sleepq is only locked if the curhtread is the last reader. By the time
the lock gets acquired new ones could have arrived. The previous code
would unlock and loop back. This results spurious relocking of sleepq.
This is a step towards xadd-based unlock routine.
In order to go to sleep threads set waiter flags, but that can spuriously
fail e.g. when a new reader arrives. Instead of unlocking everything and
looping back, re-evaluate the new state while still holding the lock necessary
to go to sleep.
- Add a new KTR_STRUCT_ARRAY ktrace record type which dumps an array of
structures.
The structure name in the record payload is preceded by a size_t
containing the size of the individual structures. Use this to
replace the previous code that dumped the kevent arrays dumped for
kevent(). kdump is now able to decode the kevent structures rather
than dumping their contents via a hexdump.
One change from before is that the 'changes' and 'events' arrays are
not marked with separate 'read' and 'write' annotations in kdump
output. Instead, the first array is the 'changes' array, and the
second array (only present if kevent doesn't fail with an error) is
the 'events' array. For kevent(), empty arrays are denoted by an
entry with an array containing zero entries rather than no record.
- Move kevent decoding tables from truss to libsysdecode.
This adds three new functions to decode members of struct kevent:
sysdecode_kevent_filter, sysdecode_kevent_flags, and
sysdecode_kevent_fflags.
kdump uses these helper functions to pretty-print kevent fields.
- Move structure definitions for freebsd11 and freebsd32 kevent
structures to <sys/event.h> so that they can be shared with userland.
The 32-bit structures are only exposed if _WANT_KEVENT32 is defined.
The freebsd11 structures are only exposed if _WANT_FREEBSD11_KEVENT is
defined. The 32-bit freebsd11 structure requires both.
- Decode freebsd11 kevent structures in truss for the compat11.kevent()
system call.
- Log 32-bit kevent structures via ktrace for 32-bit compat kevent()
system calls.
- While here, constify the 'void *data' argument to ktrstruct().
Reviewed by: kib (earlier version)
MFC after: 1 month
Differential Revision: https://reviews.freebsd.org/D12470
Upon successful completion, the execve() system call invokes
exec_setregs() to initialize the registers of the initial thread of the
newly executed process. What is weird is that when execve() returns, it
still goes through the normal system call return path, clobbering the
registers with the system call's return value (td->td_retval).
Though this doesn't seem to be problematic for x86 most of the times (as
the value of eax/rax doesn't matter upon startup), this can be pretty
frustrating for architectures where function argument and return
registers overlap (e.g., ARM). On these systems, exec_setregs() also
needs to initialize td_retval.
Even worse are architectures where cpu_set_syscall_retval() sets
registers to values not derived from td_retval. On these architectures,
there is no way cpu_set_syscall_retval() can set registers to the way it
wants them to be upon the start of execution.
To get rid of this madness, let sys_execve() return EJUSTRETURN. This
will cause cpu_set_syscall_retval() to leave registers intact. This
makes process execution easier to understand. It also eliminates the
difference between execution of the initial process and successive ones.
The initial call to sys_execve() is not performed through a system call
context.
Reviewed by: kib, jhibbits
Differential Revision: https://reviews.freebsd.org/D13180
An assertion was modified to use the found value, but it was not updated to
handle a race where blocked threads appear after the entrance to the func.
Move the assertion down to the area protected with sleepq lock where the
lock is read anyway. This does not affect coverage of the assertion and
is consistent with what rw locks are doing.
Reported by: Shawn Webb
The pair is of use only in debug or LOCKPROF kernels, but was passed (zeroed)
for many locks even in production kernels.
While here whack the tid argument from wlock hard and xlock hard.
There is no kbi change of any sort - "external" primitives still accept the
pair.
Mainly focus on files that use BSD 3-Clause license.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Special thanks to Wind River for providing access to "The Duke of
Highlander" tool: an older (2014) run over FreeBSD tree was useful as a
starting point.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Special thanks to Wind River for providing access to "The Duke of
Highlander" tool: an older (2014) run over FreeBSD tree was useful as a
starting point.
Initially, only tag files that use BSD 4-Clause "Original" license.
RelNotes: yes
Differential Revision: https://reviews.freebsd.org/D13133
Previously, symlinks in FreeBSD were artificially limited to PATH_MAX-2.
Add a short test case to verify the change.
Submitted by: Gaurav Gangalwar <ggangalwar AT isilon.com>
Reviewed by: kib
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D12589
This shortens the lock hold time while not affecting corretness.
All the woken up threads end up competing can lose the race against
a completely unrelated thread getting the lock anyway.
clock_ts_to_ct has a KASSERT that the converted year fits into four
digits. By default (sysctl debug.allow_insane_settime is 0) the kernel
disallows a time too far in the future, using a value of 9999 366-day
years. However, clock_settime is epoch-relative and the assertion will
fail with a tv_sec corresponding to some 8030 years.
Avoid trying to be too clever, and just use a limit of 8000 365-day
years past the epoch.
Submitted by: Heqing Yan <scottieyan@gmail.com>
Reported by: Syzkaller (https://github.com/google/syzkaller)
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
kern.poweroff_on_panic which, when enabled, instructs a system to
power off on a panic instead of a reboot.
kern.powercyle_on_panic which, when enabled, instructs a system to
power cycle, if possible, on a panic instead of a reboot.
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D13042
The PT_DETACH case above the sendsig: label already looped over all
threads clearing flags in td_dbgflags. Reuse this loop to clear
TDB_SUSPEND and move the logic out of the sendsig: block.
Most of the conditionals in the 'sendsig:' block are now only different
for PT_ATTACH vs other continue requests. Pull the PT_ATTACH-specific
logic up into the PT_ATTACH case and simplify the 'sendsig:' block. This
also permits moving the unlock of proctree_lock above the sendsig: label
since PT_KILL doesn't hold the lock and and the other cases all fall
through to the label.
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D13073
This fixes a panic when attaching to an already-stopped process after
r325028. While here, clean up a few other things in the control flow
of the 'sendsig' section:
- Only check for P_STOPPED_TRACE rather than either of P_STOPPED_SIG
or P_STOPPED_TRACE for most ptrace requests. The signal handling
code in kern_sig.c never sets just P_STOPPED_SIG for a traced
process, so if P_STOPPED_SIG is stopped, P_STOPPED_TRACE should be
set anyway. Remove a related debug printf. Assuming P_STOPPED_TRACE
permits simplifications in the 'sendsig:' block.
- Move the block to clear the pending thread state up into a new
block conditional on P_STOPPED_TRACE and handle delivering pending
signals to the reporting thread and clearing the reporting thread's
state in this block.
- Consolidate case to send a signal to the process in a single case
for PT_ATTACH. The only case that could have been in the else before
was a PT_ATTACH where P_STOPPED_SIG was not set, so both instances
of kern_psignal() collapse down to just PT_ATTACH.
Reported by: pho, mmel
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D12837
This introduces a facility to EVENTHANDLER(9) for explicitly defining a
reference to an event handler list. This is useful since previously all
invokers of events had to do a locked traversal of the global list of
event handler lists in order to find the appropriate event handler list.
By keeping a pointer to the appropriate list an invoker can avoid this
traversal completely. The pointer is initialized with SYSINIT(9) during
the eventhandler stage. Users registering interest in events do not need
to know if the event is backed by such a list, since the list is added
to the global list of lists. As with lists that are not pre-defined it
is safe to register for the events before the list has been created.
This converts the process_* and thread_* events to using the new
facility, as these are events whose locked traversals end up showing up
significantly in ports build workflows (and presumably other workflows
with many short lived threads/procs). It may be advantageous to convert
other events to using the new facility.
The el_flags field is now unused, but leave it be so that this revision
can be MFC'd.
Reviewed by: bdrewery, markj, mjg
Approved by: rstone (mentor)
In collaboration with: ian
MFC after: 4 weeks
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D12814
