OpenBSD src changes summary for 2017-03-16 ==========================================
bin/ksh bin/pax gnu/usr.bin/binutils-2.17 lib/libc lib/libcrypto lib/libutil libexec/spamd regress/lib sbin/sysctl share/man sys/arch/amd64/include sys/arch/arm64/arm64 sys/arch/arm64/include sys/arch/i386/include sys/arch/luna88k/cbus sys/arch/luna88k/conf sys/arch/luna88k/dev sys/arch/luna88k/include sys/arch/luna88k/stand/boot sys/dev/pci sys/dev/pv sys/dev/wscons sys/net sys/netinet6 usr.bin/newsyslog usr.bin/tmux usr.sbin/dhcrelay usr.sbin/httpd usr.sbin/pcidump usr.sbin/pkg_add usr.sbin/syslogd == bin =============================================================== 01/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/bin ksh ~ sh.1 > document the "hash" built-in, moved to (posix) base in issue 7; (jmc@) ~ sh.1 > document the ulimit built-in, which is now marked xsi by posix; > noted by scott cheloha (jmc@) ~ sh.1 > document the "type" built-in, marked as xsi by posix; (jmc@) pax ~ tables.c > buffer increase factor belongs in the nmemb field of reallocarray() > ok guenther (deraadt@) == gnu =============================================================== 02/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/gnu usr.bin/binutils-2.17 ~ include/opcode/i386.h > Fix handling of the SVM instructions such that they accept a register as > an operand instead of a memory address. This matches what more recent > binutils versions do and allows us to write inline assembly that's accepted > by both clang and gas. > ok mlarkin@, visa@ (kettenis@) == lib =============================================================== 03/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libc ~ stdio/fvwrite.c > When reallocating the buffer for asprintf(), just round up to the > nearest page instead of doubling the old size until it is large > enough. OK deraadt@ (millert@) ~ gen/sysctl.3 > bit more consistent; (jmc@) ~ stdio/asprintf.c ~ stdio/vasprintf.c > Only reallocate the buffer to fit for medium-size allocations where > we expanded the buffer to a single page. The final realloc() can > be expensive for large buffers and is not realled needed. OK deraadt@ > (millert@) ~ gen/sysctl.3 > sync the vfs bits as best i can; (jmc@) libcrypto ~ buffer/buffer.c > Use calloc() instead of malloc() followed by manually zeroing fields. > ok beck@ inoguchi@ (jsing@) ~ buffer/buffer.c > Convert BUF_MEM_grow() and BUF_MEM_grow_clean() to recallocarray(), > ensuring that the buffer contents are zeroed on allocation and not leaked > when resizing. > It is worth noting that BUF_MEM_grow_clean() already did this manually by > avoiding realloc(). > ok beck@ inoguchi@ (jsing@) libutil ~ fmt_scaled.c > Fix overly-conservative overflow checks on mulitplications and add checks > on additions. This allows scan_scaled to work up to +/-LLONG_MAX > (LLONG_MIN > will still be flagged as a range error). ok millert@ (dtucker@) == libexec =========================================================== 04/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/libexec spamd ~ spamd.8 > use one way to show filter rules, not two. the bits and pieces of the > spamd setup are complex enough without freestyling the pf rules; > while here, Bk/Ek no longer required (jmc@) ~ spamd.8 > define the role of spamd-setup a little better; (jmc@) == regress =========================================================== 05/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress lib ~ libutil/fmt_scaled/fmt_test.c > Enable tests for +/-LLONG_MAX and add tests for other boundary conditions. > ok millert@ (dtucker@) == sbin ============================================================== 06/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin sysctl ~ sysctl.c > Fix a regression preventing ``kern.profiling'' to return the profiling > state even with a correct kernel. > Reported by jmc@, ok tb@ (mpi@) ~ sysctl.8 > remove the list of available sysctls: the sysctls are fully > described in sysctl(3) and a list of available sysctls on any particular > machine is best retrieved using sysctl; text tweaked by schwarze > ok bluhm millert deraadt schwarze (jmc@) == share ============================================================= 07/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man5/spamd.conf.5 > spamd.conf is the configuration file for spamd-setup, not spamd; (jmc@) == sys =============================================================== 08/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/amd64/include ~ cpu.h > Typo, from miod@ (mpi@) arch/arm64/arm64 ~ pmap.c > Remove some unused variables, unused types, duplicated/unused function > prototypes and duplicate defines. > Also remove PMAP_CACHE_PTE, which is unused and misleading since the page > tables are mapped as normal write-back cached memory. > Fix a bunch on KNF issues as well. > ok patrick@ (kettenis@) arch/arm64/include ~ pmap.h ~ pte.h > Remove some unused variables, unused types, duplicated/unused function > prototypes and duplicate defines. > Also remove PMAP_CACHE_PTE, which is unused and misleading since the page > tables are mapped as normal write-back cached memory. > Fix a bunch on KNF issues as well. > ok patrick@ (kettenis@) arch/i386/include ~ cpu.h > Typo, from miod@ (mpi@) arch/luna88k/cbus ~ i82365_cbus.c ~ necsb.c > Use a power of two and left shift, rather than integer multiplication, for > the bus_space access stride. This allows better instruction scheduling by > the compiler. > ok aoyama@ (miod@) arch/luna88k/conf ~ ld.script > Fix passing entropy from boot loader to the kernel: > - make sure the kernel has a PT_OPENBSD_RANDOMIZE phdr. > - load /etc/random.conf in all cases, not only manual boot (oops). > ok aoyama@ (miod@) arch/luna88k/dev ~ spc.c > Use a power of two and left shift, rather than integer multiplication, for > the bus_space access stride. This allows better instruction scheduling by > the compiler. > ok aoyama@ (miod@) ~ mb89352.c > Unroll two small bus_space_{read_write}_multi operations in a hot code > path. > This causes a tiny speedup. > ok aoyama@ (miod@) arch/luna88k/include ~ bus.h > Use a power of two and left shift, rather than integer multiplication, for > the bus_space access stride. This allows better instruction scheduling by > the compiler. > ok aoyama@ (miod@) arch/luna88k/stand/boot ~ boot.c ~ init_main.c > Fix passing entropy from boot loader to the kernel: > - make sure the kernel has a PT_OPENBSD_RANDOMIZE phdr. > - load /etc/random.conf in all cases, not only manual boot (oops). > ok aoyama@ (miod@) dev/pci ~ pcireg.h > Print PCIe Extended Capabilities, from Simon Mages > ok kettenis mlarkin (deraadt@) dev/pv ~ vioscsi.c > vioscsi: Don't call virtio_dequeue_commit on not-queued slots > Also, add some debug asserts all over the place. (sf@) ~ vioscsi.c > vioscsi: return XS_NO_CCB if out of space > If there is not enough space in the vring, don't return > XS_DRIVER_STUFFUP but XS_NO_CCB, so that the operation will > be retried later. (sf@) ~ vioscsi.c > vioscsi: Negotiate features during attach > While vioscsi does not support any virtio-scsi features, we need to call > virtio_negotiate_features() in order to negotiate device-independent > features like VIRTIO_F_RING_INDIRECT_DESC. (sf@) dev/wscons ~ wsconsio.h ~ wsmouse.c ~ wsmouseinput.h ~ wstpad.c > Simplify wstpad option handling: > - Remove unnecessary abstraction > - Make it possible to support new features/options without ABI break > - Remove some complexity and hard limits > - Simplify & keep documentation close to options > ok bru@ (mpi@) net ~ rtsock.c > Do not allow to change the gateway of an existing entry if the AF > family is incorrect. > Prevent bgpd(8) to corrupt ARP entries as reported by Joe Holden. > ok benno@, claudio@ (mpi@) netinet6 ~ in6.c > Userland expects pltime to be a time stamp not number of seconds. > pltime conversion was missing from rev 1.193 which did the vltime > conversion only. > OK mpi (florian@) == usr.bin =========================================================== 09/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin newsyslog ~ newsyslog.c > syslogd(8) -Z generates log files with time in ISO format and UTC. > newsyslog(8) still used the traditional BSD syslog timstamps in > local time zone. Convert the latter to the new format unconditionally. > It is usefull to have a distinct timestamp including the year at > the beginning of every logfile. > OK deraadt@ jung@ (bluhm@) tmux ~ cmd-kill-pane.c > Redraw after killp -a. (nicm@) == usr.sbin ========================================================== 10/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin dhcrelay ~ dhcpd.h ~ dhcrelay.c > Move the server struct to header file and make it compatible with IPv6. > ok reyk@ (rzalamena@) httpd ~ httpd.conf.5 ~ server_http.c > Expand $HTTP_HOST in redirects. > From Rivo Nurges Rivo.Nurges AT smit.ee, thanks! > OK reyk (florian@) pcidump ~ pcidump.c > Print PCIe Extended Capabilities, from Simon Mages > ok kettenis mlarkin (deraadt@) pkg_add ~ OpenBSD/PkgAdd.pm > tied files should be treated like deleted files wrt checksums. (espie@) syslogd ~ ttymsg.c > There was a file descripotor leak in the syslogd(8) ttymsg() error > path. Before returning early with an error, close the newly opened > file descriptor. > OK deraadt@ (bluhm@) ~ syslogd.c > Convert the global variable fd_tls to a local one. Use separate > callbacks for TCP and TLS accept(2) instead of looking at the value > of the listen file descriptor. > OK millert@ (bluhm@) ~ Makefile ~ syslogd.c ~ syslogd.h + log.c + log.h > Start to replace the home grown syslogd(8) internal debug and logging > functions with a more common log.c implementation. Of course > openlog(3) cannot be used, so adapt the log.[ch] initially copied > from ospfd(8) to syslogd's special needs. As the messages are > limited to ERRBUFSIZE anyway, malloc(3) in the error logging code > can be avoided. Changing all log calls to the new API will be done > in a separate step. > OK millert@ (bluhm@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
