kernel/linux.git/drivers/net/ethernet, branch v3.15.2

sfc: PIO:Restrict to 64bit arch and use 64-bit writes.

2014-06-26T19:17:33+00:00

[ Upstream commit daf37b556e437ec1ea1a597dcfeff338068380e1 ] Fixes:ee45fd92c739 ("sfc: Use TX PIO for sufficiently small packets") The linux net driver uses memcpy_toio() in order to copy into the PIO buffers. Even on a 64bit machine this causes 32bit accesses to a write- combined memory region. There are hardware limitations that mean that only 64bit naturally aligned accesses are safe in all cases. Due to being write-combined memory region two 32bit accesses may be coalesced to form a 64bit non 64bit aligned access. Solution was to open-code the memory copy routines using pointers and to only enable PIO for x86_64 machines. Not tested on platforms other than x86_64 because this patch disables the PIO feature on other platforms. Compile-tested on x86 to ensure that works. The WARN_ON_ONCE() code in the previous version of this patch has been moved into the internal sfc debug driver as the assertion was unnecessary in the upstream kernel code. This bug fix applies to v3.13 and v3.14 stable branches. Signed-off-by: Shradha Shah Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

sh_eth: fix SH7619/771x support

2014-06-26T19:17:33+00:00

[ Upstream commit d8b0426af5b67973585712c9af36b86f6ea97815 ] Commit 4a55530f38e4 (net: sh_eth: modify the definitions of register) managed to leave out the E-DMAC register entries in sh_eth_offset_fast_sh3_sh2[], thus totally breaking SH7619/771x support. Add the missing entries using the data from before that commit. Signed-off-by: Sergei Shtylyov Acked-by: Yoshihiro Shimoda Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

sh_eth: use RNC mode for packet reception

2014-06-26T19:17:33+00:00

[ Upstream commit 530aa2d0d9d55ab2775d47621ddf4b5b15bc1110 ] The current behaviour of the sh_eth driver is not to use the RNC bit for the receive ring. This means that every packet recieved is not only generating an IRQ but it also stops the receive ring DMA as well until the driver re-enables it after unloading the packet. This means that a number of the following errors are generated due to the receive packet FIFO overflowing due to nowhere to put packets: net eth0: Receive FIFO Overflow Since feedback from Yoshihiro Shimoda shows that every supported LSI for this driver should have the bit enabled it seems the best way is to remove the RMCR default value from the per-system data and just write it when initialising the RMCR value. This is discussed in the message (http://www.spinics.net/lists/netdev/msg284912.html). I have tested the RMCR_RNC configuration with NFS root filesystem and the driver has not failed yet. There are further test reports from Sergei Shtylov and others for both the R8A7790 and R8A7791. There is also feedback fron Cao Minh Hiep[1] which reports the same issue in (http://comments.gmane.org/gmane.linux.network/316285) showing this fixes issues with losing UDP datagrams under iperf. Tested-by: Sergei Shtylyov Signed-off-by: Ben Dooks Acked-by: Yoshihiro Shimoda Acked-by: Simon Horman Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman

net: ec_bhf: Add runtime dependencies

2014-06-03T00:02:28+00:00

The ec_bhf driver is specific to the Beckhoff CX embedded PC series. These are based on Intel x86 CPU. So we can add a dependency on X86, with COMPILE_TEST as an alternative to still allow for broader build-testing. Signed-off-by: Jean Delvare Cc: Darek Marcinkiewicz Cc: David S. Miller Signed-off-by: David S. Miller

net/mlx4_core: Reset RoCE VF gids when guest driver goes down

2014-05-30T23:57:52+00:00

Reset the GIDs assigned to a VF in the port RoCE GID table when that guest goes down (either crashes or goes down cleanly). As part of this fix, we refactor the RoCE gid table driver copy, moving it to the mlx4_port_info structure (together with the MAC and VLAN tables). As with the MAC and VLAN tables, we now use a mutex per port for the GID table so that modifying the driver copy and modifying the firmware copy of a port GID table becomes an atomic operation (thus avoiding driver-copy/FW-copy mismatches). Signed-off-by: Jack Morgenstein Signed-off-by: Or Gerlitz Signed-off-by: David S. Miller

emac: aggregation of v1-2 PLB errors for IER register

2014-05-30T23:29:57+00:00

Aggreagation of version 1-2 because of version 1 can hit PLB errors too. If it's not set so we missing events for PLB bits and driver can't process those interrupts. Signed-off-by: Ivan Mikhaylov Signed-off-by: David S. Miller

emac: add missing support of 10mbit in emac/rgmii

2014-05-30T23:29:57+00:00

In chips of emac/rgmii b'000' for 0/1 channel isn't suitable which resulted in non working network interface in this mode. Signed-off-by: Ivan Mikhaylov Signed-off-by: David S. Miller

qlcnic: info leak in qlcnic_dcb_peer_app_info()

2014-05-24T18:01:17+00:00

This function is called from dcbnl_build_peer_app(). The "info" struct isn't initialized at all so we disclose 2 bytes of uninitialized stack data. We should clear it before passing it to the user. Fixes: 48365e485275 ('qlcnic: dcb: Add support for CEE Netlink interface.') Signed-off-by: Dan Carpenter Signed-off-by: David S. Miller

stmmac: Remove unbalanced clk_disable call

2014-05-22T19:16:23+00:00

The stmmac_open call was calling clk_disable_unprepare on phy init failure, but it never calls clk_prepare_enable, this causes a WARN_ON in the clk framework to trigger if for some reason phy init fails. Signed-off-by: Hans de Goede Acked-by: Giuseppe Cavallaro Acked-by: Chen-Yu Tsai Signed-off-by: David S. Miller

bnx2x: Convert return 0 to return rc

2014-05-19T00:51:03+00:00

These "return 0;" uses seem wrong as there are rc variables where error return values are set but unused. Signed-off-by: Joe Perches Acked-by: Dmitry Kravkov Signed-off-by: David S. Miller