summaryrefslogtreecommitdiffstats
path: root/sys/i386
Commit message (Collapse)AuthorAgeFilesLines
* Remove an extra #includescottl2004-11-211-1/+0
|
* MFC amd64:scottl2004-11-211-29/+45
| | | | | | | | Consolidate all of the bounce tests into the BUS_DMA_COULD_BOUNCE flag. Allocate the bounce zone at either tag creation or map creation to help avoid null-pointer derefs later on. Track total pages per zone so that each zone can get a minimum allocation at tag creation time instead of being defeated by mis-behaving tags that suck up the max amount.
* Maintain the broken state of backwards compatibilty for a.out (anddas2004-11-201-2/+3
| | | | | | | PECOFF!) core dumps. None of the old versions of gdb I tried were able to read a.out core dumps before or after this change. Reviewed by: arch@
* Remove references to U area and garbage collect includes.das2004-11-201-4/+2
| | | | Reviewed by: arch@
* Remove UAREA_PAGES.das2004-11-201-1/+0
| | | | Reviewed by: arch@
* U areas are going away, so don't allocate one for process 0.das2004-11-202-16/+4
| | | | Reviewed by: arch@
* Revert part of rev 1.57. The tag boundary is honored by splitting thescottl2004-11-191-9/+5
| | | | segment, not by bouncing.
* Initiate deorbit burn sequence for 80386 support in FreeBSD: Removejhb2004-11-1610-262/+9
| | | | 80386 (I386_CPU) support from the kernel.
* Add acpi_sony to the list of drivers that are built.imp2004-11-151-0/+3
|
* After discussions with Nate, repo copy the acpi assist drivers fromimp2004-11-153-1710/+0
| | | | | | | | | | i386 to dev/acpi_support. In theory, these devices could be found other than in i386 machines only as amd64 becomes more popular. These drivers don't appear to do anything i386 specific, so move them to dev/acpi_support. Move config lines to files so that those architectures that don't support kernel modules can build them into the kernel. At the same time, rename acpi_snc to acpi_sony to follow the lead of all the other specialty devices.
* o Change the strcmp() to a strncmp() to allow us to attach to 'E'philip2004-11-121-69/+67
| | | | | | | | | | models of laptops, which are essentially the same as the normal ones, as far as acpi_asus is concerned[1] o Use the above as an excuse to reshuffle the mess I made of the probe function when I originally wrote it. Reported by: Soeren Larsen <soeren@whiteswan.dk>
* Minor whitespace nitpicking to reduce my diffs of Real Changes[tm]philip2004-11-121-9/+6
|
* Spell _KERNEL correctly so that UP kernels are actually optimized again.jhb2004-11-121-1/+1
| | | | Submitted by: pjd
* - Use the SMP style ops for atomic_load/store() in userland so thatjhb2004-11-121-2/+1
| | | | | | | libraries and binaries will work on both UP and SMP machines. - Remove unnecessary gcc memory barrier from the UP atomic_store() op. Submitted by: bde
* - Place the gcc memory barrier hint in the right place in the 80386 versionjhb2004-11-111-8/+9
| | | | | | | | | | of atomic_store_rel(). - Use the 80386 versions of atomic_load_acq() and atomic_store_rel() that do not use serializing instructions on all UP kernels since a UP machine does need to synchronize with other CPUs. This trims lots of cycles from spin locks on UP kernels among other things. Benchmarked by: rwatson
* pbio has moved to dev/pbioimp2004-11-111-463/+0
| | | | Prodded by: peter
* cosmetic changes.nyan2004-11-101-6/+1
|
* Zero the tag when it's allocated. Also fix a printf format problem. Thisscottl2004-11-091-2/+3
| | | | should fix the problems introduced several hours ago.
* First pass at replacing the single global bounce pool with sub-pools that arescottl2004-11-091-36/+140
| | | | | | | | | | | | | | appropriate for different tag requirements. With the former global pool, bounce pages might get allocated that are appropriate for one tag, but not appropriate for another, but the system had no way to distinguish between them. Now zones with distinct attributes are created to hold pages, and each tag that requires bouncing is associated with a zone. New zones are created as needed if no existing zones can meet the requirements of the tag. Stats for each zone are tracked via the hw.busdma sysctl node. This should help drivers that are failing with mysterious data corruption. MFC After: 1 week
* Add support for gadgets on Asus L4R and M6R notebooks.philip2004-11-081-0/+22
|
* Introduce two new options, "CPU private" and "no wait", to sf_buf_alloc().alc2004-11-081-3/+5
| | | | | | Change the spelling of the "catch" option to be consistent with the new options. Implement the "no wait" option. An implementation of the "CPU private" for i386 will be committed at a later date.
* Add power profile support so that the LCD changes brightness levels basednjl2004-11-071-20/+86
| | | | | | | on the AC line state. Submitted by: OGAWA Takaya <t-ogawa@triaez.kaisei.org> MFC after: 1 week
* Begin an invasion of i386-land by amd64.peter2004-11-061-0/+11
| | | | | | Expose some of the amd64-specific sysarch functions to allow alternative implementations of the %fs/%gs code for TLS, threads, etc. USER_LDT does not exist on the amd64 kernel, so we have to implement things other ways.
* Fix support for the Asus-compatible gadgets in Samsung P30/P35 laptops.philip2004-11-051-9/+47
| | | | | PR: 73380 Submitted by: Sebastian Schulze Struchtrup <seb@struchtrup.com>
* Don't use atomic ops to increment interrupt stats. This was only done onscottl2004-11-031-3/+3
| | | | amd64 and i386 anyways. The stats are only kept for informational purposes.
* Streamline busdma a bit. Inline _bus_dmamap_load_buffer, optimize somescottl2004-11-021-18/+19
| | | | tests, replace a passed td with a passed pmap to eliminate some deferences.
* Reduce annoying SCSI probing delay from 15 to 5 seconds in all GENRIC kernels.andre2004-11-021-1/+1
| | | | Discussed on: -current
* Add support for Asus M6N laptopsphilip2004-11-021-0/+11
| | | | Submitted by: Andreas Dieling <snow@quantentunnel.de>
* - Change the ddb paging "support" to use a variable (db_lines_per_page) tojhb2004-11-013-3/+3
| | | | | | | | | | | | | | | | | control the number of lines per page rather than a constant. The variable can be examined and changed in ddb as '$lines'. Setting the variable to 0 will effectively turn off paging. - Change db_putchar() to force out pending whitespace before outputting newlines and carriage returns so that one can rub out content on the current line via '\r \r' type strings. - Change the simple pager to rub out the --More-- prompt explicitly when the routine exits. - Add some aliases to the simple pager to make it more compatible with more(1): 'e' and 'j' do a single line. 'd' does half a page, and 'f' does a full page. MFC after: 1 month Inspired by: kris
* Allow individual application processors to be disabled from the loaderjhb2004-11-011-1/+16
| | | | | | | via hints for 'lapicX'. For example, to disable the CPU with the local APIC ID of 7, use 'hint.lapic.7.disabled=1'. MFC after: 1 month
* Add TUNABLE_LONG and TUNABLE_ULONG, and use the latter for thedes2004-10-311-4/+3
| | | | | | | hw.pci.host_mem_start tunable. Add comments to TUNABLE_INT and TUNABLE_QUAD recommending against their use. MFC after: 3 weeks
* Whitespace cleanupdes2004-10-311-8/+8
|
* Implement per-CPU SYSMAPs, i.e., CADDR* and CMAP*, to reduce lockalc2004-10-291-33/+50
| | | | contention within pmap_zero_page() and pmap_copy_page().
* Preserve dcons(4) buffer passed by loader(8).simokawa2004-10-241-0/+16
|
* Hook the hptmv driver up to the build.scottl2004-10-242-0/+6
|
* Add some basic KTR tracing to busdma on i386. This is likely notrwatson2004-10-231-10/+60
| | | | | | | | | the final set of traces -- someone with more busdma background will probably want to review and expand this, as well as port to other platforms. This tracing is sufficient to identify key busdma events on i386, and in particular to draw attention to bounce buffering events that may have a substantial performance impact.
* Remove a "needs Giant" flag from the /dev/apm compat device.njl2004-10-221-1/+0
| | | | MFC after: 2 weeks
* Add new function ttyinitmode() which sets our systemwide defaultphk2004-10-181-6/+1
| | | | | | | | | | | | | | modes on a tty structure. Both the ".init" and the current settings are initialized allowing the function to be used both at attach and open time. The function takes an argument to decide if echoing should be enabled. Echoing should not be enabled for regular physical serial ports unless they are consoles, in which case they should be configured by ttyconsolemode() instead. Use the new function throughout.
* When sf_buf_alloc() replaces a virtual-to-physical mapping, it needn'talc2004-10-161-1/+15
| | | | | | | | | invalidate the TLB(s) if the old mapping wasn't used by the CPU. With network interfaces that implement checksum off-loading, the old mapping is almost never used by the CPU, only by the device driver for setting up the DMA operation. Reviewed by: tegge@
* Let nexus print our flags for us. Also, clean up an obfuscated if stmt.njl2004-10-141-10/+7
|
* Print flags in the nexus for child devices.njl2004-10-141-0/+2
|
* Remove local hacks to set flags now that the device probe does this for us.njl2004-10-142-11/+2
| | | | | | | Tested on every device except sio_pci and the pc98 fd.c. Perhaps something similar should be done for the "disabled" hints also. MFC after: 2 weeks
* Add zero flags argument to sysctl calls.phk2004-10-121-1/+1
|
* Add missing zero flag arguments to sysctl calls.phk2004-10-123-3/+3
| | | | Add missing pointy hat to peter@
* Make the lower range of the memory area 0x80000000 again. Alsoimp2004-10-111-1/+15
| | | | | | introduce hw.{pci,acpi}.host_mem_start tunable to change this. MFC: ASAP
* Match surrounding style, not style(msmith).njl2004-10-111-16/+16
|
* Move the code for halting the CPU (acpi_cpu_c1) into machdep files.njl2004-10-112-1/+8
| | | | | | This removes the last MD portion of acpi_cpu.c. MFC after: 2 weeks
* Fix conflicts I didn't fix before I committed my busspace changes.imp2004-10-111-4/+2
| | | | Noticed by: ru@ (and likely tinderbox, I haven't checked)
* Convert to newbus. (chances are we could now move this to dev/pbioimp2004-10-101-46/+61
| | | | | | since I believe it is now MI, but that hasn't been done yet). Reviewed by: dds
* style(9)obrien2004-10-091-57/+51
|
OpenPOWER on IntegriCloud