03:03 +00:00
a297cfac57
Fix fpsetmask(3) to be able to clear bits previously set.
2011-10-19 15:24:58 +00:00
a92b0ce490
don't use %edi!
2011-09-30 23:45:41 +00:00
6dd90d6d37
make rounding work
2011-09-30 23:42:00 +00:00
26ca52544d
PR/44293: Paul Goyette: Fix the mess of the rounding code.
2011-09-30 17:42:34 +00:00
64eddf5284
even simpler.
2011-09-29 22:55:45 +00:00
cfa5bcb221
even shorter
2011-09-29 22:53:13 +00:00
cfb62a8b7a
add a simpler version
2011-09-29 22:46:00 +00:00
38179c4fc2
add a simpler version.
2011-09-29 22:45:20 +00:00
41d70a0317
Instead of using a PICGOT relocation for map, use just a pc-relative one
...
since it is guaranteed to be close enough. Fixes issue with binutils-2.21.1a
(probably a bug).
2011-09-28 15:11:42 +00:00
39079c5f77
Explicitly add back __resumecontext wrapper
2011-09-20 08:42:29 +00:00
0a2b758384
Mark _resumecontext as dead. ANSIfy all declarations for it.
2011-09-19 21:24:57 +00:00
f31d8e575a
rename sparc64 BLOCK_SIZE and BLOCK_ALIGN to have SPARC64_ prefixes.
...
for the assembler files, define the old names to the new names
since using the new names cause ugliness due to longer identifer
names, and reduces churn.
fixes build issues in dtv and vaguely makes <machine/psl.h> slightly
less name-space invasive.
2011-07-12 07:51:33 +00:00
85f1d6f3df
Don't use "f" with non-fp types. Use a union to go between
...
double and uint64_t. Makes clang happy.
2011-07-10 21:18:47 +00:00
23282762ff
Fix so that float128_to_uint64_round_to_zero is defined for sparc64.
...
(isn't tha a namespace violation?)
2011-07-10 04:52:23 +00:00
f73fcccaa1
Don't build tf (long double) if building for O32.
2011-07-09 04:12:01 +00:00
cee1c10ee1
Add ieee754 versions of the compiler runtime functions
...
__fixuns{sf,df,tf}{si,di}.
Add an ieee754 version of the compiler runtime function __floatuntfdi
2011-07-09 02:30:27 +00:00
85c07878cd
Don't define a new enum. Just #define round_* to their FP_Rx instead.
2011-07-07 07:14:57 +00:00
c2405fa58b
Add __floatunsidf __floatunsisf __floatunsitf routines.
...
XXX i think they are correct but not sure.
2011-07-04 08:02:34 +00:00
80229e0feb
Fix softfloat for mips*el
2011-06-27 06:40:26 +00:00
5ed322d90a
apply some -Wno-error/-Wno-enum-compare/-Wno-array-bounds.
...
this should be looked at more closely.
2011-06-22 05:23:53 +00:00
bdd01b22a1
Functions that return a struct or union, must set %eax to the value
...
of the original address of the callers area before it returns.
(System V ABI, i386 supplement, fourth edition, page 3-14)
2011-06-12 20:29:25 +00:00
f3fc9a4693
Use excb before and after the mt_fpcr/mf_fpcr instructions as specified
...
in the alpha ARM 4.7.8.1 (Accessing the FPCR):
Because Alpha floating-point hardware can overlap the execution of
a number of floating-point instructions, accessing the FPCR must
be synchronized with other floating-point instructions. An EXCB
instruction must be issued both prior to and after accessing the
FPCR to ensure that the FPCR access is synchronized with the
execution of previous and subsequent floating-point instructions;
otherwise synchronization is not ensured.
Use unions to convert between double and uint64_t to avoid casting.
2011-06-12 05:44:36 +00:00
a4fc3a054f
make sure the infinity returned by HUGE_VALL has the "implicit"
...
bit set, otherwise it is invalid
This code is unlikely to be hit because gcc (and clang) use
their builtins for these special values.
2011-06-06 17:02:29 +00:00
33c7476cb2
Revert previous and make frach be 0xc0000000 for NAN's generated by strtold()
...
as suggested by tron.
2011-06-05 14:43:12 +00:00
093aa7c0f7
Make NAN tests consistent with other platforms and FreeBSD.
2011-06-04 15:45:41 +00:00
178153b0c2
use _ALIGN consistently.
2011-05-25 02:11:16 +00:00
bde00486af
make this compile with DEBUG (_ALIGN was not defined).
2011-05-24 22:39:47 +00:00
b53ad3df44
Use a plain float constant, no need to deal with messy 0r prefix
2011-05-22 20:11:47 +00:00
9af4d11fa4
Use "_end" instead of "end" for consistency with other architectures.
2011-05-13 23:14:36 +00:00
251de2dee0
Save a few global registers during set/longjmp - at least the application
...
registers and the only currently used system register (%g7).
Sparc now passes the setjmp tests (and should be able to build perl again).
2011-04-30 23:41:12 +00:00
9881f8de04
Fix previous: the way we calculate, we need to test the size of sigjmp_buf
2011-04-30 23:39:08 +00:00
c670087695
Add another compile time assert: catch overflows of jmp_buf
2011-04-30 23:30:27 +00:00
b549c67469
Save/restore some global registers (sigsetjmp compatible). Does not realy
...
make a difference currently, but is a hard to spot blackhole if something
ever starts using the application registers.
2011-04-30 19:43:17 +00:00
44283e6e19
Rearange jmp_buf content again: storing the sigsetjmp() save_mask midway
...
inside the environment is not a good idea, we may overwrite it when
fetching other details of the sigjmp environment.
Instead move it to the very end again.
Do not bother to save volatile (caller saved) global registers.
Fixes PR port-sparc64/44918.
2011-04-30 19:39:38 +00:00
e148e1968d
Use correct CALLFRAME_SIZ depending on ABI.
2011-04-29 08:05:22 +00:00
d21307a5d2
Fix offsets used in the assembly code to save global registers into a
...
jmp_buf, add a compile time assert to catch this kind of divergence
in the C code.
Move the fixed "%npc = %pc + 4" computation to longjmp, this frees
the sigcontext sc_npc value. Use this space to save the savemask
for siglongjmp (we have run out of space behind the globals).
Fixes the tests/lib/libc/setjmp/t_threadjmp tests, and also hopefully
fixes PR port-sparc64/44902.
2011-04-27 21:08:48 +00:00
e9f4b70f7d
rm -f ${.TARGET} before trying to > write to it; some how mine ended up
...
being mode 444 and the build failed.
2011-04-12 03:35:47 +00:00
037499057e
make infinityl match what gdtoa thinks.
2011-04-10 13:51:18 +00:00
962e9ad63f
fix long double infinity.
2011-04-06 21:55:36 +00:00
ecf8c9b0f0
ABI police: avoid cluttering global registers %g2 and %g7. Not quite there
...
yet for %g2, but the target is not very well defined. %g7 should be clean
now.
2011-03-28 11:19:12 +00:00
2587e6e437
add fpgetprec/fpsetprec
2011-03-26 19:51:41 +00:00
d2c615416b
Avoid using %g7
2011-03-23 20:54:35 +00:00
0c3a269fd8
Remove the sparc-v7 compatibility wrapper functions (.div, .mul, ... .urem),
...
the compiler never generates (or did generate in the past) 64bit code to call
them, and they are still present in our 32bit compat library.
2011-03-22 12:24:53 +00:00
a46656e6ee
Add ld formats for the unit-tests.
2011-03-20 23:16:07 +00:00
c5d79d45e3
Properly support the MIPS TLS ABI.
2011-03-15 07:40:18 +00:00
69358e3723
Add PPC TLS support. Add -D_LIBC_SOURCE for tls.c and _lwp.c (powerpc)
...
so that __lwp_gettcb_fast and __lwp_settcb inlines are define.
2011-03-12 07:55:09 +00:00
7e1a38b9b3
Reorder insns to fix more load delay hazard botches on R3000.
...
Fixes floating point exceptions in _longjmp() during /etc/rc and
now NWS-3470D boots up to multi user with -current userland binaries.
Also fill BDslots in error path properly.
2011-03-09 16:10:29 +00:00
49f1dcc8e5
While we use hardware (for float and double), but cooperate with softfloat
...
(to do long double), we need to not only handle the hardware exception
mask and cummulated flags, but also update/query the softfloat variables.
Since this may go away sometime (to make us psABI compliant), ifdef it
properly.
2011-03-06 10:32:47 +00:00
74ed8d21f3
fix my whitespace copypaste
2011-02-25 11:55:28 +00:00
ba7613308e
Fix LE softfloat (mostly theoretical) and comment.
...
from Alessandro Forin
2011-02-25 11:53:07 +00:00
1631a78097
Allow storing and receiving the LWP private pointer via ucontext_t
...
on all platforms except VAX and IA64. Add fast access via register for
AMD64, i386 and SH3 ports. Use this fast access in libpthread to replace
the stack based pthread_self(). Implement skeleton support for Alpha,
HPPA, PowerPC, SPARC and SPARC64, but leave it disabled.
Ports that support this feature provide __HAVE____LWP_GETPRIVATE_FAST in
machine/types.h and a corresponding __lwp_getprivate_fast in
machine/mcontext.h.
This material is based upon work partially supported by
The NetBSD Foundation under a contract with Joerg Sonnenberger.
2011-02-24 04:28:41 +00:00
2599cc89ba
Don't use r0 to restore LR, use r7 instead.
2011-01-29 02:21:20 +00:00
5f7e80a834
Add support for the Extensible MIPS ("eMIPS") platform. The
...
NetBSD/emips port runs on Xilinx and Beecube FPGA systems and the
Giano system simulator.
eMIPS is a platform developed at Microsoft Research for researching
reconfigurable computing. eMIPS allows dynamic loading and scheduling
of application-specific circuits for the purpose of accelerating
computations based on the current workload.
NetBSD eMIPS support for NetBSD 4.x was written at Microsoft Research
by Alessandro Forin and Neil Pittman. Microsoft Corporation has
donated full copyright to The NetBSD Foundation.
Platform support for eMIPS is the first part of Microsoft's
contribution. The second part includes the hardware accelerator
framework and will be proposed on tech-kern soon.
2011-01-26 01:18:43 +00:00
787e823fd1
Move IDs into .ident
...
Use END(x)
Remove a.out support
Make CERROR hidden
2011-01-25 02:38:15 +00:00
86b3b196ed
Correctly define inta using a helper macro for the #
2011-01-23 14:08:53 +00:00
0da37b7b85
Remove CPUFLAGS?= since it's no longer needed.
2011-01-20 18:41:39 +00:00
8ef507a844
Fix some bugs introduced when switching to new GOT/PLT format.
2011-01-19 02:47:01 +00:00
57d93d8715
Cleanup the o32 compat build test.
2011-01-18 08:12:05 +00:00
dfe29b1db4
Change to deal with callframe changes
2011-01-18 01:23:24 +00:00
bd081d44b9
Make the MIPS N32/N64 ABIs properly support 128-bit long doubles. With this
...
change, we should be fully conformant with the N32 and N64 ABIs.
Add {fpclassify,infinity,isnan,ininf,signbit}l_ieee754.c back to lib/libc/gen.
Note that infinityl_ieee754.c will work with either 64-bit, 80-bit, or
128-bit long doubles.
2011-01-17 23:53:03 +00:00
7d92b19968
Make __cerror hidden.
2011-01-17 23:34:44 +00:00
51db1c5cf3
Kill _NOREGNAMES. Everything should be using %rX now. If it doesn't it
...
soon will.
2011-01-17 08:23:53 +00:00
b79441d450
Switch to the PIC_GOTSETUP/PIC_TOCSETUP macros for PIC code. Now libc
...
will be built to use secureplt by default.
2011-01-16 02:43:10 +00:00
cf88c3890b
Use END(foo) everywhere.
...
Make __cerror hidden.
Use non-PLT calls to __cerror.
Use assym.h when appropriate.
Use addi to adjust stack instead of loading it.
Add __RCSIDs
Force -D_NOREGNAMES for all .S files.
[this is all in preperation for secure plt support]
2011-01-15 07:31:11 +00:00
958d64ed82
Generate symbol constants for UC_GREGS_* and CALLFRAME*.
2011-01-14 06:15:27 +00:00
2c2c9e807e
Change __cerror to be marked hidden and change callers to avoid the PLT.
2011-01-14 06:12:16 +00:00
be1c859466
We can't assume syscall retains t0, unless it's documented as ABI.
2010-12-12 09:03:06 +00:00
f3f411dad2
brk(2) should update __curbrk, not __minbrk or garbage.
2010-12-12 09:00:37 +00:00
015f9bbf7a
Pullup fixes from matt-nb5-mips64.
...
This makes setjmp/longjmp work again.
2010-09-03 17:22:51 +00:00
4c92852a80
remove almost all the ability to build netbsd with an a.out target.
...
we're ELF now, and there are many missing checks against OBJECT_FMT.
if we ever consider switching, the we can figure out what new ones
we need but for now it's just clutter.
this doesn't remove any of the support for exec_aout or any actually
required-for-boot a.out support, only the ability to build a netbsd
release in a.out format. ie, most of this code has been dead for
over a decade.
i've tested builds on vax, amd64, i386, mac68k, macppc, sparc, atari,
amiga, shark, cats, dreamcast, landisk, mmeye and x68k. this covers
the 5 MACHINE_ARCH's affected, and all the other arch code touched.
it also includes some actual run-time testing of sparc, i386 and
shark, and i performed binary comparison upon amiga and x68k as well.
some minor details relevant:
- move shlib.[ch] from ld.aout_so into ldconfig proper, and cut them
down to only the parts ldconfig needs
- remove various unused source files
- switch amiga bootblocks to using elf2bb.h instead of aout2bb.h
2010-07-06 05:59:50 +00:00
3bdcb2e4b5
add a nop to provide enough space for the linker and long calls.
...
fixes link errors with 'cerror' and libposix.
from dennis.c.ferguson@gmail.com in PR#43010.
2010-03-22 02:17:23 +00:00
093c642fac
replace kernel-only 'stmd' and 'lmd' operations in userland with equivalent
...
sequences of code, as done in eg, powerpc64/gen/_setjmp.S rev 1.2.
from dennis.c.ferguson@gmail.com in PR#43009.
2010-03-22 01:39:11 +00:00
19499f3f8d
update all of these makefiles for the .so -> .pico change.
...
most of them are most harmless, but the libgcc parts are quite essential.
before this change, all the special rules for .pico files were not applied,
and exception handling wasn't enabled. this caused c++ exceptions not to
work on sparc64.
this fixes the build of boost-headers (it was correctly calling exception
support broken!), which in turn makes all the things that depend upon it
to actually work again on sparc64.
2010-03-14 00:26:09 +00:00
6fc217346b
Add memset to libkern on vax, moving existing memset.S to src/common.
...
While kernel source code is supposed to use <machine/macros.h> via
<sys/param.h> and pick an inline asm version from there, gcc itself
can generate direct calls to memset() for structure initialization, so
the symbol must be provided.
Fixes INSTALL kernel, for which -Os caused gcc to use memset() call to
initialize a struct in uvm_fault_internal().
2010-03-12 09:12:34 +00:00
56da4fa7b5
Move AMD64's bswap64 implementation from libc to src/common and share it
...
with the kernel.
2010-01-14 02:09:46 +00:00
5ddaffd107
Deal with PLABELs in _lwp_makecontext
2010-01-07 12:31:10 +00:00
42b55ff43a
Fix typo.
2009-12-14 02:53:52 +00:00
2288207161
Merge from matt-nb5-mips64.
2009-12-14 01:07:41 +00:00
09afef2063
Rename ${SRCS.{alpha,arm,...}.gen} to ${LSRCS.{alpha,arm,...}.gen} because
...
they ended up in ${LSRCS}, not ${SRCS}. ${SRCS.*} namespace will be used
for more useful things. No functional changes intended.
2009-12-06 07:12:17 +00:00
619e0bd90e
Set ${SRCS.softfloat}.
2009-12-06 07:05:50 +00:00
ad4d8ed92e
Add no/op routines for m68000 for the sticky, mask, and round settings.
2009-11-28 23:46:38 +00:00
32381fa0dd
Follow upstream license changes for files with Michael Shalayeff's
...
copyright. In most cases, this means dropping the 3rd and 4th clauses.
2009-11-03 05:07:25 +00:00
43f47d3c56
G/C bcopy.m4 and references to it.
2009-10-21 19:30:46 +00:00
c9394f466a
SystemV-R4 ABI for M68k returns pointers in %a0, so we have to make sure
...
that CERROR returns -1 in %a0 in addition to %d0 and %d1, to make functions
like mmap(2), mremap(2), shmat(2) or sbrk(2) return -1 in case of an error.
A side effect of this bug was a segfault caused by jemalloc, when mmap()
failed.
2009-10-03 22:28:33 +00:00
a245205b96
Change NO_SRCS lines to specify .S in order to exclude the .c file from
...
SRCS but add it to LSRCS (sort of makes sense!).
2009-08-11 17:30:43 +00:00
d3c3d024a7
bzero.S defines memset(), so exclude memset.c
2009-08-03 06:26:41 +00:00
af129c0bc8
bzero.S also defines memset(), so exclude memset.c
2009-08-03 06:25:34 +00:00
8190cfe5fa
Don't remove imaxdiv.c
2009-07-31 20:54:35 +00:00
10ad5ffa71
Add the 'abs', 'div' and 'erand48' sources to stdlib/Makefile.inc.
...
Change the arch files only include the .S files.
This adds imaxdiv() to some archs where it was missing.
2009-07-31 20:39:59 +00:00
f0a25cf095
Stop bcopy.c and bzero.c being compiled
2009-07-30 21:26:36 +00:00
0d34d14bb9
If SRCS contains any .S files then remove the corresponding .c file from
...
SRCS and add to LSRCS (for lint).
Change the 'string' Makefiles so that the arch/*/string/Makefile.inc
need only specify the .S files for that architecture and not the .c
files for all the files they don't override.
2009-07-30 20:57:15 +00:00
6dffe8d42b
Remove references to index.c and rindex.c (strchr.c and strrchr.c
...
now provide the definitions).
2009-07-18 09:44:30 +00:00
03c902aa70
Change all archs so that strchr.[cS] and strrchr.[cS] exist and generate
...
duplicate symbols for index() and rindex().
libc and kernel build ok for affected archs (i386, amd64, vax, sun3,
evbmips-eb) all seem ok.
(vax might lose an asm index() - but comments imply it is worse than the C)
2009-07-17 19:37:54 +00:00
9d87375b3e
Fix stack alignment for the thread stack. The AMD64 stack has 16 Bytes
...
alignment, but this is active before the return address is pushed onto
the stack. Fixes random crashes in va_start when the SSE registers are
saved.
2009-07-02 09:57:09 +00:00
c1faa994fd
Reduce diff to makecontext.
2009-07-01 16:54:41 +00:00
2a6e000238
add _lwp_makecontext for the vax.
2009-06-03 01:02:28 +00:00
5da54c35c9
Based on input from Eduardo Horvath (eeh@), replace 'stmd' and 'lmd',
...
which apparently don't exist as instructions, with a sequence of
std / ld instructions.
Also remove the attempted include of ldstm.S which we don't have.
2009-03-08 18:57:12 +00:00
c79c3eed1d
Make NAN a compile time constant (with help from the compiler). Suggested
...
by krister. Fixes PR 40695.
Make references to the old construct (which we can't remove for binary
compatibility reasons) emit a linker warning.
2009-02-22 01:34:01 +00:00
b85aa1ea53
fix -Wsign-compare issue
2009-02-16 03:23:29 +00:00
86223c379c
The explicit integer bit is "don't care" for infinity, and now that it's
...
included in ext_frach, it needs to be masked off when testing for an
infinity value. Fixes the ieeefp/infinity regression test on the 68060
which clears the explicity integer bit when loading an infinity value.
2009-01-30 07:00:45 +00:00
14c5fc5d63
Sprinkle the "memory" clobber register so the fpsr manipulation actually
...
happens.
Fixes the ieeefp/except regresion test.
2009-01-25 16:44:41 +00:00
461a86f9bd
merge christos-time_t
2009-01-11 02:45:45 +00:00
10b0d64e49
Remove __LIBC12_SOURCE__
2008-11-24 16:10:27 +00:00
531d468612
Redo rev.1.13. We must refer to CURBRK via GOT in case of PIC.
2008-11-06 13:42:55 +00:00
2f0d0e12bf
define and use LIBCDIR in terms of .CURDIR. fixes compat/lib/libc build.
2008-10-26 07:43:07 +00:00
cac8e44915
Add C99 functions imaxabs and imaxdiv.
2008-08-04 21:29:27 +00:00
01b36ae8ad
Use WEAK_ALIAS
2008-08-04 20:26:38 +00:00
e796a34c9c
Use WEAK_ALIAS.
...
Add imaxabs weak alias.
2008-08-04 20:21:26 +00:00
7781a976a7
Avoid calling a syscall when argument to sbrk(2) is zero.
...
OK by <ad>.
2008-07-02 20:07:43 +00:00
5d1469bd77
Backout previous: the license sweep touched these files in error, so
...
restore the old license.
2008-05-10 15:31:03 +00:00
f1a537e1d9
Fix nested comment and move to 2 clause license
2008-05-02 18:34:17 +00:00
ce099b4099
Remove clause 3 and 4 from TNF licenses
2008-04-28 20:22:51 +00:00
690e5f1c0b
add matching END(). found via -Wa,--fatal-warnings
2008-04-22 04:47:44 +00:00
6f4524c1a6
Fix a stupid typo in a comment. Pointed out by Juergen Hannken-Illjes.
2008-04-18 08:05:05 +00:00
1ad3697abd
Back out the ppc assembler changes for memcpy/memmove/memcmp on evbppc.
...
It breaks the explora because the cpu (403) cannot deal with unaligned
accesses. Reported by Juergen Hannken-Illjes.
2008-04-18 05:34:04 +00:00
c82cfbf571
Use SYSTRAP macro, rather than inlining the swi instruction.
...
This makes sure the SWI will be in the NetBSD assigned range of swi
numbers.
2008-03-30 21:22:05 +00:00
0db6a6b023
When compiling for _STANDALONE, omit a bunch of headers which aren't really
...
needed and which trigger build problems.
(These two files should probably be unified and put under common/...)
2008-03-18 20:11:43 +00:00
71ca0c37fa
As mrg@ informs me, delay slots are indented 1 and not 2 spaces.
2008-03-10 18:31:08 +00:00
bbed82985f
Fix the problem that CURBRK can't be referenced directly via R_SPARC_GOT13
...
here anymore (apparently libc grew too big). Use sethi and %hi/%lo instead.
2008-03-10 18:28:43 +00:00
f908c50890
Make use of ppc tuned assembler.
2008-02-21 17:37:00 +00:00
d638eaddc1
m68k/Makefile.inc:
...
- add __mmap.S to SRCS.
- add _lwp_getprivate.S mremap.S to ASM.
sys/Makefile.inc:
- look for __glue.S as well as glue.S.
- introduce a new ASM_MD and set it to "_lwp_getprivate.S mremap.S".
this is used for assembler files that may have non-default
implementations.
now m68k and m68000 mostly build again.
2008-02-09 02:41:06 +00:00
8f4340d1d4
Modify the high part of the fraction so that we recognize our __infinityl
...
value.
XXX - seems like we could share this with LE archs
2008-01-10 00:34:27 +00:00
d8de5b102b
Remove support for NetBSD/pc532.
2008-01-09 11:25:58 +00:00
e094ff6106
Add missing stubs for syscalls returning addresses.
...
We can now run on a kernel that does not copy the syscall return value
to register A0.
As discussed in PR port-m68k/37631.
2007-12-31 07:39:19 +00:00
24ced765a1
Add a comment explaining stack layout and register contents.
2007-12-22 14:10:03 +00:00
fd903b1729
Implement ptrace.
2007-12-02 15:48:47 +00:00
819071140e
Fix include files and change use LEAF_ENTRY_NOPROFILE function to fix
...
the unwind entry.
2007-12-02 14:04:19 +00:00
3ba44b945b
Reorder the code for i386 and amd64 syscall stubs so that the 'jc' on error
...
is a forwards jump instead of a backwards jump that a P4 will mispredect (and
probably immediatly get a cache-line miss due to the way the code gets
aligned).
2007-11-23 07:36:05 +00:00
4f3467d61e
Remove unused include.
...
Fix unwind info with S/ENTRY/LEAF_&/
2007-11-18 14:11:25 +00:00
8ed2bd6ec6
Generate correct unwind info for syscall stubs by using LEAF_ENTRY instead
...
of ENTRY(foo,0) so that gdb can backtrace through the stubs.
2007-11-18 13:25:39 +00:00
51a2159851
Don't unconditionally clear the direction flag. The ABI says it must always
...
be clear when making a function call, and 'cld' takes about 50 clock cyles
on the P4.
2007-11-12 18:41:59 +00:00
5265977a9e
When converting long double values to integer types, explicitly use the
...
"round to zero" variants of the softfloat conversion functions.
Add a variant to convert long double to unsigned long - the "to long"
variant checked for overflows that do not apply to unsigned results.
This fixes the regress/lib/libc/convfp tests for sparc64.
2007-11-08 15:50:19 +00:00
bd8185f2eb
syscall() has to be implemented in the kernel - so is a normal system call here.This has always been broken, so I suspect everything is using __syscall().
2007-11-05 20:30:07 +00:00
f1273d89a6
Set up the tail of the instruction address queue so that
...
we don't have to rely on luck for swapcontext(3) to work.
This fixes a problem pointed out to me by Chuck Silvers.
2007-10-28 09:38:06 +00:00
a34b3f871d
In handwritten asm use -fPIC code instead of -fpic, our libc has grown
...
large enough for GOT to be larger than 8k.
While here kill redundant PIC ifdefs in setjmp.S - sparc "call"
instruction is piccy by itself.
Tested by martin@
2007-10-08 13:06:00 +00:00
8ce1f4fff2
fix typos
2007-04-29 20:23:34 +00:00
65ec9d356d
Don't clobber callee save registers r12 & r13 in swapcontext. They
...
were saved in the stack, but the subsequent swapcontext would return
to the earlier stack frame and never restore them. Fixes puffs
with default compiler flags on amd64.
Thanks to Mihai Chelaru for providing access to an amd64 machine
for debugging this problem.
2007-04-12 12:33:26 +00:00
5764a76889
Remove support for NetBSD/{,evb}sh5.
2007-04-08 09:35:21 +00:00
3cac6dbb47
If I686_LIBC is defined, make syscalls using the 'sysenter' instruction.
2007-03-09 14:30:45 +00:00
a99fcd719d
Remove an obsolete comment.
2007-03-08 13:46:31 +00:00
d417d9f5d4
From Christos: gdtoa fixes for m68k. M68k ports should build now, but
...
printing extended precision is a little off.
2007-02-17 17:50:52 +00:00
12287894d6
ext_int bit is no more.
2007-02-02 23:18:57 +00:00
0449ecc079
since the int bit is merged, do the explicit math.
2007-02-02 23:14:09 +00:00
55ac93d329
Remove more duplicate #includes, and a few spurious whitespaces at EOL
...
From Slava Semushin <slava.semushin@gmail.com>
2007-01-17 23:24:22 +00:00
1c4ebce529
Provide a full frame stack space in the clone'd child.
2006-12-28 17:46:03 +00:00
55a9a999dd
Use CCFSZ instead of 96 for standard save operations.
2006-12-28 10:02:43 +00:00
f666b179b1
After discussing this further with uwe, rearange stack usage slightly and
...
replace a comment with words suggested by him.
2006-12-27 13:40:56 +00:00
520e5c8f6a
When setting up the __clone()'s userland stack, make sure to allocate
...
an initial frame. Fixes PR 33075.
2006-12-25 21:44:37 +00:00
2ed5a47025
Pull over rev. 1.5 of the i386 version (../../i386/resumecontext.S),
...
quoting Matthias Drochner:
> -check for a NULL uc_link before calling setcontext(2); do a real exit(3)
> instead with exit handlers and all that
> -If setcontext(2) returns (which it shouldn't), do an _exit(-1) rather
> than relying on a sensible return value to use as exit code. Makes
> it less likely that an ucontext corruption goes unnoticed.
This fixes regress/lib/libc/context for m68k.
Thanks to Steve Woodford for optimization hints.
2006-12-13 10:48:02 +00:00
e56cdf1dd0
Sync all resumecontext.c versions with the changes originally made
...
by Matthias Drochner to i386 and alpha:
-check for a NULL uc_link before calling setcontext(2); do a real exit(3)
instead with exit handlers and all that
-If setcontext(2) returns (which it shouldn't), do an _exit(-1) rather
than relying on a sensible return value to use as exit code. Makes
it less likely that an ucontext corruption goes unnoticed.
2006-12-08 16:09:02 +00:00
2cb7656bf1
Pull over rev. 1.5 of the i386 version (../../i386/resumecontext.S),
...
quoting Matthias Drochner:
> -check for a NULL uc_link before calling setcontext(2); do a real exit(3)
> instead with exit handlers and all that
> -If setcontext(2) returns (which it shouldn't), do an _exit(-1) rather
> than relying on a sensible return value to use as exit code. Makes
> it less likely that an ucontext corruption goes unnoticed.
This also fixes regress/lib/libc/context for sparc64.
2006-12-08 15:51:45 +00:00
58bd20f660
if setcontext(2) returns, _exit(2) with a return code of -1 instead
...
of relying on a sensible return value
2006-11-08 21:16:40 +00:00
ac9d7f2faf
-check for a NULL uc_link before calling setcontext(2); do a real exit(3)
...
instead with exit handlers and all that
-If setcontext(2) returns (which it shouldn't), do an _exit(-1) rather
than relying on a sensible return value to use as exit code. Makes
it less likely that an ucontext corruption goes unnoticed.
2006-11-08 20:28:03 +00:00
1ad9454efb
Towards ia64 libc. Round #2
2006-09-23 17:39:34 +00:00
64f11261d1
ia64 libc MD stuff #1 . Pushing in everything before my harddisk pops.
2006-09-22 17:59:47 +00:00
768b7d52a9
add ia64 libc MI bits.
2006-09-22 12:09:24 +00:00
503611ba29
libc stubs added
2006-09-10 21:22:33 +00:00
0f58c090bc
Rename these functions to the __<name>14 names.
...
XXX: still need to be converted to use sigprocmask().
2006-09-04 01:08:11 +00:00
e8ef331ffa
Fix PR 34112, where a program could mysteriously exit on return from a
...
signal handler, by deriving the context pointer from the stack pointer
instead of trying to read it from where the sigaction function may have
overwritten it with arbitrary bits.
Fix provided by Matthias Drochner in the PR.
2006-08-02 04:18:48 +00:00
60ce5dccc9
Add missing ppc64 elements
2006-07-07 21:07:35 +00:00
c41313666f
checkin libc asm stubs, #1
2006-07-07 07:49:19 +00:00
549728f08a
build __sigaction14_sigtramp.c
2006-07-07 06:56:34 +00:00
fefc484d05
rcsid & white space
2006-07-07 06:53:42 +00:00
db64036e6c
fix @toc reference macro
2006-07-07 06:50:06 +00:00
7108da6a37
Add missing sigaction()
2006-07-07 06:40:45 +00:00
7008de4c81
bzero() and memset() for ppc64.
2006-07-07 04:54:52 +00:00
4669390119
rewrite for ppc elf64 reloc
2006-07-06 17:05:04 +00:00
5844b0794f
we need the current vfork
2006-07-06 17:04:22 +00:00
3fbcafa1c9
convert remaining ppc32 code to ppc64
2006-07-05 18:28:43 +00:00
0584742537
convert PIC linkage
2006-07-05 18:27:08 +00:00
50500ff1b2
revise for ppc64
2006-07-05 18:12:54 +00:00
a9ea5c11ba
Adjust types and casts for LP64 compilation.
...
Split "long *sp;" into "__greg_t sp, *spp;", which reduces the cast
density and could be used to clean up the ppc32 version as well.
2006-07-05 18:10:36 +00:00
a87ca7ecec
fix typo
2006-07-05 18:07:22 +00:00
dfb9c8767b
Finish ppc64 conversion
2006-07-05 18:06:47 +00:00
9ed70a7fa2
adjust casts for LP64
2006-07-05 18:05:45 +00:00
3379c1c805
replaced by code in libc/compat/gen
2006-07-03 13:20:19 +00:00
46b4d67318
build {frexp,ldexp,modf} in the "compat" subtree
...
(Drop the modf assembler version. We have the same code in libm.)
(Drop the ldexp inline-assembler version. The same code is in libm as
scalbn; the ldexp there is just a wrapper providing error handling.)
2006-07-03 13:13:53 +00:00
5ea27734ec
build frexp_ieee754 and ldexp_ieee754 in the "compat" subtree
...
(There is an assembler version of modf which needs to be checked.)
2006-07-03 13:06:12 +00:00
2a270967fb
This is a new port - don't ever start with frexp/ldexp/modf in libc, we
...
have it in libm already.
2006-07-03 13:03:49 +00:00
451c99c7a2
build frexp_ieee754 and ldexp_ieee754 in the "compat" subtree
...
(There is an assembler version of modf which needs to be checked.)
2006-07-03 13:00:54 +00:00
4cbc875060
build frexp_ieee754 in the "compat" subtree
...
(There are assembler versions of ldexp and modf which need to be checked.
I'm in doubt they are metter than those in libm.)
2006-07-03 12:58:55 +00:00
c48c54fd05
build frexp_ieee754 in the "compat" subtree
...
(The Makefile logics for ldexp and modf is too involved for me to deal
with this now.)
2006-07-03 12:54:34 +00:00
d48f14661d
ppc64 build framework first pass
2006-07-01 16:37:20 +00:00
fa073991b3
build frexp/ldexp/modf in the "compat" part (code is identical)
...
(These are the easy cases where no assembler implementations exist.)
2006-06-28 14:46:32 +00:00
0ed7304bd6
build frexp/ldexp/modf in the "compat" part
...
(code is identical)
2006-06-28 14:22:15 +00:00
4502979a72
Build just the compatibility versions of frexp()/ldexp()/modf(), the
...
real implementation is in libm.
2006-06-27 18:19:57 +00:00
46daac62de
G/c KMSRCS and KMINCLUDES now that we have common/lib/libc
2006-06-17 18:04:23 +00:00
e69cf86f15
remove GCC2 support
2006-06-02 19:47:15 +00:00
15fc3900ea
add a local prototype for fabs()
2006-05-21 02:51:15 +00:00
96473fa808
change __weak_extern to __weak_reference so that gcc4 works.
2006-05-18 17:54:19 +00:00
8726b401d3
if SOFTFLOAT_FOR_GCC is not defined, don't declare float_detect_tininess.
2006-05-16 20:55:51 +00:00
640873a430
make sure there is a prototype in scope.
2006-05-16 05:29:16 +00:00
340a97942e
if SOFTFLOAT_FOR_GCC is not defined, don't declare float_detect_tininess.
2006-05-11 23:53:14 +00:00
fd07545586
in makecontext(), sp is unsigned * so don't assign signed *'s to it.
2006-05-11 00:34:44 +00:00
a640fe8c43
It's "its."
2006-04-24 19:00:29 +00:00
13ab2e2a10
Now that bcopy.S doesn't contain the actual implemenation, move it
...
back to libc.
2006-04-23 00:08:18 +00:00
0db1500b43
retire HAVE_GCC3/HAVE_GCC4 and introduce HAVE_GCC that is set to 2, 3 or 4.
2006-04-07 19:38:58 +00:00
6f5914a550
__sigtramp1.S moved to the new compat layout so we should try and build
...
it here.
2006-04-05 12:23:51 +00:00
f81322cf18
build erand48_ieee754.c on everything but vax
2006-03-22 22:03:58 +00:00
dcaa7a3be5
It's .endif, not .fi.
2006-03-21 22:50:13 +00:00
4b173cfc48
Don't build extended-precision routines on m68000; from Havard.
2006-03-20 09:52:23 +00:00
bc89c06cbf
Add strtof(3) and strtold(3); welcome libc 12.137.
2006-03-15 17:35:17 +00:00
7c5c37b133
move to the new compat layout.
2006-03-11 19:56:17 +00:00
ab2a637b34
Welcome to the new compat layout.
2006-03-11 19:34:40 +00:00
2b52112d5a
Remove files moved to the new compat layout.
2006-03-11 19:16:07 +00:00
edc071d5b4
remove files moved to the new compat layout.
2006-03-11 19:02:16 +00:00
22d2b3199c
Welcome to the new compat layout.
2006-03-09 16:20:27 +00:00
5d1e8b2745
Fix some typos.
2006-02-25 02:28:55 +00:00
1ad8067cb3
Fix typos, reported by Alexey Dobriyan ("Gathered from Linux"),
...
forwarded by jmc@openbsd.
2006-02-25 00:58:34 +00:00
516a950568
Remove unused local variable (re-declared in inner conditional scope).
2006-02-22 08:49:45 +00:00
b1732b7a63
make these consistent.
2006-02-19 05:06:02 +00:00
8b121c704a
Seperate out the compat support for VAX. Also, make references to the
...
sigcontext sigtramp code weak so compat can be compiled out.
2006-02-18 22:54:51 +00:00
b09fbb025f
*setjmp14 are not compat, plain setjmp/sigsetjmp are the compat versions.
...
Exchange the files.
2006-02-18 00:45:22 +00:00
bfb62ce415
Move PowerPC compat files from arch/powerpc to compat/arch/powerpc.
2006-02-18 00:21:27 +00:00
804e6c96a2
Add machine-dependent definitions for gdtoa.
2006-01-25 15:33:28 +00:00
2e1a547276
Cosmetics (pointed out by uwe): use trivial restore, not the explicit
...
version.
2006-01-08 20:03:19 +00:00
286956e76d
Use __errno in the _REENTRANT case (pointed out by Valeriy E. Ushakov).
...
While there, indent some delay slot instructions.
2006-01-08 12:05:33 +00:00
5cee9283d0
Provide _REENTRANT version of errno = 0;
2006-01-06 20:38:01 +00:00
0e79e13e9c
Use JUMP_CERROR instead of messing with errno directly.
...
Simplify.
2006-01-06 07:07:34 +00:00
3a310524fb
Drop unnecessary nops.
2006-01-06 06:19:20 +00:00
ec3c02b8ff
Simplify.
2006-01-06 05:23:46 +00:00
6ea6e06ed1
Change !_REENTRANT && !PIC case to use r4 too.
2006-01-06 05:14:39 +00:00
026eb7b69e
Open code _SYSCALL_NOERROR to shave off a memory load and a branch.
...
Makes this congruent to vforks. Add .size directive.
2006-01-06 05:11:29 +00:00
73375bbc4d
Add .size directive.
2006-01-06 04:38:10 +00:00
b46734f0ca
Change cerror to take errno in r4, instead of r0.
...
In PIC code fetch cerror address from GOT to avoid
unresolved text relocations in shared libraries.
Hide the code to jump to cerror inside JUMP_CERROR macro.
Change syscall code to use JUMP_CERROR.
Die text relocations! DIE!
2006-01-06 03:58:31 +00:00
b1648afca2
This can be simplified to RSYSCALL(__syscall), no special treatment necessary.
2006-01-06 01:57:38 +00:00
63e9eca689
This can be simplified to PSEUDO(exect,execve).
2006-01-06 01:53:30 +00:00
63f84ca820
This can be simplified to RSYSCALL(syscall), no special treatment necessary.
2006-01-06 01:27:23 +00:00
he