summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* MFC: r287405:imp2015-09-081-7/+41
| | | | Make out of memory behavior less pathological.
* Merge r286649:rodrigc2015-09-082-10/+6
| | | | | | Fix and re-enable UTF-8 tests. X-Merge with: r287393
* MFC r287437sbruno2015-09-081-5/+3
| | | | | | | | | | | | | r249170 was just plain wrong. The effect of the change is to always delete a logic volume on status change which is NOT what we want here. The original code is correct in that when the volume changes status the driver will only delete the volume if the status is one of the fatal errors. A drive failure in a mirrored volume is NOT a situtation where the volume should dissapear. Reported on freebsd-scsi@: https://lists.freebsd.org/pipermail/freebsd-scsi/2015-September/006800.html
* MFC r287369:andrew2015-09-081-1/+1
| | | | | | | | | | | | | | | | | | | Ensure we use calculate_first_tls_offset, even if the main program doesn't have TLS program header. This is needed on architectures with Variant I tls, that is arm, arm64, mips, and powerpc. These place the thread control block at the start of the buffer and, without this, this data may be trashed. This appears to not be an issue on mips or powerpc as they include a second adjustment to move the thread local data, however this is on arm64 (with a future change to fix placing this data), and should be on arm. I am unable to trigger this on arm, even after changing the code to move the data around to make it more likely to be hit. This is most likely because my tests didn't use the variable in offset 0. Reviewed by: kib MFC after: 1 week Sponsored by: ABT Systems Ltd
* MFC r287360:kib2015-09-082-28/+33
| | | | | Fix t_spawnattr test for attributes handling by posix_spawn(3). Connect it to the build.
* MFC r287123:delphij2015-09-081-2/+3
| | | | | | | Finish r89633 and completely remove the remaining of VERSION. PR: bin/202308 Submitted by: John Hein <z7dr6ut7gs snkmail com>
* MFC r287093:delphij2015-09-081-1/+1
| | | | | | | Instead of doing an no-op (|= 0), actually clear the flags in acl_clear_flags_np. Reported by: Pascal Drecker <pascal freebsd drecker com>
* MFC r287021:delphij2015-09-081-1/+1
| | | | Fix build when LEAPSECONDS is enabled.
* MFC r287020: MFV r287019: leapseconds from tzdata2015f.delphij2015-09-081-58/+18
|
* MFC r287310:kib2015-09-061-1/+1
| | | | | Use P1B_PRIO_MAX to designate max posix priority for the RR/FIFO scheduler types.
* MFC r287309:kib2015-09-062-94/+71
| | | | | Remove single-use macros obfuscating malloc(9) and free(9) calls. Style.
* MFC r287432: Fix copy-paste bug introduced in r275458.mav2015-09-061-1/+1
|
* MFC r287292:kib2015-09-0522-121/+104
| | | | | | | | | | Switch libc from using _sig{procmask,action,suspend} symbols, which are aliases for the syscall stubs and are plt-interposed, to the libc-private aliases of internally interposed sigprocmask() etc. MFC r287300: Use libthr interposed functions instead of syscalls, in posix_spawn()' child.
* MFC r277853sbruno2015-09-041-1/+1
| | | | | | | Check for invalid length or more than max length for the interpreter, instead of the validity of the string pointer holding the interpreter. Submitted by: sson
* MFC r276834sbruno2015-09-042-1/+2
| | | | | | | | | | | | | Update Features2 to display SDBG capability of processor. This is showing up on Haswell-class CPUs From the Intel SDM, "Table 3-20. Feature Information Returned in the ECX Register" 11 | SDBG | A value of 1 indicates the processor supports IA32_DEBUG_INTERFACE MSR for silicon debug. Submitted by: jiashiun@gmail.com
* MFC r286238sbruno2015-09-041-3/+3
| | | | | | | | A misplaced #endif in ixgbe_ioctl() causes interface MTU to become zero when INET and INET6 are undefined. PR: 162028 Submitted by: hoomanfazaeli@gmail.com pluknet
* MFC r287368:gjb2015-09-041-2/+2
| | | | | | | | Remove '-' separating OSRELEASE and SNAPSHOT_DATE for vagrant builds, and prepend it to SNAPSHOT_DATE to prevent a trailing '-' in the final box name for a release build. Sponsored by: The FreeBSD Foundation
* MFC r287345:delphij2015-09-041-10/+4
| | | | | | | | Drop group privileges after opening the kvm descriptor, otherwise, the code would not drop privileges as expected. While there also add checks for the drop and bail out immediately if we failed.
* Put timeout directory in correct placerodrigc2015-09-031-2/+2
|
* MFC r286964asomers2015-09-0310-0/+257
| | | | Add ATF functional tests for fstyp(8). No ZFS or GELI tests yet.
* MFC 283622:jhb2015-09-034-4/+8
| | | | | Add <sys/user.h> to the SYNOPSIS of the kinfo_get*() functions since these functions all return types that are defined in that header.
* MFC: r287183, r287264, r287265imp2015-09-031-0/+4
| | | | Export kern.features.invariants when kernel is compiled with invariants.
* MFC 281941:jhb2015-09-034-9/+17
| | | | | | | | | Watchdog drivers need to support rearming the watchdog in contexts which are not permitted to sleep. Only use the IPMI watchdog with backends which poll driver-initiated requests to meet this requirement. In practice this means that watchdogs will no longer be used on systems that use the SSIF backend.
* MFC r287208, r287216:ume2015-09-036-5/+8
| | | | Make it buildable with WITH_OPENLDAP, again.
* MFC: r286490,r286491,r287125bapt2015-09-021-16/+6
| | | | | | | | | Per rfc3629 value greater than 0x10ffff should be rejected (r286490,r286491) Make UTF-8 parsing and generation more strict. (r287125 by ed) - in mbrtowc() we need to disallow codepoints above 0x10ffff. - In wcrtomb() we need to disallow codepoints between 0xd800 and 0xdfff.
* MFC: r268745,r268746,r268747,r268748,r268749,r268861,r268750,r268751,r268763bapt2015-09-027-0/+727
| | | | | | | | r273769,r273771,r276771,r278810 New BSDL timeout(1) utility compatible with GNU timeout Relnotes: yes
* MFC r286887: Using the error return code documented in the comment.avatar2015-09-011-1/+1
| | | | | | | | Though there is no direct midi_uninit() caller amongst existing drivers at this moment, a quick experiment indicates that EBUSY gives users more precise error message once drivers start to honour this result. For example, emu_midi_detach() should check the result of mpu401_uninit() and block module unloading if there is any MIDI I/O in progress.
* MFC r286892asomers2015-09-013-5/+42
| | | | | | | | | Serve /etc/eui64 via NIS. The C library already knows how to lookup eui64 entries from NIS. For example, fwcontrol(8) does it. But /var/yp/Makefile.dist doesn't build the eui64 maps, and ypinit(8) doesn't push them to slaves. This change fixes that.
* MFC r286910:delphij2015-09-011-10/+5
| | | | | | | | - ANSIfy - Remove the redundant _PATH_RSH definition (paths.h at r96194); - Use pid_t for PIDs - Note that we are at the same level of OpenBSD's counterpart of revision 1.7 (r94757).
* MFC r286894:delphij2015-09-012-0/+10
| | | | | | Set curvnet context inside the RPC code in more places. Reviewed by: melifaro
* MFC r272282,r272363,r272383:bdrewery2015-08-312-2/+26
| | | | | | | | | | | | | | r272282: Search for the nearest PORTSDIR where Mk/bsd.ports.mk exists, from .CURDIR. This will only take effect if PORTSDIR is not set, as previously supported. r272363: Always resolve PORTSDIR to absolute paths using realpath(1). r272383: Revise r272363 by collapsing the tests into a for loop. Relnotes: yes
* MFC r282551: Remove historical GNUC testemaste2015-08-312-8/+0
| | | | | | | The requirement is for a GCC-compatible compiler and not necessarily GCC itself. However, we currently expect any compiler used for building the whole of FreeBSD to be GCC-compatible and many things will break if not; there's no longer a need to have an explicit test for this in rtld.
* MFC r286836:delphij2015-08-311-1/+1
| | | | | so_vnet is constant after creation and no locking is necessary, document this fact.
* MFC r286834:delphij2015-08-312-16/+2
| | | | Use arc4random_buf(). While there, unifdef the code for !HAVE_ARC4RANDOM.
* MFC r285844: ar: add -U (unique) option to disable -D (deterministic) modeemaste2015-08-312-7/+34
| | | | | | | | | This is required in order for us to support deterministic mode by default. If multiple -D or -U options are specified on the command line, the final one takes precedence. GNU ar also uses -U for this. PR: 196929 Sponsored by: The FreeBSD Foundation
* MFC r279084,280713: setmode(): Use sysctl kern.proc.umask instead of umask()jilles2015-08-291-9/+33
| | | | | | | | | if possible. The kern.proc.umask.<pid> sysctl allows querying the umask without temporarily modifying it. r280713 is the actual change, while r279084 is a whitespace change.
* MFC r286888: Using consistent coding style to deal with error inside the loop.avatar2015-08-291-2/+2
|
* MFC r287025: Remove some code duplication by using biofinish().mav2015-08-291-6/+2
|
* MFC r286814, r286816: Remove UMA allocation of ATA requests.mav2015-08-292-36/+3
| | | | | | After CAM replaced old ATA stack, this driver processes no more then one request at a time per channel. Using UMA after that is overkill, so replace it with simple preallocation of one request per channel.
* MFC r286773:hselasky2015-08-298-30/+73
| | | | | Improve the realtime properties of USB transfers for embedded systems like RPI-B and RPI-2.
* MFC r286799:hselasky2015-08-292-1/+10
| | | | | | | Fix race in USB PF which can happen if we stop tracing exactly when the kernel is tapping an USB transfer. This leads to a NULL pointer access. The solution is to only trace while the USB bus lock is locked.
* MFC r283067, r286118, r285638, r285935, r286778, r286780 and r286802:hselasky2015-08-293-408/+374
| | | | | | | | | | | | - Make the FIFO configuration a bit more flexible for the DWC OTG in device side mode. - Limit the number of times we loop inside the DWC OTG poll handler to avoid starving other fast interrupts. Fix a comment while at it. - Optimise the DWC OTG host mode driver's transmit path - Optimise the DWC OTG host mode driver's receive path - Minor code refactor to avoid duplicating code. - Handle NYET high speed tokens and predict NAK'ing is up next. - Fixes for HIGH speed ISOCHRONOUS traffic.
* MFC r287012:jamie2015-08-295-21/+78
| | | | | | | Make pkill/pgrep -j ARG take jname, not just jid. PR: 201588 Submitted by: Daniel Shahaf <danielsh at apache.org>
* MFC: r286790rmacklem2015-08-281-2/+5
| | | | | | | | For the case where an NFSv4.1 ExchangeID operation has the client identifier that already has a confirmed ClientID, the nfsrv_setclient() function would not fill in the clientidp being returned. As such, the value of ClientID returned would be whatever garbage was on the stack. This patch fixes the problem by filling in these fields.
* MFC r286344: find: Fix segfault with very long path in -exec/-ok ... {} \;.jilles2015-08-282-15/+25
| | | | | | | | | | | | | If the resulting argument is longer than MAXPATHLEN, realloc() was called to extend the space, but the new pointer was not correctly stored. Different from what OpenBSD has done, rewrite brace_subst() to calculate the necessary space first and realloc() at most once. As before, the e_len fields are not updated in case of a realloc. Therefore, a following long argument will do another realloc. PR: 201750
* MFC r286720:ae2015-08-281-3/+6
| | | | | | Use g_conf_printf_escaped() to escape illegal symbols in file name. PR: 202289
* MFC r285252:markj2015-08-281-1/+1
| | | | Fix an incorrect assertion in witness.
* MFC r286167:markj2015-08-271-2/+6
| | | | Avoid dereferencing curthread->td_proc->p_cred in DTrace probe context.
* MFC r286418:markj2015-08-271-2/+2
| | | | | ipv4_is_zeronet() and ipv4_is_loopback() expect an address in network order, but IN_ZERONET and IN_LOOPBACK expect it in host order.
* MFC r286169:markj2015-08-271-5/+17
| | | | | | Perform bounds checking when constructing a format string. PR: 201657
OpenPOWER on IntegriCloud