summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2018-03-29ath10k: fix TDLS peer TX data failure issue on encryped APYingying Tang
For WPA encryption, QCA6174 firmware(version: WLAN.RM.4.4) will unblock data when M4 was sent successfully. For other encryption which didn't need 4-way handshake firmware will unblock the data when peer authorized. Since TDLS is 3-way handshake host need send authorize cmd to firmware to unblock data. Signed-off-by: Yingying Tang <yintang@qti.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29ath10k: avoid to set WEP key for TDLS peerYingying Tang
TDLS peer do not need WEP key. Setting WEP key will lead to TDLS setup failure. Add fix to avoid setting WEP key for TDLS peer. Signed-off-by: Yingying Tang <yintang@qti.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29ath10k: enable TDLS peer inactivity detectionYingying Tang
Enable TDLS peer inactivity detetion feature. QCA6174 firmware(version: WLAN.RM.4.4) support TDLS link inactivity detecting. Set related parameters in TDLS WMI command to enable this feature. Signed-off-by: Yingying Tang <yintang@qti.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29ath10k: enable TDLS peer buffer STA featureYingying Tang
Enable TDLS peer buffer STA feature. QCA6174 firmware(version: WLAN.RM.4.4) support TDLS peer buffer STA, it reports this capability through wmi service map in wmi service ready event. Set related parameter in TDLS WMI command to enable this feature. Signed-off-by: Yingying Tang <yintang@qti.qualcomm.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29wcn36xx: dequeue all pending indicator messagesDaniel Mack
In case wcn36xx_smd_rsp_process() is called more than once before hal_ind_work was dispatched, the messages will end up in hal_ind_queue, but wcn36xx_ind_smd_work() will only look at the first message in that list. Fix this by dequeing the messages from the list in a loop, and only stop when it's empty. This issue was found during a review of the driver. In my tests, that race never actually occured. Signed-off-by: Daniel Mack <daniel@zonque.org> Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29wcn36xx: Fix firmware crash due to corrupted buffer addressLoic Poulain
wcn36xx_start_tx function retrieves the buffer descriptor from the channel control queue to start filling tx buffer information. However, nothing prevents this same buffer to be concurrently accessed in a concurent tx call, leading to potential buffer coruption and firmware crash (observed during iperf test). The channel control queue should only be accessed and updated with the channel lock. Fix this issue by using a local buffer descriptor which will be copied in the thread-safe wcn36xx_dxe_tx_frame. Note that buffer descriptor size is few bytes so the introduced copy overhead is insignificant. Moreover, this allows to keep the locked section minimal. Signed-off-by: Loic Poulain <loic.poulain@linaro.org> Signed-off-by: Ramon Fried <rfried@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29wcn36xx: turn off probe response offloadingRamon Fried
It appears that the WCN36xx firmware doesn't actually respond to probe requests. Until it's resolved, switch the probe response responsibility to the 802.11 layer to allow creation of hidden SSID AP's. Signed-off-by: Ramon Fried <rfried@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29ath10k: add memory dump support QCA9984Anilkumar Kolli
QCA9984/QCA99X0/QCA4019 chipsets have 8 memory regions, dump all of them to the firmware coredump file. Some of the regions need to be read using ioread() so add new region types for them. Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org> [kvalo: refactoring etc] Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29ath10k: refactor ath10k_pci_dump_memory() in preparation for QCA9984 supportKalle Valo
As QCA9984 needs two region types refactor the code to make it easier add the new types. No functional changes. Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29ath10k: suppress "Unknown eventid: 36925" warningsSathishkumar Muruganandam
FW has Smart Logging feature enabled by default for detecting failures and processing FATAL_CONDITION_EVENTID (36925 - 0x903D) back to host. Since ath10k doesn't implement the Smart Logging and FATAL CONDITION EVENT processing yet, suppressing the unknown event ID warning by moving this under ATH10K_DBG_WMI. Simulated the same issue by having associated STA powered off when ping flood was running from AP backbone. This triggerd STA KICKOUT in AP followed by FATAL CONDITION event 36925. Issue was reproduced and verified in below DUT ------------------------------------------------ AP mode of OpenWRT QCA9984 running 6.0.8 with FW ver 10.4-3.5.3-00053 Signed-off-by: Sathishkumar Muruganandam <murugana@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29ath10k: add FW API 6 firmware image for QCA9377Ryan Hsu
Firmware WLAN.TF.2.1-00014-QCARMSWP-1 now supports reading the board ID information and also required 9 IRAM bank, which older ath10k version don't have the support will fail to be enabled, so in order to maintain the backward compatibility, we need to update the FW API to 6. Signed-off-by: Ryan Hsu <ryanhsu@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29ath10k: enable QCA6174/QCA9377 to read the chip temperatureRyan Hsu
The firmware of QCA6174/QCA9377 already support the feature, just enable it to be able to handle the get_temperature command and process the event. You can read the temperature by using the hwmon interface, cat /sys/class/ieee80211/phy*/device/hwmon/hwmon2/temp1_input Verified with the following hardware and software combination, QCA6174, only firmware-4.bin doesn't support this, otherwise all support. QCA9377, all the firmwares upstreamed support this command Signed-off-by: Ryan Hsu <ryanhsu@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29Revert "ath10k: send (re)assoc peer command when NSS changed"Karthikeyan Periyasamy
This reverts commit 55884c045d31a29cf69db8332d1064a1b61dd159. When Ath10k is in AP mode and an unassociated STA sends a VHT action frame (Operating Mode Notification for the NSS change) periodically to AP this causes ath10k to call ath10k_station_assoc() which sends WMI_PEER_ASSOC_CMDID during NSS update. Over the time (with a certain client it can happen within 15 mins when there are over 500 of these VHT action frames) continuous calls of WMI_PEER_ASSOC_CMDID cause firmware to assert due to resource exhaust. To my knowledge setting WMI_PEER_NSS peer param itself enough to handle NSS updates and no need to call ath10k_station_assoc(). So revert the original commit from 2014 as it's unclear why the change was really needed. Now the firmware assert doesn't happen anymore. Issue observed in QCA9984 platform with firmware version:10.4-3.5.3-00053. This Change tested in QCA9984 with firmware version: 10.4-3.5.3-00053 and QCA988x platform with firmware version: 10.2.4-1.0-00036. Firmware Assert log: ath10k_pci 0002:01:00.0: firmware crashed! (guid e61f1274-9acd-4c5b-bcca-e032ea6e723c) ath10k_pci 0002:01:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe ath10k_pci 0002:01:00.0: kconfig debug 1 debugfs 1 tracing 0 dfs 1 testmode 1 ath10k_pci 0002:01:00.0: firmware ver 10.4-3.5.3-00053 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast crc32 4c56a386 ath10k_pci 0002:01:00.0: board_file api 2 bmi_id 0:4 crc32 c2271344 ath10k_pci 0002:01:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal otp max-sta 512 raw 0 hwcrypto 1 ath10k_pci 0002:01:00.0: firmware register dump: ath10k_pci 0002:01:00.0: [00]: 0x0000000A 0x000015B3 0x00981E5F 0x00975B31 ath10k_pci 0002:01:00.0: [04]: 0x00981E5F 0x00060530 0x00000011 0x00446C60 ath10k_pci 0002:01:00.0: [08]: 0x0042F1FC 0x00458080 0x00000017 0x00000000 ath10k_pci 0002:01:00.0: [12]: 0x00000009 0x00000000 0x00973ABC 0x00973AD2 ath10k_pci 0002:01:00.0: [16]: 0x00973AB0 0x00960E62 0x009606CA 0x00000000 ath10k_pci 0002:01:00.0: [20]: 0x40981E5F 0x004066DC 0x00400000 0x00981E34 ath10k_pci 0002:01:00.0: [24]: 0x80983B48 0x0040673C 0x000000C0 0xC0981E5F ath10k_pci 0002:01:00.0: [28]: 0x80993DEB 0x0040676C 0x00431AB8 0x0045D0C4 ath10k_pci 0002:01:00.0: [32]: 0x80993E5C 0x004067AC 0x004303C0 0x0045D0C4 ath10k_pci 0002:01:00.0: [36]: 0x80994AAB 0x004067DC 0x00000000 0x0045D0C4 ath10k_pci 0002:01:00.0: [40]: 0x809971A0 0x0040681C 0x004303C0 0x00441B00 ath10k_pci 0002:01:00.0: [44]: 0x80991904 0x0040688C 0x004303C0 0x0045D0C4 ath10k_pci 0002:01:00.0: [48]: 0x80963AD3 0x00406A7C 0x004303C0 0x009918FC ath10k_pci 0002:01:00.0: [52]: 0x80960E80 0x00406A9C 0x0000001F 0x00400000 ath10k_pci 0002:01:00.0: [56]: 0x80960E51 0x00406ACC 0x00400000 0x00000000 ath10k_pci 0002:01:00.0: Copy Engine register dump: ath10k_pci 0002:01:00.0: index: addr: sr_wr_idx: sr_r_idx: dst_wr_idx: dst_r_idx: ath10k_pci 0002:01:00.0: [00]: 0x0004a000 15 15 3 3 ath10k_pci 0002:01:00.0: [01]: 0x0004a400 17 17 212 213 ath10k_pci 0002:01:00.0: [02]: 0x0004a800 21 21 20 21 ath10k_pci 0002:01:00.0: [03]: 0x0004ac00 25 25 27 25 ath10k_pci 0002:01:00.0: [04]: 0x0004b000 515 515 144 104 ath10k_pci 0002:01:00.0: [05]: 0x0004b400 28 28 155 156 ath10k_pci 0002:01:00.0: [06]: 0x0004b800 12 12 12 12 ath10k_pci 0002:01:00.0: [07]: 0x0004bc00 1 1 1 1 ath10k_pci 0002:01:00.0: [08]: 0x0004c000 0 0 127 0 ath10k_pci 0002:01:00.0: [09]: 0x0004c400 1 1 1 1 ath10k_pci 0002:01:00.0: [10]: 0x0004c800 0 0 0 0 ath10k_pci 0002:01:00.0: [11]: 0x0004cc00 0 0 0 0 ath10k_pci 0002:01:00.0: CE[1] write_index 212 sw_index 213 hw_index 0 nentries_mask 0x000001ff ath10k_pci 0002:01:00.0: CE[2] write_index 20 sw_index 21 hw_index 0 nentries_mask 0x0000007f ath10k_pci 0002:01:00.0: CE[5] write_index 155 sw_index 156 hw_index 0 nentries_mask 0x000001ff ath10k_pci 0002:01:00.0: DMA addr: nbytes: meta data: byte swap: gather: ath10k_pci 0002:01:00.0: [455]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [456]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [457]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [458]: 0x594a0038 0 0 0 1 ath10k_pci 0002:01:00.0: [459]: 0x580c0a42 0 0 0 0 ath10k_pci 0002:01:00.0: [460]: 0x594a0060 0 0 0 1 ath10k_pci 0002:01:00.0: [461]: 0x580c0c42 0 0 0 0 ath10k_pci 0002:01:00.0: [462]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [463]: 0x580c0c42 0 0 0 0 ath10k_pci 0002:01:00.0: [464]: 0x594a0038 0 0 0 1 ath10k_pci 0002:01:00.0: [465]: 0x580c0a42 0 0 0 0 ath10k_pci 0002:01:00.0: [466]: 0x594a0060 0 0 0 1 ath10k_pci 0002:01:00.0: [467]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [468]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [469]: 0x580c1c42 0 0 0 0 ath10k_pci 0002:01:00.0: [470]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [471]: 0x580c1c42 0 0 0 0 ath10k_pci 0002:01:00.0: [472]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [473]: 0x580c1c42 0 0 0 0 ath10k_pci 0002:01:00.0: [474]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [475]: 0x580c0642 0 0 0 0 ath10k_pci 0002:01:00.0: [476]: 0x594a0038 0 0 0 1 ath10k_pci 0002:01:00.0: [477]: 0x580c0842 0 0 0 0 ath10k_pci 0002:01:00.0: [478]: 0x594a0060 0 0 0 1 ath10k_pci 0002:01:00.0: [479]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [480]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [481]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [482]: 0x594a0038 0 0 0 1 ath10k_pci 0002:01:00.0: [483]: 0x580c0842 0 0 0 0 ath10k_pci 0002:01:00.0: [484]: 0x594a0060 0 0 0 1 ath10k_pci 0002:01:00.0: [485]: 0x580c0642 0 0 0 0 ath10k_pci 0002:01:00.0: [486]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [487]: 0x580c0642 0 0 0 0 ath10k_pci 0002:01:00.0: [488]: 0x594a0038 0 0 0 1 ath10k_pci 0002:01:00.0: [489]: 0x580c0842 0 0 0 0 ath10k_pci 0002:01:00.0: [490]: 0x594a0060 0 0 0 1 ath10k_pci 0002:01:00.0: [491]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [492]: 0x58174040 0 1 0 0 ath10k_pci 0002:01:00.0: [493]: 0x5a946040 0 1 0 0 ath10k_pci 0002:01:00.0: [494]: 0x59909040 0 1 0 0 ath10k_pci 0002:01:00.0: [495]: 0x5ae5a040 0 1 0 0 ath10k_pci 0002:01:00.0: [496]: 0x58096040 0 1 0 0 ath10k_pci 0002:01:00.0: [497]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [498]: 0x580c0642 0 0 0 0 ath10k_pci 0002:01:00.0: [499]: 0x5c1e0040 0 1 0 0 ath10k_pci 0002:01:00.0: [500]: 0x58153040 0 1 0 0 ath10k_pci 0002:01:00.0: [501]: 0x58129040 0 1 0 0 ath10k_pci 0002:01:00.0: [502]: 0x5952f040 0 1 0 0 ath10k_pci 0002:01:00.0: [503]: 0x59535040 0 1 0 0 ath10k_pci 0002:01:00.0: [504]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [505]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [506]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [507]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [508]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [509]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [510]: 0x594a0010 0 0 0 1 ath10k_pci 0002:01:00.0: [511]: 0x580c0042 0 0 0 0 ath10k_pci 0002:01:00.0: [512]: 0x5adcc040 0 1 0 0 ath10k_pci 0002:01:00.0: [513]: 0x5cf3d040 0 1 0 0 ath10k_pci 0002:01:00.0: [514]: 0x5c1e9040 64 1 0 0 ath10k_pci 0002:01:00.0: [515]: 0x00000000 0 0 0 0 Signed-off-by: Karthikeyan Periyasamy <periyasa@codeaurora.org> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29ath9k: fix DFS detector synchronizationTimothy Redaelli
some userspace programs (e.g. hostapd) need to set the regulatory domain before selecting the operating channel. Synchronize DFS detector regardless of the value of ah->curchan, to avoid situations where wireless scan can't be done on some 5GHz sub-bands, because dfs_region is constantly UNSET. Acked-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Timothy Redaelli <tredaelli@redhat.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2018-03-29nl80211: Add SOCKET_OWNER support to START_APDenis Kenzior
Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2018-03-29nl80211: Add SOCKET_OWNER support to JOIN_MESHDenis Kenzior
Signed-off-by: Denis Kenzior <denkenz@gmail.com> [johannes: fix race with wdev lock/unlock by just acquiring once] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2018-03-29nl80211: Add SOCKET_OWNER support to JOIN_IBSSDenis Kenzior
Signed-off-by: Denis Kenzior <denkenz@gmail.com> [johannes: fix race with wdev lock/unlock by just acquiring once] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2018-03-29Merge tag 'iwlwifi-next-for-kalle-2018-03-28' of ↵Kalle Valo
git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next iwlwifi patches for 4.14 * Fix a UBSAN warning; * Improvement in the net-stack/driver log syncing * An RCU lock fix in the new rate-scaling code; * Support for new PCI IDs for the 9000 family; * Support for a new API version; * Initial work for supporting the new ETSI rules; * Support for advanced dwell in scanning; * Support for OCE in scanning; * Some debugging improvements; * Other small improvements and cleanups here and there.
2018-03-29cfg80211: Support all iftypes in autodisconnect_wkDenis Kenzior
Currently autodisconnect_wk assumes that only interface types of P2P_CLIENT and STATION use conn_owner_nlportid. Change this so all interface types are supported. Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2018-03-29cfg80211: enable use of non-cleared DFS channels for DFS offloadDmitry Lebed
Currently channel switch/start_ap to DFS channel cannot be done to non-CAC-cleared channel even if DFS offload if enabled. Make non-cleared DFS channels available if DFS offload is enabled. CAC will be started by HW after channel change, start_ap call, etc. Signed-off-by: Dmitry Lebed <dlebed@quantenna.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2018-03-29cfg80211: fix CAC_STARTED event handlingDmitry Lebed
Exclude CAC_STARTED event from !wdev->cac_started check, since cac_started will be set later in the same function. Signed-off-by: Dmitry Lebed <dlebed@quantenna.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2018-03-29mac80211: Use proper chan_width enum in sta opmode eventtamizhr@codeaurora.org
Bandwidth change value reported via nl80211 contains mac80211 specific enum value(ieee80211_sta_rx_bw) and which is not understand by userspace application. Map the mac80211 specific value to nl80211_chan_width enum value to avoid using wrong value in the userspace application. And used station's ht/vht capability to map IEEE80211_STA_RX_BW_20 and IEEE80211_STA_RX_BW_160 with proper nl80211 value. Signed-off-by: Tamizh chelvam <tamizhr@codeaurora.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2018-03-29mac80211: Use proper smps_mode enum in sta opmode eventtamizhr@codeaurora.org
SMPS_MODE change value notified via nl80211 contains mac80211 specific value(ieee80211_smps_mode) and user space application will not know those values. This patch add support to map the mac80211 enum value to nl80211_smps_mode which will be understood by the userspace application. Signed-off-by: Tamizh chelvam <tamizhr@codeaurora.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2018-03-29cfg80211: fix data type of sta_opmode_info parametertamizhr@codeaurora.org
Currently bw and smps_mode are u8 type value in sta_opmode_info structure. This values filled in mac80211 from ieee80211_sta_rx_bandwidth and ieee80211_smps_mode. These enum values are specific to mac80211 and userspace/cfg80211 doesn't know about that. This will lead to incorrect result/assumption by the user space application. Change bw and smps_mode parameters to their respective enums in nl80211. Signed-off-by: Tamizh chelvam <tamizhr@codeaurora.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2018-03-28Merge branch 'nfp-bpf-updates'Alexei Starovoitov
Jakub Kicinski says: ==================== This set adds support for update and delete calls from the datapath, as well as XADD instructions (32 and 64 bit) and pseudo random numbers. The XADD support depends on verifier enforcing alignment which Daniel recently added. XADD uses NFP's atomic engine which requires values to be in big endian, therefore we need to keep track of which parts of the values are used as atomics and byte swap them accordingly. Pseudo random numbers are generated using NFP's HW pseudo random number generator. Jiong tackles initial implementation of packet cache, which he describes as follows: Memory reads on NFP would first fetch data from memory to transfer-in registers, then move them from transfer-in to general registers. Given NFP is rich on transfer-in registers, they could serve as memory cache. This patch tries to identify a sequence of packet data read (BPF_LDX) that are executed sequentially, then the total access range of the sequence is calculated and attached to each read instruction, the first instruction in this sequence is marked with an cache init flag so the execution of it would bring in the whole range of packet data for the sequence. All later packet reads in this sequence would fetch data from transfer-in registers directly, no need to JIT NFP memory access. Function call, non-packet-data memory read, packet write and memcpy will invalidate the cache and start a new cache range. Cache invalidation could be improved in the future, for example packet write doesn't need to invalidate the cache if the the write destination won't be read again. ==================== Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: improve wrong FW response warningsJakub Kicinski
When FW responds with a message of wrong size or type make sure the type is checked first and included in the wrong size message. This makes it easier to figure out which FW command failed. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: add support for bpf_get_prandom_u32()Jakub Kicinski
NFP has a prng register, which we can read to obtain a u32 worth of pseudo random data. Generate code for it. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: add support for atomic add of unknown valuesJakub Kicinski
Allow atomic add to be used even when the value is not guaranteed to fit into a 16 bit immediate. This requires the value to be pulled as data, and therefore use of a transfer register and a context swap. Track the information about possible lengths of the value, if it's guaranteed to be larger than 16bits don't generate the code for the optimized case at all. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: expose command delay slotsJakub Kicinski
Allow callers to control the delay slots of commands, instead of giving them just a wait/nowait choice. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: add basic support for atomic addsJakub Kicinski
Implement atomic add operation for 32 and 64 bit values. Depend on the verifier to ensure alignment. Values have to be kept in big endian and swapped upon read/write. For now only support atomic add of a constant. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28bpf: add parenthesis around argument of BPF_LDST_BYTES()Jakub Kicinski
BPF_LDST_BYTES() does not put it's argument in parenthesis when referencing it. This makes it impossible to pass pointers obtained by address-of operator (e.g. BPF_LDST_BYTES(&insn)). Add the parenthesis. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: add map deletes from the datapathJakub Kicinski
Support calling map_delete_elem() FW helper from the datapath programs. For JIT checks and code are basically equivalent to map lookups. Similarly to other map helper key must be on the stack. Different pointer types are left for future extension. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: add map updates from the datapathJakub Kicinski
Support calling map_update_elem() from the datapath programs by calling into FW-provided helper. Value pointer is passed in LM pointer #2. Keeping track of old state for arg3 is not necessary, since LM pointer #2 will be always loaded in this case, the trivial optimization for value at the bottom of the stack can't be done here. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: add helper for basic map call checksJakub Kicinski
Add a verifier helper for performing the basic state checks before a call to a map helper. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: add helper for validating stack pointersJakub Kicinski
Our implementation has restriction on stack pointers for function calls. Move the common checks into a helper for reuse. The state has to be encapsulated into a structure to support parameters other than BPF_REG_2. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: rename map_lookup_stack() to map_call_stack_common()Jakub Kicinski
We will reuse most of map call code gen for other map calls. Rename the lookup gen function and use meta->func_id instead of hard-coding lookup. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com> Reviewed-by: Jiong Wang <jiong.wang@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: detect packet reads could be cached, enable the optimisationJiong Wang
This patch is the front end of this optimisation, it detects and marks those packet reads that could be cached. Then the optimisation "backend" will be activated automatically. Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: support unaligned read offsetJiong Wang
This patch add the support for unaligned read offset, i.e. the read offset to the start of packet cache area is not aligned to REG_WIDTH. In this case, the read area might across maximum three transfer-in registers. Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28nfp: bpf: read from packet data cache for PTR_TO_PACKETJiong Wang
This patch assumes there is a packet data cache, and would try to read packet data from the cache instead of from memory. This patch only implements the optimisation "backend", it doesn't build the packet data cache, so this optimisation is not enabled. This patch has only enabled aligned packet data read, i.e. when the read offset to the start of cache is REG_WIDTH aligned. Signed-off-by: Jiong Wang <jiong.wang@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2018-03-28Merge branch 'bpf-raw-tracepoints'Daniel Borkmann
Alexei Starovoitov says: ==================== v7->v8: - moved 'u32 num_args' from 'struct tracepoint' into 'struct bpf_raw_event_map' that increases memory overhead, but can be optimized/compressed later. Now it's zero changes in tracepoint.[ch] v6->v7: - adopted Steven's bpf_raw_tp_map section approach to find tracepoint and corresponding bpf probe function instead of kallsyms approach. dropped kernel_tracepoint_find_by_name() patch v5->v6: - avoid changing semantics of for_each_kernel_tracepoint() function, instead introduce kernel_tracepoint_find_by_name() helper v4->v5: - adopted Daniel's fancy REPEAT macro in bpf_trace.c in patch 6 v3->v4: - adopted Linus's CAST_TO_U64 macro to cast any integer, pointer, or small struct to u64. That nicely reduced the size of patch 1 v2->v3: - with Linus's suggestion introduced generic COUNT_ARGS and CONCATENATE macros (or rather moved them from apparmor) that cleaned up patch 6 - added patch 4 to refactor trace_iwlwifi_dev_ucode_error() from 17 args to 4 Now any tracepoint with >12 args will have build error v1->v2: - simplified api by combing bpf_raw_tp_open(name) + bpf_attach(prog_fd) into bpf_raw_tp_open(name, prog_fd) as suggested by Daniel. That simplifies bpf_detach as well which is now simple close() of fd. - fixed memory leak in error path which was spotted by Daniel. - fixed bpf_get_stackid(), bpf_perf_event_output() called from raw tracepoints - added more tests - fixed allyesconfig build caught by buildbot v1: This patch set is a different way to address the pressing need to access task_struct pointers in sched tracepoints from bpf programs. The first approach simply added these pointers to sched tracepoints: https://lkml.org/lkml/2017/12/14/753 which Peter nacked. Few options were discussed and eventually the discussion converged on doing bpf specific tracepoint_probe_register() probe functions. Details here: https://lkml.org/lkml/2017/12/20/929 Patch 1 is kernel wide cleanup of pass-struct-by-value into pass-struct-by-reference into tracepoints. Patches 2 and 3 are minor cleanups to address allyesconfig build Patch 4 refactor trace_iwlwifi_dev_ucode_error from 17 to 4 args Patch 5 introduces COUNT_ARGS macro Patch 6 introduces BPF_RAW_TRACEPOINT api. the auto-cleanup and multiple concurrent users are must have features of tracing api. For bpf raw tracepoints it looks like: // load bpf prog with BPF_PROG_TYPE_RAW_TRACEPOINT type prog_fd = bpf_prog_load(...); // receive anon_inode fd for given bpf_raw_tracepoint // and attach bpf program to it raw_tp_fd = bpf_raw_tracepoint_open("xdp_exception", prog_fd); Ctrl-C of tracing daemon or cmdline tool will automatically detach bpf program, unload it and unregister tracepoint probe. More details in patch 6. Patch 7 - trivial support in libbpf Patches 8, 9 - user space tests samples/bpf/test_overhead performance on 1 cpu: tracepoint base kprobe+bpf tracepoint+bpf raw_tracepoint+bpf task_rename 1.1M 769K 947K 1.0M urandom_read 789K 697K 750K 755K ==================== Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28selftests/bpf: test for bpf_get_stackid() from raw tracepointsAlexei Starovoitov
similar to traditional traceopint test add bpf_get_stackid() test from raw tracepoints and reduce verbosity of existing stackmap test Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28samples/bpf: raw tracepoint testAlexei Starovoitov
add empty raw_tracepoint bpf program to test overhead similar to kprobe and traditional tracepoint tests Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28libbpf: add bpf_raw_tracepoint_open helperAlexei Starovoitov
add bpf_raw_tracepoint_open(const char *name, int prog_fd) api to libbpf Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28bpf: introduce BPF_RAW_TRACEPOINTAlexei Starovoitov
Introduce BPF_PROG_TYPE_RAW_TRACEPOINT bpf program type to access kernel internal arguments of the tracepoints in their raw form. >From bpf program point of view the access to the arguments look like: struct bpf_raw_tracepoint_args { __u64 args[0]; }; int bpf_prog(struct bpf_raw_tracepoint_args *ctx) { // program can read args[N] where N depends on tracepoint // and statically verified at program load+attach time } kprobe+bpf infrastructure allows programs access function arguments. This feature allows programs access raw tracepoint arguments. Similar to proposed 'dynamic ftrace events' there are no abi guarantees to what the tracepoints arguments are and what their meaning is. The program needs to type cast args properly and use bpf_probe_read() helper to access struct fields when argument is a pointer. For every tracepoint __bpf_trace_##call function is prepared. In assembler it looks like: (gdb) disassemble __bpf_trace_xdp_exception Dump of assembler code for function __bpf_trace_xdp_exception: 0xffffffff81132080 <+0>: mov %ecx,%ecx 0xffffffff81132082 <+2>: jmpq 0xffffffff811231f0 <bpf_trace_run3> where TRACE_EVENT(xdp_exception, TP_PROTO(const struct net_device *dev, const struct bpf_prog *xdp, u32 act), The above assembler snippet is casting 32-bit 'act' field into 'u64' to pass into bpf_trace_run3(), while 'dev' and 'xdp' args are passed as-is. All of ~500 of __bpf_trace_*() functions are only 5-10 byte long and in total this approach adds 7k bytes to .text. This approach gives the lowest possible overhead while calling trace_xdp_exception() from kernel C code and transitioning into bpf land. Since tracepoint+bpf are used at speeds of 1M+ events per second this is valuable optimization. The new BPF_RAW_TRACEPOINT_OPEN sys_bpf command is introduced that returns anon_inode FD of 'bpf-raw-tracepoint' object. The user space looks like: // load bpf prog with BPF_PROG_TYPE_RAW_TRACEPOINT type prog_fd = bpf_prog_load(...); // receive anon_inode fd for given bpf_raw_tracepoint with prog attached raw_tp_fd = bpf_raw_tracepoint_open("xdp_exception", prog_fd); Ctrl-C of tracing daemon or cmdline tool that uses this feature will automatically detach bpf program, unload it and unregister tracepoint probe. On the kernel side the __bpf_raw_tp_map section of pointers to tracepoint definition and to __bpf_trace_*() probe function is used to find a tracepoint with "xdp_exception" name and corresponding __bpf_trace_xdp_exception() probe function which are passed to tracepoint_probe_register() to connect probe with tracepoint. Addition of bpf_raw_tracepoint doesn't interfere with ftrace and perf tracepoint mechanisms. perf_event_open() can be used in parallel on the same tracepoint. Multiple bpf_raw_tracepoint_open("xdp_exception", prog_fd) are permitted. Each with its own bpf program. The kernel will execute all tracepoint probes and all attached bpf programs. In the future bpf_raw_tracepoints can be extended with query/introspection logic. __bpf_raw_tp_map section logic was contributed by Steven Rostedt Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28macro: introduce COUNT_ARGS() macroAlexei Starovoitov
move COUNT_ARGS() macro from apparmor to generic header and extend it to count till twelve. COUNT() was an alternative name for this logic, but it's used for different purpose in many other places. Similarly for CONCATENATE() macro. Suggested-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28net/wireless/iwlwifi: fix iwlwifi_dev_ucode_error tracepointAlexei Starovoitov
fix iwlwifi_dev_ucode_error tracepoint to pass pointer to a table instead of all 17 arguments by value. dvm/main.c and mvm/utils.c have 'struct iwl_error_event_table' defined with very similar yet subtly different fields and offsets. tracepoint is still common and using definition of 'struct iwl_error_event_table' from dvm/commands.h while copying fields. Long term this tracepoint probably should be split into two. Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28net/mac802154: disambiguate mac80215 vs mac802154 trace eventsAlexei Starovoitov
two trace events defined with the same name and both unused. They conflict in allyesconfig build. Rename one of them. Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28net/mediatek: disambiguate mt76 vs mt7601u trace eventsAlexei Starovoitov
two trace events defined with the same name and both unused. They conflict in allyesconfig build. Rename one of them. Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28treewide: remove large struct-pass-by-value from tracepoint argumentsAlexei Starovoitov
- fix trace_hfi1_ctxt_info() to pass large struct by reference instead of by value - convert 'type array[]' tracepoint arguments into 'type *array', since compiler will warn that sizeof('type array[]') == sizeof('type *array') and later should be used instead The CAST_TO_U64 macro in the later patch will enforce that tracepoint arguments can only be integers, pointers, or less than 8 byte structures. Larger structures should be passed by reference. Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
2018-03-28iwlwifi: wrt: add fw force restart via triggersShahar S Matityahu
We can set triggers that cause a debug data collection when something of interest happens (e.g. when too many probes are lost conscutively). Normally, this triggers don't cause the FW to be restarted, but in some cases that may be desired, so we recover from the problem. To support this, add a flag that indicates that the FW should be restarted when the trigger fires. Signed-off-by: Shahar S Matityahu <shahar.s.matityahu@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
generated by cgit (git 2.34.1) at 2025-09-12 13:05:21 +0000