summaryrefslogtreecommitdiffstats
path: root/sys/arch/sparc64/fpu (follow)
Commit message (Collapse)AuthorAgeFilesLines
* arithemtic -> arithmeticmiod2011-09-171-2/+2
|
* Revert art@'s moving around of the KERNEL_LOCK()/KERNEL_UNLOCK() calls,guenther2011-07-111-1/+9
| | | | | | as it causes hangs in some ports, including libsigsegv's configure script confirmed by krw@, landry@
* There is a bunch of places in the kernel entry points where we don'tart2011-07-071-9/+1
| | | | | | | | | | | hold the kernel lock, but still need call one function that needs it. Instead of grabbing the lock all over the place, move the locks into the affected functions: trapsignal, scdebug*, ktrsyscall, ktrsysret, systrace_redirect and ADDUPROF. In the cases we already hold the biglock we'll just recurse. kettenis@, beck@ ok
* Clean up after P_BIGLOCK removal.art2011-07-061-9/+9
| | | | | | | KERNEL_PROC_LOCK -> KERNEL_LOCK KERNEL_PROC_UNLOCK -> KERNEL_UNLOCK oga@ ok
* Fix comments as well.kettenis2010-05-091-3/+3
|
* Rename round() to fpu_round(). This matches the m68k code from which thiskettenis2010-05-091-9/+9
| | | | | | | code is derived and makes gcc4 happy by avoiding a conflict with the builtin for the standard C99 round(3). ok miod@
* Make sure we grab the kernel lock before invoking trapsignal().miod2010-01-011-1/+9
| | | | ok kettenis@
* First pass at removing clauses 3 and 4 from NetBSD licenses.ray2008-06-261-8/+1
| | | | | | | | | 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@
* Make fpproc per-cpu. This turns fpproc into a macro, so adjust fpu/fpu.ckettenis2007-09-091-17/+17
| | | | to avoid collisions.
* There was hope, at one time, of sharing fpu/ with sparc. That's stilljason2006-06-215-39/+5
| | | | | | possible, but not with #ifdef SUN4U all over the place. The define should be _v9 or __sparc64__ since it will equally well apply to all v9 machines. We'll burn that bridge before we cross it.
* Make the array of fpu exception types/codes big enough.kettenis2006-05-141-7/+7
| | | | | Fix a typo too, such that we actually read beyond element 0 in all cases. ok miod@
* Remove the advertising clause in the UCB license which Berkeleymillert2003-06-0211-66/+22
| | | | rescinded 22 July 1999. Proofed by myself and Theo.
* one more... nuke clause 3 & 4jason2003-06-021-6/+1
|
* Remove a large chunk of `#ifdef notyet' codejason2002-08-121-53/+1
|
* Don't depend on curproc, pass the proc * down to the functions that need it (fmovr/fmovcc).jason2002-08-032-22/+25
|
* 'sez' is not an english word; ok deraadt@jsyn2002-07-101-2/+2
|
* First round of __P removal in sysmillert2002-03-143-47/+47
|
* define FPE_STATE so that DEBUG will compilejason2001-10-051-1/+2
|
* Mostly rewritten decode of fpu emulation:jason2001-09-151-278/+683
| | | | | | o handles decode of all 64 32 bit registers (based on type) o easier to read (at least imo). o needs some optimization (clarity was the first goal here)
* More accurate compare (from netbsd)jason2001-09-101-13/+9
|
* More debugging stuff.jason2001-09-102-9/+23
|
* Add more debugging stuff (including fpstate dumping)jason2001-09-101-7/+38
| | | | Also, and more importantly, applying the same mask to rs2 and rd is bad ju-ju for the F?TO? instructions.
* Add some debugging stuff: a function to dump all of the fpstate64 fields.jason2001-09-101-1/+4
|
* only one prototype for fpu_implode necessary.jason2001-09-091-2/+1
|
* in fpu_ftox() make sure the high bits are the return value and res[1] getsjason2001-09-091-5/+10
| | | | the low bits.
* From NetBSD: debugging stuff to dump fpn'sjason2001-09-081-1/+18
|
* Import netbsd's fpu handling (modified to be separate from sparc); this isjason2001-09-0812-0/+3301
for paranoia's sake (don't want to break sparc32 while hacking on it). Real work by art.
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.