OpenBSD src changes summary for 2016-09-16 ==========================================
distrib/i386 distrib/miniroot distrib/sets gnu/usr.bin/binutils-2.17 gnu/usr.bin/perl lib/libcurses lib/libkvm libexec/ld.so regress/usr.bin sbin/dhclient share/man sys/arch/amd64/amd64 sys/arch/amd64/conf sys/arch/arm/include sys/arch/hppa/hppa sys/arch/i386/i386 sys/conf sys/ddb sys/dev sys/dev/acpi sys/dev/pv sys/dev/usb sys/dev/wscons sys/dev/wsfont sys/kern sys/lib/libsa sys/net sys/sys sys/uvm usr.bin/mg usr.bin/tmux usr.bin/wc usr.sbin/procmap == distrib =========================================================== 01/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib i386 - ramdiskB/Makefile - ramdiskB/list.local - ramdiskC/Makefile - ramdiskC/list.local > these build directories were left around after we moved to 1 floppy. > (deraadt@) - ramdiskA/Makefile - ramdiskA/list.local > oh, this directory can also go (deraadt@) miniroot ~ install.sub > Use 'umount -f' so failing to umount root does not blow up install/upgrade > while root (sic) causes are probed. > Problem reported by RD Thrush via bugs@ and reproduced locally. (krw@) ~ install.sub > Revert last. -f is a bad bad idea here. > ok deraadt@ (krw@) sets ~ lists/comp/mi ~ lists/man/mi > sync (deraadt@) ~ lists/base/md.alpha ~ lists/base/md.amd64 ~ lists/base/md.armv7 ~ lists/base/md.hppa ~ lists/base/md.i386 ~ lists/base/md.landisk ~ lists/base/md.loongson ~ lists/base/md.luna88k ~ lists/base/md.macppc ~ lists/base/md.octeon ~ lists/base/md.sgi ~ lists/base/md.socppc ~ lists/base/md.sparc64 ~ lists/comp/mi ~ lists/man/mi > sync (deraadt@) == gnu =============================================================== 02/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/gnu usr.bin/binutils-2.17 ~ ltconfig > Hardcode STRIP=/usr/bin/strip so that install -s works > ok natano@ (guenther@) usr.bin/perl ~ installperl > Don't check if target directory is writeable. This gets in the way of > noperm builds. > ok deraadt some time ago (natano@) == lib =============================================================== 03/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libcurses ~ Makefile + legacy_coding.3 > Include the legacy_coding(3) page which documents use_legacy_coding(), > probably missed at one update or another. Reported by Anthony Coulter, > discussed with jmc. (nicm@) libkvm ~ kvm_proc.c > handle fallout of moving the address maps to RBT code. > because userland doesnt have subr_tree, and the tree traversal in > this file is simple, this inlines the logic that the functions in > the kernel do. (dlg@) == libexec =========================================================== 04/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/libexec ld.so ~ m88k/rtld_machine.c > Remove the fallback to the __plt_{start,end} symbols, as the dynamic tags > with the same info (DT_88K_PLT{START,END} have been around since 5.4 > testing by aoyama@ (guenther@) ~ m88k/rtld_machine.c > Implement the DT_RELACOUNT optimization. Missed previously because I > didn't recognize that RELOC_BBASED_32 was the m88k name for a pure-relative > relocation > testing by aoyama@ (guenther@) == regress =========================================================== 05/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress usr.bin ~ ssh/misc/kexfuzz/kexfuzz.c > fix for newer modp DH groups (diffie-hellman-group14-sha256 etc) (djm@) ~ ssh/misc/kexfuzz/README > add a note on kexfuzz' limitations (djm@) == sbin ============================================================== 06/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin dhclient ~ dhclient.c > Revert use of rdaemon() until it doesn't break install/upgrade. > Problem cause deduced by kili@, reproduced by & ok rpe@. (krw@) == share ============================================================= 07/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man4/Makefile > Hook up Hyper-V man pages (mikeb@) == sys =============================================================== 08/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/amd64/amd64 ~ pmap.c > move the vm_page struct from being stored in RB macro trees to RBT > functions > vm_page structs go into three trees, uvm_objtree, uvm_pmr_addr, and > uvm_pmr_size. all these have been moved to RBT code. > this should give us a decent chunk of code space back. (dlg@) ~ db_trace.c > teach ddb(4) about CTF. currently it only loads the CTF and uses it on > amd64 > to lookup the number of function parameters. however having this basic > facility allows us to expand it's usage. > currently hidden behind the (disabled) DDBCTF kernel option as some of the > required tools are not available in base yet. in addition to that one > also needs recent bootblocks that load the .SUNW_ctf kernel section. > discussed with mpi@ over many a cider and ale in cambridge > feedback and ok guenther@ mpi@ (jasper@) arch/amd64/conf ~ GENERIC > Enable Hyper-V guest drivers (mikeb@) arch/arm/include ~ exec.h > Define PT_ARM_EXIDX. > ok guenther@ (kettenis@) arch/hppa/hppa ~ pmap.c > move the vm_page struct from being stored in RB macro trees to RBT > functions > vm_page structs go into three trees, uvm_objtree, uvm_pmr_addr, and > uvm_pmr_size. all these have been moved to RBT code. > this should give us a decent chunk of code space back. (dlg@) arch/i386/i386 ~ pmap.c > move the uvm_map_addr RB tree from RB macros to the RBT functions > this tree is interesting because it uses all the red black tree > features, specifically the augment callback thats called on tree > topology changes, and it poisons and checks entries as theyre removed > from and inserted back into the tree respectively. > ok stefan@ (dlg@) ~ pmap.c ~ pmapae.c > move the vm_page struct from being stored in RB macro trees to RBT > functions > vm_page structs go into three trees, uvm_objtree, uvm_pmr_addr, and > uvm_pmr_size. all these have been moved to RBT code. > this should give us a decent chunk of code space back. (dlg@) conf ~ GENERIC ~ files > teach ddb(4) about CTF. currently it only loads the CTF and uses it on > amd64 > to lookup the number of function parameters. however having this basic > facility allows us to expand it's usage. > currently hidden behind the (disabled) DDBCTF kernel option as some of the > required tools are not available in base yet. in addition to that one > also needs recent bootblocks that load the .SUNW_ctf kernel section. > discussed with mpi@ over many a cider and ale in cambridge > feedback and ok guenther@ mpi@ (jasper@) ddb ~ db_hangman.c > drop unneeded casting noise > pointed out by guenther@ in a separate diff (jasper@) ~ db_elf.c ~ db_extern.h + db_ctf.c + db_ctf.h + db_elf.h > teach ddb(4) about CTF. currently it only loads the CTF and uses it on > amd64 > to lookup the number of function parameters. however having this basic > facility allows us to expand it's usage. > currently hidden behind the (disabled) DDBCTF kernel option as some of the > required tools are not available in base yet. in addition to that one > also needs recent bootblocks that load the .SUNW_ctf kernel section. > discussed with mpi@ over many a cider and ale in cambridge > feedback and ok guenther@ mpi@ (jasper@) dev + rndis.h > Introduce a common header files for RNDIS protocol definitions > that both urndis(4) and hvn(4) can use. (mikeb@) dev/acpi ~ acpicpu.c > Suppress debugging output about CSD entries that say nothing we care about > and improve the output for those we might. > output examples from mikeb@ > ok kettenis@ (guenther@) dev/pv ~ if_hvn.c ~ rndisreg.h > Switch over to the generic RNDIS header (mikeb@) dev/usb ~ if_urndis.c ~ if_urndisreg.h > Switch urndis(4) over to the generic RNDIS header (mikeb@) dev/wscons ~ wsconsio.h TAGGED OPENBSD_6_0 > MFC: limit the number of fonts that can be loaded > ok millert mlarkin deraadt (jcs@) ~ wsconsio.h TAGGED OPENBSD_5_9 > MFC: limit the number of fonts that can be loaded > ok millert mlarkin deraadt (jcs@) dev/wsfont ~ wsfont.c TAGGED OPENBSD_6_0 > MFC: limit the number of fonts that can be loaded > ok millert mlarkin deraadt (jcs@) ~ wsfont.c TAGGED OPENBSD_5_9 > MFC: limit the number of fonts that can be loaded > ok millert mlarkin deraadt (jcs@) kern ~ subr_tree.c > remove a trailing \ > i mustnt have cleaned this up properly when i copied the tree.h code > from Ilya Kaliman (dlg@) ~ subr_hibernate.c > move the vm_page struct from being stored in RB macro trees to RBT > functions > vm_page structs go into three trees, uvm_objtree, uvm_pmr_addr, and > uvm_pmr_size. all these have been moved to RBT code. > this should give us a decent chunk of code space back. (dlg@) ~ vfs_bio.c ~ vfs_subr.c > move buf_rb_bufs from RB macros to RBT functions > i had to shuffle the order of some header bits cos RBT_PROTOTYPE > needs to see what RBT_HEAD produces. (dlg@) ~ vfs_cache.c ~ vfs_subr.c > move the namecache_rb_tree from RB macros to RBT functions. > i had to shuffle the includes a bit. all the knowledge of the RB > tree is now inside vfs_cache.c, and all accesses are via cache_* > functions. (dlg@) lib/libsa ~ cread.c > unifdef SAVE_MEMORY which is no longer set nor used > ok deraadt@ millert@ (jasper@) net ~ if_trunk.c > Reconfigure interface capabilities after switching trunkproto; ok mpi > (mikeb@) ~ switchofp.c > Fix the data storage size check. When we exceed the amount of available > storage within an mbuf packet header we have to allocate a cluster. > Pointed out by mikeb@ > ok mikeb@ (rzalamena@) sys ~ buf.h ~ vnode.h > move buf_rb_bufs from RB macros to RBT functions > i had to shuffle the order of some header bits cos RBT_PROTOTYPE > needs to see what RBT_HEAD produces. (dlg@) ~ namei.h ~ vnode.h > move the namecache_rb_tree from RB macros to RBT functions. > i had to shuffle the includes a bit. all the knowledge of the RB > tree is now inside vfs_cache.c, and all accesses are via cache_* > functions. (dlg@) ~ vnode.h > move RBT_PROTOTYPE under #ifdef _KERNEL (dlg@) uvm ~ uvm_addr.c ~ uvm_fault.c ~ uvm_map.c ~ uvm_map.h ~ uvm_mmap.c ~ uvm_unix.c > move the uvm_map_addr RB tree from RB macros to the RBT functions > this tree is interesting because it uses all the red black tree > features, specifically the augment callback thats called on tree > topology changes, and it poisons and checks entries as theyre removed > from and inserted back into the tree respectively. > ok stefan@ (dlg@) ~ uvm_addr.c ~ uvm_addr.h ~ uvm_map.h > move uaddr_free_rbtree from RB macros to RBT functions (dlg@) ~ uvm_aobj.c ~ uvm_device.c ~ uvm_map.c ~ uvm_object.c ~ uvm_object.h ~ uvm_page.c ~ uvm_page.h ~ uvm_pmemrange.c ~ uvm_pmemrange.h ~ uvm_vnode.c > move the vm_page struct from being stored in RB macro trees to RBT > functions > vm_page structs go into three trees, uvm_objtree, uvm_pmr_addr, and > uvm_pmr_size. all these have been moved to RBT code. > this should give us a decent chunk of code space back. (dlg@) ~ uvm_pmemrange.c ~ uvm_pmemrange.h > move uvm_pmemrange_addr from RB macros to RBT functions (dlg@) ~ uvm_addr.c > fix spelling of RBT_NEXT in a comment (dlg@) ~ uvm_pmemrange.c > fix some DEBUG code so its using the right rb tree code (dlg@) ~ uvm_map.h > put RBT_PROTOTYPE inside #ifdef _KERNEL (dlg@) == usr.bin =========================================================== 09/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin mg - theo.c ~ Makefile ~ main.c > like the sparc port, ancient wisdom is ancient. consign to the archives. > ok beck (tedu@) tmux ~ tmux.h ~ window.c > Swap watermarks from high (4096) to low (128) when we get full buffers > into the read callback several times in succession; swap back when we > see empty buffers several times. This hopefully limits how much programs > that print a lot for a long period can monopolize tmux (like large, fast > compiling), without penalizing programs that print a lot briefly (like > most curses applications). Helps a lot for me, the actual numbers may > need tweaking later. (nicm@) wc ~ wc.c > - Removed unnecessary string.h include > - Changed 'format_and_print' argument type to int64_t and casting > inside the function > - Declaring 'print_counts', 'format_and_print', and 'cnt' as static > - Remove unnecessary cast for NULL, and (void) casts from printfs, > 'mbtowc' and 'format_and_print' calls > - In 'cnt', change bufsz type from ssize_t to size_t to avoid > converting between pointers to integer types with different sign > when calling getline (catched when compiling with Clang) > - Use return instead of exit in main > OK jung@ (fcambus@) == usr.sbin ========================================================== 10/10 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin procmap ~ Makefile ~ procmap.c > procmap fumbles with uvm_map_addr structures, which are now in RBTs > it also does proper traversal of the tree (ie, it does FOREACH) > which in turn uses MIN and NEXT operations to iterate over the whole > tree. theyre complicated and need code. > so for now this pulls in subr_tree.c from the kernel and builds it > as part of procmap. that allows for traversal of the RBT using the > same code that the kernel uses. > it is a bit ugly though because procmap updates the pointers between > items in the tree so they point at local copies instead of kernel > addresses. its made worse because RBT code has pointers between > rb_entry structs, not between the nodes. > im putting this in now to unbreak the tree. it can be polished after > coffee/naps. (dlg@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
