| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2018-05-11 | Compile the EFI bootloader with -mfloat-abi=soft to prevent the compiler from | Mark Kettenis | |
| using FPU instructions. Makes the bootloader work again after we switched clang to use -mfloat-abi=softfp by default. ok jsg@, patrick@ | |||
| 2018-05-09 | Fix a linking error in the IP26 kernel. This is a regression caused | Visa Hankala | |
| by the addition of db_save_stack_trace(). sys/arch/mips64/mips64/trap.c:(.text+0x410): undefined reference to `u_intr' Pointed out by deraadt@ | |||
| 2018-05-09 | Avoid locking two vm_pages simultaneously in pmap_enter(). To achieve | Visa Hankala | |
| that, remove any existing mapping before locking the new vm_page. This fix prevents a deadlock. Issue pointed out by witness(4). | |||
| 2018-05-09 | vm_page lookup does not need locking. Hoist it outside the pmap lock. | Visa Hankala | |
| 2018-05-09 | Implement stack trace saving on mips64. | Visa Hankala | |
| OK mpi@ | |||
| 2018-05-08 | Print xHCI revision in dmesg. | Martin Pieuchot | |
| ok patrick@, kettenis@ | |||
| 2018-05-07 | Use speed from device tree for serial console on armv7 too. | Mark Kettenis | |
| ok visa@, patrick@ | |||
| 2018-05-06 | Use speed from device tree for serial console. Makes it possible to boot | Mark Kettenis | |
| and install systems where the firmware uses a non-standard speed. This is important for various boards that use a Rockchip SoC where the vendor thinks using 1500000 bps as the serial console speed is a good idea (it isn't). ok deraadt@, visa@, patrick@ | |||
| 2018-05-05 | Enable octcrypto(4). | Visa Hankala | |
| 2018-05-04 | Add a common rules file for ofw sources to help keep the configurations | Visa Hankala | |
| of fdt-enabled platforms in sync. OK deraadt@ | |||
| 2018-05-04 | Fix WITNESS build on arm64. | Visa Hankala | |
| OK guenther@ | |||
| 2018-05-04 | Add stack trace saving for arm64. | Visa Hankala | |
| OK guenther@ | |||
| 2018-05-04 | Initialize st_count in db_save_stack_trace(). | Visa Hankala | |
| OK guenther@ | |||
| 2018-05-04 | Leave out an unnecessary symbol lookup. | Visa Hankala | |
| OK guenther@ | |||
| 2018-05-03 | Add support for SMCCC 1.1 which provides proper support for the firmware-based | Mark Kettenis | |
| workaround for branch target injection attacks (CVE 2017-5715). ok patrick@, jsg@ | |||
| 2018-05-02 | Fix build. The power domain API is needed by r1.9 of xhci_fdt.c. | Visa Hankala | |
| 2018-05-02 | Add imxpd(4) which is a power domain controller driver that | Patrick Wildt | |
| essentially calls into ATF to make it supply power. ok kettenis@ | |||
| 2018-05-02 | Implement a power domain framework to turn on/off so-called power | Patrick Wildt | |
| domains. This mechanism is used by the newer i.MX8M SoCs so that drivers can call into ATF to supply power to e.g. a USB port. ok kettenis@ | |||
| 2018-05-02 | Remove proc from the parameters of vn_lock(). The parameter is | Visa Hankala | |
| unnecessary because curproc always does the locking. OK mpi@ | |||
| 2018-05-01 | vmmci(4) depends on pvbus(4) being available, unbreak | Peter Hessler | |
| from mlarkin@ | |||
| 2018-04-30 | vmd(8): unbreak i386 | Mike Larkin | |
| 2018-04-28 | replace add_*_randomness with enqueue_randomness() | Jasper Lievisse Adriaanse | |
| this gets rid of the source annotation which doesn't really add anything other than adding complexitiy. randomess is generally good enough that the few extra bits that the source type would add are not worth it. ok mikeb@ deraadt@ | |||
| 2018-04-28 | Clean up the parameters of VOP_LOCK() and VOP_UNLOCK(). It is always | Visa Hankala | |
| curproc that does the locking or unlocking, so the proc parameter is pointless and can be dropped. OK mpi@, deraadt@ | |||
| 2018-04-27 | rasops: implement scrollback | Joshua Stein | |
| activate it for efifb and inteldrm ok kettenis | |||
| 2018-04-27 | use nitems() | Jasper Lievisse Adriaanse | |
| verified md5 sum on amd64 ok mlarkin@ | |||
| 2018-04-27 | -fretpoline for non-RAMDISK kernels. Skylake product needs it, and | Theo de Raadt | |
| therefore all other x86 cpus must pay the performance price of the hazard workaround until Skylake disappears from the ecosystem eventually like 486. This returns your cpu's performance to pre-inflated performance. ok mlarkin guenther | |||
| 2018-04-27 | vmm(4): pass through ELCRx ports to vmd(8) | Mike Larkin | |
| These ports are used for Edge/Level control on the legacy PIC and will be needed for a subsequent commit. | |||
| 2018-04-27 | Fix the prologue detection to match amd64 opcodes, not i386 opcodes | Philip Guenther | |
| ok mlarkin@ | |||
| 2018-04-26 | vmm(4): passthrough port 0x61 to vmd(8) | Mike Larkin | |
| ok guenther | |||
| 2018-04-26 | Oups, we only need vmmci(4) for the ramdisk, not the whole vmm(4). | Peter Hessler | |
| noticed by deraadt@ and mlarkin@ | |||
| 2018-04-26 | enable vmmci(4) on the ramdisks, so bsd.rd plays nicer when it powers off | Peter Hessler | |
| OK mlarkin@ | |||
| 2018-04-26 | vmm(4): Rearrange some interrupt window exiting code | Mike Larkin | |
| Simplify some code and reduce the number of places we are handling VINTR exits. ok brynet | |||
| 2018-04-26 | Prefer leaq+%rip-relative over movabsq | Philip Guenther | |
| xrstor_resume must not have profile prologue, so use NENTRY Don't use _C_LABEL() with some pure-ASM labels | |||
| 2018-04-26 | Reorder trapframe/intrframe to put %rbp next to %rip and make it | Philip Guenther | |
| behave like a real call frame, thus vastly simplifying the ddb back trace logic. based on whinging from deraadt@ ok jasper@ mpi@ phessler@ | |||
| 2018-04-26 | vmm(4): remove some unnecessary kernel lock code from the SVM guest loop | Mike Larkin | |
| that is not needed; this code deals with delaying the relocking of the kernel lock until after interrupts are processed during external interrupt exiting, but this is handled differently on SVM. External interrupts are automatically handled by the CPU as soon as stgi() is performed after exit. (The original code came from the VMX/Intel guest loop.) ok guenther@ | |||
| 2018-04-26 | vmm(4): ensure SVM_INTERCEPT_INTR is always enabled before entering | Mike Larkin | |
| guest VM. | |||
| 2018-04-26 | Drop into ddb(4) if pmap_tlb_shoot*() take too much time in MP_LOCKDEBUG | Martin Pieuchot | |
| kernels. While here sync all MP_LOCKDEBUG/while loops. ok mlarkin@, visa@ | |||
| 2018-04-25 | Simplify the handling of faults in iretq and xrstor by doing the | Philip Guenther | |
| handling purely in ASM instead of fixing up the state to call into C...all in order to return back into a chunk of custom ASM which was longer than the direct solution! ok mlarkin@ | |||
| 2018-04-25 | vmm(4): clarify whose FPU state we are saving (change in a comment) | Mike Larkin | |
| no code change | |||
| 2018-04-25 | Add a way to reattach efifb(4). This will be used when radeondrm(4) | Jonathan Gray | |
| attaches and claims glass console but after root is mounted if firmware is not found efifb gets reattached to have a useable glass console. | |||
| 2018-04-24 | Move code into inline functions to hide #ifdefs and improve readability. | Philip Guenther | |
| Constipate a couple variables. Delete IDTVEC() macro and code handling 32bit-only traps (panic instead). ok mlarkin@ | |||
| 2018-04-24 | vmm(4): fix lockups on SVM hosts when booting Linux guests | Mike Larkin | |
| The previous version of the code used "disable_intr" as a protection around the FPU state saving routine, which left host RFLAGS.IF=0 when entering the guest. If the guest ended up in a non-exitable condition, the machine could hang due to missed TLB flush IPIs (and eventual kernel lock contention and starvation). tested by brynet@ ok guenther@ | |||
| 2018-04-24 | <machine/i82489reg.h> isn't used here | Philip Guenther | |
| ok mlarkin@ | |||
| 2018-04-24 | account for FPU context switch (uvm statistic only) | Mike Larkin | |
| 2018-04-24 | make the check for XSAVE the same as it is later in the file, for | Mike Larkin | |
| consistency. ok guenther note - there is no equivalent diff here for i386 since XSAVE is not used there. | |||
| 2018-04-24 | Enable gpioiic(4) and gpioow(4). From Artturi Alm. | Mark Kettenis | |
| 2018-04-24 | Enable em(4) on loongson. | Visa Hankala | |
| 2018-04-24 | Do not lock the kernel when expanding the working buffer. | Visa Hankala | |
| malloc(9) and free(9) are MP-safe nowadays. OK mpi@ | |||
| 2018-04-23 | Move efi_cleanup() to before the mem_pass() since efi_cleanup() might | YASUOKA Masahiko | |
| change the memory map. diff from Ryan Lennox ok kettenis | |||
| 2018-04-22 | Add reference counting that prevents freeing of busy aesni sessions. | Visa Hankala | |
| The early freeing has become possible because aesni_process() and crypto_freesession() are no longer serialized by the kernel lock. The flaw has caused kernel panics with IPsec traffic. Issue seen by some, fix tested by mabi on bugs@ OK mikeb@, mpi@ | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |