summaryrefslogtreecommitdiffstats
path: root/lib/libc/arch (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* OR r13, not r9 to match SYS.h; fixes syscall(2) on mvme88k; OK miod@millert2003-12-081-3/+3
|
* Only a clown like me could put cerror in the data section by mistake...miod2003-10-161-4/+3
|
* These files are not necessary anymore.miod2003-09-012-74/+0
|
* In setjmp() and sigsetjmp(), be sure to return with r14 and r15 unchanged, asmiod2003-08-062-4/+11
| | | | they are part of the call preserved register set, and gcc -O relies upon this.
* Fix the *longjmp() behaviour - it is legal to reuse a jmp_buf several times.miod2003-08-013-36/+15
| | | | Gets us a working perl 5.8.
* warn about unsafe APIs at link time. Conditional on libc/Makefile definingderaadt2003-07-244-4/+33
| | | | | APIWARN; disabled by default. In use by many developers for quite some time, now they have a common knob to enable/disable this
* enable the libc portions of the quad float stuff (nothing will call thisjason2003-07-211-1/+3
| | | | until a forthcoming gcc patch is in place).
* implementation of __ftoul (float -> unsigned long) (based on __dtoul above it)jason2003-07-211-6/+53
|
* quad float emulation stuff. Mostly based on freebsd (which is based onjason2003-07-2117-0/+3898
| | | | NetBSD's kernel emulation stuff). Not enabled yet.
* proto repairsderaadt2003-07-151-1/+3
|
* Use a logical comparison (unsigned) for pointers. ok drahn@tdeval2003-07-101-2/+2
|
* pull in headersderaadt2003-06-253-6/+9
|
* Sync with share/misc/license.template and add missing DARPA creditmillert2003-06-171-8/+8
| | | | where applicable.
* oopsderaadt2003-06-114-24/+4
|
* ansification; checked by pvalderaadt2003-06-114-4/+24
|
* terms 3 & 4 cleanup based on "terms" filederaadt2003-06-031-6/+1
|
* Use an ISC-tyle license for all my code; it is simpler and more permissive.millert2003-06-031-22/+11
|
* three four kill ...mickey2003-06-0212-83/+23
|
* Remove the advertising clause in the UCB license which Berkeleymillert2003-06-02283-1747/+615
| | | | rescinded 22 July 1999. Proofed by myself and Theo.
* assembler versions of insque and remque for m68k, also from 4.4BSDmillert2003-05-092-0/+116
|
* insque and remque are VAX instructions, no need for C functions; From 4.4BSDmillert2003-05-092-0/+98
|
* Add Doug Gwyn's portable alloca() for platforms w/o an assemblermillert2003-05-027-8/+14
| | | | | | | version. Normally, gcc provides its own alloca() (unless -fno-builtin is specified). The C version is decidedly suboptimal--we really need proper assembler versions but this will work as a stop-gap measure.
* Indent delay slot insnsjason2003-04-228-43/+43
|
* ELF fixups, ELF PIC syscall support, asm changes as requested by assembler.drahn2003-04-1712-35/+84
|
* duplicate words and spelling fixes in commentsdavid2003-03-103-8/+8
| | | | ok miod@
* Provide real ieee fp*() functions; from NetBSD.miod2003-03-015-48/+56
| | | | ok millert@
* indent delay slotjason2003-02-181-2/+2
|
* Obey calling convention and use the right opcode (fabss -> fabsd); Fabulous...jason2003-02-081-4/+2
| | | | (based on netbsd pr 20140)
* Align this char array accessed as double, for paranoia reasons.drahn2003-01-291-3/+4
|
* Waiter! I asked for a working vfork!miod2003-01-161-3/+3
|
* Alignment requirements are satisfied in ENTRY(), no need to add more here.miod2003-01-161-2/+1
|
* minbrk is exposed, gmon uses itmickey2003-01-151-6/+6
|
* Get static rcsid correctly, better style for .S files, trim whitespace.miod2003-01-0730-156/+167
|
* Properly align __infinity - m88k is very strict wrt double alignment.miod2003-01-071-6/+7
| | | | Fixes - oh what a surprise - bus errors in awk.
* Get the rounding modes table correct, eventually.miod2003-01-071-4/+10
|
* One more pass to get thinks correct wrt weak symbols and syscalls.miod2003-01-071-35/+27
|
* Fix *longjmp() return values.miod2003-01-043-59/+57
|
* Adapt to recent changes in libc/libc_r wrt weak symbols, so that programsmiod2003-01-026-50/+66
| | | | | | | can link against libc without unresolved symbols again. Anyone who correctly guessed that my m88k hard drive has been recovered, wins a strawberry lollipop.
* Rename cerror to __cerror and curbrk to __curbrk, to avoid namespacemiod2003-01-029-41/+42
| | | | pollution. Conforming to art@'s evil plans.
* Begone.miod2003-01-023-94/+0
|
* Typosmiod2003-01-023-13/+13
|
* Make the mul/div/rem (and u... versions) weak so that ld.so can overridedrahn2002-11-233-3/+18
| | | | | it with optimized versions. And allow ld.so to compile with strong versions with extra symbols so it can find them.
* Oops, unbrek libc_pmiod2002-11-181-2/+2
|
* Move the weak definition from before the syscall's entry point tomillert2002-11-061-13/+11
| | | | | after. This avoids a bunch of reloc warnings by working around a deficiency in the a.out loader. miod@ OK
* From NetBSD; Use _C_LABEL, not SYSNAME to call sigaltstack, sigblock,millert2002-11-061-5/+5
| | | | | | sigreturn. sigblock has not been a syscall since 4.3BSD and using SYSCALL directly seems like a bad idea in general. Fixes a build problem on vax. OK miod@
* thread safe libc -- 2nd try. OK miod@, millert@marc2002-11-052-37/+38
| | | | Thanks to miod@ for m68k and vax fixes
* Rename cerror to __cerror and curbrk to __curbrk, to avoid namespacemiod2002-11-058-25/+25
| | | | pollution. Conforming to art@'s evil plans.
* Rename cerror to __cerror and curbrk to __curbrk, to avoid namespacemiod2002-11-059-28/+28
| | | | pollution. Conforming to art@'s evil plans.
* back out previous patch.. there are still some vax/m68k issuesmarc2002-11-032-2/+27
|
* libc changes for thread safety. Tested on:marc2002-11-032-27/+2
| | | | | | | alpha (millert@), i386 (marc@), m68k (millert@ and miod@), powerpc (drahn@ and dhartmei@), sparc (millert@ and marc@), sparc64 (marc@), and vax (millert@ and miod@). Thanks to millert@, miod@, and mickey@ for fixes along the way.
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.