The skb is usually started by a padding which allows the protocols in the
network stack to add their headers in front of the payload. The skb can be
reallocated in case the preallocated padding is not large enough. This can for
example happen in the function __skb_cow which will check the requested extra
headroom and allocate more buffer when the requested headroom is bigger than
the available one. The extra buffer is aligned again to the multiple of the
NET_SKB_PAD of the target architecture.
The macro used to create the multiple of the NET_SKB_PAD is written in a way
which allows only values power two as alignment parameter. The currently used
value of 48 bytes can not be written as n ** 2 but as 2 ** 4 + 2 ** 5. The
extra buffer is therefore not always the multiple of 48 but can be 16, 64, 80,
128, 144 and so on. The generated values are also not monotonic (48 requested
bytes are mapped to 80 allocated bytes and 49 requested bytes are mapped to 64
allocated bytes).
These unexpected small values result in more reallocations of the buffer. This
was noticed prominently during tests between two QCA9558 720 MHz devices which
were connected via ethernet to PCs and had a HT40 802.11n 3x3 link between each
other. The throughput PC-to-PC during iperf TCP runs increased reliable from
186 Mibit/s to 214 Mibit/s in one direction and from 195 Mibit/s to 220 Mibit/s
in the other direction. This is a performance increase of ~14% just by reducing
the amount of reallocations.
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 37948
- introduces persistent null session to allow access to procedures without login
- implements session.login procedure to support user logins via json-rpc
SVN-Revision: 37941
SPARC builds are currently broken because we do not define the MUSL
dynamic linker name, fix that.
Signed-off-by: Florian Fainelli <florian@openwrt.org>
SVN-Revision: 37938
Currently, the module causes an oops at least on rt5350. These patches
have been accepted upstream at:
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net.git
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
SVN-Revision: 37931
- adds support for service triggering on config commit
- adds uci transaction support
- adds daemon reload support without loosing session data
- exports headers for use by external plugins
- drops LuCI2 in favor to an out of tree plugin
SVN-Revision: 37921
BCM5365 (and probably other older variants) use a different phy id, so
the phy driver never attached for them.
Fix this by adding the appropriate phy id to the fixup and the phy
driver.
Reported-by: Russell Senior <russell@personaltelco.net>
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 37906
This fixes the following error:
keyspan: Unknown symbol ezusb_fx1_ihex_firmware_download (err 0)
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 37899
This enables usage of the ath10k driver on OpenWrt. Only 2.0 PCI hardware
is supported. The ath10k firmware is currently only available
through github so use that as its source for now.
This goes build tested, not run time tested.
[Hauke: I did some minor changes]
Cc: ath10k@lists.infradead.org
Cc: Kalle Valo <kvalo@qca.qualcomm.com>
Cc: Kathy Giori <kgiori@qca.qualcomm.com>
Cc: Mathieu Olivari <mathieu@qca.qualcomm.com>
Cc: Jerome Proffit <jproffit@qca.qualcomm.com>
Cc: Sujith Manoharan <c_manoha@qca.qualcomm.com>
Cc: Matthew McClintock <mmcclint@qca.qualcomm.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 37898
b53_no_ops has no elements and b53_port_ops has one element, this makes
the code access some random memory when trying to access the mib
counter functions.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 37895
