summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2017-06-21rtlwifi: Revise special packet notification to be readable format.Ping-Ke Shih
We extend types of special packets to a enumeration from boolean value. Signed-off-by: Ping-Ke Shih <pkshih@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Yan-Hsuan Chuang <yhchuang@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rtlwifi: Add TX report and disable key to force wait until report acked.Ping-Ke Shih
When using EAPOL to do a PTK rekey, there is a possible race condition. When msg 3/4 is received, the supplicant will send msg 4/4 and install the new key immediately; however, the driver must make sure that msg 4/4 is sent before installing the new key. We use TX report to ensure it is sent. Signed-off-by: Ping-Ke Shih <pkshih@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Yan-Hsuan Chuang <yhchuang@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rtlwifi: extend debug_comp to u64Ping-Ke Shih
The number of debugging conditions now exceeds the capabilities of a 32-bit word. Signed-off-by: Ping-Ke Shih <pkshih@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Yan-Hsuan Chuang <yhchuang@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rtlwifi: Correct power save capability while init mac80211Ping-Ke Shih
Since the driver itself will enter power saving mode dynamically according to the traffic, we set hw capability SUPPORTS_PS and SUPPORTS_DYNAMIC_PS in case of fwctrl_lps. The process IEEE80211_CONF_CHANGE_PS in op_config is used by SW-LPS only, so we add constraints to avoid errors. Signed-off-by: Ping-Ke Shih <pkshih@realtek.com> Signed-off-by: vincent_fann <vincent_fann@realtek.com> Signed-off-by: shaofu <shaofu@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rtlwifi: btcoex: rtl8723be: fix ant_sel not workPing-Ke Shih
To make ant_sel work, we should call power_on_setting to set antenna correctly. Signed-off-by: Ping-Ke Shih <pkshih@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Yan-Hsuan Chuang <yhchuang@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rtlwifi: btcoex: set correct interface type and parameter.Ping-Ke Shih
This commit fixies two problems. The first one is interface types (e.g. PCI) that are used to switch antenna, and the second is to add wifi_only parameter to give correct state. Signed-off-by: Ping-Ke Shih <pkshih@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Yan-Hsuan Chuang <yhchuang@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rtlwifi: btcoex: call bind to setup btcoexPing-Ke Shih
New btcoex add a function 'bind' to connect adapter, so we should call it during initializing. Signed-off-by: Ping-Ke Shih <pkshih@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Yan-Hsuan Chuang <yhchuang@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rsi: add tx frame for common device configurationPrameela Rani Garnepudi
After successful loading of firmware, a CARD READY indication is received by host. Common device configuration parameters are sent to the device after this. It includes information like device operating mode (Wi-Fi alone or BT coex), power save related parameters, GPIO information etc. As device supports BT coex, this frame is send in COEX queue initially. Based on the operating mode, CARD READY indication is received from each protocol module in firmware i.e. WLAN, BT. Signed-off-by: Prameela Rani Garnepudi <prameela.j04cs@gmail.com> Signed-off-by: Amitkumar Karwar <amit.karwar@redpinesignals.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rsi: configure new boot parameters to devicePrameela Rani Garnepudi
Boot parameters are changed in new firmware. Also three new sdio sleep parameters are added for ultra low power save. Signed-off-by: Prameela Rani Garnepudi <prameela.j04cs@gmail.com> Signed-off-by: Amitkumar Karwar <amit.karwar@redpinesignals.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rsi: receive path enhancement for RS9113Prameela Rani Garnepudi
RS9113 chipset supports Coex feature. Initial frame exchanges during device initialization happens on coex queue. This patch adds the handling for coex queue. Signed-off-by: Prameela Rani Garnepudi <prameela.j04cs@gmail.com> Signed-off-by: Amitkumar Karwar <amit.karwar@redpinesignals.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rsi: Register interrupt handler before firmware loadPrameela Rani Garnepudi
Before firmware load, sometimes false interrupts are received. System hang is observed if interrupt handler is not registered to receive these interrupts. Hence interrupt handler registration is moved before firmware load. We will drop these false interrupts as these are not from the device. Signed-off-by: Prameela Rani Garnepudi <prameela.j04cs@gmail.com> Signed-off-by: Amitkumar Karwar <amit.karwar@redpinesignals.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rsi: use enum for FSM statesAmitkumar Karwar
Currently macros are used for FSM states. We will replace it with enum so that new state can be added easily without worrying about macro value. Signed-off-by: Amitkumar Karwar <amit.karwar@redpinesignals.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rsi: add usb RS9113 chipset supportPrameela Rani Garnepudi
usb device id and vendor id for RS9113 chipset are included. So it can be used through usb interface. Signed-off-by: Prameela Rani Garnepudi <prameela.j04cs@gmail.com> Signed-off-by: Amitkumar Karwar <amit.karwar@redpinesignals.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21mwifiex: debugfs: remove redunant check of mwifiex_dfs_dirShawn Lin
debugfs_remove already check mwifiex_dfs_dir, so remove it. Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21mwifiex: fixes the unexpected be printed log by defaultCaesar Wang
This patch uses WARN level is not printed by default. In some cases, some boards have always met the unused log be printed as follows. ... [23193.523182] mwifiex_pcie 0000:01:00.0: mwifiex_get_cfp: cannot find cfp by band 2 & channel=13 freq=0 [23378.633684] mwifiex_pcie 0000:01:00.0: mwifiex_get_cfp: cannot find cfp by band 2 & channel=13 freq=0 Due to we used the wifi default area was US and didn't support 12~14 channels. As Frequencies: * 2412 MHz [1] (30.0 dBm) * 2417 MHz [2] (30.0 dBm) * 2422 MHz [3] (30.0 dBm) * 2427 MHz [4] (30.0 dBm) * 2432 MHz [5] (30.0 dBm) * 2437 MHz [6] (30.0 dBm) * 2442 MHz [7] (30.0 dBm) * 2447 MHz [8] (30.0 dBm) * 2452 MHz [9] (30.0 dBm) * 2457 MHz [10] (30.0 dBm) * 2462 MHz [11] (30.0 dBm) * 2467 MHz [12] (disabled) * 2472 MHz [13] (disabled) * 2484 MHz [14] (disabled) Also, as the commit 1b499cb72f26b ("mwifiex: disable channel filtering feature in firmware"), it proved to be a feature to get better scan result from overlapping channel. Even there could be AP from overlapping channel (might be 12/13/14 in this case), it will be filtered depend on reg domain rules. e.g: ... if (ch->flags & IEEE80211_CHAN_DISABLED) continue; So it should not been an ERROR, use the WARN level to instead it for now. Signed-off-by: Caesar Wang <wxt@rock-chips.com> Acked-by: Xinming Hu <huxm@marvell.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21rtlwifi: fix REG_USTIME_TSF register definitionKevin Lo
The REG_USTIME_TSF (US Time Tuning for TSF) definition of Realtek chips should be 0x55C. Signed-off-by: Kevin Lo <kevlo@kevlo.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-21brcmfmac: Use separate firmware for revision 0 of the brcm43430 chipHans de Goede
The brcm43430 chip needs different firmware files for chip revision 0 and 1. The file currently in linux-firmware is for revision 1 only. This commit makes brcmfmac request brcmfmac43430a0-sdio.bin instead of brcmfmac43430-sdio.bin for revision 0 chips. Note that the behavior for revision 1 chips is not changed, ideally those would load brcmfmac43430a1-sdio.bin, but that will break existing setups. Signed-off-by: Hans de Goede <hdegoede@redhat.com> Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13wlcore: spi: remove unnecessary variableGustavo A. R. Silva
Remove unnecessary variable and refactor the code. Addresses-Coverity-ID: 1365000 Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13wlcore: add wl1285 compatibleSebastian Reichel
Motorola Droid 4 uses a WL 1285C. With differences between chips not being public let's add explicit binding for wl1285 instead of relying on wl1283 being very similar. Reviewed-by: Rob Herring <robh@kernel.org> Acked-by: Kalle Valo <kvalo@codeaurora.org> Acked-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13mwifiex: Replace semaphore async_sem with mutexBinoy Jayan
The semaphore 'async_sem' is used as a simple mutex, so it should be written as one. Semaphores are going away in the future. Signed-off-by: Binoy Jayan <binoy.jayan@linaro.org> Reviewed-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13rtlwifi: btcoex: let btcoex get wifi rssi and link statusYan-Hsuan Chuang
Instead of rssi status, the btcoex also needs to get the link status of the wifi. In addition, some of the rssi status can be merged into link status. Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Ping-Ke Shih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13rtlwifi: btcoex: remove unused display functionsYan-Hsuan Chuang
These display functions are useless and will not be called in the future. Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Ping-Ke Shih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13rtlwifi: btcoex: bind BT coex information with wifi driverYan-Hsuan Chuang
When initializing, gather BT information in struct btcoexist and provide them to wifi driver. Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Ping-Ke Shih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13rtlwifi: btcoex: hook external functions for newer chipsYan-Hsuan Chuang
Hook the external functions for newer ICs such as 8821a and 8192e. Rename ex_halbtc8192e2ant_halt_notify to ex_btc8192e2ant_halt_notify. Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Ping-Ke Shih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13rtlwifi: btcoex: add settings before the hardware is readyYan-Hsuan Chuang
When the hardware is turned on and in the initialization stage, the PTA circuit is unstable. Hence we need to force some hardware settings to make sure the PTA circuit work correctly, otherwise it may affect the user's experience. Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Ping-Ke Shih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13rtlwifi: btcoex: hook external PnP notify by chip for wifi driverYan-Hsuan Chuang
Hook the chip-specific PnP notify functions for the wifi driver to notify btcoexistence. Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Ping-Ke Shih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13rtlwifi: btcoex: settings before firmware is downloadedYan-Hsuan Chuang
The btcoex is sometimes unstable because there are some unexpected behaviors before the firmware has been downloaded successfully. Therefore we force the antenna path settings to avoid this, then let the firmware control the btcoexistence when the firmware is ready. Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Ping-Ke Shih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13rtlwifi: btcoex: rename ex_halbtc*ant to ex_btc*antYan-Hsuan Chuang
These external functions are for BT-coexistence, so remove the "hal" prefix for consistancy. Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Ping-Ke Shih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13rtlwifi: btcoex: add macros to check chip typeYan-Hsuan Chuang
For some external functions that have hardware dependency, we need to know the type of the hardware before invoking them. Signed-off-by: Yan-Hsuan Chuang <yhchuang@realtek.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Ping-Ke Shih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13qtnfmac: check band before allocating cmd_skb to avoid resource leakColin Ian King
The current code allocates cmd_skb and then will leak this if band->band is an illegal value. It is simpler to sanity check the band first before allocating cmd_skb so that we don't have to free cmd_skb if an invalid band occurs. Detected by CoverityScan, CID#1437561 ("Resource Leak") Signed-off-by: Colin Ian King <colin.king@canonical.com> Reviewed-by: Igor Mitsyanko <igor.mitsyanko.os@quantenna.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13mwifiex: make function mwifiex_ret_pkt_aggr_ctrl staticColin Ian King
function mwifiex_ret_pkt_aggr_ctrl can be made static as it does not need to be in global scope. Cleans up sparse warning: "symbol 'mwifiex_ret_pkt_aggr_ctrl' was not declared. Should it be static?" Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13brcmfmac: Fix glom_skb leak in brcmf_sdiod_recv_chainPeter S. Housel
An earlier change to this function (3bdae810721b) fixed a leak in the case of an unsuccessful call to brcmf_sdiod_buffrw(). However, the glom_skb buffer, used for emulating a scattering read, is never used or referenced after its contents are copied into the destination buffers, and therefore always needs to be freed by the end of the function. Fixes: 3bdae810721b ("brcmfmac: Fix glob_skb leak in brcmf_sdiod_recv_chain") Fixes: a413e39a38573 ("brcmfmac: fix brcmf_sdcard_recv_chain() for host without sg support") Cc: stable@vger.kernel.org # 4.9.x- Signed-off-by: Peter S. Housel <housel@acm.org> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13brcmfmac: Fix grammar issue in error messageMartin Michlmayr
Fix grammar issue in error message about ISO3166. Signed-off-by: Martin Michlmayr <tbm@cyrius.com> Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13brcmfmac: add scheduled scan support for specified BSSIDsArend Van Spriel
Add support to handle scheduled scan request containing BSSID in the matchsets. The firmware can send event upon finding BSSIDs and SSIDs. To get these in one event the bit REPORT_SEPARATELY needed to be removed from the flags in brcmf_pno_config(). Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13brcmfmac: add mutex to protect pno requestsArend Van Spriel
The request references kept in pno are accessed in user-space context and in firmware event handler context. As such we need to protect it with a lock. As both context allow sleep a mutex seems appropriate. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-13brcmfmac: add support multi-scheduled scanArend Van Spriel
This change adds support for multi-scheduled scan in the driver. It currently relies on g-scan support in firmware and will set struct wiphy::max_sched_scan_reqs accordingly. This is limited to 16 concurrent requests. The firmware currently has a limit of 64 channels that can be configured for all requests in total regardless whether there are duplicates. So if a request uses 35 channels there are 29 channels left for another request. When user-space does not specify any channels cfg80211 will add all channels defined by the wiphy instance to the request, which makes reaching the limit rather easy for dual-band devices. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-06-12Merge tag 'iwlwifi-next-for-kalle-2017-06-06' of ↵Kalle Valo
git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next First batch of iwlwifi driver patches 4.13 * Loads of FW API documentation improvements (for tools and htmldoc); * Continued work for the new A000 family; * Bumped the maximum supported FW API to 31; * Improve the differentiation between 8000, 9000 and A000 families; * A lot of fixes and cleanups here and there; kvalo: There were conflicts iwl_mvm_stop_device() and iwl_mvm_tcool_set_cur_state(). The former was easy but latter needed more thought. Apparently the mutex was taken too late, so I fixed so that the mutex is taken first and then check for iwl_mvm_firmware_running().
2017-06-12Merge tag 'wireless-drivers-next-for-davem-2017-06-12' of ↵David S. Miller
git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers-next Kalle Valo says: ==================== wireless-drivers-next patches for 4.13 The first pull request for 4.13. We have a new driver qtnfmac, but also rsi driver got a support for new firmware and supporting ath10k SDIO devices was started. Major changes: ath10k * add initial SDIO support (still work in progress) rsi * new loading for the new firmware version rtlwifi * final patches for the new btcoex support rt2x00 * add device ID for Epson WN7512BEP qtnfmac * new driver for Quantenna QSR10G chipsets ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-12Merge branch 'udp-reduce-cache-pressure'David S. Miller
Paolo Abeni says: ==================== udp: reduce cache pressure In the most common use case, many skb fields are not used by recvmsg(), and the few ones actually accessed lays on cold cachelines, which leads to several cache miss per packet. This patch series attempts to reduce such misses with different strategies: * caching the interesting fields in the scratched space * avoid accessing at all uninteresting fields * prefetching Tested using the udp_sink program by Jesper[1] as the receiver, an h/w l4 rx hash on the ingress nic, so that the number of ingress nic rx queues hit by the udp traffic could be controlled via ethtool -L. The udp_sink program was bound to the first idle cpu, to get more stable numbers. On a single numa node receiver: nic rx queues vanilla patched kernel delta 1 1850 kpps 1850 kpps 0% 2 2370 kpps 2700 kpps 13.9% 16 2000 kpps 2220 kpps 11% [1] https://github.com/netoptimizer/network-testing/blob/master/src/udp_sink.c v1 -> v2: - replaced secpath_reset() with skb_release_head_state() - changed udp_dev_scratch fields types to u{32,16} variant, replaced bitfield with bool v2 -> v3: - no changes, tested against apachebench for performances regression ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-12udp: try to avoid 2 cache miss on dequeuePaolo Abeni
when udp_recvmsg() is executed, on x86_64 and other archs, most skb fields are on cold cachelines. If the skb are linear and the kernel don't need to compute the udp csum, only a handful of skb fields are required by udp_recvmsg(). Since we already use skb->dev_scratch to cache hot data, and there are 32 bits unused on 64 bit archs, use such field to cache as much data as we can, and try to prefetch on dequeue the relevant fields that are left out. This can save up to 2 cache miss per packet. v1 -> v2: - changed udp_dev_scratch fields types to u{32,16} variant, replaced bitfiled with bool Signed-off-by: Paolo Abeni <pabeni@redhat.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-12udp: avoid a cache miss on dequeuePaolo Abeni
Since UDP no more uses sk->destructor, we can clear completely the skb head state before enqueuing. Amend and use skb_release_head_state() for that. All head states share a single cacheline, which is not normally used/accesses on dequeue. We can avoid entirely accessing such cacheline implementing and using in the UDP code a specialized skb free helper which ignores the skb head state. This saves a cacheline miss at skb deallocation time. v1 -> v2: replaced secpath_reset() with skb_release_head_state() Signed-off-by: Paolo Abeni <pabeni@redhat.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-12net: factor out a helper to decrement the skb refcountPaolo Abeni
The same code is replicated in 3 different places; move it to a common helper. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-12sh_eth: add support for changing MTUNiklas Söderlund
The hardware supports the MTU to be changed and the driver it self is somewhat prepared to support this. This patch hooks up the callbacks to be able to change the MTU from user-space. Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Acked-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-11bpf, arm64: take advantage of stack_depth trackingDaniel Borkmann
Make use of recently implemented stack_depth tracking for arm64 JIT, so that stack usage can be reduced heavily for programs not using tail calls at least. Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-11Merge tag 'mlx5-updates-2017-06-11' of ↵David S. Miller
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux Saeed Mahameed says: ==================== mlx5-updates-2017-06-11 This series provides updates to mlx5 header rewrite feature, from Or Gerlitz. and three more small updates From maor and eran. ------- Or says: Packets belonging to flows which are different by matching may still need to go through the same header re-writes (e.g set the current routing hop MACs and issue TTL decrement). To minimize the number of modify header IDs, we add a cache for header re-write IDs which is keyed by the binary chain of modify header actions. The caching is supported for both eswitch and NIC use-cases, where the actual conversion of the code to use caching comes in separate patches, one per use-case. Using a per field mask field, the TC pedit action supports modifying partial fields. The last patch enables offloading that. ------- From Maor, update flow table commands layout to the latest HW spec. From Eran, ethtool connector type reporting updates. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-10net: ethernet: ti: cpdma: do not enable host error misc irqGrygorii Strashko
CPSW driver does not handle this interrupt, so there are no reasons to enable it in hardware. Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-10net: ethernet: ti: cpsw: enable HWTSTAMP_FILTER_PTP_V1_L4_EVENT filterGrygorii Strashko
CPSW driver supports PTP v1 messages, but for unknown reasons this filter is not advertised. As result, ./tools/testing/selftests/networking/timestamping/timestamping utility can't be used for testing of CPSW RX timestamping with option SOF_TIMESTAMPING_RX_HARDWARE, because it uses HWTSTAMP_FILTER_PTP_V1_L4_SYNC filter. Hence, fix it by advertising HWTSTAMP_FILTER_PTP_V1_L4_XXX filters in CPSW driver. Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-10Merge branch 'bpf-misc-updates'David S. Miller
Daniel Borkmann says: ==================== Misc BPF updates This set contains a couple of misc updates: stack usage reduction for perf_sample_data in tracing progs, reduction of stale data in verifier on register state transitions that I still had in my queue and few selftest improvements as well as bpf_set_hash() helper for tc programs. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-10bpf: add bpf_set_hash helper for tc progsDaniel Borkmann
Allow for tc BPF programs to set a skb->hash, apart from clearing and triggering a recalc that we have right now. It allows for BPF to implement a custom hashing routine for skb_get_hash(). Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-10bpf: remove cg_skb_func_proto and use sk_filter_func_proto directlyDaniel Borkmann
Since cg_skb_func_proto() doesn't do anything else than just calling into sk_filter_func_proto(), remove it and set sk_filter_func_proto() directly for .get_func_proto callback. Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit (git 2.34.1) at 2025-05-31 19:38:38 +0000