aboutsummaryrefslogtreecommitdiff
path: root/drivers/net
AgeCommit message (Collapse)AuthorFilesLines
2014-07-15drivers: net: cpsw: disable coalesce when rx_coalesce_usecs is zeroMugunthan V N1-3/+7
instead of return error on zero rx_coalesce_usecs, disable coalesce Signed-off-by: Mugunthan V N <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15bonding: fix bond_option_mode_set warningNikolay Aleksandrov1-1/+2
During the conversion to "static" functions this one got left out, only its prototype was converted, thus resulting in: drivers/net/bonding//bond_options.c:674:5: warning: symbol 'bond_option_mode_set' was not declared. Should it be static? Fix it by making it static and also break the line in two as it was too long. CC: Stephen Hemminger <[email protected]> CC: Jay Vosburgh <[email protected]> CC: Veaceslav Falico <[email protected]> CC: Andy Gospodarek <[email protected]> CC: David S. Miller <[email protected]> Signed-off-by: Nikolay Aleksandrov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15bonding: permit enslaving interfaces without set_mac supportVeaceslav Falico1-9/+10
Currently we exit if the slave isn't the first slave, doesn't support mac address setting and fail_over_mac isn't FOM_ACTIVE. It's wrong because we only require ndo_set_mac_address in case bonding is in active-backup mode and FOM isn't FOM_ACTIVE. To fix this - only exit with an error if we're in a/b mode and have fail_over_mac != FOM_ACTIVE. Also, maintain current behaviour on the first slave (forcibly change fom to FOM_ACTIVE) to not break anyone's configuration. CC: Jay Vosburgh <[email protected]> CC: Andy Gospodarek <[email protected]> Signed-off-by: Veaceslav Falico <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15sfc: add extra RX drop counters for nodesc_trunc and noskb_dropEdward Cree9-10/+56
Added a counter rx_noskb_drop for failure to allocate an skb. Summed the per-channel rx_nodesc_trunc counters earlier so that they can be included in rx_dropped. Signed-off-by: Edward Cree <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15net: ppp: reset nextseq counter when enabling SC_MULTILINKChristoph Schulz1-0/+2
If using a demand-dialled PPP unit for a PPP multilink master, the pppd daemon needs to reset the sequence counter between two connections. This allows the daemon to reuse the PPP unit instead of destroying and recreating it. As there is no API to reset the counter, this patch resets the counter whenever the SC_MULTILINK flag is set. Signed-off-by: Christoph Schulz <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15bonding: add proper __rcu annotation for current_arp_slaveEric Dumazet2-7/+10
Using __rcu annotation actually helps to spot all accesses to bond->current_arp_slave are correctly protected, with LOCKDEP support. Signed-off-by: Eric Dumazet <[email protected]> Reviewed-by: Nikolay Aleksandrov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15bonding: add proper __rcu annotation for curr_active_slaveEric Dumazet4-45/+63
RCU was added to bonding in linux-3.12 but lacked proper sparse annotations. Using __rcu annotation actually helps to spot all accesses to bond->curr_active_slave are correctly protected, with LOCKDEP support. Signed-off-by: Eric Dumazet <[email protected]> Acked-by: Veaceslav Falico <[email protected]> Reviewed-by: Nikolay Aleksandrov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15bonding: use rcu_access_pointer() in bonding_show_mii_status()Eric Dumazet1-1/+2
curr_active_slave is rcu protected, and bonding_show_mii_status() only wants to check if pointer is NULL or not. Signed-off-by: Eric Dumazet <[email protected]> Acked-by: Veaceslav Falico <[email protected]> Reviewed-by: Nikolay Aleksandrov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15bonding: get rid of bond_option_active_slave_get()Eric Dumazet3-11/+16
Only keep bond_option_active_slave_get_rcu() helper. bond_fill_info() uses a new bond_option_active_slave_get_ifindex() helper. Signed-off-by: Eric Dumazet <[email protected]> Acked-by: Veaceslav Falico <[email protected]> Reviewed-by: Nikolay Aleksandrov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15Merge tag 'linux-can-next-for-3.17-20140715' of ↵David S. Miller2-41/+29
git://gitorious.org/linux-can/linux-can-next Marc Kleine-Budde says: ==================== pull-request: can-next 2014-07-15 this is a pull request of 4 patches for net-next/master. Prabhakar Lad contributes a patch that converts the c_can driver to use the devm api. The remaining four patches by Nikita Edward Baruzdin improve the SJA1000 driver with loopback testing support and introduce a new testing mode presume ack, for successful transmission even if no ACK is received. ==================== Signed-off-by: David S. Miller <[email protected]>
2014-07-15amd-xgbe: Remove the adjustments needed for fixed speedLendacky, Thomas2-14/+0
With the addition of entries in the phy speed/duplex settings array to support KR and KX mode, the work-around to add/remove baseT settings to run at a fixed speed is no longer needed. Signed-off-by: Tom Lendacky <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15phy: Expand phy speed/duplex settings arrayLendacky, Thomas1-1/+21
Expand the phy speed/duplex settings array to support more than just baseT features. This change adds entries to support the following additional speed/duplex/media types: SUPPORTED_10000baseKR_Full SUPPORTED_10000baseKX4_Full SUPPORTED_2500baseX_Full SUPPORTED_1000baseKX_Full Additionally, it changes the 10GbE baseT entry from using the hardcoded value 10000 to the SPEED_10000 define. Signed-off-by: Tom Lendacky <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15cxgb4/iw_cxgb4: work request logging featureHariprasad Shenai3-0/+22
This commit enhances the iwarp driver to optionally keep a log of rdma work request timining data for kernel mode QPs. If iw_cxgb4 module option c4iw_wr_log is set to non-zero, each work request is tracked and timing data maintained in a rolling log that is 4096 entries deep by default. Module option c4iw_wr_log_size_order allows specifing a log2 size to use instead of the default order of 12 (4096 entries). Both module options are read-only and must be passed in at module load time to set them. IE: modprobe iw_cxgb4 c4iw_wr_log=1 c4iw_wr_log_size_order=10 The timing data is viewable via the iw_cxgb4 debugfs file "wr_log". Writing anything to this file will clear all the timing data. Data tracked includes: - The host time when the work request was posted, just before ringing the doorbell. The host time when the completion was polled by the application. This is also the time the log entry is created. The delta of these two times is the amount of time took processing the work request. - The qid of the EQ used to post the work request. - The work request opcode. - The cqe wr_id field. For sq completions requests this is the swsqe index. For recv completions this is the MSN of the ingress SEND. This value can be used to match log entries from this log with firmware flowc event entries. - The sge timestamp value just before ringing the doorbell when posting, the sge timestamp value just after polling the completion, and CQE.timestamp field from the completion itself. With these three timestamps we can track the latency from post to poll, and the amount of time the completion resided in the CQ before being reaped by the application. With debug firmware, the sge timestamp is also logged by firmware in its flowc history so that we can compute the latency from posting the work request until the firmware sees it. Signed-off-by: Steve Wise <[email protected]> Signed-off-by: Hariprasad Shenai <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15cxgb4/iw_cxgb4: display TPTE on errorsHariprasad Shenai2-0/+67
With ingress WRITE or READ RESPONSE errors, HW provides the offending stag from the packet. This patch adds logic to log the parsed TPTE in this case. cxgb4 now exports a function to read a TPTE entry from adapter memory. Signed-off-by: Steve Wise <[email protected]> Signed-off-by: Hariprasad Shenai <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15cxgb4/iw_cxgb4: use firmware ord/ird resource limitsHariprasad Shenai4-0/+25
Advertise a larger max read queue depth for qps, and gather the resource limits from fw and use them to avoid exhaustinq all the resources. Design: cxgb4: Obtain the max_ordird_qp and max_ird_adapter device params from FW at init time and pass them up to the ULDs when they attach. If these parameters are not available, due to older firmware, then hard-code the values based on the known values for older firmware. iw_cxgb4: Fix the c4iw_query_device() to report these correct values based on adapter parameters. ibv_query_device() will always return: max_qp_rd_atom = max_qp_init_rd_atom = min(module_max, max_ordird_qp) max_res_rd_atom = max_ird_adapter Bump up the per qp max module option to 32, allowing it to be increased by the user up to the device max of max_ordird_qp. 32 seems to be sufficient to maximize throughput for streaming read benchmarks. Fail connection setup if the negotiated IRD exhausts the available adapter ird resources. So the driver will track the amount of ird resource in use and not send an RI_WR/INIT to FW that would reduce the available ird resources below zero. Signed-off-by: Steve Wise <[email protected]> Signed-off-by: Hariprasad Shenai <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15iw_cxgb4: Detect Ing. Padding Boundary at run-timeHariprasad Shenai2-0/+4
Updates iw_cxgb4 to determine the Ingress Padding Boundary from cxgb4_lld_info, and take subsequent actions. Signed-off-by: Steve Wise <[email protected]> Signed-off-by: Hariprasad Shenai <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15net: rtnetlink - make create_link take name_assign_typeTom Gundersen1-3/+8
This passes down NET_NAME_USER (or NET_NAME_ENUM) to alloc_netdev(), for any device created over rtnetlink. v9: restore reverse-christmas-tree order of local variables Signed-off-by: Tom Gundersen <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15net: set name_assign_type in alloc_netdev()Tom Gundersen45-58/+76
Extend alloc_netdev{,_mq{,s}}() to take name_assign_type as argument, and convert all users to pass NET_NAME_UNKNOWN. Coccinelle patch: @@ expression sizeof_priv, name, setup, txqs, rxqs, count; @@ ( -alloc_netdev_mqs(sizeof_priv, name, setup, txqs, rxqs) +alloc_netdev_mqs(sizeof_priv, name, NET_NAME_UNKNOWN, setup, txqs, rxqs) | -alloc_netdev_mq(sizeof_priv, name, setup, count) +alloc_netdev_mq(sizeof_priv, name, NET_NAME_UNKNOWN, setup, count) | -alloc_netdev(sizeof_priv, name, setup) +alloc_netdev(sizeof_priv, name, NET_NAME_UNKNOWN, setup) ) v9: move comments here from the wrong commit Signed-off-by: Tom Gundersen <[email protected]> Reviewed-by: David Herrmann <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15net: mvpp2: Fix a typo in the licenseEzequiel Garcia1-1/+1
The proper string for this license is "GPL v2", instead of "GPLv2". This commit fixes that. Reported-by: Arnd Bergmann <[email protected]> Signed-off-by: Ezequiel Garcia <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2014-07-15wireless: zd1211rw: new url for fw, remove experimentalXose Vazquez Perez1-2/+2
Cc: Daniel Drake <[email protected]> Cc: Ulrich Kunitz <[email protected]> Cc: John W. Linville <[email protected]> Cc: [email protected] Signed-off-by: Xose Vazquez Perez <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15b43: N-PHY: add missing TX gain table for radio 0x2057 rev 5Rafał Miłecki1-8/+50
Signed-off-by: Rafał Miłecki <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15brcmfmac: add brcmf_p2p_detach() call in brcmf_cfg80211_detach()Arend van Spriel1-0/+1
The function brcmf_p2p_detach() was only called in error flow of the brcmf_cfg80211_attach() routine, but it also needs to be called upon brcmf_cfg80211_detach(). Reviewed-by: Hante Meuleman <[email protected]> Reviewed-by: Pieter-Paul Giesberts <[email protected]> Signed-off-by: Arend van Spriel <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15brcmfmac: rework wiphy structure setupArend van Spriel1-268/+298
Instead of waiting for IFF_UP of the primary net device to determine the band and channel information of the wiphy structure, this is now done during driver initialization in brcmf_cfg80211_attach(). The channel information is obtained from the device and the 2G band is updated when 40MHz bandwidth is enabled for that band. Before this change the band and channel objects were common between multiple brcmfmac devices in the system, which make that information rather unreliable. That is also fixed with this reworked implementation. Reviewed-by: Hante Meuleman <[email protected]> Reviewed-by: Pieter-Paul Giesberts <[email protected]> Reviewed-by: Franky (Zhenhui) Lin <[email protected]> Reviewed-by: Daniel (Deognyoun) Kim <[email protected]> Signed-off-by: Arend van Spriel <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15brcmfmac: moving some functions aroundArend van Spriel1-144/+144
Just reordering the functions in preparation of subsequent changes. Signed-off-by: Arend van Spriel <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15brcmfmac: introduce feature and quirk handlingArend van Spriel6-27/+244
Introducing a new source module that will be responsible for identifying features and quirks related to the device being handled. Reviewed-by: Hante Meuleman <[email protected]> Reviewed-by: Franky (Zhenhui) Lin <[email protected]> Reviewed-by: Daniel (Deognyoun) Kim <[email protected]> Reviewed-by: Pieter-Paul Giesberts <[email protected]> Signed-off-by: Arend van Spriel <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15brcmfmac: move attach and detach functions in wl_cfg80211.cArend van Spriel1-108/+107
Preparing for another patch move the functions in separate commit. Reviewed-by: Hante Meuleman <[email protected]> Reviewed-by: Daniel (Deognyoun) Kim <[email protected]> Reviewed-by: Pieter-Paul Giesberts <[email protected]> Signed-off-by: Arend van Spriel <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15brcmfmac: Make firmware path a module parameterDaniel Kim4-20/+41
This patch makes firmware path a module parameter so that firmware and nvram files can be loaded from the specified path. Signed-off-by: Daniel Kim <[email protected]> Signed-off-by: Arend van Spriel <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15brcmfmac: rework debugfs functions in the driverArend van Spriel4-235/+207
Reworked the debugfs functions in the driver making it easier for other driver parts to add a debugfs entry and keeping the information they want to expose in debugfs private, ie. not in a header. This is accomplished by providing the function brcmf_debugfs_add_entry() in which the caller provides a read function in which they provide the content. The debugfs function will take care of creating the debugfs entry and cleaning up upon removal. Reviewed-by: Hante Meuleman <[email protected]> Reviewed-by: Franky (Zhenhui) Lin <[email protected]> Reviewed-by: Daniel (Deognyoun) Kim <[email protected]> Reviewed-by: Pieter-Paul Giesberts <[email protected]> Signed-off-by: Arend van Spriel <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15brcmfmac: make use of seq_file API for debugfs entriesArend van Spriel2-201/+170
The use of seq_file simplifies the debugfs code. Simpler is better. Reviewed-by: Hante Meuleman <[email protected]> Reviewed-by: Franky (Zhenhui) Lin <[email protected]> Reviewed-by: Pieter-Paul Giesberts <[email protected]> Signed-off-by: Arend van Spriel <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15brcmfmac: Cleanup used device IDs.Hante Meuleman5-85/+107
This patch cleans up used broadcom IDs, device IDs for all the bus layers and uses consistent naming for all IDs. Reviewed-by: Arend Van Spriel <[email protected]> Reviewed-by: Franky (Zhenhui) Lin <[email protected]> Reviewed-by: Pieter-Paul Giesberts <[email protected]> Reviewed-by: Daniel (Deognyoun) Kim <[email protected]> Signed-off-by: Hante Meuleman <[email protected]> Signed-off-by: Arend van Spriel <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15mwifiex: correct a typo in mwifiex_ret_tdls_operBing Zhao1-1/+1
This patch fixes this typo. Reported-by: Paul Stewart <[email protected]> Signed-off-by: Bing Zhao <[email protected]> Signed-off-by: Avinash Patil <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15mwifiex: declare sta_ptr in smaller scopeBing Zhao1-1/+2
sta_ptr is used only in an 'if' branch in this function. Move it to the smaller scope where it is used. Reported-by: Paul Stewart <[email protected]> Signed-off-by: Bing Zhao <[email protected]> Signed-off-by: Avinash Patil <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15mwifiex: define TDLS idle timeout macro with unitsBing Zhao2-2/+2
The unit of this timeout is in seconds. Reported-by: Paul Stewart <[email protected]> Signed-off-by: Bing Zhao <[email protected]> Signed-off-by: Avinash Patil <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15mwifiex: fix a cut-n-paste error in adhoc-startBing Zhao1-1/+1
The 'else if' branch never gets the chance as its condition matches 'if' branch's. Reported-by: David Binderman <[email protected]> Signed-off-by: Amitkumar Karwar <[email protected]> Signed-off-by: Bing Zhao <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15mwifiex: fix corner case system hang issueAmitkumar Karwar1-5/+11
Sometimes pending internal scan commands are delayed to give preference to Tx traffic. 'scan_processing' flag has been checked at the beginning of delay timer routine to know if in the meantime scan operation has been cancelled. There is a corner case where pending scan commands are emptied after scan_processing flag check is passed. In this case wrong pointer returned by list_first_entry() is passed to list_del() which causes system hang. This patch fixes the issue by adding list_empty() check. Signed-off-by: Amitkumar Karwar <[email protected]> Signed-off-by: Bing Zhao <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15mwifiex: access rx_reorder_tbl_ptr only while holding lockAmitkumar Karwar1-7/+20
This patch fixes a bug in which rx_reorder_tbl_ptr is accessed without holding spinlock at few places. Signed-off-by: Amitkumar Karwar <[email protected]> Signed-off-by: Bing Zhao <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wireless: rt2x00: add new rt2800usb devicesXose Vazquez Perez1-0/+4
0x0b05 0x17e8 RT5372 USB 2.0 bgn 2x2 ASUS USB-N14 0x0411 0x0253 RT5572 USB 2.0 abgn 2x2 BUFFALO WLP-U2-300D 0x0df6 0x0078 RT???? Sitecom N300 Cc: Ivo van Doorn <[email protected]> Cc: Helmut Schaa <[email protected]> Cc: John W. Linville <[email protected]> Cc: [email protected] Cc: [email protected] Signed-off-by: Xose Vazquez Perez <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore/wl18xx/wl12xx: convert bitmaps to unsigned longsEliad Peller6-17/+17
The *_bit operations expect unsigned longs. Instead of casting the pointers, simply define various bitmaps as unsigned long (instead of u32). Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wl18xx: make sure fw_status->priv exists before derefEliad Peller1-2/+12
In some corner cases with specific timings, we might try dequeueing tx before we got information about the link status (e.g. due to recovery during tx). Instead of NULL dereference, assume all the links in this case have low priorities. Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wl18xx: change the number of WLAN addrs per chipArik Nemtsov1-1/+1
Each 18xx chip contains only 2 real MAC addresses usable for WLAN, forcing us to use the LAA bit approach to obtain a third MAC address. Signed-off-by: Arik Nemtsov <[email protected]> Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore: don't switch channels on disconnected STA vifsArik Nemtsov1-0/+7
Sending the FW a channel switch command on a disconnected vif may result in a beacon loss event. Avoid this corner case. Signed-off-by: Arik Nemtsov <[email protected]> Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore: register vendor commandsEliad Peller1-0/+4
All the smart config code is in place now, so register the relevant vendor commands. Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore: increase max roc duration to 30 secondsEliad Peller1-1/+1
we don't have any actual limitation in the driver, so increase it arbitrarily to 30 seconds. The long ROC is needed for the smart config.flow. Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore/wl18xx: handle smart config eventsEliad Peller5-1/+93
add defintions and handling for smart config events (SMART_CONFIG_SYNC_EVENT_ID and SMART_CONFIG_DECODE_EVENT_ID) parse the relevant info and send it to userspace as vendor event. Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore: handle smart config vendor commandsEliad Peller3-1/+221
userspace can ask to perform various smart config actions via custom vendor commands. Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore/wl18xx: add smart config commandsEliad Peller5-1/+127
These commands configures the fw to set key, enter smart config mode, and exit it. Add relevant hw ops as well. Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore: add smart config definitionsEliad Peller2-0/+17
Add definitions for the smart config commands. Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore: use correct LAA bitEliad Peller1-1/+1
The LAA bit is second bit of the MSB, not of the third byte. Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wl18xx: fix last tx rate calculationEliad Peller1-2/+5
The last tx rate calculation didn't take into account the different indices of 11a and 11g rates tables. Add the required alignment (count only from the first 11a rate in case of 11a) Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2014-07-15wlcore: user smaller sqn padding for GEMEliad Peller3-4/+13
On recovery, we increase the current seq num by WL1271_TX_SQN_POST_RECOVERY_PADDING in order to compensate for packets we might have missed during recovery. It seems that some GEM APs have issues when the gap is too big, so use a smaller padding in this case. Signed-off-by: Eliad Peller <[email protected]> Signed-off-by: John W. Linville <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.