imp [Mon, 12 Feb 2018 06:51:20 +0000 (06:51 +0000)]
Turn devmatch on by default.
Turn devmatch on by default. However, use 'start' instead of
'onestart' in the devmatch.conf file so the setting of
'devmatch_enable' is honored. Give an example of what to put in
devd.conf if you want to disable just the run-time part of devmatch.
imp [Mon, 12 Feb 2018 04:45:26 +0000 (04:45 +0000)]
Switch to using devmatch to autoload drivers. Remove usb.conf
as obsolete because devmatch gets its information from the same
place as the genration scripts.
imp [Sun, 11 Feb 2018 17:45:38 +0000 (17:45 +0000)]
Consistent macro indentation is the hobgoblin of small minds
Line up the macro definitions and names here like the machine/stdarg.h
files that this replaced. This is easier to read and also makes it
easier to match up with other includes. Also two space indent va_end
to match the rest of the surrounding if block.
dteske [Sun, 11 Feb 2018 03:02:29 +0000 (03:02 +0000)]
Fix typo in dtrace_tcp(4)
Using args[2]->tcps_state as-documented results in error:
operator -> cannot be applied to pointer to type "void"
This error is accurate as the synopsis for tcp:::state-change is:
tcp:::state-change(void *, csinfo_t *, void *, tcpsinfo_t *, void *,
tcplsinfo_t *);
args[2] refers to the third argument which is always NULL (as-
documented). The to-state for the TCP connection state transition is
actually in the fourth argument, args[3]->tcps_state.
jhibbits [Sat, 10 Feb 2018 17:17:15 +0000 (17:17 +0000)]
Fix uninitialized warning, and work around a bug in gcc over clobbering
Summary:
r329077 caused gcc to emit uninitialized use warnings. Attempting to
fix those warnings yielded the following warnings:
usr.bin/tftp/main.c: In function 'main':
usr.bin/tftp/main.c:181: warning: variable 'el' might be clobbered by
'longjmp' or 'vfork'
usr.bin/tftp/main.c:182: warning: variable 'hist' might be clobbered by
'longjmp' or 'vfork'
This is a known bug in gcc, found at
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=24239
Work around that by simply marking hist and el as static.
eugen [Sat, 10 Feb 2018 17:09:51 +0000 (17:09 +0000)]
ppp(8): fix code producing debugging logs
Fix several cases when long buffer is copied to shorter one
using snprintf that results in contents truncation and
clobbering unsaved errno value and creation of misleading logs.
mav [Sat, 10 Feb 2018 00:55:46 +0000 (00:55 +0000)]
Teach mps(4) and mpr(4) drivers to autotune chain frames.
This is a first part of the change. It makes the drivers to calculate
the required number of chain frames to satisfy worst case scenarios, but
it does not change existing overly strict limits on them. The next step
will be to rewrite the allocator to not require megabytes of physically
contiguous address space, that may be problematic if done after boot,
after doing which the limits can be removed. Until that this code can
just correct user set limits, if they are set too high.
Sponsored by: iXsystems, Inc.
Differential Revision: https://reviews.freebsd.org/D14261
emaste [Sat, 10 Feb 2018 00:22:35 +0000 (00:22 +0000)]
Promote llvm-cov to a standalone option
Introduce WITH_/WITHOUT_LLVM_COV to match GCC's WITH_/WITHOUT_GCOV.
It is intended to provide a superset of the interface and functionality
of gcov.
It is enabled by default when building Clang, similarly to gcov and GCC.
This change moves one file in libllvm to be compiled unconditionally.
Previously it was included only when WITH_CLANG_EXTRAS was set, but the
complexity of a new special case for (CLANG_EXTRAS | LLVM_COV) is not
worth avoiding a tiny increase in build time.
Reviewed by: dim, imp
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D142645
nwhitehorn [Fri, 9 Feb 2018 20:09:32 +0000 (20:09 +0000)]
Fix PowerMac G5 thermal management, plus likely other bugs, introduced in
r328113 and affecting SMP systems.
The way the time is set on PowerMacs is racy and relies on all the
CPUs in the system setting a register simultaneously in a rendezvous. A
few-cycle delay can result in out-of-sync times, which can break the
scheduler and result in calls like mtx_sleep() and pause() never timing out
if the thread is migrated while sleeping. r328113 added a call to a no-op
function between the beginning of the rendezvous and setting the time that
was only called on APs and added enough cycles to cause a problematic offset.
For some reason, the fan-management code was the first place this appeared.
mckusick [Fri, 9 Feb 2018 19:50:47 +0000 (19:50 +0000)]
Merge biodone_finish() back into biodone(). The primary purpose is
to make the order of operations clearer to avoid the race condition
that was fixed in r328914. In particular, this commit corrects a
similar race that existed in the soft updates callback.
Doing some sleuthing through the SVN repository, it appears that
bufdone_finish() was added to support XFS:
Changes imported from XFS for FreeBSD project:
- add fields to struct buf (needed by XFS)
- 3 private fields: b_fsprivate1, b_fsprivate2, b_fsprivate3
- b_pin_count, count of pinned buffer
- add new B_MANAGED flag
- add breada() function to initiate asynchronous I/O on read-ahead blocks.
- add bufdone_finish(), bpin(), bunpin_wait() functions
Patches provided by: kan
Reviewed by: phk
Silence on: arch@
cem [Fri, 9 Feb 2018 19:46:51 +0000 (19:46 +0000)]
tftp(1): Fix libedit state corruption involving signals
This bug was first reported 14 years ago. The problem was understood 8.5
years ago. A patch that is functionally identical to this one was proposed
almost 8 years ago and languished in the PR system / Bugzilla.
PR: 63197
Submitted by: lxv AT omut.org, fernando.apesteguia AT gmail.com
Reported by: freebsd AT nbritton.org
Each contiguous range of blocks is printed on a line.
The distance metric is the size of the gap from the end of the
previous set of blocks to the beginning of the next set of blocks.
Short distances are desirable.
jtl [Fri, 9 Feb 2018 17:46:33 +0000 (17:46 +0000)]
On bootup, the amd64 pmap initialization code creates page-table
mappings for the pages used for the kernel and some initial allocations
used for the page table. It maps the kernel and the blocks used for
these initial allocations using 2MB pages.
However, if the kernel does not end on a 2MB boundary, it still maps the
last portion using a 2MB page, but reports that the unused 4K blocks
within this 2MB allocation are free physical blocks. This means that
these same physical blocks could also be mapped elsewhere - for example,
into a user process. Given the proximity to the kernel text and data
area, it seems wise to avoid allowing someone to write data to physical
blocks also mapped into these virtual addresses.
(Note that this isn't a security vulnerability: the direct map makes
most/all memory on the system mapped into kernel space. And, nothing
in the kernel should be trying to access these pages, as the virtual
addresses are unused. It simply seems wise to avoid reusing these
physical blocks while they are mapped to virtual addresses so close
to the kernel text and data area.)
Consequently, let's reserve the physical blocks covered by the
page-table mappings for these initial allocations.
royger [Fri, 9 Feb 2018 10:20:16 +0000 (10:20 +0000)]
loader: fix endianness conversion
r328536 broke symbol loading on amd64 at least (and probably other
arches). r328826 contained the problem to ppc only by adding
pre-processors guards.
Fix this properly by moving the endianness conversion to separate
helper functions, and make the conversion more robust by using sizeof
instead of having to manually code the size of each field.
Finally list the fields in each structure in a macro in order to avoid
code repetition.
glebius [Fri, 9 Feb 2018 04:45:39 +0000 (04:45 +0000)]
Fix boot_pages exhaustion on machines with many domains and cores, where
size of UMA zone allocation is greater than page size. In this case zone
of zones can not use UMA_MD_SMALL_ALLOC, and we need to postpone switch
off of this zone from startup_alloc() until full launch of VM.
o Always supply number of VM zones to uma_startup_count(). On machines
with UMA_MD_SMALL_ALLOC ignore it completely, unless zsize goes over
a page. In the latter case account VM zones for number of allocations
from the zone of zones.
o Rewrite startup_alloc() so that it will immediately switch off from
itself any zone that is already capable of running real alloc.
In worst case scenario we may leak a single page here. See comment
in uma_startup_count().
o Hardcode call to uma_startup2() into vm_mem_init(). Otherwise some
extra SYSINITs, e.g. vm_page_init() may sneak in before.
o While here, remove uma_boot_pages_mtx. With recent changes to boot
pages calculation, we are guaranteed to use all of the boot_pages
in the early single threaded stage.
imp [Fri, 9 Feb 2018 00:36:55 +0000 (00:36 +0000)]
Set script.lang in the environment to either 'forth' or 'simple' to
reflect what scripting language was compiled into the loader. I
anticipate that being able to find this out quickly from the OK prompt
will be useful in troubleshooting in the future.
vangyzen [Fri, 9 Feb 2018 00:13:05 +0000 (00:13 +0000)]
Fix ICMPv6 redirects
icmp6_redirect_input() validates that a redirect packet came from the
current gateway for the respective destination. To do this, it compares
the source address, which has an embedded scope zone id, to the next-hop
address, which does not. If the address is link-local, which should be
the case, the comparison fails and the redirect is ignored.
Insert the scope zone id into the next-hop address so the comparison
is accurate.
Unsurprisingly, this fixes 35 UNH IPv6 conformance test cases.
mckusick [Thu, 8 Feb 2018 23:06:58 +0000 (23:06 +0000)]
The goal of this change is to prevent accidental foot shooting by
folks running filesystems created on check-hash enabled kernels
(which I will call "new") on a non-check-hash enabled kernels (which
I will call "old). The idea here is to detect when a filesystem is
run on an old kernel and flag the filesystem so that when it gets
moved back to a new kernel, it will not start getting a slew of
check-hash errors.
Back when the UFS version 2 filesystem was created, it added a file
flag FS_INDEXDIRS that was to be set on any filesystem that kept
some sort of on-disk indexing for directories. The idea was precisely
to solve the issue we have today. Specifically that a newer kernel
that supported indexing would be able to tell that the filesystem
had been run on an older non-indexing kernel and that the indexes
should not be used until they had been rebuilt. Since we have never
implemented on-disk directory indicies, the FS_INDEXDIRS flag is
cleared every time any UFS version 2 filesystem ever created is
mounted for writing.
This commit repurposes the FS_INDEXDIRS flag as the FS_METACKHASH
flag. Thus, the FS_METACKHASH is definitively known to have always
been cleared. The FS_INDEXDIRS flag has been moved to a new block
of flags that will always be cleared starting with this commit
(until they get used to implement some future feature which needs
to detect that the filesystem was mounted on a kernel that predates
the new feature).
If a filesystem with check-hashes enabled is mounted on an old
kernel the FS_METACKHASH flag is cleared. When that filesystem is
mounted on a new kernel it will see that the FS_METACKHASH has been
cleared and clears all of the fs_metackhash flags. To get them
re-enabled the user must run fsck (in interactive mode without the
-y flag) which will ask for each supported check hash whether it
should be rebuilt and enabled. When fsck is run in its default preen
mode, it will just ignore the check hashes so they will remain
disabled.
The kernel has always disabled any check hash functions that it
does not support, so as more types of check hashes are added, we
will get a non-surprising result. Specifically if filesystems get
moved to kernels supporting fewer of the check hashes, those that
are not supported will be disabled. If the filesystem is moved back
to a kernel with more of the check-hashes available and fsck is run
interactively to rebuild them, then their checking will resume.
Otherwise just the smaller subset will be checked.
A side effect of this commit is that filesystems running with
cylinder-group check hashes will stop having them checked until
fsck is run to re-enable them (since none of them currently have
the FS_METACKHASH flag set). So, if you want check hashes enabled
on your filesystems after booting a kernel with these changes, you
need to run fsck to enable them. Any newly created filesystems will
have check hashes enabled. If in doubt as to whether you have check
hashes emabled, run dumpfs and look at the list of enabled flags
at the end of the superblock details.
imp [Thu, 8 Feb 2018 22:59:51 +0000 (22:59 +0000)]
Fix build of userboot.so
Since it's not possible to unset a variable easily, create a new
variable 'PIC' to signal that we are creating a shared object that we
want to install. defs.mk refains from defining NO_PIC and ITNERALLIB
when PIC is defined. This unbreaks userboot.so building.
dim [Thu, 8 Feb 2018 21:11:48 +0000 (21:11 +0000)]
Pull in r324594 from upstream clang trunk (by Alexander Ivchenko):
Fix for #31362 - ms_abi is implemented incorrectly for values >=16
bytes.
Summary:
This patch is a fix for following issue:
https://bugs.llvm.org/show_bug.cgi?id=31362 The problem was caused by
front end lowering C calling conventions without taking into account
calling conventions enforced by attribute. In this case win64cc was
no correctly lowered on targets other than Windows.
This fixes clang 6.0.0 assertions when building the emulators/wine and
emulators/wine-devel ports, and should also make it use the correct
Windows calling conventions. Bump __FreeBSD_version to make the fix
easy to detect.
imp [Thu, 8 Feb 2018 17:07:27 +0000 (17:07 +0000)]
Move to tabs for indentation and to 8-space notches, per style(9).
4 space indentation with a mix of tabs and spaces is a hassle. Update
to project-standard hard-tabs with 8-space indentation in these files.
This matches the new code coming in better as well.
jhibbits [Thu, 8 Feb 2018 05:18:30 +0000 (05:18 +0000)]
Temporarily widen count for interrupt rate calculations on 32-bit archs
If the interrupt count is very high (greater than ~42M), notably on one-shot
execution on long running systems, the intermediate multiplication step in the
rate calculation will overflow the width of a 32-bit architecture long (32
bits), causing the rest of the calculation to calculate with a truncated value,
and report very low rates (sometimes 0).
avg [Wed, 7 Feb 2018 21:51:59 +0000 (21:51 +0000)]
exec_map_first_page: fix an inverse condition introduced in r254138
While the bug itself was serious, as we could either pass a non-busied
page to vm_pager_get_pages() or leak a busy page, it could only be
triggered under a very rare condition where the page is already inserted
into the object, but it is not valid yet.
Rework ipfw dynamic states implementation to be lockless on fast path.
o added struct ipfw_dyn_info that keeps all needed for ipfw_chk and
for dynamic states implementation information;
o added DYN_LOOKUP_NEEDED() macro that can be used to determine the
need of new lookup of dynamic states;
o ipfw_dyn_rule now becomes obsolete. Currently it used to pass
information from kernel to userland only.
o IPv4 and IPv6 states now described by different structures
dyn_ipv4_state and dyn_ipv6_state;
o IPv6 scope zones support is added;
o ipfw(4) now depends from Concurrency Kit;
o states are linked with "entry" field using CK_SLIST. This allows
lockless lookup and protected by mutex modifications.
o the "expired" SLIST field is used for states expiring.
o struct dyn_data is used to keep generic information for both IPv4
and IPv6;
o struct dyn_parent is used to keep O_LIMIT_PARENT information;
o IPv4 and IPv6 states are stored in different hash tables;
o O_LIMIT_PARENT states now are kept separately from O_LIMIT and
O_KEEP_STATE states;
o per-cpu dyn_hp pointers are used to implement hazard pointers and they
prevent freeing states that are locklessly used by lookup threads;
o mutexes to protect modification of lists in hash tables now kept in
separate arrays. 65535 limit to maximum number of hash buckets now
removed.
o Separate lookup and install functions added for IPv4 and IPv6 states
and for parent states.
o By default now is used Jenkinks hash function.
imp [Wed, 7 Feb 2018 18:33:53 +0000 (18:33 +0000)]
Cull Atmel board configs no longer relevant.
Remove most of the Atmel at91 boards. Most of them are no longer
relevant or used by people. Kept ATMEL since it should work on all the
boards that still work (I've not confirmed this, since I don't have
all these boards). Also kept SAM9G20EK, since I have several boards
that it is used on. If I've deleted a kernel in error, please let me
know.
glebius [Wed, 7 Feb 2018 18:32:51 +0000 (18:32 +0000)]
Fix three miscalculations in amount of boot pages:
o Most of startup zones have struct uma_slab embedded into the slab,
so provide macro UMA_SLAB_SPACE and use it instead of UMA_SLAB_SIZE,
when calculating how many pages would certain kind of allocations
require. Some zones are offpage, so we might have a positive inaccuracy.
o The keg for the zone of zones is allocated "dynamically", so we
need +1 when calculating amount of pages for kegs. [1]
o The zones of zones and zones of kegs have arbitrary alignment of 32,
and this also needs to be accounted for. [2]
While here, spread more comments and improve diagnostic messages.
arichardson [Wed, 7 Feb 2018 16:58:01 +0000 (16:58 +0000)]
Fix compilation of mips_postboot_fixup() with a C11 compiler
The _Alignas specifier must come before the declaration and not after. It
works if _Alignas() expands to __attribute__(aligned(x)) which was the only
case I tested before.
markj [Wed, 7 Feb 2018 16:57:10 +0000 (16:57 +0000)]
Dequeue wired pages lazily.
Previously, wiring a page would cause it to be removed from its page
queue. In the common case, unwiring causes it to be enqueued at the tail
of that page queue. This change modifies vm_page_wire() to not dequeue
the page, thus avoiding the highly contended page queue locks. Instead,
vm_page_unwire() takes care of requeuing the page as a single operation,
and the page daemon dequeues wired pages as they are encountered during
a queue scan to avoid needlessly revisiting them later. For pages in
PQ_ACTIVE we do even better, since a requeue is unnecessary.
The change improves scalability for some common workloads. For instance,
threads wiring pages into the buffer cache no longer need to modify
global page queues, and unwiring is usually done by the bufspace thread,
so concurrency is not as much of an issue. As another example, many
sysctl handlers wire the output buffer to avoid faults on copyout, and
since the buffer is likely to be in PQ_ACTIVE, we now entirely avoid
modifying the page queue in this case.
The change also adds a block comment describing some properties of
struct vm_page's reference counters, and the busy lock.
imp [Wed, 7 Feb 2018 16:28:26 +0000 (16:28 +0000)]
Add a note about why we have the conditional before including
bsd.compiler.mk. It's so fmake from older 9.x systems still
works (still a supported build config, and having the note here
will let us know when we can cull it more easily).
Also pull in a related change from include to sinclude from
arichardson@'s cross building work, as well as it's companion in
Makefile.inc1 with a note about why we do the odd thing there.
adrian [Wed, 7 Feb 2018 09:37:22 +0000 (09:37 +0000)]
[ath] Use the BSSID address logic for STA VAPs too.
For DWDS VAPs on ath(4) we need to ensure that the STA vap and hostap VAP
have different MAC addresses. If the STA code path doesn't utilise the
address assign / reclaim path then it doesn't update the bitmap with which
address was allocated.
This should fix a bunch of corner issues I've been seeing with DWDS STA + AP
VAPs that I was working around with manual MAC address assignment.
kevans [Wed, 7 Feb 2018 01:54:13 +0000 (01:54 +0000)]
if_awg: Skip emac reset if configured for internal PHY
On the OrangePi One at least, emac reset when an ethernet cable is not
plugged in seems to break ethernet. Soft reset will fail, even with
increasing the delay and retries to wait for up to 20 seconds. This can be
reproduced across at least two different OrangePi One's by simply leaving
ethernet cable unplugged when awg attaches. Whether it's plugged in or not
through u-boot process makes no difference.
Skipping the reset in this configuration doesn't seem to cause any problems,
tried across many many reboots with and without ethernet cable plugged in.
Tested on: OrangePi One
Tested on: Other boards (manu)
Reviewed by: manu
Differential Revision: https://reviews.freebsd.org/D13974
imp [Tue, 6 Feb 2018 23:12:16 +0000 (23:12 +0000)]
Avoid find -s, use find | sort instead.
find -s was introduced to make the metalog more
deterministic. However, find -s is not portable. find | sort is
portable and accomplishes the same goals, even if it isn't
pedantically the same. TZS is the same before / after the change so
any fussy differences between the two are moot and there won't be
METALOG churn across this change.
jeff [Tue, 6 Feb 2018 22:10:07 +0000 (22:10 +0000)]
Use per-domain locks for vm page queue free. Move paging control from
global to per-domain state. Protect reservations with the free lock
from the domain that they belong to. Refactor to make vm domains more
of a first class object.
glebius [Tue, 6 Feb 2018 22:06:59 +0000 (22:06 +0000)]
Fix boot_pages calculation for machines that don't have UMA_MD_SMALL_ALLOC.
o Call uma_startup1() after initializing kmem, vmem and domains.
o Include 8 eight VM startup pages into uma_startup_count() calculation.
o Account for vmem_startup() and vm_map_startup() preallocating pages.
o Account for extra two allocations done by kmem_init() and vmem_create().
o Hardcode the place of execution of vm_radix_reserve_kva(). Using SYSINIT
allowed several other SYSINITs to sneak in before it, thus bumping
requirement for amount of boot pages.