* 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.
sched_clock() rather than using callouts. This means we no longer have to
take the load of the callout thread into consideration while balancing and
should make the balancing decisions simpler and more accurate.
Tested on: x86/UP, amd64/SMP
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
global mutex, accept_mtx, which serializes access to the following
fields across all sockets:
so_qlen so_incqlen so_qstate
so_comp so_incomp so_list
so_head
While providing only coarse granularity, this approach avoids lock
order issues between sockets by avoiding ownership of the fields
by a specific socket and its per-socket mutexes.
While here, rewrite soclose(), sofree(), soaccept(), and
sonewconn() to add assertions, close additional races and address
lock order concerns. In particular:
- Reorganize the optimistic concurrency behavior in accept1() to
always allocate a file descriptor with falloc() so that if we do
find a socket, we don't have to encounter the "Oh, there wasn't
a socket" race that can occur if falloc() sleeps in the current
code, which broke inbound accept() ordering, not to mention
requiring backing out socket state changes in a way that raced
with the protocol level. We may want to add a lockless read of
the queue state if polling of empty queues proves to be important
to optimize.
- In accept1(), soref() the socket while holding the accept lock
so that the socket cannot be free'd in a race with the protocol
layer. Likewise in netgraph equivilents of the accept1() code.
- In sonewconn(), loop waiting for the queue to be small enough to
insert our new socket once we've committed to inserting it, or
races can occur that cause the incomplete socket queue to
overfill. In the previously implementation, it was sufficient
to simply tested once since calling soabort() didn't release
synchronization permitting another thread to insert a socket as
we discard a previous one.
- In soclose()/sofree()/et al, it is the responsibility of the
caller to remove a socket from the incomplete connection queue
before calling soabort(), which prevents soabort() from having
to walk into the accept socket to release the socket from its
queue, and avoids races when releasing the accept mutex to enter
soabort(), permitting soabort() to avoid lock ordering issues
with the caller.
- Generally cluster accept queue related operations together
throughout these functions in order to facilitate locking.
Annotate new locking in socketvar.h.
It was based on the pty(4) driver which as a tty side an a non-tty side.
Nmdm(4) seems to have inherited two symmetric sides from pty but
unfortunately they are not quite ttys. Running a getty one one
side and tip on the other failed to produce NL->CRNL mapping for
instance.
Rip out the basically bogus cdevsw->{read,write} functions and rely
on ttyread() and ttywrite() which does the same thing.
Use taskqueue_swi_giant to run a task for either side to do what
needs to be done. (Direct calling is not an option as it leads to
recursion.) Trigger the task from the t_oproc and t_stop methods.
Default the ports to not ECHO. Since we neither rate limiting nor
emulation, two ports echoing each other is a really bad idea, which
can only be properly mitigated by rate limiting, rate emulation or
intelligent detection. Rate emulation would be a neat feature.
Ditch the modem-line emulation, if needed for some app, it needs
to be thought much more about how it interacts with the open/close
logic.
. Implement option -c, all partition sizes will be calculated
in cylinders as opposed to sectors. Since the Sun label is
inherently cylinder-based, this makes the job a little easier.
. Implement option -h, print the label in `human readable'
size/offset format.
. Implement SVR4-compatible VTOC-style elements. They are
fully optional, defaulting to the current behaviour where no
VTOC-style table will be written to disk. However, if
desired, the full functionality of the partitioning menu of
Solaris' format(1m) is now offered (and even more).
. When editing the label, do not loop around edit_label() where
a new template file is generated for each turn, this used to
be annoying in that any possible syntax error caused a
complaint, but then the template was created anew, so the
user had to perform all their editing again. Rather loop
inside edit_label(), similar to bsdlabel(8), so in case of
errors, the user will be presented their previous template
file again.
. If VTOC-style elements are present, the overlap checks are
made less stringent. Overlaps will still be warned about,
but overlaps of `unmountable' partitions against other ones
are no longer fatal. That way, e. g. VxVM encapsulated
disk labels can be fully edited in FreeBSD (but not in
Solaris ;-).
. In print_label(), generate the editing hints only if the -e
flag is in effect. Additionally, print a hint about the
total number of sectors in the (hardware) medium.
. When editing a label, allow for changing the geometry
emulation (and textual name) by modifying the "text:" line
on top. That way, a more effective emulation can be
chosen.
. When editing/reading a label, additionally allow for the
suffixes `s' (512-byte sectors), and `c' (cylinders) in the
partition size field.
. Finally, turn the stub man page into something that really
explains the entire thing.
APIC interrupt pin based on the settings in the corresponding interrupt
source structure.
- Use ioapic_program_intpin() in place of manual frobbing of the intpin
configuration in ioapic_program_destination() and ioapic_register().
- Use ioapic_program_intpin() to implement suspend/resume support for I/O
APICs.
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.
descriptors out of fdrop_locked() and into vn_closefile(). This
removes all knowledge of vnodes from fdrop_locked(), since the lock
behavior was specific to vnodes. This also removes the specific
requirement for Giant in fdrop_locked(), it's now only required by
code that it calls into.
Add GIANT_REQUIRED to vn_closefile() since VFS requires Giant.
nextpkt within the m_hdr was not being initialized to NULL for
!M_PKTHDR cases. *Maybe* this will fix weird socket buffer
inconsistency panics, but we'll see.
