OpenBSD src changes summary for 2017-10-08 to 2017-10-15 inclusive ==================================================================
bin/pax distrib/miniroot distrib/sets etc/mtree/4.4BSD.dist etc/rc gnu/usr.bin/clang include/Makefile include/pthread.h include/pthread_np.h include/sched.h include/semaphore.h lib/libarch lib/libc lib/libpthread lib/librthread lib/libssl lib/libtls libexec/ld.so libexec/spamd regress/lib regress/sbin regress/sys regress/usr.sbin sbin/dhclient share/man sys/arch/alpha/alpha sys/arch/alpha/tc sys/arch/amd64/amd64 sys/arch/amd64/conf sys/arch/amd64/include sys/arch/amd64/isa sys/arch/amd64/pci sys/arch/amd64/stand/efiboot sys/arch/amd64/stand/libsa sys/arch/arm64/conf sys/arch/armv7/conf sys/arch/macppc/conf sys/arch/mips64/mips64 sys/conf sys/ddb sys/dev sys/dev/acpi sys/dev/fdt sys/dev/hid sys/dev/ic sys/dev/pci sys/dev/sdmmc sys/dev/tc sys/dev/usb sys/kern sys/lib/libsa sys/net sys/netinet sys/netinet6 sys/sys sys/ufs/ffs usr.bin/awk usr.bin/bc usr.bin/ctags usr.bin/mg usr.bin/ssh usr.bin/tmux usr.sbin/apmd usr.sbin/bgpctl usr.sbin/bgpd usr.sbin/pkg_add usr.sbin/smtpd usr.sbin/spamdb usr.sbin/syspatch == bin =============================================================== 01/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/bin pax ~ tar.1 > Better document what the -v flag actually does and that you can > specify it more than once. Adapted from a diff by and OK schwarze@ > (millert@) == distrib =========================================================== 02/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib miniroot ~ install.sub > Allow to boot into the existing bsd.rd and to download, verify > and upgrade to the bsd.rd of the next release. > The installer detects this bsd.rd only upgrade if the answer to > the 'Server directory?' question points to the sets of the next > release. > It then limits the set selection to the bsd.rd file, verifies it > using the on-disk signify key of the next release and makes a > backup of the existing bsd.rd before installing it. > Idea from and OK deraadt@ > OK benno@ (rpe@) ~ install.sub > If $MODE.site fails, store random.seed and error out. Based on a diff > for auto$MODE by rpe in response to a question by mpi. > ok rpe mpi (tb@) ~ dot.profile > Add a comment for slaacd(8) start and replace the if-then-fi block > with a [[ test ]] && command construct. (rpe@) ~ dot.profile > Tweak comments (rpe@) ~ dot.profile > Comment trap handlers (rpe@) ~ dot.profile > Move extraction of rootdisk information into the DONEPROFILE if-block. > OK tb@ (rpe@) sets ~ lists/base/mi ~ lists/comp/mi > sync (visa@) ~ lists/comp/mi > sort (visa@) ~ lists/comp/mi > sync (deraadt@) == etc =============================================================== 03/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/etc mtree/4.4BSD.dist ~ mtree/4.4BSD.dist > Make cacheflush(3) and get_fpc_csr(3) manual pages available on octeon. > OK deraadt@ (visa@) rc ~ rc > Move comment line and spacing. (rpe@) ~ rc > The testprogram for ld.so reordering is executed in tmpdir. > Move tmpdir for reordering library from /tmp to /usr/lib. > This allows to have /tmp mounted noexec. > prompted by reports on misc@ > OK deraadt@ tj@ tb@ (rpe@) == gnu =============================================================== 04/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/gnu usr.bin/clang ~ libLLVMARMCodeGen/Makefile > Add missing file. > ok patrick@ (kettenis@) == include =========================================================== 05/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/include Makefile ~ Makefile > Move the thread-related .h files to /usr/src/include/, since the > implementation is now spread between libc and librthread. No changes > to the content > ok mpi@ (guenther@) pthread.h + pthread.h > Move the thread-related .h files to /usr/src/include/, since the > implementation is now spread between libc and librthread. No changes > to the content > ok mpi@ (guenther@) pthread_np.h + pthread_np.h > Move the thread-related .h files to /usr/src/include/, since the > implementation is now spread between libc and librthread. No changes > to the content > ok mpi@ (guenther@) sched.h + sched.h > Move the thread-related .h files to /usr/src/include/, since the > implementation is now spread between libc and librthread. No changes > to the content > ok mpi@ (guenther@) semaphore.h + semaphore.h > Move the thread-related .h files to /usr/src/include/, since the > implementation is now spread between libc and librthread. No changes > to the content > ok mpi@ (guenther@) == lib =============================================================== 06/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libarch ~ mips64/get_fpc_csr.3 > Fix typo. (visa@) ~ mips64/Makefile > Make cacheflush(3) and get_fpc_csr(3) manual pages available on octeon. > OK deraadt@ (visa@) libc ~ sys/ktrace.2 > Document fktrace(2) more completely, including separation of errors. > Add a few errors that were previously undocumented. > ok jmc@ (guenther@) ~ string/bzero.3 > Add STANDARDS: denis@ spotted that it was missing. > OK deraadt@ jca@ jmc@ (schwarze@) ~ arch/mips64/gen/ldexp.S > Do not use an uninitialized value when determining the sign > of a denormal result of ldexp(3). > The bug was found when investigating why denormal results > of pow(3) can have an incorrect sign on loongson. > pow(3) misbehaviour reported and fix tested by juanfra@ > No objection from deraadt@ (visa@) libpthread - include/Makefile.inc - include/pthread.h - include/pthread_np.h - include/sched.h - include/semaphore.h > Move the thread-related .h files to /usr/src/include/, since the > implementation is now spread between libc and librthread. No changes > to the content > ok mpi@ (guenther@) librthread ~ Makefile > Move the thread-related .h files to /usr/src/include/, since the > implementation is now spread between libc and librthread. No changes > to the content > ok mpi@ (guenther@) libssl ~ d1_both.c ~ d1_clnt.c ~ d1_srvr.c ~ s3_lib.c ~ ssl_both.c ~ ssl_locl.h > Convert ssl3_send_change_cipher_spec() to use CBB and make it handle DTLS, > which allows us to drop dtls1_send_change_cipher_spec() entirely. > ok inoguchi@ (jsing@) ~ ssl_clnt.c > Fix some style/whitespace/indentation issues in ssl3_connect(). (jsing@) ~ ssl_srvr.c > Fix some style/whitespace/indentation issues in ssl3_accept(). (jsing@) ~ d1_clnt.c > Reduce non-functional differences between dtls1_connect() and > ssl3_connect() - synchronise comments, whitespace, line wrapping, etc. > (jsing@) ~ d1_srvr.c > Reduce non-functional differences between dtls1_accept() and > ssl3_accept() - synchronise comments, whitespace, line wrapping, etc. > (jsing@) ~ d1_clnt.c ~ d1_meth.c ~ ssl_clnt.c ~ ssl_locl.h > Merge dtls1_connect() into ssl3_connect(), removing a large amount of > duplicated code. For now this is essentially adds a diff of the two > functions with 'if (SSL_IS_DTLS(s))' - further clean up and improvement > will follow. > ok inoguchi@ (jsing@) ~ ssl_clnt.c > ((remove) (some) (unnecessary) (parentheses)) > Part of a diff from doug@ (jsing@) ~ ssl_lib.c ~ ssl_locl.h ~ ssl_srvr.c > Make ssl_bytes_to_cipher_list() take a CBS, rather than a pointer and > length, since the caller has already been converted to CBS. A small amount > of additional clean up whilst here. (jsing@) ~ ssl_locl.h ~ ssl_srvr.c ~ t1_lib.c > Fully convert ssl3_send_server_hello() to CBB. > Based on a diff from doug@ (jsing@) ~ ssl_clnt.c ~ ssl_lib.c ~ ssl_locl.h ~ t1_lib.c > Convert ssl3_client_hello() to CBB. > As part of this, change ssl_cipher_list_to_bytes() to take a CBB argument, > rather than a pointer/length. Some additional clean up/renames while here. > Based on a diff from doug@ (jsing@) ~ ssl_locl.h > Drop prototypes for ssl23_*() functions, which no longer exist. (jsing@) ~ d1_meth.c ~ d1_srvr.c ~ ssl_locl.h ~ ssl_srvr.c > Fold dtls1_accept() into ssl_accept(), removing a lot of duplicated code. > With review/feedback from inoguchi@ (jsing@) ~ ssl_clnt.c ~ ssl_locl.h > Rename ssl3_client_hello() to ssl3_send_client_hello() for consistency. > (jsing@) libtls ~ man/tls_conn_version.3 ~ man/tls_load_file.3 > hyphenate DER/PEM-encoded, for consistency; (jmc@) == libexec =========================================================== 07/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/libexec ld.so ~ alpha/rtld_machine.c ~ powerpc/rtld_machine.c ~ sparc64/rtld_machine.c > __plt_{start,end} are no longer generated or necessary, having been > superseded by the generic ?WX and RELRO support. Stop trying to look > up and use them. > ok kettenis@ deraadt@ (guenther@) spamd ~ spamd.c > Check for asprintf() returning -1 instead of assuming that the > pointer will remain unchanged when fails. OK deraadt@ (millert@) == regress =========================================================== 08/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress lib ~ libssl/unit/cipher_list.c > Revise regress now that ssl_bytes_to_cipher_list() takes a CBS. (jsing@) ~ libssl/unit/cipher_list.c > Update regress now that ssl_cipher_list_to_bytes() takes a CBB. > Based on a diff from doug@ (jsing@) ~ libc/Makefile + libc/ldexp/Makefile + libc/ldexp/ldexp_test.c > Add a regression test for ldexp(3). (visa@) sbin ~ slaacd/Makefile > don't create pyc files (florian@) ~ slaacd/Makefile + slaacd/IfInfo.py + slaacd/Slaacctl.py + slaacd/process_ra.py > Generate a router advertisement with scapy and check that slaacd > receives it by parsing slacctl show interface. > This is a first stab, more things should be checked. (florian@) sys - net/pf_divert/args-udp-packet-in-init.pl - net/pf_divert/args-udp-packet-in-resp.pl - net/pf_divert/args-udp-packet-out-init.pl - net/pf_divert/args-udp-packet-out-resp.pl ~ net/pf_divert/Makefile ~ net/pf_divert/Packet.pm > The divert-packet socket option IP_DIVERTFL, IPPROTO_DIVERT_RESP, > IPPROTO_DIVERT_INIT feature is gone. Remove the corresponding > tests. (bluhm@) ~ kern/pledge/generic/Makefile ~ kern/pledge/generic/manager.c > Use a regex(3) to find the new pledge violation string in the dmesg. > (bluhm@) usr.sbin + vmd/Makefile + vmd/config/Makefile + vmd/config/vmd-fail-boot-name-too-long.conf + vmd/config/vmd-fail-boot-name-too-long.ok + vmd/config/vmd-fail-disk-path-too-long.conf + vmd/config/vmd-fail-disk-path-too-long.ok + vmd/config/vmd-fail-kernel-keyword.conf + vmd/config/vmd-fail-kernel-keyword.ok + vmd/config/vmd-fail-too-few-ram.conf + vmd/config/vmd-fail-too-few-ram.ok + vmd/config/vmd-fail-too-many-disks.conf + vmd/config/vmd-fail-too-many-disks.ok + vmd/config/vmd-fail-too-many-ifs.conf + vmd/config/vmd-fail-too-many-ifs.ok + vmd/config/vmd-fail-vm-name-too-long.conf + vmd/config/vmd-fail-vm-name-too-long.ok + vmd/config/vmd-pass-boot-keyword.conf + vmd/config/vmd-pass-boot-keyword.ok + vmd/config/vmd-pass-memory-just-enough.conf + vmd/config/vmd-pass-memory-just-enough.ok + vmd/config/vmd-pass-memory-round.conf + vmd/config/vmd-pass-memory-round.ok > New import: > Add regression tests for vmd(8) config parsing. ~ Makefile > Link vmd regress to build on amd64 and i386. (bluhm@) == sbin ============================================================== 09/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin dhclient ~ parse.c > Simply parse_date() by trusting strptime() more and omitting > pointlessly precise error messages in favour of 'expecting > UTC time'. (krw@) ~ parse.c > Tweak parse_ip_addr() to emit a single message on > failure ("expecting IPv4 address") and to properly > handle the terminating ';' in error situations. (krw@) ~ clparse.c ~ dhcpd.h ~ parse.c > Tweak parse_lease_time() to emit a single message on > failure ("expecting unsigned 32-bit decimal value") and to > properly handle the terminating ';' in error situations. > Make parse_lease_time() return an int to indicate success > or failure as its friends do. Also avoid swapping endianess > twice. > Use == 1 vs != 0 when checking parse_ip_addr() return > value. (krw@) ~ parse.c > Tweak parse_cidr() to be more like its friends. > i.e. handle ';' better, and issue only one error > message ("expecting IPv4 CIDR block"). > One gratuitous whitespace tweak tossed in. (krw@) ~ clparse.c ~ parse.c > Tweak parse_boolean() to be more like its friends. > i.e. handle ';' better, and issue error > message ("expecting boolean.") itself. (krw@) ~ options.c > Repair printing of classless-static-routes to leases > file. Broken due to incorrect translation from a netmask > to the count of leading 1's. Use brute force counting > until something better is found. > Issue discovered and diffs tested by Anthony Coulter via > bugs@. Thanks! (krw@) ~ clparse.c ~ parse.c > Tweak parse_decimal() and its invocations to emit a > single error message ("expecting integer between x and y") > and to properly handle ';' in error cases. (krw@) ~ clparse.c ~ dhcpd.h ~ parse.c > Tweak parse_date() again, this time to factor > code in the same way as the other parse_*() > functions. Whitespace fix in comment while > passing. Also conform to idiom that original > data is not changed in the error path. (krw@) ~ clparse.c > Eliminate unneeded intermediate variable. Fix > some 'return;'s that should be 'break;'s. (krw@) ~ parse.c > Works better when both files in a diff are > committed. (krw@) ~ clparse.c ~ dhcpd.h ~ parse.c > Make parse_string() toe the new line by returning > 0 for failure, 1 for success, emitting a > single error message ("expecting string.") and > handling ';' better. > Don't leak memory when encountering pathological > config or lease files containing repeated instances > of an option or command with string data. (krw@) ~ clparse.c > Stop leaks of SSID strings. Whitespace fix. (krw@) ~ clparse.c > Don't leak string that overflows the option data buffer. (krw@) ~ clparse.c > Tweak parse_option_decl() to return 0/1 to indicate success, > to not touch existing data on error paths, to handle ';' > better. (krw@) ~ clparse.c > Tweak parse_option_list() to return 0/1 to indicate success, > to not touch existing data on error paths, to handle ';' > better, emit single error message ("expecting comma > delimited list of option names."). (krw@) ~ clparse.c ~ dhcpd.h ~ parse.c > Refactor parsing of hex sequences. Rename parse_X() > to parse_hex_octets() and have it return 0/1 to indicate success, > to not touch existing data on error paths, to handle ';' > better, emit single error message ("expecting colon > delimited list of hex octets."). Fold parse_hex() into > parse_hex_octets(). > Simplify logic in parse_option_decl() to eliminate duplicate > overflow error checking and error messages as a result. (krw@) ~ clparse.c > Whitespace fix. (krw@) == share ============================================================= 10/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man4/umb.4 > add em7345; from christoph r. murauer (jmc@) ~ man4/pf.4 > some style fixes from ross l richardson; > checked by/ok bluhm (jmc@) ~ man9/mbuf.9 > Copy comment for M_LOOP from header file to mbuf(9). (bluhm@) ~ man5/pf.conf.5 > Integrate the description of flow queues into the main body of text > (mikeb@) ~ man5/pf.conf.5 > tweak previous; (jmc@) == sys =============================================================== 11/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/alpha/alpha ~ db_trace.c > ansify function prototypes (to match db_stack_trace_print() in the same > file) (jasper@) arch/alpha/tc ~ ioasic.c > Make the tc alpha blinkenlichten start by default. > Restores the balance of serenity and peace of mind of the last > real alpha user, to some extent. > From miod@ (mpi@) arch/amd64/amd64 ~ locore.S ~ locore0.S ~ machdep.c > remove a unused variable > ok tom, kettenis, deraadt (mlarkin@) ~ acpi_machdep.c ~ autoconf.c ~ bios.c ~ bus_dma.c ~ codepatch.c ~ consinit.c ~ cpu.c ~ fpu.c ~ i8259.c ~ identcpu.c ~ intr.c ~ ioapic.c ~ lapic.c ~ machdep.c ~ mainbus.c ~ mem.c ~ mpbios.c ~ pmap.c ~ process_machdep.c ~ sys_machdep.c ~ trap.c ~ tsc.c ~ via.c ~ vm_machdep.c ~ wscons_machdep.c > reduce the amount of includes in arch/amd64 > ok mpi@ deraadt@ (jsg@) ~ locore0.S > locore0: remove an unused variable > ok deraadt (mlarkin@) arch/amd64/conf ~ GENERIC > Add bwfm(4), a driver for Broadcom FullMAC WiFi controllers. The > FullMAC, in comparison to SoftMAC, does most WiFi handling in the > firmware that's running on the controller. This means we have to > work around the net80211 stack while still implementing all the > WiFi interfaces to userland. > This driver is still in early development. So far it can connect > to open WiFis over the USB bus. SDIO and PCIe support, for devices > like the Raspberry Pi 3 or the Macbooks, is not yet implemented. > Also mbufs on the transmit path leak and are not yet freed. > ok stsp@ (patrick@) arch/amd64/include ~ tcb.h > I was one bit off in the "is canonical" test for FS.base (TCB pointer) > values, > so it was still possible to fault the kernel. > Problem observed by Maxime Villard > ok deraadt@ bluhm@ (guenther@) ~ tcb.h TAGGED OPENBSD_6_2 > I was one bit off in the "is canonical" test for FS.base (TCB pointer) > values, > so it was still possible to fault the kernel. > Problem observed by Maxime Villard > from guenther@; ok deraadt@ bluhm@ > OpenBSD 6.2 errata 001 (bluhm@) ~ tcb.h TAGGED OPENBSD_6_1 > I was one bit off in the "is canonical" test for FS.base (TCB pointer) > values, > so it was still possible to fault the kernel. > Problem observed by Maxime Villard > from guenther@; ok deraadt@ bluhm@ > OpenBSD 6.1 errata 032 (bluhm@) arch/amd64/isa ~ clock.c ~ isa_machdep.c > reduce the amount of includes in arch/amd64 > ok mpi@ deraadt@ (jsg@) arch/amd64/pci ~ pchb.c ~ pci_machdep.c ~ pciide_machdep.c > reduce the amount of includes in arch/amd64 > ok mpi@ deraadt@ (jsg@) arch/amd64/stand/efiboot ~ efiboot.c ~ efiboot.h > Add "machine gop" command to support changing graphic output mode. > The selected mode is used by efifb(4). > original diff from Klemens Nanni, tested by Andrew Daugherity (yasuoka@) arch/amd64/stand/libsa ~ cmd_i386.c > Add "machine gop" command to support changing graphic output mode. > The selected mode is used by efifb(4). > original diff from Klemens Nanni, tested by Andrew Daugherity (yasuoka@) arch/arm64/conf ~ Makefile.arm64 > Use gap.o on arm64. After the recent llvm/lld 5.0.0 update the linker > script can be used. ok kettenis@ (jsg@) arch/armv7/conf ~ GENERIC ~ RAMDISK > Enable rkiic(4) and rkpmic(4) for Rockchip RK3288 boards. > ok patrick@, jsg@ (kettenis@) ~ GENERIC > Add bwfm(4), a driver for Broadcom FullMAC WiFi controllers. The > FullMAC, in comparison to SoftMAC, does most WiFi handling in the > firmware that's running on the controller. This means we have to > work around the net80211 stack while still implementing all the > WiFi interfaces to userland. > This driver is still in early development. So far it can connect > to open WiFis over the USB bus. SDIO and PCIe support, for devices > like the Raspberry Pi 3 or the Macbooks, is not yet implemented. > Also mbufs on the transmit path leak and are not yet freed. > ok stsp@ (patrick@) arch/macppc/conf ~ files.macppc > input files should not be .o; from Scott Cheloha (deraadt@) arch/mips64/mips64 ~ cache_octeon.c > Try to make CPU cache size values more correct, especially on OCTEON II > and III. The logic does not look nice, but the parameters do not follow > the standard config register layout anyway. > Remove unnecessary default values, and assume that the Config1 > and Config2 registers are available on every CPU. > Tested on CN5020, CN6120, CN7130 and CN7360. (visa@) conf ~ newvers.sh TAGGED OPENBSD_6_2 > 6.2-stable (bluhm@) ~ files TAGGED OPENBSD_6_2 > Add bwfm(4), a driver for Broadcom FullMAC WiFi controllers. The > FullMAC, in comparison to SoftMAC, does most WiFi handling in the > firmware that's running on the controller. This means we have to > work around the net80211 stack while still implementing all the > WiFi interfaces to userland. > This driver is still in early development. So far it can connect > to open WiFis over the USB bus. SDIO and PCIe support, for devices > like the Raspberry Pi 3 or the Macbooks, is not yet implemented. > Also mbufs on the transmit path leak and are not yet freed. > ok stsp@ (patrick@) ddb ~ db_ctf.c > missing prototype for db_ctf_pprint > ok mpi@ (jasper@) ~ db_ctf.c > - use db_printf() > - refuse to pretty-print if there's no CTF > ok mpi@ (jasper@) ~ db_ctf.c > return a missing return (was left out of previous) (jasper@) dev ~ video.c > Fix debug printf, from Dave Voutila. (mpi@) dev/acpi ~ dsdt.c > Add missing size to free(); ok deraadt@ mpi@ (anton@) dev/fdt ~ rkiic.c > Enable rkiic(4) and rkpmic(4) for Rockchip RK3288 boards. > ok patrick@, jsg@ (kettenis@) ~ syscon.c > Print the name of the device tree node like we do for simplebus(4). > This makes it easier to match dmesg output to the device tree. > ok patrick@ (kettenis@) dev/hid ~ hidmt.c ~ hidmtvar.h > Adapt hidmt to the MT interface of wsmouse, and add the compat-mode setup. > ok mpi@, jcs@ (bru@) ~ hidmt.c ~ hidmtvar.h > hidmt: rename internal hidmt_input struct to hidmt_data, avoid clash > with hidmt_input function (jcs@) ~ hidmt.c ~ hidmtvar.h > hidmt: add support for hybrid packet mode used by some touchpads (jcs@) ~ hidmt.c ~ hidmtvar.h > hidmt: send input mode set report as 2 bytes, matches linux > tested by a few (jcs@) dev/ic ~ rtsx.c ~ rtsxvar.h > Add ADMA support to rtsx(4) to make use of DMA buffers provided by > sdmmc(4). > Data transfers use a little bit less CPU time now. > Tested by me on RT5209 and on RTS525A by jcs@ > ok mpi@ visa@ (stsp@) ~ rtsx.c > Fix build on 32 bit; paddr in rtsx_xfer_dma needs to be 64bit on all > platforms. > Found the hard way by deraadt@ (stsp@) + bwfm.c + bwfmreg.h + bwfmvar.h > Add bwfm(4), a driver for Broadcom FullMAC WiFi controllers. The > FullMAC, in comparison to SoftMAC, does most WiFi handling in the > firmware that's running on the controller. This means we have to > work around the net80211 stack while still implementing all the > WiFi interfaces to userland. > This driver is still in early development. So far it can connect > to open WiFis over the USB bus. SDIO and PCIe support, for devices > like the Raspberry Pi 3 or the Macbooks, is not yet implemented. > Also mbufs on the transmit path leak and are not yet freed. > ok stsp@ (patrick@) ~ ar5211.c ~ ar5212.c > Initialize status return code in ar5k_ar5211_reset() and > ar5k_ar5212_reset(), > in the same way as ar5k_ar5210_reset() does it. > Error messages were displaying uninitialized values: > "ath_reset: unable to reset hardware; hal status 3497439432" > ok mpi@ (stsp@) ~ bwfm.c ~ bwfmreg.h > Set default join preferences to prefer APs based on the RSSI, with > a boost set to prefer 5GHz APs. (patrick@) dev/pci ~ drm/radeon/radeon_combios.c > Turns out that the previous commit resolves the problem on Clang but at > the same time breaks the sparc64 build due to a very similar issue. To > make it work, stop returning a huge struct and instead make the caller > pass a pointer to the struct it allocated only once on the stack. > Similar diff from sashan@ > "yeah just make it work, and add it to the mop list" deraadt@ (patrick@) ~ drm/radeon/radeon_combios.c > The previous commit was deemed too invasive, so back it out and replace > it with a simple ifdef that uses the inline keyword if being compiled > with clang and doesn't use the keyword if being compiled with gcc. > ok kettenis@ (patrick@) dev/sdmmc ~ files.sdmmc + if_bwfm_sdio.c > Add bwfm(4), a driver for Broadcom FullMAC WiFi controllers. The > FullMAC, in comparison to SoftMAC, does most WiFi handling in the > firmware that's running on the controller. This means we have to > work around the net80211 stack while still implementing all the > WiFi interfaces to userland. > This driver is still in early development. So far it can connect > to open WiFis over the USB bus. SDIO and PCIe support, for devices > like the Raspberry Pi 3 or the Macbooks, is not yet implemented. > Also mbufs on the transmit path leak and are not yet freed. > ok stsp@ (patrick@) ~ sdmmc_io.c > Apparently one of the main concepts in the SDMMC I/O subsystem is that > the driver attached to an SDIO card always holds the lock and only > releases it once it detaches. Now with that in mind, some time ago > sdmmc_io_function_disable() and sdmmc_io_function_ready() were changed > to only assert the write lock and not take it, but not all of the tree > was converted. Change sdmmc_io_function_enable() as well, and remove > the enter/exit dance in the interrupt code. Apparently there is no > SDIO driver yet/anymore which would trigger those issues. > ok kettenis@ (patrick@) dev/tc ~ if_le_ioasic.c ~ ioasicreg.h > Implement media auto selection for DEC 3000. > diff from miod@ (mpi@) dev/usb ~ umcs.c > Correct logic check preventing the device to attach. > Fix regression introdruced in previous. > Found the hardway and diff from weerd@ (mpi@) ~ files.usb ~ usbdevs + if_bwfm_usb.c > Add bwfm(4), a driver for Broadcom FullMAC WiFi controllers. The > FullMAC, in comparison to SoftMAC, does most WiFi handling in the > firmware that's running on the controller. This means we have to > work around the net80211 stack while still implementing all the > WiFi interfaces to userland. > This driver is still in early development. So far it can connect > to open WiFis over the USB bus. SDIO and PCIe support, for devices > like the Raspberry Pi 3 or the Macbooks, is not yet implemented. > Also mbufs on the transmit path leak and are not yet freed. > ok stsp@ (patrick@) ~ usbdevs.h ~ usbdevs_data.h > regen (patrick@) ~ if_bwfm_usb.c > Similar to other USB based backend drivers, have a list of TX/RX > buffers. By storing the pointer of the TX mbuf in the TX buffer, > we can make sure to free the mbuf properly after TX completed. > This also seems to speed up the WiFi scan. > previous version ok stsp@ (patrick@) kern ~ kern_lock.c > Make _kernel_lock_held() always succeed after panic(9). > ok visa@ (mpi@) ~ uipc_domain.c > Reduces the scope of the NET_LOCK() in sysctl(2) path. > Exposes per-CPU counters to real parrallelism. > ok visa@, bluhm@, jca@ (mpi@) ~ kern_event.c > Move kq_kev from struct kqueue to the stack. > It turns this set of events per-thread without having to lock anything. > From Dragonfly 10f6680a4f6684751aaae0965abfe140f19e9231 > ok kettenis@, visa@, bluhm@ (mpi@) ~ kern_event.c > Move `kq_count' increase/decrease close to the corresponding TAILQ_* > insert/remove operation. > No functionnal change for the moment. However this helps to make this > code mp-safe. > Note that markers are still not, and wont be, counted. > ok visa@, jsing@, bluhm@ (mpi@) ~ uipc_mbuf.c > Move sysctl_mq() where it can safely mess with mbuf queue internals. > ok visa@, bluhm@, deraadt@ (mpi@) ~ kern_rwlock.c > Use a temporary variable in rw_status() to dereference only once the > volatile member of the struct. > Not forcing a memory read on every access, 3 in this function, might > reduce cache traffic in some cases. > Micro-optimization and diff provided by Mateusz Guzik. > ok visa@ (mpi@) ~ kern_pledge.c > Print the word pledge in the kernel log when there is a violation. > This should make it easier to figure out what is going on. Note > that the pledgecode it shows is only a guess which pledge(2) might > help. > OK deraadt@ semarie@ (bluhm@) ~ sys_process.c > Split sys_ptrace() by request type: > - control operations: trace_me, attach, detach, step, kill, continue. > Manipulate process relation/state or send a signal > - kernel-state get/set: thread list, event mask, trace state. > About the process and don't require target to be stopped, need copyin/out > - user-state get/set: memory, register, window cookie. > Often thread-specific, require target to be stopped, need copyin/out > sys_ptrace() changes to handle request checking, copyin/out to > kernel buffers with size check and zeroing, and dispatching to the > routines above for the real work. This simplfies the permission checks > and copyin/out handling and will simplify lock handling in the future. > Inspired in part by FreeBSD. > ok mpi@ visa@ (guenther@) lib/libsa ~ stand.h > Revert: libsa lacks memmove currently; a more thorough rename/change is > needed > problem noted by naddy@ (guenther@) net ~ if.c ~ if_etherip.c ~ pfkeyv2.c ~ rtsock.c > Reduces the scope of the NET_LOCK() in sysctl(2) path. > Exposes per-CPU counters to real parrallelism. > ok visa@, bluhm@, jca@ (mpi@) ~ if.c > ifioctl() cleanups. > Only one semantic chance for SIOCSIFXFLAGS, rtm_ifchg() is now only called > when the flags changed, not always. > ok bluhm@, visa@ (mpi@) ~ if.c > Fix a singed vs unsigned comparison resulting in an overflow of the > routing socket. > When clang became the default compiler, `if_flags' from `struct ifnet' > was changed from "short" to "unsigned short", to silence a warning. > Sadly the copy of these flags on the stack was still a "short" which > made the flags comparison always true, which in turn made ifioctl() > generates a RTM_INFO message for many ioctl(2). > Since my last commit, the flag comparison is done for every ioctl(2). > This made the kernel generate at least one routing message per ioctl, > resulting in a lot of RTM_DESYNC. > RTM_DESYNC problem reported by krw@ thanks to dhclient(8)'s noisiness! > ok krw@, patrick@ (mpi@) ~ if.c ~ if_var.h > Move sysctl_mq() where it can safely mess with mbuf queue internals. > ok visa@, bluhm@, deraadt@ (mpi@) netinet ~ igmp.c ~ ip_carp.c ~ ip_divert.c ~ ip_ether.c ~ ip_gre.c ~ ip_icmp.c ~ ip_input.c ~ ip_ipip.c ~ ipsec_input.c ~ tcp_usrreq.c ~ udp_usrreq.c > Reduces the scope of the NET_LOCK() in sysctl(2) path. > Exposes per-CPU counters to real parrallelism. > ok visa@, bluhm@, jca@ (mpi@) ~ ip_ipsp.c > Remove lie talking about splsoftclock(). (mpi@) netinet6 ~ raw_ip6.c > ramdisks do not contain MROUTING, so the local variable becomes unused. > wrap it. (deraadt@) ~ icmp6.c ~ ip6_divert.c ~ ip6_input.c > Reduces the scope of the NET_LOCK() in sysctl(2) path. > Exposes per-CPU counters to real parrallelism. > ok visa@, bluhm@, jca@ (mpi@) sys ~ eventvar.h > Move kq_kev from struct kqueue to the stack. > It turns this set of events per-thread without having to lock anything. > From Dragonfly 10f6680a4f6684751aaae0965abfe140f19e9231 > ok kettenis@, visa@, bluhm@ (mpi@) ~ exec_elf.h > Add Section Attirbute flag defines, some default values for n_type > and a couple of missing EM_* required to build devel/libdwarf and > devel/valgrind. > ok deraadt@, jasper@ (mpi@) ~ sysctl.h > Move sysctl_mq() where it can safely mess with mbuf queue internals. > ok visa@, bluhm@, deraadt@ (mpi@) ufs/ffs ~ ffs_softdep.c > At reboot all vnodes are flushed, but when it sleeps, the update > process has a chance to create new dirty vnodes. Resolving soft > dependencies adds vnodes to the dirty list. > In softdep_flushfiles() vnodes and softdeps are flushed in a loop. > But if they sleep, it is not guaranteed that all vnodes have been > flushed when the softdep worklist flush reports that nothing has > to be done. > The solution is to do a final vnode flush after the softdep worklist > has been flushed. Then the dirty list is empty and the final check > in dounmount_leaf() does not panic with "unmount: dangling vnode". > OK mikeb@ (bluhm@) == usr.bin =========================================================== 12/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin awk ~ main.c > Move the assignment to cmdname up before its first use. > From Ori Bernstein. > ok deraadt (tb@) ~ awk.h ~ b.c ~ lib.c ~ main.c ~ parse.c ~ run.c ~ tran.c > rename dprintf macro to DPRINTF to avoid collision with libc posix dprintf. > (amazing to me that the posix name was chosen though it specifically > collides with code Kernighan maintained since the V7). > ok millert (deraadt@) bc ~ bc.1 > be a bit clearer about -l: original diff from jan stary, but i've > tweaked it a bit; > ok otto (jmc@) ~ bc.1 > Jan Stary convinced me 4 * a(1) is a better example (otto@) ctags ~ print.c > Ensure that the tag and line number are always separated by at least a > single > space. > ok deraadt@ (anton@) mg ~ cscope.c ~ grep.c > Replace fgetln(3) with getline(3) which is portable and less error prone. > From Scott Cheloha (scottcheloha AT gmail), thanks! > OK bcallah (florian@) ssh ~ sshd_config.5 > clarify the order in which config statements are used. > ok jmc@ djm@ (benno@) ~ servconf.c TAGGED OPENBSD_6_2 > MFC: > revision 1.313 > date: 2017/10/04 18:49:30; author: djm; state: Exp; lines: +4 -4; > commitid: FQf3mxZmjBcd8ygd; > fix (another) problem in PermitOpen introduced during the channels.c > refactor: the third and subsequent arguments to PermitOpen were being > silently ignored; ok markus@ > ---------------------------- > ok deraadt@ (djm@) ~ sshkey.c TAGGED OPENBSD_6_2 > BIO_get_mem_data() is supposed to take a char* as pointer argument, > so don't pass it a const char* (djm@) ~ ssh.1 TAGGED OPENBSD_6_2 > mention SSH_USER_AUTH in the list of environment variables (djm@) ~ sshd_config.5 TAGGED OPENBSD_6_2 > trim permitrootlogin description somewhat, to avoid ambiguity; > original diff from walter alejandro iglesias, tweaked by sthen and myself > ok sthen schwarze deraadt (jmc@) ~ packet.c TAGGED OPENBSD_6_2 > log debug messages sent to peer; ok deraadt markus (djm@) tmux ~ tmux.1 > Fix description of history_size, from Campbell Barton. (nicm@) ~ tty-keys.c > kind should be S-Down not S-Up. (nicm@) ~ mode-tree.c ~ window-tree.c > Add C-n and C-p keys for tree mode, and choose the right initial line > when no panes. (nicm@) ~ mode-tree.c > Clear to end of line properly with UTF-8 present. (nicm@) ~ window-tree.c > Box around label in preview. (nicm@) ~ format.c ~ server-fn.c ~ server.c ~ tmux.h ~ window.c > Show exit status and time in the remain-on-exit pane text, mostly from > Timo Boettcher in GitHub issue 1103. (nicm@) == usr.sbin ========================================================== 13/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin apmd ~ apmd.c > Kill unused defines (jca@) bgpctl ~ bgpctl.8 ~ bgpctl.c ~ parser.c > Add "ssv" variant which turns a rib route output into a single > semicolon-seperated line, so that grep and such can operate easier; > then use tr ';' '\n' to convert it to normal form later. This helps > with scanning vast swathes of routes.. > ok benno claudio job (deraadt@) bgpd ~ bgpd.h > Add "ssv" variant which turns a rib route output into a single > semicolon-seperated line, so that grep and such can operate easier; > then use tr ';' '\n' to convert it to normal form later. This helps > with scanning vast swathes of routes.. > ok benno claudio job (deraadt@) pkg_add ~ OpenBSD/PackageRepository.pm > correct calls to this constructor *must* include a state. If they don't > you won't notice until the repository wants to output an error message, > which is bad. > if code breaks thanks to this commit, the call site needs to be fixed. > (espie@) ~ pkg_sign.1 > thinko > noticed by Bryan Vyhmeister, thx (espie@) ~ OpenBSD/PkgCreate.pm > clean-up option handling slightly, some more to be done. > Add a -S option that computes signatures, so that we may remove > code from bsd.port.mk that does the exact same thing, and no longer > work since the addition of @version. > For now, -S is exclusive of other package creation options, as it exits > as soon as it computes what it needs. (espie@) ~ pkg_add.1 > in -u mode, the pkgname is optional (espie@) ~ pkg_add.1 > reorder the description so that important parts come first. > Ditch "old style" signature checks that are no longer relevant at all. > (espie@) ~ OpenBSD/FwUpdate.pm > Add bwfm(4) to fw_update(1). > ok espie@ sthen@ (patrick@) smtpd ~ queue_fs.c > Use clock_gettime(CLOCK_REALTIME) instead of > gettimeofday+TIMEVAL_TO_TIMESPEC > ok jung@ (guenther@) spamdb ~ spamdb.8 ~ spamdb.c > Allow GREY entries to be removed by IP address. OK beck@ (millert@) ~ spamdb.c > add -G to usage(); (jmc@) syspatch ~ syspatch.sh > Don't return 1 if the syspatch release directory does not exist; it just > means that no patch is available yet. (ajacoutot@) ~ syspatch.sh > Better error if installurl points to a local directory. (ajacoutot@) ~ syspatch.sh > Loosen up installurl(5) check: it's OK to have a trailing slash. > (ajacoutot@) ~ syspatch.sh > Simplify and explicitely state when we cannot access our mirror. > (ajacoutot@) ~ syspatch.sh > Don't check for mirror availability but directly try to fetch SHA256.sig. > That > check was not safe (index needed to be enabled, top URL may not be > available > etc.). > based on a comment from tj@ (ajacoutot@) =============================================================================== _______________________________________________ owc mailing list [email protected] http://www.squish.net/mailman/listinfo/owc
