]>
Zhao Yanbai Git Server - minix.git/log
Ben Gras [Fri, 4 Feb 2011 13:42:54 +0000 (13:42 +0000)]
include - throw out gettiminglocks stuff from include
Ben Gras [Fri, 4 Feb 2011 13:42:17 +0000 (13:42 +0000)]
kernel - doesn't do lock timings any more
Ben Gras [Fri, 4 Feb 2011 13:34:43 +0000 (13:34 +0000)]
is - no more getlocktimings.
Erik van der Kouwe [Wed, 2 Feb 2011 14:53:49 +0000 (14:53 +0000)]
Fix comment mutilated by find/replace
Ben Gras [Mon, 31 Jan 2011 16:53:53 +0000 (16:53 +0000)]
/usr/Makefile - automatically update pkgin database too.
Ben Gras [Mon, 31 Jan 2011 00:58:50 +0000 (00:58 +0000)]
/usr/Makefile - automatically install git
Ben Gras [Sun, 30 Jan 2011 23:40:51 +0000 (23:40 +0000)]
some extra math functions from netbsd for netpbm.
Ben Gras [Sun, 30 Jan 2011 23:36:45 +0000 (23:36 +0000)]
always define PKG_TOOLS_BIN
Ben Gras [Fri, 28 Jan 2011 11:50:47 +0000 (11:50 +0000)]
<ansi.h> -> <minix/ansi.h>
Ben Gras [Fri, 28 Jan 2011 11:35:02 +0000 (11:35 +0000)]
<ansi.h> -> <minix/ansi.h>
Ben Gras [Fri, 28 Jan 2011 00:02:27 +0000 (00:02 +0000)]
take pkg_install out of the base system and pre-install it using the
pkgsrc binary packages.
rationale:
. pkg_install (which is the pkg_* tools) is entangled with pkgsrc,
not with minix, so tracking it from pkgsrc (easier than with
base system) makes more sense
. simplifies upstreaming minix specific changes for pkg_* tools
. reduce pkgsrc-in-basesystem maintenance burden
Ben Gras [Thu, 27 Jan 2011 23:59:02 +0000 (23:59 +0000)]
update package names in packages.install to new pkgsrc versions
Arun Thomas [Thu, 27 Jan 2011 12:18:33 +0000 (12:18 +0000)]
Remove code for kernel task stack initialization
We no longer have kernel tasks, so this code is unnecessary
Ben Gras [Wed, 26 Jan 2011 17:02:54 +0000 (17:02 +0000)]
add minix3.org pkgsrc mirror as backup source
Ben Gras [Wed, 26 Jan 2011 14:36:19 +0000 (14:36 +0000)]
added uint typedef declaration in <minix/types.h>
. needed for xmlrpc (needed for cmake (needed for compiler-rt))
Ben Gras [Wed, 26 Jan 2011 14:35:34 +0000 (14:35 +0000)]
added <complex.h> (needed for compiler-rt)
David van Moolenbroek [Tue, 18 Jan 2011 10:18:08 +0000 (10:18 +0000)]
Kernel: fix clearing IPC references resulting in system crash
Erik van der Kouwe [Tue, 11 Jan 2011 11:03:37 +0000 (11:03 +0000)]
Servers request TSC freq from kernel rather than each one measuring it individually
David van Moolenbroek [Fri, 7 Jan 2011 17:10:12 +0000 (17:10 +0000)]
RS: don't share executable images with a dead process
David van Moolenbroek [Fri, 7 Jan 2011 17:09:30 +0000 (17:09 +0000)]
RS: move setuid() hack to where it belongs.. again.
David van Moolenbroek [Fri, 7 Jan 2011 17:07:46 +0000 (17:07 +0000)]
DS: fix regression in ds_subscribe regex construction
David van Moolenbroek [Fri, 7 Jan 2011 17:06:24 +0000 (17:06 +0000)]
DS: update subscriptions for deleted keys upon process exit
David van Moolenbroek [Fri, 7 Jan 2011 17:04:43 +0000 (17:04 +0000)]
Kernel: fix notification delivery to non-ANY receivers
David van Moolenbroek [Tue, 21 Dec 2010 11:00:13 +0000 (11:00 +0000)]
remove fsck1 altogether
David van Moolenbroek [Tue, 21 Dec 2010 10:44:45 +0000 (10:44 +0000)]
change bitchunk_t from 16-bit to 32-bit
David van Moolenbroek [Mon, 20 Dec 2010 23:52:53 +0000 (23:52 +0000)]
sys/queue.h: add LIST_FOREACH_SAFE
Arun Thomas [Fri, 17 Dec 2010 13:47:11 +0000 (13:47 +0000)]
Fix multiboot for ACK-built images
Move the profiling buffer to the end of the data segment
David van Moolenbroek [Thu, 16 Dec 2010 09:46:26 +0000 (09:46 +0000)]
Kernel: statistical profiling fixes
- create name entries for forked processes as well;
- create name entries only for system processes.
Ben Gras [Wed, 15 Dec 2010 14:43:59 +0000 (14:43 +0000)]
vfs - use a static buffer instead of malloc()+free(), solving
recently appeared ENOMEM problems during exec().
Erik van der Kouwe [Sat, 11 Dec 2010 18:20:31 +0000 (18:20 +0000)]
Workaround for circular dependency problem in pkgadd
Arun Thomas [Fri, 10 Dec 2010 23:04:04 +0000 (23:04 +0000)]
libc and csu updates for ELF
Arun Thomas [Fri, 10 Dec 2010 22:20:12 +0000 (22:20 +0000)]
csu/build support for ELF
David van Moolenbroek [Fri, 10 Dec 2010 15:21:59 +0000 (15:21 +0000)]
mkfifo.c: fix ack warning
Arun Thomas [Fri, 10 Dec 2010 15:18:30 +0000 (15:18 +0000)]
unstack: ELF support
Arun Thomas [Fri, 10 Dec 2010 15:18:28 +0000 (15:18 +0000)]
file(1): Add ELF support
Arun Thomas [Fri, 10 Dec 2010 09:33:48 +0000 (09:33 +0000)]
Fix libname in bsd.prog.mk
Arun Thomas [Fri, 10 Dec 2010 09:27:56 +0000 (09:27 +0000)]
VFS/RS support for ELF
David van Moolenbroek [Wed, 8 Dec 2010 14:54:08 +0000 (14:54 +0000)]
RS: fix IPC privilege computation bug
Take into account the ALL and ALL_SYS cases when constructing proper
symmetrical IPC send masks. Fix system.conf accordingly, to keep
userland processes from sending to several non-interface servers and
drivers. Also fix IS's F4 formatting.
David van Moolenbroek [Tue, 7 Dec 2010 12:16:31 +0000 (12:16 +0000)]
system.conf: base ipc permissions on process names rather than labels
From now on, the "ipc" directive in system.conf refers to process names
instead of labels, similar to the "control" directive. The old, more
fine-grained approach is deemed unnecessary and cumbersome at this time.
As side effects, this patch unbreaks late IPC permission computation as
well as the filter driver.
David van Moolenbroek [Tue, 7 Dec 2010 10:32:42 +0000 (10:32 +0000)]
Kernel/RS: fix permission computation with 32+ system processes
David van Moolenbroek [Fri, 3 Dec 2010 13:19:15 +0000 (13:19 +0000)]
RS: do not restart file systems
Arun Thomas [Wed, 1 Dec 2010 16:24:50 +0000 (16:24 +0000)]
vfs: terminate string in rdlink_direct
Fixes test56 when compiled with GCC.
Erik van der Kouwe [Mon, 29 Nov 2010 08:30:30 +0000 (08:30 +0000)]
Fix missing parenthesis (suggested by Antoine Leca)
Ben Gras [Thu, 25 Nov 2010 15:25:17 +0000 (15:25 +0000)]
mkboot: include git id in image name if available.
Dirk Vogt [Wed, 24 Nov 2010 16:30:13 +0000 (16:30 +0000)]
removed unneeded global var
Dirk Vogt [Wed, 24 Nov 2010 16:29:50 +0000 (16:29 +0000)]
don't send FS_READY anymore
Ben Gras [Wed, 24 Nov 2010 13:44:42 +0000 (13:44 +0000)]
Fix for objdir feature of BSD make
contributed by Antoine Leca.
Dirk Vogt [Tue, 23 Nov 2010 19:34:56 +0000 (19:34 +0000)]
decouple file system server start/termination from mount/umount
Arun Thomas [Tue, 23 Nov 2010 11:12:06 +0000 (11:12 +0000)]
pkgsrc: Update git repo info
Arun Thomas [Tue, 23 Nov 2010 11:05:09 +0000 (11:05 +0000)]
Add machine/param.h
Arun Thomas [Fri, 19 Nov 2010 12:36:44 +0000 (12:36 +0000)]
For GCC/clang, don't build in ACK int64 routines
Arun Thomas [Fri, 19 Nov 2010 10:33:26 +0000 (10:33 +0000)]
Allow for overriding GCC var in test Makefile
Arun Thomas [Fri, 19 Nov 2010 10:28:17 +0000 (10:28 +0000)]
kernel: Use LIBDIR var in Makefile
Instead of using a hard-coded path, use the LIBDIR variable so
the library path can be overridden.
Tomas Hruby [Thu, 18 Nov 2010 16:37:40 +0000 (16:37 +0000)]
Inlining 64bit integer handling functions
- if gcc/clang is used the C functions that use long long can be
inlined to get better performance (yes, it makes difference)
Tomas Hruby [Thu, 18 Nov 2010 15:12:01 +0000 (15:12 +0000)]
kernel/debug.c includes u64.h (fix)
Dirk Vogt [Wed, 17 Nov 2010 15:10:20 +0000 (15:10 +0000)]
libdriver without mandatory driver main loop
Dirk Vogt [Wed, 17 Nov 2010 14:53:07 +0000 (14:53 +0000)]
Added possibility to inject input events to tty
M include/Makefile
A include/minix/input.h
M include/minix/com.h
M drivers/tty/keyboard.c
M drivers/tty/tty.c
M drivers/tty/tty.h
M include/minix/syslib.h
M lib/libsys/Makefile
A lib/libsys/input.c
Ben Gras [Wed, 17 Nov 2010 14:29:32 +0000 (14:29 +0000)]
revert r8783, as those entries are auto-generated now
Ben Gras [Wed, 17 Nov 2010 14:28:23 +0000 (14:28 +0000)]
. auto-generate per-directory .gitignore files based on $(CLEANFILES)
. gitignore .gitignore files except for the root one.
Ben Gras [Wed, 17 Nov 2010 14:27:23 +0000 (14:27 +0000)]
Updated some CLEANFILES macros in Makefiles.
Arun Thomas [Mon, 15 Nov 2010 15:59:59 +0000 (15:59 +0000)]
Update gitignore
Arun Thomas [Mon, 15 Nov 2010 11:10:02 +0000 (11:10 +0000)]
Remove redundant sha2 code from inet
Arun Thomas [Fri, 12 Nov 2010 18:38:10 +0000 (18:38 +0000)]
GCC/clang: int64 routines in C
Erik van der Kouwe [Thu, 11 Nov 2010 14:12:48 +0000 (14:12 +0000)]
Remove global variable 'i'
Ben Gras [Thu, 11 Nov 2010 02:00:12 +0000 (02:00 +0000)]
update for releases with git.
Arun Thomas [Sun, 7 Nov 2010 23:35:29 +0000 (23:35 +0000)]
Use int64 functions consistently
Instead of manipulating the u64_t type directly, use the
ex64hi()/ex64lo()/make64() functions.
Arun Thomas [Sat, 6 Nov 2010 20:43:49 +0000 (20:43 +0000)]
Import mkfifo(1) from NetBSD
Arun Thomas [Sat, 6 Nov 2010 20:41:51 +0000 (20:41 +0000)]
Import mktemp(1) from NetBSD
Arun Thomas [Sat, 6 Nov 2010 20:40:15 +0000 (20:40 +0000)]
Move minix/paths.h to paths.h
Also, merge in relevant macros from NetBSD
Ben Gras [Thu, 4 Nov 2010 21:11:33 +0000 (21:11 +0000)]
.gitignore: also ignore CVS and .svn dirs
- to make importing from cvs and svn repos smoother
Arun Thomas [Thu, 4 Nov 2010 01:13:59 +0000 (01:13 +0000)]
Remove revision IDs from files
Ben Gras [Tue, 2 Nov 2010 22:43:25 +0000 (22:43 +0000)]
add top-level .gitignore.
Ben Gras [Tue, 2 Nov 2010 22:05:40 +0000 (22:05 +0000)]
stdio: split vsprintf and vsnprintf
- workaround for linking problems
Ben Gras [Tue, 2 Nov 2010 22:02:50 +0000 (22:02 +0000)]
stdio/freopen.c: fill __iotab table with stream pointer
- lets fclose()d and then freopen()ed streams be fclose()d again
without error
Ben Gras [Tue, 2 Nov 2010 22:01:47 +0000 (22:01 +0000)]
stdio/fclose.c: reset _buf
- if the stream gets freopen()ed, avoid buffer being seen as valid
Ben Gras [Tue, 2 Nov 2010 21:27:04 +0000 (21:27 +0000)]
kernel: limit stacktrace depth to 50.
- Unlikely to discard useful information and avoids long
system lockups on machines configured to print system output to serial.
Tomas Hruby [Tue, 26 Oct 2010 21:08:00 +0000 (21:08 +0000)]
/proc/cpuinfo
- when /proc/cpuinfo is read procfs retrievs information about cpus
from the kernel, formats it and prints it
Tomas Hruby [Tue, 26 Oct 2010 21:07:50 +0000 (21:07 +0000)]
sys_getcpuinfo()
Tomas Hruby [Tue, 26 Oct 2010 21:07:27 +0000 (21:07 +0000)]
Kernel keeps information about each cpu
- kernel maintains a cpu_info array which contains various
information about each cpu as filled when each cpu boots
- the information contains idetification, features etc.
Tomas Hruby [Mon, 25 Oct 2010 16:21:23 +0000 (16:21 +0000)]
SMP - reduced TLB flushing
- flush TLB of processes only if the page tables has been changed and
the page tables of this process are already loaded on this cpu which
means that there might be stale entries in TLB. Until now SMP was
always flushing TLB to make sure everything is consistent.
Tomas Hruby [Mon, 25 Oct 2010 08:32:14 +0000 (08:32 +0000)]
Fixed unixbench Run script
- perl path has changed
Tomas Hruby [Thu, 21 Oct 2010 17:07:12 +0000 (17:07 +0000)]
if verbore=1 tell us who registers which irq handler
- a useful piece of information when debugging
Tomas Hruby [Thu, 21 Oct 2010 17:07:09 +0000 (17:07 +0000)]
ACPI pci-to-pci bridges
- every pci device which implements _PRT acpi method is considered to
be a pci-to-pci bridge
- acpi driver constructs a hierarchy of pci-to-pci bridges
- when pci driver identifies a pci-to-pci bridge it tells acpi driver
what is the primary and the secondary bus for this device
- when pci requests IRQ routing information from acpi, it passes the
bus number too to be able to identify the device accurately
Tomas Hruby [Thu, 21 Oct 2010 17:07:07 +0000 (17:07 +0000)]
Zero no more hardwired as BSP apic id
- the BSP apic id is written in the io apic redirection entries to
deliver the interrupts to BSP
Tomas Hruby [Thu, 21 Oct 2010 17:07:05 +0000 (17:07 +0000)]
VM implicitly aligns mappings to a page
- problem when a PCI BAR is not aligned
- together with David v. Moolenboek
Tomas Hruby [Thu, 21 Oct 2010 17:07:01 +0000 (17:07 +0000)]
single shot timer interrupts fix
- accidentaly this wasn't part of the SMP merge and the implementation
remained uncomplete with the timer keeping ticking periodically
- APIC timer is set for a signel shot and restarted everytime it
expires. This way we can keep the AP's trully idle
- the timer is restarted a little later before leaving to userspace
- LAPIC_TIMER_ICR is written before LAPIC_LVTTR so the newest value is
used
Tomas Hruby [Tue, 19 Oct 2010 17:07:21 +0000 (17:07 +0000)]
spurious and error interrupt apic handlers
- fixed spurious and error interrupt handlers
- not to hog the system the warning isn't reported every time, just
once every 100 times, similarly for the spurious PIC interrupts
Tomas Hruby [Tue, 19 Oct 2010 17:07:19 +0000 (17:07 +0000)]
BSP apic id
- BSP apic id used uninitialized causes problems
Tomas Hruby [Tue, 19 Oct 2010 17:07:17 +0000 (17:07 +0000)]
panic in dummy apic handlers
- as panic can do the same as printf now, let's rather panic when a
dummy apic interrupt vector handler is hit
Tomas Hruby [Tue, 19 Oct 2010 17:07:11 +0000 (17:07 +0000)]
Fixed BKL statistics
Tomas Hruby [Tue, 19 Oct 2010 10:30:15 +0000 (10:30 +0000)]
PCI driver debug output fix
- bus number (busnr) must be used instead of internal busind
Tomas Hruby [Fri, 15 Oct 2010 22:21:01 +0000 (22:21 +0000)]
A klib.S include fix
- by Antoine Leca
Tomas Hruby [Fri, 15 Oct 2010 22:20:01 +0000 (22:20 +0000)]
acpi driver Makefile fix
- by Antoine Leca
Ben Gras [Fri, 15 Oct 2010 11:25:40 +0000 (11:25 +0000)]
<minix/hash.h>
Ben Gras [Fri, 15 Oct 2010 09:10:14 +0000 (09:10 +0000)]
vm - hash table for block cache
Ben Gras [Fri, 15 Oct 2010 09:09:29 +0000 (09:09 +0000)]
vm - fix for not placing region outside requested mmap addr
Erik van der Kouwe [Fri, 15 Oct 2010 08:53:22 +0000 (08:53 +0000)]
e1000: add 82574L ethernet adapter (thx Niek for your comments)
Ben Gras [Tue, 12 Oct 2010 11:26:47 +0000 (11:26 +0000)]
profile command - correct logic error
- this caused profile to complain about frequency when
no command (start, stop) was given.
Ben Gras [Mon, 11 Oct 2010 15:35:43 +0000 (15:35 +0000)]
mfs - no more WRITE_IMMED
Ben Gras [Sun, 10 Oct 2010 22:33:23 +0000 (22:33 +0000)]
change SSIZE_MAX to INT_MAX instead of 32767.
With this change, suggested by Gautam Tirumala, ports for pkgin and
pkg_install are cleaner and so easier to upstream. Presumably other
ports will be smoother too.
There doesn't seem to be a reason SSIZE_MAX was so small to begin with.