summaryrefslogtreecommitdiffstats
path: root/sys/i386
Commit message (Collapse)AuthorAgeFilesLines
* Add a new loader tunable, hw.hasbrokenint12, to indicate that BIOSiwasaki2002-11-091-46/+78
| | | | | | | | | | has broken int 12H. If hw.hasbrokenint12="1" in loader environment, kernel never use BIOS INT 12 call to determine base memory size. Otherwise, kernel use INT 12 in old behaviour. This should fix kernel panic problem caused by 1.544 changes. MFC after: 1 day
* Print real / avail memory in megabytes rather than kilobytes.des2002-11-091-4/+4
|
* Fix instances of macros with improperly parenthasized arguments.alfred2002-11-091-1/+1
| | | | Verified by: md5
* Make 3dfx i386-only. The memrange API it uses may be defined in an MIjhb2002-11-081-0/+12
| | | | header, but it is only implemented on i386.
* Move rc(4) over to MI notes and enable it as a MI module.jhb2002-11-081-5/+0
|
* Make xrpu(4) i386-only. Consumers of i386_btop() are not MI.jhb2002-11-081-1/+2
|
* adjust critical section to only wrap around vm86_bioscall().davidxu2002-11-081-2/+2
|
* use critical_enter/exit to add a critical section around BIOS call,davidxu2002-11-081-5/+9
| | | | | | this unbreaks WITNESS. Pointed out by: jhb
* Move the definitions of the hw.physmem, hw.usermem and hw.availpagestmm2002-11-071-37/+0
| | | | | | | | | | | sysctls to MI code; this reduces code duplication and makes all of them available on sparc64, and the latter two on powerpc. The semantics by the i386 and pc98 hw.availpages is slightly changed: previously, holes between ranges of available pages would be included, while they are excluded now. The new behaviour should be more correct and brings i386 in line with the other architectures. Move physmem to vm/vm_init.c, where this variable is used in MI code.
* Properly parenthesize the DBREG_DRX macro's variables to allow foralfred2002-11-071-1/+1
| | | | DBREG_DRX(&dbregs, n) usage.
* Move sr(4) over to i386-only as it is yet another user of kvtop().jhb2002-11-071-0/+6
|
* Simplify and optimize pmap_object_init_pt(). More specifically,alc2002-11-071-65/+32
| | | | | | take advantage of the fact that the vm object's list of pages is now ordered to reduce the overhead of finding the desired set of pages to be mapped. (See revision 1.215 of vm/vm_page.c.)
* Move firewire back to being MI.jhb2002-11-071-7/+0
|
* 1.Fix smp race between kernel vm86 BIOS calling and userland vm86 mode code,davidxu2002-11-077-21/+18
| | | | | | | | | | remove global variable in_vm86call, set vm86 calling flag in PCB flags. 2.Fix vm86 BIOS calling preempted problem by changing vm86_lock mutex type from MTX_DEF to MTX_SPIN. vm86pcb is not remembered in thread struct, when the thread calling vm86 BIOS is preempted by interrupt thread, and later switching back to the thread would cause incorrect context be loaded into CPU registers, this leads to kernel crash.
* lnc(4) uses kvtop() and is thus i386-only for now.jhb2002-11-061-0/+7
|
* Make firewire i386-only for now. It doesn't quite handle machines withjhb2002-11-061-0/+7
| | | | | 64-bit address spaces yet. Pointy hat to myself for sticking it in the MI NOTES file to begin with.
* ed(4) uses kvtop() and is thus i386-only. It has several other warningsjhb2002-11-061-0/+9
| | | | | | | related to sizeof(int) != sizeof(void *), but kvtop() is much harder to fix. Approved by: imp
* Move the drm code to the i386 MD NOTES file. It can be added to other MDjhb2002-11-061-0/+20
| | | | | NOTES files if desired, but this code is not MI on FreeBSD. The Alpha support is Linux specific and does not compile on FreeBSD.
* Move digi to the i386 MD NOTES until it stops using inb() and outb().jhb2002-11-061-0/+13
| | | | Please use bus_space functions instead.
* - Move comments regarding flags for dgb(4) over to the MD NOTES file wherejhb2002-11-061-6/+11
| | | | | | dgb(4) lives. - Move dgb(4) back to where it used to be relative to other drives in the old NOTES/LINT file.
* Make the ar(4) driver i386-only for now. It has lots of sizeof(int) ==jhb2002-11-061-0/+7
| | | | | | | sizeof(void *) assumptions and doesn't use busdma yet (it uses kvtop() which is not an MI interface). Recommended by: jake, mux
* Remove what was a temporary bogus assignment of bits of siginfo_t, as it doesjmallett2002-11-061-4/+0
| | | | | | not look like the prerequisites to fill it in properly will be in the tree for the upcoming release, but it's mostly done, so there is no need for these to stay around to remind us.
* - Convert to newbus, bus_space etc.mdodd2002-11-053-1705/+0
| | | | | | - Move to MI space. Tested on: i386
* Fix typo. ioport_rid should be irq_rid.davidxu2002-11-051-1/+1
|
* MFp4:imp2002-11-022-20/+68
| | | | | | | | | | | | | | | o It turns out that we always need to try to route the interrupts for the case where the $PIR tells us there can be only one. Some machines require this, while others fail when we try to do this (bogusly, imho). Since we have no apriori way of knowing which is which, we always try to do the routing and hope for the best if things fail. o Add some additional comments that state the obvious, but amplify it in non-obvious ways (judging from the questions I've gotten). This should un-break older laptops that still have to use PCIBIOS to route interrupts. Tested by: sam
* Use 0xffffffff instead of -1 for id to compare against.imp2002-11-022-20/+22
| | | | | | | Use exact width types, since this is a MD file and won't be used elsewhere. Fix a couple of resulting printf breakages Bug found by: phk using Flexlint
* bzero() the sigframe before we fill it. This was not done at all inmarcel2002-11-021-3/+4
| | | | linux_rt_sendsig() and only done for the fpstate in linux_sendsig().
* Merge PC98 changes.mdodd2002-11-022-32/+62
|
* Handle hints for the atspeaker device.mdodd2002-10-282-1/+27
| | | | Document same.
* Note that the sched_lock protects md_ldt of struct mdproc.jhb2002-10-251-1/+1
|
* Finish fixing the 5.x FPU code for dealing with signal handlers.peter2002-10-251-0/+1
| | | | Obtained from: bde
* Split 4.x and 5.x signal handling so that we can keep 4.x signalpeter2002-10-257-30/+339
| | | | | | | | | | | | | | | | handling clean and functional as 5.x evolves. This allows some of the nasty bandaids in the 5.x codepaths to be unwound. Encapsulate 4.x signal handling under COMPAT_FREEBSD4 (there is an anti-foot-shooting measure in place, 5.x folks need this for a while) and finish encapsulating the older stuff under COMPAT_43. Since the ancient stuff is required on alpha (longjmp(3) passes a 'struct osigcontext *' to the current sigreturn(2), instead of the 'ucontext_t *' that sigreturn is supposed to take), add a compile time check to prevent foot shooting there too. Add uniform COMPAT_43 stubs for ia64/sparc64/powerpc. Tested on: i386, alpha, ia64. Compiled on sparc64 (a few days ago). Approved by: re
* Change method to determine base memory size.iwasaki2002-10-251-43/+55
| | | | | | | | | | Try INT 15H/E820H first, then fall back to the old compatibility method (INT 12H). This is a workaround for newer machines which have broken INT 12H BIOS service implementation. Reviewed by: -current ML MFC after: 3 days
* Use the correct values for LDBL_*. Libc doesn't completely supportimp2002-10-251-9/+10
| | | | | | | | | | | | | | long doubles at the moment (printf truncates them to doubles). However, long doubles to appear to work to the ranges listed in this commit on both -stable (4.5) and -current. There may be some slight rounding issues with long doubles, but that's an orthogonal issue to these constants. I've had this in my local tree for 3 months, and in my company's local tree for 15 months with no ill effects. Obtained from: NetBSD Not likely to like it: bde
* Extract out KSE specific code from machine specific codejulian2002-10-241-38/+2
| | | | | | | | so that there is ony one copy of it. Fix that one copy so that KSEs with no mailbox in a KSE program are not a cause of page faults (this can legitmatly happen). Submitted by: (parts) davidxu
* Rename 'device acpica' to 'device acpi'.jhb2002-10-241-1/+1
| | | | Approved by: msmith, iwasaki
* device rc no longer requires a count since jhb newbussified the rcrwatson2002-10-241-1/+1
| | | | device driver.
* - New-bussify the rc(4) device driver.jhb2002-10-233-1738/+0
| | | | | | | | | | | | - Add detach support to the driver so that you can kldunload the module. Note that currently rc_detach() fails to detach a unit if any of its child devices are open, thus a kldunload will fail if any of the tty devices are currently open. - sys/i386/isa/ic/cd180.h was moved to sys/dev/ic/cd180.h as part of this change. Requested by: rwatson Tested by: rwatson
* npx(4) doesn't honor port hints so don't provide an example one.jhb2002-10-231-1/+0
|
* Very minor whitespace/style nit.markm2002-10-231-2/+1
|
* - No need for pmtimer hint anymore.jhb2002-10-221-2/+0
| | | | - npx doesn't need an 'at' hint anymore.
* No need for pmtimer hint anymore.jhb2002-10-221-1/+0
|
* Add an identify method that creates a pmtimer0 device if it doesn't alreedyjhb2002-10-221-2/+19
| | | | exist. Hints are no longer needed to instantiate a pmtimer(4) device.
* Use if_printf() instead of printf() to print diagnositic messages in thebrooks2002-10-211-9/+6
| | | | form "rdp#: blah".
* Use if_printf(ifp, "blah") instead ofbrooks2002-10-211-22/+15
| | | | printf("%s%d: blah", ifp->if_name, ifp->if_unit).
* Change the definition of the debugging registers to be an array, sophk2002-10-203-58/+55
| | | | | | | that we can index into it, rather than do pointer gymnastics on a structure containing 8 elements. Verified by: MD5 hash on the produced .o files.
* Remove a boatload of '&' which are surplus to the requirements.phk2002-10-202-82/+82
| | | | Validated by: md5 hash is unchanged.
* Revert last commit, there actually was a -1 waaaaay down in pcireg_cfgread().phk2002-10-201-0/+2
|
* Hide inline assembly if lint is defined.phk2002-10-201-1/+1
|
* "id" is never going to be -1 when it is unsigned.phk2002-10-201-2/+0
| | | | Spotted by: FlexeLint
OpenPOWER on IntegriCloud