hardclock() tick should be run on every active CPU, or on only one.
On my tests, avoiding extra interrupts because of this on 8-CPU Core i7
system with HZ=10000 saves about 2% of performance. At this moment option
implemented only for global timers, as reprogramming per-CPU timers is
too expensive now to be compensated by this benefit, especially since we
still have to regularly run hardclock() on at least one active CPU to
update system uptime. For global timer it is quite trivial: timer runs
always, but we just skip IPIs to other CPUs when possible.
Option is enabled by default now, keeping previous behavior, as periodic
hardclock() calls are still used at least to implement setitimer(2) with
ITIMER_VIRTUAL and ITIMER_PROF arguments. But since default schedulers don't
depend on it since r232917, we are much more free to experiment with it.
MFC after: 1 month
3rd argument of ifa->ifa_rtrequest is now ``rt_addrinfo *'' instead of
``sockaddr *''. While here, un-document RTM_RESOLVE cmd argument for
ifa_rtrequest() that became a stub after separating L2 tables in r186119.
MFC after: 1 week
Winbond Super I/O chips.
With minor efforts it should be possible the extend the driver to support
further chips/revisions available from Winbond. In the simplest case
only new IDs need to be added, while different chipsets might require
their own function to enter extended function mode, etc.
Sponsored by: Sandvine Incorporated ULC (in 2011)
Reviewed by: emaste, brueffer
MFC after: 2 weeks
that it is better to error out when people attempt to build using the
wrong bsd.*.mk files, than to silently ignore the problem.
This means, that after this commit, if you want to build kernel modules
by hand (or via a port) from a head source tree, you *must* make sure
the files in /usr/share/mk are in sync with that tree. If that isn't
possible, for example when you are running on an older FreeBSD branch,
you can:
- Run "make buildenv" from your head source tree, to have the correct
environment setup. (It's advisable to have run "make buildworld", or
at a minimum "make toolchain" first.)
- Alternatively, set MAKESYSPATH to the share/mk directory under your
head source tree. If your build tools are too old, other problems may
still occur.
- Alternatively, use "make -m" and specify the share/mk directory under
your head source tree. Again, build tools that are too old may still
result in trouble.
MFC after: 2 weeks
kernel modules using their old installed /usr/share/mk/bsd.*.mk files,
instead of the updated ones in their source tree. This leads to errors
like:
"sys/conf/kmod.mk", line 111: Malformed conditional (${MK_CLANG_IS_CC} == "no" && ${CC:T:Mclang} != "clang")
Obviously, these errors will go away after a "make installworld", or
alternatively, by using "make buildenv" before attempting to manually
build modules.
However, since it is apparently an expected use case to build using old
.mk files, change the way we test for clang, so it also works when the
MK_CLANG_IS_CC macro doesn't exist.
Note the conditional expressions are becoming rather unreadable now, but
I will attempt to fix that on a followup commit.
MFC after: 2 weeks
- Document the following routines: pci_alloc_msi(), pci_alloc_msix(),
pci_find_cap(), pci_get_max_read_req(), pci_get_vpd_ident(),
pci_get_vpd_readonly(), pci_msi_count(), pci_msix_count(),
pci_pending_msix(), pci_release_msi(), pci_remap_msix(), and
pci_set_max_read_req().
- Group the functions into five sub-sections: raw configuration access,
locating devices, device information, device configuration, and
message signaled interrupts.
- Discourage use of pci_disable_io() and pci_enable_io() in device drivers.
The PCI bus driver handles this automatically as resources are activated.
MFC after: 2 weeks
interface supported by mvs(4) are 88SX, while AHCI-like chips are 88SE.
PR: kern/165271
Submitted by: Jia-Shiun Li <jiashiun@gmail.com>
MFC after: 1 week
not ACPI-specific at all, but deal with PCI power states. Also,
pci_set_powerstate() fails with EOPNOTSUPP if a request is made that the
underlying device does not support rather than falling back to somehow
setting D0.
pci_cfg_save() and pci_cfg_restore() for device drivers to use when
saving and restoring state (e.g. to handle device-specific resets).
Reviewed by: imp
MFC after: 2 weeks
installs clang as /usr/bin/cc, /usr/bin/c++ and /usr/bin/cpp.
Note this does *not* disable building and installing gcc, which will
still be available as /usr/bin/gcc, /usr/bin/g++ and /usr/bin/gcpp. If
you want to disable gcc completely, you must use WITHOUT_GCC.
MFC after: 2 weeks
get rid of testing explicitly for clang (using ${CC:T:Mclang}) in
individual Makefiles.
Instead, use the following extra macros, for use with clang:
- NO_WERROR.clang (disables -Werror)
- NO_WCAST_ALIGN.clang (disables -Wcast-align)
- NO_WFORMAT.clang (disables -Wformat and friends)
- CLANG_NO_IAS (disables integrated assembler)
- CLANG_OPT_SMALL (adds flags for extra small size optimizations)
As a side effect, this enables setting CC/CXX/CPP in src.conf instead of
make.conf! For clang, use the following:
CC=clang
CXX=clang++
CPP=clang-cpp
MFC after: 2 weeks
USERSPACE:
1. add support for devices with different number of rx and tx queues;
2. add better support for zero-copy operation, adding an extra field
to the netmap ring to indicate how many buffers we have already processed
but not yet released (with help from Eddie Kohler);
3. The two changes above unfortunately require an API change, so while
at it add a version field and some spares to the ioctl() argument
to help detect mismatches.
4. update the manual page for the two changes above;
5. update sample applications in tools/tools/netmap
KERNEL:
1. simplify the internal structures moving the global wait queues
to the 'struct netmap_adapter';
2. simplify the functions that map kring<->nic ring indexes
3. normalize device-specific code, helps mainteinance;
4. start exploring the impact of micro-optimizations (prefetch etc.)
in the ixgbe driver.
Use 'legacy' descriptors on the tx ring and prefetch slots gives
about 20% speedup at 900 MHz. Another 7-10% would come from removing
the explict calls to bus_dmamap* in the core (they are effectively
NOPs in this case, but it takes expensive load of the per-buffer
dma maps to figure out that they are all NULL.
Rx performance not investigated.
I am postponing the MFC so i can import a few more improvements
before merging.
- Remove all attempts to guess physical temperature using DiodeOffset.
There are too many reports that it varies wildly depending on motherboard.
Instead, if it is known to scale well and its offset is known from other
temperature sensors on board, the user may set "dev.amdtemp.0.sensor_offset"
tunable to compensate the difference. Document the caveats in amdtemp(4).
- Add a quirk for Socket AM2 Revision G processors. These processors are
known to have a different offset according to Linux k8temp driver.
- Warn about Family 10h Erratum 319. These processors have broken sensors.
- Report temperature in more logical orders under dev.amdtemp node. For
example, "dev.amdtemp.0.sensor0.core0" is now "dev.amdtemp.0.core0.sensor0".
- Replace K8, K10 and K11 with official processor names in amdtemp(4).
- Convert to our standard SYNOPSIS language
- Consistently use "The .Nm driver"
- Rewrite .nf .fi block with mdoc
- Add pci dependency to SYNOPSIS
- Start .Bl list with new paragraph
Approved by: luigi
1. Add new functionality to the force_depend method to incorporate the
tests for whether the service is enabled and/or already running.
2. Add a new option to bypass checking only that the service is enabled
at boot time, and always check if it is running.
3. Use this new functionality to greatly simplify the rc.d scripts that
use force_depend.
4. Add a force_depend for statd in lockd
5. Remove the check that either nfs_server or nfs_client is _enable'd
from statd and lockd. This was always overkill, and prevented using
the {one|force}start options, as well as stop'ing on the command line.
6. The yp* scripts had some of their arguments in various weird orders.
Bring them into line with the model.
7. If mountd fails to create /var/db/mountdtab, err out.
Ideas, suggestions, and/or review from delphij and jilles.
Pointy hats are completely my responsibility however.
The reasoning behind this, is that if we are consistent in our
documentation about the uint*_t stuff, people will be less tempted to
write new code that uses the non-standard types.
I am not going to bump the man page dates, as these changes can be
considered style nits. The meaning of the man pages is unaffected.
MFC after: 1 month
Add support for updating the devfs mount (currently only changing the
ruleset number is supported).
Check mnt_optnew with vfs_filteropt(9).
This new option sets the specified ruleset number as the active ruleset
of the new devfs mount and applies all its rules at mount time. If the
specified ruleset doesn't exist, a new empty ruleset is created.
MFC after: 1 month
side effect of something else using them. If they haven't been loaded
already but you want to use them, say for configuring a jail, you're out
of luck.
So add a knob to always load the default rulesets. While I'm here document
the other devfs_ knobs in rc.conf.5.
This should have been included in change 230592.
The supported hardware section still needs some updating to reflect
the current reality.
MFC after: 3 days
TCP_KEEPCNT, that allow to control initial timeout, idle time, idle
re-send interval and idle send count on a per-socket basis.
Reviewed by: andre, bz, lstewart
kernel modules that include binary-only code.
More fine-grained control is provided via MK_SOURCELESS_HOST (for native code
that runs on host CPU) and MK_SOURCELESS_UCODE (for microcode).
Reviewed by: julian, delphij, freebsd-arch
Approved by: kib (mentor)
MFC after: 2 weeks
