- 30 May, 2018 5 commits
-
-
Colin Ian King authored
The subtraction of two struct ieee80211_wmm_rule pointers leaves a result that is automatically scaled down by the size of the size of pointed-to type, hence the division by sizeof(struct ieee80211_wmm_rule) is bogus and should be removed. Detected by CoverityScan, CID#1467777 ("Extra sizeof expression") Fixes: 77e30e10 ("iwlwifi: mvm: query regdb for wmm rule if needed") Signed-off-by:Colin Ian King <colin.king@canonical.com> Signed-off-by:
Luca Coelho <luciano.coelho@intel.com>
-
Shaul Triebitz authored
Make sure the rx_allocator worker is canceled before running the rx_init routine. rx_init frees and re-allocates all rxb's pages. The rx_allocator worker also allocates pages for the used rxb's. Running rx_init and rx_allocator simultaniously causes a kernel panic. Fix that by canceling the work in rx_init. Signed-off-by:
Shaul Triebitz <shaul.triebitz@intel.com> Signed-off-by:
-
Emmanuel Grumbach authored
Our hardware has a limited amount of buffer descriptors for each Tx packet. Because of that, there is a short piece of code that makes sure that that we don't push too many subframes in an A-MSDU because of subframes needs 2 buffer descriptors. This code also takes into account the number of fragment of the skb since we also need a buffer descriptor for each fragment in the skb. This piece of code though didn't check that the resulting number of subframes wasn't 0. A user reported that using NFS client, he could get skbs that are so fragmented that the code mentioned above returned 0 for the number of subframes making skb_gso_segment fail and subconsequently iwlwifi would WARN. Fix this by make sure that num_subframes is at least 1. This fixes: https://bugzilla.kernel.org/show_bug.cgi?id=199209 Fixes: a6d5e32f ("iwlwifi: mvm: send large SKBs to the transport") Signed-off-by:
Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by:
-
Aviya Erenfeld authored
Some versions of the FW don't support channel switch in TDLS. Add a condition that checks it. Fixes: 307e4723 ("iwlwifi: mvm: configure TDLS peers to FW") Signed-off-by:
Aviya Erenfeld <aviya.erenfeld@intel.com> Signed-off-by:
-
Sara Sharon authored
In some cases we may get from FW errored frames with UNKNOWN security type. This may happen in unsecured aggregation flow, where the first packet had a CRC error in the WEP bit, which was followed by a failure to decrypt and was dropped. The next frames in the aggregation "inherit" the bad metadata of the first packet. Make sure to drop such frames since RADA and other offloads will not operate correctly which may have unexpected results. In case of AP it also causes to TX AMSDU frames to the peers, resulting with assert 0x104B. Signed-off-by:
Sara Sharon <sara.sharon@intel.com> Signed-off-by:
-
- 29 May, 2018 26 commits
-
-
Golan Ben Ami authored
Recently we have switched the csr addresses and values configuration from a single configuration to all devices to a per-device configuration. Doing that, the configuration for 6300 devices wasn't set. This missing definition introduced a kernel panic once trying to access the csr's. Add the missing 6300 csr configuration. While at it, add a checker that the csr values were indeed configured, and bail out more gracefully if not. Fixes: a8cbb46f ("iwlwifi: allow different csr flags for different device families") Signed-off-by:
Golan Ben Ami <golan.ben.ami@intel.com> Signed-off-by:
-
Gustavo A. R. Silva authored
Remove and refactor some code in order to avoid having identical code for different branches. Notice that the logic has been there since 2014. Addresses-Coverity-ID: 1426199 ("Identical code for different branches") Signed-off-by:Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by:
Kalle Valo <kvalo@codeaurora.org>
-
Xinming Hu authored
Active scan is not allowed on radar channel, instead using passvie scan with more time. Signed-off-by:
Xinming Hu <huxm@marvell.com> Signed-off-by:
-
Xinming Hu authored
Signed-off-by:
-
Gustavo A. R. Silva authored
In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. Signed-off-by:
-
Ganapathi Bhat authored
If firmware does not support embedded supplicant, then it in turn will not support GT rekey offloading. If this is the case, then driver must not advertise WOWLAN flags related to GTK rekey and it must also skip sending the GT_REKEY_OFFLOAD_CFG command. Signed-off-by:
Shrenik Shikhare <shrenik@marvell.com> Signed-off-by:
Cathy Luo <cluo@marvell.com> Signed-off-by:
Ganapathi Bhat <gbhat@marvell.com> Signed-off-by:
-
Ganapathi Bhat authored
Race condition is observed during rmmod of mwifiex_usb: 1. The rmmod thread will call mwifiex_usb_disconnect(), download SHUTDOWN command and do wait_event_interruptible_timeout(), waiting for response. 2. The main thread will handle the response and will do a wake_up_interruptible(), unblocking rmmod thread. 3. On getting unblocked, rmmod thread will make rx_cmd.urb = NULL in mwifiex_usb_free(). 4. The main thread will try to resubmit rx_cmd.urb in mwifiex_usb_submit_rx_urb(), which is NULL. To fix, wait for main thread to complete before calling mwifiex_usb_free(). Signed-off-by:
-
Ganapathi Bhat authored
This is a follow-up patch for commit 21c5c83c ("mwifiex: support sysfs initiated device coredump"). Let us avoid exporting mwifiex_send_cmd and instead use a utility function mwifiex_fw_dump_event to achive the work. Signed-off-by:
-
Colin Ian King authored
Trivial fix to spelling mistake in rsi_dbg message text Signed-off-by:
-
Eyal Reizer authored
the wl pointer can be null In case only wlcore_sdio is probed while no WiLink module is successfully probed, as in the case of mounting a wl12xx module while using a device tree file configured with wl18xx related settings. In this case the system was crashing in wl1271_suspend() as platform device data is not set. Make sure wl the pointer is valid before using it. Signed-off-by:
Eyal Reizer <eyalr@ti.com> Signed-off-by:
-
YueHaibing authored
add pci_disable_device in error handling while init_atmel_card failed. Signed-off-by:
YueHaibing <yuehaibing@huawei.com> Signed-off-by:
-
Ping-Ke Shih authored
In former patch, I enqueu all C2H commands and processed by a workqueue. In case C2H_BT_INFO will issue a H2C command to set BT reg, and wait for a C2H ack. But it is totally impossible that C2H workqueue waits for a C2H command, so kernel log warn rtlwifi: :<0> btmpinfo wait (req_num=0) timeout Since the C2H ack command C2H_BT_MP can be safely processed in interrupt context, add a fast command path to deal with the command. Signed-off-by:
Ping-Ke Shih <pkshih@realtek.com> Signed-off-by:
-
Ping-Ke Shih authored
The format of C2H data is ID(1 byte) + Length(1 byte) + value, and it is more readable to use macros to access C2H data. Signed-off-by:
-
Ping-Ke Shih authored
We use 'struct rtl_c2hcmd' to store C2H commands originally, and the code is slightly complex to enqueue and dequeue and also wastes time to allocate and memcpy data. Since C2H commands are asynchronous events, they can be processed in work queue, so RX ISR enqueues C2H result in removal of rtl_c2h_packet_handler(). Signed-off-by:
-
Ping-Ke Shih authored
Similar to rx_command_packet, we can call rtl_c2h_content_parsing so the hal_op isn't necessary. Signed-off-by:
-
Ping-Ke Shih authored
Because the hal_op rx_command_packet does C2H handler if rx packet type is C2H, and the handler have been moved to base.c so we can call the handler directly. Signed-off-by:
-
Ping-Ke Shih authored
Merge duplicate C2H handler and implement the handler in base.c. Signed-off-by:
-
Ping-Ke Shih authored
We're going to merge C2H handler into one, but one special case is to handle RA_REPORT that implements in individual IC folder. So this commit adds a hal_op for caller in common code. Signed-off-by:
-
Ping-Ke Shih authored
The caller of hal_op rx_command_packet will assert function pointer before calling, so we can remove dummy functions safely. Signed-off-by:
-
Ping-Ke Shih authored
The IDs are defined by driver and map to the fw C2H IDs, but they aren't used now result in removal. Signed-off-by:
-
Ping-Ke Shih authored
Move C2H definition to wifi.h, because the definitions of 8192ee, 8723be and 8821ae are the same. Signed-off-by:
-
Ping-Ke Shih authored
Current chips use packet-based C2H commands whose IDs differ from old ones, so this commit simply gives C2H_V0_ as prefix of command IDs. Signed-off-by:
-
Ping-Ke Shih authored
Move duplicate definitions from def.h of ic folder to wifi.h Signed-off-by:
-
Ping-Ke Shih authored
To use keep-alive mechanism in mac80211 stack, since driver supports reporting accurate nullfunc frame tx ack now. Signed-off-by:
Tzu-En Huang <tehuang@realtek.com> Signed-off-by:
-
Tzu-En Huang authored
In order to realize the keep-alive mechanism in mac80211 stack, reporting accurate tx ack status for nullfunc frame is added in this commit. If current frame is nullfunc frame, we ask firmware to report by filling TX report bit in TX descriptor. After this frame DMA done, TX interrupt is triggered but TX status is unknown at this moment, so enqueue this skb into tx_report->queue. Finally, C2H report will be received if the frame is transmitted successfully or retried over, and then we report to mac80211 with IEEE80211_TX_STAT_ACK flag only if it's successful. Otherwise, if failure or timeout (one second), we report to mac80211 without this flag. Signed-off-by:
-
Tony Lindgren authored
We can have pm_runtime_get_sync() return 1, and we can have pm_runtime_put_sync() return -EBUSY. See rpm_suspend() and rpm_resume() for more information. Fix the issue by returning 0 from wl12xx_sdio_power_on() on success. And use pm_runtime_put() instead of pm_runtime_put_sync() for wl12xx_sdio_power_off(), then the MMC subsystem will idle the bus when suitable. Otherwise wlcore can sometimes get confused and may report bogus errors and WLAN connection can fail. Note that while wlcore checks the return value for wl1271_power_on(), the return value is ignored for wl1271_power_off(). Let's fix them both though to avoid further confusion in the future. Fixes: 60f36637 ("wlcore: sdio: allow pm to handle sdio power") Signed-off-by:
Tony Lindgren <tony@atomide.com> Signed-off-by:
-
- 23 May, 2018 9 commits
-
-
Franky Lin authored
Attempt to dump dongle memory for debug upon receiving firmware halt message through dongle to host mail box interrupt. Reviewed-by:
Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by:
Franky Lin <franky.lin@broadcom.com> Signed-off-by:
-
Franky Lin authored
PCIe dongle firmware signals a halt/trap through mailbox interrupt. Trigger a memory dump upon receiving such signal could help to provide useful information for issue debug. Reviewed-by:
-
Franky Lin authored
In patch "brcmfmac: add support for sysfs initiated coredump", a new scenario of brcmf_debug_create_memdump was added in which the user of the function might not necessarily provide prefix data. Hence the function should not assume the data is always valid and should perform a check before copying. Reviewed-by:
-
Arend Van Spriel authored
Since commit 3c47d19f ("drivers: base: add coredump driver ops") it is possible to initiate a device coredump from user-space. This patch adds support for it adding the .coredump() driver callback. As there is no longer a need to initiate it through debugfs remove that code. Signed-off-by:
-
Arend Van Spriel authored
The driver already supports device coredump initiated by firmware event. Since commit 3c47d19f ("drivers: base: add coredump driver ops") it is also possible to initiate it from user-space through sysfs. This patch adds support for SDIO and PCIe devices. Reviewed-by:
Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by:
Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by:
-
Arend Van Spriel authored
The only user of ALLFFMAC is the flowring module so no need to expose it in a header file. Reviewed-by:
-
Sanjay Konduri authored
Auto rate table sent to firmware is getting corrupted as memset to zeros is not done. Added memset to skb data before filling auto rate table. Signed-off-by:
Sanjay Konduri <sanjay.konduri@redpinesignals.com> Signed-off-by:
Sushant Kumar Mishra <sushant.mishra@redpinesignals.com> Signed-off-by:
-
Sanjay Konduri authored
Observed crash in some scenarios when assertion has occurred, this is because hw structure is freed and is tried to get accessed in some functions where null check is already present. So, avoided the crash by making the hw to NULL after freeing. Signed-off-by:
-
Rafał Miłecki authored
This allows reading all capabilities as reported by a firmware. They are printed using native (raw) names, just like developers like it the most. It's how firmware reports support for various features, e.g. supported modes, supported standards, power saving details, max BSS-es. Access to all that info is useful for trying new firmwares, comparing them and debugging features AKA bugs. Signed-off-by:
Rafał Miłecki <rafal@milecki.pl> Reviewed-by:
-
