| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | The x86 atomic operations are already locked, so they do not need an | John Baldwin | 2000-10-28 | 1 | -2/+0 |
| * | Declare or #define per-cpu globals in <machine/globals.h> in all cases. | Bruce Evans | 2000-10-27 | 6 | -31/+0 |
| * | - Add atomic_cmpset_{acq_,rel_,}_long | John Baldwin | 2000-10-25 | 1 | -0/+52 |
| * | - Overhaul the software interrupt code to use interrupt threads for each | John Baldwin | 2000-10-25 | 1 | -2/+1 |
| * | Define the mtx_legal2block() macro used in the witness code that managed | John Baldwin | 2000-10-20 | 1 | -0/+2 |
| * | Catch up to moving headers: | John Baldwin | 2000-10-20 | 1 | -1/+1 |
| * | - Make the mutex code almost completely machine independent. This greatly | John Baldwin | 2000-10-20 | 1 | -512/+23 |
| * | - Expand the set of atomic operations to optionally include memory barriers | John Baldwin | 2000-10-20 | 1 | -2/+108 |
| * | Axe the barrier_{read,write,rw}() helper functions as this method of | John Baldwin | 2000-10-20 | 3 | -72/+0 |
| * | Add PCI BIOS function codes for IRQ routing fetch and route. | Mike Smith | 2000-10-19 | 1 | -0/+2 |
| * | Add in a simple API for memory barriers to machine/bus.h: | John Baldwin | 2000-10-18 | 3 | -0/+69 |
| * | Add types and prototypes. | Warner Losh | 2000-10-16 | 2 | -2/+46 |
| * | Move DELAY() from <machine/clock.h> to <sys/systm.h> | Poul-Henning Kamp | 2000-10-15 | 1 | -1/+0 |
| * | Removed unused include of <machine/lock.h>. The locking interface stopped | Bruce Evans | 2000-10-12 | 1 | -4/+0 |
| * | Moved the definitions of AST_PENDING and AST_RESCHED to the correct place. | Bruce Evans | 2000-10-12 | 1 | -0/+6 |
| * | Work around a bug by adding struct tags. gcc-2.95 apparently gets the | Bruce Evans | 2000-10-06 | 1 | -2/+2 |
| * | - Change fast interrupts on x86 to push a full interrupt frame and to | John Baldwin | 2000-10-06 | 4 | -25/+1 |
| * | currentldt is now a "special" global-data variable, and as such, there | John Baldwin | 2000-10-06 | 1 | -3/+0 |
| * | Interrupt frames don't include the saved cpl anymore since cpl is dead. | John Baldwin | 2000-10-06 | 1 | -2/+0 |
| * | - Heavyweight interrupt threads on the alpha for device I/O interrupts. | John Baldwin | 2000-10-05 | 1 | -0/+1 |
| * | Replace loadandclear() with atomic_readandclear_int(). | John Baldwin | 2000-10-05 | 1 | -11/+0 |
| * | Add atomic_readandclear_int and atomic_readandclear_long. | John Baldwin | 2000-10-05 | 1 | -2/+34 |
| * | Make the gd_currentldt member in struct globaldata unconditional so | John Baldwin | 2000-10-05 | 1 | -3/+1 |
| * | Reduce userland namespace polution. | Jason Evans | 2000-10-04 | 1 | -1/+4 |
| * | Move the i386 PCI attachment code out of i386/isa back into i386/pci. | Mike Smith | 2000-10-02 | 1 | -0/+5 |
| * | More updates to the ACPI code: | Mike Smith | 2000-09-30 | 1 | -0/+13 |
| * | This is the first snapshot of the FreeBSD/ia64 kernel. This kernel will | Doug Rabson | 2000-09-29 | 1 | -0/+80 |
| * | First shot at identifying the Pentum 4 acording to our reading of the | Peter Wemm | 2000-09-29 | 1 | -2/+3 |
| * | Get out the roto-rooter and clean up the abuse of nexus ivars by the | Peter Wemm | 2000-09-28 | 1 | -0/+55 |
| * | Fix the assmebly mutex macros to handle saving/restoring interrupt state | John Baldwin | 2000-09-24 | 1 | -7/+21 |
| * | Move MAXCPU from machine/smp.h to machine/param.h to fix breakage | Paul Saab | 2000-09-23 | 1 | -10/+0 |
| * | #include <sys/proc.h> in order to get curproc. This seems to be the lesser | Jason Evans | 2000-09-23 | 1 | -3/+2 |
| * | Remove the NCPU, NAPIC, NBUS, NINTR config options. Make NAPIC, | Paul Saab | 2000-09-22 | 2 | -48/+65 |
| * | Teach MTX_EXIT_RECURSE that the recursion count is a 32-bit integer, | John Baldwin | 2000-09-22 | 1 | -3/+3 |
| * | Add a couple of debug register helper functions to assist in setting | Brian S. Dean | 2000-09-21 | 2 | -0/+10 |
| * | Make LINT compile. | Poul-Henning Kamp | 2000-09-16 | 1 | -1/+1 |
| * | Remove the mtx_t, witness_t, and witness_blessed_t types. Instead, just | John Baldwin | 2000-09-14 | 1 | -25/+23 |
| * | Clean up process accounting some more. Unfortunately, it is still not | John Baldwin | 2000-09-12 | 1 | -18/+14 |
| * | When doing statistics for statclock on other CPU's, use the other CPUs' | John Baldwin | 2000-09-11 | 1 | -3/+3 |
| * | Style cleanups. No functional changes. | Jason Evans | 2000-09-09 | 1 | -42/+37 |
| * | Add file and line arguments to WITNESS_ENTER() and WITNESS_EXIT, since | Jason Evans | 2000-09-09 | 1 | -14/+14 |
| * | Rename mtx_enter(), mtx_try_enter(), and mtx_exit() and wrap them with cpp | Jason Evans | 2000-09-08 | 1 | -68/+80 |
| * | Remove an unneeded extern declaration of cp_time. | John Baldwin | 2000-09-08 | 1 | -2/+0 |
| * | Really fix USER_LDT. (Don't use currentldt as an L-value.) | Jake Burkholder | 2000-09-08 | 1 | -1/+1 |
| * | Test for both SMP and I386_CPU being set before generating an error. | John Baldwin | 2000-09-07 | 1 | -1/+1 |
| * | Major update to the way synchronization is done in the kernel. Highlights | Jason Evans | 2000-09-07 | 7 | -74/+910 |
| * | Introduce atomic_cmpset_int() and atomic_cmpset_long() from SMPng a | Poul-Henning Kamp | 2000-09-06 | 1 | -0/+72 |
| * | Increase the default NAPIC from 1 to 2 as a bandaid until we allocate | Mike Smith | 2000-08-18 | 1 | -1/+1 |
| * | Prepare for a cleanup of pmap module API pollution introduced by the | Tor Egge | 2000-08-16 | 1 | -0/+1 |
| * | Clean up some low level bootstrap code: | Peter Wemm | 2000-08-11 | 3 | -12/+3 |
