summaryrefslogtreecommitdiffstats
path: root/sys/arm64
Commit message (Expand)AuthorAgeFilesLines
* Rework the exception entry/return functions to make them valid frames to beandrew2015-12-012-36/+44
* Rework the atomic code to reduce the repetition. This merges some of theandrew2015-12-011-586/+204
* Print useful information when we hit a data abort we can't handle. Thisandrew2015-12-011-0/+9
* Shorten conditional branch code.kib2015-11-301-2/+1
* Remove unneeded instructions.kib2015-11-301-2/+0
* Remove sv_prepsyscall, sv_sigsize and sv_sigtbl members of the structkib2015-11-281-3/+0
* Correct arm64 gic_v3 sizeof argumentemaste2015-11-261-1/+1
* Add support for moving the DMAP range. This is needed as some AMD SoCsandrew2015-11-242-9/+30
* Revert r291142.skra2015-11-231-2/+2
* Fix BUS_DMA_MIN_ALLOC_COMP flag logic. When bus_dmamap_t map is beingskra2015-11-211-2/+2
* Avoid a NULL pointer dereference in bounce_bus_dmamap_unload() whenmarius2015-11-211-0/+5
* Avoid a NULL pointer dereference in bounce_bus_dmamap_sync() when themarius2015-11-201-1/+1
* Fix typo in message from arm64 ITS workaroundemaste2015-11-091-1/+1
* Sync arm64 GENERIC whitespace/comments with amd64emaste2015-11-081-4/+4
* arm64: add igb(4) to GENERICemaste2015-11-071-0/+1
* Mark the arm64 nexus devices to be attached early. This allows multipassandrew2015-11-061-4/+4
* Fix an alignment check that is wrong in half the busdma implementations.ian2015-11-021-2/+2
* Make the arm64_cpu driver quiet as it adds no new information.andrew2015-11-021-4/+7
* Mark functions as such. This means we call them directly rather than haveandrew2015-10-271-1/+1
* Fix bus numbering in ThunderX ITS quirkzbb2015-10-251-2/+4
* Add support for unspecified ranges on ThunderX systemzbb2015-10-251-3/+39
* arm64: remove exception instruction length assertionemaste2015-10-231-14/+0
* Remove unclear comment about address truncation in busdma. Add (hopefully mu...jah2015-10-231-1/+0
* Add support for CloudABI on ARM64.ed2015-10-221-0/+162
* Use pmap_quick* functions in arm64 busdma to make bounce buffer synchronizati...jah2015-10-211-45/+56
* arm64: Enable CTF for DTrace supportemaste2015-10-211-0/+1
* Use 4 levels of page tables when enabling the MMU. This will allow us toandrew2015-10-192-6/+76
* Add ThunderX VNIC to arm64/GENERIC kernelzbb2015-10-181-0/+2
* Correctly align the stack. The early csu assumed we passed the aux vectorandrew2015-10-181-1/+7
* Replace build_section_pagetable with build_l1_block_pagetable as it takesandrew2015-10-171-27/+43
* Rename build_block_pagetable to build_l2_block_pagetable in preperationandrew2015-10-171-3/+3
* Add checks for kernel VA accesses to the copyin(9) and relatedkib2015-10-173-1/+52
* Properly set the return value for casueword to 0 upon success.ed2015-10-151-0/+2
* Build changes that allow the modules on arm64.kib2015-10-081-1/+0
* Implement in-kernel relocator for the arm64 module linker.kib2015-10-081-5/+64
* Make the copyright notice in the file to match reality. Use thekib2015-10-081-19/+11
* When trying to execute from a misaligned address raise a SIGBUS with theandrew2015-10-041-0/+5
* When asserting IL bit in exception syndrome register, print the rawkib2015-10-041-1/+1
* Use pmap_load more consistently. While here try to only load the data onceandrew2015-10-011-42/+48
* Add the ENTRY/END entries around the exception handlers.andrew2015-10-011-6/+12
* Exploit r288122 to address a cosmetic issue. Since PV chunk pages don'talc2015-09-261-1/+1
* Remove apostrophe from AP's for consistency (arm64 mp_machdep)emaste2015-09-251-1/+1
* Add pass device to arm64 GENERIC (for smartctl)emaste2015-09-221-0/+3
* Call ast when handling irq from userspace, otherwise we could misskib2015-09-221-2/+8
* Don't restore interrupts when we are about to disable them in the nextandrew2015-09-221-3/+0
* Re-check for new ast after ast was handled. We should not return tokib2015-09-221-4/+4
* arm64 ThunderX PCIe workaround: enumerate only one slot for nowemaste2015-09-221-0/+8
* Do not execute exception handlers with disabled interrupts.kib2015-09-181-16/+25
* Clear exclusive monitors when handling data aborts, the monitors arekib2015-09-182-0/+18
* Don't read the floating-point registers for now. We will need to enable theandrew2015-09-181-0/+3
OpenPOWER on IntegriCloud