- 28 May, 2013 23 commits
-
-
Simon Horman authored
In order to mitigate ongoing incresase in the size of struct skbuff use 16 bit integer offsets rather than pointers for inner_*_headers. This appears to reduce the size of struct skbuff from 0xd0 to 0xc0 bytes on x86_64 with the following all unset. CONFIG_XFRM CONFIG_NF_CONNTRACK CONFIG_NF_CONNTRACK_MODULE NET_SKBUFF_NF_DEFRAG_NEEDED CONFIG_BRIDGE_NETFILTER CONFIG_NET_SCHED CONFIG_IPV6_NDISC_NODETYPE CONFIG_NET_DMA CONFIG_NETWORK_SECMARK Signed-off-by:
Simon Horman <horms@verge.net.au> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Florian Fainelli authored
The Broadcom BCM63xx PHY driver is for the SoC internal PHYs, flag these as internal PHY devices. Signed-off-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
Florian Fainelli authored
libphy currently always reports a PHY as an external transceiver from the ethtool output. This is inaccurate, because some drivers should be able to tell that a PHY device is an internal transceiver of an Ethernet MAC. Add a new flag (PHY_IS_INTERNAL) which can be set by PHY drivers just like other flags, and a corresponding helper: phy_is_internal() which can be used by networking drivers to query if a given PHY device is internal. Signed-off-by:
-
Florian Fainelli authored
Add a comment which explains the real meaning of XCVR_INTERNAL (PHY and Ethernet MAC in the same package/die) and XCVR_EXTERNAL (PHY and Ethernet MAC in a different package/die). Most if not all of the drivers setting their transceiver type already do it the way the comment describes it. Signed-off-by:
Ben Hutchings <bhutchings@solarflare.com> Signed-off-by:
-
Jingoo Han authored
The driver core clears the driver data to NULL after device_release or on probe failure, since commit 0998d063 (device-core: Ensure drvdata = NULL when no driver is bound). Thus, it is not needed to manually clear the device driver data to NULL. Signed-off-by:
Jingoo Han <jg1.han@samsung.com> Signed-off-by:
-
Jingoo Han authored
The driver core clears the driver data to NULL after device_release or on probe failure, since commit 0998d063 (device-core: Ensure drvdata = NULL when no driver is bound). Thus, it is not needed to manually clear the device driver data to NULL. Signed-off-by:
-
Jingoo Han authored
The driver core clears the driver data to NULL after device_release or on probe failure, since commit 0998d063 (device-core: Ensure drvdata = NULL when no driver is bound). Thus, it is not needed to manually clear the device driver data to NULL. Signed-off-by:
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Signed-off-by:
-
Jingoo Han authored
The driver core clears the driver data to NULL after device_release or on probe failure, since commit 0998d063 (device-core: Ensure drvdata = NULL when no driver is bound). Thus, it is not needed to manually clear the device driver data to NULL. Signed-off-by:
Nicolas Ferre <nicolas.ferre@atmel.com> Acked-by:
Rob Herring <rob.herring@calxeda.com> Acked-by:
Roland Stigge <stigge@antcom.de> Acked-by:
Mugunthan V N <mugunthanvnm@ti.com> Reviewed-by:
H Hartley Sweeten <hsweeten@visionengravers.com> Tested-by:
-
David S. Miller authored
Fabio Estevam says: ==================== This series provides support for controlling the PHY regulator in suspend/resume. Fabio Estevam (5): fec: Place the phy regulator in the private structure fec: Invert the order of error path sequence fec: Disable the PHY regulator on error and removal fec: Remove irqs first fec: Handle the regulator in suspend/resume ==================== Signed-off-by: -
Fabio Estevam authored
In order to save power, let's disable the regulator in the suspend function and enable it in resume. Tested on a mx28evk board. Signed-off-by:
Fabio Estevam <fabio.estevam@freescale.com> Signed-off-by:
-
Fabio Estevam authored
During probe the clocks are enabled prior than the acquiring the interrupts. In the remove function we need to do the opposite: first remove the interrupts and then disable the clocks. Signed-off-by:
-
Fabio Estevam authored
In the case of error during probe, disable the PHY regulator. Do the same in fec_drv_remove(). Signed-off-by:
-
Fabio Estevam authored
Currently when fec_enet_init fails it jumps to 'failed_init' error path, which will attemp to free the interrupts. This is wrong because at this point the interrupts have not even been acquired. Swap failed_init/failed_irq to fix the error path. Signed-off-by:
-
Fabio Estevam authored
Instead of using a local reg_phy structure, let's put it inside the private structure, so that we are able to have access to the regulator structure even when we are outside fec_probe(). This is in preparation for controlling the FEC PHY regulator in the suspend and resume functions. Signed-off-by:
-
Somnath Kotur authored
For padded packets, Lancer computes incorrect checksum. The workaround is to trim such packets. This workaround is mainly for IPv4 packets. Signed-off-by:
Kalesh AP <kalesh.purayil@emulex.com> Signed-off-by:
Somnath Kotur <somnath.kotur@emulex.com> Signed-off-by:
-
Somnath Kotur authored
In Lancer, packets that are 32 bytes or less may cause a transmit stall. The work-around is to pad such packets to a 36 byte length. Signed-off-by:
-
Somnath Kotur authored
Removing the be_cmd_get_fw_ver() query from be_get_drvinfo() and invoking the same from be_setup() and/or post firmware download. Signed-off-by:
-
Yuval Mintz authored
This changes the PCI power management scheme of the bnx2x driver to be similar to those of most network drivers - the driver will now changes the power state into D3hot whenever the driver will be removed, instead of whenever an interface is unloaded. This change enables the driver to access its eeprom via ethtool callbacks even when interfaces are unloaded (such access requires the function to be in D0active). Signed-off-by:
Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by:
Ariel Elior <ariele@broadcom.com> Signed-off-by:
Eilon Greenstein <eilong@broadcom.com> Signed-off-by:
-
Yuval Mintz authored
Implement the PCI shutdown callback to support un-orderly shutdown, i.e., shutdowns in which the remove callback will not be called. Due to the lack of this functionality, when an un-orderly shutdown occurred, it was possible for wake-on-lan to remain disabled. This is now fixed as the callback introduces the correct place in which `system_state' can be queried and wake-on-lan be configured accordingly. Signed-off-by:
-
Dmitry Kravkov authored
Commit 0eb43b4b "bnx2x, bnx2fc: Use per port max exchange resources" has changed the number of available FCoE exchanges, even in scenarios when some of the functions has no FCoE support; This needlessly degraded the available resources. Remedy this by calculating the maximal number of functions that may actually utilize said connection. Signed-off-by:
Dmitry Kravkov <dmitry@broadcom.com> CC: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by:
-
Ariel Elior authored
A PF should ack the firmware when receiving unknown messages through the VF's mailbox. This prevents the VF mailbox from being stuck in case of a VF sending a message unknown to the PF (e.g. VF with more advanced version). Signed-off-by:
-
Dmitry Kravkov authored
This adds the print of the PCI gen3 link speed (8GHz), as well as correcting the same print for 57712 boards (the print erroneously showed a 2.5GHz speed). Signed-off-by:
-
Ariel Elior authored
Hypervisor/Supervisor should set the VF's MAC prior to its load; Using a randomly generated MAC as a default is a bad practice. Signed-off-by:
-
- 26 May, 2013 6 commits
-
-
Joe Perches authored
case TCP_FIN_WAIT1 can also be simplified by reversing tests and adding breaks; Add braces after case and move automatic definitions. Signed-off-by:
Joe Perches <joe@perches.com> Signed-off-by:
-
Joe Perches authored
case TCP_SYN_RECV: can have another indentation level removed by converting if (acceptable) { ...; } else { return 1; } to if (!acceptable) return 1; ...; Reflow code and comments to fit 80 columns. Another pure cleanup patch. Signed-off-by:Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by:
-
Eric Dumazet authored
Remove one level of indentation 'introduced' in commit c3ae62af (tcp: should drop incoming frames without ACK flag set) if (true) { ... } @acceptable variable is a boolean. This patch is a pure cleanup. Signed-off-by:
Eric Dumazet <edumazet@google.com> Cc: Yuchung Cheng <ycheng@google.com> Signed-off-by:
-
Jiri Pirko authored
Now when upper device is changed, event is not propagated via RT Netlink to userspace. Userspace might never now about the change. Fix this by adding upper-device-change notifier event. Signed-off-by:
Jiri Pirko <jiri@resnulli.us> Signed-off-by:
-
Jingoo Han authored
Use the wrapper functions for getting and setting the driver data using platform_device instead of using dev_{get,set}_drvdata() with &pdev->dev, so we can directly pass a struct platform_device. Also, unnecessary dev_set_drvdata() is removed, because the driver core clears the driver data to NULL after device_release or on probe failure. Signed-off-by: -
Lorenzo Colitti authored
This adds the ability to send ICMPv6 echo requests without a raw socket. The equivalent ability for ICMPv4 was added in 2011. Instead of having separate code paths for IPv4 and IPv6, make most of the code in net/ipv4/ping.c dual-stack and only add a few IPv6-specific bits (like the protocol definition) to a new net/ipv6/ping.c. Hopefully this will reduce divergence and/or duplication of bugs in the future. Caveats: - Setting options via ancillary data (e.g., using IPV6_PKTINFO to specify the outgoing interface) is not yet supported. - There are no separate security settings for IPv4 and IPv6; everything is controlled by /proc/net/ipv4/ping_group_range. - The proc interface does not yet display IPv6 ping sockets properly. Tested with a patched copy of ping6 and using raw socket calls. Compiles and works with all of CONFIG_IPV6={n,m,y}. Signed-off-by:Lorenzo Colitti <lorenzo@google.com> Signed-off-by:
-
- 25 May, 2013 11 commits
-
-
David S. Miller authored
Shahed Shaikh says: ==================== This patch series includes following changes: * Support for spoof check configuration per VF using iproute2 tool. * Set HW mask for 8300 adapter in INT-x mode to stop generating interrupts. * Updated IRQ name for 8200 and 8300 Series adapter as per format used by other multiqueue drivers. * Remove qlcnic_config_npars module parameter. * Initialize trans_work and idc_aen_work at VF probe. * Convert netsted if-else into switch-case statement. * Change in diagnostics routine. * Modify reset recovery path in diag mode by providing diagnostics routines enough time to unwind before proceeding with reset recovery. * Implement GET_LED_STATUS command for 82xx adapter and refactor qlcnic_store_beacon() to split 8200 and 8300 specific calls. * Some cleanup and log enhancements ==================== Signed-off-by: -
Shahed Shaikh authored
Signed-off-by:
Shahed Shaikh <shahed.shaikh@qlogic.com> Signed-off-by:
-
Manish Chopra authored
Signed-off-by:
Manish Chopra <manish.chopra@qlogic.com> Signed-off-by:
-
Manish Chopra authored
Signed-off-by:
-
Himanshu Madhani authored
o GET_LED_STATUS command will be used by driver to get current beacon state from 82xx adapter. o Refactored qlcnic_store_beacon() to split 8200 and 8300 specific calls. Signed-off-by:
Himanshu Madhani <himanshu.madhani@qlogic.com> Signed-off-by:
-
Sony Chacko authored
Provide diagnostics routines enough time to unwind before proceeding with reset recovery. Signed-off-by:
Sony Chacko <sony.chacko@qlogic.com> Signed-off-by:
-
Sony Chacko authored
Test and set diagnostics mode bit before starting diagnostics tests. Stop diagnostics tests if adapter is resetting. Signed-off-by:
-
Jitendra Kalsaria authored
Signed-off-by:
Jitendra Kalsaria <jitendra.kalsaria@qlogic.com> Signed-off-by:
-
Sucheta Chakraborty authored
o work_struct should be initialized before cancel_delayed_work call to destroy it. Signed-off-by:
Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by:
-
Shahed Shaikh authored
qlcnic_config_npars module parameter is used to configure NPAR operational modes where NPAR function is used for PCI passthrough. Removing this paramter as PCI passthrough is not supported for NPAR functions. Signed-off-by:
-
Himanshu Madhani authored
o Updated IRQ name for 8200 and 8300 Series adapter as per format used by other multiqueue drivers. Signed-off-by:
-
