o Simplify the logic by removing a lot of unnecesary nesting
o Reduce the amount of local variables
o Zero-out the allocated structure and get rid of
all the unnecessary setting to 0 and NULL;
Refactor _pthread_mutex_destroy
o Simplify the logic by removing a lot of unnecesary nesting
o No need to check pointer that the mutex attributes points
to. Checking passed in pointer is enough.
a list in the thread structure to keep track of the locks and
how many times they have been locked. This list is checked
on every lock and unlock. The traversal through the list is
O(n). Most applications don't hold so many locks at once that
this will become a problem. However, if it does become a problem
it might be a good idea to review this once libthr is
off probation and in the optimization cycle.
This fixes:
o deadlock when a thread tries to recursively acquire a
read lock when a writer is waiting on the lock.
o a thread could previously successfully unlock a lock it did not own
o deadlock when a thread tries to acquire a write lock on
a lock it already owns for reading or writing [ this is admittedly
not required by POSIX, but is nice to have ]
- Update and improve the documentation for %[aA]
o Like %[eE], %[aA] may round the result if a precision is specified.
o Grammar police: Fix a split infinitive.
o The FreeBSD implementation does better than the minimum required
by C99 (literal translation of the mantissa). The digit before
the hexadecimal-point is never 0 unless the number itself is 0.
o Clarify that the exponent field represents a decimal exponent of 2.
o Discuss the fact that multiple valid representations are possible.
o Remove the entry in the BUGS section claiming that %[aA] is not
implemented.
- Remove the entry in the BUGS section claiming that the ' flag for
printing thousands separators is unimplemented for floating-point.
- Remove the entry in the BUGS section claiming that the L modifier
reduces the precision to "double" before conversion.
on the release media -- only put what is different in the crypto
version compared to the base version. This reduces PAM entries
in /usr/lib in the "crypto" distribution to:
libpam.a
libpam.so@
libpam.so.2
pam_krb5.so@
pam_krb5.so.2
pam_ksu.so@
pam_ksu.so.2
pam_ssh.so@
pam_ssh.so.2
The libpam.so* is still redundant (it is identical to the "base"
version), but we can't set DISTRIBUTION differently for libpam.a
and libpam.so.
(The removal of libpam.so* from the crypto distribution could be
addressed by the release/scripts/crypto-make.sh script, but then
we'd also need to remove redundant PAM headers, and I'm not sure
this is worth a hassle.)
these are not fully implemented and ifdef'd out, the bugs have
never manifested themselves. Specifically:
- Fix a memory leak in the case where %a follows another
floating-point format.
- Make the %a/%A code behave like %e/%E with respect to
precision.
- It is no longer valid to assume that '-' and '0x' are
mutually exclusive.
- Address other minor issues.
Makes it possible to have multiple packet aliasing instances in a
single process by moving all static and global variables into an
instance structure called "struct libalias".
Redefine a new API based on s/PacketAlias/LibAlias/g
Add new "instance" argument to all functions in the new API.
Implement old API in terms of the new API.
For pshared semaphore, this commit still does not enable cancellation
point, I think there should be a pthread_enter_cancellation_point_np
for libc to implement a safe cancellation point.
code and simply return EINVAL (which is allowed by the standard) in
all those pthread functions that previously initialized it.
o Refactor the pthread_rwlock_[try]rdlock() and pthread_rwlock_[try]wrlock()
functions. They are now completeley condensed into rwlock_rdlock_common()
and rwlock_wrlock_common(), respectively.
o If the application tries to destroy an rwlock that is currently
held by a thread return EBUSY where it previously went ahead and
freed all resources associated with the lock.
o Refactor _pthread_rwlock_init() to make it look (relatively) sane.
o When obtaining a read lock on an rwlock the check for whether it
would exceed the maximum allowed read locks should happen *before*
we obtain the lock.
o The pthread_rwlock_* functions shall *never* return EINTR, so make
sure to requeue/resuspend the thread if it encounters such an error.
o Make a note that pthread_rwlock_unlock() needs to ensure it holds a
lock on an rwlock it tries to unlock. It will be implemented in a
separate commit because it requires some additional rwlock infrastructure.
associated floppy if needed into a static split_openfile() function.
- Use this function in splitfs_open() to open the first chunk rather
than using open() directly. This allows the first chunk to be located
on a different disk than the actual foo.split file.
getpwent(3) or getpwuid(3) when using NIS adjunct maps. The bug was
present in the internal `nis_passwd' function. The lookup in the
adjunct map used the name passed into `nis_passwd', however no name
was of course supplied by getpwent or getpwuid. Correctly use the
name from the `struct pwd' that was found instead.
PR: bin/59962
Submitted by: Gabriel Gomez <ggomez@fing.edu.uy>
in contributed sources with just a hack made possible
by bsd.sys.mk,v 1.33. This is better because it just
nulls out the warning flags rather than adding gcc(1)
specific -w option to CFLAGS.
must first attach to the traced process. If the tracing process
exits without detaching, the traced process will be killed rather
than continued. For the duration of the tracing session, the traced
process is reparented to the tracing process (with resulting expected
behaviors). It is permissible to trace more than one other process
at a time. When using waitpid() to monitor the behavior of the traced
process, signals are intercepted: they may optionally then be
forwarded using ptrace(). Signals are generated normally by and for
the process, but also by the tracing facility (SIGTRAP).
Product of: Suffering
Sponsored by: DARPA, AFRL
at it, use the ANSI C generic pointer type for the second argument,
thus matching the documentation.
Remove the now extraneous (and now conflicting) function declarations
in various libc sources. Remove now unnecessary casts.
Reviewed by: bde
incorrectly when encountering `large' groups (many members and/or many
long member names). The reporter tracked this down to the glibc NSS
module compatibility code (nss_compat.c): it would prematurely record
that a NSS module was finished iterating through its database in some
cases.
Two aspects are corrected:
1. nss_compat.c recorded that a NSS module was finished iterating
whenever the module reported something other than SUCCESS. The
correct logic is to continue iteration when the module reports
either SUCCESS or RETURN. The __nss_compat_getgrent_r and
__nss_compat_getpwent_r routines are updated to reflect this.
2. An internal helper macro __nss_compat_result is used to map glibc
NSS status codes to BSD NSS status codes (e.g. NSS_STATUS_SUCCESS ->
NS_SUCCESS). It provided the obvious mapping.
When a NSS routine is called with a too-small buffer, the
convention in the BSD NSS code is to report RETURN. (This is used
to implement reentrant APIs such as getpwnam_r(3).) However, the
convention in glibc for this case is to set errno = ERANGE and
overload TRYAGAIN. __nss_compat_result is updated to handle this
case.
PR: bin/60287
Reported by: Lachlan O'Dea <odela01@ca.com>
on a rwlock while there are writers waiting. We normally favor
writers but when a reader already has at least one other read lock,
we favor the reader. We don't track all the rwlocks owned by a
thread, nor all the threads that own a rwlock -- we just keep
a count of all the read locks owned by a thread.
PR: 24641
waiting on a locked mutex. This involves passing a struct timespec
from the pthread mutex locking interfaces all the way down to the
function that suspends the thread until the mutex is released.
The timeout is assumed to be an absolute time (i.e. not relative to
the current time).
Also, in _thread_suspend() make the passed in timespec const.
o Remove some code duplication between _thread_init(), which is run once
to initialize libthr and the intitial thread, and pthread_create(), which
initializes newly created threads, into a new function called from both
places: init_td_common()
o Move initialization of certain parts of libthr into a separate
function. These include:
- Active threads list and it's lock
- Dead threads list and it's lock & condition variable
- Naming and insertion of the initial thread into the
active threads list.
ó++ ABI document at http://www.codesourcery.com/cxx-abi/abi.html#dso-dtor
The ABI was initially defined for ia64, but GCC3 and Intel compilers
have adopted it on other platforms.
This is the patch from PR bin/59552 with a number of changes by
me.
PR: bin/59552
Submitted by: Bradley T Hughes (bhughes at trolltech dot com)
C++ ABI document at http://www.codesourcery.com/cxx-abi/abi.html#dso-dtor
The ABI was initially defined for ia64, but GCC3 and Intel compilers
have adopted it on other platforms.
This is the patch from PR bin/59552 with a number of changes by
me.
PR: bin/59552
Submitted by: Bradley T Hughes (bhughes at trolltech dot com)
work before anyways, and I didn't want to fix broken code I had no
way of testing. It was necessary however, in order to get rid of GIANT_LOCK.
Pthread priorities will have to wait a little longer to get fixed.
problems: (1) The wrong flag was being checked for in the attribute
(2) The pthread's state was not being set to indicate it was
suspended.
Noticed by: Igor Sysoev <is@rambler-co.ru>
call (pam_get_authtok() will return the previous token if try_first_pass
or use_first_pass is specified). Incidentally fix an ugly bug where the
buffer holding the prompt was freed immediately before use, instead of
after.
likely to be non-zero. When leaving the cancellation point, check
the return value against -1 to see if cancellation should be
checked. While I'm here, make the same change to connect() just
to be consisitent.
Pointed out by: davidxu
_thr_leave_cancellation_point to _thr_cancel_leave, add a parameter
to _thr_cancel_leave to indicate whether cancellation point should be
checked, this gives us an option to not check cancallation point if
a syscall successfully returns to avoid any leaks, current I have
creat(), open() and fcntl(F_DUPFD) to not check cancellation point
after they sucessfully returned.
Replace some members in structure kse with bit flags to same some
memory.
Conditionally compile THR_ASSERT to nothing if _PTHREAD_INVARIANTS is
not defined.
Inline some small functions in thr_cancel.c.
Use __predict_false in thr_kern.c for some executed only once code.
Reviewd by: deischen
flags. We now create asynchronous contexts or syscall contexts only.
Syscall contexts differ from the minimal ABI dictated contexts by
having the scratch registers saved and restored because that's where
we keep the syscall arguments and syscall return values.
Since this change affects KSE, have it use kse_switchin(2) for the
"new" syscall context.
Instead of just deleting it, turn the original page into a general
overview of the multibyte character conversion functions, somewhat
similar to stdio(3).
UTS with the stack correctly aligned. Also, while here, use an indirect
jump rather than the pushq/ret hack.
This fixes threaded apps that use floating point for me, although
it hasn't solved all the problems. It is an improvement though.
Preservation of the 128 byte red zone hasn't been resolved yet.
Approved by: re (scottl)
ABI-required stack alignment. C code expects that the push of the
return address disturbed the 16 byte alignment and it will take corrective
measures to fix it before making another call. Of course, if its wrong
to start with, then all hell breaks loose. Essentially we "fix" this
by making the stack alignment odd to start with.
This was one of the things that broke on libkse with apps that use
floating point/varargs/etc.
Approved by: re (scottl)
we can end up with some threads with a non-16-byte-aligned stack. This
causes some interesting side effects, including general protection
faults leading to a SIGBUS when doing floating point or varargs. This
should be just a verbose NOP for the other platforms.
Approved by: re (scottl)
to sendfile(2) being erroneously automatically restarted after a signal
is delivered. Fixed by converting ERESTART to EINTR prior to exiting.
Updated manual page to indicate the potential EINTR error, its cause
and consequences.
Approved by: re@freebsd.org
through branch predict as suggested in INTEL IA32 optimization guide.
2.Allocate siginfo arrary separately to avoid pthread to be allocated at
2K boundary, which hits L1 address alias problem and causes context
switch to be slow down.
3.Simplify context switch code by removing redundant code, code size is
reduced, so it is expected to run faster.
Reviewed by: deischen
Approved by: re (scottl)
in init_main_thread. Also don't initialize lock and lockuser again for initial
thread, it is already done by _thr_alloc().
Reviewed by: deischen
Approved by: re (scottl)
initialization overhead, there's a problem in that we never call
imalloc() and thus malloc_init() for zero-sized allocations. As a
result, malloc(0) returns NULL when it's the first or only malloc in
the program. Any non-zero allocation will initialize the malloc code
with the side-effect that subsequent zero-sized allocations return a
non-NULL pointer. This is because the pointer we return for zero-
sized allocations is calculated from malloc_pageshift, which needs
to be initialized at runtime on ia64.
The result of the inconsistent behaviour described above is that
configure scripts failed the test for a GNU compatible malloc. This
resulted in a lot of broken ports.
Other, even simpler, solutions were possible as well:
1. initialize malloc_pageshift with some non-zero value (say 13 for
8KB pages) and keep the runtime adjustment.
2. Stop using malloc_pageshift to calculate ZEROSIZEPTR.
Removal of the runtime adjustment was chosen because then ia64 is the
same as any other platform. It is not to say that using a page size
obtained at runtime is bad per se. It's that there's currently a high
level of gratuity for its existence and the moment it causes problems
is the moment you need to get rid of it. Hence, it's not unthinkable
that this commit is (partially) reverted some time in the future when
we do have a good reason for it and a good way to achieve it.
Approved by: re@ (rwatson)
Reported by: kris (portmgr@) -- may the ports be with you
that they will be installed before application constructors are invoked.
Its possible to link applications such that this fails, application code
is invoked before they are installed, but, well, Don't Do That.
Approved by: re (jhb)
was rejected as a range error, while any values less than LONG_MIN
were silently substituted with LONG_MIN. Furthermore, on some
platforms `time_t' has less range than `long' (e.g. alpha), which may
give incorrect results when parsing some strings.
context of sockets, and document EINVAL as a possible failure mode
based on the object selected, not just the label provided.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
SO_PEERLABEL. This provides an interface to query the label of a
socket peer without embedding implementation details of mac_t in
the application. Previously, sizeof(*mac_t) had to be specified
by an application when performing getsockopt().
Document mac_get_peer(3), and expand documentation of the other
mac_get(3) functions. Note that it's possible to get EINVAL back
from mac_get_fd(3) when pointing it at an inappropriate object.
NOTE: mac_get_fd() and mac_set_fd() support for sockets will
follow shortly, so the documentation is slightly ahead of the
code.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
mac_free(3), which is used only for variables of type mac_t in
the FreeBSD implementation.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
constants NG_*SIZ that include the trailing NUL byte. This change
is mostly mechanical except for the replacement of a couple of snprintf()
and sprintf() calls with strlcpy.
is accessed for the first time as a result of an application looking
up label configuration information. Previously, the check and read
were kicked off by mac_prepare_(typename)() functions; since
mac_prepare_type() may now be directly employed by a user process,
push the check and initialization into that function.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
Replace occurences of the magic constant 2 with an offsetof macro
call that computes the size of the leading members of the sockaddr.
Use strlcpy instead of sprintf where appropriate. Document the new changes
in the man page.
- In __sigreturn call sigprocmask() to restore our signal state rather than
returning through sigreturn(). jmp to ___sigreturn to restore our register
state following this.
Requested by: pete
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
set NAS-IP-Address attribute in requests generated by the pam_radius
module. This attribute is mandatory for some Radius servers out there.
Reviewed by: des
MFC after: 2 weeks
on whether the parent chunk is of type whole. This also applies to
MBR slices for non-GPT disks. Since most of the GPT handling is
conditionally compiled, do the same with the partition naming.
This fixes a braino that caused slices to be named as GPT partitions
and generally messing up an install.
Pointy hat: marcel
string files (__SSTR flag set). This is necessary because __sputc()
does not respect the __SALC flag, and crashes trying to flush the buffer
instead of resizing it.
PR: 59167
sorting strings with common prefixes by noting
when all the strings land in just one bin.
Testing shows significant speedups (on the order of
30%) on strings with common prefixes and no slowdowns on any
of my test cases.
Submitted by: Markus Bjartveit Kruger <markusk@pvv.ntnu.no>
PR: 58860
Approved by: gordon (mentor)
by a parent that is a session leader (e.g., login shell) by ignoring
SIGHUP in before calling fork(2) and then restoring SIGHUP's action
after setsid(3). Based on the patch by Martin Kammerhofer
<mkamm@gmx.net>.
PR: bin/25462
Reviewed by: bde, alex.neyman@auriga.ru
signal handling mode, there is no chance to handle the signal, something
must be wrong in the library, just call kse_thr_interrupt to dump its core.
I have the code for a long time, but forgot to commit it.
Catch up with renaming of "Japanese" to "ja_JP.eucJP". Comment out the
statement that EUC is provided for compatibility with UNIX-based systems;
this is not a very good opening paragraph.
- fixed a length of the sadb extension in the case of pfkey_send_x5().
- used getprotobynumber() for printing a upper layer protocol name.
- modified the output format against the change of the setkey syntax
about a icmp6 type/code.
- don't enumerate reserved fields. use memset.
Obtained from: KAME
Aside from the POSIX requirements for pthread_atfork(), when
fork()ing, take the malloc lock to keep malloc state consistent
in the child.
Reviewed by: davidxu
it around an application's fork() call. Our new thread libraries
(libthr, libpthread) can now have threads running while another
thread calls fork(). In this case, it is possible for malloc
to be left in an inconsistent state in the child. Our thread
libraries, libpthread in particular, need to use malloc internally
after a fork (in the child).
Reviewed by: davidxu
mbstate_t object that they ignore. The zeroing is fairly expensive, and it
will never be necessary in these functions; when we support state-dependent
encodings, we will pass in a pointer to the file's mbstate_t object, and
only zero it at the time the file gets opened.
tcpdump -y ieee802_11 will work in the basic senses, including the
code compilation for filters (where you may specify "link[]" to refer
to parts of the 802.11 header, as well as treat it like a normal
Ethernet header). Previously, it was just too far off to do anything
useful for us.
* While I'm here, fix some compile problems that will result from lex
and yacc namespace polution when linking with -lpcap. The namespace
is now "pcapyy*" instead of "yy*", and it tests fine with world and
some external applications that may or may not use "yy*".
index referencing it. We need to know the original type and name
so that we know what to put in the table when we reconstruct it.
o Clear the table entries before we rebuild it to avoid that we
end up with stale data.
o Sequentially populate the table entries from the chunks. For the
chunks that have an index (now referencing the saved copy) we
use the saved type and name. This way we can handle unknown types
better. In all cases we update the start and end LBAs.
rather than generating an error. This is consistent with other tools
printing user and group names, and means you can read the ACL using
our tools rather than being up a creek.
PR: 56991
Submitted by: Michael Bretterklieber <mbretter@a-quadrat.at>
filling in the GPT entry. Both are already in sector numbers (LBA)
and exactly what we need for the entry. We now write a structurally
correct GPT partitioning.
part of the disk. The first appears to be a typo and instead of
dividing the media size with the sector size, we multiplied. The
second is an off-by-1 error that's the result of mixing up count
and index. The code in question is only applicable for virgin disks
and is used to create the "whole" chunk, which covers only the GPT
usable portion of the disk.
mbrtowc() and wcrtomb() directly. GB18030, GBK and UTF2 are left
unconverted; GB18030 will be done eventually, but GBK and UTF2 may just
be removed, as they are subsets of GB18030 and UTF-8 respectively.
platforms except ia64 and use Int_Open_Disk() in open_ia64_disk.c
on ia64. We need to know more than GEOM can provide us so we're
forced to read from the disk. Move uuid_type() to open_ia64_disk.c
and remove all references on non-ia64.
o Pass the GEOM conftxt to Int_Open_Disk() so that only Open_Disk()
needs to know about GEOM and libdisk can more easily be used with
media not handled by GEOM.
o Create an ia64 specific definiton of struct disk on ia64, because
we don't need/have most of the fields other platforms need and
other fields not applicable on platforms other than ia64.
o Do not compile change.c on ia64. It's too PC specific.
o In Fixup_Names() in create_chunk.c, try all partition numbers
that are valid for the GPT disk. We have the total number of
partitions that can be allocated in the disk structure on ia64.
Also, use the GPT partition naming if we're creating one under
a chunk of type "whole". It's a GPT partition in that case.
o In Create_Chunk(), compile-out the PC specific code on ia64 that
checks BIOS geometry restrictions.
o In Debug_Disk() in disk.c, dump the ia64 specific fields.
o Save the partition index in the chunk on ia64 so that we can
preserve it when we write the data back to disk. This avoids that
partitions get moved around or swapped after installing FreeBSD,
which may render a disk unusable.
Cyl_Aligned(), Prev_Cyl_Aligned() and Next_Cyl_Aligned() into
tautologies on ia64. GPT removes all notion of tracks, heads and
sectors per track, so there are no alignment considerations.
doesn't have any meaning and only results in lines longer than 80
characters.
o In Delete_Chunk2(), also look for chunks of type "part" under
chunks of type "whole" on ia64. They're not only under chunks of
type "freebsd" there.
as wrappers around the deprecated 4.4BSD rune functions. This paves the
way for state-dependent encodings, which the rune API does not support.
- Add __emulated_sgetrune() and __emulated_sputrune(), which are
implementations of sgetrune() and sputrune() in terms of
mbrtowc() and wcrtomb().
- Rename the old rune-wrapper mbrtowc() and wcrtomb() functions to
__emulated_mbrtowc() and __emulated_wcrtomb().
- Add __mbrtowc and __wcrtomb function pointers, which point to the
current locale's conversion functions, or the __emulated versions.
- Implement mbrtowc() and wcrtomb() as calls to these function pointers.
- Make the "NONE" encoding implement mbrtowc() and wcrtomb() directly.
All of this emulation mess will be removed, together with rune support,
in FreeBSD 6.
when the current implementation won't use it, anyway. Just pass NULL.
This will need to be changed when state-dependent encodings are
supported, but there's no need to take the performance hit
in the meantime.
in KAME implementation, even when no policy is installed
into kernel, getaddrinfo(3) sorts addresses. Since it
causes POLA violation, I modified to don't sort addresses
when no policy is installed into kernel,
Obtained from: KAME
This enable us to use /dev/fwmem* as a core file.
e.g.
ps -M /dev/fwmem0.0 -N kernel.debug
dmesg -M /dev/fwmem0.0 -N kernel.debug
gdb -k -c /dev/fwmem0.0 kernel.debug
You need to set target EUI64 in hw.firewire.fwmem.eui64_hi/lo before
opening the device. On the target arch, (PCI) bus address must be
equivalent to physical address.
(We cannot use this for sparc64 because of IOMMU.)
No objection in: -audit
send strhash(3) off to sleep with the fishes. Nothing in our tree uses it.
It has no documentation. It is nonstandard and in spite of the filename
strhash.c and strhash.h, it lives in application namespace by providing
compulsory global symbols hash_create()/hash_destroy()/hash_search()/
hash_traverse()/hash_purge()/hash_stats() regardless of whether you
#include <strhash.h> or not. If it turns out that there is a huge
application for this after all, I can repocopy it somewhere safer and
we can revive it elsewhere. But please, not in libc!
that are only in libc.so.5. This broke some 4.X applications linked
to libm and run under 5.X.
Background:
In C99, isinf() and isnan() cannot be implemented as regular
functions. We use macros that call libc functions in 5.X, but for
libm-internal use, we need to use the old versions until the next
time libm's major version number is bumped.
Submitted by: bde
Reported by: imp, kris
documented naming scheme (unfortunately the documentation isn't in the
tree as far as I can tell); no repocopy is required as there is no
history to preserve.
- replace simple and almost-correct implementation with slightly hackish
but definitely correct implementation (tested on i386, alpha, sparc64)
which requires pulling in fpmath.h and the MD _fpmath.h from libc.
- try not to make a mess of the Makefile in the process.
- enterprising minds are encouraged to implement more C99 long double
functions.
(aka RFC2292bis). Though I believe this commit doesn't break
backward compatibility againt existing binaries, it breaks
backward compatibility of API.
Now, the applications which use Advanced Sockets API such as
telnet, ping6, mld6query and traceroute6 use RFC3542 API.
Obtained from: KAME
the denormal/unnormal trap, is not a standard IEEE trap. We did
not exclude it from being returned by fpgetmask(), nor did we make
sure that fpsetmask() didn't clobber it. Since the non-IEEE trap
is not part of fp_except_t, users of ifpgetmask()/fpsetmask() would
be confronted with unexpected behaviour, one of which is a SIGFPE
for denormal/unnormal FP results.
This commit makes sure that we don't leak the denormal/unnormal mask
bit in fp_except_t and also that we don't clobber it.
closer to reality. More work remains to be done. st_mtime should
be the most complete based on IEEE Std 1003.1, 2003 Edition, a
review of ufs_vnops.c, and some experimentation.
about the fpu code here. It should be using fxsave/fxrstor instead of
saving/restoring the control word. The SSE registers are used a lot in
gcc generated code on amd64. I'm not sure how this all fits together
though.
section alignnment of 16 bytes for amd64 and this breaks file(1).
Before:
./cp: ELF 64-bit LSB executable, AMD x86-64, version 1 (FreeBSD), for \
FreeBSD 127.7.9, statically linked, stripped
after: ^^^^^^^
./ls: ELF 64-bit LSB executable, AMD x86-64, version 1 (FreeBSD), for \
FreeBSD 5.0.1, dynamically linked (uses shared libs), stripped
The reason for this is that the NOTE sections are not contiguous
internally. If the note section has an alignment of 16, then anything
that looks for the data is supposed to round up the payload start to
the next multiple of the alignment. But FreeBSD/amd64 broke because the
structure is declared as a single structure, not a (header,payload) group,
where the payload had an explicit alignment roundup.
The alternative is to change things like file(1) to ignore the ELF payload
alignment rules for the PT_NOTE section only for FreeBSD.
- fix hard sentence breaks
- sprinkle a few .Vt's where neccessary
- remove incorrect use of `\-'
- proper quoting using .Dq, instead of manual ``...''
Approved by: des@ (mentor)
Reviewed by: ru@
On ia64, where there's no libc_r at all, libkse is now the default
thread library by virtue of these links.
The reasons for this change are:
1. libkse is slated to become the default thread library anyway,
2. active development and maintenance is only present for libkse,
3. GNOME and KDE, both in the process of being supported on ia64,
work better with KSE; even on ia64.
there to support sysinstall, and enabling DEBUG creates spurious
console output that can't be read anyway... This slightly cleans up
the visual impression of the system install by not spamming the console
during the labeling of the disks.
(fstp stores a mem32 value, fstpl stores a mem64 value)
This fixes ghostscript for 'make release' on amd64. Ghostscript for some
reason thinks it is a good idea to use -fno-builtin, which means it is
vulnerable to bugs in libc that are normally hidden by the builtin gcc
functions. Oops.
can clear the pointer to mutex, not the thread doing mutex
handoff. Because _mutex_lock_backout does not hold scheduler
lock while testing THR_FLAGS_IN_SYNCQ and then reading mutex
pointer, it is possible mutex owner begin to unlock and
handoff the mutex to the current thread, and mutex pointer
will be cleared to NULL before current thread reading it, so
current thread will end up with deferencing a NULL pointer,
Fix the race by making mutex waiters to clear their mutex pointers.
While I am here, also save inherited priority in mutex for
PTHREAD_PRIO_INERIT mutex in mutex_trylock_common just like what
we did in mutex_lock_common.
Skinny is the protocol used by Cisco IP phones to talk to Cisco Call
Managers. With this code, one can use a Cisco IP phone behind a FreeBSD
NAT gateway.
Currently, having the Call Manager behind the NAT gateway is not supported.
More information on enabling Skinny support in libalias, natd, and ppp
can be found in those applications' manpages.
PR: 55843
Reviewed by: ru
Approved by: ru
MFC after: 30 days
for interrupted field.
Also in _thr_sig_handler, retrieve current signal mask from kernel not
from ucp, the later is pre-unioned mask, not current signal mask.
pthread_md.h. This commit only moves the definition; it does not
change it for any of the platforms. This more easily allows 64-bit
architectures (in particular) to pick a slightly larger stack size.
THR_SETCONTEXT as PANIC(). The THR_SETCONTEXT macro is currently not
used, which means that the definition we had could be wrong, overly
pessimistic or unknowingly right. I don't like the odds...
The new _ia64_break_setcontext() and corresponding kernel fixes make
KSE mostly usable. There's still a case where we don't properly
restore a context and end up with a NaT consumption fault (typically
an indication for not handling NaT collection points correctly),
but at least now mutex_d works...
state for amd64 was twice as large as necessary. Peter
recently fixed this, so the comment no longer applies.
Also, since the size of struct mcontext changed, adjust
the threads library version of get&set context to match.
FYI, any change layout/size change to any arch's struct
mcontext will likely need some minor changes in libpthread.
always widen the imputed netmask if it is narrower than the specified octets.
fixes a strange behaviour where inet_net_pton would always return 4 (bits)
for multicast addresses no matter how many octets were specified.
negotiated with Paul Vixie, original author of this function.
PR: standards/53151
Submitted by: Max Laier <max@love2party.net>
Optained from: OpenBSD
to avoid potential memory leak, also fix a bug in pthread_create, contention
scope should be inherited when PTHREAD_INHERIT_SCHED is set, and also check
right field for PTHREAD_INHERIT_SCHED, scheduling inherit flag is in sched_inherit.
2. Execute hooks registered by atexit() on thread stack but not on scheduler
stack.
3. Simplify some code in _kse_single_thread by calling xxx_destroy functions.
Reviewed by: deischen
should be a value past to pthread_attr_setguardsize, not a rounded up value.
Also fix a stack size matching bug in thr_stack.c, now stack matching code
uses number of pages but not bytes length to match stack size, so for example,
size 512 bytes and size 513 bytes should both match 1 page stack size.
Reviewed by: deischen
a shared library or any other dyanmic allocated data block, once
pthread_once_t is initialized, a mutex is allocated, if we unload the
shared library or free those data block, then there is no way to deallocate
the mutex, result is memory leak.
To fix this problem, we don't use mutex field in pthread_once_t, instead,
we use its state field and an internal mutex and conditional variable in
libkse to do any synchronization, we introduce a third state IN_PROGRESS to
wait if another thread is already in invoking init_routine().
Also while I am here, make pthread_once() conformed to pthread cancellation
point specification.
Reviewed by: deischen
specified directory is not found in the mount list. Before the
MNT_BYFSID changes, unmount(2) used to return ENOENT for a nonexistent
path and EINVAL for a non-mountpoint, but we can no longer distinguish
between these cases. Of the two error codes, EINVAL was more likely
to occur in practice, and it was the only one of the two that was
documented.
Update the manual page to match the current behaviour.
Suggested by: tjr
Reviewed by: tjr
no peer address information is desired.
PR: 56044
Submitted by: Felix Opatz <felix@zotteljedi.de> and
Bernd Luevelsmeyer <bdluevel@heitec.net>
MFC after: 1 month
value for getcontext() in a preserved register rather than on the stack.
The second time around, the stack value would likely have changed so we
can't depend on it for the return value.
otherwise the return from the syscall stub for getcontext will pop off
the return value for the caller to the getcontext stub and it will appear
as though the setcontext() syscall returned instead of the getcontext().
The same bug exists on amd64, a fix is coming there too.
The bug can be demonstrated with this test code fragment:
main()
{
ucontext_t top;
if (getcontext(&top) == 0) {
write(2, "PING!\n", 6);
/* Cause a return value of 1 from getcontext this time */
top.uc_mcontext.mc_eax = 1;
setcontext(&top);
err(1, "setcontext() returned");
}
write(2, "PONG!\n", 6);
_exit(0);
}
instead of long types for low-level locks.
Add prototypes for some internal libc functions that are
wrapped by the library as cancellation points.
Add memory barriers to alpha atomic swap functions (submitted
by davidxu).
Requested by: bde
the default label support in /etc/mac.conf. Rather than maintain
each default label type in an explicit global variable in mac.c,
keep a list of defaults loaded from the configuration file.
Generalize the parsing so that we support both the older:
default_file_labels foo
default_ifnet_labels foo
default_process_labels foo
And also a new:
default_labels file foo
default_labels ifnet foo
default_labels process foo
We now accept arbitrary object classes in the first argument. If
the same object is specified more than once, we discard the
earlier definition in favor of the later one.
Add a new API, mac_prepare_type(), which accepts a mac_t to
prepare, as well as an object name in the second argument, which
will pull a default label set for the object out of the
configuration loaded by mac_init_internal(). This permits the libc
to adapt to new objects known about by applications but not by libc
at compile-time.
Also liberalize the error handling a bit: if we're using implicit
initialization (i.e., the application didn't explicitly initialize
the MAC code), ignore syntax errors and only use valid lines. In
the future, we may want to add explicit warnings and do this a
bit more consistently.
While here, add support for a MAC_CONFFILE environmental variable,
which may be used to specify an alternative mac.conf configuration
file if the application isn't running with modified privilege
(issetugid()).
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
critical region, we wrap some syscalls for thread cancellation point, and
when syscalls returns, we call _thr_leave_cancellation_point, at the time
if a signal comes in, it would be buffered, and when the thread leaves
_thr_leave_cancellation_point, buffered signals will be processed, to avoid
messing up normal syscall errno, we should save and restore errno around
signal handling code.
gethostname()'s old and new signatures without requiring a library
bump. Note that programs which called gethostname() with a negative
argument were already broken, since the same type conversion was done
by the old implementation. Add a note in the Makefile so that whoever
next bumps the libc revision will delete the kluge at the same time
(as it will no longer be necessary). This is only operative on 64-bit
platforms.
Submitted by: marcel
when the buffer is not long enough to hold the current host name.
POSIX does not standardize error returns for gethostname(), so it
doesn't matter which one we use, but ENAMETOOLONG is at least a little
more intuitive, and mi suggests the existence of prior art. I've been
running with this change for a while on my home machine with no
effect. At the same time, I've updated the prototype for
gethostname() to use the correct standard type (size_t) for the
namelen argument.
All of the in-tree callers fall into one of the following categories:
1) Call perror() or equivalent when gethostname() fails.
2) Ignore gethostname()'s return value entirely, potentially resulting
in data corruption if the buffer is too small.
3) Fall back to a (possibly sensible) default value if gethostname()
fails.
Many of the callers I examined shows signs of confusion about the
correct sizing of the host name buffer. gethostname(3) now has more
information about this, as well as updated standards information.
PR: 48114
Submitted by: mi (in part)
yet, so we can protect some locking code from being interrupted by signal
handling. When KSE mode is turned on, reset the thread flag to scope process
except we are running in 1:1 mode which we needn't turn it off.
Also remove some unused member variables in structure kse.
Tested by: deischen
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.
have execute permissions. Run "perl verify" instead. Replace all
occurences of the hardcoding of ./verify with $(VERIFY) to allow
it to be overridden as well.
- All those diffs to syscalls.master for each architecture *are*
necessary. This needed clarification; the stub code generation for
mlockall() was disabled, which would prevent applications from
linking to this API (suggested by mux)
- Giant has been quoshed. It is no longer held by the code, as
the required locking has been pushed down within vm_map.c.
- Callers must specify VM_MAP_WIRE_HOLESOK or VM_MAP_WIRE_NOHOLES
to express their intention explicitly.
- Inspected at the vmstat, top and vm pager sysctl stats level.
Paging-in activity is occurring correctly, using a test harness.
- The RES size for a process may appear to be greater than its SIZE.
This is believed to be due to mappings of the same shared library
page being wired twice. Further exploration is needed.
- Believed to back out of allocations and locks correctly
(tested with WITNESS, MUTEX_PROFILING, INVARIANTS and DIAGNOSTIC).
PR: kern/43426, standards/54223
Reviewed by: jake, alc
Approved by: jake (mentor)
MFC after: 2 weeks
otherwise masks all signals until fork() returns, in child process,
we reset library state before restoring signal masks until we reach
a safe to point.
Reviewed by: deischen
didn't provide a constant for one of them (non-IEEE denormal trap),
in an attempt to not support it probably, it's not we are left with
the lower 5 bits.
o Properly mask the passed or returned fp_except_t. Not doing so
causes instant core dumps by trying to write an invalid value to
ar.fpsr. Now that we're masking, stop using exclusive-or to invert
bits.
This fixes the illegal instruction fault encountered when building
mozilla.
o fix the len argument of memcmp(3) to be the size of the node field
of the uuid structure, not the size of the uuid structure itself.
We're comparing the node fields...
o uuid_compare(3) is specified to return -1, 0 or 1, depending on
the outcome of the comparison. memcmp(3) returns the difference
between the first differing bytes. Hence, we cannot ever return
the return value of memcmp(3) as-is.
PR: standards/55370
Submitted by: Konstantin Oznobihin <bork@rsu.ru>
happens, the context of the interrupted thread is exported to
userland. Unlike most contexts, it will be an async context and
we cannot easily use our existing functions to set such a
context.
To avoid a lot of complexity that may possibly interfere with
the common case, we simply let the kernel deal with it. However,
we don't use the EPC based syscall path to invoke setcontext(2).
No, we use the break-based syscall path. That way the trapframe
will be compatible with the context we're trying to restore and
we save the kernel a lot of trouble. The kind of trouble we did
not want to go though ourselves...
However, we also need to set the threads mailbox and there's no
syscall to help us out. To avoid creating a new syscall, we use
the context itself to pass the information to the kernel so that
the kernel can update the mailbox. This involves setting a flag
(_MC_FLAGS_KSE_SET_MBOX) and setting ifa (the address) and isr
(the value).
wctob() in terms of wcrtomb() instead of sputrune(). There should be
no functional differences, but there may be a small performance hit
because we make an extra function call.
The aim here is to have as few functions as possible calling
s{get,put}rune() to make it easier to remove them in the future.
TCB. We know that the thread pointer points to &tcb->tcb_tp, so all
we have to do is subtract offsetof(struct tcb, tcb_tp) from the
thread pointer to get to the TCB. Any reasonably smart compiler will
translate accesses to fields in the TCB as negative offsets from TP.
In _tcb_set() make sure the fake TCB gets a pointer to the current
KCB, just like any other TCB. This fixes a NULL-pointer dereference
in _thr_ref_add() when it tried to get the current KSE.
makecontext(). We only supply 3, not 4. This is mostly harmless,
except that on ia64 the garbage can include NaT bits, resulting
in NaT consumption faults.
that the TLS is 16-byte aligned, as well as guarantee that the thread
pointer is 16-byte aligned as it points to struct ia64_tp. Likewise,
struct tcb and struct ksd are also guaranteed to be 16-byte aligned
(if they weren't already).
archs that can (or are required to) have per-thread registers.
Tested on i386, amd64; marcel is testing on ia64 and will
have some follow-up commits.
Reviewed by: davidxu
context functions. We don't need to enter the kernel anymore. The
contexts are compatible (ie a context created by getcontext() can
be restored by _ia64_restore_context()).
While here, make the use of THR_ALIGNBYTES and THR_ALIGN a no-op.
They are going to be removed anyway.
We write 1 for r8 in the context so that _ia64_restore_context()
will return with a non-zero value. _ia64_save_context() always
return 0.
o In _ia64_restore_context(), don't restore the thread pointer. It
is not normally part of the context. Also, restore the return
registers. We get called for contexts created by getcontext(),
which means we have to restore all the syscall return values.
Also change "Auto mode" to use a "special" value
instead of 0, and define and document it.
I had thought libpthread had already been switched to use auto mode but
it appears that patch hasn't been committed yet.
Discussed with: Davidxu
First of all, it should be written as: tr 'a-z' 'A-Z'
ranges not encolosed in [] according to POSIX, so [] just included
in the replacement.
Second, it should be written: tr '[:lower:]' '[:upper:]'
since a-z and A-Z may have different length in some locales.
that we can flush the register stack prior to entering the kernel.
This avoids having dirty registers and saves us from having to
manually write them to the backing store from within the kernel.
In that respect, flushing the RSE is both functionally required as
well as performance optimal.
On average we had 18 dirty registers when getcontext(2) was called
from libthr. Since libthr does not switch back to a context created
by getcontext(2), not having dealt with the dirty registers was
harmless.
on the corresponding .proc directive, or the .endp must not have a
name at all.
While here, remove an artificial dependency in Ovfork.S by performing
manual register renaming.
the userland version of [gs]etcontext to switch between a thread
and the UTS scheduler (and back again). This also fixes a bug
in i386 _thr_setcontext() which wasn't properly restoring the
context.
Reviewed by: davidxu
almost identical.
* Merge strchr(3) and strrchr(3) to strchr(3) since the two functions
are almost identical.
* Make the wording of index(3) and strchr(3) more similar.
* mdoc(7) cleanup.
Submitted by: SUZUKI Koichi <metal@gc5.so-net.ne.jp>, keramida, myself
PR: docs/32054
Reviewed by: ru
Approved by: ceri (mentor)
switching anymore, so there's no need to save and restore GP. This
change breaks threaded applications linked against libc_r. Pull the
tier 2 card again: relink. This will link against libthr instead.
do not also provide a __generic_XXX version as well. This is how we
used to runtime select the generic vs i387 versions on the i386 platform.
This saves a pile of #defines in the src/math_private.h file to undo the
__generic_XXX renames in some of the *.c files.
fp emulator, stop doing the runtime selection of hardware or emulated
floating point operations on i386. Note that I have not suppressed the
duplicate compiles yet.
While here, fix the alpha. It has provided specific copysign/copysignf
functions since the beginning of time, but they have never been used.
This eliminates ping-ponging of locks, where the idle KSE wakes
up only to find the lock it needs is being held. This gives
little or no gain to M:N mode but greatly speeds up 1:1 mode.
Reviewed & Tested by: davidxu
on that platform, invert the test for the platforms on
which libthr is built. Amd64 and powerpc are the only
platforms excluded.
Compile tested on: amd64, alpha
don't probe the server at all for passwd.by* maps. This fixes
interoperability with the Services For UNIX NIS server (which is
really a front end to Captive^WActiveDirectory). This server
incorrectly returns success for all YPPROC_MASTER requests,
even for maps that don't exist, which makes it impossible to
(ab)use it to probe for the existence of the master.passwd.by*
maps.
This is a little kludgey, but basically restores the original
behavior of getpwent.c as it is in -stable, and works around both
the lack of YPPROC_ORDER on NIS+ servers as well as the broken
YPPROC_MASTER on Services For UNIX servers.
handed-off/signaled to a higher priority thread. Note that when
there are idle KSEs that could run the higher priority thread,
we still add the preemption point because it seems to take the
kernel a while to schedule an idle KSE. The drawbacks are that
threads will be swapped more often between CPUs (KSEs) and
that there will be an extra userland context switch (the idle
KSE is still woken and will probably resume the preempted
thread). We'll revisit this if and when idle CPU/KSE wakeup
times improve.
Inspired by: Petri Helenius <pete@he.iki.fi>
Reviewed by: davidxu
is system bound thread and when it is blocked, no upcall is generated.
o Add ability to libkse to allow it run in pure 1:1 threading mode,
defining SYSTEM_SCOPE_ONLY in Makefile can turn on this option.
o Eliminate code for installing dummy signal handler for sigwait call.
o Add hash table to find thread.
Reviewed by: deischen
Previously, there were two copies of telnet; a non-crypto version
that lived in the usual places, and a crypto version that lived in
crypto/telnet/. The latter was built in a broken manner somewhat akin
to other "contribified" sources. This meant that there were 4 telnets
competing with each other at build time - KerberosIV, Kerberos5,
plain-old-secure and base. KerberosIV is no longer in the running, but
the other three took it in turns to jump all over each other during a
"make buildworld".
As the crypto issue has been clarified, and crypto _calls_ are not
a problem, crypto/telnet has been repo-copied to contrib/telnet,
and with this commit, all telnets are now "contribified". The contrib
path was chosen to not destroy history in the repository, and differs
from other contrib/ entries in that it may be worked on as "normal"
BSD code. There is no dangerous crypto in these sources, only a
very weak system less strong than enigma(1).
Kerberos5 telnet and Secure telnet are now selected by using the usual
macros in /etc/make.conf, and the build process is unsurprising and
less treacherous.
don't call it according to the runtime specification and especially
WRT to gp this can cause trouble. The gcc 3.3.1 import broke the
ia64 runtime because the compiler saved gp prior to us being able
to set it properly. Restoring gp after the calls would then invalidate
gp and cause segmentation faults later on.
By rewriting _start() as an assembly function, we also avoided even
more gcc dependences, by trying to use gcc specific features to work
around the problem.
This version of _start() does not reference _DYNAMIC. We register the
cleanup function when it's a non-NULL pointer. The kernel will always
pass a NULL pointer and dynamic linkers may pass a non-NULL pointer.
The machine independent code to set __progname now unfortunately is
written in assembly. So be it.
its waitset, but if the signal is not masked by the thread, the signal
can interrupt the thread and signal action can be invoked by the thread,
sigwait should return with errno set to EINTR.
Also save and restore thread internal state(timeout and interrupted)
around signal handler invoking.
