Commit graph

403 commits

Author SHA1 Message Date
John Crispin
aecac14cc4 ramips: Enable rx of 1536 bytes ethernet frames on MT7621
The default switch frame size (with FCS + header) is 1536 bytes. But the
GMAC only accepted frames up to 1522 bytes. Setting it to 1536 allows to
receive ethernet frames using the full of MTU 1500 + an extra VLAN header +
VLAN header added by the switch.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>

SVN-Revision: 47117
2015-10-05 10:26:34 +00:00
John Crispin
7b306e3eb3 ramips: Fix too small rx buffer
The driver assumes that the maximum received buffer for non-jumbo frames is
1536 bytes. But the allocation of the rx fragment doesn't reflect that. It
currently allocates fragments which will only be large enough to be used as
rx buffer with the size of 1534 bytes. This is problematic because the GMAC
will now try to write to 2 bytes which don't belong to its receive buffer
when a large enough ethernet frame is received.

This may already be a problem on existing chips but will at least become a
problem when the 1536 byte rx modus is enabled on MT7621a. It is required
on this SoC to receive ethernet frames which use their full 1500 bytes MTU
and a VLAN header next to the switch VLAN tag.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>

SVN-Revision: 47116
2015-10-05 10:26:28 +00:00
John Crispin
6b4985b105 ramips: Fix setting of rx buffer length
The length of the DMA rx buffer was always set to 0 because the function
for extracting the length was used to calculate the value for setting it.
Instead the macro has to be split in a get and set function similar to the
TX_DMA_(GET_|)PLEN(0|1) macro.

No problem was noticed on MT7621a before this was changed and thus maybe it
was hidden by different problem which is not yet fixed.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>

SVN-Revision: 47115
2015-10-05 10:26:21 +00:00
John Crispin
1aab21df9f ramips: Allow to receive vlan over untag ports on MT7530
The MT7530 switch driver with enable_vlan set will automatically set all
ports to the user port mode. The hardware will remove the incoming vlan tag
on these ports and use it for its internal vlan. This is usually not wanted
and makes it impossible to communicate via vlan over the switch in both
directions.

It is possible to configure a switch port to "transparent mode" when this
port is only used as untag in the switch VLANs. This will disable the VLAN
untagging of packets when they were received on this port. The tagging on
"tag" ports based on the vlan id is still working.

The transparent port mode cannot be used when a port is both used in a VLAN
as "tag" and in another one as "untag" port.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>

SVN-Revision: 47114
2015-10-05 10:26:09 +00:00
Felix Fietkau
1e6479aa87 ramips: collect rx related members to fe_rx_ring struct
Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 46295
2015-07-10 19:41:48 +00:00
Felix Fietkau
279be7be09 ramips: add xmit_more support
use pktgen to verify on rt3662. can improve transmit rate.
pkt_size 1500
burst 1 : 807Mb/sec
burst 8 : 984Mb/sec

pkt_size 60
burst 1 : 57Mb/sec
burst 8 : 236Mb/sec

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 46294
2015-07-10 19:41:42 +00:00
John Crispin
2238b75285 ralink: add support for mt7621 switch counters
Signed-off-by: Cristian Morales Vega <cristian@samknows.com>

SVN-Revision: 46050
2015-06-18 18:21:52 +00:00
Felix Fietkau
4480fbce62 ramips: improve tx clean up and add fe_tx_ring struct
if there is any new tx need to clean up. do it in next napi poll.
collect tx related members to fe_tx_ring struct. for better
cache usage and more readable.

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 45895
2015-06-05 08:59:57 +00:00
Felix Fietkau
2a66c9886f ramips: change ethernet napi interrupt sequence
when open device. first ready napi software rx.
then enable hardware interrupt.
final start software tx queue to send data.

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 45894
2015-06-05 08:59:48 +00:00
Felix Fietkau
0d043d3ac0 ramips: fix ethernet vlan tx offload support check when delete
Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 45893
2015-06-05 08:59:42 +00:00
Felix Fietkau
e890a3bc9a ramips: make ethernet register map u16 instead of u32
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 44940
2015-03-22 15:11:14 +00:00
Felix Fietkau
37912d66c2 ramips: increase NAPI weight for mt7621 to improve routing performance a bit more
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 44918
2015-03-21 12:08:07 +00:00
John Crispin
c7229b96b7 ralink: support change tx/rx ring size
Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44741
2015-03-13 08:38:08 +00:00
John Crispin
2c94eef5df ralink: use random generate mac address instead of default one.
Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44740
2015-03-13 08:38:01 +00:00
John Crispin
0f4431758b ralink: fix when tx done is 0 also need to clean interrupt status
Signed-off-by: michael lee <igvtee@gmail.com>
Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 44401
2015-02-11 13:05:14 +00:00
John Crispin
07352ca2c7 ralink: fix hw status almost full not work on mt7620 and mt7621
the old FE_INT_STATUS register becomes two registers.
FE_INT_STATUS and INT_STATUS. so the hw status almost full
must change to read from FE_INT_STATUS register.
tx/rx done read from INT_STATUS register.

mt7620 datasheet define CNT_GDM1_AF at BIT(29).
but after test it should be BIT(13). why?

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44371
2015-02-09 19:34:49 +00:00
John Crispin
654bc380ec ralink: add 3.18 support
keep default as 3.14, mt7621 gic need to be ported to 3.18

Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 44349
2015-02-09 12:13:55 +00:00
Felix Fietkau
10c8d8c483 ramips: fix ethernet driver tx completion polling
- budget is decremented with completed frames, so don't check if done is
smaller
- ACK the interrupt before processing further frames to fix a small race
condition.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 44234
2015-02-01 08:20:37 +00:00
John Crispin
c56cf3edc7 ralink: check ethernet MAC address whether it is valid
The mac address usually write in factory block. but sometime user erase this block , the mac address will change to ff:ff:ff:ff:ff:ff.
This patch is purpose to fix this issue.

Signed-off-by: wengbj <linux.c@foxmail.com>

SVN-Revision: 44166
2015-01-28 12:06:22 +00:00
Felix Fietkau
bef829efa5 ralink: fix ethernet feature TSO not work
* fix TSO features verify on mt7621 firewrt board
* improve tx clean up. no need to access uncached
  memory. also use TX_DTX register instead of
  read tx ring DONE bit
* mt7621 need napi weight 64 to get more performance
* remove netif_receive_skb, after kernel version
  3.7 tcp4_gro_receive can handle tcp checksum.
  on rt2880 use iperf tcp LAN to WAN throughput test.
  with gro 135 Mbits/sec. without gro 80.4Mbits/sec.

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44118
2015-01-24 22:45:28 +00:00
Felix Fietkau
3ec294a825 ralink: init tx/rx vlan offload register for mt7621
on mt7621 don't have tx vlan vid registers.
so set FE_REG_FE_DMA_VID_BASE to 0.
set rx vlan offload register to disable.

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44117
2015-01-24 22:45:23 +00:00
Felix Fietkau
e634e07893 ralink: use fe_max_buf_size to get rx buffer size
the rx_buf_size now is 1534 when mtu is 1500.
the ethernet frame with vlan tag and FCS is 1522.
so the buffer is enough.

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44116
2015-01-24 22:45:16 +00:00
Felix Fietkau
5d4837706a ramips: fix ethernet vlan tx offload support check
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 44115
2015-01-24 22:45:10 +00:00
Felix Fietkau
9e7cc83edc ralink: fix ethernet rx buffer size breakage
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 44047
2015-01-18 20:17:41 +00:00
Felix Fietkau
a4460ca5a7 ralink: improve check rx sg function. use check frags instead of function call
Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44046
2015-01-18 20:17:28 +00:00
Felix Fietkau
b3ca42ec8f ralink: correct handle hardware rx 2bytes offset
Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44045
2015-01-18 20:17:18 +00:00
Felix Fietkau
03ea0cf6f1 ralink: use fe_reset to control all reset
Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44044
2015-01-18 20:17:07 +00:00
Felix Fietkau
50588ef192 ralink: improve tx_timeout function
* use default timeout value
* print more debug ring info
* move timeout reset function to workqueue

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44043
2015-01-18 20:16:50 +00:00
Felix Fietkau
a5b578c2d9 ralink: reduce access to uncached tx/rx dma ring buffer
Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 44041
2015-01-18 20:16:35 +00:00
Felix Fietkau
00db07be58 ramips: fix tx bql related race condition in the ethernet driver
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 43995
2015-01-17 13:13:55 +00:00
Felix Fietkau
acfc6c153b ramips: use a larger number of rx descriptors on mt7621 to work around rx overrun related packet loss
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 43953
2015-01-13 00:34:45 +00:00
Felix Fietkau
05d4b8c79b ramips: remove interrupt coalescing, it is unnecessary with napi polling and could reduce throughput
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 43952
2015-01-13 00:34:38 +00:00
Felix Fietkau
10c2bdbc4f ramips: disable ethernet checksum offloads for mt7621, they are broken
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 43811
2015-01-02 21:53:24 +00:00
Felix Fietkau
51b1f0ca5c ramips: use netif_receive_skb instead of napi_gro_receive when rx csum offload is unavailable
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 43810
2015-01-02 21:53:18 +00:00
John Crispin
e1f6d0dd0a ralink: fix tx vlan offload and hardware status
hardware status and tx vlan offload support on all targets
except rt5350. so i modify the IS_ENABLE condition only for
mt7621.
support mt7621 hardware status reference by SDK. but i don't
have mt7621. if not work just set mt7621 FE_REG_FE_COUNTER_BASE
to 0 to let software count.

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 43303
2014-11-19 09:19:57 +00:00
John Crispin
a5f562993e ralink: support netconsole
Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 43302
2014-11-19 09:19:49 +00:00
John Crispin
9c5dfd9546 ralink: improve tx vlan offload
hardware tx vlan offload only support max 16 vids
now use add/delete vlan interface to update vlan id table
when duplicate vlan id index detect.
disable hardware tx vlan offload support.

Signed-off-by: michael lee <igvtee@gmail.com>

SVN-Revision: 43301
2014-11-19 09:19:43 +00:00
John Crispin
caa29dea2c ralink: update ethernet driver to use new ralink_soc variable
Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43247
2014-11-14 16:52:52 +00:00
John Crispin
bf8bb5b611 ralink: fix the 10mbit bug on mt7621
a missing "val =" caused the AN bit in the phy0 reg to be flushed.

Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43246
2014-11-14 16:52:47 +00:00
John Crispin
194ca6127e ralink: improve mt7530 support
the SDK does a bit of extra init that we did not do yet when using an external mt7530.

Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43245
2014-11-14 16:52:42 +00:00
John Crispin
062828aee0 ralink: mt7621 has a different base addr for PVID
Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43243
2014-11-14 16:52:30 +00:00
John Crispin
86e3315ece ralink: add support for mt7621 ethernet
somehow all switch ports still come up as 10mbit.

Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43237
2014-11-12 14:54:50 +00:00
John Crispin
e399931376 ralink: fix mdio polling of external phys if only 1 phy exists
Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43235
2014-11-11 11:46:09 +00:00
John Crispin
9007e2244f ralink: mt7628 does not actually need this fix for the ethernet driver
Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43206
2014-11-07 10:51:00 +00:00
John Crispin
c32ad22e40 ralink: improve r43200
r43200 tries to detect if the fixup is needed or not. control the behaviour via
OF instead and disable unused ports.

Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43201
2014-11-06 12:19:53 +00:00
John Crispin
0cc0ba2b3a ralink: only apply ephy1 fixup if the phy is present
this broke e1700 lan2 as the fixup was applied to the phy inside the mt7530.

Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43200
2014-11-06 11:11:44 +00:00
John Crispin
f8404aaf9e ralink: add mt7628 support to the ethernet driver
Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43196
2014-11-06 09:31:19 +00:00
John Crispin
26504289df ralink: cleanup ethernet Makefile/Kconfig
Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43195
2014-11-06 09:31:09 +00:00
John Crispin
d524fe46fd ralink: move ethernet driver to files/
Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 43194
2014-11-06 09:31:03 +00:00
John Crispin
332b94fbd5 ralink: refresh patches
Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 39949
2014-03-18 19:21:56 +00:00