Age | Commit message (Collapse) | Author | |
---|---|---|---|
2007-05-28 | Move the MSIZE, MCLSHIFT, MCLBYTES and the MCLOFSET | Thordur I. Bjornsson | |
mbuf constants from MD param.h to MI param.h. Besides being the same on every arch, things will most probly break if any arch has different values then the others. The NMBCLUSTERS constants needs to be MD though; ok miod@,krw@,claudio@ | |||
2007-05-22 | protect more variables from existing outside of locore, should fix | Martin Reindl | |
cpuinfo breakage (but my make build is still running) ok gwk@ | |||
2007-05-16 | splassert_ctl defaults to 1 now, so dont wrap the checks for | Thordur I. Bjornsson | |
splassert_ctl > 0 in __predict_false(). ok deraadt@ | |||
2007-05-16 | The world of __HAVEs and __HAVE_NOTs is reducing. All architectures | Artur Grabowski | |
have cpu_info now, so kill the option. eyeballed by jsg@ and grange@ | |||
2007-05-15 | Remove the MI implementation of mutexes and remove the __HAVE_MUTEX | Artur Grabowski | |
option. Every architecture implements mutexes now. | |||
2007-05-15 | switch m68k to __HAVE_CPUINFO | Martin Reindl | |
help miod@, art@ ok deraadt@ | |||
2007-05-01 | Provide <machine/lock.h> on all platforms, so that MI code may #include it | Miod Vallat | |
unconditionnaly. | |||
2007-04-13 | While splitting flags and pqflags might have been a good idea in theory | Artur Grabowski | |
to separate locking, on most modern machines this is not enough since operations on short types touch other short types that share the same word in memory. Merge pg_flags and pqflags again and now use atomic operations to change the flags. Also bump wire_count to an int and pg_version might go int as well, just for alignment. tested by many, many. ok miod@ | |||
2007-04-10 | ``it's'' -> ``its'' when the grammar gods require this change. | Miod Vallat | |
2007-04-04 | Mechanically rename the "flags" and "version" fields in struct vm_page | Artur Grabowski | |
to "pg_flags" and "pg_version", so that they are a bit easier to work with. Whoever uses generic names like this for a popular struct obviously doesn't read much code. Most architectures compile and there are no functionality changes. deraadt@ ok ("if something fails to compile, we fix that by hand") | |||
2007-03-21 | Real atomic_{set,clear}bits_int implementation, no performance hit on 060 | Miod Vallat | |
unless used on non-aligned data; ok deraadt@ | |||
2007-03-04 | Oops, bring back pmap_collect1() and the check for pmap_kernel() in | Miod Vallat | |
pmap_collect() as we do an explicit pmap_collect(pmap_kernel()) in dire memory situations. | |||
2007-02-26 | Zero rval[1] in setregs. | Miod Vallat | |
2007-02-19 | only make this interface available to the kernel for now, discussed witha | Theo de Raadt | |
rt and such; tested and ok miod drahn | |||
2007-02-17 | If we __HAVE_PMAP_DIRECT, handle these mappings in pmap_extract(). | Miod Vallat | |
2007-02-17 | In pmap_collect(): | Miod Vallat | |
- do not bother checking for pmap_kernel() and operating on it, as pmap_collect() will not be invoked for P_SYSTEM process. - preserve wired pages while purging. | |||
2007-02-06 | Add machine/atomic.h to all architectures and define two operations | Artur Grabowski | |
right now that are supposed to be atomic with respect to interrupts and SMP: atomic_setbits_int and atomic_clearbits_int. All architectures other than i386 and amd64 get dummy implementations since at first we'll be replacing operations that are done with "a |= bit" and "a &= ~bit" today. More proper implementations will follow kettenis@, miod@ ok | |||
2007-02-05 | Oops, forgot to commit these files when adding md mutex code; spotted by nick@ | Miod Vallat | |
2007-02-03 | Simple single-processor mutex implementation, simpler than the MI code by | Miod Vallat | |
use of MD spl code bowels. No functional change. | |||
2007-01-28 | optimize m68k writeback(): | Miod Vallat | |
- it can really only be invoked from trap(), not from other userret() callers, so it is safe to hardcode its docachepush parameter to 1. - use pmap_kenter_pa()/pmap_kremove() for the temporary mapping instead of pmap_enter()/pmap_remove(). optimize m68k userret(): - define PROC_PC for m68k systems. - only check want_resched when processing T_ASTFLT traps. - provide two version of userret(), one which will also invoke writeback() on 68040 if required, which is only invoked from trap(), and regular userret(). This speeds up system call returns. | |||
2006-11-30 | s/completly/completely/ | Michael Knudsen | |
ok brad | |||
2006-11-29 | Remove MAPPEDCOPY. It does not work and I am not convinced it is worth | Miod Vallat | |
fixing. | |||
2006-08-22 | Implement pmap_steal_memory() if HAVE_PMAP_DIRECT; no change except for | Miod Vallat | |
faster uvm initialization. | |||
2006-07-09 | Make this compile on platforms which define __HAVE_PMAP_DIRECT. | Miod Vallat | |
2006-07-06 | PMAP_MD_RWLOW kluge is no longer necessary now that mac68k does TRT. | Miod Vallat | |
2006-06-24 | Add pmap_enter_cache(), similar to pmap_kenter_cache() but for managed | Miod Vallat | |
pages, and implement pmap_enter() as a particular case of it. | |||
2006-06-20 | Fix PMAP_DEBUG-only tests I botched in revision 1.13. | Miod Vallat | |
2006-06-17 | Introduce pmap_kenter_cache(), similar to pmap_kenter_pa() but allowing | Miod Vallat | |
the pte cache bits to be specified. Will be used very soon. | |||
2006-06-12 | Oops, typo | Miod Vallat | |
2006-06-11 | Protect a variable only used for the HP MMU if #ifdef M68K_MMU_HP blocks. | Miod Vallat | |
2006-06-11 | Move common prototypes to <m68k/{cpu,param}.h> instead of incomplete | Miod Vallat | |
declarations in <machine/{cpu,param}.h> and elsewhere. | |||
2006-06-11 | Factorize spl handling functions and software interrupt routines between | Miod Vallat | |
m68k-based platforms. | |||
2006-06-11 | Move pmap_aliasmask declaration and initialization to MD code. No functional | Miod Vallat | |
change. | |||
2006-06-11 | Use bracketed include paths. | Miod Vallat | |
2006-06-07 | Pass M_CANFAIL to malloc() in sendsig(), and if it fails, kill the process; | Miod Vallat | |
this is better than panic'ing due to low memory condition. | |||
2006-04-16 | Change child_return() to record a proper ktrace record for vfork child | Miod Vallat | |
processes. ok art@ uwe@ (some time ago, and I forgot to commit this chunk) | |||
2006-04-09 | #ifdef lint wraps for va_start | Theo de Raadt | |
2006-03-19 | remove unused bdbtofsb(bn) macro | Martin Reindl | |
found by drahn@ | |||
2006-02-22 | Remove unused _{ins,rem}que functions - they were not even implemented on | Miod Vallat | |
all architectures. | |||
2006-01-30 | Better siginfo field values for fpe failures. | Miod Vallat | |
2006-01-16 | Sync m68k fpe code with NetBSD; but remove unimplemented opcodes stubs, so | Miod Vallat | |
that we trigger SIGILL instead of propagating false results. Tested by martin@ on Macintosh LCIII. | |||
2006-01-16 | Correctly define extended doubles, the way the 6888x use them; from NetBSD. | Miod Vallat | |
2006-01-13 | Make size_t, ssize_t, intptr_t and uintptr_t consistent amongst our | Todd C. Miller | |
architectures. They are now defined as unsigned long, long, long and unsigned long respectively. | |||
2006-01-10 | Remove duplicate _C_LABEL define. OK miod@ | Todd C. Miller | |
2006-01-08 | Expose struct sigcontext, struct sigstat and struct sigaltstack for XPG4.2 | Todd C. Miller | |
2006-01-06 | Don't redefine SIZE_MAX | Todd C. Miller | |
2006-01-06 | Adapt things to use __type_t instead of _BSD_TYPE_T_ | Todd C. Miller | |
Add new sys/_types.h header Include machine/_types.h or sys/_types.h where applicable | |||
2006-01-06 | Merge machine/ansi.h and machine/types.h into machine/_types.h and | Todd C. Miller | |
rename the types using the __ protected namespace (e.g. __size_t). Idea from FreeBSD. | |||
2006-01-05 | don't include sys/errno.h, pointed out by Miod | Martin Reindl | |
2006-01-04 | Switch to the C version of random() on m68k platforms. It is almost as fast | Miod Vallat | |
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@ |