Commit Graph

185231 Commits

Author SHA1 Message Date
Sean Bruno
b68831134f Install include files for netpfil/pf when requested by the Makefile
Reviewed by:	glebius
2013-11-01 00:32:26 +00:00
Marcel Moolenaar
fde9169e11 The PAL_PTCE_INFO function returns the counts and strides of the
outer and inner loop as 32-bit integers mux'd in 64-bit return
values. Change our data types for the count and stride to match
and simplify/adjust accordingly.
Note that with this change the defaults of the ptc.e parameters
have changed. Since all hardware is supposed to support the PAL
call, there should be no impact. Even ski is unaffected, because
the TC is re-initialized without considering the virtual address.
So, as long as we call ptc.e at least once, we're good. That's
what the new defaults do.
Most processor implementations need but a single ptc.e to purge
the entire TC anyway...
2013-11-01 00:21:38 +00:00
Ian Lepore
034e9ed611 Add the Soc- / machine-dependent parts of imx6 support. 2013-11-01 00:21:09 +00:00
Alexander Motin
9813c936c4 Fix lock recursion, triggered by smartctl -a /dev/adaX. 2013-11-01 00:14:15 +00:00
Ian Lepore
8ef18b3b15 Convert the if/else list of compatible devices to the table-driven
ofw_bus_search_compatible() routine.  In addition to converting existing
strings to table entries, also add compat strings for the whole imx family.
2013-10-31 23:57:33 +00:00
Ian Lepore
3d9d5b4313 Add definitions for the register and data that describes the SoC type.
This isn't in the chip reference manuals, it was found in u-boot and
various old mailing list threads.
2013-10-31 23:08:30 +00:00
Marcel Moolenaar
3598d9555a Purge the translation cache of APs before we unleash them. To that
end, make pmap_invalidate_all() global and have it only handle the
local CPU -- i.e. no rendezvous. We do not use pmap_invalidate_all
other than during initialization.
Note that the BSP already purges its TC -- it was missing for APs
only. Nonetheless, this so far seems to eliminate random problems.
2013-10-31 23:06:04 +00:00
Ian Lepore
f955880f7b Revamp the SoC identity numbering scheme to be more in line with the way
Freescale numbers the chips in the ID registers.
2013-10-31 23:05:05 +00:00
Marcel Moolenaar
ec9b648ef8 Respect the kern.smp.disabled tunable. When we're scanning the MADT in
ia64_probe_sapics(), we also create PCPU structures for any Local SAPICs
we encounter. When SMP is disabled, this leaves us with partially setup
PCPU structures, which typically results in panics when we're iterating
over CPUs. When SMP is disabled, we now prevent the creation of the
PCPU structures.
2013-10-31 22:46:03 +00:00
Andreas Tobler
12e7de228a Ouch, do the shift properly. An uin64_t needs the right postfix when shifting. 2013-10-31 20:42:47 +00:00
Maksim Yevmenkin
f7a3a2a57c Rate limit (to once per minute) "Listen queue overflow" message in
sonewconn().

Reviewed by:	scottl, lstewart
Obtained from:	Netflix, Inc
MFC after:	2 weeks
2013-10-31 20:33:21 +00:00
Gleb Smirnoff
c1fd62d4c1 Remove 'netstat -i' from crashinfo, as this no longer is supported. 2013-10-31 20:30:56 +00:00
Sean Bruno
3055dc3e48 Integrate github revision 0446821e49224c6a29be28760cc9c4cc14f56d8a as
suggested by upstream

https://github.com/the-tcpdump-group/libpcap/issues/327

Quiesce assignment warnings by use of const
2013-10-31 20:03:59 +00:00
Mikolaj Golub
b24e24eb7d Add myself as a ports committer
Approved by:	bdrewery (mentor)
2013-10-31 19:55:36 +00:00
Sean Bruno
a9f8a8772a Queisce warning about attempting to add char * together and use explicit
array indexing to indicate what is meant to be done
2013-10-31 18:53:39 +00:00
Sean Bruno
9113c7c613 Queisce warning about attempting to add char * together and use explicit
array indexing to indicate what is meant to be done
2013-10-31 18:53:13 +00:00
Ed Maste
ce086c793b Fix up FreeBSD tag for files not from a vendor branch
Unexpand the tag, remove the fbsd:nokeywords property and add the
svn:keywords property.  This should eliminate the gratuituous diffs
that appear on these files in projects branches.

Sponsored by:	The FreeBSD Foundation
2013-10-31 18:44:40 +00:00
Andre Oppermann
5b74cfe42f Make struct ifnet readable and comprehensible again by grouping
and ordering related variables, fields and locks next to each
other.  Add more comments to variables.

Over time 'ifnet' has accumlated a lot of additional pointers and
functionality in an unstructured way making it quite hard to read
and understand while obfuscating relationships between fields and
variables.

Quantify the structure size and how bloated it has become.

This is only a mechanical change in preparation for upcoming
work to make ifnet opaque to drivers and to separate out the
interface queuing.

Sponsored by:	The FreeBSD Foundation
2013-10-31 15:46:10 +00:00
Ian Lepore
7352de58e8 Add sdhci driver glue for imx family SoCs. This should support both uSDHC
(newer SoCs) and eSDHC (older SoCs), but the eSDHC support is untested and
likely to need some tweaking.
2013-10-31 15:27:39 +00:00
Ian Lepore
9808ebfa39 Add stubbed-out imx6 support for clocks and power management. This
contains little more than a few stub functions required to keep the
linker happy, but it's enough to let early imx6 development proceed.
2013-10-31 15:04:23 +00:00
Ian Lepore
c8c16b3df2 Add support for the USB PHY on imx6 SoCs. Pretty minimal at this point,
but enough to get usb host mode working.
2013-10-31 14:52:06 +00:00
Adrian Chadd
c48efaecf6 Allow the Arduino Leonardo to work by supporting CDC=0 devices.
CDC=0 simply means "no command codes", CDC=1 means "AT command codes."
There's no driver change required!  It's purely to tell the application
layer whether to speak AT commands or not.  Things are all still serial.

PR:		usb/183505
Reviewed by:	hps
MFC after:	1 week
2013-10-31 13:54:51 +00:00
Baptiste Daroussin
c92d635ad3 After around 20 years of duty it is time for pkg_install to retire 2013-10-31 13:00:35 +00:00
Baptiste Daroussin
8cf0e32801 Adjust the manpage category for the pkg bootstrap 2013-10-31 12:05:37 +00:00
Pawel Jakub Dawidek
b19d096363 Make lpathconf(2) support optional to make it compile again on Linux.
Submitted by:	Hashem Nasarat @riseup.net
2013-10-31 11:38:25 +00:00
Hans Petter Selasky
645826bef4 Fix a potential memory leak.
MFC after:	1 week
2013-10-31 11:03:36 +00:00
Konstantin Belousov
d2812536e8 Remove the dependency on procfs. Use sysctl KERN_PROC_PATHNAME and
KERN_PROC_PID to obtain the parent process pathname and command, used
to determine the calling shell.

Submitted by:	Stefan Neudorf
PR:	bin/183484
MFC after:	1 week
2013-10-31 09:29:37 +00:00
Hans Petter Selasky
35a2489852 Fix some types. Compilation for external target complains that "u_int"
is not idential to "uint32_t" when defining set channel prototype
functions. The WLAN channel range should be from 0 to 65535
inclusivly, and u_int should be fine for this purpose.
2013-10-31 09:20:30 +00:00
Konstantin Belousov
4c66d8f978 Fix long line and record proper commit message for r257427:
Add the '-k' option for getopt() call and usage.

Submitted by:	Stefan Neudorf
PR:	bin/183494
MFC after:	1 week
2013-10-31 09:03:42 +00:00
Konstantin Belousov
de111151f5 The limit for the swap space is spelled 'swapsize', at least since
tcsh 6.17.00 import.

Submitted by:	Stefan Neudorf
PR:	bin/183480
MFC after:	1 week
2013-10-31 08:58:32 +00:00
Konstantin Belousov
83bce6073a The limit for the swap space is spelled 'swapsize', at least since
tcsh 6.17.00 import.

Submitted by:	Stefan Neudorf
PR:	bin/183480
MFC after:	1 week
2013-10-31 08:22:29 +00:00
Neel Natu
a1a4cbea58 Make the virtual ioapic available unconditionally in a bhyve virtual machine.
This is in preparation for moving the ioapic device model from userspace to
vmm.ko.

Reviewed by:	grehan
2013-10-31 05:44:45 +00:00
Neel Natu
513c8d338d Rename the VMM_CTRx() family of macros to VCPU_CTRx() to highlight that these
tracepoints are vcpu-specific.

Add support for tracepoints that are global to the virtual machine - these
tracepoints are called VM_CTRx().
2013-10-31 05:20:11 +00:00
Gleb Smirnoff
a9b3c1bf05 Provide a crutch that prevents watchdog to interrupt dumping
on a box with IPMI enabled.

Okay from:	jhb
Sponsored by:	Netflix
Sponsored by:	Nginx, Inc.
2013-10-31 05:13:53 +00:00
Mark Johnston
be09cd5706 Fix some lingering build failures caused by fixing implicit inclusion of
if_var.h. Also explicitly include lock.h and mutex.h in if_kr.c rather than
depending on if_var.h to bring them in.
2013-10-31 05:00:50 +00:00
Ian Lepore
425a5c7900 Do not EOI an interrupt until the point after the filter handlers / before
threaded handlers.

It's not easy to see from the diffs of this change exactly how it
accomplishes the above.  The arm_mask_irq() and arm_unmask_irq() functions
are, respectively, the pre_thread and post_thread hooks.  Not seen in
these diffs, the arm_post_filter() routine also EOIs.  The post_filter
routine runs after filter handlers if there will be no threaded handlers,
so it just EOIs.  The pre_thread routine masks the interrupt (at the
controller, not the source) and EOIs.  So one way or another, the EOI
happens at the point where filter handlers are done.
2013-10-31 03:23:25 +00:00
Ian Lepore
adcea15135 Don't iterate through the bits of the pending interrupt register if the
whole register is zero.  Most of the registers will be zero most of the time.
2013-10-31 03:12:48 +00:00
Mark Johnston
57170f49f2 Remove references to an unused fasttrap probe hook, and remove the
corresponding x86 trap type. Userland DTrace probes are currently handled
by the other fasttrap hooks (dtrace_pid_probe_ptr and
dtrace_return_probe_ptr).

Discussed with:	rpaulo
2013-10-31 02:35:00 +00:00
Mark Johnston
9f6f3311f4 Fix a typo introduced in r257338. 2013-10-31 02:27:16 +00:00
Adrian Chadd
cea884bf7d Don't base the rate table selection based on the channel mode;
it needs to check whether there are rate entries in there or not.

PR:		kern/183428
2013-10-31 02:21:48 +00:00
Rui Paulo
74678000dc Enable USB. 2013-10-31 02:14:28 +00:00
Ian Lepore
54fe60971c Reset the timer interrupt status register at the top rather than bottom of
the interrupt handler.  If the event callback starts a new short timeout,
the timer can fire before returning from the event callback, and clearing
the interrupt status after that loses the interrupt and hangs until the
counter wraps.  Fixing all of this removes the need for the do-nothing
loop at the top of the handler which really just waited for the counter to
roll over and reach the one-shot count again.

Also add a missing return(0) in the periodic timer start case.
2013-10-31 02:11:35 +00:00
Adrian Chadd
7f0660a04b Don't treat the node as an 11n node if 11n rates are disabled.
For now, the AMRR code only knows about _either_ MCS or non-MCS rates.
It doesn't know how to downgrade (ie, doing 11b CCK rates if MCS0 isn't
reliable.)

PR:		kern/183428
2013-10-31 02:04:53 +00:00
Kevin Lo
30dfa97944 Add RT2860_BCN_OFFSET1 and RT2860_MAX_LEN_CFG register initialization to
match with the vendor driver.  While here, remove unused RT2860_DEF_MAC
definition.
2013-10-31 02:03:30 +00:00
Kevin Lo
1365315d41 Initialize BBP68 to improve rx sensitivity. 2013-10-31 02:02:14 +00:00
Kevin Lo
c513ecfd12 Enable DC filter in RT3071 Version E. 2013-10-31 02:00:58 +00:00
Kevin Lo
5ed33b90ea Add manpage for urtwnfw, the Realtek RTL8188CU/RTL8192CU firmware
module.  Also fix a few nits in urtwn.4.

Reviewed by:	rpaulo
2013-10-31 01:57:05 +00:00
Ian Lepore
eb756ebd94 Expand the list of compatible devices this driver works with. Increase
the target frequency from 1 to 10 MHz because these SoCs are plenty fast
enough to benefit from the extra event timer resolution.
2013-10-31 01:45:55 +00:00
Sean Bruno
707dd2b1d7 Queisce warning about undeclared function usage.
yp_get_default_domain is defined in workaround.c but is not declared
in any header file.  Tie the declaration to the same #define conditional
used when the function is called, NETGROUP
2013-10-30 22:45:35 +00:00
Sean Bruno
5375e4f833 Quiesce warning, which could be a bug IMO, by correctly defining the host_info
structure name
2013-10-30 22:43:40 +00:00