libgomp: Update to version 4.3.5.
This is a partial revert of r282115, to bring the fromal upstream
libgomp from GCC 4.3.5 Release under LGPLv2.1+.
This is only brought to ease the ongoing development of the CPU
affinity support.
This shall not be MFC'd.
While GCC 4.3 changed license, the library components remained under
LGPLv2.1 all the during all the existence of the branch.
From gcclibs, only libcpp and libgomp have some value for us in the base
gcc 4.2.1: bring updated versions from GCC 4.3.5 as reference as we have
already been including some of the fixes.
These not expected to become part of any FreeBSD release.
MACHINE_CPUARCH or MACHINE_ARCH, not MACHINE. The latter is for kernel
only things. Also, I think this should be unconditional since all our
architectures have long double support, but I don't have time to test
that thoroughly so just add a comment to that effect.
devd.conf execution of third-party software, that needs libraries
from /usr/local. Since devd is launched before ldconfig script, if
the hardware that has associated software is attached on boot, then
execution would fail.
Differential Revision: https://reviews.freebsd.org/D2332
Reviewed by: imp
The old comment in zfs_rezget explains what situation the code handles,
the new comment also describes how the situation can arise.
Also, re-join a line that became sufficiently shorti some time ago.
Differential Revision: https://reviews.freebsd.org/D2352
Reviewed by: delphij, smh
MFC after: 12 days
/dev/zfs always has per-open data, so when it is missing the file
descriptor is for some other file. Returning ENOENT in this case
is confusing as a variety of other conditions (like a missing dataset)
may result in the same error. It's better to consistently return
EBADF for any problems with the file descriptor.
Note that zfs_onexit_fd_hold() is used with 'automatic cleanup fd'
- when that fd is closed, typically because a process is terminated,
some cleanup action is taken by ZFS driver. E.g. a temporary
snapshot hold is released.
Perhaps, it would even be worthwhile changing devfs_get_cdevpriv()
to return EBADF if there is no associated data.
Differential Revision: https://reviews.freebsd.org/D2370
Reviewed by: delphij, smh
MFC after: 12 days
1) Advertise the actual min / max speeds the hardware is capable
of supporting given the reference clock used by the board.
2) Rather than attempting to extend the hardware's timeout register
in software (the hardware doesn't have sufficient bits to directly
support long timeouts), simply implement the same timeout approach
used in the SDXC driver.
3) Set the timeout for a linked command (e.g. STOP TRANSMISSION) based
on the previous multiblock read / write.
The changes have been smoke tested on both the ODROID-C1 and the VSATV102-M6
using the following cards:
* PQI 2GB microSD
* SanDisk 2GB microSD
* PQI 8GB SDHC (not a microSD so only tested on the ATV-102)
* PNY 8GB microSDHC
* SanDisk Ultra 32GB microSDHC
Submitted by: John Wehle
pages which pass a NULL virtual address. If the BUS_DMA_KEEP_PG_OFFSET
flag is set, use the physical address to compute the page offset
instead. The physical address should always be valid when adding
bounce pages and should contain the same page offset like the virtual
address.
Submitted by: Svatopluk Kraus <onwahe@gmail.com>
MFC after: 1 week
Reviewed by: jhb@
This is not likely to make it into a release and is basically disabled
but should still be useful for testing.
Obtained from: GCC pre-4.3 (rev. 123494, 125542; LGPLv2.1+)
ACPI C3 ends up doing a lot more work before entering sleep, some of which
requires grabbing a global ACPI hardware serialising mutex.
Because of this, the more CPU cores you have, the more that lock contends
under load, reaching close to the #1 lock contention (after VM, which is being
worked on.)
Tested:
* Sandy bridge Xeon, 2 socket * 8 core
* Ivy bridge Xeon v2, 2 socket * 8 core
* Westmere-EX, 4 socket * 10 core
* Ivybridge desktop
* Sandybridge mobile
* Ivybridge mobile
MFC after: 2 weeks
