do not participate in the global symbols namespace, but rtld locks are
still replaced and functions are interposed. In particular,
__pthread_map_stacks_exec is resolved to the libc version. If a
library is loaded later, which requires adjustment of the stack
protection mode, rtld calls into libc __pthread_map_stacks_exec due to
the symbols scope. The libc version might recurse into binder and
recursively acquire rtld bind lock, causing the hang.
Make libc __pthread_map_stacks_exec() interposed, which synchronizes
rtld locks and version of the stack exec hook when libthr loaded,
regardless of the symbol scope control or symbol resolution order.
The __pthread_map_stacks_exec() symbol is removed from the private
version in libthr since libc symbol now operates correctly in presence
of libthr.
Reported and tested by: markj
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
wraps sendmsg(2) and recvmsg(2) into batch send and receive operation.
The goal of this implementation is only to provide API compatibility
with Linux.
The cancellation behaviour of the functions is not quite right, but
due to relative rare use of cancellation it is considered acceptable
comparing with the complexity of the correct implementation. If
functions are reimplemented as syscalls, the fix would come almost
trivial. The direct use of the syscall trampolines instead of libc
wrappers for sendmsg(2) and recvmsg(2) is to avoid data loss on
cancellation.
Submitted by: Boris Astardzhiev <boris.astardzhiev@gmail.com>
Discussed with: jilles (cancellation behaviour)
MFC after: 1 month
intended behaviour in its man page. Simplify tty_drain() to match.
Don't call ttydevsw methods in tty_flush() if the device is gone
since we now sometimes call it then.
The flushing was supposed to be implemented by passing the FNONBLOCK
flag to VOP_CLOSE() for revoke(). The tty driver is one of the few
that can block in close and was one of the fewer that knew about this.
This almost worked in FreeBSD-1 and similarly in Net/2. These
versions only almost worked because there was and is considerable
confusion between IO_NDELAY and FNONBLOCK (aka O_NONBLOCK). IO_NDELAY
is only valid for VOP_READ() and VOP_WRITE(). For other VOPs it has
the same value as O_SHLOCK. But since vfs_subr.c and tty.c
consistently used the wrong flag and the O_SHLOCK flag is rarely set,
this mostly worked. It also gave the feature than applications could
get the non-blocking close by abusing O_SHLOCK.
This was first broken then fixed in 1995. I changed only the tty
driver to use FNONBLOCK, as a hack to get non-blocking via the normal
flag FNONBLOCK for last closes. I didn't know about revoke()'s use
of IO_NDELAY or change it to be consistent, so revoke() was broken.
Then I changed revoke() to match.
This was next broken in 1997 then fixed in 1998. Importing Lite2 made
the flags inconsistent again by undoing the fix only in vfs_subr.c.
This was next broken in 2008 by replacing everything in tty.c and not
checking any flags in last close. Other bugs in draining limited the
resulting unbounded waits to drain in some cases.
It is now possible to fix this better using the new FREVOKE flag.
Just restore flushing for revoke() for now. Don't restore or undo any
hacks for ordinary last closes yet. But remove dead code in the
1-second relative timeout (r272789). This did extra work to extend
the buggy draining for revoke() for as long as possible. The 1-second
timeout made this not very long by usually flushing after 1 second.
Submitted by: bde
MFC after: 2 weeks
* Fix __FreeBSD_version check.
* Update history section in man page.
An MFC of this commit to stable/10 will allow using the new system calls
instead of the fallback.
MFC after: 3 days
up to now.
The new sendfile is the code that Netflix uses to send their multiple tens
of gigabits of data per second. The new implementation features asynchronous
I/O, when I/O operations are launched, but not awaited to be complete. An
explanation of why such behavior is beneficial compared to old one is
going to be too long for a commit message, so we will skip it here.
Additional features of new syscall are extra flags, which provide an
application more control over data sent. The SF_NOCACHE flag tells
kernel that data shouldn't be cached after it was sent. The SF_READAHEAD()
macro allows to specify readahead size in pages.
The new syscalls is a drop in replacement. No modifications are required
to applications. One can take nginx binary for stable/10 and run it
successfully on head. Although SF_NODISKIO lost its original sense, as now
sendfile doesn't block, and now means something completely different (tm),
using the new sendfile the old way is absolutely safe.
Celebrates: Netflix global launch!
Sponsored by: Nginx, Inc.
Sponsored by: Netflix
Relnotes: yes
(CLOCK_REALTIME case) system calls is non negative.
This commit hides a kernel panic in atrtc_settime() as the clock_ts_to_ct()
does not properly convert negative tv_sec.
ps. in my opinion clock_ts_to_ct() should be rewritten to properly handle
negative tv_sec values.
Differential Revision: https://reviews.freebsd.org/D4714
Reviewed by: kib
MFC after: 1 week
Depending on system configuration and parameters, clock_gettime() and
gettimeofday() may not be system calls. If so, passing an invalid pointer
will cause a signal and not an [EFAULT] error.
From a standards perspective, this is OK since passing an invalid pointer is
undefined behaviour.
MFC after: 1 week
system call information such as system call arguments. Initially this
will consist of pulling duplicated code out of truss and kdump though it
may prove useful for other utilities in the future.
This commit moves the shared utrace(2) record parser out of kdump into
the library and updates kdump and truss to use it. One difference from
the previous version is that the library version treats unknown events
that start with the "RTLD" signature as unknown events. This simplifies
the interface and allows the consumer to decide how to handle all
non-recognized events. Instead, this function only generates a string
description for known malloc() and RTLD records.
Reviewed by: bdrewery
Differential Revision: https://reviews.freebsd.org/D4537
This removes the need for manually changing this flag for Google Chrome
users. It also improves compatibility with Linux applications running under
Linuxulator compatibility layer, and possibly also helps in porting software
from Linux.
Generally speaking, the flag allows applications to create the shared memory
segment, attach it, remove it, and then continue to use it and to reattach it
later. This means that the kernel will automatically "clean up" after the
application exits.
It could be argued that it's against POSIX. However, SUSv3 says this
about IPC_RMID: "Remove the shared memory identifier specified by shmid from
the system and destroy the shared memory segment and shmid_ds data structure
associated with it." From my reading, we break it in any case by deferring
removal of the segment until it's detached; we won't break it any more
by also deferring removal of the identifier.
This is the behaviour exhibited by Linux since... probably always, and
also by OpenBSD since the following commit:
revision 1.54
date: 2011/10/27 07:56:28; author: robert; state: Exp; lines: +3 -8;
Allow segments to be used even after they were marked for deletion with
the IPC_RMID flag.
This is permitted as an extension beyond the standards and this is similar
to what other operating systems like linux do.
MFC after: 1 month
Relnotes: yes
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D3603
This uses the kdump(1) utrace support code directly until a common library
is created.
This allows malloc(3) tracing with MALLOC_CONF=utrace:true and rtld tracing
with LD_UTRACE=1. Unknown utrace(2) data is just printed as hex.
PR: 43819 [inspired by]
Reviewed by: jhb
MFC after: 2 weeks
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D3819
Note that the mountlist manipulations are somewhat fragile, and not very
pretty. The reason for this is to avoid changing vfs_mountroot(), which
is (obviously) rather mission-critical, but not very well documented,
and thus hard to test properly. It might be possible to rework it to use
its own simple root mount mechanism instead of vfs_mountroot().
Reviewed by: kib@
MFC after: 1 month
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D2698
are aliases for the syscall stubs and are plt-interposed, to the
libc-private aliases of internally interposed sigprocmask() etc.
Since e.g. _sigaction is not interposed by libthr, calling signal()
removes thr_sighandler() from the handler slot etc. The result was
breaking signal semantic and rtld locking.
The added __libc_sigprocmask and other symbols are hidden, they are
not exported and cannot be called through PLT. The setjmp/longjmp
functions for x86 were changed to use direct calls, and since
PIC_PROLOGUE only needed for functional PLT indirection on i386, it is
removed as well.
The PowerPC bug of calling the syscall directly in the setjmp/longjmp
implementation is kept as is.
Reported by: Pete French <petefrench@ingresso.co.uk>
Tested by: Michiel Boland <boland37@xs4all.nl>
Reviewed by: jilles (previous version)
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
having some children, the children' reaper is not reset to the parent.
This allows for the situation where reaper has children but not
descendands and the too strict asserts in the reap_status() fire.
Remove the wrong asserts, add some clarification for the situation to
the procctl(2) REAP_STATUS.
Reported and tested by: feld
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
The comment in the libc/sys symbol map referenced the generated symbols
for the syscall trampolines. Such comment was out of place in the secure
symbol map so remove the stale comment and attempt to clarify the old one
to avoid risks of confusion.
Pointed out by: kib
As part of the code refactoring to support FORTIFY_SOURCE we want
a new subdirectory "secure" to keep the files related to security.
Move the stack protector functions to this new directory.
No functional change.
Differential Review: https://reviews.freebsd.org/D3333
It looks like EVFILT_READ and EVFILT_WRITE trigger under the same
conditions as poll()'s POLLRDNORM and POLLWRNORM as described by POSIX.
The only difference is that POLLRDNORM has to be triggered on regular
files unconditionally, whereas EVFILT_READ only triggers when not EOF.
Introduce a new flag, NOTE_FILE_POLL, that can be used to make
EVFILT_READ and EVFILT_WRITE behave identically to poll(). This flag
will be used by cloudlibc's poll() function.
Reviewed by: jmg
Differential Revision: https://reviews.freebsd.org/D3303
of the timehands, from the kern_tc.c implementation to vdso. Add
comments giving hints where to look for the algorithm explanation.
To compensate the removal of rmb() in userspace binuptime(), add
explicit lfence instruction before rdtsc. On i386, add usual
complications to detect SSE2 presence; assume that old CPUs which do
not implement SSE2 also execute rdtsc almost in order.
Reviewed by: alc, bde (previous version)
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Summary:
Back in 2005, maxim@ attempted to fix shutdown() to return ENOTCONN in case the socket was not connected (r150152). This had to be rolled back (r150155), as it broke some of the existing programs that depend on this behavior. I reapplied this change on my system and indeed, syslogd failed to start up. I fixed this back in February (279016) and MFC'ed it to the supported stable branches. Apart from that, things seem to work out all right.
Since at least Linux and Mac OS X do the right thing, I'd like to go ahead and give this another try. To keep old copies of syslogd working, only start returning ENOTCONN for recent binaries.
I took a look at the XNU sources and they seem to test against both SS_ISCONNECTED, SS_ISCONNECTING and SS_ISDISCONNECTING, instead of just SS_ISCONNECTED. That seams reasonable, so let's do the same.
Test Plan:
This issue was uncovered while writing tests for shutdown() in CloudABI:
https://github.com/NuxiNL/cloudlibc/blob/master/src/libc/sys/socket/shutdown_test.c#L26
Reviewers: glebius, rwatson, #manpages, gnn, #network
Reviewed By: gnn, #network
Subscribers: bms, mjg, imp
Differential Revision: https://reviews.freebsd.org/D3039
SIGCHLD signal, should keep full 32 bits of the status passed to the
_exit(2).
Split the combined p_xstat of the struct proc into the separate exit
status p_xexit for normal process exit, and signalled termination
information p_xsig. Kernel-visible macro KW_EXITCODE() reconstructs
old p_xstat from p_xexit and p_xsig. p_xexit contains complete status
and copied out into si_status.
Requested by: Joerg Schilling
Reviewed by: jilles (previous version), pho
Tested by: pho
Sponsored by: The FreeBSD Foundation
This is based on work done by jeff@ and jhb@, as well as the numa.diff
patch that has been circulating when someone asks for first-touch NUMA
on -10 or -11.
* Introduce a simple set of VM policy and iterator types.
* tie the policy types into the vm_phys path for now, mirroring how
the initial first-touch allocation work was enabled.
* add syscalls to control changing thread and process defaults.
* add a global NUMA VM domain policy.
* implement a simple cascade policy order - if a thread policy exists, use it;
if a process policy exists, use it; use the default policy.
* processes inherit policies from their parent processes, threads inherit
policies from their parent threads.
* add a simple tool (numactl) to query and modify default thread/process
policities.
* add documentation for the new syscalls, for numa and for numactl.
* re-enable first touch NUMA again by default, as now policies can be
set in a variety of methods.
This is only relevant for very specific workloads.
This doesn't pretend to be a final NUMA solution.
The previous defaults in -HEAD (with MAXMEMDOM set) can be achieved by
'sysctl vm.default_policy=rr'.
This is only relevant if MAXMEMDOM is set to something other than 1.
Ie, if you're using GENERIC or a modified kernel with non-NUMA, then
this is a glorified no-op for you.
Thank you to Norse Corp for giving me access to rather large
(for FreeBSD!) NUMA machines in order to develop and verify this.
Thank you to Dell for providing me with dual socket sandybridge
and westmere v3 hardware to do NUMA development with.
Thank you to Scott Long at Netflix for providing me with access
to the two-socket, four-domain haswell v3 hardware.
Thank you to Peter Holm for running the stress testing suite
against the NUMA branch during various stages of development!
Tested:
* MIPS (regression testing; non-NUMA)
* i386 (regression testing; non-NUMA GENERIC)
* amd64 (regression testing; non-NUMA GENERIC)
* westmere, 2 socket (thankyou norse!)
* sandy bridge, 2 socket (thankyou dell!)
* ivy bridge, 2 socket (thankyou norse!)
* westmere-EX, 4 socket / 1TB RAM (thankyou norse!)
* haswell, 2 socket (thankyou norse!)
* haswell v3, 2 socket (thankyou dell)
* haswell v3, 2x18 core (thankyou scott long / netflix!)
* Peter Holm ran a stress test suite on this work and found one
issue, but has not been able to verify it (it doesn't look NUMA
related, and he only saw it once over many testing runs.)
* I've tested bhyve instances running in fixed NUMA domains and cpusets;
all seems to work correctly.
Verified:
* intel-pcm - pcm-numa.x and pcm-memory.x, whilst selecting different
NUMA policies for processes under test.
Review:
This was reviewed through phabricator (https://reviews.freebsd.org/D2559)
as well as privately and via emails to freebsd-arch@. The git history
with specific attributes is available at https://github.com/erikarn/freebsd/
in the NUMA branch (https://github.com/erikarn/freebsd/compare/local/adrian_numa_policy).
This has been reviewed by a number of people (stas, rpaulo, kib, ngie,
wblock) but not achieved a clear consensus. My hope is that with further
exposure and testing more functionality can be implemented and evaluated.
Notes:
* The VM doesn't handle unbalanced domains very well, and if you have an overly
unbalanced memory setup whilst under high memory pressure, VM page allocation
may fail leading to a kernel panic. This was a problem in the past, but it's
much more easily triggered now with these tools.
* This work only controls the path through vm_phys; it doesn't yet strongly/predictably
affect contigmalloc, KVA placement, UMA, etc. So, driver placement of memory
isn't really guaranteed in any way. That's next on my plate.
Sponsored by: Norse Corp, Inc.; Dell
Use a constant array for the MIB. Newer LLVM decided that mib[] warranted
stack protections, with the obvious crash after the setup was done.
As a positive side effect, code size shrinks a bit.
I'm not sure why this hasn't bitten us yes, but it is certainly possible and
there are no real drawbacks to this change anyway.
Submitted by: pfg
Obtained from: NetBSD
MFC after: 1 week
- Note that ftruncate(2) can operate on shared memory objects and cross
reference shm_open(2).
- Note that ftruncate(2) does not change the file position pointer (aka
seek pointer) of the file descriptor.
- ftruncate(2) will fail with EINVAL for all sorts of other fd types than
just sockets, so instead note that it fails for all but regular files and
shared memory objects.
- Note that ftruncate(2) also appeared in 4.2BSD along with truncate(2).
(Or at least the manpage for both appeared in 4.2, I did not check the
kernel code itself to see if either predated 4.2.)
PR: 199472 (2)
Submitted by: andrew@ugh.net.au (2)
MFC after: 1 week
domain, not a file descriptor. Use 'domain' instead of the original 'd'
for this argument to match socket(2).
PR: 199491
Reported by: sp55aa@qq.com
MFC after: 1 week
