summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2019-10-19Do not crash trying to fix layout size if only one cell, from Azat Khuzhin.Nicholas Marriott
2019-10-19The size limit of the shutdown reason was relaxed to 255 characters butClaudio Jeker
instead of mentioning the limit, remove it. bgpctl will properly error out if the reason is too long. OK job@ sthen@
2019-10-19Unlock the vnode in vop_inactive of deadfs. The unlock operation isVisa Hankala
a no-op at the moment, but future work will introduce an actual lock that has to be released. OK mpi@ anton@
2019-10-18Don't check that the ipv6 source address of a neighbor advertisment isSebastian Benoit
from a neighbor's address. Sthen@ dug out RFC 4861 4.4 that says that the source address is "An address assigned to the interface from which the advertisement is sent." -- which can be from a network that the receiver does not know about. Indeed my provider sends such a neighbor advertisment for my default gateway, which breaks my uplink. Claudio@ added this check for symetry with NetBSD code when he added the same check to nd6_ns_input(), where it is needed to fix CVE-2008-2476. See also OpenBSD 4.2 errata 15. ok claudio@, kn@
2019-10-18add some checks to avoid UID_MAX (-1) here. this is not problematic withTed Unangst
the current code, but it's probably safer this way. ok deraadt
2019-10-18setting uid to -1 won't work with setresuid, so detect that conditionTed Unangst
and return an error instead. may prevent some unset/missing confusion. ok deraadt millert
2019-10-18Add support for dynamic queue allocation (DQA) to iwm(4).Stefan Sperling
Required for new firmware images because the command queue index has changed. The driver remains compatible with our current firmware images for now. Tested by benno, jan, kevlo, florian, jmatthew, Tracey Emery ok jmatthew
2019-10-18Plug mem leak in error path; ok florian@ who also spotted the rad caseOtto Moerbeek
2019-10-18Another case of using yacc effectively to avoid repeating code; ok florian@Otto Moerbeek
2019-10-18Reafctor forwarder clause to avoid having to write out the same codeOtto Moerbeek
a couple of times by declaring optional stuff the proper way. ok @florian
2019-10-18Bump datasize-cur for pbuild on mips64, to make room for modernity.Visa Hankala
OK deraadt@
2019-10-18Add mips64el to LP64_ARCH.Visa Hankala
OK patrick@
2019-10-17Enable sxidog(4).Mark Kettenis
ok jsg@
2019-10-17Avoid including an armv7 MD header.Mark Kettenis
ok jsg@
2019-10-17drm/amd/display: num of sw i2c/aux engines less than num of connectorsJonathan Gray
From Hersen Wu 0e8e4fbf8d8905071c045f2922de55adbe1a6abe in mainline linux Avoids a NULL deref of aux_engine in dc_link_aux_transfer() with 'ASRock A320M-HDV R4.0' and Raven Ridge reported by semarie@ on bugs@. ok semarie@
2019-10-17Use -1 to indicate an invalid uid/gid, not UID_MAX and GID_MAX.Todd C. Miller
This is the userland portion. OK deraadt@ sashan@
2019-10-17Fix some compiler warings in ifconfig(8). Move all prototypes andAlexander Bluhm
variables used in multiple .c files into common ifconfig.h. Basically this renames brconfig.h to ifconfig.h and also uses it for sff.c. Fix missing prototypes. Global variable name s is bad as it shadows local variables. Call it sock and use it everywhere. OK deraadt@
2019-10-17Cache flush operations on arm64 were being incorrectly treated as writeDale Rahn
operations, however they should be treated as read per the design. Switch to using bit defines, correct said defines. Fixes cache flushing causing Firefox to abort. ok kettenis@ kurt@
2019-10-17Sync RSA_padding_check_PKCS1_OAEP_mgf1().Joel Sing
Update RSA_padding_check_PKCS1_OAEP_mgf1() with code from OpenSSL 1.1.1d (with some improvements/corrections to comments). This brings in code to make the padding check constant time. ok inoguchi@ tb@
2019-10-17Fix -N flag to actually shut down the (entire) socket when the inputBob Beck
goes away. This allows for using nc in cases where the network server will no longer expect anything after eof, instead of hanging waiting for more input from our end. Additionaly, shut down if tls is in use if either side of the socket goes away, since we higher level TLS operations (tls_read and write) will require the socket to be both readable and writable as we can get TLS_WANT_POLLIN or TLS_WANT_POLLOUT on either operation. deraadt@ buying it. found by sthen@
2019-10-17Provide err_clear_last_constant_time() as a way of clearing an error fromJoel Sing
the top of the error stack in constant time. This will be used by upcoming RSA changes. From OpenSSL 1.1.1d. ok inoguchi@ tb@
2019-10-17sgi: iockbc(4), mkbc(4): tsleep(9) -> tsleep_nsec(9); ok kn@ visa@cheloha
2019-10-17Use -1 to indicate an invalid uid/gid, not UID_MAX and GID_MAX.Todd C. Miller
This is clearer and more consistent with the rest of the kernel. OK deraadt@ sashan@
2019-10-17syncTheo de Raadt
2019-10-17in6_setsockaddr and in6_setpeeraddr can't fail, so let them return void.David Gwynne
this also brings them in line with the AF_INET equivalents. ok visa@ bluhm@
2019-10-16Make sxidog(4) MI.Mark Kettenis
2019-10-16Move sxidog(4) such that we can use it on arm64.Mark Kettenis
ok patrick@
2019-10-16tweak previous; help/ok claudioJason McIntyre
2019-10-16Link regress ifconfig to build.Alexander Bluhm
2019-10-16Test various combinations of setting interface addresses with Alexander Bluhm
ifconfig and check the expected result. This covers the ifconfig(8) program and the SIOCAIFADDR and SIOCDIFADDR ioctl(2).
2019-10-16Add an output format for bird and one doing CSV. Also update the manpageClaudio Jeker
to include all the changes.
2019-10-16Clean up header includes. Most are no longer required.Claudio Jeker
2019-10-16Set the maxlength to either the prefixlen itself or the value from the ROA.Claudio Jeker
This way maxlength will always be set to a reasonable value and there is no need to use a conditional to print the value.
2019-10-16tsleep(9) -> tsleep_nsec(9)Martin Pieuchot
ok cheloha@, visa@
2019-10-16tsleep(9) -> tsleep_nsec(9)Martin Pieuchot
ok cheloha@, visa@, akoshibe@
2019-10-16potential NULL dereference for revoked hostkeys; reported byDamien Miller
krishnaiah bommu
2019-10-16free buf before return; reported by krishnaiah bommuDamien Miller
2019-10-16vmd(8): provide some additional info in a debug msgMike Larkin
Print the guest %rip when it tries to do I/O to a nonexistent port. Also convert the message to a DPRINTF so that it doesn't leak guest address information into any logging the host might be doing under normal non-debug conditions.
2019-10-16Add a few newer _OSI strings to the parser.Mike Larkin
Early in the release cycle is the right time to do this. "yes please" jcs@. ok deraadt@
2019-10-16regenMike Larkin
2019-10-16Fix a bunch of lines that had whitespace at the endMike Larkin
2019-10-16Remove trailing whitespace from a lineMike Larkin
2019-10-16Fix reversed gotos; CID 1453190.Daniel Dickman
ok jmatthew@
2019-10-15Include the .SUNW_ctf section in bsd.gdbMartin Pieuchot
Once the section has been built from the DWARF symbols also add it to the debug kernel. That makes ddb(4) print the correct number of args in function backtraces in such kernel as well. While here make comment fit in 80 columns. ok jasper@
2019-10-15Fix db_stack_dump() w/ custom addr & implement db_save_stack_trace().Martin Pieuchot
Substract the BIAS from the `addr' argument since the unwinding algorithm assumes the given frame address doesn't include it. In other words, account for the BIAS added by __builtin_frame_address(). Inputs from and ok kettenis@
2019-10-15Reduce the number of places where `p_priority' and `p_stat' are set.Martin Pieuchot
This refactoring will help future scheduler locking, in particular to shrink the SCHED_LOCK(). No intended behavior change. ok visa@
2019-10-15Rewrite options_array_set to be clearer and remove a spurious warningNicholas Marriott
with newer GCC. From Ben Boeckel.
2019-10-15Add support for percentage sizes for resize-pane ("-x 10%"). Also changeNicholas Marriott
split-window and join-pane -l to accept similar percentages and deprecate -p. From Anindya Mukherjee.
2019-10-15Try to match the devid against the bootpath if link->port_wwn doesn'tJonathan Matthew
work. mpii(4) controllers tend not to provide the right wwn for sata devices, so this helps if we happen to be booting off one of those. ok dlg@ kettenis@
2019-10-15Fix compiler's view of the alignment of xhci structs. Because the structVisa Hankala
fields are naturally aligned, it is enough to remove __packed. This lets the compiler generate saner machine code. In particular, now the compiler does not use lwl/lwr and swl/swr instructions on mips64 for changing xhci descriptors. The descriptors are properly aligned in memory, so lwl/lwr and swl/swr end up doing a redundant load or store. Feedback from guenther@ deraadt@ Tested on amd64 by krw@ OK mpi@ krw@ kettenis@
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-14 20:12:44 +0000