summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2014-02-09do a better job in wipe_info: close files and reap children.Marc Espie
also wipe every state variable... fixes reget update_info in fringe cases (this should fix nigel@'s problems...)
2014-02-09use linux style memory allocations in radeonJonathan Gray
2014-02-09use linux style memory allocations in ttmJonathan Gray
2014-02-09add some more linux compat functions for memory allocationJonathan Gray
2014-02-09regenPhilip Guenther
2014-02-09nanosleep() isn't actually NOLOCK safe yet: tsleep() still requires kernelPhilip Guenther
lock for PCATCH and ktrace(CSW) handling ok kettenis@
2014-02-08Only create DBDIR if it does not exist.Antoine Jacoutot
2014-02-08Do not bzero() the available memory in pmap_bootstrap(); allocations inMiod Vallat
pmap_bootstrap explicitely bzero them, and there is no need to clear the remaining memory. ok mpi@
2014-02-08For some reason (lack of testing being my #1 candidate), IP32 PROM version 4.18Miod Vallat
ends up with the CKSEG1 function pointer values being wrong. All of them. Wrong, as in, off-by-0x60. Of course, invoking the advertized function pointers leads to interesting results, from bogus panics to invalid pointer dereferences. Attempt to identify this revision by: - checking that all five function pointer values, as set up by bios_ident(), match the 4.18 bogus values; - instructions at said pointer match the 4.18 values. If the test is positive, then the pointer values are replaced with the correct values. This allows O2 systems with 4.18 PROM to correctly powerdown, reboot and halt. Found the hard way by matthieu@ after a PROM upgrade, verified on a second system by me.
2014-02-08Replace ARC_Call2(), used to invoke the `return-to-PROM' routines with aMiod Vallat
PROM-compatible stack, with a sligthly different ARC_Call_Via_Pointer(), which uses a global function pointer variable instead of computing the address from the ARCBios parameter block; said variable being set in bios_ident(). No functional change... yet. Tested on IP22(24), IP30 and IP32.
2014-02-08Use extent_alloc_with_descr(9) and add a mutex to protect the extent.Mark Kettenis
This should make bus_dmamap_load(9) and bus_dmamap_unload(9) "mpsafe".
2014-02-08Add extent_alloc_with_descr(9) as a wrapper aroundMark Kettenis
extent_alloc_subregion_with_descr(9) mirroring extent_alloc(9). ok jmc@
2014-02-08Create dhclient lease files and dhcpd lease files with permissions 0640,Kenneth R Westerback
rather than 0000 and 0664 respectively. ok kettenis@ tedu@
2014-02-08installboot has grown up, reflect new simplified usage in softraid exampleChris Cappuccio
2014-02-08be less verbose for pkg_delete -XMarc Espie
2014-02-08scrape update_info when we have full plists.Marc Espie
2014-02-08Be conservative about the resources the controller advertises forChris Cappuccio
"type D" Marvel 9580. From Ching Huang, Areca. ok dlg@
2014-02-08Stop disablng/enabling interrupts in the interrupt handler forChris Cappuccio
"chip type D" which is Marvell 9580. None of the other types do this and OpenBSD doesn't interrupt during the interrupt routine anyways. From Ching Huang, Areca. ok dlg@
2014-02-08more Va -> Cm fixes from Jan Stary <hans at stare.cz>, thanks!Alexandre Ratchov
2014-02-08copy the correct number of channels in join/expandAlexandre Ratchov
2014-02-08remove debug bits: dont fill with junk unused channelsAlexandre Ratchov
2014-02-08use the correct integer width in calls to AMSG_ISSET()Alexandre Ratchov
2014-02-08styleMarc Espie
2014-02-08Fix REST (restart transfer) for offsets large than 2GB. OK okan@Todd C. Miller
2014-02-08Some (if not all) G5 systems use a different layout for the physical memoryMiod Vallat
information (property `reg' of the `/memory' node). Fortunately the available physical memory information still uses the same format, so this only affects the computation of physmem. Detect this case and parse the information correctly, converting to the format expected by pmap, ignoring physical memory beyond 4GB. Compute physmem from all the physical memory information, even memory not usable by the kernel. Let pmap not recompute physmem in pmap_bootstrap() if physmem is != 0 upon entry. This should allow G5 systems fitted with more than 2GB of physical memory to report the correct amount of memory, even though the kernel will only use the lower 2GB. Prompted by a dmesg@ submission by Greg Marsh, owner of a 3.5GB G5 help and tweaks kettenis@, ok mpi@
2014-02-08don't allow unsigned packages by default.Marc Espie
as kettenis@ says "test it now", always possible to revert if we don't want it for the release.
2014-02-08also limit the number of currently opened packages locally: each uncompressMarc Espie
gobbles memory. (this was probably already an issue with external gunzip, but not THAT visible because of separate processes).
2014-02-08Initializing the time from an uninitialized variable isn't a very good idea.Mark Kettenis
While there, afnsify the inittodr() prototype.
2014-02-08Don't unmask interrupts that should be left blocked because we're at aMark Kettenis
higher spl. Fixes booting my n1200 with a usb disk plugged in.
2014-02-08fudge verbose state so that makewhatis is only verbose if pkg_add/delete -vvMarc Espie
or more.
2014-02-08Add option MIPS_PTE64 to IP27/IP35 kernels, and no longer ignore memoryMiod Vallat
beyond 16GB physical.
2014-02-08Allow page table entries to be either 32 bits wide (the existing state of theMiod Vallat
code), or 64 bits wide, if option MIPS_PTE64. 64-bit ptes allow for physical memory beyond 16GB (34 bits) to be addressable by the pmap code.
2014-02-08document -DunsignedMarc Espie
2014-02-08only copy wmesg if wchan, after chatting with guenther and millert.Ted Unangst
(and no need to copy empty string, it's already been zero filled.)
2014-02-07Initializing the time from an uninitialized variable isn't a very good idea.Mark Kettenis
While there, fix some indentation and adjust things for 64-bit time_t.
2014-02-07make sure a free bpf exists before dhclient is run, and remove someAlexander Hall
prior workarounds until we have clonable bpfs ok rpe@ krw@
2014-02-07Fix remote printing; broken in rev 1.50. OK guenther@ stsp@Todd C. Miller
2014-02-07Make carp send IPv6 neighbour advertisements with the 'router' flagStefan Sperling
if we're acting as router. Otherwise clients lose the default route learned from router advertisements during carp failover. Patch by Ole Myhre; ok bluhm@
2014-02-07Update to sendmail 8.14.8. This touches a lot of files due to theTodd C. Miller
Sendmail, Inc -> Proofpoint name change. See RELEASE_NOTES for actual changes.
2014-02-07-Wold-style-declaration is not supported by gcc 3.Miod Vallat
2014-02-07Create -L leases files with the same 0000 permissions as regularKenneth R Westerback
leases.IF files. Inconsistency pointed out by Jiri B.
2014-02-07When using SM_PATH, fetch the SHA256 file only once since we will not beAntoine Jacoutot
using multiple repos. ok rpe@
2014-02-07Log failures to fchmod() or fchown() the file written inKenneth R Westerback
priv_write_file().
2001-09-11sendmail 8.12.0 with $Id tags converted to $SendmailTodd C. Miller
2014-02-07reclaim: we don't need to keep tracking temporary files/directories weMarc Espie
removed.
2014-02-07add qla; ok dlgJason McIntyre
2014-02-07Nuke unused #define CLIENT_PATH. A remnant of dhclient-script.Kenneth R Westerback
2014-02-07historyDavid Gwynne
2014-02-07don't store full terminal, we just need a few control sequencesMarc Espie
2014-02-07Borrow the sparc installboot a.out nlist code, for /usr/mdec/bootxx is stillMiod Vallat
built as an a.out binary.
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-12 21:31:19 +0000