copyright.
When all member nations of the Buenos Aires Convention adopted the Berne
Convention, the phrase "All rights reserved" became unnecessary to assert
copyright. Remove it from files under my or Panasas's copyright. The files
related to jedec_dimm(4) also bear avg@'s copyright; he has approved this
change.
Approved by: avg
Sponsored by: Panasas
When using poll(), select() or kevent() on netmap file descriptors,
netmap executes the equivalent of NIOCTXSYNC and NIOCRXSYNC commands,
before collecting the events that are ready. In other words, the
poll/kevent callback has side effects. This is done to avoid the
overhead of two system call per iteration (e.g., poll() + ioctl(NIOC*XSYNC)).
When the kqueue subsystem invokes the kqueue(9) f_event callback
(netmap_knrw), it holds the lock of the struct knlist object associated
to the netmap port (the lock is provided at initialization, by calling
knlist_init_mtx).
However, netmap_knrw() may need to wake up another netmap port (or even
the same one), which means that it may need to call knote().
Since knote() needs the lock of the struct knlist object associated to
the to-be-wake-up netmap port, it is possible to have a lock order reversal
problem (AB/BA deadlock).
This change prevents the deadlock by executing the knote() call in a
per-selinfo taskqueue, where it is possible to hold a mutex.
Reviewed by: aleksandr.fedorov_itglobal.com
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D18956
bus_teardown_intr(9) before pci_release_msi(9).
- Ensure that iflib(4) and associated drivers pass correct RIDs to
bus_release_resource(9) by obtaining the RIDs via rman_get_rid(9)
on the corresponding resources instead of using the RIDs initially
passed to bus_alloc_resource_any(9) as the latter function may
change those RIDs. Solely em(4) for the ioport resource (but not
others) and bnxt(4) were using the correct RIDs by caching the ones
returned by bus_alloc_resource_any(9).
- Change the logic of iflib_msix_init() around to only map the MSI-X
BAR if MSI-X is actually supported, i. e. pci_msix_count(9) returns
> 0. Otherwise the "Unable to map MSIX table " message triggers for
devices that simply don't support MSI-X and the user may think that
something is wrong while in fact everything works as expected.
- Put some (mostly redundant) debug messages emitted by iflib(4)
and em(4) during attachment under bootverbose. The non-verbose
output of em(4) seen during attachment now is close to the one
prior to the conversion to iflib(4).
- Replace various variants of spelling "MSI-X" (several in messages)
with "MSI-X" as used in the PCI specifications.
- Remove some trailing whitespace from messages emitted by iflib(4)
and change them to consistently start with uppercase.
- Remove some obsolete comments about releasing interrupts from
drivers and correct a few others.
Reviewed by: erj, Jacob Keller, shurd
Differential Revision: https://reviews.freebsd.org/D18980
Effectively all i386 kernels now have two pmaps compiled in: one
managing PAE pagetables, and another non-PAE. The implementation is
selected at cold time depending on the CPU features. The vm_paddr_t is
always 64bit now. As result, nx bit can be used on all capable CPUs.
Option PAE only affects the bus_addr_t: it is still 32bit for non-PAE
configs, for drivers compatibility. Kernel layout, esp. max kernel
address, low memory PDEs and max user address (same as trampoline
start) are now same for PAE and for non-PAE regardless of the type of
page tables used.
Non-PAE kernel (when using PAE pagetables) can handle physical memory
up to 24G now, larger memory requires re-tuning the KVA consumers and
instead the code caps the maximum at 24G. Unfortunately, a lot of
drivers do not use busdma(9) properly so by default even 4G barrier is
not easy. There are two tunables added: hw.above4g_allow and
hw.above24g_allow, the first one is kept enabled for now to evaluate
the status on HEAD, second is only for dev use.
i386 now creates three freelists if there is any memory above 4G, to
allow proper bounce pages allocation. Also, VM_KMEM_SIZE_SCALE changed
from 3 to 1.
The PAE_TABLES kernel config option is retired.
In collaboarion with: pho
Discussed with: emaste
Reviewed by: markj
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D18894
This fixes BIO_ORDERED semantics while also improving performance by:
- sleeping also before BIO_ORDERED bio, as defined, not only after;
- not queueing BIO_ORDERED bio to taskqueue if no other bios running;
- waking up sleeping taskqueue explicitly rather then rely on polling.
On Samsung SSD 970 PRO this shows sync write latency, measured with
`diskinfo -wS`, reduction from ~2ms to ~1.1ms by not sleeping without
reason till next HZ tick.
On the same device ZFS pool with 8 ZVOLs synchronously writing 4KB blocks
shows ~950 IOPS instead of ~750 IOPS before. I suspect ZFS does not need
BIO_ORDERED on BIO_FLUSH at all, but that will be next question.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
Because of a typo, the code was mistakenly resetting the
vtnrx_vq pointer rather than vtntx_tq.
Reviewed by: bryanv
MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D19015
On sync-kloop stop, send a wake-up signal to the kloop, so that
waiting for the timeout is not needed.
Also, improve logging in netmap_freebsd.c.
MFC after: 3 days
Enforce net80211 rates for control / management / multicast / EAPOL frames
and allow to override rate for unicast frames via ifconfig(8) 'ucastrate'
option; by default it still uses f/w rate adaptation for unicast frames.
MFC after: 1 week
* There's no reason to have a while() loop here, because:
- if msleep returns 0, that means we were woken up by the interrupt handler,
and we are going to exit immediately as sc_fw_chunk_done will now be 1
(there is nothing else that sleeps on sc_fw.)
- if msleep doesn't return 0 (i.e. it returned ETIMEDOUT) then we will
exit immediately because of the if-test.
So, just use a single msleep() and then check sc_fw_chunk_done as before.
* The comment said we were sleeping for 5 seconds, but the msleep was only
for 1. Before r314065, this was 1 second and so was the comment,
and in that commit the comment was changed and the function call wasn't.
Possibly fixes failures to initialize uCode on certain devices.
Submitted by: Augustin Cavalier (waddlesplash gmail.com)
Obtained from: Haiku 132990ecdcb072f2ce597b5d497ff3e5b1f09c20
MFC after: 10 days
Wrap ieee80211_add_channel_list_2ghz into another function
which supplies default (1-14) channel list to it and drop
its copies from drivers.
Checked with RTL8188EE, country US / JP / KR / UA.
MFC after: 2 weeks
- Drain offload transmit queues when RATELIMIT is enabled but
TCP_OFFLOAD is not.
- Expose the per-VI nofldtxq and first_ofld_txq sysctls when
RATELIMIT is enabled but TCP_OFFLOAD is not.
- Clear offload transmit queue stats as part of a 'cxgbetool clearstats'
request when RATELIMIT is enabled but TCP_OFFLOAD is not.
Reviewed by: np
MFC after: 2 weeks
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D18966
Replace in-place implementation with system-wide one; since it
guarantees non-zero result drop all less-than-one checks from
drivers and net80211.
MFC after: 2 weeks
* This hopefully avoids some firmware panics, I was occasionally seeing,
when iwm disconnects upon losing signal to an access point at some point.
* This is synchronizing the if_iwm_time_event.c file a bit more from the
corresponding Linux iwlwifi/mvm/time-event.c.
Taken-From: Linux iwlwifi
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (e8cb71584a6a72232c13151d60e57f7f229220eb)
* This is a mix of the OpenBSD Git 7fd9664469d1b717a307eebd74aeececbd3c41cc
change, and syncing with the Linux iwlwifi code.
Taken-From: Linux iwlwifi, and OpenBSD
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (706a3044afd27c3fecfdf57bec1695310e53e228)
* This avoids firmware resets in all the cases in iwm_newstate(). Instead
iwm_bring_down_firmware() is called, which tears down all the STA
connection state, according to the sc->sc_firmware_state value.
* Improve the behaviour of the LED blinking a bit, so it only blinks when
there really is a wireless scan going on.
* Print the newstate arg in debug output of iwm_newstate(), to help in
debugging.
This is inspired by the firmware state maintaining change in OpenBSD's iwm,
by stsp@openbsd.org (OpenBSD Git 0ddb056fb7370664b1d4b84392697cb17d1a414a).
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (8a41b10ac639d0609878696808387a6799d39b57)
* While there remove unused IWM_UCODE_TLV_CAPA_LMAC_UPLOAD definition,
which isn't defined in iwlwifi.
Taken-From: Linux iwlwifi
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (fd4f9de8bc72ea961e50829b45b59d0549040b7d)
* Remove outdated notifications IWM_SCAN_ABORT_CMD,
IWM_SCAN_START_NOTIFICATION and IWM_SCAN_RESULTS_NOTIFICATION.
* Remove unused enum iwm_scan_complete_status.
* Use the updated FW Api version 3 of struct iwm_scan_results_notif.
* No functional change, since struct iwm_scan_results_notif is never
accessed in iwm at the moment.
Taken-From: Linux iwlwifi commits 1083fd7391e989be52022f0f338e9dadc048b063
and 75118fdb63496e4611ab50380499ddd62b9de69f.
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (c947b0b8dc96dabefd63f7b70d53695e36c7b64f)
* Rename some structs and struct members for firmware handling.
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (4b1006a6e4d0f61d48c67b46e1f791e30837db67)
* There is (almost) nothing to do in suspend/resume if if_iwm has failed
during initialization (e.g. because of firmware load failure) and was
already uninitialized by iwm_detach_local().
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (67b5e090efb225654815fed91020db6cfc16bb19)
* We should load the firmware exactly once before the driver really
initializes the hardware the first time, and unload it at detach time.
There is no need to retrieve the firmware during execution of
iwm_mvm_load_ucode_wait_alive(), we should make sure we already have the
firmware data at hand before that.
* The existing sc_preinit_hook code fails to deal with the case where
if_iwm is loaded by the loader (or is statically linked) and the
firmware needs to be loaded from disk. So we can just call
iwm_read_firmware() from iwm_attach() directly.
* A separate solution will have to be added to properly defer the firmware
loading during bootup, until the necessary filesystem is mounted.
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (0104ee1f4cb6a2313c00c2526c6ae98d42e5041d)
* Doing the iwm_prepare_card_hw() call in iwm_attach() only on Family 8000
hardware matches the code in Linux iwlwifi.
* While there remove DEFAULT_MAX_TX_POWER definition which is unused, and
has a value different from IWL_DEFAULT_MAX_TX_POWER in iwlwifi.
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (e8560f8dc58df12a7c79a6bb4e6ccb156e001085)
* Rather than providing a non-zero index into the firmware RS table,
we should always use index 0 and update the firmware RS table whenever
our chosen tx rate for data-frames changes.
* Send IWM_LQ_CMD updates when the tx rate gets updated by the net80211
rate control (which is after we tell the tx status to the net80211
rate-control in iwm_mvm_rx_tx_cmd_single()).
* Disregard frames transferred with a different tx rate than the currently
selected rate for the rate-control calculations. This way we avoid
counting management frames (which are sent at a slow, and fixed rate),
as well as frames we added to the tx queue just before a new IWM_LQ_CMD
update took effect.
Submitted by: Augustin Cavalier <waddlesplash@gmail.com> (Haiku)
Obtained from: DragonFlyBSD (5d6b465e288ac5b52d7115688d4e6516acbbea1c)
From Krzysztof:
Ensure that the entire data buffer passed from the NVM update tool is copied in
to kernel space and copied back out to user space using copyin() and copyout().
PR: 234104
Submitted by: Krzysztof Galazka <krzysztof.galazka@intel.com>
Reported by: Finn <ixbug@riseup.net>
MFC after: 5 days
Sponsored by: Intel Corporation
Differential Revision: https://reviews.freebsd.org/D18817
From Jake:
In r341156 ("Fix first-packet completion", 2018-11-28) a hack to work
around a delta calculation determining how many descriptors were used
was added to ixl_isc_tx_credits_update_dwb.
The same fix was also applied to the em and igb drivers in r340310, and
to ix in r341156.
The hack checked the case where prev and cur were equal, and then added
one. This works, because by the time we do the delta check, we already
know there is at least one packet available, so the delta should be at
least one.
However, it's not a complete fix, and as indicated by the comment is
really a hack to work around the real bug.
The real problem is that the first time that we transmit a packet,
tx_cidx_processed will be set to point to the start of the ring.
Ultimately, the credits_update function expects it to point to the
*last* descriptor that was processed. Since we haven't yet processed any
descriptors, pointing it to 0 results in this incorrect calculation.
Fix the initialization code to have it point to the end of the ring
instead. One way to think about this, is that we are setting the value
to be one prior to the first available descriptor.
Doing so, corrects the delta calculation in all cases. The original fix
only works if the first packet has exactly one descriptor. Otherwise, we
will report 1 less than the correct value.
As part of this fix, also update the MPASS assertions to match the real
expectations. First, ensure that prev is not equal to cur, since this
should never happen. Second, remove the assertion about prev==0 || delta
!= 0. It looks like that originated from when the em driver was
converted to iflib. It seems like it was supposed to ensure that delta
was non-zero. However, because we originally returned 0 delta for the
first calculation, the "prev == 0" was tacked on.
Instead, replace this with a check that delta is greater than zero,
after the correction necessary when the ring pointers wrap around.
This new solution should fix the same bug as r341156 did, but in a more
robust way.
Submitted by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed by: shurd@
Sponsored by: Intel Corporation
Differential Revision: https://reviews.freebsd.org/D18545
Note that the affected interface is available only to root.
admbugs: 765
Reported by: Vlad Tsyrklevich <vlad@tsyrklevich.net>
Reviewed by: emaste, ram
MFC after: 1 day
Security: Kernel memory disclosure
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D18914
