drivers already map sections into KVA as needed anyway. Note that this
will probably break the nvidia driver, but I will coordinate to get that
fixed.
MFC after: 2 weeks
to search for a specific extended capability. If the specified capability
is found for the given device, then the function returns success and
optionally returns the offset of that capability. If the capability is
not found, the function returns an error.
holiday is now celebrated on December 1st. From the PR:
December 1 was adopted as National Day in 1990, being the day of
celebration of the Great Assembly of Alba Iulia which voted for the
union of Transylvania with Romania and which symbolise the union of all
Romanians within a single state and the achievement of the unity of
Romanian national state. [1]
[1] LAW Number 10 from July 31st, 1990
Regarding the proclamation of the National Day of Romania
http://www.1decembrie.ro/en/index.php?option=com_content&task=view&id=1&Itemid=4
PR: docs/90673
Submitted by: Ion-Mihai "IOnut" Tetcu
Originally pointed out by: Cornel Ilie <cornel dot c punkt ilie at gmail punkt com>
distributed non-large args, this saves about 14 of 134 cycles for
Athlon64s and about 5 of 199 cycles for AthlonXPs.
Moved the check for x == 0 inside the check for subnormals. With
gcc-3.4 on uniformly distributed non-large args, this saves another
5 cycles on Athlon64s and loses 1 cycle on AthlonXPs.
Use INSERT_WORDS() and not SET_HIGH_WORD() when converting the first
approximation from bits to double. With gcc-3.4 on uniformly distributed
non-large args, this saves another 4 cycles on both Athlon64s and and
AthlonXPs.
Accessing doubles as 2 words may be an optimization on old CPUs, but on
current CPUs it tends to cause extra operations and pipeline stalls,
especially for writes, even when only 1 of the words needs to be accessed.
Removed an unused variable.
The purpose of this change is consistency (not performance improvement:)),
as it was hard to tell if fdrop() is MPSAFE or not when I saw it sometimes
under the Giant and sometimes without it.
Glanced at by: ssouhlal, kan
special handling when zero. This caused no PFSYNC_ACT_DEL message and thus
disfunction of pfflowd and state synchronisation in general.
Discovered by: thompsa
Good catch by: thompsa
MFC after: 7 days
We loaded PAE kernels at 2MB and non-PAE kernels at 4MB to skip
the first PSE page (PSE pages are 2MB on PAE), not at 8MB as it
was incorrectly specified in the previous commit.
Submitted by: jhb
to light by the PR. Specifically, convert these three scripts
into good rc.d citizens, making sure that their functionality
is preserved, but the rc.d framework rules are not broken.
Add support for cleanvar as a regular rc.d script in the
default rc.conf, and document this in the man page.
Add a descriptive comment to rc.conf that regarding the
three emulation/compatibility services provided by abi
so users will not be confused by these services not having
their own startup scripts.
PR: conf/84574
Submitted by: Alexander Botero-Lowry
provide enough room for decompression (up to 2.5MB is necessary). This
should be safe to do since we load i386 kernels after 8MB mark now, so
that 16MB is the minimum amount of RAM necessary to even boot FreeBSD.
This makes bzip2-support practically useable.
memory directly available to loader(8) and friends was limited to 640K on i386.
Those times have passed long time ago and now loader(8) can directly access
up to 4GB of RAM at least theoretically. At the same time, there are several
places where it's assumed that malloc() will only allocate memory within
first megabyte.
Remove that assumption by allocating appropriate bounce buffers for BIOS
calls on stack where necessary.
This allows using memory above first megabyte for heap if necessary.
function approximation for the second step. The polynomial has degree
2 for cbrtf() and 4 for cbrt(). These degrees are minimal for the final
accuracy to be essentially the same as before (slightly smaller).
Adjust the rounding between steps 2 and 3 to match. Unfortunately,
for cbrt(), this breaks the claimed accuracy slightly although incorrect
rounding doesn't. Claim less accuracy since its not worth pessimizing
the polynomial or relying on exhaustive testing to get insignificantly
more accuracy.
This saves about 30 cycles on Athlons (mainly by avoiding 2 divisions)
so it gives an overall optimization in the 10-25% range (a larger
percentage for float precision, especially in 32-bit mode, since other
overheads are more dominant for double precision, surprisingly more
in 32-bit mode).
1. The ELF-64 typedefs are now standardized, so that the libelf port
(devel/libelf) does not need to compensate for not having the
Elf64_Xword and Elf64_Sxword types.
2. ELF Symbol versioning support has been added. This also affects
the libelf port (though configure should detect this correctly).
- in preparing for the third approximation, actually make t larger in
magnitude than cbrt(x). After chopping, t must be incremented by 2
ulps to make it larger, not 1 ulp since chopping can reduce it by
almost 1 ulp and it might already be up to half a different-sized-ulp
smaller than cbrt(x). I have not found any cases where this is
essential, but the think-time error bound depends on it. The relative
smallness of the different-sized-ulp limited the bug. If there are
cases where this is essential, then the final error bound would be
5/6+epsilon instead of of 4/6+epsilon ulps (still < 1).
- in preparing for the third approximation, round more carefully (but
still sloppily to avoid branches) so that the claimed error bound of
0.667 ulps is satisfied in all cases tested for cbrt() and remains
satisfied in all cases for cbrtf(). There isn't enough spare precision
for very sloppy rounding to work:
- in cbrt(), even with the inadequate increment, the actual error was
0.6685 in some cases, and correcting the increment increased this
a little. The fix uses sloppy rounding to 25 bits instead of very
sloppy rounding to 21 bits, and starts using uint64_t instead of 2
words for bit manipulation so that rounding more bits is not much
costly.
- in cbrtf(), the 0.667 bound was already satisfied even with the
inadequate increment, but change the code to almost match cbrt()
anyway. There is not enough spare precision in the Newton
approximation to double the inadequate increment without exceeding
the 0.667 bound, and no spare precision to avoid this problem as
in cbrt(). The fix is to round using an increment of 2 smaller-ulps
before chopping so that an increment of 1 ulp is enough. In cbrt(),
we essentially do the same, but move the chop point so that the
increment of 1 is not needed.
Fixed comments to match code:
- in cbrt(), the second approximation is good to 25 bits, not quite 26 bits.
- in cbrt(), don't claim that the second approximation may be implemented
in single precision. Single precision cannot handle the full exponent
range without minor but pessimal changes to renormalize, and although
single precision is enough, 25 bit precision is now claimed and used.
Added comments about some of the magic for the error bound 4/6+epsilon.
I still don't understand why it is 4/6+ and not 6/6+ ulps.
Indent comments at the right of code more consistently.
we can cache its value in the softc. Eliminates one PCI register
write per call to bge_start().
A 1.8% speedup for UDP_RR test on my old box.
Obtained from: NetBSD(jonathan) via delphij
to be compatible with symbol versioning support as implemented by
GNU libc and documented by http://people.redhat.com/~drepper/symbol-versioning
and LSB 3.0.
Implement dlvsym() function to allow lookups for a specific version of
a given symbol.
case if memory allocation failed.
- Remove fourth argument from VLAN_INPUT_TAG(), that was used
incorrectly in almost all drivers. Indicate failure with
mbuf value of NULL.
In collaboration with: yongari, ru, sam
