summaryrefslogtreecommitdiffstats
path: root/sys/lib/libkern/arch (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* Replace a few more 32-bit instructions with the proper 64-bit equivalent.kettenis2011-11-271-22/+21
| | | | | Also fix the return value of memcpy. With these changes, this seems to work as advertised now.
* Regen.kettenis2011-09-192-4/+4
|
* Avoid sign-extension when extracting the low bits of addresses when checkingkettenis2011-09-191-3/+3
| | | | | | if the addresses are 4-byte aligned. ok jsing@
* we need to alias __movstr* as __movmem* for gcc4jsg2011-07-191-2/+2
| | | | from NetBSD. ok miod@ drahn@
* Regen.jsing2011-04-142-9/+9
|
* Use 64-bit operations for address manipulation.jsing2011-04-141-4/+4
|
* Correctly load p_addr and avoid trashing the source address.jsing2011-04-141-4/+4
|
* syncderaadt2011-03-124-32/+32
|
* In the original sparc V7 book (and in the v8 book), the divrem leaf code usedderaadt2011-03-121-2/+2
| | | | | | | | | | | | | | | | | | | | | | | local registers for a few temporaries. This was changed to use two global registers. Maybe to permit use in-kernel without conflicting with the register V7 register window handlers. (Was this done by Chris Torek? Is this related to Gordon Irlam's work? Or was it in NetBSD? Hard to tell because NetBSD removed their original cvs tree.) In V8 the ABI was tightened; more global registers became offlimits in different ways. We started supporting sun4m, and did not consider this. As a result, the global registers chosen are the wrong choice. In particular, %g7 is a poor choice for upcoming TLS work. It looks like it is safer to use %g5 and %g6 since these functions are "system software". All re-entrant parts of the system save it. On sparc64 these functions are in libc per ABI requirement, but are unused. On sparc, they occur in bootblocks (no reentrancy), kernel (reentrancy saves globals; kernel is not ABI compliant), userland libc (signal handlers save globals), and ld.so (symbol binding is not re-entrant on its own). Discussed rather extensively with guenther, kettenis, miod and drahn.
* s/DST/DEST/ to avoid warnings when building RAMDISK kernels, which definemiod2011-01-271-60/+60
| | | | TIMEZONE and DST...
* destintation -> destinationmiod2010-04-201-3/+3
|
* Allow for multiple CPUs by moving to an array of cpu_info.jsing2010-04-011-6/+2
| | | | ok kettenis@
* Provide a correct version for little endian kernels... sighmiod2009-12-122-3/+13
|
* <machine/macros.h> would provide inline version of a few of the functionsmiod2009-08-198-5/+459
| | | | | | | | | | | | traditionnaly found in libkern. However, the memcmp() flavour would behave as bcmp() with only two possible return values: zero and positive non-zero. This broke the name cache RB trees which now rely upon proper memcmp() semantics(negative value, zero, or positive value). Just give up on these macros and provide the same code as libc, in libkern. As a side effect, this no longer uses the cmpc3 instruction, which is not implemented and requires (slow) kernel emulation, on the original uVax.
* Use the MI random() for all cases; no need for a MD version since itderaadt2008-10-101-1/+0
| | | | | has never been performance sensitive. Running on all platforms, discussed with millert and kettenis, ok toby
* First pass at removing clauses 3 and 4 from NetBSD licenses.ray2008-06-268-61/+5
| | | | | | | | | Not sure what's more surprising: how long it took for NetBSD to catch up to the rest of the BSDs (including UCB), or the amount of code that NetBSD has claimed for itself without attributing to the actual authors. OK deraadt@
* These files are not used (and some of them would not compile anyway).miod2008-05-143-300/+0
|
* Catch up with dismantled libkern.kettenis2008-04-206-0/+336
|
* Remove the random() function from locore.s (which used sparcv7 instructions,kettenis2008-04-201-1/+0
| | | | | including mulscc to do multiplications) and switch to the generic random.c code.
* place a comment as to why these are otherwise emptyderaadt2007-11-2843-2/+43
|
* mips64 also needs memcpy.S, otherwise memcpy.c gets built, which conflictsjsing2007-11-271-0/+0
| | | | | | with bcopy.S ok miod@
* mips64 needs at least these to build; spotted by jsingderaadt2007-11-262-0/+0
|
* libkern, begone. Move to a new mechanism where config(8)'s "file"deraadt2007-11-25104-1392/+3519
| | | | | | | | | | | | directive can select between MI and MD versions of these files. At the same time, adjust the boot programs to pick exactly what they need, instead of the 7 or 8 mechanisms previously used. There will be some fallout from this, but testing it all by myself is a ridiculously slow process; it will be finished in-tree. Various developers were very nice and avoided making fun of me when I was gibbering in the corner..
* SCCS junk, bit by bitderaadt2007-11-247-56/+7
|
* junk RCSderaadt2007-11-241-8/+1
|
* more junk SCCSderaadt2007-11-243-12/+0
|
* junk SCCSderaadt2007-11-2411-75/+11
|
* delete unused junk RCS idsderaadt2007-11-249-36/+0
|
* Uncomment rule to build bcopy.S, and use that as our bcopy(9) implementation.kettenis2007-10-131-5/+5
|
* Make this actually work by using the right register numbers. In the conversionkettenis2007-10-131-12/+12
| | | | | from hppa the fact that t1-t4 actually number down from r22-r19 got somehow lost.
* "boundries" -> "boundaries" in various comments. Started by Diego Casati.krw2007-05-254-6/+6
|
* move hppa to __HAVE_CPUINFOmartin2007-05-141-1/+5
| | | | input from miod@, ok kettenis@
* more locc() diesderaadt2007-05-045-112/+6
|
* remove strcpy and strcat from the kernel; they are dead and unused code.reyk2007-05-047-357/+2
| | | | | | | (OpenBSD does not use strcat/strcpy in the kernel, if people do it in external modules they should update their code) ok deraadt@
* Move m88k memory copy and fill functions to libkern. The copy functions willmiod2006-11-177-1/+576
| | | | | | no longer share the same code, but will be instead be duplicated from the same source, so that bcopy() and memcpy() do not need to check for overlap, and memcpy() and memmove() are shorter.
* typoderaadt2006-11-091-1/+1
|
* import supah pieces from netbsdmickey2006-10-067-0/+997
|
* Switch to the C version of random() on m68k platforms. It is almost as fastmiod2006-01-041-2/+2
| | | | | | | | | | as the assembly version on 0[234]0, but *fifteen* times faster than it on 68060 systems, since it uses a form of muls.l which has to be emulated. And since we use random() for statclock variance, this means we were gratuitously doing an average of 100 emulation traps per second. ok deraadt@ millert@
* I'm Positive these are Type-O.miod2005-08-012-4/+4
|
* not neededderaadt2005-07-201-2/+2
|
* add moddi3, needed by new libzderaadt2005-07-201-5/+2
|
* hppa64 bitsmickey2005-04-013-0/+1740
|
* 64 bit safe over entire address rangepefo2004-10-085-69/+69
|
* Keep providing a 32bit only version of bzero() #ifdef _STANDALONE, asmiod2004-09-211-3/+20
| | | | bootblocks are compiled as 32bit code.
* slightly 64 bit improvedpefo2004-09-181-8/+8
|
* Changes for 64 bit kernel. Some functions needis to be rewritten orpefo2004-09-098-16/+16
| | | | optimized to take advantage of full 64 bit width registers.
* add missing copyrightpefo2004-08-138-9/+257
|
* mips64 nowpefo2004-08-1111-0/+388
|
* move to mips64pefo2004-08-1111-388/+0
|
* It turns out that the so-called fast ffs(3) routines were wrong, as soon asmiod2004-04-012-18/+2
| | | | | | | | | | | | more than one bit is set, as the ff1 instruction counts from the highest order. However, gcc/m88k with optimization enabled would use a correct, short sequence based upon ff1 and tweaks, to achieve the intended result, hence i did not catch this flaw initially. So revert to the C implementation - it is correct, still decently fast, and will only be used when compiling at -O0.
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.