Jonas Höppner authored 2 years ago and Tobias Kahlki committed 2 years ago

Copy the files from the 5.10 kernel, except
for imx8mm.dtsi and imx8mm-pinfunc.h which come
from the freescale devicetree directory in the 5.15 kernel.

It seems that there are circumstances when access to QBMAN through the
software portals will be delayed. Accessing some lower speeds interfaces
while also QBMAN commands are issued from the kernel will lead to
software timeouts happening in the dpaa2-eth driver.

What we have observed is that management commands like re-arming the
interrupts on a specific channel, waiting for a dequeue response to be
available etc, will take a longer time to complete.
All these commands have to wait for a valid bit to be set for the
command to be interpreted as successfully completed.

Increase the maximum number of times the Linux kernel drivers will busy
poll for a successful result of one of these commands.

Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>

Enable CONFIG_SFP for the Linux SFP infrastructure. Also, enable the
Lynx 28G SerDes PHY driver which supports runtime reconfiguration of
SerDes lanes.

Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>

There is a downstream patch which adds a new interface type -
PHY_INTERFACE_MODE_2500SGMII (which is really the same one as
PHY_INTERFACE_MODE_2500BASEX).

We backported from upstream the following phylink patch which, of
course, does not treat the PHY_INTERFACE_MODE_2500SGMII interface mode
in a switch case statement.
	34ae2c09 ("net: phylink: add generic validate implementation")

Because of this, we get the following build warning.

drivers/net/phy/phylink.c: In function ‘phylink_get_linkmodes’:
drivers/net/phy/phylink.c:322:2: warning: enumeration value ‘PHY_INTERFACE_MODE_2500SGMII’ not handled in switch [-Wswitch]
  322 |  switch (interface) {
      |  ^~~~~~

Fix it by treating the new interface mode in the switch-case statement.

Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>

This reverts commit 6809b4d8.

The patch causes kernel crashes on DPAA2. Revert the changes and
reimplement at a later time.

Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Since the blamed commit, VCAP filters can appear on more than one list.
If their action is "trap", they are chained on ocelot->traps via
filter->trap_list. This is in addition to their normal placement on the
VCAP block->rules list head.

Therefore, when we free a VCAP filter, we must remove it from all lists
it is a member of, including ocelot->traps.

There are at least 2 bugs which are direct consequences of this design
decision.

First is the incorrect usage of list_empty(), meant to denote whether
"filter" is chained into ocelot->traps via filter->trap_list.
This does not do the correct thing, because list_empty() checks whether
"head->next == head", but in our case, head->next == head->prev == NULL.
So we dereference NULL pointers and die when we call list_del().

Second is the fact that not all places that should remove the filter
from ocelot->traps do so. One example is ocelot_vcap_block_remove_filter(),
which is where we have the main kfree(filter). By keeping freed filters
in ocelot->traps we end up in a use-after-free in
felix_update_trapping_destinations().

Attempting to fix all the buggy patterns is a whack-a-mole game which
makes the driver unmaintainable. Actually this is what the previous
patch version attempted to do:
https://patchwork.kernel.org/project/netdevbpf/patch/20220503115728.834457-3-vladimir.oltean@nxp.com/



but it introduced another set of bugs, because there are other places in
which create VCAP filters, not just ocelot_vcap_filter_create():

- ocelot_trap_add()
- felix_tag_8021q_vlan_add_rx()
- felix_tag_8021q_vlan_add_tx()

Relying on the convention that all those code paths must call
INIT_LIST_HEAD(&filter->trap_list) is not going to scale.

So let's do what should have been done in the first place and keep a
bool in struct ocelot_vcap_filter which denotes whether we are looking
at a trapping rule or not. Iterating now happens over the main VCAP IS2
block->rules. The advantage is that we no longer risk having stale
references to a freed filter, since it is only present in that list.

Fixes: e42bd4ed ("net: mscc: ocelot: keep traps in a list")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
(cherry picked from commit e1846cff)
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>

This reverts commit 8146e2b6. It was
fixed differently upstream, we'll use that.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>

This reverts commit 0e9c8bdf. This was
fixed differently in upstream, we're going for that approach.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>

Downstream commit bbd95975 ("selftests: ocelot: tc_flower_chains:
reorder interfaces") got merged upstream as 93196ef9 ("selftests:
ocelot: tc_flower_chains: use conventional interface names"), with the
difference that interfaces are ordered differently and traffic flows in
the other direction.

Eliminate this difference by copy-pasting the upstream version of the
selftest.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>

In order to use micfil correctly, uart3 is disabled in relating dts
files.

Reviewed-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Signed-off-by: Chancel Liu <chancel.liu@nxp.com>
Acked-by: Jason Liu <jason.hui.liu@nxp.com>

Read entropy-delay property from device tree for
TRNG configuration.

update the value if entry is missing in DT.

Signed-off-by: Vabhav Sharma <vabhav.sharma@nxp.com>
Reviewed-by: Gaurav Jain <gaurav.jain@nxp.com>
Reviewed-by: Horia Geanta <horia.geanta@nxp.com>

Add imx6sx entropy-delay property.

TRNG self test are run to determine the correct entropy delay.
firmware test is executed with different voltage and temperature to
identify the worst case value for entropy delay. after adding a margin
value(1000),entropy delay should be at least 12000.

Signed-off-by: Vabhav Sharma <vabhav.sharma@nxp.com>
Reviewed-by: Horia Geanta <horia.geanta@nxp.com>
Reviewed-by: Gaurav Jain <gaurav.jain@nxp.com>

Add entropy delay property which defines the length (in system clocks) of
each Entropy sample taken for TRNG configuration.

Signed-off-by: Vabhav Sharma <vabhav.sharma@nxp.com>
Reviewed-by: Horia Geanta <horia.geanta@nxp.com>
Reviewed-by: Varun Sethi <v.sethi@nxp.com>

Process 'fixed-link' flags under ACPI boot and configure the MACs and
PHYs appropriately.

Signed-off-by: Vikas Singh <vikas.singh@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add the fwnode_fixed_phy_register() call. The call is similar to the existing
fixed_phy_register().

Signed-off-by: Vikas Singh <vikas.singh@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add ACPI support for the DPAA Ethernet driver when connecting
to a PHY.

Signed-off-by: AnjanaSuraj <anjana.r@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add support to get the internal PHY present on MDIO under ACPI.

Signed-off-by: Vikas Singh <vikas.singh@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add ACPI support for the FMan MAC driver so that it can be
probed in both DT and ACPI cases.

Signed-off-by: AnjanaSuraj <anjana.r@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add ACPI support for the FMan Port driver so that it can
be probed in both DT and ACPI cases.

Signed-off-by: Vikas Singh <vikas.singh@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add ACPI support for the Frame Manager driver so that it can be
probed in both DT and ACPI cases.

Signed-off-by: AnjanaSuraj <anjana.r@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Set up reserved memory regions for the QBMan devices under ACPI using
Contiguous Memory Allocations (CMA) at probe. Changing the following
default CMA kernel configs is required at build:
(a) Set CONFIG_CMA_SIZE_MBYTES to 256
(b) Set CONFIG_CMA_ALIGNMENT to 12

The qbman_init_private_mem() API is updated to indicate which device is
calling it (BMan or QMan).

Signed-off-by: Vikas Singh <vikas.singh@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add ACPI support for the Queue Manager Portal driver so that
it can be probed in both DT and ACPI cases.

Signed-off-by: Vikas Singh <vikas.singh@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add ACPI support for the Queue Manager CCSR driver so that
it can be probed in both DT and ACPI cases.

Signed-off-by: Vikas Singh <vikas.singh@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add ACPI support for the Buffer Manager Portal driver so that
it can be probed in both DT and ACPI cases.

Signed-off-by: Vikas Singh <vikas.singh@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add ACPI support for the Buffer Manager CCSR driver so that
it can be probed in both DT and ACPI cases.

Signed-off-by: Vikas Singh <vikas.singh@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

Add ACPI support for QorIQ PTP clock so that it can be probed
in both DT and ACPI cases.

Signed-off-by: AnjanaSuraj <anjana.r@puresoftware.com>
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>

DMA unmap the Scatter/Gather table before going through the array to
unmap and free each of the header and data chunks. This is so we do not
touch the data between the dma_map and dma_unmap calls.

Fixes: 3dc709e0cd47 ("dpaa2-eth: add support for software TSO")
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 0a09c5b8cb8f75344da7d90c771b84f7cdeaea04)
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>

The incorrect software annotation field was being used, swa->sg.sgt_size
instead of swa->tso.sgt_size, which meant that the SGT buffer was
unmapped with a wrong size.
This is also confirmed by the DMA API debug prints which showed the
following:

[   38.962434] DMA-API: fsl_dpaa2_eth dpni.2: device driver frees DMA memory with different size [device address=0x0000fffffafba740] [map size=224 bytes] [unmap size=0 bytes]
[   38.980496] WARNING: CPU: 11 PID: 1131 at kernel/dma/debug.c:973 check_unmap+0x58c/0x9b0
[   38.988586] Modules linked in:
[   38.991631] CPU: 11 PID: 1131 Comm: iperf3 Not tainted 5.18.0-rc7-00117-g59130eeb2b8f #1972
[   38.999970] Hardware name: NXP Layerscape LX2160ARDB (DT)

Fixes: 3dc709e0cd47 ("dpaa2-eth: add support for software TSO")
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit d5f4e19a85670b4e5697654f4a4e086e064f8a47)
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>

The TSO header was DMA unmapped before the virtual address was retrieved
and then used to free the buffer. This meant that we were actually
removing the DMA map and then trying to search for it to help in
retrieving the virtual address. This lead to a invalid virtual address
being used in the kfree call.

Fix this by calling dpaa2_iova_to_virt() prior to the dma_unmap call.

[  487.231819] Unable to handle kernel paging request at virtual address fffffd9807000008

(...)

[  487.354061] Hardware name: SolidRun LX2160A Honeycomb (DT)
[  487.359535] pstate: a0400005 (NzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[  487.366485] pc : kfree+0xac/0x304
[  487.369799] lr : kfree+0x204/0x304
[  487.373191] sp : ffff80000c4eb120
[  487.376493] x29: ffff80000c4eb120 x28: ffff662240c46400 x27: 0000000000000001
[  487.383621] x26: 0000000000000001 x25: ffff662246da0cc0 x24: ffff66224af78000
[  487.390748] x23: ffffad184f4ce008 x22: ffffad1850185000 x21: ffffad1838d13cec
[  487.397874] x20: ffff6601c0000000 x19: fffffd9807000000 x18: 0000000000000000
[  487.405000] x17: ffffb910cdc49000 x16: ffffad184d7d9080 x15: 0000000000004000
[  487.412126] x14: 0000000000000008 x13: 000000000000ffff x12: 0000000000000000
[  487.419252] x11: 0000000000000004 x10: 0000000000000001 x9 : ffffad184d7d927c
[  487.426379] x8 : 0000000000000000 x7 : 0000000ffffffd1d x6 : ffff662240a94900
[  487.433505] x5 : 0000000000000003 x4 : 0000000000000009 x3 : ffffad184f4ce008
[  487.440632] x2 : ffff662243eec000 x1 : 0000000100000100 x0 : fffffc0000000000
[  487.447758] Call trace:
[  487.450194]  kfree+0xac/0x304
[  487.453151]  dpaa2_eth_free_tx_fd.isra.0+0x33c/0x3e0 [fsl_dpaa2_eth]
[  487.459507]  dpaa2_eth_tx_conf+0x100/0x2e0 [fsl_dpaa2_eth]
[  487.464989]  dpaa2_eth_poll+0xdc/0x380 [fsl_dpaa2_eth]

Fixes: 3dc709e0cd47 ("dpaa2-eth: add support for software TSO")
Link: https://bugzilla.kernel.org/show_bug.cgi?id=215886


Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 06d129946a71f3159b3b40ee95549183edf2c79d)
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>

In the imx8mp-evk-os08a20.dts, just single os08a20 sensor is used.
The i2c2 node is used but the other i2c3 sensor's node should be removed.
If i2c3 sensor's node is not removed, this will lead a inaccurate variable.
When do grep os08a20 /sys/firmware/devicetree/base/soc@0//i2c@/*/compatible",
the return value is not correct.

Signed-off-by: Alice Yuan <alice.yuan@nxp.com>
Reviewed-by: Robby Cai <robby.cai@nxp.com>
Acked-by: Jason Liu <jason.hui.liu@nxp.com>

For sensor FXLS8974CFR3, in default, the int1 pad is pull down. But
on imx8ulp-9x9-evk board, this int1 pad is shared with other two sensor's
interrupt pad. If int1 is pull down, this will block the other shared
interrupt. So for shared interrupt pad connection, need to config all
these three interrupt pin as open-drain.

Signed-off-by: Haibo Chen <haibo.chen@nxp.com>
Reviewed-by: Clark Wang <xiaoning.wang@nxp.com>
Acked-by: Jason Liu <jason.hui.liu@nxp.com>

This patch add the support to set interrupt pin as open-drain even
this interrupt pin is not used at all. There is a specail case on
imx8ulp-9x9 evk board. Three sensors share their interrupt pin, and
this pin belong to M core side, need M image implement the interrupt
handler. So for linux on A core side, driver can't use this shared
interrupt pin. But need to make sure all the three sensors interrupt
pin is open-drain, otherwise may impact the share interrupt pad behavior.

Signed-off-by: Haibo Chen <haibo.chen@nxp.com>
Reviewed-by: Clark Wang <xiaoning.wang@nxp.com>
Acked-by: Jason Liu <jason.hui.liu@nxp.com>

Update the copyright and license info for this newly
added file.

Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>

On i.MX8ULP 9*9 EVK baord, the GPIO used for SW4 is changed
to port PTF26, so update the gpio key property to correct it.

Signed-off-by: Jacky Bai <ping.bai@nxp.com>
Reviewed-by: Ye Li <ye.li@nxp.com>

This patch fixes a bug in scatterlist processing that may cause incorrect AES block encryption/decryption.

Fixes: 2e6d793e ("crypto: mxs-dcp - Use sg_mapping_iter to copy data")
Signed-off-by: Tomas Paukrt <tomaspaukrt@email.cz>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
(cherry picked from commit 28e9b6d8)

Previously we only assigned clock rate for 44kHz series sample rate.
But meet issue on i.MX8MM 16KHz case, the calculated mclk_freq is
512KHz, but because root clock divider range limitation, the smallest
mclk rate is 768Hkz, then cause issue. so also assign mclk_freq
for 48kHz series sample rates

Fixes: 3a650ee6 ("LF-5909-2: ASoC: imx-card: fix DSD mclk frequency")
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Reviewed-by: Chancel Liu <chancel.liu@nxp.com>
Acked-by: Jason Liu <jason.hui.liu@nxp.com>

kzalloc order of parameters is wrong.

Reviewed-by: Paul Olaru <paul.olaru@nxp.com>
Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Acked-by: Jason Liu <jason.hui.liu@nxp.com>

The SC_PM_CLK_MISC0 clock is not available for the resources of ENET0/1 on
imx8dxl. Using the clock device compatible string to identify the platform
and only initiate the enetX_rgmii_rx_clk clock for non imx8dxl platform.

Signed-off-by: Shenwei Wang <shenwei.wang@nxp.com>
Reviewed-by: Ranjani Vaidyanathan <ranjani.vaidyanathan@nxp.com>
Reviewed-by: Jacky Bai <ping.bai@nxp.com>
Acked-by: Jason Liu <jason.hui.liu@nxp.com>

The LPUART hardware doesn't zero out the parity bit on the received
characters. This behavior won't impact the use cases of CS8 because
the parity bit is the 9th bit which is not currently used by software.
But the parity bit for CS7 must be zeroed out by software in order to
get the correct raw data.

Signed-off-by: Shenwei Wang <shenwei.wang@nxp.com>
Reviewed-by: Sherry Sun <sherry.sun@nxp.com>
Acked-by: Jason Liu <jason.hui.liu@nxp.com>

This reverts commit 0fc1913e.