| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2014-12-01 | Remove an unused gdt structure definition. | Mike Larkin | |
| 2014-12-01 | Remove an old and superfluous comment, and fix some KNF style issues in | Mike Larkin | |
| some other comments. | |||
| 2014-11-30 | SPARC T4 and later have a pause instruction to voluntarily pause a virtual | Mark Kettenis | |
| processor in order to give other strands a chance to run. Use it in __mp_lock_spin_hook() to avoid wasting CPU cycles if we're waiting for the kernel or scheduler locks. This is instruction is patched in, just like we already do for the sleep instruction on SPARC64 VI processors. We look at the hwcap-list property of the cpu nodes in the machine description to decide whether the pause instruction is available. | |||
| 2014-11-30 | Unbreak zzz/ZZZ resume - ensure NX is re-enabled on resume (if enabled). | Mike Larkin | |
| Also transformed some complicated code that was setting EFER into something much simpler. Tested on a variety of real machines, emulators/VMs, MP, UP, zzz, ZZZ, NX on, NX off ... ok deraadt@ | |||
| 2014-11-30 | Mask out EFER_LMA when restoring saved EFER on zzz/ZZZ resume as it's a | Mike Larkin | |
| read only bit. Also fix some comments describing EFER bits that were obviously wrong. ok deraadt@ | |||
| 2014-11-30 | rxr ioctl handling. | Brad Smith | |
| 2014-11-27 | Missing comparison caused NX to always be enabled during boot, even on CPUs | Mike Larkin | |
| that may have had it disabled in BIOS. ok deraadt@ | |||
| 2014-11-26 | Introduce softraid boot support to sparc64 ofwboot. | Stefan Sperling | |
| sparc64 machines should now be able to boot from softraid raid1 and crypto. Tested on sun blade100, sunfire v240, and sun t1000 ldom guests. Lots of encouragement from deraadt@ | |||
| 2014-11-26 | Add a new ELF segment .openbsd.bootdata to the sparc64 kernel. | Stefan Sperling | |
| This can be used to pass boot parameters to the kernel which can't be passed safely via the Open Firmware interface, such as softraid volume IDs and keys. The kernel already reads the arguments if available but ofwboot won't provide them until further changes are committed there. With support from deraadt, kettenis and matthew. ok deraadt@ | |||
| 2014-11-26 | Add softraid support to sparc64 first-level boot blocks. | Stefan Sperling | |
| Written with use of the forth by jedi jsing@ For now this only supports 'a' RAID partitions, no other partition letters. | |||
| 2014-11-26 | Add softraid boot support code files ported by me to sparc64 from amd64. | Stefan Sperling | |
| Not hooked to the build yet. ok kettenis@ | |||
| 2014-11-26 | Copy some Open Firmware interface functions from the kernel to ofwboot. | Stefan Sperling | |
| Will be needed soon. ok kettenis@ | |||
| 2014-11-26 | Hook ofwbootfd to the build. | Stefan Sperling | |
| 2014-11-26 | Add a separate ofwboot build for the sparc64 floppies. | Stefan Sperling | |
| 2014-11-25 | A few reliability fixes: | Miod Vallat | |
| - protect pv lists with splvm. - try to return as soon as possible from pmap_rmproc(). - correctly maintain wired pages statistics. - invoke pmap_remove_pcb() at splsched to avoid racing the scheduler. | |||
| 2014-11-25 | Brag about what OS we are running on the LCD panel, if there is one. | Miod Vallat | |
| 2014-11-25 | Speed up page zeroing by using a loop of dcbz/dcbzl instead of bzero(). | Martin Pieuchot | |
| While here, use the direct map for pmap_copy_page() and remove the now unused stolen page addresses. No objection from the usual suspects, "it works, commit" deraadt@ | |||
| 2014-11-25 | disable qla. there is a boot-time race | Theo de Raadt | |
| 2014-11-24 | Actually, the previous change didn't quite work on the SPARC T5-2. But if | Mark Kettenis | |
| we change the order of the xxx_setvalidand xxx_setstate hypervisor calls things actually work. | |||
| 2014-11-24 | Add missing pci_msiq_setstate and pci_msi_setstate calls. Makes MSI work | Mark Kettenis | |
| on the SPARC T5-2. Make error messages a bit more consistent while there. | |||
| 2014-11-24 | Do not attempt to dereference NULL pointers when encountering an empty SCSI | Miod Vallat | |
| bus on a dual-scsi controller abstraction, while walking the KL component tree. | |||
| 2014-11-24 | The hypervisor on a SPARC T5-2 needs stricter alignment of the MSI message | Mark Kettenis | |
| queue. | |||
| 2014-11-24 | If we use the vintr cookie to point at the interrupt handler struct, we | Mark Kettenis | |
| shouldn't link it into the table. | |||
| 2014-11-24 | Use the pointer to "struct intrhand" as the vintr cookie. | Mark Kettenis | |
| 2014-11-24 | On sun4v, interpret the first data word of a device interrupt message as a | Mark Kettenis | |
| pointer to "struct intrhand" if it is larger than MAXINTNUM. To be used with the cookie based "vintr" scheme. | |||
| 2014-11-24 | use the correct capitalization for Realtek. | Brad Smith | |
| 2014-11-23 | enable xhci(4) in ramdisks | Jonathan Gray | |
| ok deraadt@ | |||
| 2014-11-23 | Make sure the direction bit is cleared on entry to the kernel by syscall | Philip Guenther | |
| or interrupt, as specified by the ABIs. Our current gcc doesn't assume this bit of the ABI, but future compilers may and some ASM can be simplified by assuming it. in snaps for a bit | |||
| 2014-11-22 | the per-cpu storage page does not need to be executable in this | Theo de Raadt | |
| fashion. Anyways, we don't do MP here yet, so #ifdef it all away. with miod | |||
| 2014-11-22 | Be more careful draining the error register and reenabling traps in the sun4m | Miod Vallat | |
| NMI handler; apparently ROSS modules depend upon this to behave correctly. From NetBSD | |||
| 2014-11-22 | Driver for the ECC memory controller found on some sun4m (ss20), so that | Miod Vallat | |
| recoverable memory errors no longer panic the kernel. From NetBSD but the manual page. | |||
| 2014-11-22 | Previous diff changed the location of the ACPI S3/S4 trampoline, which has | Mike Larkin | |
| a common #defined location for both i386 and amd64. This diff fixes i386 to match. Also fix a tab/space issue in amd64 hibernate_var.h discussed with deraadt | |||
| 2014-11-22 | { on seperate line at start of function or I go crazy | Theo de Raadt | |
| 2014-11-22 | Do not use PROT_MASK as a non-mask | Theo de Raadt | |
| 2014-11-22 | Split the MP trampoline into two pages, one for code and one for data/stack | Mike Larkin | |
| and then protect the code page as RX and the data/stack page as RW (NX). ok deraadt@ | |||
| 2014-11-22 | Switch to using MI installboot, after a conversation with stsp. He will | Theo de Raadt | |
| need this soon. install tested by jsg | |||
| 2014-11-22 | spell PROT_EXEC correctly | Theo de Raadt | |
| 2014-11-22 | mop up a barely started project... getting in the way of grepping the tree! | Theo de Raadt | |
| 2014-11-21 | We can map 'fuse ROM' area with read only. | Kenji Aoyama | |
| 2014-11-21 | remove unused #defines and labels. | Mike Larkin | |
| ok deraadt, guenther | |||
| 2014-11-21 | Since we enable NX in locore now, no need to do it again later in the | Mike Larkin | |
| bootstrap sequence ok deraadt@ | |||
| 2014-11-21 | Add NX (if available) to the mid-level page tables (PDE, etc). | Mike Larkin | |
| ok deraadt@ | |||
| 2014-11-20 | Fix comment. | Kenji Aoyama | |
| 2014-11-20 | When removing the identity mapping in low memory used during bootstrap, | Mike Larkin | |
| there is no reason to keep the NX bit around on null PTEs (PTEs that have been removed). | |||
| 2014-11-20 | Don't attempt to clear/disable %tick_cmpr. This register is no longer present | Mark Kettenis | |
| on newer sun4v implementations (possibly SPARC T3 and later, definitely not there on SPARC T5) and it should not be necessary on older systems. Tested by dlg@ | |||
| 2014-11-20 | Disentagle bcopy/memcpy/memmove. | Theo de Raadt | |
| The situation was: memcpy swaps registers and drops into bcopy, which does the overlap check, before getting around to business. But memcpy is not supposed to handle overlapped arguments special, so we don't need all this complex register swapping and tests up front -- when memcpy is the most common operation. Refactor all these upside downupside down. Survived multiple make builds, so probably good, and faster. ok dlg | |||
| 2014-11-20 | Move previous PTE permission fixup code into locore, and fixup some more | Mike Larkin | |
| ranges while we're there. ok deraadt@, tested by many and in snaps | |||
| 2014-11-20 | Revert the previous changes that enabled various PTE permissions fixups | Mike Larkin | |
| in pmap.c - equivalent code will shortly be committed to locore.S that accomplishes the same thing. ok deraadt@ | |||
| 2014-11-19 | Only keep {recv,send}udp() in the pxe-specific code, the remainder of what | Miod Vallat | |
| used to be local net.c (renamed to pxe_udp.c) can now be obtained from libsa's net.c rather than being duplicated. No functional change. | |||
| 2014-11-19 | Remove some unused i386 pmap functions. Also fix two typos in comments | Mike Larkin | |
| while I had the hood open. ok deraadt@, mpi@ | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |