OpenBSD src changes summary for 2017-02-04 ==========================================
include/arpa/inet.h sys/arch/arm64/arm64 sys/arch/arm64/include sys/arch/arm64/stand/efiboot sys/dev/pci sys/dev/usb sys/kern sys/netinet sys/sys usr.bin/mandoc usr.bin/netstat usr.bin/tmux usr.sbin/dhcpd usr.sbin/smtpd usr.sbin/vmd == include =========================================================== 01/04 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/include arpa/inet.h ~ arpa/inet.h > Move the typedefs for in_{addr,port}_t from <sys/types.h> to > <netinet/in.h> and <arpa/inet.h> > ok and ports test naddy@ (thanks!) > ok krw@ beck@ millert@ (guenther@) ~ arpa/inet.h > The macro versions of htonl et al don't require them, but POSIX says > <arpa/inet.h> needs to provide uint16_t and uint32_t. > ok millert@ krw@ naddy@ (guenther@) == sys =============================================================== 02/04 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/arm64/arm64 ~ pmap.c ~ machdep.c > Remove gdb waitcount debug code and dummy print. (patrick@) ~ machdep.c > A bit of code and comment cleanup. (patrick@) ~ pmap.c > Read the data cache size instead of hardcoding some value. We might > want to start caching that value somewhere though. (patrick@) arch/arm64/include ~ atomic.h > Implement atomic_{set,clear}bits_int and membars for arm64. > With feedback from Dale Rahn, ok patrick@. (jsg@) arch/arm64/stand/efiboot ~ efiboot.c > In reality our kernel does not need to be aligned at a 256M boundary, > but simply at 2M boundary. This might have been a requirement from > armv7, but it's not true for arm64. This also helps reduce memory > holes below the kernel. (patrick@) dev/pci ~ if_iwm.c > Prevent netlock related deadlock with the X server during iwm(4) scans. > The issue appears at least with 7265 and 8260 chips as confirmed by > stsp@, tb@ and myself. > Fixed by unlocking NET_LOCK() at the beginning of iwmioctl() and > relocking it at the end. > Guidance and ok mpi@, ok stsp@. (pirofti@) ~ if_iwm.c > Prefer <sys/endian.h> to <machine/endian.h> > ok stsp@ (guenther@) dev/usb ~ dwc2/dwc2.c > Prefer <sys/endian.h> to <machine/endian.h> > ok visa@ (guenther@) kern ~ kern_proc.c > Update usage for ps/w to match previous commit (guenther@) netinet ~ in.h > Move the typedefs for in_{addr,port}_t from <sys/types.h> to > <netinet/in.h> and <arpa/inet.h> > ok and ports test naddy@ (thanks!) > ok krw@ beck@ millert@ (guenther@) sys ~ types.h > Move the typedefs for in_{addr,port}_t from <sys/types.h> to > <netinet/in.h> and <arpa/inet.h> > ok and ports test naddy@ (thanks!) > ok krw@ beck@ millert@ (guenther@) == usr.bin =========================================================== 03/04 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin mandoc ~ man_term.c ~ mdoc_term.c > Do not fix the default indent for all subsequent files; some may use > a different macro language and hence require a different indent. > You can see the effect with "man -a 1 host hostname". (schwarze@) netstat ~ mbuf.c > Include netinet/in.h for in_addr_t now it has moved out of sys/types.h. > gcc warned about this, clang fatally errored. (jsg@) tmux ~ log.c > Missing va_end, from Anton Lindqvist. (nicm@) == usr.sbin ========================================================== 04/04 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin dhcpd ~ dhcpd.conf.5 > Appropriately replace "domain name" with "hostname"; less misleading > tb@ was initially concerned about next-server but there were more > similar occurrences. Simple solution - "hostname" - proposed by jmc@ > ok from both (jca@) smtpd ~ rfc2822.c > Cast char to unsigned char for isspace() > ok gilles@ (guenther@) vmd ~ loadfile_elf.c > missing function prototype (mlarkin@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
