Commit Graph

246 Commits

Author SHA1 Message Date
Konstantin Belousov
adc6846785 Remove duplicate .note.GNU-stack section declaration. bcopy already
made the neccessary provisions.

Reported by:	arundel
2011-02-04 21:04:00 +00:00
Konstantin Belousov
93ab758670 Add section .note.GNU-stack for assembly files used by 386 and amd64. 2011-01-07 16:08:40 +00:00
Jilles Tjoelker
220856f376 libc: Remove the i386 assembler version of strlen(3).
On anything modern, the C version, which processes a word at a time, is much
faster. The Intel optimization manual explicitly warns against using REP
prefixes with SCAS or CMPS, which is exactly what the assembler version
does.

A simple test on a Phenom II showed the C version, compiled with -O2, to be
about twice as fast determining the length of 100000 strings between 0 and
255 bytes long.

MFC after:	2 weeks
2010-10-01 13:10:11 +00:00
Dimitry Andric
3c1594be0a Retire the amd64 and i386 specific inline assembly versions of ldexp.c,
as they are slower than the generic version in C, at least on modern
hardware.  This leaves us with just five implementations.

Suggested by:	bde
Approved by:	rpaulo (mentor)
2010-09-30 19:20:21 +00:00
Rui Paulo
7e78863707 Workaround LLVM bug #4434:
Reorder inline assembly arguments temp2, temp, value and texp to follow
the st(0), st(1), etc. style.
Also mark the temp2 variable as volatile to workaround another clang
bug.
This allows clang to buildworld FreeBSD/i386.

Submitted by:	dim
2010-09-21 20:23:19 +00:00
Ed Schouten
e71280c7bc Remove hand-written labs/abs implementations. GCC is smart enough.
It turns out GCC generates code that's a couple of bytes big bigger, but
performs no branching whatsoever.

Submitted by:	Christoph Mallon
2009-06-23 09:04:59 +00:00
Ed Schouten
f863f9cbd9 Remove unneeded stores back into the function arguments.
Submitted by:	Christoph Mallon
2009-06-22 10:56:08 +00:00
Konstantin Belousov
8b61c10e48 Unconditionally build wrappers for i386_get_ioperm(), i386_set_ioperm(),
i386_get_ldt() and i386_set_ldt().

In collaboration with:	pho
Reviewed by:	jhb
2009-04-01 19:23:46 +00:00
Peter Wemm
9b16c7add1 On i386, the primary function that SYSCALL() generates is with the
__sys_ prefix.  Make END() match.  This didn't cause a compile error, but
the function size is attached to the .weak symbol, not the real one.
2008-11-15 22:23:07 +00:00
Ed Schouten
a1b5a8955e Mark uname(), getdomainname() and setdomainname() with COMPAT_FREEBSD4.
Looking at our source code history, it seems the uname(),
getdomainname() and setdomainname() system calls got deprecated
somewhere after FreeBSD 1.1, but they have never been phased out
properly. Because we don't have a COMPAT_FREEBSD1, just use
COMPAT_FREEBSD4.

Also fix the Linuxolator to build without the setdomainname() routine by
just making it call userland_sysctl on kern.domainname. Also replace the
setdomainname()'s implementation to use this approach, because we're
duplicating code with sysctl_domainname().

I wasn't able to keep these three routines working in our
COMPAT_FREEBSD32, because that would require yet another keyword for
syscalls.master (COMPAT4+NOPROTO). Because this routine is probably
unused already, this won't be a problem in practice. If it turns out to
be a problem, we'll just restore this functionality.

Reviewed by:	rdivacky, kib
2008-11-09 10:45:13 +00:00
Peter Wemm
ed820052d0 Add missing END() macros, as per rev 184547 for amd64. The lack of these
is mostly harmless, but it does upset some of valgrind's functionality.
2008-11-02 01:28:47 +00:00
David Schultz
4110421449 We should also save and restore the MXCSR as on amd64, but detecting
whether the CPU supports SSE or not here is rather odious.
2008-06-28 17:58:06 +00:00
Bruce Evans
d2012f3333 Add an alternative view of the bits in an 80-bit long double (64+16
instead of 32+32+15+1) on all arches that have such long doubles (amd64,
ia64 and i386).  Large objects should be be accessed in large units,
and the 32+32+15+1[+padding] decomposition asks for almost the opposite
of that, sometimes resulting in very slow accesses depending on how
well the compiler ignores what we ask for and converts to the best
units for the given machine.  E.g., on Athlons, there is a 10-20 cycle
penalty for accessing the middle 32-bit word immediately after an
80-bit store.

Whether actually using the alternative view is better is very machine-
dependent.  A 32+32+16 view is probably best with old 32-bit systems
and gcc through 4.2.1.  The compiler should mostly avoid the view and
generate best accesses, but gcc-4.2.1 is far from doing that.  I think
64+16 is best for now.  Similarly for doubles -- they should be using
64+0 especially on 64-bit machines, but fdlibm uses 32+32 extensively
for them.  Fortunately, in 64-bit mode for doubles, gcc already ignores
the 32+32-bit view and generates best accesses in many cases.
2008-01-17 16:39:07 +00:00
David Schultz
7cd4a83267 Since nan() is supposed to work the same as strtod("nan(...)", NULL),
my original implementation made both use the same code. Unfortunately,
this meant libm depended on a vendor header at compile time and previously-
unexposed vendor bits in libc at runtime.

Hence, I just wrote my own version of the relevant vendor routine. As it
turns out, mine has a factor of 8 fewer of lines of code, and is a bit more
readable anyway. The strtod() and *scanf() routines still use vendor code.

Reviewed by:	bde
2007-12-18 23:46:32 +00:00
David Schultz
39e7abef0e Export gdtoa's __ULto{x,Q}_D2A routine in a private namespace so
libm can use it.
2007-12-16 21:15:57 +00:00
David Schultz
199cdab56f Arrange so that the NaN returned by strtod("nan", NULL) is the same as
the NaN returned by strtod("nan()", NULL).
2007-12-16 21:15:09 +00:00
David Schultz
9c90f85a6b In scanf, round according to the current rounding mode. 2007-12-03 07:17:33 +00:00
Peter Wemm
65a6d893ba Classify mmap, lseek, pread, pwrite, truncate, ftruncate as pseudo
syscalls, unless WITHOUT_SYSCALL_COMPAT is defined.  The default case
will have the .c wrappers still.  If you define WITHOUT_SYSCALL_COMPAT,
the .c wrappers will go away and libc will make direct syscalls.

After 7-stable starts, the direct syscall method will be default.

Approved by:  re (kensmith)
2007-07-04 23:23:01 +00:00
Peter Wemm
eabc04d472 Adjust the syscall stub macros to be consistent in their meaning. In
particular:
SYSCALL() makes a syscall, with errno handling, and continues execution
directly after the macro in the non-error case.
RSYSCALL() is just like SYSCALL(), but returns after success.
Both SYSCALL(name) and RSYSCALL(name) export  "__sys_name" as a strong
symbol, with "_name" and "name" as weak aliases.
PSEUDO() is just like RSYSCALL(), but skipping the "name" weak alias.  It
still does "__sys_name" and "_name".

Change i386 to add errno handling to PSEUDO.  The same for amd64 and
sparc64, with appear to have copied the behavior.
ia64 was correct (as was alpha).  Just remove some apparently unused
variants of the macros. (untested!)
I believe powerpc is correct.
Fix arm to not export "name" from the PSEUDO case.  Remove apparently
extra unused variants.  (untested!)

The errno problem manifested on i386/amd64/sparc64 by having "PSEUDO"
classified syscalls return without setting errno.  eg: "addr = mmap()"
could return with "addr" = 22 instead of setting errno to 22 and
returning -1.

Approved by: re (kensmith)
2007-07-04 23:18:38 +00:00
Daniel Eischen
2665faf497 Some libc symbol map cleanups.
net: endhostdnsent is named _endhostdnsent and is
  private to netdb family of functions.

  posix1e: acl_size.c has been never compiled in,
  so there's no "acl_size".

  rpc: "getnetid" is a static function.

  stdtime: "gtime" is #ifdef'ed out in the source.

  some symbols are specific only to some architectures,
  e.g., ___tls_get_addr is only defined on i386.

  __htonl, __htons, __ntohl and __ntohs are no longer
  functions, they are now (internal) defines in
  <machine/endian.h>.

Submitted by:	ru
2007-05-31 13:01:34 +00:00
Daniel Eischen
0c58e013aa Remove .mcount from gmon's Symbol map and add it to the appropriate
arch.  It can be named differently depending on the arch (.mcount,
_mcount).

Submitted by:	marius
2007-05-13 14:16:55 +00:00
Daniel Eischen
5f864214bb Use C comments since we now preprocess these files with CPP. 2007-04-29 14:05:22 +00:00
Alexander Kabaev
445eba04a8 Clean-ip TLS symbol versions. [_]__tls_get_addr function is part of
the platform ABI and as such does not belong in FBSDprivate.

__libc_tls_* functions do not have to be visible to outside world
at all.
2007-04-09 22:48:08 +00:00
Warner Losh
c879ae3536 Per Regents of the University of Calfornia letter, remove advertising
clause.

# If I've done so improperly on a file, please let me know.
2007-01-09 00:28:16 +00:00
David Schultz
58d6b4605f Handle inf/nan correctly. 2007-01-06 21:50:04 +00:00
David Schultz
266cb5ad57 The distinction between quiet and signaling NaN formats is
machine-dependent; these files tell the latest version of gdtoa
what to do.
2007-01-03 05:00:03 +00:00
Olivier Houchard
05b432d2d1 Instead of re-implementing hton[ls] and friends for each arch, add a new MI
file, net/ntoh.c, which just implement them using the inline functions from
<sys/endian.h>.

Suggested by:	bde
2006-11-06 22:07:47 +00:00
Ruslan Ermilov
08e9f8c84f Forgot to bump the document date. 2006-10-13 20:34:44 +00:00
Ruslan Ermilov
531ba7c084 Some word smithing adopted from the PR.
PR:		docs/54451
Originally by:	Andreas Fuchs
Adopted by:	jhb
MFC after:	3 days
2006-10-13 20:28:58 +00:00
Ruslan Ermilov
a73a3ab56b Markup fixes. 2006-09-17 21:27:35 +00:00
Ruslan Ermilov
3c03c7095e GC dead code. If we want to stay polite to the foreign compilers,
we can find another way to issue an #error, but using a preprocessed
assembler for that purpose and clobbering libc.a with an empty .o
just for the sake of #error reporting is way too much of a burden.
2006-09-05 16:21:26 +00:00
Konstantin Belousov
3d5fa0356e Replace absolute addressing in the call instructions with position-independend
calls. This eliminates TEXTREL from libc, making its text segment relocatable.

PR:	i386/85242
Approved by:	kan (mentor)
MFC after:	1 month
2006-06-05 14:59:33 +00:00
Brian S. Dean
de76bc3609 A few minor wording changes and a clarification about error conditions. 2006-05-03 17:33:29 +00:00
Daniel Eischen
6fad3aaf15 Add each directory's symbol map file to SYM_MAPS. 2006-03-13 01:15:01 +00:00
Daniel Eischen
cce72e8860 Add symbol maps and initial symbol version definitions to libc.
Reviewed by:	davidxu
2006-03-13 00:53:21 +00:00
Stefan Farfeleder
613100918d Include a couple of headers to ensure consistency between the prototype and
the function definition.
2005-09-12 19:52:42 +00:00
Peter Wemm
3b4399f6a7 Clean out the leftovers from the i386_set_gsbase() TLS conversion.
Like on libthr, there is an i386_set_gsbase() stub implementation here
to avoid libc.so.5 issues.  This should likely be a weak symbol and I
expect this will be fixed soon.

Approved by:	re
2005-06-29 23:15:36 +00:00
Peter Wemm
3bbf3512ad Provide stub functions for i386_set_ldt() and i386_get_ldt() even when
compiling as an amd64 support binary.  They will return EINVAL on an amd64
kernel, but this simplifies other #ifdefs that were getting a bit nasty.
2005-04-26 20:51:15 +00:00
Peter Wemm
8fa4081fe3 No longer use _amd64_set_gsbase(). Use i386_set_gsbase() even when
building for amd64.
2005-04-26 20:46:11 +00:00
Alan Cox
c0862430d5 Eliminate an unpredictable branch from bcmp().
Reviewed by: bde
2005-04-21 23:07:20 +00:00
Peter Wemm
4453c6dc67 Attempt i386_set_gsbase() before using the user_ldt code. Unimplemented
sysarch() calls return EINVAL, not SIGSYS.. so we can trivially adapt.
2005-04-14 00:02:37 +00:00
Peter Wemm
53193c146a Add stubs for the %fs/%gs base management calls. 2005-04-14 00:01:35 +00:00
Peter Wemm
8d6f45f534 Fix strict-alias warnings by removing excessive (and wrong) casts. 2005-04-07 04:33:15 +00:00
David Schultz
1dfab5edec Define LDBL_NBIT to be a mask indicating the position of the integer
bit in a long double.  For architectures that don't have such a bit,
LDBL_NBIT is 0.  This makes it possible to say `mantissa & ~LDBL_NBIT'
in places that previously used an #ifdef to select the right expression.
The optimizer should dispense with the extra arithmetic when LDBL_NBIT
is 0.
2005-03-07 04:55:22 +00:00
David Schultz
f154b03b25 Update my email address. 2005-02-06 03:23:31 +00:00
David Schultz
bd15659f64 Eliminate gdtoa.mk and move its contents to ${MACHINE_ARCH}/Makefile.inc.
The purpose of having a separate file involved an abandoned scheme that
would have kept contrib/gdtoa out of the include path for the rest of libc.
2005-01-15 05:23:58 +00:00
Daniel Eischen
6c68cb596e Make sure the first argument to the user function is 16-byte aligned.
Submitted by:	tegge
2004-12-05 21:22:08 +00:00
John Baldwin
d0565c1493 Remove 80386 support from libc. 2004-11-16 21:01:31 +00:00
Peter Wemm
5bc7bd5ff2 When building the custom i386 libraries for use on amd64 kernels, we
can't use the i386_set_ldt() family of routines, because they are not
implemented.  Instead, use the recently exposed direct access sysarch
routines for setting what %fs and %gs point to.

Use this for the i386 TLS _set_tp() routine, but only when compiling to
run as a 32 bit support binary for amd64 kernels.
2004-11-06 03:28:26 +00:00
David E. O'Brien
62f696fcfa Sync with sys/i386/include/endian.h: use the single instruction 'bswap'. 2004-10-18 17:19:36 +00:00