Extended atomics are supported with RC and XRC QP types, but Linux commit
a60109dc9a95 added an unneeded check to to_mlx5_access_flags().
This broke XRC QPs.
The following ib_atomic_bw invocation over XRC reproduces the issue:
ib_atomic_bw -d mlx5_1 --connection=XRC --atomic_type=FETCH_AND_ADD
It is safe to remove such checks because the QP type was already checked
in ib_modify_qp_is_ok(), which was previously called from
mlx5_ib_modify_qp().
Linux commit:
13f8d9c16693afb908ead3d2a758adbe6a79eccd
Reviewed by: kib
Sponsored by: Mellanox Technologies // NVIDIA Networking
(cherry picked from commit cf88b86e49)
The maximum page size in the mkey context is 2GB.
Until today, we didn't enforce this requirement in the code, and therefore,
if we got a page size larger than 2GB, we have passed zeros in the
log_page_shift instead of the actual value and the registration failed.
This patch limits the driver to use compound pages of 2GB for mkeys.
Linux commit:
762f899ae7875554284af92b821be8c083227092
Reviewed by: kib
Sponsored by: Mellanox Technologies // NVIDIA Networking
(cherry picked from commit 565cb4e8cc)
The patch simplifies mlx5_ib_cont_pages and fixes the following
issues in the original implementation:
First issues is related to alignment of the PFNs. After the check
base + p != PFN, the alignment of the PFN wasn't checked. So the PFN
sequence 0, 1, 1, 2 would result in a page_shift of 13 even though
the 3rd PFN is not 8KB aligned.
This wasn't actually a bug because it was supported by all the
existing mlx5 compatible device, but we don't want to require
this support in all future devices.
Another issue is because the inner loop didn't advance PFN so
the test "if (base + p != pfn)" always failed for SGE with
len > (1<<page_shift).
Linux commit:
d67bc5d4e3e100d762c0f57ea67f28bc219698a6
Reviewed by: kib
Sponsored by: Mellanox Technologies // NVIDIA Networking
(cherry picked from commit 21bc3710a4)
- Upon error more completion events than requested may be generated,
particularly when using the completion event factor feature.
- Count number of event errors in the transmit path.
Reviewed by: kib
Sponsored by: Mellanox Technologies // NVIDIA Networking
(cherry picked from commit 4f4739a77b)
On some environments, such as certain SRIOV VF configurations, RoCE is
not supported for mlx5 Ethernet ports. Currently, the driver will not
open IB device on that port.
This is problematic, since we do want user-space RAW Ethernet (RAW_PACKET
QPs) functionality to remain in place. For that end, enhance the relevant
driver flows such that we do create a device instance in that case.
Linux commit:
ca5b91d63192ceaa41a6145f8c923debb64c71fa
Reviewed by: kib
Sponsored by: Mellanox Technologies // NVIDIA Networking
(cherry picked from commit 626cb01d44)
Make the mlx5e_mode_table[] array one dimensional, because there is only
one entry, 10G ER/LR, which share the same protocol bit.
This patch only adds support for basic sub-type distinguishing for the
extended protocol bits. Use verbose ifconfig eeprom output to get actual
media type.
Remove write only "connector_type" variable while at it.
Reviewed by: kib
Sponsored by: Mellanox Technologies // NVIDIA Networking
(cherry picked from commit a888087fba)
Reserve one page for the DMA subsystem, that may need it when the I/O
buffer is not page aligned.
Without this change, writes with the maximum allowed size failed, if:
- physical memory was fragmented, making it necessary to use one DMA
segment for each page
- the buffer to be written was not page aligned, causing the DMA
subsystem to need one extra segment
In the scenario above, the DMA subsystem would run out of segments,
resulting in a write with no SG segments, that would fail.
Reviewed by: imp
Sponsored by: Instituto de Pesquisas Eldorado (eldorado.org.br)
Differential Revision: https://reviews.freebsd.org/D30798
(cherry picked from commit 9c2c635319)
Last an(4) devices have been End Of Life and End Of Sale in 2007.
Time to remove this driver.
Differential Revision: https://reviews.freebsd.org/D30678
Reviewed by: imp (earlier version), adrian (earlier version)
MFC after: 3 days
Sponsored by: Diablotin Systems
(cherry picked from commit ab30bb8270)
If a regulator hasn't been enable by a driver but is enabled in hardware
(most likely enabled by U-Boot), regulator_status will returns that it
is enabled and so any call to regulator_disable will panic as it wasn't
enabled by one of our drivers.
Sponsored by: Diablotin Systems
Differential Revision: https://reviews.freebsd.org/D30293
(cherry picked from commit f52072b06d)
This allow us to powerup/down the card and enabling/disabling the
regulators if any.
Sponsored by: Diablotin Systems
Differential Revision: https://reviews.freebsd.org/D30292
(cherry picked from commit ce41765c21)
This helper can be used to enable/disable the regulator and starting
the power sequence of sd/sdio/eMMC cards.
Sponsored by: Diablotin Systems
Differential Revision: https://reviews.freebsd.org/D30291
(cherry picked from commit 03d4e8bb65)
If a sd/emmc node have a pwrseq property parse it and get the corresponding
driver.
This can later be used to powerup/powerdown the SDIO card or eMMC.
Sponsored by: Diablotin Systems
Differential Revision: https://reviews.freebsd.org/D30289
(cherry picked from commit b0387990a7)
This driver is used to power up sdio card or eMMC.
It handle the reset-gpio, clocks and needed delays for powerup/powerdown.
Sponsored by: Diablotin Systems
Differential Revision: https://reviews.freebsd.org/D30288
(cherry picked from commit 5b2a81f58d)
For the discovery phase of SD/eMMC we need to do some transaction in a async
way.
The classic CAM XPT_{GET,SET}_TRAN_SETTING cannot be used in a async way.
This also allow us to split the discovery phase into a more complete state
machine and we don't mtx_sleep with a random number to wait for completion
of the tasks.
For mmc_sim we now do the SET_TRAN_SETTING in a taskqueue so we can call
the needed function for regulators/clocks without the cam lock(s). This part is
still needed to be done for sdhci.
We also now save the host OCR in the discovery phase as it wasn't done before and
only worked because the same ccb was reused.
Reviewed by: imp, kibab, bz
Differential Revision: https://reviews.freebsd.org/D30038
(cherry picked from commit af2253f61c)
A lot more generic cam related things are done in mmc_sim so this simplify
the driver a lot.
Differential Revision: https://reviews.freebsd.org/D27487
Reviewed by: kibab
(cherry picked from commit f1cc48e5da)
The pwm utility cant set the only flag defined (PWM_POLARITY_INVERTED) so this
patch add the option -I (capital letter i) to send it to the drivers.
None of existing PWM driver have implemented support for flags.
But soon:ish I will put up an review of a pwm driver using TI OMAP DMTimer.
Differential Revision: https://reviews.freebsd.org/D29137
MFC after: 2 weeks
(cherry picked from commit 17b14d8f77)
Fix the types of period and duty in share/man/man9/pwmbus.9 to match the one in sys/dev/pmw/pwmbus.c.
Reviewed By: rpokala
Differential Revision: https://reviews.freebsd.org/D29139
MFC after: 3 days
(cherry picked from commit 7d4a5de84d)
To guard against the ill effects of a spurious interrupt during
construction (or one that was bogusly pending), enable interrupts after
the qpair is completely constructed. Otherwise, we can die with null
pointer dereferences in nvme_qpair_process_completions. This has been
observed in at least one pre-release NVMe drive where the MSIX interrupt
fired while the queue was being created, before we'd started the NVMe
controller card.
The alternative of only turning on the interrupts after the rest was
tried, but was insufficient to work around this bug and made the code
more complicated w/o benefit.
Reviewed by: mav, chuck
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D31182
(cherry picked from commit fc9a084023)
ni_txseqs is kept as 16-bit counter, but we need to trim the upper four
bits as they may have special meanings for the firmware / hardware.
For instance, bit 15 enables hardware / firmware generation of sequence
numbers that overrides sequence numbers programmed by the driver.
Reviewed by: adrian
(cherry picked from commit 7544c1d20d)
Given all the code does operate on struct ifnet, the last step in this
longer series of changes now is to rename struct net_device to
struct ifnet (that is what it was defined to in the LinuxKPi code).
While mlx4 and OFED are "shared" code the decision was made years ago
to not write it based on the netdevice KPI but the native ifnet KPI
for most of it. This commit simply spells this out and with that
frees "struct netdevice" to be re-done on LinuxKPI to become a more
native/mixed implementation over time as needed by, e.g., wireless
drivers.
Sponsored by: The FreeBSD Foundation
Reviewed by: hselasky
Differential Revision: https://reviews.freebsd.org/D30515
(cherry picked from commit 1411f52fac)
The LinuxKPI net_device actually is an ifnet; in order to further
clean that up so we can extend "net_device" replace the few macros
inline in mlx4.
Sponsored by: The FreeBSD Foundation
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D30476
(cherry picked from commit 60afad6fc3)
This removes all unused bits from linux/netdevice.h and migrates two
inline functions into the mlx4 and ofed code respectively.
This gets the mlx4/ofed (struct ifnet) specific bits down to 7 lines
in netdevice.h.
Sponsored by: The FreeBSD Foundation
Reviewed by: hselasky, kib
Differential Revision: https://reviews.freebsd.org/D30461
(cherry picked from commit c35034b338)
