Qualcomm claims this improves the D-cache footprint. Origina commit message below:
From: Ben Menchaca <ben.menchaca@qca.qualcomm.com>
Date: Fri, 7 Jun 2013 10:57:28 -0500
Subject: [ag71xx] cluster/align structs for cache perf
Cluster the frequently used, per-packet structures in ag71xx near
to each other, and cacheline-align them. Some other re-ordering
occurred to move "warmer" structures near the per-packet structures.
Signed-off-by: Ben Menchaca <ben.menchaca@qca.qualcomm.com>
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Because the oldest supported kernel version on the ar71xx target is 4.4,
the condition that the kernel version is smaller than 4.2.0 is always
false. Remove the obsolete check from ag71xx_main.c to clean up the code
a bit.
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
Qualcomm claims this reduces cache misses. Original commit message below:
From: Ben Menchaca <ben.menchaca@qca.qualcomm.com>
Date: Tue, 11 Jun 2013 12:18:46 -0500
Subject: [ag71xx] reduce NAPI weight
In an attempt to increase our cache warmth, we are decreasing NAPI.
This increases the warmth of the reused SKBs.
Signed-off-by: Ben Menchaca <ben.menchaca@qca.qualcomm.com>
Signed-off-by: Rosen Penev <rosenp@gmail.com>
The motivation for this was misguided. It turns out tuning the NAPI weight could be useful for testing purposes. Therefore reverting.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
This reverts commit 13e5e47369.
This commit causes a severe regression in LAN->WAN routing performance
for several devices. This appears to be caused by the extra requirement
to validate the SKB checksum early in the rx path, which the ethernet
hardware does not do
Signed-off-by: Felix Fietkau <nbd@nbd.name>
This add support for kernel 4.9 to the ar71xx target.
It was compile tested with the generic, NAND and mikrotik subtarget.
Multiple members of the community tested it on their boards and did not
report any major problem so far.
Especially the NAND part received some changes to adapt to the new
kernel APIs. The serial driver hack used for the Arduino Yun was not
ported because the kernel changed there a lot.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
On a TL-WN710N, this patch increases iperf performance from ~92.5 to ~93.5 mbps. Keep in mind the WN710N is a 100mbps device. I expect greater numbers from gigabit devices.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
ag71xx_ethtool_set_ringparam() will return an uninitialized value on
success.
Found-by: Coverity Scan #1330877
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Should fix LAN speed issues on some devices. This is an updated version
of the previously reverted commit with the same name.
It improves the check for MACs connected to a built-in switch
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Fully reset the chip like on a full up/down, but without the PHY
statemachine restart.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 48228
no-op since 2.6.35
removed in Kernel 4.1
see https://lwn.net/Articles/380931/
Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
SVN-Revision: 46280
I don't see that we're in an atomic context so there's no need to
busy-wait. Therefore replace the delay with sleep calls.
See also Documentation/timers/timers-howto.txt. It states:
"In general, use of mdelay is discouraged and code should
be refactored to allow for the use of msleep."
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
SVN-Revision: 43539
Replace the fixed wait time of 1s with polling for BMCR_RESET
to be cleared on all PHYs.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
SVN-Revision: 43538
This improves performance when doing concurrent rx/tx on a single
ethernet MAC, e.g. when routing between VLANs.
Fixes#13072
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 42328
The r39147 commit introduces a regression: at lease on some routers
with ar8216 switch large packets get lost if 802.1q tagged port is
used on the interface connected to the aforementioned switch.
The r39147 changes code in the way so interface is set to accept
packets no longer than max ethernet frame length for a given mtu.
Unfortunately ar8216 has a feature: it sends two additional bytes
as a packet header and those this header needs to be added to the
max frame length. Otherwise long enough packets get lost.
The problem only manuifests itself if interface is used in vlan
tagged mode. If interface is untagged then ar8216's header fits
into space used by 802.1q tag and not packets are lost.
Include two additional bytes in the max frame length calculation
to fix the issue.
This patch is tested and works with Trendnet TEW-632BRP.
Signed-off-by Nikolay Martynov <mar.kolya@gmail.com>
Patchwork: http://patchwork.openwrt.org/patch/4656/
[juhosg:
- simplify the patch to include the additional bytes of the
switch header unconditionally,
- change subject and update commit message]
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 39219
The currently used bitmask of the maximum frame length field
is wrong for both models. On AR724x/AR933x the largest frame
size is 2047 bytes, on the AR934x it is 16383 bytes.
Make the MTU setup code model specific, and use the correct
bitmask for both models. Also change the value to the maximum.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 39148
Set the MAX_FRAME_LEN register to zero in ag71xx_hw_init()
and write the correct value into that from the ag71xx_open()
and ag71xx_fast_reset() functions.
Also recalculate the RX buffer size based on the actual
maximum frame length value to optimize memory allocation.
Additionaly, disallow to change the MTU value while the
interface it running.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 39147
