This changes struct kinfo_filedesc and kinfo_vmentry such that they are
same on both 32 and 64 bit platforms like i386/amd64 and won't require
sysctl wrapping.
Two new OIDs are assigned. The old ones are available under
COMPAT_FREEBSD7 - but it isn't that simple. The superceded interface
was never actually released on 7.x.
The other main change is to pack the data passed to userland via the
sysctl. kf_structsize and kve_structsize are reduced for the copyout.
If you have a process with 100,000+ sockets open, the unpacked records
require a 132MB+ copyout. With packing, it is "only" ~35MB. (Still
seriously unpleasant, but not quite as devastating). A similar problem
exists for the vmentry structure - have lots and lots of shared libraries
and small mmaps and its copyout gets expensive too.
My immediate problem is valgrind. It traditionally achieves this
functionality by parsing procfs output, in a packed format. Secondly, when
tracing 32 bit binaries on amd64 under valgrind, it uses a cross compiled
32 bit binary which ran directly into the differing data structures in 32
vs 64 bit mode. (valgrind uses this to track file descriptor operations
and this therefore affected every single 32 bit binary)
I've added two utility functions to libutil to unpack the structures into
a fixed record length and to make it a little more convenient to use.
parentheses.
Fixed alignment issue in gr_dup() in its assignment of gr_mem using a
struct to force alignment without performing alignment mathematics. This
was noticed recently with libutil was built with WARNS=6 on platform such
as sparc64.
Added checks to gr_dup(), gr_equal() and gr_make() to prevent segfaults
when examining struct group's with the struct members pointing to NULL's.
With fix of alignment issue, restore WARNS?=6.
Reviewed by: des
MFC after: 1 week
struct sockaddr * that it casts internally to the appropriate type based on
sa_family. However, struct sockaddr has very lax alignment requirements,
which causes the compiler to complain when you cast a struct sockaddr * to,
say, a struct sockaddr_in6 *.
I find it reasonable to assume that the pointer we received is in fact
correctly aligned. Therefore, we can work around the compiler warnings by
casting to void * before casting to the desired type. For readability's
sake, this is done with macros.
The same technique should prove useful in other parts of the tree that
deal with socket addresses.
MFC after: 3 weeks
As discussed on the commits list, there is no need to call revoke()
inside openpty(). On RELENG_6 and RELENG_7 unlockpt() will call
revoke(). On HEAD we create pseudo-terminals on demand, so there is no
need to revoke the slave device node.
This change should never be MFC'd, because the implementation we have in
RELENG_6 and RELENG_7 should work flawlessly with older versions of
libc.
Discussed with: jhb
MFC after: never
- Pass O_NOCTTY to posix_openpt(2). This makes the implementation work
consistently on implementations that make the PTY the controlling TTY
by default.
- Call unlockpt() before opening the slave device. POSIX mentions that
de slave device should only be opened after grantpt() and unlockpt()
have been called.
- Replace some redundant code by a label.
In theory we could remove a lot of code from openpty() on FreeBSD
-CURRENT, because grantpt(), unlockpt() and revoke() are not needed in
our implementation. We'd better keep them there. This makes the code
still work with older FreeBSD releases and even makes it work on other
non-BSD operating systems.
I've compiled openpty() on Linux. You only need to remove the revoke()
call, because revoke() on Linux always returns -1. Apart from that, it
seems to work like it should.
Reviewed by: jhb
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
after similar calls related to struct pwd in libutil/pw_util.c:
- gr_equal()
Perform a deep comparison of two struct grp's. It does a thorough, yet
unoptimized comparison of all the members regardless of order.
- gr_make()
Create a string (see group(5)) from a struct grp.
- gr_dup()
Duplicate a struct grp. Returns a value that is a single contiguous
block of memory.
- gr_scan()
Create a struct grp from a string (as produced by gr_make()).
MFC after: 3 weeks
Significant changes:
- rev. 1.11: Use PRId64 instead of a cast to long long and %lld to print
an int64_t.
- rev. 1.12: Fix a bug that humanize_number() produces "1000" where it
should be "1.0G" or "1.0M". The bug reported by Greg Troxel.
PR: 118461
PR: 102694
Approved by: rwatson (mentor)
Obtained from: NetBSD
MFC after: 1 month
Specifically, remove the BUGS section and note that openpty(3) now always
does the various security-related steps. Also, update the error return
value section. The PR below is for the original bug rather than the doc
updates.
MFC after: 1 week
PR: bin/9770
kick off any other users on the device line before using it since
openpty(3) is documented to do this. Note that grantpt(3) does not
call revoke(2), it only adjusts permissions and ownership.
MFC after: 3 days
success and zero pid from pidfile_read(). Return EAGAIN instead. Sleep
up to three times for 5 ms while waiting for pidfile to be written.
mount(8) does the kill(mountpid, SIGHUP). If mountd pidfile is truncated,
that would result in the SIGHUP delivered to the mount' process group
instead of the mountd.
Found and analyzed by: Peter Holm
Tested by: Peter Holm, kris
Reviewed by: pjd
MFC after: 1 week
Reported by: phk
- While here, check the unit before calculating the actually number.
This way we can return EINVAL for invalid unit instead of ERANGE.
Approved by: re (kensmith)
a number in human-readable form is converted to int64_t, for example:
123b -> 123
10k -> 10240
16G -> 17179869184
First version submitted by: Eric Anderson <anderson@freebsd.org>
Approved by: re (bmah)
to pidfile_write happen, the pidfile will have nul characters prepended
due to the cached file descriptor offset...
Reviewed by: scottl
MFC after: 3 days
from /etc/login.conf, or an unterminated string buffer could result.
Probably, login_times.c should reject excessively long time strings as
unparseable, rather than truncating, which might render an invalid
string valid.
Found with: Coverity Prevent (tm)
Reviewed by: csjp
MFC after: 3 days
rather than forcing the state to LOOK. If we are in the middle of parsing
a line when we have to do a FILL we would have lost any token we were in
the middle of parsing and would have treated the next character as being
at the start of a new line instead.
PR: kern/89181
Submitted by: Antony Mawer gnats at mawer dot org
MFC after: 1 week
host name. This is matches the documented behaviro. The previous
behavior would remove the domain name even if the result retained a dot.
This fixes rsh connections from a.example.com to example.com.
Reviewed by: ceri (at least the concept)
Further contemplation has convinced me that this was
not going to really solve the problem of environment-poisoning
without raising serious administrative headaches. There
must be a better way...
This function removes all environment variables except
the ones listed on a "whitelist."
The function accepts two whitelist arguments.
If the first is NULL, a built-in default list will be
used. This allows callers to get a variety of behaviors:
* Default screening: provide NULL for both lists
* Custom screening: provide a custom list for the first argument
* Modified default screening: provide NULL for first arg,
list of additional variables to preserve in the second arg
Idea from: Jacques Vidrine
MFC after: 2 weeks
symbols exported by newer versions of libc, and so we want applications
depending on the newer library code to be required to link against the
newer libc.
Discussed with: scottl, kris, imp
binaries in /bin and /sbin installed in /lib. Only the versioned files
reside in /lib, the .so symlink continues to live /usr/lib so the
toolchain doesn't need to be modified.
or the tty, just block selected signals in the parent like system(3) does.
Many thanks to bde for his assistance in finding the correct solution.
PR: bin/50679
proper way to fix this. The way this works is to prepend "exec " to
the editor command to eliminate the "shell in the middle" which prevents
us from properly reawakening the editor after a SIGTSTP.
PR: bin/50679
password quality, not login.conf(5).
- Move warnexpire and warnpasswd from the ``Accounting Limits''
section to ``Authentication'', and nix everything else in the
former section. The accounting knobs are not available in
the base system, and the subset of them available in ports
should be documented in the ports' manpages.
PR: 47960
Reviewed by: mike (mentor), doc
(previous variant return NULL pointer for both empty file case and error case,
so caller can't sense error properly).
It not affect existen programs because property_find() now returns NULL
for both NULL pointer and NULLified struct.
information, since we rely on the pwd entry to know what MAC labels
to set as part of the login process.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
if the user has a 'label' entry in their login class. If so, attempt
to set that label on the process as part of the credential setup. If
we're unable to parse the label, or unable to set the label, fail.
In the future, we may also want to warn if a label is set but the
kernel doesn't support MAC.
Approved by: re
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
named by its argument and use ttyslot(3) instead to determine what slot to
use. The problem is that sshd(8) calls pam_open_session(3) before forking
the child (as it should), at which point it does not have a controlling
terminal. Also, ttyslot(3) is very crude as it assumes fd 0, 1 or 2 refers
to the controlling terminal, which is usually (but not always) the case.
Instead of using ttyslot(3) to determine the slot number, look up the
specified tty in /etc/ttys ourselves (this is what ttyslot(3) does anyway).
(perforce change 9969)
Sponsored by: DARPA, NAI Labs
them to point at static strings that contain the default paths. This
makes 'vipw -d' work again (I broke it in rev 1.21; apologies for taking
so long to fix it.)
Spotted by: Olivier Houchard <doginou@cognet.ci0.org>
Sponsored by: DARPA, NAI Labs
extattr namespace routines to the libc/posix1e directory. While
the extattr calls are not strictly POSIX.1e, POSIX.1e wasn't
strictly ever approved, so I think that's OK.
Obtained from: TrustedBSD Project
is not defined in the capability type list. Provide a definition for
'bool', if a slightly less than elegant one. Note that this definition
does not include the complete scope of available behavior defined
in cgetcap(3), and could probably be improved.
of NFS home directory and root directory processing fails to include
additional groups. This doesn't impact the final credential, but does
mean that users may be denied login even when additional groups might
allow it.
reserved word, causing breakage when a C++ program included libutil.h
This change will be propagated elsewhere shortly.
Submitted by: jkh
Obtained from: TrustedBSD Project
utility functions which convert between string namespace names and
numeric constants used by the interface. Right now, two namespaces
are supported, EXTATTR_NAMESPACE_SYSTEM ("system") and
EXTATTR_NAMESPACE_USER ("user"). These functions are used by
various userland EA utilities, rather than hard coding the routines
all over the place.
Obtained from: TrustedBSD Project
that applications know how large of a buffer they must allocate before
calling property_find(). Also added a $FreeBSD$ tag while I'm here.
Approved by: jkh
getting libutil/libcrypt to work properly. I've determined that GCC
thinks it can inline all functions, including weak-symboled ones, if
it feels like it.
Create a new stub.c and move any stubs there to prevent inlining.
Thanks to jdp and William S. Duncanson for helping me finally find the
problem.
and bump __FreeBSD_version to 500012 to mark the occasion.
setproctitle() is prototyped in unistd.h as opposed to stdlib.h
where OpenBSD and NetBSD have it.
Reviewed by: peter
for crypt(3) by now. In any case:
Add crypt_set_format(3) + documentation to -lcrypt.
Add login_setcryptfmt(3) + documentation to -lutil.
Support for switching crypt formats in passwd(8).
Support for switching crypt formats in pw(8).
The simple synopsis is:
edit login.conf; add a passwd_format field set to "des" or "md5"; go nuts :)
Reviewed by: peter
via IPv6, the hostname is trimed due to the length of IPv6 address.
This change saves it as possible.
I have a grudge against the shortage of UT_HOSTSIZE.
to be applied to the value given. This does not break installed
/etc/login.conf files, since un-suffixed numbers are interpreted as
they were before.
PR: 19750
Submitted by: Paul Herman <pherman@frenchfries.net>
when connecting via IPv6, hostname was not recorded to utmp anymore.
Because, if hostname is longer than buffer size, getnameinfo() returns
with ENI_MEMORY.
Reviewed by: shin
Approved by: jkh
Now libutil depends on libc.so.4, so needs to update the major version.
Without this, old binaries which use libutil and build with libc.so.3
will coredump on recent 4.0.
Solicited comment for cvs-committers and there seems to be no objection.
Approved by: jkh
string to u_long and back using two functions, flags_to_string and
string_to_flags, which co-existed with 'ls'. As time has progressed
more and more other tools have used these private functions to
manipulate the file flags.
Recently I moved these functions from /usr/src/bin/ls to libutil,
but after some discussion with bde it's been decided that they
really ought to go in libc.
There are two already existing libc functions for manipulating file
modes: setmode and getmode. In keeping with these flags_to_string
has been renamed getflags and string_to_flags to setflags.
The manual page could probably be improved upon ;)
This fixes some nasty procfs problems for SMP, makes ps(1) run much faster,
and makes ps(1) even less dependent on /proc which will aid chroot and
jails alike.
To disable this facility and revert to previous behaviour:
sysctl -w kern.ps_arg_cache_limit=0
For full details see the current@FreeBSD.org mail-archives.
