As mentioned before, we should at least aim to have one piece of code in
both user space and kernel space that uses C11 atomics, to get some
coverage. This piece of code can be migrated trivially, so it's a good
candidate.
- Reconnect with some minor modifications, in particular now selsocket()
internals are adapted to use sbintime units after recent'ish calloutng
switch.
cancellation point. When enabling the cancellation, only process the
pending cancellation for asynchronous mode.
Reported and reviewed by: Kohji Okuno <okuno.kohji@jp.panasonic.com>
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
performance.
- Always free to the alloc bucket if there is space. This gives LIFO
allocation order to improve hot-cache performance. This also allows
for zones with a single bucket per-cpu rather than a pair if the entire
working set fits in one bucket.
- Enable per-cpu caches of buckets. To prevent recursive bucket
allocation one bucket zone still has per-cpu caches disabled.
- Pick the initial bucket size based on a table driven maximum size
per-bucket rather than the number of items per-page. This gives
more sane initial sizes.
- Only grow the bucket size when we face contention on the zone lock, this
causes bucket sizes to grow more slowly.
- Adjust the number of items per-bucket to account for the header space.
This packs the buckets more efficiently per-page while making them
not quite powers of two.
- Eliminate the per-zone free bucket list. Always return buckets back
to the bucket zone. This ensures that as zones grow into larger
bucket sizes they eventually discard the smaller sizes. It persists
fewer buckets in the system. The locking is slightly trickier.
- Only switch buckets in zalloc, not zfree, this eliminates pathological
cases where we ping-pong between two buckets.
- Ensure that the thread that fills a new bucket gets to allocate from
it to give a better upper bound on allocation time.
Sponsored by: EMC / Isilon Storage Division
real shared object and libssp_nonshared.a.
This was the last showstopper that prevented from enabling SSP for ports
by default. portmgr@ performed a buildworld which showed no significant
breakage with this patch.
Details:
On i386 for PIC objects, gcc uses the __stack_chk_fail_local hidden
symbol instead of calling __stack_chk_fail directly [1]. This happen
not only with our gcc-4.2.1 but also with the latest gcc-4.8. If you
want the very nasty details, see [2].
OTOH the problem doesn't exist on other architectures. It also doesn't
exist with Clang as the latter will somehow manage to create the
function in the object file at compile time (contrary to only
referencing it through a symbol that will be brought in at link time).
In a perfect world, when an object file is compiled with
-fstack-protector, it will be linked into a binary or a DSO with this
same flag as well, so GCC will add libssp_nonshared.a to the linker
command-line. Unfortunately, we don't control softwares in ports and we
may have such broken DSO. This is the whole point of this patch.
You can reproduce the problem on i386 by compiling a source file into an
object file with "-fstack-protector-all -fPIE" and linking it
into a binary without "-fstack-protector".
This ld script automatically proposes libssp_nonshared.a along with the
real libc DSO to the linker. It is important to understand that the
object file contained in this library will be pulled in the resulting
binary _only if_ the linker notices one of its symbols is needed (i.e.
one of the SSP symbol is missing).
A theorical performance impact could be when compiling, but my testing
showed less than 0.1% of difference.
[1] For 32-bit code gcc saves the PIC register setup by using
__stack_chk_fail_local hidden function instead of calling
__stack_chk_fail directly. See comment line 19460 in:
src/contrib/gcc/config/i386/i386.c
[2] When compiling a source file to an object file, if you use something
which is external to the compilation unit, GCC doesn't know yet if
this symbol will be inside or outside the DSO. So it expects the
worst case and routes the symbol through the GOT, which means
additional space and extra relocation for rtld(1).
Declaring a symbol has hidden tells GCC to use the optimal route (no
GOT), but on the other hand this means the symbol has to be provided
in the same DSO (namely libssp_nonshared.a).
On i386, GCC actually uses an hidden symbol for SSP in PIC objects
to save PIC register setup, as said in [1].
PR: ports/138228
PR: ports/168010
Reviewed by: kib, kan
Based on r134760:
Reset the seek pointer to 0 when a file is successfully opened,
since otherwise the initial seek offset will contain the directory
offset of the filesystem block that contained its directory entry.
This bug was mostly harmless because typically the directory is
less than one filesystem block in size so the offset would be zero.
It did however generally break loading a kernel from the (large)
kernel compile directory.
Also reset the seek pointer when a new inode is opened in read_inode(),
though this is not actually necessary now because all callers set
it afterwards.
PR: 177328
Submitted by: Eric van Gyzen
Reviewed by: iedowse
MFC after: 5 days
Store/restore the VFP registers in setjmp/longjmp on ARM EABI if VFP is
enabled in the kernel. It checks the hw.floatingpoint sysctl to see if
floating-point is available and uses this to determine if it should store
them. If it does it uses a different magic value so longjmp is able to know
if it should load them.
libusbx deprecated libusb_get_port_path and replaced it with
libusb_get_port_numbers. The latter omits an extra parameter which was
unused in the FreeBSD implementation anyway.
with merge the functions but leave out the code to save/load the VFP
registers as that requires other changes to ensure the VFP is enabled
first.
This removes storing the old fpa registers. These were never fully
supported, and the only user of this code I can find have moved to newer
CPUs which use a VFP.
as a fairly faithful implementation of the algorithm found in
PTP Tang, "Table-driven implementation of the Expm1 function
in IEEE floating-point arithmetic," ACM Trans. Math. Soft., 18,
211-222 (1992).
Over the last 18-24 months, the code has under gone significant
optimization and testing.
Reviewed by: bde
Obtained from: bde (most of the optimizations)
* Use integral numerical constants, and let the compiler do the
conversion to long double.
ld128/s_expl.c:
* Use integral numerical constants, and let the compiler do the
conversion to long double.
* Use the ENTERI/RETURNI macros, which are no-ops on ld128. This
however makes the ld80 and ld128 identical.
Reviewed by: bde (as part of larger diff)
* In the special case x = -Inf or -NaN, use a micro-optimization
to eliminate the need to access u.xbits.man.
* Fix an off-by-one for small arguments |x| < 0x1p-65.
ld128/s_expl.c:
* In the special case x = -Inf or -NaN, use a micro-optimization
to eliminate the need to access u.xbits.manh and u.xbits.manl.
* Fix an off-by-one for small arguments |x| < 0x1p-114.
Obtained from: bde
* Update the evaluation of the polynomial. This allows the removal
of the now unused variables t23 and t45.
ld128/s_expl.c:
* Update the evaluation of the polynomial and the intermediate
result t. This update allows several numerical constants to be
written as double rather than long double constants. Update
the constants as appropriate.
Obtained from: bde
and use macros to access the e component of the unions. This allows
the portions of the code in ld80 to be identical to the ld128 code.
Obtained from: bde
The names now coincide with the name used in PTP Tang's paper.
* Rename the variable from s to tbl to better reflect that
this is a table, and to be consistent with the naming scheme
in s_exp2l.c
Reviewed by: bde (as part of larger diff)
* Update Copyright years to include 2013.
ld128/s_expl.c:
* Correct and update Copyright years. This code originated from
the ld80 version, so it should reflect the same time period.
Reviewed by: bde (as part of larger diff)
I initially thought wchar_t was locale independent, but this seems to be
only the case on Linux. This means that we cannot depend on the *wc*()
routines to implement *c16*() and *c32*(). Instead, use the Citrus
libiconv that is part of libc.
I'll see if there is anything I can do to make the existing functions
somewhat useful in case the system is built without libiconv in the
nearby future. If not, I'll simply remove the broken implementations.
Reviewed by: jilles, gabor
identified, unify the code of check_deferred_signal() for all
architectures, making the variant under #ifdef x86 common.
Tested by: marius (sparc64)
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Per the NetBSD Foundation statement
Third parties are encouraged to change the license on any files which
have a 4-clause license contributed to the NetBSD Foundation to a
2-clause license.
check_deferred_signal() returns twice, since handle_signal() emulates
the return from the normal signal handler by sigreturn(2)ing the
passed context. Second return is performed on the destroyed stack
frame, because __fillcontextx() has already returned. This causes
undefined and bad behaviour, usually the victim thread gets SIGSEGV.
Avoid nested frame and the need to return from it by doing direct call
to getcontext() in the check_deferred_signal() and using a new private
libc helper __fillcontextx2() to complement the context with the
extended CPU state if the deferred signal is still present.
The __fillcontextx() is now unused, but is kept to allow older
libthr.so to be used with the new libc.
Mark __fillcontextx() as returning twice [1].
Reported by: pgj
Pointy hat to: kib
Discussed with: dim
Tested by: pgj, dim
Suggested by: jilles [1]
MFC after: 1 week
* Use ENTERI/RETURNI to allow the use of FP_PE on i386 target.
Reviewed by: das (and bde a long time ago)
Approved by: das (mentor)
Obtained from: bde (polynomial coefficients)
are workarounds for various symptoms of the problem described in clang
bugs 3929, 8100, 8241, 10409, and 12958.
The regression tests did their job: they failed, someone brought it
up on the mailing lists, and then the issue got ignored for 6 months.
Oops. There may still be some regressions for functions we don't have
test coverage for yet.
implementations visible for use by applications. The functions $F that
are now weak symbols are:
allocm, calloc, dallocm, free, malloc, malloc_usable_size,
nallocm, posix_memalign, rallocm, realloc, sallocm
The non-weak implementations of $F are exported as __$F.
Submitted by: stevek@juniper.net
Reviewed by: jasone@, kib@
Approved by: jasone@ (jemalloc)
Obtained from: juniper Networks, Inc
- Remove an unneeded variable.
- Fix whitespace bugs.
- Fix typoes in comment.
- Improve string handling a bit. Don't handroll strstr() and don't
terminate a strdup()'ed string. Instead, simply strndup() the part we
need.
If we were already provided a struct _citrus_iconv (e.g. through
iconv_open_into()), we should not call free() in case io_init_context()
fails. Instead, call it on the pointer of the allocated object, which
will be NULL in case of iconv_open_into().
The <uchar.h> header, part of C11, adds a small number of utility
functions for 16/32-bit "universal" characters, which may or may not be
UTF-16/32. As our wchar_t is already ISO 10646, simply add light-weight
wrappers around wcrtomb() and mbrtowc().
While there, also add (non-yet-standard) _l functions, similar to the
ones we already have for the other locale-dependent functions.
Reviewed by: theraven
If 'e' is used, the kernel must support the recently added pipe2() system
call.
The use of pipe2() with O_CLOEXEC also fixes race conditions between
concurrent popen() calls from different threads, even if the close-on-exec
flag on the fd of the returned FILE is later cleared (because popen() closes
all file descriptors from earlier popen() calls in the child process).
Therefore, this approach should be used in all cases when pipe2() can be
assumed present.
The old version of popen() rejects "re" and "we" but treats "r+e" like "r+".
clang on head between r239347 and r245428.
The former revision introduced CLOCK_PROCESS_CPUTIME_ID as a clock id
for the clock_gettime() function and friends, but it was only added in
<sys/time.h>, not in <time.h>. Any program including <time.h> would
therefore not be able to use CLOCK_PROCESS_CPUTIME_ID, even though the
value of _POSIX_CPUTIME indicates its existence. The latter revision
synchronized the defines again.
Work around this problem by defining the id on the command line for the
particular .cpp file that needs it. If the id ever changes value, this
hack will need to be updated.
The ability to clear a file descriptor's close-on-exec flag via
posix_spawn_file_actions_adddup2() is in fact proposed in Austin Group issue
#411.
MFC after: 1 week
As per POSIX.1-2008, posix_spawn_file_actions_add* return [EBADF] if a file
descriptor is negative, not [EINVAL]. The bug was only in the manual page;
the code is correct.
MFC after: 1 week
kernel-based POSIX semaphore descriptors to userland via procstat(1) and
fstat(1):
- Change sem file descriptors to track the pathname they are associated
with and add a ksem_info() method to copy the path out to a
caller-supplied buffer.
- Use the fo_stat() method of shared memory objects and ksem_info() to
export the path, mode, and value of a semaphore via struct kinfo_file.
- Add a struct semstat to the libprocstat(3) interface along with a
procstat_get_sem_info() to export the mode and value of a semaphore.
- Teach fstat about semaphores and to display their path, mode, and value.
MFC after: 2 weeks
There is no point in hiding, e.g. pmc.xscale(3) from a developer running
on amd64, when the target arch in question will probably never have
manual pages installed at all.
Reviewed by: sbruno, hiren
The pipe2() function is similar to pipe() but allows setting FD_CLOEXEC and
O_NONBLOCK (on both sides) as part of the function.
If p points to two writable ints, pipe2(p, 0) is equivalent to pipe(p).
If the pointer is not valid, behaviour differs: pipe2() writes into the
array from the kernel like socketpair() does, while pipe() writes into the
array from an architecture-specific assembler wrapper.
Reviewed by: kan, kib
The accept4() function, compared to accept(), allows setting the new file
descriptor atomically close-on-exec and explicitly controlling the
non-blocking status on the new socket. (Note that the latter point means
that accept() is not equivalent to any form of accept4().)
The linuxulator's accept4 implementation leaves a race window where the new
file descriptor is not close-on-exec because it calls sys_accept(). This
implementation leaves no such race window (by using falloc() flags). The
linuxulator could be fixed and simplified by using the new code.
Like accept(), accept4() is async-signal-safe, a cancellation point and
permitted in capability mode.
that the method is not supported, return an empty string.
This looks more handy for callers like procstat(1), which will not
abort after the failed call and still output some useful information.
MFC after: 3 weeks
!defined(LIBRARIES_ONLY) so it is only created once on architectures
with 32-bit compat support.
Replace ln -fhs with ${INSTALL_SYMLINK} to the link is logged in the
METALOG.
is a bit obfuscated here, as ia64 adds string source files elsewhere, so
simply exclude it here.
Reviewed by: imp
MFC after: 3 days
Sponsored by: DARPA, AFRL
Trim two now-unneeded (and likely harmful) lines from the libstand
setjmp/longjmp for MIPS.
Spotted by: jmallett
MFC after: 3 days
Sponsored by: DARPA, AFRL
Clang only supports atomic operations for ARMv6. For non-ARMv6, we still
need to emit these functions.
Clang's prototype for these functions slightly differs, as it is truly
based on GCC's documentation. It requires the use of signed types, but
also requires varargs. Still, we are not allowed to simply implement
this function directly. Cleverly work around this by implementing it
under a different name and using __strong_reference().
These APIs were relevant when memory for virtual machine allocation was
hard partitioned away from the rest of the system but that is no longer
the case. The sysctls that provided this information were garbage collected
a while back.
Obtained from: NetApp
There are no getdtablesize() bounds on the file descriptor to be duplicated;
it only has to be open. If the RLIMIT_NOFILE rlimit was decreased after
opening the file descriptor, it may be greater than or equal to
getdtablesize() but still valid.
MFC after: 1 week
Originally we disabled libcompiler-rt on MIPS and SPARC64, because of an
issue where __clzdi2 and __ctzdi2 would cause endless recursion. This
bug has been fixed in r230021 already, but for some reason we only
switched to libcompiler-rt on SPARC64 -- not MIPS.
This means we can finally use <stdatomic.h> on all our architectures.
