sigsuspend, thread shouldn't wait, in old code, it may be
ignored.
When a signal handler is invoked in sigsuspend, thread gets
two different signal masks, one is in thread structure,
sigprocmask() can retrieve it, another is in ucontext
which is a third parameter of signal handler, the former is
the result of sigsuspend mask ORed with sigaction's sa_mask
and current signal, the later is the mask in thread structure
before sigsuspend is called. After signal handler is called,
the mask in ucontext should be copied into thread structure,
and becomes CURRENT signal mask, then sigsuspend returns to
user code.
Reviewed by: deischen
Tested by: Sean McNeil <sean@mcneil.com>
on all inputs of the form x.75, where x is an even integer and
log2(x) = 21. A similar problem occurred when rounding upward.
The bug involves the following snippet copied from rint():
i>>=1;
if((i0&i)!=0) i0 = (i0&(~i))|((0x100000)>>j0);
The constant 0x100000 should be 0x200000. Apparently this case was
never tested.
It turns out that the bit manipulation is completely superfluous
anyway, so remove it. (It tries to simulate 90% of the rounding
process that the FPU does anyway.) Also, the special case of +-0 is
handled twice (in different ways), so remove the second instance.
Throw in some related simplifications from bde:
- Work around a bug where gcc fails to clip to float precision by
declaring two float variables as volatile. Previously, we
tricked gcc into generating correct code by declaring some
float constants as doubles.
- Remove additional superfluous bit manipulation.
- Minor reorganization.
- Include <sys/types.h> explicitly.
Note that some of the equivalent lines in rint() also appear to be
unnecessary, but I'll defer to the numerical analysts who wrote it,
since I can't test all 2^64 cases.
Discussed with: bde
permission), try to continue in FTS_DONTCHDIR mode. Of course this
won't work for long paths, but we can't descend more than one pathname
component beyond the directory anyway if we lack search permission.
Here is a transcript demonstrating the change, where oldls is ls(1)
linked with the old fts(3):
das@VARK:~> mkdir t && touch t/{a,b,c} && chmod u-x t
das@VARK:~> oldls t
a b c
das@VARK:~> oldls -l t
das@VARK:~> \ls t
a b c
das@VARK:~> \ls -l t
ls: a: Permission denied
ls: b: Permission denied
ls: c: Permission denied
I had forgotten about this patch until bde reminded me. He reports
using it without problems for over a year.
PR: 45723
writable. Affected callers include fwrite(), put?(), and *printf().
The issue of whether this is the right errno for funopened streams is
unresolved, but that's an obscure case, and some errno is better than
no errno.
Discussed with: bde, jkh
distinguish files from dirs (trailing '/' indicated a dir). Since
POSIX.1-1987, this convention is no longer necessary. However, there
are current tar programs that pretend to write POSIX-compliant
archives, yet store directories as "regular files", relying on this
old filename convention to save them. <sigh> So, move the check for
this old convention so it applies to all tar archives, not just those
identified as "old."
Pointed out by: Broken distfile for audio/faad port
it sees a truncated input the first time it gets called.
(In particular, files shorter than 512 bytes cannot be tar archives.)
This allows the top-level archive_read_next_header code to
generate a proper error message for unrecognized file types.
Pointed out by: numerous ports that expect tar to extract non-tar files ;-(
Thanks to: Kris Kennaway
It does not appear to be possible to cross-build arm from i386 at the
moment, and I have no ARM hardware anyway. Thus, I'm sure there are
bugs. I will gladly fix these when the arm port is more mature.
Reviewed by: standards@
features appear to work, subject to the caveat that you tell gcc you
want standard rather than recklessly fast behavior
(-mieee-with-inexact -mfp-rounding-mode=d).
The non-standard feature of delivering a SIGFPE when an application
raises an unmasked exception does not work, presumably due to a kernel
bug. This isn't so bad given that floating-point exceptions on the
Alpha architecture are not precise, so making them useful in userland
requires a significant amount of wizardry.
Reviewed by: standards@
version called the higher-level archive_read_data and
archive_read_data_skip functions, which screwed up state management of
those functions. This bit of mis-design has existed for a long time,
but became a serious issue with the recent changes to the
archive_read_data APIs, which added more internal state to the
high-level archive_read_data function. Most common symptom was a
failure to correctly read 'L' entries (long filename) from GNU-style
archives, causing the message ": Can't open: No such file or
directory" with an empty filename.
Pointed out by: Numerous port build failures
Thanks to: Kris Kennaway
that as end-of-archive. Otherwise, a short read at this point
generates an error. This accomodates broken tar writers (such as the
one apparently in use at AT&T Labs) that don't even write a single
end-of-archive block.
Note that both star and pdtar behave this way as well.
In contrast, gtar doesn't complain in either case, and as a
result, will generate no warning for a lot of trashed archives.
Pointed out by: shells/ksh93 port (Thanks to Kris Kennaway)
push extract data down into archive_read_extract.c and out
of the library-global archive_private.h; push dir-specific
mode/time fixup down into dir restore function; now that the
fixup list is file-local, I can use somewhat more natural
naming.
Oh, yeah, update a bunch of comments to match current reality.
reflect src/libexec/rtld-elf/rtld.c rev. 1.68 - the globally-loaded
objects (RTLD_GLOBAL) are searched before the local object's DAG's.
PR: 62770
Submitted by: Kimura Fuyuki <fuyuki@nigredo.org>
We approximate pi with more than float precision using pi_hi+pi_lo in
the usual way (pi_hi is actually spelled pi in the source code), and
expect (float)0.5*pi_lo to give the low part of the corresponding
approximation for pi/2. However, the high part for pi/2 (pi_o_2) is
rounded to nearest, which happens to round up, while the high part for
pi was rounded down. Thus pi_o_2+(float)0.5*pi (in infinite precision)
was a very bad approximation for pi/2 -- the low term has the wrong
sign and increases the error drom less than half an ULP to a full ULP.
This fix rounds up instead of down for pi_hi. Consistently rounding
down instead of up should work, and is the method used in e_acosf.c
and e_asinf.c. The reason for the difference is that we sometimes
want to return precisely pi/2 in e_atan2f.c, so it is convenient to
have a correctly rounded (to nearest) value for pi/2 in a variable.
a_acosf.c and e_asinf.c also differ in directly approximating pi/2
instead pi; they multiply by 2.0 instead of dividing by 0.5 to convert
the approximation.
These complications are not directly visible in the double precision
versions because rounding to nearest happens to round down.
* New read_data_block is both sparse-file aware and uses zero-copy semantics
* Push read_data_block down into specific formats (opens door to
various encoded entry bodies, such as zip or gtar -S)
* Reimplement read_data, read_data_skip, read_data_into_fd in terms
of new read_data_block.
* Update documentation
It's unfortunate that I couldn't just call the new interface
archive_read_data, but didn't want to upset the API that much.
and not tanf() because float type can't represent numbers large enough
to trigger the problem. However, there seems to be a precedent that
the float versions of the fdlibm routines should mirror their double
counterparts.
Also update to the FDLIBM 5.3 license.
Obtained from: FDLIBM
Reviewed by: exhaustive comparison
where the exponent is an odd integer and the base is negative).
Obtained from: fdlibm-5.3
Sun finally released a new version of fdlibm just a coupe of weeks
ago. It only fixes 3 bugs (this one, another one in pow() that we
already have (rev.1.9), and one in tan(). I've learned too much about
powf() lately, so this fix was easy to merge. The patch is not verbatim,
because our base version has many differences for portability and I
didn't like global renaming of an unrelated variable to keep it separate
from the sign variable. This patch uses a new variable named sn for
the sign.
[t=p_l+p_h High]. We multiply t by lg2_h, and want the result to be
exact. For the bogus float case of the high-low decomposition trick,
we normally discard the lowest 12 bits of the fraction for the high
part, keeping 12 bits of precision. That was used for t here, but it
doesnt't work because for some reason we only discard the lowest 9
bits in the fraction for lg2_h. Discard another 3 bits of the fraction
for t to compensate.
This bug gave wrong results like:
powf(0.9999999, -2.9999995) = 1.0000002 (should be 1.0000001)
hex values: 3F7FFFFF C03FFFFE 3F800002 3F800001
As explained in the log for the previous commit, the bug is normally
masked by doing float calculations in extra precision on i386's, but
is easily detected by ucbtest on systems that don't have accidental
extra precision.
This completes fixing all the bugs in powf() that were routinely found
by ucbtest.
(1) The bit for the 1.0 part of bp[k] was right shifted by 4. This seems
to have been caused by a typo in converting e_pow.c to e_powf.c.
(2) The lower 12 bits of ax+bp[k] were not discarded, so t_h was actually
plain ax+bp[k]. This seems to have been caused by a logic error in
the conversion.
These bugs gave wrong results like:
powf(-1.1, 101.0) = -15158.703 (should be -15158.707)
hex values: BF8CCCCD 42CA0000 C66CDAD0 C66CDAD4
Fixing (1) gives a result wrong in the opposite direction (hex C66CDAD8),
and fixing (2) gives the correct result.
ucbtest has been reporting this particular wrong result on i386 systems
with unpatched libraries for 9 years. I finally figured out the extent
of the bugs. On i386's they are normally hidden by extra precision.
We use the trick of representing floats as a sum of 2 floats (one much
smaller) to get extra precision in intermediate calculations without
explicitly using more than float precision. This trick is just a
pessimization when extra precision is available naturally (as it always
is when dealing with IEEE single precision, so the float precision part
of the library is mostly misimplemented). (1) and (2) break the trick
in different ways, except on i386's it turns out that the intermediate
calculations are done in enough precision to mask both the bugs and
the limited precision of the float variables (as far as ucbtest can
check).
ucbtest detects the bugs because it forces float precision, but this
is not a normal mode of operation so the bug normally has little effect
on i386's.
On systems that do float arithmetic in float precision, e.g., amd64's,
there is no accidental extra precision and the bugs just give wrong
results.
there's no need to enable support for it separately
from 'tar.' (The call to enable gnutar support is
now just an alias for the tar support, left in to
avoid API breakage.)
multibyte representation in conversion state objects, store the
accumulated wide character, set number and number of bytes remaining
to avoid having to derive them every time mbrtowc() is called.
certain flags set (e.g., schg or uappend) would fail because the flags
were restored before the hardlink was created.
To address this, I've generalized the existing machinery for deferring
directory timestamp/mode restoration and used it to defer the
restoration of highly-restrictive flags to the end of the extraction,
after any links have been created.
Pointed out by: Pawel Jakub Dawidek (pjd@)
through byte by byte with mbrtowc(). In the usual case (buffer is big
enough to contain the multibyte character, character does not straddle
buffer boundary) this results in only one call to mbrtowc() for each
wide character read.
to the initial state when a stream is opened or seeked upon. Use the
stream's conversion state object instead of a freshly-zeroed one in
fgetwc(), fputwc() and ungetwc().
This is only a performance improvement for now, but it would also be
required in order to support state-dependent encodings.
followed are: Only 3 functions (pthread_cancel, pthread_setcancelstate,
pthread_setcanceltype) are required to be async-signal-safe by POSIX. None of
the rest of the pthread api is required to be async-signal-safe. This means
that only the three mentioned functions are safe to use from inside
signal handlers.
However, there are certain system/libc calls that are
cancellation points that a caller may call from within a signal handler,
and since they are cancellation points calls have to be made into libthr
to test for cancellation and exit the thread if necessary. So, the
cancellation test and thread exit code paths must be async-signal-safe
as well. A summary of the changes follows:
o Almost all of the code paths that masked signals, as well as locking the
pthread structure now lock only the pthread structure.
o Signals are masked (and left that way) as soon as a thread enters
pthread_exit().
o The active and dead threads locks now explicitly require that signals
are masked.
o Access to the isdead field of the pthread structure is protected by both
the active and dead list locks for writing. Either one is sufficient for
reading.
o The thread state and type fields have been combined into one three-state
switch to make it easier to read without requiring a lock. It doesn't need
a lock for writing (and therefore for reading either) because only the
current thread can write to it and it is an integer value.
o The thread state field of the pthread structure has been eliminated. It
was an unnecessary field that mostly duplicated the flags field, but
required additional locking that would make a lot more code paths require
signal masking. Any truly unique values (such as PS_DEAD) have been
reborn as separate members of the pthread structure.
o Since the mutex and condvar pthread functions are not async-signal-safe
there is no need to muck about with the wait queues when handling
a signal ...
o ... which also removes the need for wrapping signal handlers and sigaction(2).
o The condvar and mutex async-cancellation code had to be revised as a result
of some of these changes, which resulted in semi-unrelated changes which
would have been difficult to work on as a separate commit, so they are
included as well.
The only part of the changes I am worried about is related to locking for
the pthread joining fields. But, I will take a closer look at them once this
mega-patch is committed.
character, as some tar implementations incorrectly include a '/' with
the prefix.
Thanks to: Divacky Roman for the UnixWare 7 tarfile that
demonstrated this issue.
code available at tuhs.org, and found out that my chronology
is a bit off. In particular, Seventh Edition already used
the "linkflag" and "linkname" fields. Also, it appears that
there was no tar in Sixth Edition, contrary to what an earlier
tar.1 manpage claimed.
A few mdoc fixes also crept in here.
the size field for a hardlink entry. Specifically, ensure that
we do obey the size field for archives that we know are pax interchange
format archives, as required by POSIX.
Also, clarify the comment explaining why this is necessary and explain
the (very unusual) conditions under which it might fail.
a non-zero size but no body, some write a non-zero size and include
a body. To distinguish these cases, look for a valid tar header immediately
following a hardlink header with non-zero size.
of fcntl(2), flock(2), and lockf(3) advisory locks.
Add such a paragraph to the flock(2) manpage for the
sake of consistency.
Reviewed by: Cyrille Lefevre and Kirk McKusick on -arch
MFC after: 2 weeks
