OpenBSD src changes summary for 2015-09-08 ==========================================
bin/ksh distrib/notes lib/libkvm lib/libtls regress/usr.bin sbin/dhclient sbin/disklabel sys/arch/alpha/alpha sys/arch/amd64/amd64 sys/arch/arm/arm sys/arch/hppa/dev sys/arch/hppa/gsc sys/arch/hppa/hppa sys/arch/i386/i386 sys/arch/i386/pci sys/arch/loongson/dev sys/arch/m88k/m88k sys/arch/macppc/dev sys/arch/powerpc/powerpc sys/arch/sgi/dev sys/arch/sgi/gio sys/arch/sgi/sgi sys/arch/sgi/xbow sys/arch/sh/sh sys/arch/solbourne/solbourne sys/arch/sparc/dev sys/arch/sparc/sparc sys/arch/sparc64/dev sys/arch/sparc64/sparc64 sys/dev sys/dev/wscons sys/dev/wsfont sys/kern sys/sys usr.bin/nc == bin =============================================================== 01/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/bin ksh ~ c_ksh.c > Remove superfluous variable initialization; from Michael McConville. (jca@) == distrib =========================================================== 02/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib notes ~ armv7/contents > Typo; Eric Lalonde (miod@) == lib =============================================================== 03/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libkvm ~ kvm.c ~ kvm_proc2.c > port the changes just made to kvm_getfiles in kvm_file2.c r1.47 to > kvm_getprocs. > basically cope with the number of procs growing between when we get > the size and when we get the list. > ok guenther@ (dlg@) libtls ~ tls_client.c ~ tls_internal.h > Remove the default HTTPS port from tls_connect() - this is a TLS library, > not a HTTPS library. > ok beck@ (jsing@) == regress =========================================================== 04/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress usr.bin ~ mandoc/char/unicode/input.out_utf8 ~ mandoc/char/unicode/latin1.in ~ mandoc/char/unicode/latin1.out_ascii ~ mandoc/char/unicode/latin1.out_html ~ mandoc/char/unicode/nogroff.out_utf8 > Apparently, some recent update of Unicode data in the base system > changed the output of wcwidth(3) for some weird Unicode characters, > causing harmless whitespace changes in mandoc(1) output; fix up the > regression suite accordingly. > The processing of the characters themselves still works correctly, > as it did before, and that's what these tests are intended to make > sure. They were never intended to check for whitespace issues. > Problem reported by jsg@. (schwarze@) == sbin ============================================================== 05/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin dhclient ~ dhclient.c > Correct strsep() usage to free() correct string. Spotted by and > original diff from Michael McConville via tech@. Thanks! > ok millert@ beck@ guenther@ jca@ (krw@) disklabel ~ editor.c > Avoid a SIGSEGV with FGJ malloc.conf flags when a template is used. > Found by jsg@. OK jsg@ deraadt@ (millert@) == sys =============================================================== 06/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/alpha/alpha ~ pmap.c > Give the pool page allocator backends more sensible names. We now have: > * pool_allocator_single: single page allocator, always interrupt safe > * pool_allocator_multi: multi-page allocator, interrupt safe > * pool_allocator_multi_ni: multi-page allocator, not interrupt-safe > ok deraadt@, dlg@ (kettenis@) arch/amd64/amd64 ~ est.c ~ mem.c > convert several malloc(9) to mallocarray(9). > ok deraadt@ guenther@ "re-wrap the long lines" kettenis@ (semarie@) ~ pmap.c > Give the pool page allocator backends more sensible names. We now have: > * pool_allocator_single: single page allocator, always interrupt safe > * pool_allocator_multi: multi-page allocator, interrupt safe > * pool_allocator_multi_ni: multi-page allocator, not interrupt-safe > ok deraadt@, dlg@ (kettenis@) arch/arm/arm ~ pmap.c ~ pmap7.c > Give the pool page allocator backends more sensible names. We now have: > * pool_allocator_single: single page allocator, always interrupt safe > * pool_allocator_multi: multi-page allocator, interrupt safe > * pool_allocator_multi_ni: multi-page allocator, not interrupt-safe > ok deraadt@, dlg@ (kettenis@) arch/hppa/dev ~ apic.c > sizes for free(), when known; ok semarie (deraadt@) arch/hppa/gsc ~ harmony.c > sizes for free(), when known; ok semarie (deraadt@) arch/hppa/hppa ~ intr.c > sizes for free(), when known; ok semarie (deraadt@) arch/i386/i386 ~ esm.c ~ est.c ~ k6_mem.c ~ mem.c > convert several malloc(9) to mallocarray(9). > ok deraadt@ guenther@ "re-wrap the long lines" kettenis@ (semarie@) ~ esm.c ~ est.c ~ ioapic.c ~ k1x-pstate.c ~ mem.c ~ powernow-k7.c ~ powernow-k8.c ~ sys_machdep.c ~ via.c > sizes for free(); ok semarie (deraadt@) arch/i386/pci ~ pcibios.c > convert several malloc(9) to mallocarray(9). > ok deraadt@ guenther@ "re-wrap the long lines" kettenis@ (semarie@) ~ glxsb.c ~ pcibios.c > sizes for free(); ok semarie (deraadt@) arch/loongson/dev ~ bonito.c > sizes for free(); ok semarie (deraadt@) arch/m88k/m88k ~ pmap.c > Give the pool page allocator backends more sensible names. We now have: > * pool_allocator_single: single page allocator, always interrupt safe > * pool_allocator_multi: multi-page allocator, interrupt safe > * pool_allocator_multi_ni: multi-page allocator, not interrupt-safe > ok deraadt@, dlg@ (kettenis@) arch/macppc/dev ~ awacs.c ~ dbdma.c ~ i2s.c ~ macintr.c ~ openpic.c > sizes for free(); ok semarie (deraadt@) arch/powerpc/powerpc ~ pmap.c > Give the pool page allocator backends more sensible names. We now have: > * pool_allocator_single: single page allocator, always interrupt safe > * pool_allocator_multi: multi-page allocator, interrupt safe > * pool_allocator_multi_ni: multi-page allocator, not interrupt-safe > ok deraadt@, dlg@ (kettenis@) arch/sgi/dev ~ impact.c ~ spdmem_mainbus.c > sizes for free(); ok semarie (deraadt@) arch/sgi/gio ~ grtwo.c ~ newport.c > sizes for free(); ok semarie (deraadt@) arch/sgi/sgi ~ eisa_machdep.c ~ ip27_machdep.c ~ l1.c > sizes for free(); ok semarie (deraadt@) arch/sgi/xbow ~ xbridge.c ~ xheart.c > sizes for free(); ok semarie (deraadt@) arch/sh/sh ~ pmap.c > Give the pool page allocator backends more sensible names. We now have: > * pool_allocator_single: single page allocator, always interrupt safe > * pool_allocator_multi: multi-page allocator, interrupt safe > * pool_allocator_multi_ni: multi-page allocator, not interrupt-safe > ok deraadt@, dlg@ (kettenis@) arch/solbourne/solbourne ~ pmap.c > Give the pool page allocator backends more sensible names. We now have: > * pool_allocator_single: single page allocator, always interrupt safe > * pool_allocator_multi: multi-page allocator, interrupt safe > * pool_allocator_multi_ni: multi-page allocator, not interrupt-safe > ok deraadt@, dlg@ (kettenis@) arch/sparc/dev ~ cs4231.c ~ fd.c > sizes for free(); ok semarie (deraadt@) arch/sparc/sparc ~ intr.c ~ machdep.c ~ vm_machdep.c > sizes for free(); ok semarie (deraadt@) arch/sparc64/dev ~ msi.c > sizes for free(); ok semarie (deraadt@) arch/sparc64/sparc64 ~ machdep.c ~ vm_machdep.c > sizes for free(); ok semarie (deraadt@) dev ~ ipmi.c ~ systrace.c > sizes for free(); ok semarie (deraadt@) dev/wscons ~ wsdisplay.c ~ wsdisplay_compat_usl.c > sizes for free(); checked by semarie (deraadt@) dev/wsfont ~ wsfont.c > sizes for free(); checked by semarie (deraadt@) kern ~ subr_pool.c > Now that msleep(9) no longer requires the kernel lock (as long as PCATCH > isn't specified) the default backend allocator implementation no longer > needs to grab the kernel lock. > ok visa@, guenther@ (kettenis@) ~ subr_pool.c > Give the pool page allocator backends more sensible names. We now have: > * pool_allocator_single: single page allocator, always interrupt safe > * pool_allocator_multi: multi-page allocator, interrupt safe > * pool_allocator_multi_ni: multi-page allocator, not interrupt-safe > ok deraadt@, dlg@ (kettenis@) sys ~ pool.h > Give the pool page allocator backends more sensible names. We now have: > * pool_allocator_single: single page allocator, always interrupt safe > * pool_allocator_multi: multi-page allocator, interrupt safe > * pool_allocator_multi_ni: multi-page allocator, not interrupt-safe > ok deraadt@, dlg@ (kettenis@) == usr.bin =========================================================== 07/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin nc ~ netcat.c > Netcat could hang during write(2) although poll(2) reports that the > socket is writeable. This happens because netcat tries to write > more than the low water mark of the socket write buffer. With a > non-blocking socket you may get a short write, otherwise it blocks. > The latter could cause a total hang of the netcat process depending > on the upper protocol. So make the network connection non-blocking. > OK claudio@ millert@ (bluhm@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
