| Commit message (Expand) | Author | Age | Files | Lines |
* | Unify SPR defines formatting, no funtional changes. | raj | 2012-05-26 | 1 | -88/+87 |
* | Update HID defines for E500mc and E5500 CPU cores. | raj | 2012-05-25 | 1 | -0/+53 |
* | Fix physical address type to vm_paddr_t also for powerpc64. | raj | 2012-05-25 | 1 | -11/+11 |
* | Missing vm_paddr_t bits which should have been part of r235936. | raj | 2012-05-25 | 1 | -4/+4 |
* | Add a missing " to get closer to compiling. | bz | 2012-05-24 | 1 | -1/+1 |
* | Atomic operation acquire barriers also need to be isync on 64-bit systems. | nwhitehorn | 2012-05-24 | 1 | -1/+1 |
* | Revert isync for ILP32 to sync as per my original change that I discussed | marcel | 2012-05-24 | 1 | -4/+4 |
* | MFp4 bz_ipv6_fast: | bz | 2012-05-24 | 1 | -0/+4 |
* | A few improvements: | marcel | 2012-05-24 | 2 | -64/+179 |
* | Fix physical address type to vm_paddr_t. | raj | 2012-05-24 | 5 | -31/+31 |
* | Remove Semihakf-ism. DEBUG is a kernel configuration option. It | marcel | 2012-05-24 | 1 | -3/+0 |
* | Just return if the size of the window is 0. This can happen when the | marcel | 2012-05-24 | 1 | -1/+7 |
* | Either the I/O port range or the memory mapped I/O range may not be | marcel | 2012-05-24 | 1 | -2/+7 |
* | o Rename kernload_ap to bp_kernelload. This to introduce a common prefix | marcel | 2012-05-24 | 3 | -19/+39 |
* | Fix the memory barriers for CPUs that do not like lwsync and wedge or cause | marcel | 2012-05-24 | 1 | -12/+22 |
* | Replace the list of PVOs owned by each PMAP with an RB tree. This simplifies | nwhitehorn | 2012-05-20 | 4 | -178/+76 |
* | Fix final bugs in memory barriers on PowerPC: | nwhitehorn | 2012-05-04 | 4 | -8/+10 |
* | Add a convenience macro for the returns_twice attribute, and apply it to | dim | 2012-04-29 | 1 | -1/+1 |
* | Fix build on 32-bit systems. | nwhitehorn | 2012-04-28 | 1 | -1/+1 |
* | After switching mutexes to use lwsync, they no longer provide sufficient | nwhitehorn | 2012-04-28 | 2 | -30/+19 |
* | Switch the default I/O memory barrier to eieio, as it should be. This | nwhitehorn | 2012-04-24 | 1 | -5/+1 |
* | Revert r234581 for this file. The lockless SLB tree code does in fact need | nwhitehorn | 2012-04-24 | 1 | -2/+2 |
* | Fix copy-and-paste error in r230400. | nwhitehorn | 2012-04-23 | 1 | -1/+1 |
* | Fix missing header for powerpc_iomb(). | nwhitehorn | 2012-04-23 | 1 | -0/+1 |
* | Provide a clearer split between read/write and acquire/release barriers. | nwhitehorn | 2012-04-22 | 1 | -19/+24 |
* | Correctly specify assembler constrains for synchronization instructions. | nwhitehorn | 2012-04-22 | 1 | -3/+3 |
* | Clarify what we are doing in r234583 a little better: eieio and isync do | nwhitehorn | 2012-04-22 | 1 | -15/+23 |
* | On non-64-bit systems (which generally don't have lwsync), use eieio and | nwhitehorn | 2012-04-22 | 1 | -0/+5 |
* | Use lwsync to provide memory barriers on systems that support it instead | nwhitehorn | 2012-04-22 | 2 | -22/+20 |
* | Remove dead code. The routines in atomic.S did not work properly anyway, and | nwhitehorn | 2012-04-22 | 2 | -147/+0 |
* | Replace eieio; sync for creating bus-space memory barriers with sync. | nwhitehorn | 2012-04-22 | 3 | -61/+75 |
* | Avoid a lock order reversal in pmap_extract_and_hold() from relocking | nwhitehorn | 2012-04-22 | 1 | -1/+32 |
* | Organize some members of ucontext_t in the same order they are in the | nwhitehorn | 2012-04-21 | 1 | -3/+3 |
* | Make sure all pending operations have completed on the existing thread | nwhitehorn | 2012-04-20 | 2 | -0/+2 |
* | We don't need kcopy() in any of the remaining places it is used, so | nwhitehorn | 2012-04-11 | 4 | -34/+3 |
* | Only manipulate the PGA_EXECUTABLE flag on managed pages. This is a proxy | nwhitehorn | 2012-04-11 | 1 | -14/+10 |
* | Fix error in r233949. Synchronizing icaches on uncacheable pages turns out | nwhitehorn | 2012-04-11 | 1 | -2/+4 |
* | Do not restore the register holding the TLS pointer when doing various | nwhitehorn | 2012-04-11 | 1 | -2/+12 |
* | Execute an initial ptesync if and only if the PTE is actually being | nwhitehorn | 2012-04-06 | 1 | -14/+7 |
* | Substantially reduce the scope of the locks held in pmap_enter(), which | nwhitehorn | 2012-04-06 | 1 | -34/+8 |
* | Reduce the frequency that the PowerPC/AIM pmaps invalidate instruction | nwhitehorn | 2012-04-06 | 3 | -57/+29 |
* | Give the kernel pmap lock a different name than user pmap locks. It has | nwhitehorn | 2012-04-06 | 1 | -2/+3 |
* | - Rename VM_MEMATTR_UNCACHED to VM_MEMATTR_WEAK_UNCACHEABLE on x86 to | jhb | 2012-03-29 | 1 | -1/+0 |
* | Allow multiple inclusion of trap.h. This has always been broken, but | nwhitehorn | 2012-03-29 | 3 | -4/+10 |
* | Add software PMC support. | fabient | 2012-03-28 | 1 | -1/+2 |
* | More PMAP performance improvements: skip 256 MB segments entirely if they | nwhitehorn | 2012-03-28 | 2 | -11/+26 |
* | Make sure to call vm_page_dirty() before the pmap lock is released to | nwhitehorn | 2012-03-27 | 1 | -2/+2 |
* | More PMAP concurrency improvements: replace the table lock and (almost) all | nwhitehorn | 2012-03-27 | 1 | -86/+100 |
* | More PMAP performance improvements: on powerpc64, when TLBIE can be run | nwhitehorn | 2012-03-25 | 1 | -4/+11 |
* | Only call vm_page_dirty() on pages that are writable in order not to | nwhitehorn | 2012-03-24 | 1 | -4/+12 |