Adrian Chadd [Sun, 12 Aug 2012 00:46:15 +0000 (00:46 +0000)]
Revert the ath_tx_draintxq() method, and instead teach it the minimum
necessary to "do" EDMA.
It was just using the TX completion status for logging information about
the descriptor completion. Since with EDMA we don't know this without
checking the TX completion FIFO, we can't provide this information.
So don't.
Adrian Chadd [Sun, 12 Aug 2012 00:37:29 +0000 (00:37 +0000)]
Break out ath_draintxq() into a method and un-methodize ath_tx_processq().
Now that I understand what's going on with this, I've realised that
it's going to be quite difficult to implement a processq method in
the EDMA case. Because there's a separate TX status FIFO, I can't
just run processq() on each EDMA TXQ to see what's finished.
i have to actually run the TX status queue and handle individual
TXQs.
So:
* unmethodize ath_tx_processq();
* leave ath_tx_draintxq() as a method, as it only uses the completion status
for debugging rather than actively completing the frames (ie, all frames
here are failed);
* Methodize ath_draintxq().
The EDMA ath_draintxq() will have to take care of running the TX
completion FIFO before (potentially) freeing frames in the queue.
The only two places where ath_tx_draintxq() (on a single TXQ) are used:
* ath_draintxq(); and
* the CABQ handling in the beacon setup code - it drains the CABQ before
populating the CABQ with frames for a new beacon (when doing multi-VAP
operation.)
So it's quite possible that once I methodize the CABQ and beacon handling,
I can just drop ath_tx_draintxq() in its entirety.
Finally, it's also quite possible that I can remove ath_tx_draintxq()
in the future and just "teach" it to not check the status when doing
EDMA.
Adrian Chadd [Sat, 11 Aug 2012 23:26:19 +0000 (23:26 +0000)]
Extend the beacon code slightly to support AP mode beaconing for the
EDMA HAL hardware.
* The EDMA HAL code assumes the nexttbtt and intval values are in TU/8
units, rather than TU. For now, just "hack" around that here, at least
until I code up something to translate it in the HAL.
* Setup some different TXQ flags for EDMA hardware.
* The EDMA HAL doesn't support setting the first rate series via
ath_hal_setuptxdesc() - instead, a call to ath_hal_set11nratescenario()
is always required. So for now, just do an 11n rate series setup
for EDMA beacon frames.
This allows my AR9380 to successfully transmit beacon frames.
However, CABQ TX and all normal data frame TX and TX completion is
still not functional and will require some more significant code churn
to make work.
Adrian Chadd [Sat, 11 Aug 2012 22:25:28 +0000 (22:25 +0000)]
Add the AR9300 HAL ID in to the 11n check routine.
I was having TX hang issues, which I root caused to having the
legacy ath_hal_setupxtxdesc() called, rather than the 11n rate scenario
setup code. This meant that rate control information wasn't being
put into frames, causing the MAC to stall/hang.
Adrian Chadd [Sat, 11 Aug 2012 22:20:28 +0000 (22:20 +0000)]
Begin fleshing out the TX FIFO support.
* Add ATH_TXQ_FIRST() for easy tasting of what's on the list;
* Add an "axq_fifo_depth" for easy tracking of how deep the current
FIFO is;
* Flesh out the handoff (mcast, hw) functions;
* Begin fleshing out a TX ISR proc, which tastes the TX status FIFO.
The legacy hardware stuffs the TX completion at the end of the final frame
descriptor (or final sub-frame when doing aggregate.) So it's feasible
to do a per-TXQ drain and process, as the needed info is right there.
For EDMA hardware, there's a separate TX completion FIFO. So the TX
process routine needs to read the single FIFO and then process the
frames in each hardware queue.
This makes it difficult to do a per-queue process, as you'll end up with
frames in the TX completion FIFO for a different TXQ to the one you've
passed to ath_tx_draintxq() or ath_tx_processq().
Testing:
I've tested the TX queue and TX completion code in hostap mode on an
AR9380. Beacon frames successfully transmit and the completion routine
is called. Occasional data frames end up in TXQ 1 and are also
successfully completed.
However, this requires some changes to the beacon code path as:
* The AR9380 beacon configuration API is now in TU/8, rather than
TU;
* The AR9380 TX API requires the rate control is setup using a call
to setup11nratescenario, rather than having the try0 series setup
(rate/tries for the first series); so the beacon won't go out.
I'll follow this up with commits to the beacon code.
Alexander Motin [Sat, 11 Aug 2012 15:08:19 +0000 (15:08 +0000)]
Allow idle threads to steal second threads from other cores on systems with
8 or more cores to improve utilization. None of my tests on 2xXeon (2x6x2)
system shown any slowdown from mentioned "excess thrashing". Same time in
pbzip2 test with number of threads more then number of CPUs I see up to 10%
speedup with SMT disabled and up 5% with SMT enabled. Thinking about
trashing I was trying to limit that stealing within same last level cache,
but got only worse results. Present code any way prefers to steal threads
from topologically closer cores.
Ed Schouten [Sat, 11 Aug 2012 12:07:24 +0000 (12:07 +0000)]
Rename aux.c to auxv.c.
On Windows, AUX is the auxiliary device, usually pointing to COM1.
Therefore it is forbidden to create a file named aux.c. To make it a bit
easier for Windows users to check out our source code, rename this file
to auxv.c.
MFC after: 1 month
Discussed with: kib
Suggested by: Eric van Gyzen <eric vangyzen net>
Dimitry Andric [Sat, 11 Aug 2012 11:13:48 +0000 (11:13 +0000)]
Change a few extern inline functions in libm to static inline, since
they need to refer to static constants, which C99 does not allow for
extern inline functions.
While here, change a comment in e_rem_pio2f.c to mention the correct
number of bits.
Andrew Turner [Sat, 11 Aug 2012 05:58:56 +0000 (05:58 +0000)]
Move the decoding of the swi instruction to the syscall function. With the
ARM EABI the syscall value will be moved to a register to ease adding thumb
support. When this happens decoding of the instruction will no longer be
required.
Alexander Motin [Fri, 10 Aug 2012 19:02:49 +0000 (19:02 +0000)]
Some minor tunings/cleanups inspired by bde@ after previous commits:
- remove extra dynamic variable initializations;
- restore (4BSD) and implement (ULE) hogticks variable setting;
- make sched_rr_interval() more tolerant to options;
- restore (4BSD) and implement (ULE) kern.sched.quantum sysctl, a more
user-friendly wrapper for sched_slice;
- tune some sysctl descriptions;
- make some style fixes.
Always initialize sc_ekey, because as of r238116 it is always used.
If GELI provider was created on FreeBSD HEAD r238116 or later (but before this
change), it is using very weak keys and the data is not protected.
The bug was introduced on 4th July 2012.
One can verify if its provider was created with weak keys by running:
# geli dump <provider> | grep version
If the version is 7 and the system didn't include this fix when provider was
initialized, then the data has to be backed up, underlying provider overwritten
with random data, system upgraded and provider recreated.
Reported by: Fabian Keil <fk@fabiankeil.de>
Tested by: Fabian Keil <fk@fabiankeil.de>
Discussed with: so
MFC after: 3 days
Take advantage of new UCOM and bus functionality so that
the device_detach() function doesn't block on UCOM device
drivers until the TTY handle is closed by the userspace
application. This is implemented by a postpone of the
softc free where the UCOM structures reside until the
TTY references are gone.
Extend the callback table of UCOM to include a
"ucom_free" function pointer which is called when
all refs on a UCOM super structure is gone.
Implement various helper functions to handle
refcounting and draining on the UCOM super
structure.
Implement macro which can be used in device
drivers to avoid module unload before all
pending TTY references are gone.
The UCOM API is backwards compatible after this
change and device drivers require no changes
to function with this change. Only a recompilation
of UCOM device drivers is required. The FreeBSD
version has been bumped in that regard.
Add new device method to free the automatically
allocated softc structure which is returned by
device_get_softc(). This method can be used to
easily implement softc refcounting. This can be
desirable when the softc has memory references
which are controlled by userspace handles for
example.
This solves the problem of blocking the caller
of device_detach() for a non-deterministic time.
Alan Cox [Fri, 10 Aug 2012 05:47:04 +0000 (05:47 +0000)]
Eliminate an unnecessary acquisition and release of the page queues lock
from pmap_pte(). PT_SET_MA() is not a queued mapping update, but instead
an immediate mapping update, so the page queues lock is not required here.
Alan Cox [Fri, 10 Aug 2012 05:00:50 +0000 (05:00 +0000)]
Merge r134393 from amd64/i386:
The machine-independent parts of the virtual memory system always pass a
valid pmap to the pmap functions that require one. Remove the checks for
NULL. (These checks have their origins in the Mach pmap.c that was
integrated into BSD. None of the new code written specifically for
FreeBSD included them.)
Xin LI [Fri, 10 Aug 2012 04:48:58 +0000 (04:48 +0000)]
RFC 2289 requires all hashes be stored in little endian format before
folding to 64 bits, while SHA1 code is big endian. Therefore, a bswap32
is required before using the value.
Without this change, the implementation does not conform to test vector
found in RFC 2289.
Alexander Motin [Thu, 9 Aug 2012 19:26:13 +0000 (19:26 +0000)]
Rework r220198 change (by fabient). I believe it solves the problem from
the wrong direction. Before it, if preemption and end of time slice happen
same time, thread was put to the head of the queue as for only preemption.
It could cause single thread to run for indefinitely long time. r220198
handles it by not clearing TDF_NEEDRESCHED in case of preemption. But that
causes delayed context switch every time preemption happens, even when not
needed.
Solve problem by introducing scheduler-specifoc thread flag TDF_SLICEEND,
set when thread's time slice is over and it should be put to the tail of
queue. Using SW_PREEMPT flag for that purpose as it was before just not
enough informative to work correctly.
On my tests this by 2-3 times reduces run time deviation (improves fairness)
in cases when several threads share one CPU.
Alexander Motin [Thu, 9 Aug 2012 18:09:59 +0000 (18:09 +0000)]
SCHED_4BSD scheduling quantum mechanism appears to be broken for some time.
With switchticks variable being reset each time thread preempted (that is
done regularly by interrupt threads) scheduling quantum may never expire.
It was not noticed in time because several other factors still regularly
trigger context switches.
Handle the problem by replacing that mechanism with its equivalent from
SCHED_ULE called time slice. It is effectively the same, just measured in
context of stathz instead of hz. Some unification is probably not bad.
Alan Cox [Thu, 9 Aug 2012 16:38:17 +0000 (16:38 +0000)]
Merge r132141 and r111272 from amd64/i386:
Reduce the size of a PV entry by eliminating pv_ptem. There is no need
to store a pointer to the page table page in the PV entry because it is
easily computed during the walk down the page table.
Eliminate the ptphint from the pmap. Long, long ago, page table pages
belonged to a vm object, and we would look up page table pages based
upon their offset within this vm object. In those days, this hint may
have had tangible benefits.
Jilles Tjoelker [Thu, 9 Aug 2012 15:11:38 +0000 (15:11 +0000)]
ftw(): Do not check the maxfds argument against OPEN_MAX.
Apart from the fact that nothing should have OPEN_MAX as a limit (as opposed
to RLIMIT_NOFILE from getrlimit() or _SC_OPEN_MAX from sysconf()), POSIX
does not require us to check this. POSIX does have a requirement on the
application that maxfds not exceed {OPEN_MAX}, but does not require the
implementation to check it ("may fail").
Jilles Tjoelker [Thu, 9 Aug 2012 15:04:06 +0000 (15:04 +0000)]
nftw(): Do not check the maxfds argument against OPEN_MAX.
Apart from the fact that nothing should have OPEN_MAX as a limit (as opposed
to RLIMIT_NOFILE from getrlimit() or _SC_OPEN_MAX from sysconf()), POSIX
does not require us to check this.
Ed Maste [Thu, 9 Aug 2012 14:46:52 +0000 (14:46 +0000)]
Improve lock and unlock symmetry
- Move destruction of per-ring locks to netmap_dtor_locked to mirror the
initialization that happens in NIOCREGIF. Otherwise unloading a netmap-
capable interface that was never put into netmap mode would try to
mtx_destroy an uninitialized mutex, and panic.
- Destroy core_lock in netmap_detach, mirroring init in netmap_attach.
- Also comment out the knlist_destroy for now as there is currently no
knlist_init.
Matt Jacob [Wed, 8 Aug 2012 18:24:33 +0000 (18:24 +0000)]
More rototilling with target mode in an attemp to get multiple
CCB at a time outstanding reliable. It's not there yet, but this
is the direction to go in so might as well commit. So far,
multiple at a time CCBs work (see ISP_INTERNAL_TARGET test mode),
but it fails if there are more downstream than the SIM wants
to handle and SRR is sort of confused when this happens, plus
it is not entirely quite clear what one does if a CCB/CTIO fails
and you have more in flight (that don't fail, say) and more queued
up at the SIM level that haven't been started yet.
Some of this is driven because there apparently is no flow control
to requeue XPT_CONTINUE_IO requests like there are for XPT_SCSI_IO
requests. It is also more driven in that the few target mode
periph drivers there are are not really set up for handling pushback-
heck most of them don't even check for errors (and what would they
really do with them anyway? It's the initiator's problem, really....).
The data transfer arithmetic has been worked over again to handle
multiple outstanding commands, so you have a notion of what's been
moved already as well as what's currently in flight. It turns that
this led to uncovering a REPORT_LUNS bug in the ISP_INTERNAL_TARGET
code which was sending back 24 bytes of rpl data instead of the
specified 16. What happened furthermore here is that sending back
16 bytes and reporting an overrun of 8 bytes made the initiator
(running FC-Tape aware f/w) mad enough to request, and keep
requesting, another FCP response (I guess it didn't like the answer
so kept asking for it again).
Andrew Turner [Wed, 8 Aug 2012 09:42:44 +0000 (09:42 +0000)]
Import compiler-rt r160957.
This is mostly a no-op other than for ARM where it adds missing
__aeabi_mem* and __aeabi_*divmod functions. Even on ARM these will remain
unused until the rest of the ARM EABI code is merged.
Alan Cox [Wed, 8 Aug 2012 05:28:30 +0000 (05:28 +0000)]
The assertion that I added in r238889 could legitimately fail when a
debugger creates a breakpoint. Replace that assertion with a narrower
one that still achieves my objective.
Jim Harris [Tue, 7 Aug 2012 23:16:11 +0000 (23:16 +0000)]
During TSC synchronization test, use rdtsc() rather than rdtsc32(), to
protect against 32-bit TSC overflow while the sync test is running.
On dual-socket Xeon E5-2600 (SNB) systems with up to 32 threads, there
is non-trivial chance (2-3%) that TSC synchronization test fails due to
32-bit TSC overflow while the synchronization test is running.
John Baldwin [Tue, 7 Aug 2012 12:40:31 +0000 (12:40 +0000)]
Don't try to stop the IPMI watchdog timer if it is not running.
Starting or stopping the IPMI watchdog is rather expensive with the
current implementation as all IPMI requests are bounced via thread.
This is not viable during shutdown or dumps, and this avoids headache
in the common case that the watchdog is not enabled. The IPMI watchdog
should probably be reworked to not use a separate thread to fix this
in the case when the watchdog timer is enabled.
As it turned out, there are some installations, where BSD label
contains partitions with type zero. And it has worked.
So, allow detect these partitions.
Alan Cox [Tue, 7 Aug 2012 04:48:14 +0000 (04:48 +0000)]
Never sleep on busy pages in vm_pageout_launder(), always skip them. Long
ago, sleeping on busy pages in vm_pageout_launder() made sense. The call
to vm_pageout_flush() specified asynchronous I/O and sleeping on busy pages
blocked vm_pageout_launder() until the flush had completed. However, in
CVS revision 1.35 of vm/vm_contig.c, the call to vm_pageout_flush() was
changed to request synchronous I/O, but the sleep on busy pages was not
removed.
Brooks Davis [Mon, 6 Aug 2012 21:24:43 +0000 (21:24 +0000)]
MFP4 change 214967:
Make the process of embedding MDROOT images less perilous by
makeing the target that links kernel and embedding the image
depend on the image. This means, if the image doesn't exist you
find out before you try to boot from it and that if you change
the image you don't have to touch some random source file to
cause a rebuild.
Dimitry Andric [Mon, 6 Aug 2012 21:02:40 +0000 (21:02 +0000)]
Make sure the compiler knows g_gate_xvlog() and g_gate_xlog() do not
return. This silences a warning from clang 3.2 about uninitialized use
of the variable 'mediasize' in sbin/ggate/shared/ggate.c.
Jack F Vogel [Mon, 6 Aug 2012 20:44:05 +0000 (20:44 +0000)]
Correct the mq_start routine to avoid out-of-order
packet delivery, always enqueue when possible. Also
correct the DEPLETED test as multiple bits might be
set. Thanks to Randall Stewart for the changes!
John Baldwin [Mon, 6 Aug 2012 19:49:57 +0000 (19:49 +0000)]
Explicitly enable busmastering on PCI-PCI bridges. Transactions initiated
on the secondary side of a bridge will not be propagated to the primary
bus unless this is enabled. Busmastering is not enabled by default (we
have relied on firmware to set this bit to date). The OS needs to set it
for any bridges not configured by system firmware.
Tested by: Steve Polyack korvus comcast net
MFC after: 2 weeks
Dimitry Andric [Mon, 6 Aug 2012 18:54:17 +0000 (18:54 +0000)]
In sys/dev/cxgbe/firmware/t4fw_interface.h, change the enum
'fw_hdr_intfver' into an anonymous enum, which avoids a clang 3.2
warning about all the enum values being the same value.
Do not add handler to event handlers list until ithread is created.
In rare event when fast and ithread interrupts share the same vector
and the fast handler was registered first, we can end up trying to
schedule the ithread that is not created yet. The kernel built with
INVARIANTS then triggers an assertion.
Change the order to create the ithread first and only then add the
handler that needs it to the interrupt event handlers list.
Luigi Rizzo [Mon, 6 Aug 2012 11:02:23 +0000 (11:02 +0000)]
move functions outside the SYSBEGIN/SYSEND block
(SYSBEGIN/SYSEND are specific to ipfw/dummynet and are used to
emulate sysctl on platforms that do not have them, and they work
by creating an array which contains all the sysctl-ed symbols.)
Luigi Rizzo [Mon, 6 Aug 2012 10:50:43 +0000 (10:50 +0000)]
use FREE_PKT instead of m_freem to free an mbuf.
The former is the standard form used in ipfw/dummynet, so that
it is easier to remap it to different memory managers depending
on the platform.
Marius Strobl [Mon, 6 Aug 2012 08:58:54 +0000 (08:58 +0000)]
- Merge from NetBSD:
When issuing a non-DMA command, make sure to set the "remaining length of
command to be transferred via DMA" (sc_cmdlen) to zero up-front, otherwise
we might get confused on command competition interrupt (no DMA active but
still data left to transfer).
- Implement handling of MSG_IGN_WIDE_RESIDUE which some targets produce, as
just rejecting these leads to a resend and disconnect loop.
Reported and tested by: mjacob
Mikolaj Golub [Sun, 5 Aug 2012 17:30:17 +0000 (17:30 +0000)]
In tcp timers, check INP_DROPPED flag a little later, after
callout_deactivate(), so if INP_DROPPED is set we return with the
timer active flag cleared.
For me this fixes negative keep timer values reported by `netstat -x'
for connections in CLOSE state.
Dimitry Andric [Sun, 5 Aug 2012 17:23:47 +0000 (17:23 +0000)]
In usr.sbin/ctladm/ctladm.c, function cctl_error_inject(), initialize
the 'retval' variable to zero, to avoid returning garbage in several
cases.
This fixes the following clang 3.2 warnings:
usr.sbin/ctladm/ctladm.c:1234:6: error: variable 'retval' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized]
if (ioctl(fd, CTL_ERROR_INJECT, &err_desc) == -1) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
usr.sbin/ctladm/ctladm.c:1243:10: note: uninitialized use occurs here
return (retval);
^~~~~~
usr.sbin/ctladm/ctladm.c:1234:2: note: remove the 'if' if its condition is always true
if (ioctl(fd, CTL_ERROR_INJECT, &err_desc) == -1) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
usr.sbin/ctladm/ctladm.c:1161:7: error: variable 'retval' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized]
if (ioctl(fd, CTL_ERROR_INJECT_DELETE, &err_desc) == -1) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
usr.sbin/ctladm/ctladm.c:1243:10: note: uninitialized use occurs here
return (retval);
^~~~~~
usr.sbin/ctladm/ctladm.c:1161:3: note: remove the 'if' if its condition is always true
if (ioctl(fd, CTL_ERROR_INJECT_DELETE, &err_desc) == -1) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
usr.sbin/ctladm/ctladm.c:1029:12: note: initialize the variable 'retval' to silence this warning
int retval;
^
= 0
Dimitry Andric [Sun, 5 Aug 2012 15:55:36 +0000 (15:55 +0000)]
In usr.bin/make/var.c, function ParseModifier(), initialize the 'error'
variable to NULL, to avoid using it uninitialized in certain cases.
This fixes the following clang 3.2 warning:
usr.bin/make/var.c:1770:10: error: variable 'error' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized]
if (vp->execute) {
^~~~~~~~~~~
usr.bin/make/var.c:1777:10: note: uninitialized use occurs here
if (error)
^~~~~
usr.bin/make/var.c:1770:6: note: remove the 'if' if its condition is always true
if (vp->execute) {
^~~~~~~~~~~~~~~~~
usr.bin/make/var.c:1768:23: note: initialize the variable 'error' to silence this warning
const char *error;
^
= NULL
Add simple test program that uses the partition tables handling code.
It is useful to test and debug how boot loader handles partition tables
metadata.