Changes to acpi_gpiobus.c handle discovering and parsing the _AEI
objects and storing necessary data in device ivars. A new gpioaei.c
file implements the device, which simply requests an interrupt when
the pin is triggered and invokes the appropriate _Exx or _Lxx ACPI
method.
This makes the GPIO "power button" work on arm64 Graviton systems,
allowing EC2 "Stop"/"Reboot" instance calls to be handled cleanly.
(Prior to this change, those requests would time out after 4 minutes
and the instance would be forcibly killed.)
Reviwed by: imp, andrew, Ahmad Khalifa
MFC after: 3 days
Sponsored by: Amazon
Differential Revision: https://reviews.freebsd.org/D47253
Co-authored-by: Andrew Turner <andrew@FreeBSD.org>
(cherry picked from commit 9709bda03c)
GPIO interrupts work just fine and will be used shortly. We still
do not support GPIO_INTR_SHAREABLE however, so leave that within
the NOT_YET scope.
Reviwed by: andrew
MFC after: 1 week
Sponsored by: Amazon
Differential Revision: https://reviews.freebsd.org/D47251
(cherry picked from commit 2d4219919a)
This allows acpi_gpiobus to override the method and fall back to the
generic gpiobus_read_ivar function if needed.
Reviewed by: andrew
MFC after: 1 week
Sponsored by: Amazon
Differential Revision: https://reviews.freebsd.org/D47250
(cherry picked from commit bc0d10d01c)
AWS Graviton [1234] systems have a bug in their ACPI where they mark
the PL061's GPIO pins as needing to be configured in PullUp mode (in
fact the PL061 has no pullup/pulldown resistors); this flag needs to
be removed in order for _AEI objects to be handled on these systems.
Reviewed by: Ali Saidi
MFC after: 1 week
Sponsored by: Amazon
Differential Revision: https://reviews.freebsd.org/D47239
(cherry picked from commit 2f3f867ac6)
ACPI sleep states are only implemented on x86 systems, so having the
ACPI power button attempt to enter "S5" (or other state as configured
via the hw.acpi.power_button_state sysctl) is not useful.
On non-x86 systems, implement the power button with a call to
shutdown_nice(RB_POWEROFF)
to shut down the system.
Reviewed by: Andrew
Tested on: Graviton 2
MFC after: 2 weeks
Sponsored by: Amazon
Differential Revision: https://reviews.freebsd.org/D47094
(cherry picked from commit f41ef9d80b)
Right now flags is set to 0 before this "=" -> "|=" change, but it will
matter when the NOT_YET section above becomes effective.
MFC after: 2 weeks
Sponsored by: Amazon
(cherry picked from commit c808132731)
This currently only implements the address space handler and attempts to
configure pins with flags obtained from ACPI.
Reviewed by: wulf
MFC after: 1 month
Pull Request: https://github.com/freebsd/freebsd-src/pull/1359
(cherry picked from commit 92adaa5862)
As explained in PR 277038, iflib calls IFDI_DETACH() and then
IFDI_QUEUES_FREE(). With igc, the latter writes to a register after it
has been unmapped.
igc_if_detach() already calls igc_release_hw_control(), and looking at
callers of igc_if_queues_free(), that appears to be sufficient. So,
just remove the igc_release_hw_control() call.
PR: 277038
Reported by: Mike Belanger <mibelanger@qnx.com>
Reviewed by: kbowling
Tested by: kbowling
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D47293
(cherry picked from commit 35d05a14ed)
This allows iavf to load on E830 devices since those devices place their MSI-X
BAR at a different location than in previous 800 series products.
Signed-off-by: Eric Joyner <erj@FreeBSD.org>
Sponsored by: Intel Corporation
Differential Revision: https://reviews.freebsd.org/D46952
(cherry picked from commit e53a21abdf)
ifmedia_add() allocates an ifmedia_entry during ena_attach.
Current code doesn't release this memory during ena_detach()
This commit calls ifmedia_removeall() to properly free the
allocated memory during ena_detach().
Also, in case ena_attach fails, we need to detach ifmedia
which was allocated within ena_setup_ifnet().
This bug was first described in:
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=278100
Reviewed by: zlei
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 449496eb28)
Large LLQ depth size is currently calculated by dividing the maximum
possible size of LLQ by 2.
In newer paltforms, starting from r8g the size of BAR2,
which contains LLQ, will be increased, and the maximum depth of
wide LLQ will be set according to a value set by the device, instead of
hardcoded division by 2.
The new value will be stored by the device in max_wide_llq_depth field
for drivers that expose ENA_ADMIN_LLQ_FEATURE_VERSION_1 or higher to
the device.
There is an assumption that max_llq_depth >= max_wide_llq_depth, since
they both use the same bar, and if it is possible to have a wide LLQ
of size max_wide_llq_depth, it is possible to have a normal LLQ of the
same size, since it will occupy half of the space.
Also moved the large LLQ case calculation of max_tx_queue_size
before its rounddown.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit d0419551d9)
This commit adds support for receiving LLQ entry size recommendation
from the device. The driver will use the recommended entry size, unless
the user specifically chooses to use regular or large LLQ entry.
Also added enum ena_llq_header_size_policy_t and llq_plociy field in
order to support the new feature.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit b1c38df05d)
This commit adds a handler for the new aenq message
ENA_ADMIN_DEVICE_REQUEST_RESET,
which in turn causes the driver to trigger reset of a new type:
ENA_REGS_RESET_DEVICE_REQUEST. Also adds counting of such occurrences in
a new statistic for it.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 705879424b)
When attaching ENA driver, ena_netmap_attach() is invoked which, in turn
calls netmap_attach which, initializes a struct netmap_adapter,
allocating the struct's netmap_ring and the struct selinfo.
When we change the interface number of queues we need to reinit the
netmap adapter struct as well, so we need to detach it in order to free
the memory allocated by netmap_attach and allocate new memory based on
the new parameters like number of rings, ring size etc...
Without detaching and attaching the netmap interface, if we're to change
the number of queues from 8 to 2 for example and try to enable netmap,
the kernel will panic since the original netmap struct within the
kernel's possession still thinks that the driver has 8 queues which will
eventually cause a non-allocated virtual address access fault.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit f9c9c01de8)
When processing packets within the rx-flow
ena_netmap_rx_load_desc doesn't know the number of descriptors, so it
sets NS_MOREFRAG to all the slots to indicate that there are more
fragments for this packet.
The code calls ena_netmap_rx_load_desc() for every descriptor in
this packet to map the relevant buffer into the netmap shared memory.
After ena_netmap_rx_load_desc() calls, we need to unset the NS_MOREFRAG
for the last fragment to indicate that this is the last fragment,
so we explicitly turn off NS_MOREFRAG flag.
Current code overrides all other flags and sets NS_BUF_CHANGED.
This patch unsets the relevant flag only.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 2f17afd19a)
Netmap index wraps around based on the number of netmap kernel ring
slots.
Currently the driver prefetches the next slot using nm_i + 1 which may
be wrong since it does not handle wrap around.
This patch fixes that by using the kernel API for fetching the next
netmap index.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit ce20b51cb7)
In case ena_com_prepare_tx() fails within the netmap tx flow,
the driver will unmap the last socket chain.
Currently, the driver unmaps the wrong socket within
ena_netmap_unmap_last_socket_chain().
Illustration of the flow:
1- ena_netmap_tx_frames()
2- ena_netmap_tx_frame()
3- ena_netmap_tx_map_slots()
3.1- Map slot
3.2- Advance to the next socket
4- ena_com_prepare_tx()
4.1- ena_com_prepare_tx() fails
5- ena_netmap_unmap_last_socket_chain()
In step 5, where the driver unmaps the socket, the netmap
index already points at the next entry, meaning we're unmapping the
wrong socket in case ena_com_prepare_tx() fails.
In order to fix that, the driver should first update the netmap index to
point at the previous entry and only then update the socket parameters.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit f236e544a2)
This commit changes the code so all global counters will have the
same line break.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 90953d2f82)
The mbuf is NULL issue happens when the device sends the driver
a completion with a wrong request id.
Trigger a reset whenever this happens.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit da73e3a7d0)
This commit adds differentiation for a reset caused by missing tx
completions, by verifying if the driver didn't receive tx
completions caused by missing interrupts.
The cleanup_running field was added to ena_ring because
cleanup_task.ta_pending is zeroed before ena_cleanup() runs.
Also ena_increment_reset_counter() API was added in order to support
only incrementing the reset counter.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit a33ec635d1)
This commit sets the default value for ena_min_poll_delay_us to 100.
This commit does not change the behavior of the driver, the delay is
calculated as MAX(ENA_MIN_ADMIN_POLL_US, delay_us), where the first
field is already defined as 100.
The second parameter, delay_us is taken from ena_min_poll_delay_us
which is currently unset - 0.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 637ff00f2f)
There can be cases when we trigger reset if an admin interrupt
is missing.
In order to identify this use-case specifically,
this commit adds a new reset reason.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 274319acb4)
RX completion descriptors may sometimes contain errors due
to corruption. Upon identifying such a case, the driver will
trigger a reset with an explicit reset reason
ENA_REGS_RESET_RX_DESCRIPTOR_MALFORMED.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 4af71159db)
TX completion descriptors may sometimes contain errors due
to corruption. Upon identifying such a case, the driver will
trigger a reset with an explicit reset reason
ENA_REGS_RESET_TX_DESCRIPTOR_MALFORMED.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 3872721846)
The driver uses different reset reasons.
Some of them are counted and presented in the driver statistics.
There are cases where statistics are counted on a ring level,
but these are zeroed after a reset procedure takes place.
This commit makes the following changes:
1. Add statistics for the unrepresented reset reasons.
2. Add reset reasons which are counted on a ring level,
to be also global for better tracking.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 89ce3f6314)
This commit updates all the license signatures to 2024.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 8d6806cd08)
This commit is part of the effort of notifying the user of non-optimal
or performance impacting practices.
A new interface is serving as a communication channel
between the device and the driver. One of the goals of this channel is
to create a new mechanism of notifying the driver and user in case of
sub-optimal configuration using a bitmap.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 8cd86b51be)
Currently we count all of the newly added and already existing
missing tx completions in each iteration of
check_missing_comp_in_tx_queue() causing duplicate counts
to missing_tx_comp stat.
This commit adds a new counter new_missed_tx within the relevant
function which only counts the newly added missing tx completions
in each iteration of check_missing_comp_in_tx_queue().
This will allow us to update missing_tx_comp stat accurately without
counting duplicates.
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 1f67704e2c)
Upstream commit [1] made if_alloc_domain() never fail, then also do the
wrappers if_alloc(), if_alloc_dev(), and if_gethandle().
Upstream commit [2] removed the NULL check conducted by the driver.
This commit also removes err_customer_metrics_alloc goto label.
Commit [2] leaves behind a floating free() statement that
deallocates customer_metrics_array. This commit places the
deallocation statement where it belongs.
[1] commit 4787572d05 ("ifnet: make if_alloc_domain() never fail")
[2] commit aa3860851b ("net: Remove unneeded NULL check for the allocated ifnet")
Approved by: cperciva (mentor)
Sponsored by: Amazon, Inc.
(cherry picked from commit 5517ca8486)
Use the appropriate function to flush correctly all entries. The old
code does not remove the element from the hash table, only from the
active queue.
Reviewed by: Peter Lei, rscheff
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D46434
(cherry picked from commit 0b45d36510)
Use the appropriate function to flush correctly all entries. The old
code does not remove the element from the hash table, only from the
active queue.
Reviewed by: Peter Lei, rscheff
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D46433
(cherry picked from commit 5203dcce25)
This prevents a panic that occurs if netlink tries to invoke the
driver's media callbacks during ifattach.
PR: 282374
Reported by: ScottD <biscuits.carry.0j@icloud.com>
(cherry picked from commit f2daf89954)
The decision to disable RDMA on E830 devices has been reversed; some
SKUs will officially launch with RDMA support.
This reverts a change introduced in "ice: Update to 1.42.5-k".
Signed-off-by: Eric Joyner <erj@FreeBSD.org>
Sponsored by: Intel Corporation
(cherry picked from commit 03e1f25dcb)
- Add extra time while waiting for an EMPR to finish on E830 and E825C
devices
- Fix debug mask sysctl help message
- Handle EAGAIN error messages from the Get Link Status AQ command by
retrying them several times
- Handle the condition where E830 extended temperature SKUs can have the
chip initialize without the PHY subsystem being ready due to very low
temperatures; the driver will print out a message when this happens
- Fix inability to set link speed using the advertise_speed sysctl when
link is down and the link_active_on_if_down sysctl is set to 0
- Fix issue where VLANs would not properly be set back up after a
hardware reset (e.g. PF reset) (Thanks jacob.e.keller@intel.com)
Signed-off-by: Eric Joyner <erj@FreeBSD.org>
Tested by: Jeffrey Pieper <jeffrey.e.pieper@intel.com>
Relnotes: yes
Sponsored by: Intel Corporation
Differential Revision: https://reviews.freebsd.org/D46951
(cherry picked from commit f377a0c7df)
- Removes duplicate E825C printout
- Fixes a package download issue with E830
- Updates mentions of E830-XXV to E830-L
- Fix FW logging on load issue by limiting the setting of FW log masks
and other global configuration to just the first PF of the controller
- Update ice_hw_autogen.h with timesync registers (unused in this
driver)
- Fix possible overflow condition in NVM access function
- Blocks RDMA load on E830 devices
Signed-off-by: Eric Joyner <erj@FreeBSD.org>
Tested by: Jeffrey Pieper <jeffrey.e.pieper@intel.com>
Relnotes: yes
Sponsored by: Intel Corporation
Differential Revision: https://reviews.freebsd.org/D46950
(cherry picked from commit 440addc642)
Summary:
- Adds E830 device support
- Adds pre-release E825C support (for the Ethernet device included in
an upcoming Xeon D platform)
- Add sysctl for E810 devices to print out PHY debug statistics (mostly
for FEC debugging)
- Adds per-TX-queue tso counter sysctl to count how many times a TSO offload
was requested for a packet, matching other Intel drivers
- Various bug fixes
Signed-off-by: Eric Joyner <erj@FreeBSD.org>
Tested by: Jeffrey Pieper <jeffrey.e.pieper@intel.com>
Relnotes: yes
Sponsored by: Intel Corporation
Differential Revisison: https://reviews.freebsd.org/D46949
(cherry picked from commit f2635e844d)
In the QEMU workaround code in if_ixv.c, the ixv driver calls
pci_find_cap(dev, PCIY_MSIX, &rid). It is not checking the return code
from that function and the function appears to always be failing. This
then causes the driver to use the rid variable uninitialized, which
will mean setting a bit at an arbitrary offset in pci config space. For
now, this seems to have no adverse impact, but it could easily cause
very subtle problems.
PR: 207037
Sponsored by: BBOX.io
(cherry picked from commit b87b3696c9)
If MTU of the if_ixv VF driver is configured between 1501 and 2030, and
if a peer box have bigger MTU and sends a packet bigger than MTU into
this guest, kernel could crash.
Dynamically calculate the receive packet size on rx_mbuf_sz as ix(4)
does.
PR: 268092
Reported by: Kumara Babu <nkumarababu@gmail.com>
Sponsored by: BBOX.io
(cherry picked from commit cdbe3de538)
Handle ged interrupts directly from the interrupt handler,
while the interrupt source is masked, so as to conform
with the acpi spec, and avoid spurious interrupts and
lockups on boot.
When an acpi ged interrupt is encountered, the spec requires
the os (as stated in 5.6.4: General Purpose Event Handling)
to leave the interrupt source masked until it runs the
EOI handler. This is not a good fit for our method of
queuing the work (including the EOI ack of the interrupt),
via the AcpiOsExecute() taskqueue mechanism.
Note this fixes a bug where an arm64 server could lock up if
it encountered a ged interrupt at boot. The lockup was
due to running on a single core (due to arm64 not using
EARLY_AP_STARTUP), and due to that core encountering a
new interrupt each time the interrupt handler unmasked
the interrupt source, and having the EOI queued on a taskqueue
which never got a chance to run. This is also possible
on any platform when using just a single processor.
The symptom of this is a lockup at boot, with:
"AcpiOsExecute: failed to enqueue task, consider
increasing the debug.acpi.max_tasks tunable" scrolling
on console.
Similarly, spurious interrupts would occur when running
with multiple cores, because it was likely that the
interrupt would fire again immediately, before the
ged task could be run, and before an EOI could be sent
to lower the interrupt line. I would typically see
3-5 copies of every ged event due to this issue.
This adds a tunable, debug.acpi.ged_defer, which can be
set to 1 to restore the old behavior. This was done
because acpi is a complex system, and it may be
theoretically possible something the ged handler does
may sleep (though I cannot easily find anthing by inspection).
MFC after: 1 month
Reviewed by: andrew, jhb, imp
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D42158
(cherry picked from commit be91b4797e)
Each netmap adapter associated with a physical adapter is attached to a
netmap memory pool. contigmalloc() is used to allocate physically
contiguous memory for the pool, but ideally we would ensure that all
such memory is allocated from the NUMA domain local to the adapter.
Augment netmap's memory pools with a NUMA domain ID, similar to how
IOMMU groups are handled in the Linux port. That is, when attaching to
a physical adapter, ensure that the associated memory pools are local to
the adapter's associated memory domain, creating new pools as needed.
Some types of ifnets do not have any defined NUMA affinity; in this case
the domain ID in question is the sentinel value -1.
Add a sysctl, dev.netmap.port_numa_affinity, which can be used to enable
the new behaviour. Keep it disabled by now to avoid surprises in case
netmap applications are relying on zero-copy optimizations to forward
packets between ports belonging to different NUMA domains.
Reviewed by: vmaffione
MFC after: 2 weeks
Sponsored by: Klara, Inc.
Differential Revision: https://reviews.freebsd.org/D46666
(cherry picked from commit 1bae9dc584)
No longer used.
Sponsored by: The FreeBSD Foundation
MFC after: 2 days
Reviewed by: dev_submerge.ch
Differential Revision: https://reviews.freebsd.org/D47269
(cherry picked from commit 9a6cf27583)
When adding a new vchan, we are looking for a parent channel which
either already has vchans (i.e CHN_F_HAS_VCHAN), or does not, but is
also not being used (i.e !CHN_F_BUSY). Since CHN_F_BUSY essentially
tells us if the channel is currently being used or not, there is no need
to check if the channel's refcount is 0 as well.
When removing a vchan, we first check if we have only 1 vchan allocated
that is also being used (so we cannot remove it at the moment), and then
we check if the vchan is not busy and remove it. Again, checking
CHN_F_BUSY is enough.
Sponsored by: The FreeBSD Foundation
MFC after: 2 days
Reviewed by: dev_submerge.ch
Differential Revision: https://reviews.freebsd.org/D47268
(cherry picked from commit 43c0b593c2)
