Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Fix another mistake in the bus_dmamem_alloc_size() thing | scottl | 2003-01-29 | 1 | -2/+2 |
| | | | | Submitted by: tmm | ||||
* | Add code to repsect the D_NOGIANT flag, should the disk device driver set it. | phk | 2003-01-29 | 1 | -11/+30 |
| | | | | | | NO_GEOM cleanup: remove ifdefs. Still untested. | ||||
* | Style: keep most typedefs in the same place. | mike | 2003-01-29 | 1 | -41/+43 |
| | |||||
* | Sort these functions as the author instructed. | phk | 2003-01-29 | 1 | -16/+13 |
| | |||||
* | Make tsc_freq a 64bit on PC98 also. | phk | 2003-01-29 | 3 | -6/+9 |
| | |||||
* | Fix some more missing dt_ prefixes for dma tag fields. | scottl | 2003-01-29 | 4 | -11/+11 |
| | |||||
* | - Probe number of IT/IR DMA contexts as specified in OHCI spec. | simokawa | 2003-01-29 | 1 | -9/+9 |
| | |||||
* | The Xircom hardware always delivers received packets with the FCS appended. | joerg | 2003-01-29 | 1 | -0/+2 |
| | | | | | Thus, mark the M_HASFCS flag so the generic ethernet layers will account for this. | ||||
* | Escape the backslash in badchars so that smbfs_pathcheck() correctly | tjr | 2003-01-29 | 1 | -1/+1 |
| | | | | | | rejects pathnames with backslashes in them (and to avoid a syntax error). Found by: FlexeLint | ||||
* | NO_GEOM cleanup: Don't inlude diskslice_machdep.c | phk | 2003-01-29 | 1 | -1/+0 |
| | |||||
* | Fix two fatal signedness errors introduced when i and j in semop() | tjr | 2003-01-29 | 1 | -5/+6 |
| | | | | | | were changed from int to size_t in the previous revision. PR: 47625 | ||||
* | Make tsc_freq a 64bit quantity. | phk | 2003-01-29 | 9 | -39/+51 |
| | | | | Inspired by: http://www.theinquirer.net/?article=7481 | ||||
* | Move timecounters notion of frequency to 64 bits. | phk | 2003-01-29 | 2 | -5/+6 |
| | | | | [WARNING: CPUs in the distant future may be closer than they appear!] | ||||
* | Fix a typo in dt_maxsize from the last commit | scottl | 2003-01-29 | 1 | -1/+1 |
| | |||||
* | Implement bus_dmamem_alloc_size() and bus_dmamem_free_size() as | scottl | 2003-01-29 | 13 | -39/+248 |
| | | | | | | | | | | | | | | | | | | | | counterparts to bus_dmamem_alloc() and bus_dmamem_free(). This allows the caller to specify the size of the allocation instead of it defaulting to the max_size field of the busdma tag. This is intended to aid in converting drivers to busdma. Lots of hardware cannot understand scatter/gather lists, which forces the driver to copy the i/o buffers to a single contiguous region before sending it to the hardware. Without these new methods, this would require a new busdma tag for each operation, or a complex internal allocator/cache for each driver. Allocations greater than PAGE_SIZE are rounded up to the next PAGE_SIZE by contigmalloc(), so this is not suitable for multiple static allocations that would be better served by a single fixed-length subdivided allocation. Reviewed by: jake (sparc64) | ||||
* | - Use ksq_load as the authoritive count of kses on the pair of kseqs for | jeff | 2003-01-29 | 1 | -84/+79 |
| | | | | | | | | sched_runnable() et all. - Remove some dead code in sched_clock(). - Define two macros KSEQ_SELF() and KSEQ_CPU() for getting the kseq of the current cpu or some alternate cpu. - Start introducing kseq_() functions, such as kseq_choose() and kseq_setup(). | ||||
* | Fix a bug with syncookies; previously, the syncache's MSS size was not | silby | 2003-01-29 | 1 | -2/+2 |
| | | | | | | | | initialized until after a syncookie was generated. As a result, all connections resulting from a returned cookie would end up using a MSS of ~512 bytes. Now larger packets will be used where possible. MFC after: 5 days | ||||
* | Set the termios speed based on the tty speed. This allows consoles to work at | benno | 2003-01-29 | 1 | -1/+1 |
| | | | | | | speeds other than TTYDEF_SPEED. Approved by: jake | ||||
* | - Fix build on alpha. | simokawa | 2003-01-29 | 2 | -4/+3 |
| | | | | - Add a missing newline in printf. | ||||
* | - Remove debugging code that didn't work on UP. | jeff | 2003-01-29 | 1 | -44/+0 |
| | |||||
* | Migrate the PCI bus logic for ohci and uhci from sys/pci to sys/dev/usb. | joe | 2003-01-29 | 3 | -709/+2 |
| | |||||
* | Check bounds for index before dereferencing memory past end of array. | phk | 2003-01-28 | 1 | -1/+1 |
| | | | | Found by: FlexeLint | ||||
* | Use VOP_SPECSTRATEGY() instead of VOP_STRATEGY(). | phk | 2003-01-28 | 2 | -2/+2 |
| | |||||
* | Avoid lock order reversal by expanding the scope of the | hsu | 2003-01-28 | 2 | -15/+6 |
| | | | | AF_INET radix tree lock to cover the ARP data structures. | ||||
* | Remove BDE_DEBUGGER. | jake | 2003-01-28 | 12 | -450/+10 |
| | | | | Discussed with: bde | ||||
* | NO_GEOM cleanup: remove subr_disklabel.c from powerpc and x86_64. | phk | 2003-01-28 | 2 | -2/+0 |
| | |||||
* | NO_GEOM cleanup: remove subr_disk{label,mbr,slice}.c from pc98 kernels. | phk | 2003-01-28 | 1 | -3/+0 |
| | |||||
* | - Allow multiple packets read/write for IR/IT to reduce system call. | simokawa | 2003-01-28 | 2 | -35/+43 |
| | | | | - Remove unused variables. | ||||
* | NO_GEOM cleanup: subr_disklabel.c no longer needed. | phk | 2003-01-28 | 1 | -1/+0 |
| | |||||
* | NO_GEOM cleanup: Remove subr_disklabel.c from alpha kernels. | phk | 2003-01-28 | 1 | -1/+0 |
| | |||||
* | NO_GEOM cleanup: Don't include subr_disklabel.c in sparc64 build | phk | 2003-01-28 | 1 | -1/+0 |
| | |||||
* | Band-XXX-aid an easy to provoke panic. | phk | 2003-01-28 | 1 | -0/+7 |
| | | | | MFC: 2 weeks | ||||
* | NO_GEOM cleanup: Don't pull subr_disk{slice,label,mbr}.c into i386 | phk | 2003-01-28 | 1 | -3/+0 |
| | | | | kernels anymore, they are not needed. | ||||
* | Add PCI revision number for 630A and 900B. Enable parity error detection | mbr | 2003-01-28 | 2 | -19/+86 |
| | | | | | | | | | | on 900B and 635(A). Re-add the enhanced PHY access register method again for older chipsets, they do not seem to work with all old chips. Reviewed by: phk MFC after: 7 days | ||||
* | Put the right fix in. Instead of deleting the declaration of __FBSDID, we | benno | 2003-01-28 | 1 | -0/+7 |
| | | | | undef it before this definition. | ||||
* | Mark some args unused so this compiles in userland. | phk | 2003-01-28 | 1 | -1/+1 |
| | |||||
* | Use a void * to carry the private data for return-call'ed ioctl requests. | phk | 2003-01-28 | 1 | -1/+1 |
| | | | | | Amongst other things this avoids a complex workaround in the userland regression bits. | ||||
* | - Allow idle's pctcpu time to be calculated. | jeff | 2003-01-28 | 1 | -6/+6 |
| | |||||
* | - Fix the ksq_load calculation. It now reflects the number of entries on the | jeff | 2003-01-28 | 1 | -27/+134 |
| | | | | | | | | run queue for each cpu. - Introduce kse stealing into the sched_choose() code. This helps balance cpus better in cases where process turnover is high. This implementation is fairly trivial and will likely be only a temporary measure until something more sophisticated has been written. | ||||
* | Do not allow a cached vnode to be shared among multiple mounts of the same | tjr | 2003-01-28 | 1 | -1/+2 |
| | | | | | | | | | kind of pseudofs-based filesystem. Fixes (at least) one problem where when procfs is mounted mupltiple times, trying to unmount one will often cause the wrong one to get unmounted, and other problem where mounting one procfs on top of another caused the kernel to lock up. Reviewed by: des | ||||
* | Bang! Bang! Bang! etc etc | phk | 2003-01-28 | 1 | -2/+0 |
| | | | | | | Remove NODEVFS and NO_GEOM from options. Approved by: trb@ | ||||
* | Bang! Bang! Bang! etc etc. | phk | 2003-01-28 | 1 | -1/+0 |
| | | | | Remove NODEVFS option. | ||||
* | A few fixes to rev 1.221 | silby | 2003-01-28 | 1 | -9/+17 |
| | | | | | | | | | - Honor the previous behavior of maxfragpackets = 0 or -1 - Take a better stab at fragment statistics - Move / correct a comment Suggested by: maxim@ MFC after: 7 days | ||||
* | Merge pmap_testbit() and pmap_is_modified(). The latter is the only caller | alc | 2003-01-28 | 2 | -58/+18 |
| | | | | of the former. | ||||
* | No longer force COMPAT_FREEBSD4 to be on. | peter | 2003-01-27 | 1 | -3/+0 |
| | |||||
* | Add the sabtty children devices as unordered. Use the unit numbr of | jake | 2003-01-27 | 1 | -5/+4 |
| | | | | | | | | | the sabtty device to create its description so that they will be unique for machines with multiple sab chips. This fixes a panic on machines with an rsc card. Tested by: obrien (e250) | ||||
* | Don't dereference null vnode pointer if controling terminal was revoked. | phk | 2003-01-27 | 1 | -3/+5 |
| | | | | Submitted by: "Peter Edwards" <pmedwards@eircom.net> | ||||
* | Add PCI id for fxp(Intel Pro/100 M Mobile Connection). | sanpei | 2003-01-27 | 1 | -0/+1 |
| | | | | | | PR: i386/41915 Submitted by: Ross A. Lippert <ripper@eskimo.com> MFC after: 1 week | ||||
* | Back the previous commit out. It didn't actually fix the problem I was | benno | 2003-01-27 | 1 | -3/+5 |
| | | | | | | seeing and the memory barrier isn't needed with the bridges we're using. Fix the function style however. | ||||
* | Put the Perfection 1660 scanner entry in the right place. (Entries | joe | 2003-01-27 | 1 | -1/+1 |
| | | | | are sorted by product id). |