Aaron LI [Fri, 21 Jun 2019 15:59:37 +0000 (23:59 +0800)]
rc: Move stop_boot() from rc.d/fsck to rc.subr and improve it
Move the stop_boot() function from rc.d/fsck to rc.subr, since we will
use it in other rc.d scripts. This follows what FreeBSD does.
Improve the stop_boot() function by bringing in the FreeBSD's version.
Minor cleanups.
Sascha Wildner [Fri, 21 Jun 2019 06:20:02 +0000 (08:20 +0200)]
kernel: Add NO_SYSCTL_DESCR option to exclude sysctl descriptions.
<sys/sysctl.h> has the code for it already, brought in in commit
8e82189d49ed5be4533e1a3d2609460e8aae81bf, but the option was not
added.
While here, remove some old no-op IPSec related options.
Sascha Wildner [Fri, 21 Jun 2019 05:34:24 +0000 (07:34 +0200)]
kernel/acpi_ec: Document sysctls better.
Sascha Wildner [Thu, 20 Jun 2019 21:29:49 +0000 (23:29 +0200)]
kernel: Use correct variable for debug.acpi.ec.timeout_backoff tunable.
François Tigeot [Thu, 20 Jun 2019 18:42:23 +0000 (20:42 +0200)]
drm/radeon: Upgrade to Linux 3.19.8
* Various bug fixes and hardware bug workarounds
* Fan control improvements, especially on Southern Islands (SI)
and Sea Islands (CI) GPUs
* Performance improvements due to better memory management
Sascha Wildner [Thu, 20 Jun 2019 17:39:01 +0000 (19:39 +0200)]
bsd-family-tree: Sync with FreeBSD (add NetBSD 8.1 and DragonFly 5.6).
Also some unrelated cleanup in the layout.
Aaron LI [Thu, 20 Jun 2019 13:44:11 +0000 (21:44 +0800)]
initrd: cd to '/' before umounting '/var'
The HOME in initrd is set to '/var/home', thus umounting '/var' may
fail due to device busy, although this failure would not block the boot
process.
Sascha Wildner [Wed, 19 Jun 2019 15:51:36 +0000 (17:51 +0200)]
kernel: Add whitespace between fw_stub.awk flags and their arguments.
It's more readable in the log file.
Jan Sucan [Mon, 5 Jun 2017 16:16:59 +0000 (16:16 +0000)]
fw_stub.awk: correct usage message
* Add information about optional version and parent for the firmware
image.
* Show that the -m option is mandatory.
Jan Sucan [Mon, 5 Jun 2017 15:54:09 +0000 (15:54 +0000)]
fw_stub.awk: fix processing of values for the options when separated by space
Matthew Dillon [Wed, 19 Jun 2019 05:59:07 +0000 (22:59 -0700)]
kernel - Change fill_kinfo_lwp() and fix top
* Change fill_kinfo_lwp(), an internal function used by kern_proc.c
and libkvm, to aggregate lwp data instead of replace it.
Note that fill_kinfo_proc() will zero the lwp sub-structure and is
already typically called before zero or more fill_kinfo_lwp() calls,
so the new aggregation essentially just works even though the API is
a bit different.
In addition, when getprocs is told to aggregate lwps the tid field
will be set to -1 since it is not applicable in the aggregation case.
* 'top' will now properly aggregate the threads belonging to a process
when thread mode 'H' is not in effect.
* Also allow top to display cpu percentages above 100%, since in the
aggregation case the sum of threads can easily exceed 100% of one core.
Requested-by: hsw
zrj [Tue, 18 Jun 2019 17:48:20 +0000 (20:48 +0300)]
openpam: Implement blacklist.
Mainly to avoid issues for users who might miss what `make upgrade`
warns about. Currently only includes previously active OPIE modules.
Do not include pam_cleartext_pass_ok.so it should not be relevant for
any DragonFly installations by now.
Matthew Dillon [Tue, 18 Jun 2019 18:54:08 +0000 (11:54 -0700)]
sshd - Change options defaults to be more secure
* Change the password_authentication default to 0. Passworded logins
are NOT considered secure by default in DragonFly.
* Put in a succinct comment for both the pam default (0) and the
password authentication default (also 0) to try to prevent them
from being improper changed back to 1 accidently.
Matthew Dillon [Tue, 18 Jun 2019 18:49:32 +0000 (11:49 -0700)]
Revert "sshd(8): Add USE_PAM handling defaults."
We really did not intend to turn on passworded logins or pam by default.
They need to be turned off by default so the sshd_config is secure by
default. PAM generally allows passworded logins which we do not consider
secure by default.
This reverts commit
19523637df5f6eb42b41f3dee51bd5d7c25d2219.
Aaron LI [Tue, 18 Jun 2019 16:30:40 +0000 (00:30 +0800)]
gpt(8): Explicitly avoid out of range in searching MBR partitions
Avoid running off the end of an array in case thing we're looking for
isn't there. This is probably impossible, but not obvious.
Obtained-from: NetBSD
Bug-report: #3015
Aaron LI [Tue, 18 Jun 2019 16:21:11 +0000 (00:21 +0800)]
jail.8: Document sysctl entries jail.{allow_raw_sockets,chflags_allowed}
Meanwhile, update the description of "jail.chflags_allowed" in the code.
Obtained-from: FreeBSD
Bug-report: #2935
Aaron LI [Mon, 17 Jun 2019 14:05:41 +0000 (22:05 +0800)]
jail: Implement read-only sysctl "jail.jailed"
Implement the read-only sysctl entry 'jail.jailed', which can be used to
determine if a process is running inside a jail (value is 1) or not
(value is 0).
NOTE: The current FreeBSD has such a sysctl entry called
'security.jail.jailed'. However, DragonFly BSD doesn't not have any
'security.jail.*' but only 'jail.*' sysctl entries.
Meanwhile, update /etc/rc to use this new sysctl entry to better deal
with the rc scripts with the 'nojail' keyword.
Also document this sysctl entry in the jail.8 man page.
This commit is based mostly on FreeBSD as well as the patch in bug
report #118.
Reviewed-by: dillon, mjg (Mateusz Guzik)
Bug-report: #118
Matthew Dillon [Mon, 17 Jun 2019 22:45:59 +0000 (15:45 -0700)]
drm - Fix system lockup bug in signal_pending_state()
* Fix a system lockup bug in signal_pending_state(). It was
possible for the routine to indicate that no signal was pending
for an interruptible state when a signal was in fact pending,
causing a live-lock in a lksleep() loop.
* The proper API is to allow any signal to interrupt when the
thread is in an interruptible state and to only allow a KILL
signal to interrupt when the thread is in a non-interruptible
state.
Testing-by: dillon, tuxillo
Reviewed-by: ftigeot
Sascha Wildner [Mon, 17 Jun 2019 17:28:18 +0000 (19:28 +0200)]
Nuke timed(8) (and timedc(8)). It is no longer used for time setting.
Adjust date(1) too for the removal (based on what FreeBSD did).
François Tigeot [Mon, 17 Jun 2019 14:20:21 +0000 (16:20 +0200)]
drm/linux: Add io_schedule_timeout()
zrj [Mon, 17 Jun 2019 06:45:42 +0000 (09:45 +0300)]
Bump __DragonFly_version for libopie removal.
While there, add entry to UPDATING.
zrj [Mon, 17 Jun 2019 06:35:29 +0000 (09:35 +0300)]
kernel: Unbundle libressl headers from src-sys.tar.bz2
This reverts commit
cebb579b84f544c775170301fea06cc74fb71ded.
No longer needed.
zrj [Mon, 17 Jun 2019 06:34:32 +0000 (09:34 +0300)]
kernel: Separate md5 from userland.
There are no good reasons to include userland header. Passing raw ctx
structure from kernel to userland or vice versa is a bug.
Requested-by: swildner
zrj [Mon, 17 Jun 2019 06:33:12 +0000 (09:33 +0300)]
kernel: Make <sys/md4.h> and <sys/md5.h> exclusive to kernel only.
With libmd removal this is now possible. Poison the sys/ headers.
While there, minor cleanup in md4c.c that is only used in smb.
zrj [Mon, 17 Jun 2019 06:30:06 +0000 (09:30 +0300)]
world: Remove libmd from the base.
Everything in base has been converted to use pure LibreSSL hash API.
The libmd now is moved to dports security/libmd for compatibility.
Also remove lone rmd160.h header that was added directly to include/.
zrj [Mon, 17 Jun 2019 06:28:42 +0000 (09:28 +0300)]
lib: Update dependencies.
No libraries depend on libmd, it can now be build in standard order.
zrj [Mon, 17 Jun 2019 06:25:13 +0000 (09:25 +0300)]
world: Remove OPIE from base.
The OPIE has been deprecated since
ed5666c1699a23a9ae3c0aca97dabaae71e26431
François Tigeot [Mon, 17 Jun 2019 13:31:05 +0000 (15:31 +0200)]
drm/linux: Restore wait_event*() functionality
* Our wait_event* routines do not (yet) use Linux wait queues.
Threads using them still have to be woken up with a different mechanism.
* Restore the wakeup*() calls directly using wait_queue_head_t pointers
* While there, also properly set the current Linux thread states in the
wait_event*() routines
Issue with the previous commit reported by zrj.
François Tigeot [Sun, 16 Jun 2019 20:37:26 +0000 (22:37 +0200)]
drm/linux: Rework wait queues
* Make the implementation much closer to the Linux one
* Do not directly try to wake up threads, but use an indirect function
call to do it, allowing drivers to override the default function
* Implement the expected default_wake_function() and
autoremove_wake_function() routines
François Tigeot [Sun, 16 Jun 2019 19:36:27 +0000 (21:36 +0200)]
drm/linux: Add LIST_HEAD_INIT()
Obtained from: NetBSD
Sascha Wildner [Sun, 16 Jun 2019 05:26:40 +0000 (07:26 +0200)]
Shorten some paths in various Makefiles.
Aaron LI [Sun, 16 Jun 2019 03:49:17 +0000 (11:49 +0800)]
rconfig(8): Update usage in program and manpage
Clarify the different usages between client and server modes.
Aaron LI [Sun, 16 Jun 2019 02:50:58 +0000 (10:50 +0800)]
rconfig(8): Allow '.', '-' and '+' in tag names
Originally, tag name only allows alphabets, digits and underscore. This
commit also allows tag names to have dot (.), minus (-) and plus (+)
symbols.
Bug-report: #2977
Matthew Dillon [Sun, 16 Jun 2019 02:24:46 +0000 (19:24 -0700)]
libkvm - Increase static buffer for devname()
* For now increase the internal static buffer for devname
to MAXPATHLEN (aka 1024). This may eventually be replaced
with SPECNAMELEN.
* This change is not exposed externally and it, or later changes,
should not break anything.
Requested-by: aly
Matthew Dillon [Sun, 16 Jun 2019 02:03:20 +0000 (19:03 -0700)]
libkvm - Increase ksw_devname from [32] to [64]
* Increase ksw_devname to 64 chars. This will help it fit longer
crypto name paths, plus drive serial numbers can be longer as well.
Requested-by: aly
Matthew Dillon [Sat, 15 Jun 2019 22:04:04 +0000 (15:04 -0700)]
kernel - Fix SMAP/SMEP caught user mode access part 2/2.
* Finish implementing SMAP exception handling support by
properly detecting it in trap() and generating a panic().
Otherwise the cpu just locks up in a page-fault loop without
any indication as to why on the console.
* To properly support SMAP, make sure AC is cleared on system calls
(it is already cleared on any interrupt or exception by the frame
push code but I missed the syscall entry code).
Sascha Wildner [Sat, 15 Jun 2019 20:37:25 +0000 (22:37 +0200)]
Update the pciconf(8) database.
June 12, 2019 snapshot from https://pci-ids.ucw.cz
Matthew Dillon [Sat, 15 Jun 2019 19:07:59 +0000 (12:07 -0700)]
kernel - Fix SMAP/SMEP caught user mode access part 1/2.
* Fix improper user mode access in hammer, caught by SMAP/SMEP.
Reported-by: Tim Darby
Aaron LI [Sat, 15 Jun 2019 11:51:21 +0000 (19:51 +0800)]
terminfo: Build and install st, st-16color and st-256color
Support the st (simple terminal) from the suckless project:
https://st.suckless.org/
Submitted-by: daftaupe
Bug-report: #3159
Aaron LI [Sat, 15 Jun 2019 10:55:11 +0000 (18:55 +0800)]
Use NULL instead of 0 for the third argument in fts_open()
Bug-report: #3138
Aaron LI [Sat, 15 Jun 2019 08:29:40 +0000 (16:29 +0800)]
Makefile: Require explicit target
Explicit target (e.g., buildworld, buildkernel) is required, otherwise
fail with an error.
Obtained-from: FreeBSD
Bug-report: #3158
Aaron LI [Sat, 15 Jun 2019 05:49:38 +0000 (13:49 +0800)]
libstand/hammer1: Fix vol_name to be vol_label
The 'vol_name' field in 'struct hammer_volume_ondisk' was changed to
'vol_label' in commit
6c39d27aedd299180e0b77179a8690d90d1e6e6d .
This fix enables libstand/hammer1.c to be build with -DTESTING.
Reported-by: daftaupe
Bug-report: #3185
Aaron LI [Sat, 15 Jun 2019 05:18:36 +0000 (13:18 +0800)]
u4b/audio: Fix panic by kfree(NULL)
DragonFly's kfree(9) doesn't allow a NULL pointer, while FreeBSD's
free(9) allows.
Reported-by: tse
Bug-report: #3192
Matthew Dillon [Fri, 14 Jun 2019 05:24:46 +0000 (22:24 -0700)]
iostat - Fix column count bug
* The maxshowdevs parameter was improperly conditionalized in the
list iteration, causing the list to be truncated early in some cases.
* Remove md*, pass*, and sg* by default, they are not usually
interesting.
* Display more columns by default if stdout is a terminal and is wide
enough.
Matthew Dillon [Fri, 14 Jun 2019 02:59:08 +0000 (19:59 -0700)]
libdevstat - Fix a bug in DS_SELECT_REMOVE
* An improper decrement cut the list scan short, causing wildcard
removals to not select all matching devices.
* This only really effects e.g. systat -vm and similar utilities.
Matthew Dillon [Thu, 13 Jun 2019 17:38:28 +0000 (10:38 -0700)]
kernel - Fix gdb / tracing
* The tstop fix broke gdb / tracing. Get gdb / tracing working
again.
Sascha Wildner [Wed, 12 Jun 2019 21:09:07 +0000 (23:09 +0200)]
Use some standard type names better.
struct sigaltstack -> stack_t
struct __siginfo -> siginfo_t
struct __sigset -> sigset_t
No functional change.
Sascha Wildner [Wed, 12 Jun 2019 20:34:10 +0000 (22:34 +0200)]
<sys/signal.h>: Adjust the type of stack_t's ss_sp from char * to void *.
François Tigeot [Wed, 12 Jun 2019 19:48:53 +0000 (21:48 +0200)]
drm/linux: Fix schedule_timeout()
Interruptible sleeps were returning wrong values.
Sascha Wildner [Tue, 11 Jun 2019 20:51:46 +0000 (22:51 +0200)]
mtree: Indent with spaces.
Sascha Wildner [Wed, 12 Jun 2019 07:47:41 +0000 (09:47 +0200)]
libc: Add missing #include "un-namespace.h" in a few files.
Sascha Wildner [Wed, 12 Jun 2019 07:46:09 +0000 (09:46 +0200)]
libc/pwcache: Remove some old RCS related stuff.
Sascha Wildner [Wed, 12 Jun 2019 06:27:21 +0000 (08:27 +0200)]
<sys/param.h>: Fix _DragonFly_version.
Justin C. Sherrill [Wed, 12 Jun 2019 01:26:44 +0000 (21:26 -0400)]
Changes for DragonFly 5.7.
Sascha Wildner [Tue, 11 Jun 2019 19:12:21 +0000 (21:12 +0200)]
libc: Use quotation marks around {un-,}namespace.h (private header).
Matthew Dillon [Tue, 11 Jun 2019 19:04:49 +0000 (12:04 -0700)]
world - Makme shm_open() more compatible with expectations
* Create a tmpfs mount in /var/run/shm and also create /var/run/shm/tmp
under it, both modes 1777, to support shm_open(). No longer base
shm_open() at the root directory.
* shm_open() and shm_unlink() now access files relative to /var/run/shm.
This fixes numerous ports and removes the need for numerous
port patches.
* Rejigger libraries using namespace.h/un-namespace.h to add _unlink() and
_unlinkat() into the mix.
Matthew Dillon [Tue, 11 Jun 2019 04:20:11 +0000 (21:20 -0700)]
installer - Default to HAMMER2 (2)
* Fix comments.
Reported-by: noob237
Matthew Dillon [Tue, 11 Jun 2019 00:53:46 +0000 (17:53 -0700)]
installer - Default to HAMMER2
* Change the installer default from HAMMER1 to HAMMER2.
* Adjust the nrelease build to print the location of the image files
when it finishes.
François Tigeot [Mon, 10 Jun 2019 19:34:39 +0000 (21:34 +0200)]
drm/linux: Add task_struct, rework sleep and wake up functions
Really implement wake_up_process(), signal_pending(),
signal_pending_state() and schedule_timeout()
Some fixes contributed by Matthew Dillon.
Sascha Wildner [Mon, 10 Jun 2019 12:30:26 +0000 (14:30 +0200)]
/etc/group: Add 'input', so it is available right away on new systems.
Pointed-out-by: zrj
Peeter Must [Mon, 3 Jun 2019 07:57:11 +0000 (10:57 +0300)]
evdev: Create device entries with the "input" group
* Add a new group "input" with an id 43.
* Make the devfs subsystem override the /dev/input/event*
devices with perm 0660 and root:input.
* This approach is standard in many Linux systems.
Sascha Wildner [Sun, 2 Jun 2019 17:23:22 +0000 (19:23 +0200)]
Remove lint checks in system headers.
Modern checkers do not (need to) pass this anymore to the build.
Tomohiro Kusumi [Sun, 2 Jun 2019 05:54:18 +0000 (14:54 +0900)]
sys/vfs/hammer: Rename hammer_flush_inode_done() -> hammer_sync_inode_done()
It's a counterpart for hammer_sync_inode().
François Tigeot [Sun, 2 Jun 2019 05:16:11 +0000 (07:16 +0200)]
drm: Sync drm_gem_object_lookup() with Linux 4.7.10
Sascha Wildner [Sat, 1 Jun 2019 15:03:30 +0000 (17:03 +0200)]
libm: Fix some -Wredundant-decls. signgam is in <math.h> already.
Sascha Wildner [Sat, 1 Jun 2019 08:23:09 +0000 (10:23 +0200)]
Correct some casts in printf arguments in various utilities.
Peeter Must [Sat, 1 Jun 2019 06:56:05 +0000 (09:56 +0300)]
kernel/evdev: create input devices with UID_ROOT and GID_WHEEL
* In preparation to follow the same scheme as in drm devices:
access rights will be set using the devfs system, see commit
82aec1d31805500239e50c9b6ed8d25802b0a17c.
Sepherosa Ziehau [Sun, 3 Mar 2019 12:51:53 +0000 (20:51 +0800)]
em/emx/igb: Merge Intel em-7.7.4 and igb-2.5.6
Most noticeably added I219 V6/V7 support for em/emx.
Sascha Wildner [Fri, 31 May 2019 15:36:37 +0000 (17:36 +0200)]
resident(8): Remove a.out support.
Sascha Wildner [Fri, 31 May 2019 15:36:08 +0000 (17:36 +0200)]
ldd(1): Remove unneeded inclusion of <a.out.h>.
Sascha Wildner [Fri, 31 May 2019 15:36:03 +0000 (17:36 +0200)]
gcore(1): Stop depending on <a.out.h>.
Sascha Wildner [Fri, 31 May 2019 15:32:08 +0000 (17:32 +0200)]
Remove symorder(1). It's no longer useful.
Sascha Wildner [Fri, 31 May 2019 13:58:40 +0000 (15:58 +0200)]
crunchide(1): Remove a.out, elf32 and ECOFF traces.
François Tigeot [Fri, 31 May 2019 05:23:05 +0000 (07:23 +0200)]
drm/radeon: Stop naming the kernel module "radeonkms"
It is named "radeon" in Linux.
Matthew Dillon [Fri, 31 May 2019 04:38:29 +0000 (23:38 -0500)]
vkernel - Adjust use of GDF_VIRTUSER
* Don't clear GDF_VIRTUSER until after exiting the critical section,
otherwise hardclock() will tick sys instead of user.
A user bound program should now show more reasonable values for
user%.
* Scrap some debug counters that are no longer applicable, cleaning
up a few cache line bounces.
Matthew Dillon [Fri, 31 May 2019 03:41:32 +0000 (22:41 -0500)]
dummynet - Poll only if operational, change default freq for vkernel
* Default frequency for vkernel is now 100 hz instead of 1000 hz.
The frequency can be changed at any time with a sysctl.
* The systimer is only configured when there are pipes or flows
present, and will be deconfigured when there aren't. This
saves us from doing any unnecessary polling.
* dummynet still has some considerable holes related to SMP
operation. It really needs a rewrite to some degree.
François Tigeot [Thu, 30 May 2019 15:42:44 +0000 (17:42 +0200)]
drm/linux: Implement i2c_bit_add_bus()
Sascha Wildner [Thu, 30 May 2019 13:03:40 +0000 (15:03 +0200)]
sysctl.8: Revert
39d69daecef529eb49d36fefa429c8ac08e7cbc1.
The manpage just says whether it's a number, a string or something
else, not the exact type.
Sascha Wildner [Thu, 30 May 2019 11:25:44 +0000 (13:25 +0200)]
<sys/sysctl.h>: Remove the unused CTL_HW_NAMES define.
Reported-by: zrj
Sascha Wildner [Thu, 30 May 2019 10:56:52 +0000 (12:56 +0200)]
kernel/sysctl: Switch kern.osrevision to showing __DragonFly_version.
It was tied to a historic define (BSD) that started as 199506 and was
sporadically bumped in the past until 200708. Revert the define back
to 199506, as it is not supposed to be bumped, and add a comment about
this (taken from NetBSD). We cannot remove these defines completely
because at least some are used by ports.
François Tigeot [Thu, 30 May 2019 08:06:06 +0000 (10:06 +0200)]
drm/linux: Add wait_event_interruptible_locked()
François Tigeot [Thu, 30 May 2019 07:51:56 +0000 (09:51 +0200)]
drm/linux: Add vmalloc() and vzalloc()
François Tigeot [Thu, 30 May 2019 07:49:40 +0000 (09:49 +0200)]
drm/linux: Add set_memory_uc()
François Tigeot [Thu, 30 May 2019 07:48:47 +0000 (09:48 +0200)]
drm/linux: Add gcd()
François Tigeot [Thu, 30 May 2019 07:47:17 +0000 (09:47 +0200)]
drm/linux: Add list_prev_entry()
Obtained-from: FreeBSD
François Tigeot [Thu, 30 May 2019 07:40:22 +0000 (09:40 +0200)]
drm/linux: Fix pci_map_page() arguments
Matthew Dillon [Wed, 29 May 2019 21:38:21 +0000 (14:38 -0700)]
vkernel - Restore vkernel build
* Finish cleaning up the vkernel pmap code so the build works again.
Tested-by: dillon, tested with a NFS boot.
Matthew Dillon [Wed, 29 May 2019 21:33:07 +0000 (14:33 -0700)]
kernel - Don't block in tstop() with locks held
* There are several places where the kernel improperly blocks on a
STOP signal while locks might be held. This is a particular problem
when PCATCH is specified e.g. in the middle of the NFS code. It is
meant to catch INTR but it also improperly allowed STOP to function
and left the vnode lock held.
Several other places in the kernel also use PCATCH and don't expect
the kernel to actually block indefinitely on a STOP.
* Don't block in STOP in these situations. Simply mark the thread as
stopped and wait until it tries to return to userland before actually
stopping.
Any kernel subsystems which desire to act on the STOP in-line instead
of upon return to userland can do so manually, as long as they release
all locks for the duration.
Sascha Wildner [Wed, 29 May 2019 12:41:04 +0000 (14:41 +0200)]
<sys/syslimits.h>: Clean up inclusion check and warning.
Sascha Wildner [Tue, 28 May 2019 19:17:34 +0000 (21:17 +0200)]
Clean up a few math related manualpages.
* In frexp.3, change the library to libc, because it is part of libc,
not libm (anymore).
* In fpclassify.3 and signbit.3, remove the LIBRARY section, because
all these are macros from <math.h>.
Matthew Dillon [Tue, 28 May 2019 18:57:59 +0000 (11:57 -0700)]
kernel - Build 'evdev' into the kernel
* Build evdev into the kernel along with its EVDEV_SUPPORT option.
Requested-by: peeter
Matthew Dillon [Mon, 27 May 2019 00:20:48 +0000 (17:20 -0700)]
kernel - Refactor scheduler weightings part 2/2.
* Change the default fork()ing mechanic from 0x80 (random cpu) to
0x20 (best cpu). We no longer need to mix it up on fork because
weight4 now works.
* The best cpu algorithm has a number of desirable characteristics
for fork() and fork()/exec().
- Will generally start the child topologically 'close' to the parent,
reducing fork/exec/exit/wait overheads, but still spread the children
out while machine load is light. If the child sticks around for
long enough, it will get spread out even more optimally. If not,
closer is better.
- Will not stack children up on the same cpu unnecessarily (e.g. parent
fork()s a bunch of times at once).
- Will optimize heavy and very-heavy load situations. If the child
have nowhere else reasonable to go, this will schedule it on a
hyper-thread sibling or even on the same cpu as the parent. Depending
on the load.
* Gives us around a 15% improvement in fork/exec/exit/wait performance.
* Once a second we clear the td_wakefromcpu hint on the currently
running thread. This allows a thread which has become cpu-bound
to start to 'wander' afield (though the scheduler will still try to
avoid moving it too far away, topologically).
Sascha Wildner [Mon, 27 May 2019 16:09:05 +0000 (18:09 +0200)]
kernel/pmap: Clean up no longer used MALLOC_DEFINE.
Sascha Wildner [Sun, 26 May 2019 17:43:17 +0000 (19:43 +0200)]
<sys/cdefs.h>: Remove the old unused __DF_VISIBLE.
Nothing depends on it anymore and nothing sets _DRAGONFLY_SOURCE or
_NETBSD_SOURCE. We handle all non-POSIX visibility with __BSD_VISIBLE
quite well.
Pointed-out-by: zrj
Sascha Wildner [Sun, 26 May 2019 13:30:40 +0000 (15:30 +0200)]
Remove expand(1) from the bootstrap tools.
Sascha Wildner [Sun, 26 May 2019 12:54:53 +0000 (14:54 +0200)]
route(8): Simplify the keywords handling.
Adapted from FreeBSD. This eliminates paste(1) as a bootstrap tool.
Sascha Wildner [Sun, 26 May 2019 12:02:07 +0000 (14:02 +0200)]
dump(8): Remove some unneeded defines.
Matthew Dillon [Sun, 26 May 2019 16:49:02 +0000 (09:49 -0700)]
kernel - VM rework part 21 - Start resynchronizing the vkernel
* Fix some minor syntax errors.
* The vkernel is still not operational in master. It will be a little
while. Even though the vkernel retains the old pmap mechanism, there
are some i's to dot and t's to cross in the expectations the mainline
kernel has of the APIs.
Reported-by: swildner
Matthew Dillon [Sun, 26 May 2019 16:47:29 +0000 (09:47 -0700)]
kernel - Backout 'Reduce token backoff'
* Return the backoff to 4096. Basically there are multiple situations
here where a smaller backoff works better, and multiple situations
where a larger backoff works better. For now, return the setting to
its former glory and don't mess with it.
Matthew Dillon [Sun, 26 May 2019 16:37:51 +0000 (09:37 -0700)]
kernel - Refactor scheduler weightings part 1/2.
* Refactor the scheduler's weightings and fix a few issues that
have cropped up due to breaking previous tunings. This gets
our pgbench results back to normal.
* There will probably be a follow-up commit with a bit more
tuning work, particular with regards to resetting the
td_wakefromcpu field which we currently do not do at all.
* Increase weight1 (keep thread on current CPU) slightly,
implement weight4, and re-tune the algorithm.
* Break-out the IPC pairing control fields into two new
sysctls, kern.usched_dfly.ipc_smt and kern.usched_dfly.ipc_same,
with the default set to -1 (auto).
ipc_smt Tries to schedule IPC pairings onto sibling
hyperthreads to avoid cache mastership changes
when the load is greater than (ncpus / 2).
ipc_same Tries to schedule IPC pairings onto the same
logical cpu to avoid both cache mastership changes
AND IPIs when the load is greater than (ncpus).
* Keep in mind that the scheduler cannot perfectly predict program
behavior. In particular, these IPC pairings can work better or worse
depending on the mix between local cpu use within each process,
verses the amount of data being transfered between them. By default
we try to localize IPC pairings to nearby cores but we do not try
to schedule them to sibling hyperthreads unless the load is high
enough for it to make sense.
* The main IPC weighting is weight2, whereas the fairness metric is
weight4. Generally speaking, weight4 should be somewhat smaller
than weight2 but still high enough to ensure that available CPUs
in the system are reasonably well utilized. Also note that the
fairness metric (weight4) is based on priority-weighted load.