summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Output machdep.guessed_bootdev as an integer rather than try to formatphk2003-04-031-54/+0
| | | | | | it according to ancient and obsolete rules. This removes one more user of <sys/diskslice.h>
* Use __FBSDID rather than rcsid[].obrien2003-04-03102-529/+262
|
* Fix typo: On the relevant architectures, floppy drives are handled bybmah2003-04-032-2/+2
| | | | | | | the fdc(4) driver, not the fd(4) file descriptor files. Submitted by: Peter B <pb@ludd.luth.se> MFC after: 1 week
* Remove useless initialization.mux2003-04-031-1/+0
|
* Add support for the HP Smart Array 6400 EMps2003-04-031-3/+4
| | | | | Change the interrupt mask for the Smart Array 6xxx controllers after discussions w/ HP.
* Retire the DIOCGMBR ioctl before anybody starts to use it.phk2003-04-032-6/+0
|
* Remove BIO_SETATTR from non-GEOM part of kernel as well.phk2003-04-032-2/+0
|
* Remove all references to BIO_SETATTR. We will not be using it.phk2003-04-038-31/+0
|
* Update the initializer for GEOM_MBREXT, I overlooked it previously.phk2003-04-031-3/+2
|
* Add #define for DOSPTYP_PMBR, and use it.phk2003-04-032-1/+2
|
* Add optimized block copy and zero functions using vis instructions, whichjake2003-04-032-0/+140
| | | | | | | | | | can do 64 bytes at a time and don't allocate lines in the L2 cache. These assume that everything is 64 byte aligned, and that there's more than 128 bytes of data (best for whole pages). The block load and store instructions don't follow normal memory ordering rules and require either a memory barrier or move between registers before the data can actually be used. This implementation correctly shuffles around 3 out of the 4 sets of registers in order to avoid memory barriers expect for the last 2 blocks.
* Fix fxp(4), this changed shouldn't have crept in.mux2003-04-031-2/+2
| | | | | Pointy hat to: mux Reported by: Pawel Worach <pawel.worach@nordea.com>
* Stop logging ``authpriv'' messages to the world-readable fileyar2003-04-031-1/+1
| | | | | | | | | /var/log/messages. Such messages are for the eyes of authorized personnel only. PR: conf/48170 Discussed in: freebsd-security MFC after: 2 weeks
* Add support for saving and restoring kernel floating point state. The statejake2003-04-032-1/+45
| | | | | | | | will be saved if we context switch as a result of an interrupt which occured while using the floating point registers in the kernel (which actually can't happen right now). This allows fp disabled traps in the kernel, which normally shouldn't happen, so make sure the trapping code is what we expect it is.
* - Add space for kernel floating point registers to the pcb. These will bejake2003-04-033-10/+18
| | | | | | | | | used to support block copy and zero operations in the kernel which use the floating point registers. - While I'm changing the size, improve the layout of struct pcb, sort by size, then alphabetical etc. - Add some assertions to validate assumptions made about how the pcb is allocated.
* - Generally improve register usage in cpu_switch. Use the 'in' registersjake2003-04-031-71/+63
| | | | | | for temporaries relating to the state of the new process instead of the outs, so that functions can be called without fear of clobbering them. - Use savefpctx instead of rolling our own.
* <sys/disk.h> -> <geom/geom_disk.h>des2003-04-031-1/+1
|
* Don't assume the fp state is at offset 0 in the pcb.jake2003-04-031-1/+1
|
* Fix typos (don't use * when taking the size of an array).jake2003-04-031-2/+2
|
* Remove all the bogus volatile qualifiers from the structs definitionsmux2003-04-032-148/+146
| | | | and associated evil casts to discard them.
* MFi386: revision 1.188nyan2003-04-031-0/+2
|
* MFi386: revision 1.561nyan2003-04-032-6/+66
|
* Don't pull in geom_enc.c any more.phk2003-04-031-3/+1
|
* #include <sys/endian.h> as needed.phk2003-04-032-0/+2
|
* Remove geom_enc.c, a superset of these functions are now available inphk2003-04-033-135/+0
| | | | <sys/endian.h>
* Use <sys/endian.h> instead of geom_enc.c for endianess-agnostification.phk2003-04-035-116/+119
|
* Use sys/endian.h instead of geom_enc.c for endian-agnostfication.phk2003-04-032-17/+19
|
* Pointy hat commit:phk2003-04-031-3/+3
| | | | | Don't Cut&Paste from big endian to little endian function without subsequent adjustments.
* Make sure we don't ignore error codes.phk2003-04-031-32/+24
|
* Use "unsigned char" instead of "u_char" to avoid trouble.phk2003-04-031-12/+12
|
* Add support for Moxa Technologies Co., Ltd. PCI I/O Card 4S RS232phk2003-04-031-0/+13
| | | | Submitted by: sst@cybercity.dk
* Add inline functions {be,le}{16,32,64}{enc,dec}() for encoding decodingphk2003-04-032-1/+147
| | | | into byte strings of unknown alignment.
* NO_GEOM cleanup: Remove dkmakeminor().phk2003-04-031-32/+0
|
* Don't use dkmakeminor(), create our own minor encoding instead.phk2003-04-031-3/+2
|
* Don't use dkmakeminor(), it has nothing to do with CDroms.phk2003-04-031-6/+2
|
* 2.x->3.0 upgrade tools are no longer needed. 5.x is unlikely to fitimp2003-04-033-321/+0
| | | | | | on a 2.1/2.2 root disk anyway. OK'd by: peter
* Kill upgrade target docs. It too is stale.imp2003-04-031-1/+0
| | | | Submitted by: ru
* Kill aout to elf upgrade path. From an old comment in the makefile:imp2003-04-032-301/+0
| | | | | | | | | | | | # If -DWANT_AOUT is specified, a `make world' with OBJFORMAT=elf will # update the legacy support for aout. This includes all libraries, ld.so # and boot objects. This part of build should be regarded as # deprecated and you should _not_ expect to be able to do this past the # release of 4.0. You have exactly one major release to move entirely # to elf. Now that we're past 5.0, these aren't needed at all anymore and it is well past its freshness date.
* MFp4 @27667: WARNS=5 cleanup on i386.jmallett2003-04-035-15/+17
| | | | | | | | | | | | | | | | | | | Remove the unused FILE\ *tf from print_mesg args, and the bogus passing in of an uninitialised FILE* for it. Call a timeval 'now' instead of 'clock' due to shadowing. Remove a nested localtime declaration. Make the delete invite argument match the ID type, u_int32_t. Use const for pointers to const items. Cast to long where printing as such. Include netinet/in.h for htonl/htons. Reviewed by: imp
* - Pass a ucontext_t to _set_curthread. If non-NULL the new thread is setjake2003-04-035-11/+18
| | | | | | | | as curthread in the new context, so that it will be set automatically when the thread is switched to. This fixes a race where we'd run for a little while with curthread unset in _thread_start. Reviewed by: jeff
* - Keep seperate statistics and run queues for different scheduling classes.jeff2003-04-031-48/+134
| | | | | | | | - Treat each class specially in kseq_{choose,add,rem}. Let the rest of the code be less aware of scheduling classes. - Skip the interactivity calculation for non TIMESHARE ksegrps. - Move slice and runq selection into kseq_add(). Uninline it now that it's big.
* Commit a partial lazy thread switch mechanism for i386. it isn't as lazypeter2003-04-0234-271/+900
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | as it could be and can do with some more cleanup. Currently its under options LAZY_SWITCH. What this does is avoid %cr3 reloads for short context switches that do not involve another user process. ie: we can take an interrupt, switch to a kthread and return to the user without explicitly flushing the tlb. However, this isn't as exciting as it could be, the interrupt overhead is still high and too much blocks on Giant still. There are some debug sysctls, for stats and for an on/off switch. The main problem with doing this has been "what if the process that you're running on exits while we're borrowing its address space?" - in this case we use an IPI to give it a kick when we're about to reclaim the pmap. Its not compiled in unless you add the LAZY_SWITCH option. I want to fix a few more things and get some more feedback before turning it on by default. This is NOT a replacement for Bosko's lazy interrupt stuff. This was more meant for the kthread case, while his was for interrupts. Mine helps a little for interrupts, but his helps a lot more. The stats are enabled with options SWTCH_OPTIM_STATS - this has been a pseudo-option for years, I just added a bunch of stuff to it. One non-trivial change was to select a new thread before calling cpu_switch() in the first place. This allows us to catch the silly case of doing a cpu_switch() to the current process. This happens uncomfortably often. This simplifies a bit of the asm code in cpu_switch (no longer have to call choosethread() in the middle). This has been implemented on i386 and (thanks to jake) sparc64. The others will come soon. This is actually seperate to the lazy switch stuff. Glanced at by: jake, jhb
* - Don't overrun the ldt buffer.jeff2003-04-021-2/+2
| | | | Submitted by: gordan@freebsd.org
* When kill(-1) returns ESRCH, it could be because the current processrwatson2003-04-021-1/+1
| | | | | | doesn't have a process group, which can occur if you're working with a custom init that doesn't set up a full tty context. Rather than refusing to reboot, ignore ESRCH from the kill attempt in reboot(8).
* Lock the process before sending it a SIGIO. Not doing so is a panic(2)jhb2003-04-021-2/+7
| | | | implementation with INVARIANTS.
* Add handling for cancelled events in the g_call_me() methods.phk2003-04-028-14/+43
|
* Change events to have an array of "void *" references, and give thephk2003-04-0211-98/+90
| | | | | | | | | | | event posting functions varargs to fill these. Attribute g_call_me() to appropriate g_geom's where necessary. Add a flag argument to g_call_me() methods which will be used to signal cancellation of events in the future. This commit should be a no-op.
* If stat() on the terminal specified in utmp fails due to ENOENT, don'trwatson2003-04-023-2/+21
| | | | | | | | | print a warning, and set the idletime variable for the entry to -1; then pick up the -1 later in sprint() and lprint() and ignore those idle times by printing just whitespace. When third party applications, such as kdm, insert utmp entries, they sometimes use strings like ":0", which can't be stat()'d and currently result in warnings that are not helpful to the user.
* Drop the silly notion that fd* entries have both `a' and `b'ru2003-04-021-11/+0
| | | | partitions; this does nothing except upsetting disklabel(8).
* Back out support for RFC3514.mdodd2003-04-0211-87/+4
| | | | RFC3514 poses an unacceptale risk to compliant systems.
OpenPOWER on IntegriCloud