Looking at the current upstream driver implementation, it seems like the
TX/RX flow control is enabled only if the flow control pause option is
resolved from the device/link partner advertisements (or otherwise set).
On the other hand, our current in-tree driver force enables TX/RX
flow control by default, thus possibly leading to TX timeouts if the
other end sends pause frames (which are not properly handled?):
WARNING: CPU: 3 PID: 0 at net/sched/sch_generic.c:320 dev_watchdog+0x1ac/0x324
NETDEV WATCHDOG: eth0 (mtk_soc_eth): transmit queue 0 timed out
Disabling the flow control on PORT 5 MAC seems to fix this issues as the
pause frames are then filtered out. While at it, I'm removing the if
condition completely as suggested, since this code is run only on mt7621
SoC, so there is no need to check for the silicon revisions.
Ref: https://lists.openwrt.org/pipermail/openwrt-devel/2017-November/009882.html
Ref: https://forum.openwrt.org/t/mtk-soc-eth-watchdog-timeout-after-r11573/50000/12
Suggested-by: Felix Fietkau <nbd@nbd.name>
Reported-by: Rosen Penev <rosenp@gmail.com>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit c8f8e59816eca49d776562d2d302bf990a87faf0)
This ioctl is currently routed through generic interface code.
dev_ioctl
dev_ethtool
__ethtool_get_link_ksettings
phy_ethtool_ioctl
Cc: Felix Fietkau <nbd@nbd.name>
Cc: John Crispin <john@phrozen.org>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Some broken ISPs (e.g. Comcast) send DHCPv6 packets with hop limit=0.
This trips up the TTL=0 check in the PPE if enabled.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Current code directly writes the FOE entry to hash_val+1 position
when hash collision occurs. However, it is found that this behavior
will cause the cache and the hardware FOE table to be inconsistent.
For example, there are three flows, and their hashed values are all
equal to 100. The first flow is written to the position of 100. The
second flow is written to the position of 100+1. Then, the logic of
the current code will also write the third flow to 100+1.
At this time, the cache has flow 1 and 2; and the hardware FOE table
has flow 1 and 3, where these two parts store different contents.
So it is necessary to check whether the hash_val+1 is also occupied
before writing. If hash_val+1 is also occupied, we won’t bind th
third flow to the FOE table.
Addition to that, we also cancel the processing of foe_entry removal
because the hardware has auto age-out ability. The hardware will
periodically iterate through the FOE table to find out the time-out
entry and set it as INVALID.
Signed-off-by: HsiuWen Yen <y.hsiuwen@gmail.com>
Sometimes the tuples might be hashed to the same FOE entry.
When this hash collision problem occurs, some of the
connections will not be bound and consequently the CPU
idle rate cannot reach 100%. Therefore, two-way hashing
is adopted to alleviate this problem.
Signed-off-by: HsiuWen Yen <y.hsiuwen@gmail.com>
Some boards have external switches different than mt7530.
This patch allow to use mdio-mode without 0x1f register.
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
The pinmux for all SoCs using this driver is now set via the pinmux. It
makes this code obsolete.
Some of the code targeting the mt76x8 SoCs is still required. The sd
card pins share the pads with the EPHY. These pads need to be switched
to digital mode if the pins are used for sd cards.
The eMMC 8-bit mode has to be enabled via pinmux instead of a kernel
option. The uart2 group need to be set to function "sdxc d5 d4", pwm1
to "sdxc d6" and pwm0 to "sdxc d7" to do so. It can't be done by as
part of a default pinmux, as it would break the normal operation of
uart2.
Signed-off-by: Mathias Kresin <dev@kresin.me>
A buffer is split into multiple descriptors if it exceeds 16 KB.
Apply the same split for the skb head as well (to deal with corner cases
on fraglist support)
Signed-off-by: Felix Fietkau <nbd@nbd.name>
This reverts commit 3a8efaef00.
The change reportedly breaks UART2 on some boards. Furthermore it uses
bitwise logic on an uninitialized variable and fails to explain what it
is fixing exactly.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
To share mdio addr for IntPHY and ExtPHY,
as described in the documentation (MT7620_ProgrammingGuide.pdf).
(refer: http://download.villagetelco.org/hardware/MT7620/MT7620_ProgrammingGuide.pdf)
when port4 setup to work as gmac mode, dts like:
&gsw {
mediatek,port4 = "gmac";
};
we should set SYSCFG1.GE2_MODE==0x0 (RGMII).
but SYSCFG1.GE2_MODE may have been set to 3(RJ-45) by uboot/default
so we need to re-set it to 0x0
before this changes:
gsw: 4FE + 2GE may not work correctly and MDIO addr 4 cannot be used by ExtPHY
after this changes:
gsw: 4FE + 2GE works and MDIO addr 4 can be used by ExtPHY
Signed-off-by: Chen Minqiang <ptpt52@gmail.com>
When PHY's are defined on the MDIO bus in the DTS, gigabit support was
being masked out for no apparent reason, pegging all such ports to 10/100.
If gigabit support must be disabled for some reason, there should be a
"max-speed" property in the DTS.
Reported-by: James McKenzie <openwrt@madingley.org>
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
Mediatek has a reference platform that pairs an MT7620A with an MT7530W,
where the latter responds on MDIO address 0x1f while both chips respond on
0x0 to 0x4. The driver special-cases this arrangement to make sure it's
talking to the right chip, but two different ways in two different places.
This patch consolidates the detection without the current requirement of
both tests to be separately satisfied in the DTS.
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
There is another thing about crc to do when initialize SD card on
MT7628.
This commit is to fix this init issue.
Signed-off-by: LoveSy <shana@zju.edu.cn>
The code has some remaining issues that cause ethernet hangs, so
disable it for now until we can get it fixed
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Using the NAPI or netdev frag cache along with other drivers can lead to
32 KiB pages being held for a long time, despite only being used for
very few page fragment.
This can happen if the ethernet driver grabs one or two fragments for rx
ring refill, while other drivers use (and free up) the remaining
fragments. The 32 KiB higher-order page can only be freed once all users
have freed their fragments, which only happens after the rings of all
drivers holding the fragments have wrapped around.
Depending on the traffic patterns, this can waste a lot of memory and
look a lot like a memory leak
Signed-off-by: Felix Fietkau <nbd@nbd.name>
GRO stores packets as fraglist. If they are routed back to the ethernet
device, they need to be re-segmented if the driver does not support
sending fraglists.
Add the missing support for that, along with a missing feature flag that
allows full routed GRO->TSO offload.
Considerably reduces CPU utilization for routing
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Since kernel 4.10 commit 61e84623ace3 ("net: centralize net_device
min/max MTU checking"), the range of mtu is [min_mtu, max_mtu], which
is [68, 1500] by default.
It's necessary to set a max_mtu if a mtu > 1500 is supported.
Signed-off-by: Mathias Kresin <dev@kresin.me>
This reverts commit 8a570921b5.
This seems to have been accidentally reverted. This fixes mt7620 and
mt7628.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Acked-by: Daniel Golle <daniel@makrotopia.org>
Mostly whitespace cleanups. Some unneeded code was removed.
MMC init was also moved to the probe function as in
6069bdd
The cleanup commits are over 100, making it hard to do them individually.
Tested on GnuBee PC1 with an SD card being used as swap.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
v2: Fixed compile issue with mt7620
In case of error, the function devm_ioremap_resource() returns ERR_PTR()
and never returns NULL. The NULL test in the return value check should
be replaced with IS_ERR().
Fixes: f079b6406348 ("staging: mt7621-eth: add gigabit switch driver (GSW)")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In case of error, the function of_phy_connect() returns NULL pointer not
ERR_PTR(). The IS_ERR() test in the return value check should be
replaced with NULL test.
Fixes: e3cbf478f846 ("staging: mt7621-eth: add the drivers core files")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In case of error, the function devm_ioremap_resource() returns ERR_PTR()
and never returns NULL. The NULL test in the return value check should
be replaced with IS_ERR().
Fixes: e3cbf478f846 ("staging: mt7621-eth: add the drivers core files")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is an amagalmation of two upstream commits dealing with whitespace
and dead code removal. I'm synching instead of having two separate commits
as they go out of order compared to previous commits here.
Tested on GnuBee PC1.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Mostly whitespace cleanups. Some unneeded code was removed.
MMC init was also moved to the probe function as in
6069bdd087
The cleanup commits are over 100, making it hard to do them individually.
Tested on GnuBee PC1 with an SD card being used as swap.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
These resources are extracted from devicetree, so they aren't
needed here.
Signed-off-by: NeilBrown <neil@brown.name>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Code currently defines:
and then compiles code only if they are defined.
We might want to disable some of these via devicetree one
day, but for now just remove the #defines and the
conditions - all the code for different ports is
easy to identify.
Signed-off-by: NeilBrown <neil@brown.name>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
As the Interrupts for the PCI adapters are listed in
devicetree we shouldn't need to have them explicit in the code.
The simplest way to do this is to use of_irq_parse_and_map_pci()
and specify an interrupt-map which identifies the different
PCI hosts by bus/slot numbers.
This has the advantage that the hwirq number are mapped to virq
numbers for us, so the ugly hack can go.
Signed-off-by: NeilBrown <neil@brown.name>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove the owner assignment form the platform driver as
platform_driver_register() already initializes the owner.
Found using coccinelle.
Signed-off-by: Christian Lütke-Stetzkamp <christian@lkamp.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The stats update needs to protect against being interrupted by a tasklet
The u64 stats seqlock needs to be initialized
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Supports IPv4 flow offloading on MT7621 for Routing, SNAT and DNAT
Supported are regular ethernet->ethernet connections, including one
802.1q VLAN and/or PPPoE encapsulation
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
of_dma_configure() sets a default ->dma_mask of
DMA_BIT_MASK(32), claiming devices can DMA from
the full 32bit address space.
The mtk-mmc driver does not support access to
highmem pages, so it is really limited to the
bottom 512M (actually 448M due to 64M of IO space).
Setting ->dma_mask to NULL causes mmc_setup_queue()
to fall-back to using BLK_BOUNCE_HIGH to tell the
block layer to use a bounce-buffer for any highmem
pages requiring IO.
Signed-off-by: NeilBrown <neil@brown.name>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Petr Štetiar