OpenBSD src changes summary for 2016-11-13 to 2016-11-20 inclusive ==================================================================
Makefile bin/ksh distrib/miniroot distrib/sets etc/Makefile etc/examples/pkg.conf etc/group etc/mail/aliases etc/master.passwd gnu/usr.bin/texinfo lib/libcrypto lib/libpcap lib/libssl regress/sys regress/usr.sbin sbin/ifconfig share/man share/mk sys/arch/alpha/conf sys/arch/amd64/amd64 sys/arch/amd64/conf sys/arch/armv7/conf sys/arch/hppa/conf sys/arch/i386/conf sys/arch/landisk/conf sys/arch/loongson/conf sys/arch/loongson/dev sys/arch/loongson/include sys/arch/loongson/loongson sys/arch/luna88k/conf sys/arch/macppc/conf sys/arch/octeon/conf sys/arch/octeon/dev sys/arch/sgi/conf sys/arch/socppc/conf sys/arch/socppc/stand/boot sys/arch/sparc64/conf sys/conf sys/dev/ic sys/dev/pci sys/dev/pv sys/dev/usb sys/kern sys/lib/libkern sys/net sys/netinet sys/netinet6 sys/netmpls sys/nfs sys/sys usr.bin/calendar usr.bin/printf usr.bin/tmux usr.sbin/dhcpd usr.sbin/httpd usr.sbin/makefs usr.sbin/ospfd usr.sbin/smtpd usr.sbin/snmpd usr.sbin/switchctl usr.sbin/switchd usr.sbin/syspatch usr.sbin/tcpdump usr.sbin/vmd == Makefile ========================================================== 01/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/Makefile Makefile > Enable builds with a dedicated user that cannot elevate privileges or write > to /usr/src or /usr/xenocara. > Change /usr/{,x}obj to owner build:wobj with mode 770 and install the > systemwide makefiles before starting a build. The root of the noperm fs > containing DESTDIR should also be owned by build:wobj. > Developers will need to add their users to group wobj to be able to write > to /usr/{,x}obj/. > "push forward" deraadt; testing, input & ok rpe (tb@) == bin =============================================================== 02/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/bin ksh ~ ksh.1 > Markup improvements: > 1. Mark up reserved words intended to appear at the beginning of command > lines with .Ic. > 2. Mark up syntax elements intended to appear in the middle of command > lines with .Cm. > This is not only more systematic, but also makes internal search > commands like :tif<enter>, :tstop<enter>, and :t[[<enter> work in > man(1). (schwarze@) ~ sh.1 > Remove an escaped blank space and insert an No macro to keep \&; at a > distance from 'Ar command'. > This is schwarze's correct version of an initial incorrect diff from me. > (tb@) == distrib =========================================================== 03/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib miniroot ~ install.sub > Ensure, that the root disk question is added to the response file > mail. > found by and OK tb@ > OK halex@ krw@ (rpe@) sets ~ lists/comp/mi > sync (deraadt@) ~ lists/base/mi > sync (deraadt@) ~ lists/man/mi > sync (deraadt@) ~ lists/comp/mi > sync (tb@) ~ makeetcset ~ maketars > Fix up some permissions in RELEASEDIR and /var/sysmerge. > help, testing & ok rpe (tb@) == etc =============================================================== 04/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/etc Makefile ~ Makefile > Enable builds with a dedicated user that cannot elevate privileges or write > to /usr/src or /usr/xenocara. > Change /usr/{,x}obj to owner build:wobj with mode 770 and install the > systemwide makefiles before starting a build. The root of the noperm fs > containing DESTDIR should also be owned by build:wobj. > Developers will need to add their users to group wobj to be able to write > to /usr/{,x}obj/. > "push forward" deraadt; testing, input & ok rpe (tb@) ~ Makefile > Fix up some permissions in RELEASEDIR and /var/sysmerge. > help, testing & ok rpe (tb@) examples/pkg.conf ~ examples/pkg.conf > sync (sthen@) group ~ group > Introduce the build user and the wobj group that will soon be used as > defaults for building the system from source. > ok deraadt (tb@) mail/aliases ~ mail/aliases > Introduce the build user and the wobj group that will soon be used as > defaults for building the system from source. > ok deraadt (tb@) master.passwd ~ master.passwd > Introduce the build user and the wobj group that will soon be used as > defaults for building the system from source. > ok deraadt (tb@) == gnu =============================================================== 05/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/gnu usr.bin/texinfo ~ Makefile.bsd-wrapper > Set permissions of /usr/share/info/dir explicitly, so they don't depend on > the umask. > ok millert (tb@) == lib =============================================================== 06/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libcrypto ~ x509/x509_lu.c > Fix previous change to X509_STORE_add_{cert,crl} to not free the input > object in the error path - we don't own it. (miod@) ~ man/BIO_f_buffer.3 > Add Copyright and license. > In the SYNOPSIS, show function prototypes rather than macro definitions. > (schwarze@) ~ man/BIO_f_cipher.3 > Add Copyright and license. > Delete the useless statement that a void function does not return a value. > (schwarze@) ~ man/ECDSA_SIG_new.3 > various cleanup; (jmc@) ~ man/BIO_f_md.3 > Add Copyright and license. > For references to functions documented elsewhere, use .Xr, not .Fn. > Move historic information to the HISTORY section. (schwarze@) ~ man/BIO_f_null.3 > Copyright and license (schwarze@) ~ man/BIO_find_type.3 > Add Copyright and license. > BIO_method_type() is a function, not a macro. (schwarze@) ~ man/EC_GFp_simple_method.3 ~ man/EC_GROUP_copy.3 ~ man/EC_GROUP_new.3 ~ man/EC_KEY_new.3 ~ man/EC_POINT_add.3 ~ man/EC_POINT_new.3 > various cleanup; (jmc@) ~ man/ERR.3 ~ man/ERR_GET_LIB.3 ~ man/ERR_error_string.3 ~ man/ERR_get_error.3 ~ man/ERR_load_crypto_strings.3 ~ man/ERR_load_strings.3 ~ man/ERR_print_errors.3 ~ man/ERR_put_error.3 > various cleanup; (jmc@) ~ man/BIO_new.3 > Add Copyright and license. > Document that BIO_free(3), BIO_vfree(3), and BIO_free_all(3) accept NULL. > Delete the useless statement that void functions return no value. > (schwarze@) ~ man/BIO_read.3 > Add Copyright and license. > Bring in two clarifying sentences from OpenSSL. (schwarze@) ~ man/BIO_s_accept.3 > Add Copyright and license. > Bring in a bug fix and the new RETURN VALUES section from OpenSSL. > (schwarze@) ~ man/BIO_s_bio.3 > Add Copyright and license. > In the SYNOPSIS, provide prototypes rather than #define directives. > Bring in improvements to the EXAMPLES section from OpenSSL. (schwarze@) ~ man/BIO_s_connect.3 > Add Copyright and license. > Bring in one clarifying sentence from OpenSSL. (schwarze@) ~ man/BIO_s_fd.3 > Add Copyright and license. > Mention that BIO_set_fd(3) and BIO_get_fd(3) are macros. > Delete off-topic sentences about the return values of BIO_seek(3), > BIO_reset(3), and BIO_tell(3). (schwarze@) ~ man/BIO_push.3 ~ man/BIO_s_file.3 ~ man/BIO_s_mem.3 ~ man/BIO_s_null.3 ~ man/BIO_s_socket.3 > Copyright and license. (schwarze@) ~ man/Makefile + man/BIO_get_ex_new_index.3 > import from OpenSSL; > the referenced CRYPTO_get_ex_new_index(3) will follow shortly (schwarze@) ~ man/BIO_get_ex_new_index.3 > fix double word; (jmc@) ~ man/BIO_set_callback.3 > Add Copyright and license. > In the SYNOPSIS, provide prototypes rather than #define directives. > Name arguments consistently. > Remove lies about functions being macros. (schwarze@) ~ man/BIO_should_retry.3 > Add Copyright and license. > In the SYNOPSIS, show function prototypes, not #define directives. > (schwarze@) ~ man/BN_BLINDING_new.3 > Add Copyright and license. > Mention that BN_BLINDING_free() accepts NULL. > Adjust AUTHORS section to OpenBSD conventions. (schwarze@) ~ man/BN_CTX_new.3 > Add Copyright and license. > Mention that BN_CTX_free(3) accepts NULL. > Delete the useless statement that void functions return no value. > (schwarze@) ~ man/BN_CTX_start.3 > Add Copyright and license. > Delete the useless statement that a void function returns no value. > (schwarze@) ~ man/BN_add.3 > Add Copyright and license. > Merge a sentence from OpenSSL saying that BN_sub(3) > can do in-place manipulation. (schwarze@) ~ man/BN_add_word.3 > Add Copyright and license. > Delete duplicate specification of the error return value of BN_mod_word(3). > Fix a markup error in the remaining copy of the same return value. > (schwarze@) ~ man/BN_bn2bin.3 > Add Copyright and license. > More details about BN_hex2bn(3) lifted from the OpenSSL manual, > but worded less clumsily. (schwarze@) ~ man/BN_cmp.3 > Copyright and license (schwarze@) ~ man/BN_copy.3 > Add Copyright and license. > Merge documentation of BN_with_flags(3) from OpenSSL. (schwarze@) ~ man/BN_generate_prime.3 > Add Copyright and license. > In the SYNOPSIS, show prototypes rather than #define directives. > Clarify that BN_generate_prime_ex(3) takes the minimum length, > not the exact length. (schwarze@) ~ man/BN_mod_inverse.3 > Copyright and license (schwarze@) ~ man/BN_mod_mul_montgomery.3 ~ man/BN_mod_mul_reciprocal.3 ~ man/BN_new.3 > Add Copyright and license. > Mention that some *_free(3) functions accept NULL. > Delete useless statements about void functions not retuning a value. > (schwarze@) ~ man/BN_num_bytes.3 > Copyright and license (schwarze@) ~ man/BN_rand.3 > Add Copyright and license. > Merge one additional sentence stating a restriction on arguments from > OpenSSL. (schwarze@) ~ man/BN_set_bit.3 > Add Copyright and license. > Mention that shifting by negative numbers of bits is not allowed. > (schwarze@) ~ man/BN_swap.3 ~ man/BN_zero.3 > Copyright and license (schwarze@) ~ man/BUF_MEM_new.3 > Add Copyright and license. > Merge documentatin of BUF_MEM_grow_clean(3) and BUF_reverse(3) from > OpenSSL. > Mention that BUF_MEM_free(3) accepts NULL. (schwarze@) ~ man/CONF_modules_free.3 > Add Copyright and license. > Drop the empty RETURN VALUES section. (schwarze@) ~ man/CONF_modules_load_file.3 > Add Copyright and license. > Merge some typo fixes, a paragraph documenting another flag, > and some additional explanations and EXAMPLES from OpenSSL. (schwarze@) ~ man/DES_set_key.3 > Add the correct Copyright notices and licenses. > This is the first manual page i encounter that contains both parts > of text derived from the old SSLeay documentation, hence covered > by Tim Hudson's Copyright and available under Eric Young's original > SSLeay license, and parts changed or added later, hence covered by > the OpenSSL Copyright and license. Both licenses have to be included. > (schwarze@) ~ man/BUF_MEM_new.3 > grammar fix; (jmc@) libpcap ~ gencode.c ~ pcap.c ~ shlib_version > Add new DLT_OPENFLOW link-type to allow using tcpdump to debug switch(4), > eg. tcpdump -y openflow -i switch0 > Includes a minor bump for libpcap. > Feedback and OK rzalamena@ (reyk@) ~ gencode.c > Extend the DLT_OPENFLOW header to include the switch datapath id. > OK rzalamena@ (reyk@) libssl ~ man/BIO_f_ssl.3 > Add Copyright and license. > In the SYNOPSIS, show function prototypes rather than macro definitions. > Delete the useless NOTES section header. (schwarze@) ~ s3_pkt.c > Use defines instead of magic numbers and comments. (jsing@) ~ s3_srvr.c > Convert ssl3_get_client_kex_dhe() to CBS and perform some general code > cleanup, including the removal of a dead code path. > ok beck@ (jsing@) == regress =========================================================== 07/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress sys ~ net/rdomains/Makefile + net/rdomains/lo.ok > Regression test for creating a rdomain default loopback interface. > Even if a lo(4) interface will now be created for every rdomain, we > must ensure that condiguring such interface via netstart(8) still > work, for example: > $ cat /etc/hostname.lo1 > rdomain 1 > 127.0.0.1 > inet6 ::1/128 > alias 88.76.0.3/32 (mpi@) ~ net/rdomains/lo.ok > Check for correct flags now that ::1 is set automagically. (mpi@) ~ netinet/carp/carp_subr > Fix rdomain_is_used() now that lo(4) is created automagically. (mpi@) ~ net/rtable/delete/main.c ~ net/rtable/fullfeed/main.c > rtable_init() now creates the default routing table. (mpi@) ~ net/etherip/etherip_subr ~ net/vxlan/vxlan_subr > Sync rdomain_is_used() (mpi@) ~ net/pf_divert/Makefile ~ net/pf_divert/Remote.pm ~ net/pf_divert/remote.pl > Mark pf states created by this regress test with a label. Remove > all such states before basic testing to make sure that these subtests > are independent. (bluhm@) ~ net/pf_divert/Makefile ~ net/pf_divert/README > Test multiple combinations of divert connections. The socket cleanup > of the first connection must remove the pf state so that the second > connection can succeed. Disable two raw IP tests that do not work > with the current implementation in the kernel. (bluhm@) ~ netinet6/frag6/Makefile + netinet6/frag6/frag6_hop.py > Send fragment with Hop-by-Hop extension header after the fragment > header. I must not be processed. (bluhm@) ~ netinet6/frag6/Makefile ~ netinet6/frag6/frag6.py ~ netinet6/frag6/frag6_ext.py ~ netinet6/frag6/frag6_hop.py ~ netinet6/frag6/frag6_opt.py ~ netinet6/frag6/frag6_overatomic.py ~ netinet6/frag6/frag6_overdrop.py ~ netinet6/frag6/frag6_overhead.py ~ netinet6/frag6/frag6_overhead0.py ~ netinet6/frag6/frag6_overtail.py ~ netinet6/frag6/frag6_padding.py ~ netinet6/frag6/frag6_permute.py ~ netinet6/frag6/frag6_refrag.py ~ netinet6/frag6/frag6_shortatomic.py ~ netinet6/frag6/frag6_timeout.py ~ netinet6/frag6/frag6_udpatomic.py ~ netinet6/frag6/frag6_udpheader.py ~ netinet6/frag6/frag6_udppayload.py ~ netinet6/frag6/frag6_zerofirst.py ~ netinet6/frag6/frag6_zerosecond.py + netinet6/frag6/LICENSE > Put all make targets in a loop and print a little more what each > IPv6 fragment header test is doing. (bluhm@) ~ net/rtable/srp_compat.h > SRPL_FIRST() and SRPL_FOLLOW(). (mpi@) usr.sbin ~ syslogd/Makefile > Create the sys/syscall.ph header file on demand for the sendsyslog(2) > tests since Perl headers are not installed by default anymore. (bluhm@) ~ switchd/OFP.pm ~ switchd/run.pl > Remove code duplication for receiving/sending OFP packets and simplify > packet-in building. > ok reyk@ (rzalamena@) ~ switchd/OFP.pm ~ switchd/args-packet-jumbo.pm ~ switchd/run.pl > Add support for big reads in perl and fix the jumbo test, also remove > message type specific code from OFP encode. > ok reyk@ (rzalamena@) ~ switchd/args-packet-jumbo.pm ~ switchd/run.pl > Add support for OpenFlow 1.3.5 tests and make jumbo test use the new > version. > ok reyk@ (rzalamena@) == sbin ============================================================== 08/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin ifconfig ~ brconfig.c > Make the ifconfig switch0 output nicer using a single line and the > same keys as the actual commands (datapath, maxgroup, flowmax). > flowmax should eventually renamed to maxflow for consistency. > OK rzalamena@ (reyk@) == share ============================================================= 09/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man4/cdce.4 > update the link for communication devices; the actual document link > is quite lengthy and the original doc has been split into five parts, > so i've just linked to the man doc page; (jmc@) ~ man8/autoinstall.8 > Change the hash version string from $2a$ to $2b$. > ok rpe (tb@) ~ man8/autoinstall.8 > Improve the description of the hostname information. > OK landry@ krw@ (rpe@) ~ man9/malloc.9 > M_COUNTERS is a thing (dlg@) ~ man9/counters_alloc.9 > document counters_inc, counters_add, and counters_pkt. (dlg@) ~ man9/counters_alloc.9 > grammar; (jmc@) ~ man9/rtable_add.9 ~ man4/rdomain.4 > Automatically create a default lo(4) interface per rdomain. > In order to stop abusing lo0 for all rdomains, a new loopback interface > will be created every time a rdomain is created. The unit number will > be the same as the rdomain, i.e. lo1 will be attached to rdomain 1. > If this loopback interface is already in use it wont be possible to create > the corresponding rdomain. > In order to know which lo(4) interface is attached to a rdomain, its index > is stored in the rtable/rdomain map. > This is a long overdue since the introduction of rtable/rdomain. It also > fixes a recent regression due to resetting the rdomain of an incoming > packet reported by semarie@, Andreas Bartelt and Nils Frohberg. > ok claudio@ (mpi@) ~ man4/rdomain.4 > tweak previous; (jmc@) ~ man4/switch.4 > Update the switch(4) man page to tell that the device handles streams and > not packets anymore. > ok reyk@, jmc@ (rzalamena@) ~ man4/man4.loongson/ykbec.4 > Add support for lid state detection in ykbec(4). > Tested on a Lemote Yeeloong 8101B. > OK visa@, looks good to deraadt@ (fcambus@) ~ man4/man4.loongson/Makefile + man4/man4.loongson/htb.4 + man4/man4.loongson/leioc.4 > Add htb(4) and leioc(4). (visa@) ~ man4/com.4 ~ man4/pci.4 > Add loongson references. (visa@) ~ man8/release.8 > Update for builds with a dedicated build user. Eliminate some more > customization variables and refer to mk.conf(5) instead. > input & ok tj (tb@) ~ man5/mk.conf.5 > Update for builds with a dedicated user: BUILDUSER now defaults to build > and document WOBJGROUP and WOBJUMASK. Also: more consistent quotes. > ok jmc (tb@) ~ man8/release.8 > Add that RELEASEDIR must be writable by the build user. > ok tj (tb@) ~ man4/man4.loongson/htb.4 > Document a limitation on interrupt routing. (visa@) ~ man9/srpl_rc_init.9 > Rename SRPL_ENTER() to SRPL_FIRST() and SRPL_NEXT() to SRPL_FOLLOW(). > This allows us to introduce SRPL_NEXT() that can be used to start > iterating on an arbitrary member of an srp list, hence without calling > SRPL_ENTER(). > ok dlg@, jmatthew@ (mpi@) ~ man9/Makefile ~ man9/srpl_rc_init.9 + man9/SRPL_EMPTY_LOCKED.9 > Move LOCKED macros to teir own manual. > This will allow us to better document them and add examples. > Disccussed with dlg@ (mpi@) mk ~ bsd.obj.mk ~ bsd.own.mk > Enable builds with a dedicated user that cannot elevate privileges or write > to /usr/src or /usr/xenocara. > Change /usr/{,x}obj to owner build:wobj with mode 770 and install the > systemwide makefiles before starting a build. The root of the noperm fs > containing DESTDIR should also be owned by build:wobj. > Developers will need to add their users to group wobj to be able to write > to /usr/{,x}obj/. > "push forward" deraadt; testing, input & ok rpe (tb@) == sys =============================================================== 10/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/alpha/conf ~ Makefile.alpha > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/amd64/amd64 ~ vmm.c > Avoid an uninitialised value in rev 1.93. > ok martijn@ (jsg@) arch/amd64/conf ~ Makefile.amd64 > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/armv7/conf ~ Makefile.armv7 > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/hppa/conf ~ Makefile.hppa > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/i386/conf ~ Makefile.i386 > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/landisk/conf ~ Makefile.landisk > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/loongson/conf ~ Makefile.loongson > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) ~ files.loongson > Add drivers for PCI host bridge and built-in UARTs on Loongson 3A, > and put the pieces together by platform glue. > Feedback from miod@ (visa@) ~ GENERIC ~ RAMDISK > Turn Loongson 3A bits on. > The work on the platform was made possible by a hardware donation > from Loongson Technology. (visa@) arch/loongson/dev ~ kb3310.c > Add support for lid state detection in ykbec(4). > Tested on a Lemote Yeeloong 8101B. > OK visa@, looks good to deraadt@ (fcambus@) ~ mainbus.c + com_leioc.c + htb.c + htbreg.h + htbvar.h + leioc.c + leiocreg.h + leiocvar.h > Add drivers for PCI host bridge and built-in UARTs on Loongson 3A, > and put the pieces together by platform glue. > Feedback from miod@ (visa@) arch/loongson/include ~ intr.h > Add drivers for PCI host bridge and built-in UARTs on Loongson 3A, > and put the pieces together by platform glue. > Feedback from miod@ (visa@) ~ asm.h > Use the ErrorEPC register for curcpu(). (visa@) arch/loongson/loongson ~ machdep.c + generic3a_machdep.c > Add drivers for PCI host bridge and built-in UARTs on Loongson 3A, > and put the pieces together by platform glue. > Feedback from miod@ (visa@) ~ locore.S > Use the ErrorEPC register for curcpu(). (visa@) ~ generic3a_machdep.c > Fix end of interrupt routine to make pciide(4) work. > Clue stick from miod@ (visa@) arch/luna88k/conf ~ Makefile.luna88k > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/macppc/conf ~ Makefile.macppc > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/octeon/conf ~ Makefile.octeon > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/octeon/dev ~ octeon_intr.c > Drop pointless type casts. (visa@) arch/sgi/conf ~ Makefile.sgi > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/socppc/conf ~ Makefile.socppc > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) arch/socppc/stand/boot ~ fdt.c > Fix build now that ftd.h move to a MI directory. (mpi@) arch/sparc64/conf ~ Makefile.sparc64 > Clean up the kernel Makefile's: > - Remove the 'lint' target. lint has been removed with OpenBSD 5.2. > - Remove the 'tags' target. It does nothing of value. > - Replace 'clean::' with 'clean:', as requested by espie and millert, > and remove files from the 'clean' target, that are never generated. > - Don't create a file called 'depend' in 'make depend', but just do > nothing instead. > ok mpi tb (natano@) conf ~ files > Remove radix_mpath dragons. > This code insn't used since ART is the default. > ok vgross@ (mpi@) dev/ic ~ nvme.c ~ nvmereg.h > Mask non relevant bits when pritting version number. > Makes gonzalo@'s Macbookair7,1 NVMe report the correct version. > ok dlg@ (mpi@) ~ nvme.c > Do not use bus_space_{read,write}_8() even on LP64 archs, > some (broken) controllers require ordered split transfers. > From linux a310acd7a7ea53533886c11bb7edd11ffd61a036 > Tested by gonzalo@, ok dlg@ (mpi@) ~ re.c ~ rtl81x9reg.h > serialise posts to the txstart register. > without this it is possible for re_txeof to let the chip look at > and complete tx descriptors that re_start has written to the ring > but not accounted for in the software producer index. in this > situation no future interrupts may be generated for packets on the > ring, and tx will stall. > the serialisation is implemented by only doing the writes from tasks > via ifq_serialize. > this is part of a bigger diff that fixes lockups reported by haesbaert. > (dlg@) ~ re.c ~ rtl81x9reg.h > move toward having the a tx packet descriptor per tx ring entry. > previously you could have 64 packets on a ring with up to 1024 > descriptors. this mismatch makes the accounting for free space > complicated and allows for a data race with OACTIVE being set. > it will also let us push more packets. (dlg@) ~ re.c > dont restrict the number of tx descriptors to what can fit on a cpu page. > (dlg@) ~ re.c > allocate dmamaps for every descriptor in the tx ring. > previously there was a limit of 64 packets on the tx ring. soon we > will use as many descriptors as we want to fill the ring. (dlg@) ~ re.c > rework re_start and re_txeof to be more like other drivers. > the important change is to no longer account for packets and > descriptors separately, instead we just deal with available descriptors > on the ring. > this allows us to guarantee that oactive will only be set after > re_start has queued at least one packet, which in turn guarantees > that the chip will interrupt in the future and run re_txeof to clear > oactive. > without this change it was possible for a call of re_start to almost > fill the descriptor ring, followed by a second call that could fail > to post a packet because it had run out of descriptors and would > then set oactive. if re_txeof ran concurrently it could complete > all the previously posted packets without noticing that oactive had > been set and therefore needed to be cleared. in that situation > oactive would be set, but nothing would be set up to clear it > in the immediate future. > this fixes a lockup reported by haesbaert. (dlg@) dev/pci ~ ixgbe_type.h > Add new register definitions from the upstream; no binary change (mikeb@) ~ ixgbe_type.h > Add new structures and new fields; no functional change (mikeb@) + ixgbe_x550.c > Intel X550 specific module; not hooked up to the build yet > Obtained from the Intel ixgbe driver in FreeBSD. (mikeb@) ~ ixgbe.c ~ ixgbe_82598.c ~ ixgbe_82599.c ~ ixgbe_type.h ~ ixgbe_x540.c > Remove unused bits; no functional change (mikeb@) ~ ixgbe.c ~ ixgbe_82598.c ~ ixgbe_82599.c ~ ixgbe_phy.c ~ ixgbe_x540.c > Style and typo fixes; no binary change (mikeb@) ~ ixgbe_82598.c > Update 82598 specific module > A custom method to enable RX DMA is introduced because of the > upcoming API change in the common code. Otherwise it's just > a copy of the ixgbe_enable_rx_dma_generic. (mikeb@) ~ if_iwm.c > Make iwm(4) pass Tx/Rx MCS rates to bpf so they can be examined with > tcpdump. > ok mpi@ (stsp@) ~ ixgbe.c ~ ixgbe.h ~ ixgbe_82598.c ~ ixgbe_82599.c ~ ixgbe_phy.c ~ ixgbe_type.h ~ ixgbe_x540.c > Bring in new PHY bits, some of the 82599 specific stuff becomes shared > (mikeb@) ~ ixgbe.c ~ ixgbe.h ~ ixgbe_type.h ~ ixgbe_x540.c > EEPROM code refactoring for X540 (mikeb@) ~ ixgbe.c ~ ixgbe.h ~ ixgbe_82599.c ~ ixgbe_phy.c ~ ixgbe_type.h ~ ixgbe_x540.c > Autoconfiguration and firmware semaphore refactoring (mikeb@) ~ ixgbe_phy.c ~ ixgbe_type.h > Support for I2C combined messages as required by X550 (mikeb@) ~ ixgbe_phy.c > A firmware block check missed in previous commits (mikeb@) ~ ixgbe_type.h ~ ixgbe_x550.c > Remove virtualization-related bits (mikeb@) ~ ixgbe_phy.c > Remaining PHY bits; call SFP module detection code for X550 as well > (mikeb@) ~ if_ix.c > Minor code restructuring > - separate functions for delay value calculation and figuring out > whether or not we're doing SFP. > - MAC type detection is now done by ixgbe_set_mac_type; > - call {enable,disable}_tx_laser conditionally; > - unused TSO code bites the dust; > - default to "IFM_ETHER | IFM_AUTO" when we can't select any other > media type. (mikeb@) ~ if_ix.c > Reduce difference in ixgbe_initialize_receive_units to FreeBSD; > no binary change. (mikeb@) ~ if_ix.c ~ ixgbe_type.h > Support for new GPI signals, including X550 external PHY interrupt > This factors out the code configuring General Purpose Interrupts into > a separate function and provides LASI (Link Alarm Status Interrupt) > handler used by controllers in the X550 family lacking integrated PHY. > (mikeb@) ~ if_ix.c > Recognize active SPF+ DA modules as IFM_10G_SFP_CU > Reported and tested by Hrvoje Popovski, thanks! (mikeb@) ~ if_ix.c > Sync some changes for ixgbe_initialize_receive_units > Call a chip specific method to disable RX unit; > DPF (Drop [Unicast] Pause Frames) and PMCF (Pass MAC Control Frames) > bits should only be enabled on 82598 since others don't document them. > (mikeb@) ~ if_iwmreg.h > Set iwm's RTS retry limit to a more reasonable value. Prevents small frames > from getting stuck in the firmware's Tx queue for a long time behind other > frames large enough to trigger RTS but not making it out (e.g. due to bad > channel conditions). > ok tb@ (stsp@) ~ if_iwm.c > While setting up the basic rate bitmask for iwm's firmware, if the AP does > not specify basic rates for either the CCK or OFDM set, add just the most > basic rate to that set (1 Mbit/s in case of CCK, 6 Mbit/s in case of OFDM). > This behaviour matches what code comments seem to imply. > The previous code would add all possible basic rates in such cases. > So if all basic rates were CCK only, the code would add all possible OFDM > basic > rates on top. Then the firmware would send some frames at too high rates, > e.g. > RTS frames would be sent at 24Mbit/s which is a bit risky on noisy > channels. > ok tb@ (stsp@) dev/pv ~ hyperv.c ~ hypervvar.h > Minor cleanup, remove unused structure members (mikeb@) ~ if_hvn.c > Bump the number of TX descriptors to 256 > By increasing the number of available TX descriptors we also get > a larger NVS message ring. Altogether this results in about 30% > performance improvement. > While here, offset all TX descriptor IDs by a fixed amount so that > the Transaction ID 0 to descriptor ID mapping produces a non-zero > result making error handling more reliable. (mikeb@) ~ if_hvn.c > Always attempt to restart the transmission when OACTIVE is set > There's not much point in adding an additional condition to the > ifq_restart call so just get rid of those bits. Also while here, > remove a debugging message and make sure to cleanup the whole RNDIS > packet before sending it out to the hypervisor. (mikeb@) dev/usb ~ if_umb.c > Stop looking for further NCM datagrams if either length or offset of > the next datagram is zero. > Tested by otto@ > ok mpi@ kettnis@ otto@ (gerhard@) ~ uaudio.c > Use memcpy() to copy structures into freshly malloc'ated buffers, > as source and destination never overlap in this case. From > Michael W. Bombardieri <mb at ii.net>. (ratchov@) ~ usbdevs > Add Sierra Wireless EM7455 LTE module. > from Bryan Vyhmeister, ok sthen@ (gerhard@) ~ usbdevs_data.h ~ usbdevs.h > regen (gerhard@) kern ~ kern_pledge.c > Fix typo in comment: it's vm.loadavg, not kern.loadavg. > From patrick keshishian (tb@) ~ uipc_socket.c > Remove splnet() from socket kqueue code. > splnet() was necessary when link state changes were executed from > hardware interrupt handlers, nowdays all the changes are serialized > by the KERNEL_LOCK() so assert that it is held instead. > ok mikeb@ (mpi@) ~ init_main.c > Automatically create a default lo(4) interface per rdomain. > In order to stop abusing lo0 for all rdomains, a new loopback interface > will be created every time a rdomain is created. The unit number will > be the same as the rdomain, i.e. lo1 will be attached to rdomain 1. > If this loopback interface is already in use it wont be possible to create > the corresponding rdomain. > In order to know which lo(4) interface is attached to a rdomain, its index > is stored in the rtable/rdomain map. > This is a long overdue since the introduction of rtable/rdomain. It also > fixes a recent regression due to resetting the rdomain of an incoming > packet reported by semarie@, Andreas Bartelt and Nils Frohberg. > ok claudio@ (mpi@) ~ uipc_socket2.c > Bring back the SB_LOCK and SB_WANT flags to lock the socket buffers > in process context. The read/write lock introduced in rev 1.64 > would create lock ordering problems with the upcoming SOCKET_LOCK() > mechanism. The current tsleep() in sblock() must be replaced with > rwsleep(&socketlock) later. The sb_flags are protected by > KERNEL_LOCK(). They must not be accessed from interrupt context, > but nowadays softnet() is not an interrupt anyway. > OK mpi@ (bluhm@) lib/libkern ~ arch/mips64/memmove.S > Copy data by using 64-bit loads and stores rather than 32-bit > operations. This about doubles the routine's throughput. > No binary change on the 32-bit bootblocks > Discussed with miod@ long ago (visa@) net ~ if_bridge.c ~ if_etherip.c ~ pf.c > turn ipstat into a set of percpu counters. > each counter is identified by an enum value which correspond to the > original members of the ipstat struct. > ipstat_inc(ips_foo) replaces ipstat.ips_foo++ for the actual updates. > ipstat_inc is a thin wrapper around counters_inc. > counters are still returned to userland via the ipstat struct for now. > ok mpi@ mikeb@ (dlg@) - radix_mpath.c - radix_mpath.h ~ radix.c ~ rtable.c ~ rtable.h > Remove radix_mpath dragons. > This code insn't used since ART is the default. > ok vgross@ (mpi@) ~ radix.h > rnh_multipath can go, multipath support is dead. (mpi@) ~ if.c ~ if_loop.c ~ if_var.h ~ route.c ~ rtable.c ~ rtable.h > Automatically create a default lo(4) interface per rdomain. > In order to stop abusing lo0 for all rdomains, a new loopback interface > will be created every time a rdomain is created. The unit number will > be the same as the rdomain, i.e. lo1 will be attached to rdomain 1. > If this loopback interface is already in use it wont be possible to create > the corresponding rdomain. > In order to know which lo(4) interface is attached to a rdomain, its index > is stored in the rtable/rdomain map. > This is a long overdue since the introduction of rtable/rdomain. It also > fixes a recent regression due to resetting the rdomain of an incoming > packet reported by semarie@, Andreas Bartelt and Nils Frohberg. > ok claudio@ (mpi@) ~ if.c > It should be possible to create the default loopback interface. > Make rdomain's default loopback test pass again. (mpi@) ~ if.c > Automatically set ::1 on all default lo(4) interfaces. > Prodded by claudio@ (mpi@) ~ route.c > Reflect interface priorities when inserting RTF_CONNECTED routes. > When multiple RTF_CLONING routes exist for a given subnet, bringing > one of the interfaces up/down could make impossible to insert new > ARP/NDP entries. > In this case the first RTF_CONNECTED route of the multipath list no > longer corresponded to the wired interface because it had the same > priority as the carp(4) route. > This is another regression of supporting multiple RTF_CONNECTED routes. > Fix a bug with a CARP setup reported by stsp@ > ok stsp@ (mpi@) ~ if_pfsync.c ~ pf.c ~ pfvar.h > Instead of passing an extra mbuf pointer to pf_route(), it should > just use pd->m. Then pf_test() can also operate on pd.m and set > the *m0 value in the caller just before it returns. > OK sashan@ (bluhm@) ~ route.c > Properly initialize ``ifa''. > Note that dereferencing ``rt_ifa'' after calling rtfree(9) is generally > not safe. In this case we rely on the fact that rtredirect() is still > serialized with interface ioctl(2)s. > This function cries for a rewrite. > Reported by and ok jsg@ (mpi@) ~ pf_if.c ~ pf_ioctl.c > Kill recursive splsoftnet()s. > ok bluhm@ (mpi@) ~ bpf.c > Allow bpf_allocbufs() to fail when allocating memory. > This will help trading the KERNEL_LOCK for a mutex. > ok bluhm@ (mpi@) ~ bpf.h ~ if_switch.c ~ if_switch.h ~ switchofp.c > Add new DLT_OPENFLOW link-type to allow using tcpdump to debug switch(4), > eg. tcpdump -y openflow -i switch0 > Includes a minor bump for libpcap. > Feedback and OK rzalamena@ (reyk@) ~ bpf.c > Use goto in bpf{read,write}() to ease review of locked sections. > While here properly account for used reference in bpfwrite(). > ok bluhm@ (mpi@) ~ if_ppp.c > Kill outdated comment talking about sofnet. (mpi@) ~ if_spppsubr.c > if_link_state_change() doesn't need to be called under splsoftnet(), > it just schedule a task. (mpi@) ~ if_gre.c > Convert gre_send_keepalive() to timeout_set_proc(9) as it calls > ip_output(). (mpi@) ~ pf.c > The pf fragment reassembly code accepted IPv6 hop-by-hop headers > after fragment headers. Add an extra check that the hop-by-hop > header is always the first extension header after the IPv6 header. > Found by Antonios Atlasis; OK sthen@ mpi@ (bluhm@) ~ if_etherip.c > Remove splsoftnet()s in the ioctl(2) path. > Since soft-interrupt for network processing are no more, the > KERNEL_LOCK() is what protect two concurrent threads from > accessing data structures. > ok goda@ (mpi@) ~ bpf.h > Add DLT_USER1..15 with the same values as the upstream libpcap repository. > Suggested by reyk@ as a quick fix to unbreak ports/net/wireshark. (sthen@) ~ pipex.c > turn ipstat into a set of percpu counters. > each counter is identified by an enum value which correspond to the > original members of the udpstat struct. > udpstat_inc(udps_foo) replaces udpstat.udps_foo++ for the actual > updates. udpstat_inc is a thin wrapper around counters_inc. > counters are still returned to userland via the udpstat struct for > now. > ok mpi@ mikeb@ deraadt@ (dlg@) ~ ofp.h > Fix some style issues in header file: > - Use the right amount of zeros to match the header bit length; > - Use hex instead of bit shifting; (rzalamena@) ~ ofp.h > OpenFlow has a very simple finite state machine that is only > indirectly documented in the specification. Add the states to the > header as it is important for all implementations. Both switch(4) and > switchd(8) have to be updated to follow the fsm instead of answering > any requests at any time. > OK rzalamena@ (reyk@) ~ ofp.h > Add more documentation to header definitions that were missing. > (rzalamena@) ~ ofp.h > remove 'OpenFlow' prefix from version comments (reyk@) ~ ofp.h > Revert previous, I confused the columns (reyk@) ~ ofp.h > Add more maps to print. As discussed with rzalamena@ (reyk@) ~ switchofp.c > Fix a panic introduced with the memory leak correction. Use the saved > length instead of the packet instruction length to free the old > instruction. > ok reyk@ (rzalamena@) ~ ofp.h ~ switchofp.c > Define a group_id map and rename the defines accordingly. > Requested by rzalamena@ (reyk@) ~ rtable.c > Rename SRPL_ENTER() to SRPL_FIRST() and SRPL_NEXT() to SRPL_FOLLOW(). > This allows us to introduce SRPL_NEXT() that can be used to start > iterating on an arbitrary member of an srp list, hence without calling > SRPL_ENTER(). > ok dlg@, jmatthew@ (mpi@) ~ rtable.c > Make rtable_iterate(9) mpsafe by using the new SRPL_NEXT(9). > ok dlg@, jmatthew@ (mpi@) ~ if_switch.c ~ if_switch.h ~ ofp.h ~ switchofp.c > Extend the DLT_OPENFLOW header to include the switch datapath id. > OK rzalamena@ (reyk@) netinet ~ ip_icmp.c ~ ip_input.c ~ ip_output.c ~ ip_var.h ~ raw_ip.c > turn ipstat into a set of percpu counters. > each counter is identified by an enum value which correspond to the > original members of the ipstat struct. > ipstat_inc(ips_foo) replaces ipstat.ips_foo++ for the actual updates. > ipstat_inc is a thin wrapper around counters_inc. > counters are still returned to userland via the ipstat struct for now. > ok mpi@ mikeb@ (dlg@) ~ ip_input.c > use M_COUNTERS to allocate counters. > suggested by mpi@ and mikeb@ (dlg@) ~ ip_output.c > Automatically create a default lo(4) interface per rdomain. > In order to stop abusing lo0 for all rdomains, a new loopback interface > will be created every time a rdomain is created. The unit number will > be the same as the rdomain, i.e. lo1 will be attached to rdomain 1. > If this loopback interface is already in use it wont be possible to create > the corresponding rdomain. > In order to know which lo(4) interface is attached to a rdomain, its index > is stored in the rtable/rdomain map. > This is a long overdue since the introduction of rtable/rdomain. It also > fixes a recent regression due to resetting the rdomain of an incoming > packet reported by semarie@, Andreas Bartelt and Nils Frohberg. > ok claudio@ (mpi@) ~ ip_ipsp.c > TDB timers need a context process, so convert them to timeout_set_proc(9). > ok phessler@, bluhm@ as part of a larger diff (mpi@) ~ tcp_input.c > Use __func__ in panic strings to reduce noise when grepping. (mpi@) ~ tcp_input.c ~ tcp_var.h > Kill recursive splsoftnet()s. > While here keep local definitions local. > ok bluhm@ (mpi@) ~ ip_icmp.c > Inherit route label when creating dynamic routes for path MTU. > From Rivo Nurges; OK claudio@ mpi@ phessler@ (bluhm@) ~ ip_icmp.c > Bring icmp6_mtudisc_clone() in line with icmp_mtudisc_clone(). The > IPv4 dynamic route inherits the priority. Only clone from a valid > IPv6 route. Do not use splsoftnet() in IPv6. Some stylistic changes > to make the functions similar. > OK mpi@ (bluhm@) ~ tcp_input.c > Be consistent and do not use braces for single line statements. > Prodded by and ok bluhm@ (mpi@) ~ ip_output.c ~ udp_usrreq.c ~ udp_var.h > turn ipstat into a set of percpu counters. > each counter is identified by an enum value which correspond to the > original members of the udpstat struct. > udpstat_inc(udps_foo) replaces udpstat.udps_foo++ for the actual > updates. udpstat_inc is a thin wrapper around counters_inc. > counters are still returned to userland via the udpstat struct for > now. > ok mpi@ mikeb@ deraadt@ (dlg@) ~ ip_carp.c > Rename SRPL_ENTER() to SRPL_FIRST() and SRPL_NEXT() to SRPL_FOLLOW(). > This allows us to introduce SRPL_NEXT() that can be used to start > iterating on an arbitrary member of an srp list, hence without calling > SRPL_ENTER(). > ok dlg@, jmatthew@ (mpi@) ~ if_ether.c > Make rtable_iterate(9) mpsafe by using the new SRPL_NEXT(9). > ok dlg@, jmatthew@ (mpi@) netinet6 ~ ip6_input.c ~ ip6_output.c > Automatically create a default lo(4) interface per rdomain. > In order to stop abusing lo0 for all rdomains, a new loopback interface > will be created every time a rdomain is created. The unit number will > be the same as the rdomain, i.e. lo1 will be attached to rdomain 1. > If this loopback interface is already in use it wont be possible to create > the corresponding rdomain. > In order to know which lo(4) interface is attached to a rdomain, its index > is stored in the rtable/rdomain map. > This is a long overdue since the introduction of rtable/rdomain. It also > fixes a recent regression due to resetting the rdomain of an incoming > packet reported by semarie@, Andreas Bartelt and Nils Frohberg. > ok claudio@ (mpi@) ~ nd6.c > Kill recursive splsoftnet(). > ok phessler@, bluhm@ as part of a larger diff. (mpi@) ~ icmp6.c > Inherit route label when creating dynamic routes for path MTU. > From Rivo Nurges; OK claudio@ mpi@ phessler@ (bluhm@) ~ icmp6.c > Bring icmp6_mtudisc_clone() in line with icmp_mtudisc_clone(). The > IPv4 dynamic route inherits the priority. Only clone from a valid > IPv6 route. Do not use splsoftnet() in IPv6. Some stylistic changes > to make the functions similar. > OK mpi@ (bluhm@) ~ ip6_output.c ~ udp6_output.c > turn ipstat into a set of percpu counters. > each counter is identified by an enum value which correspond to the > original members of the udpstat struct. > udpstat_inc(udps_foo) replaces udpstat.udps_foo++ for the actual > updates. udpstat_inc is a thin wrapper around counters_inc. > counters are still returned to userland via the udpstat struct for > now. > ok mpi@ mikeb@ deraadt@ (dlg@) netmpls ~ mpls_raw.c > Return EOPNOTSUPP instead of calling a function to do only that. > ok phessler@ (mpi@) nfs ~ nfs_vfsops.c > Switch nfs_timer() to timeout_set_proc(9) as it might end up doing a > PRU_SEND. (mpi@) sys ~ malloc.h > add a type for counters. > ok mpi@ visa@ mikeb@ (dlg@) ~ percpu.h > add wrappers around common operations on counters. > specifically, counters_inc, counters_add, and counters_pkt. the latter > increments both a packet counter and adds to byte counter instead of just > one counter at a time. > part of a bigger diff thats ok mpi@ mikeb@ (dlg@) ~ socket.h > Automatically create a default lo(4) interface per rdomain. > In order to stop abusing lo0 for all rdomains, a new loopback interface > will be created every time a rdomain is created. The unit number will > be the same as the rdomain, i.e. lo1 will be attached to rdomain 1. > If this loopback interface is already in use it wont be possible to create > the corresponding rdomain. > In order to know which lo(4) interface is attached to a rdomain, its index > is stored in the rtable/rdomain map. > This is a long overdue since the introduction of rtable/rdomain. It also > fixes a recent regression due to resetting the rdomain of an incoming > packet reported by semarie@, Andreas Bartelt and Nils Frohberg. > ok claudio@ (mpi@) ~ socketvar.h > Bring back the SB_LOCK and SB_WANT flags to lock the socket buffers > in process context. The read/write lock introduced in rev 1.64 > would create lock ordering problems with the upcoming SOCKET_LOCK() > mechanism. The current tsleep() in sblock() must be replaced with > rwsleep(&socketlock) later. The sb_flags are protected by > KERNEL_LOCK(). They must not be accessed from interrupt context, > but nowadays softnet() is not an interrupt anyway. > OK mpi@ (bluhm@) ~ srp.h > Rename SRPL_ENTER() to SRPL_FIRST() and SRPL_NEXT() to SRPL_FOLLOW(). > This allows us to introduce SRPL_NEXT() that can be used to start > iterating on an arbitrary member of an srp list, hence without calling > SRPL_ENTER(). > ok dlg@, jmatthew@ (mpi@) == usr.bin =========================================================== 11/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin calendar ~ calendars/calendar.history ~ calendars/calendar.ushistory > hyrogen bomb related fixes, prompted by Carson Chittom; dates moved to > UTC. ok tb@ jmc@ (otto@) printf ~ printf.c > Delete useless call to setlocale(3). > While the C library function printf(3) is used here and is > locale-dependent, the printf(1) utility does not use the locale > dependent parts %lc and %ls. While POSIX requires LC_NUMERIC > support, we intentionally don't implement that in printf(3). > In summary, no functional change. > While here, sort headers and use the usual __dead usage() idiom. > Based on a patch from Jan Stary <hans at stare dot cz>. > OK bentley@ (schwarze@) ~ printf.1 > Fix two documentation bugs: > 1. Specifying a precision for the %s format character limits the number > of bytes printed, not the number of characters. > 2. Document the intentional POSIX violation of ignoring the locale. > Feedback and OK jmc@, bentley@. (schwarze@) tmux ~ cmd-rotate-window.c > rotate-window needs to unzoom, problem reported by Sean Haugh. (nicm@) ~ tmux.1 ~ window-copy.c > Add copy-pipe-and-cancel, from Artem Fokin. (nicm@) ~ server-client.c ~ tmux.h ~ tty-keys.c ~ tty.c > Initial attempt to make use of left and right margins if the terminal > supports them (that is, if it advertises itself as a VT420 - probably > just xterm). These are the vertical equivalent of the scroll region and > allow much faster scrolling of panes that do not take up the full width > of the terminal. (nicm@) ~ tty.c > And of course I just find a bug, disable previous for now. (nicm@) ~ server-client.c ~ tmux.h ~ tty.c > Turn on margins, with a couple of fixes (only limit to the pane for line > feeds, and do not move cursor to end for full width panes). (nicm@) ~ cmd-find.c ~ cmd-select-pane.c ~ tmux.h ~ window.c > The target validity check used window_pane_visible but that may be false > if the pane is zoomed, so instead add a new function to just check if > the pane is actually on screen (most commands still want to accept panes > invisible by zoom). Also reject panes outside the window for various > special targets. Problem reported by Sean Haugh. (nicm@) ~ window.c > Initialize modeprefix to 1 properly in window_pane_create, from Artem > Fokin. (nicm@) ~ cmd-new-session.c ~ server-client.c > Do not stop dragging when the wheel is pressed, from Artem Fokin. (nicm@) ~ cmd-rename-session.c > Don't error if renaming a session to its current name, from shik dot > chen at gmail dot com. (nicm@) ~ tty.c > Fix calculation of whether we need a region for drawing a cell (only if > full width and at the edge of the region), otherwise clear the region > entirely. (nicm@) ~ format.c ~ xmalloc.c ~ xmalloc.h > Key running commands for #() by the unexpanded command, and run them > again if the expanded form changes (otherwise at most once per second as > usual). Fixes issues reported by Gregory Pakosz. (nicm@) == usr.sbin ========================================================== 12/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin dhcpd ~ dispatch.c ~ icmp.c ~ options.c ~ pfutils.c > Replace bzero(3) with memset(3) > "Looks good" to deraadt@ (mestre@) httpd ~ httpd.conf.5 > specify ordering and precedence of location { } sections; > patch from grunk@; feedback and OK jmc@; OK florian@ (schwarze@) ~ httpd.h ~ parse.y ~ server.c > Move OCSP loading into a separate function - it is not part of the keypair > and this way we can give a separate specific error message. > ok beck@ reyk@ (jsing@) ~ server.c > Check the return value of tls_config_set_protocols(), now that it returns > an int. (jsing@) makefs ~ ffs.c ~ ffs.h ~ makefs.8 > Pick up the FS image size, the sector size and the bsize and fsize > parameters from the disklabel. > discussed with deraadt (natano@) ~ ffs.c > disklabel -w doesn't set bstart/bend, so makefs shouldn't do that either. > (natano@) ospfd ~ printconf.c > Print metric and type for "redistribute" in ospfd -v. > From Remi Locherer, ok claudio (sthen@) smtpd ~ bounce.c ~ filter.c ~ ioev.c ~ ioev.h ~ mda.c ~ mta_session.c ~ smtp_session.c > pass the user pointer as parameter to the io callback instead of having > the user dereference the io structure. > ok millert@ gilles@ (eric@) ~ ioev.c ~ ioev.h ~ mta_session.c ~ smtp_session.c > remove IO_TLSVERIFIED which is not a io event, and inline necessary code > where the callback functions are called for this event. > ok gilles@ (eric@) ~ ioev.c ~ ioev.h ~ smtpd.h ~ util.c > Add io api functions for dealing with buffered data, as wrapper around > their iobuf counterparts. > ok gilles@ sunil@ (eric@) ~ mta_session.c ~ smtp_session.c > fix regression introduced in previous commit > spotted by Heiko Zimmermann > ok gilles@ (eric@) ~ bounce.c ~ filter.c ~ ioev.c ~ ioev.h ~ mda.c ~ mta_session.c ~ smtp_session.c > add dedicated functions to set fd and callback on a struct io. > simplify io_init() prototype. > ok sunil@ gilles@ (eric@) snmpd ~ control.c ~ parse.y ~ snmpd.h ~ snmpe.c ~ traphandler.c > Add support for multiple listening sockets > One can now specify multiple "listen on" statements. The default is to > listen on 0.0.0.0 and ::, which means better handling of dual-stack > setups. ok sthen@ on a previous version, input and ok reyk@. (jca@) switchctl ~ Makefile ~ parser.c ~ parser.h ~ switchctl.c + ofpclient.c > Add a simple openflow client to dump switch information. The > implementation re-uses the openflow code from switchd. Not finished > yet, but it is better to continue working on it in the tree. Later > versions will also include a way to add flows from the command line - > this is being worked on for switchd and some code of the > "ASCII-to-flow" parser will be shared here as well. > As discussed with and lightly tested by rzalamena@ (reyk@) ~ parser.c > swap dump arguments to expect the address first (reyk@) ~ switchctl.8 > Describe the 'dump' command (reyk@) ~ ofpclient.c > Add a socket_setport() util function, in addition to socket_getport(), > and use it in switchctl to set a default port for "dump". > No functional change for switchd itself. (reyk@) ~ switchctl.8 > - Ar arg Ar arg is better written as Ar arg arg > - sort (jmc@) ~ Makefile > Add ofp_validate() for all supported versions. Will be used by switchctl, > no functional change. (reyk@) ~ ofpclient.c > Unbreak the tree: OFP_GROUP_ANY was renamed to OFP_GROUP_ID_ANY. > fix found joint with rpe, ok reyk (tb@) switchd ~ switchd.h ~ util.c > Add a socket_setport() util function, in addition to socket_getport(), > and use it in switchctl to set a default port for "dump". > No functional change for switchd itself. (reyk@) ~ proc.c ~ proc.h > Sync switchd(8) and vmd(8) with the latest proc.c changes. This sync > basically imports the imsg_flush() fix from reyk@ in httpd(8). > ok reyk@ (rzalamena@) ~ ofp13.c ~ ofp_common.c ~ switchd.h > Added the missing code to build flow-mod messages and to install > table-miss by default for switch(4). > ok reyk@ (rzalamena@) ~ ofp_common.c ~ switchd.h > Add ofp_validate() for all supported versions. Will be used by switchctl, > no functional change. (reyk@) ~ ofp13.c > Accept OFP_XM_T_ETH_TYPE. > OK rzalamena@ (reyk@) ~ ofp13.c ~ ofp_common.c ~ switchd.h ~ types.h > Add an abstraction layer / API to create flows including all matches > and instructions. This makes it easier to integrate flow features in > switchd and switchctl later. > Written and committed during a long flight. > OK rzalamena@ (reyk@) ~ ofp_common.c > In oflowmod, automatically open the instructions block with the first > instruction. This allows to make the grammar more flexible. (reyk@) ~ ofp10.c ~ ofp13.c ~ ofrelay.c > Fix some styles issues: break big lines and remove some useless spaces. > (rzalamena@) ~ genmap.sh > Fix file reference in generated comment OK rzalamena@ (reyk@) ~ genmap.sh > Define the struct constmap in each generated C file. (reyk@) ~ ofp13.c > Fix the flowmod header to send the correct command and buffer id. > OK rzalamena@ (reyk@) ~ ofp13.c > Small code improvements: > - Free ibufs on validation failures; > - Don't use htons() for flow-mod command; > ok reyk@ (rzalamena@) ~ genmap.sh > Exclude the description field from the generated maps by default. (reyk@) ~ ofp_map.h > Add more maps to print. As discussed with rzalamena@ (reyk@) ~ ofp_map.h ~ switchd.h ~ types.h > Remove leading _ from include guards as this violates the reserved space. > (reyk@) ~ genmap.sh ~ ofp_map.h ~ switchd.h ~ types.h > Define constmap in ofp_map.h to be shared along with ofp_map.c (reyk@) ~ ofp13.c ~ ofp_map.h > Define a group_id map and rename the defines accordingly. > Requested by rzalamena@ (reyk@) ~ ofp13.c > Make switchd(8) use more strings just like tcpdump(8). While here remove > action field display from features request. (rzalamena@) ~ ofp10.c ~ ofp13.c > The special LOCAL and CONTROLLER ports are valid source ports that > should be learned by the cache. Fixes "addlocal" ports on switch(4). > OK rzalamena@ (reyk@) syspatch ~ syspatch.sh > uniq -> sort -u (ajacoutot@) ~ syspatch.sh > Check for available space before installing a patch. (ajacoutot@) ~ syspatch.sh > Double the required size to make sure we have enough space for install(1) > safe > copy, the rollback tarball and in case we add new files. (ajacoutot@) ~ syspatch.sh > checkfs: ignore nonexistent files and extend comment. There's actually no > need > to double the size since files are replaced one by one. > create_rollback: create release directory with explicit permissions. > ls_installed: create release directory if it does not exist (temporary XXX) > for > early testers without a 6.1 hier(7); issue reported by Marc Benslahdine > (ajacoutot@) ~ syspatch.8 > Small wording tweak and a couple of fixes from jmc@ (ajacoutot@) ~ syspatch.8 ~ syspatch.sh > /bsd.rollbackXX -> bsd.syspatchXX to make it more obvious where it comes > from > and less prone to conflict. > Simplify ls_installed while here. (ajacoutot@) ~ syspatch.sh > If we install a new kernel, add /bsd twice (for size checking) when: > - we are on an MP system (/bsd.mp does not exist there) > - /bsd.syspatchXX is not present (create_rollback will add it) (ajacoutot@) ~ syspatch.sh > Merge comments altogether, it's easier to understand this way. > And drop some duplicate while here. (ajacoutot@) ~ syspatch.sh > Drop priv. to the _pkgfetch (for now) user when running ftp(1) and > signify(1). > Similar to what the installer does. > Listing installed patches and displaying the usage are now the only actions > allowed to a non-privileged user. > discussed with deraadt@ (ajacoutot@) ~ syspatch.sh > Re-add the root user checks in the quirks; that part is to be removed soon. > (ajacoutot@) tcpdump ~ interface.h ~ print-ofp.c ~ print-tcp.c ~ tcpdump.c > Add new DLT_OPENFLOW link-type to allow using tcpdump to debug switch(4), > eg. tcpdump -y openflow -i switch0 > Includes a minor bump for libpcap. > Feedback and OK rzalamena@ (reyk@) ~ Makefile ~ print-ofp.c + ofp_map.c + ofp_map.h > Updates tcpdump's OpenFlow printer to use the string maps from switchd > (manually synced). Also make the output a bit more like other printers. > OK rzalamena@ (reyk@) ~ print-ofp.c > miss_send_len can also be a CONTROLLER_MAXLEN string not just a number. > Requested by rzalamena@ (reyk@) ~ print-ofp.c > Don't print the field 'actions' on features reply, because in the new > OpenFlow 1.3.5 it is a reserved field that is not being used. (rzalamena@) ~ print-ofp.c > Add support for printing the missing OpenFlow set-queue action. > (rzalamena@) ~ print-802_11.c > Make tcpdump indicate basic rates listed in beacons with an asterisk. > ok tb@ (stsp@) ~ print-ofp.c > Extend the DLT_OPENFLOW header to include the switch datapath id. > OK rzalamena@ (reyk@) vmd ~ proc.c ~ proc.h > Sync switchd(8) and vmd(8) with the latest proc.c changes. This sync > basically imports the imsg_flush() fix from reyk@ in httpd(8). > ok reyk@ (rzalamena@) ~ i8259.c > Remove some old code that isn't needed anymore. (mlarkin@) =============================================================================== _______________________________________________ owc mailing list [email protected] http://www.squish.net/mailman/listinfo/owc
