| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2014-12-03 | stale header: this code hasn't used hash.h in a while | Ted Unangst | |
| 2014-12-02 | In sys_socket() and doaccept(), gotta skip setting the close-on-exec | Philip Guenther | |
| flag if the falloc() fails reported by Carlin Bingham (cb (at) viennan.net) ok deraadt@ tedu@ | |||
| 2014-12-01 | clist's chained, and could get fairly long which is why there was a | Theo de Raadt | |
| dynamic clamping mechanism. In the (new? has it already been 20 years?) world of tty ring buffers, c_cn is the maximum... There could be some ugly limit elsewhere, which may cause a deadlock (dug as deep as my patience allows), so please report any sort of new console or xterm issues that show up, such as tty lockups or high cpu utilization.. | |||
| 2014-11-21 | simple lock is long dead | Ted Unangst | |
| 2014-11-19 | panic should be reserved for unrecoverable errors, not things we merely | Ted Unangst | |
| hope don't happen. also, always check for errors. ok deraadt | |||
| 2014-11-19 | missed a file when removing KERN_VNODE. | Ted Unangst | |
| 2014-11-19 | delete the KERN_VNODE sysctl. it fails to provide any isolation from the | Ted Unangst | |
| kernel struct vnode defintion, and the only consumer (pstat) still needs kvm to read much of the required information. no great loss to always use kvm until there's a better replacement interface. ok deraadt millert uebayasi | |||
| 2014-11-18 | Nuke yet more obvious #include duplications. | Kenneth R Westerback | |
| ok miod@ | |||
| 2014-11-18 | Disable the page zeroing thread on MULTIPROCESSOR mips64 kernels as well. | Miod Vallat | |
| Regression spotted by tobiasu@. XXX I wonder if the page zeroing thread shouldn't perform explicit XXX pmap_update(pmap_kernel()) calls after each page zeroing... but that XXX might not be enough. | |||
| 2014-11-18 | unifdef SETUIDSCRIPT and FDSCRIPT. we always use them. ok deraadt | Ted Unangst | |
| 2014-11-18 | move arc4random prototype to systm.h. more appropriate for most code | Ted Unangst | |
| to include that than rdnvar.h. ok deraadt dlg | |||
| 2014-11-18 | include unistd.h instead of picking it up accidentally. noted by jsg | Ted Unangst | |
| 2014-11-16 | Delete a comment that was obsoleted by the switch to UVM | Philip Guenther | |
| vmcmd_map_zero() just returns the result of uvm_map(), so make it a tail call | |||
| 2014-11-16 | Replace a plethora of historical protection options with just | Theo de Raadt | |
| PROT_NONE, PROT_READ, PROT_WRITE, and PROT_EXEC from mman.h. PROT_MASK is introduced as the one true way of extracting those bits. Remove UVM_ADV_* wrapper, using the standard names. ok doug guenther kettenis | |||
| 2014-11-16 | Rework the __thrsigdivert (aka sigwait()) handling: instead of interfering | Philip Guenther | |
| in ptsignal(), which broke ptrace() in various circumstances, act more like sigsuspend() by updating the signal mask and picking off waited for signals when one occurs. Don't always restart when an unwaited-for-but-handled signal occurs, as that screws with both timeout calculation and cancellation. main problem noted by jmatthew@ ok tedu@ | |||
| 2014-11-15 | hoist the slowdown handling up to the pool_do_get callers. this lets | David Gwynne | |
| us handle the slowdown where we already give up pr_mtx and gets rid of an ugly goto. ok tedu@ who i think has more tweaks coming | |||
| 2014-11-14 | prefer sizeof(*ptr) to sizeof(struct) for malloc and free | Ted Unangst | |
| 2014-11-14 | move the slowdown back up. it needs to take place after the allocated page | Ted Unangst | |
| has been added to the pool, else it doesn't help because the memory isn't available. lost in locking rework. tested blambert sthen | |||
| 2014-11-12 | take a few more ticks to actually throttle down. hopefully helps in | Ted Unangst | |
| situations where e.g. web browsing is cpu intense but intermittently idle. subject to further refinement and tuning. | |||
| 2014-11-10 | Grab the pool mutex in sysctl_dopool(), but only for pools for which | Mark Kettenis | |
| pool_setipl(9) has been called. This avoids the panic introduced in rev 1.139 (which was subsequently backed out) while still effectively guaranteeing a consistent snapshot. Pools used from interrupt handlers should use the appropriate pool IPL. ok dlg@, deraadt@ | |||
| 2014-11-08 | some systems install sensors where numbering introduces a gap which | Jasper Lievisse Adriaanse | |
| falsely triggered the added panic. revert the panic for now while a better solution is being worked on as reported by Bjorn Ketelaars on misc@ via jsing@ | |||
| 2014-11-06 | printf debugging worked! figured out the bug (in free()) just after | Ted Unangst | |
| committing the printfs. revert. | |||
| 2014-11-06 | need to calculate correct size before doing the free checks. the biggest | Ted Unangst | |
| malloc bucket isn't precise, it can have anything in it. should fix recent panics. sorry for inconvenience. ok deraadt millert | |||
| 2014-11-06 | Make better use of the value of psectionslen instead of recomputing | Todd C. Miller | |
| it multiple times. Also remove an unused variable. OK deraadt@ tedu@ | |||
| 2014-11-06 | resort to printf debugging hints until we can figure this out. ok deraadt | Ted Unangst | |
| 2014-11-06 | let ramdisks compile | Theo de Raadt | |
| 2014-11-05 | need to move lock up to prevent more than one malloc. ok guenther | Ted Unangst | |
| 2014-11-05 | don't use loop variable (i) for not loop things. use a new var. | Ted Unangst | |
| 2014-11-05 | use memname to print string of type. stolen from deraadt | Ted Unangst | |
| 2014-11-05 | also print type when free size is wrong | Ted Unangst | |
| 2014-11-05 | No reason to have things like the hibernate allocation area and chunk | Mike Larkin | |
| ordering regions mapped executable, so remove those permissions. ok deraadt@ | |||
| 2014-11-05 | change the mbuf pool wait channel name from mbpl to mbufpl. "mb" | David Gwynne | |
| isnt descriptive enough for me. ok deraadt@ | |||
| 2014-11-04 | panic when a sensordev gets installed twice, instead of blowing up amazingly | Jasper Lievisse Adriaanse | |
| later on ok deraadt@ | |||
| 2014-11-03 | include sys/unistd.h where needed instead of indirect reliance. ok jsg | Ted Unangst | |
| 2014-11-03 | correct test logic. ok guenther | Ted Unangst | |
| 2014-11-03 | Put the socket splicing fields into a seperate struct sosplice that | Alexander Bluhm | |
| gets only allocated when needed. This way struct socket shrinks from 472 to 392 bytes on amd64. When splicing gets active, another 88 bytes are allocated for struct sosplice. OK dlg@ | |||
| 2014-11-03 | deobfuscate by pulling le conversions up. use mallocarray. | Ted Unangst | |
| 2014-11-03 | pass size argument to free() | Theo de Raadt | |
| ok doug tedu | |||
| 2014-11-02 | Unmap the hibernate hiballoc page after we are done with it. | Mike Larkin | |
| ok deraadt, kettenis | |||
| 2014-11-02 | tweak free panic messages too | Ted Unangst | |
| 2014-11-02 | tweak panic messages for consistency | Ted Unangst | |
| 2014-11-01 | add a few sizes to free | Ted Unangst | |
| 2014-11-01 | remove color support. discussed with dlg and mikeb | Ted Unangst | |
| 2014-10-26 | Now that threads hold their own reference to their ucreds, sys_ktrace() | Philip Guenther | |
| doesn't need to hold an extra reference during the call ok deraadt@ millert@ | |||
| 2014-10-25 | Do not launch the page zeroing thread on MULTIPROCESSOR m88k systems. This | Miod Vallat | |
| causes a deadlock between reaper and zerothread I am currently investigating. | |||
| 2014-10-22 | Extend domknodat() to support S_IFIFO directly and remove domkfifoat(). | Todd C. Miller | |
| Both sys_mkfifo(2) and sys_mkfifoat(2) could be replace by libc wrappers using mknodat(2) at a later date if we desire. OK guenther@ | |||
| 2014-10-22 | Remove some unused/unneeded code. | Mike Larkin | |
| 2014-10-22 | Use the global piglet address variables as sparingly as possible to avoid | Mike Larkin | |
| redundant copies of the same information. No functional change. Also add some comments as to how these globals are used. | |||
| 2014-10-18 | Don't assume that ep_taddr and ep_daddr are page-aligned. It is possible to | Mark Kettenis | |
| construct ELF executables for which ep_daddr ends up not being properly aligned. Sanitize the addresses before setting up the address space for the new executable. Should fix the panic discovered by Alejandro Hernandez. ok miod@ | |||
| 2014-10-17 | cpu_setperf and perflevel must remain exposed, otherwise a bunch of | Theo de Raadt | |
| MD code needs excess #ifndef SMALL_KERNEL | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |