- 03 Apr, 2023 12 commits
-
-
Cai Huoqing authored
Remove pci_clear_master to simplify the code, the bus-mastering is also cleared in do_pci_disable_device, like this: ./drivers/pci/pci.c:2197 static void do_pci_disable_device(struct pci_dev *dev) { u16 pci_command; pci_read_config_word(dev, PCI_COMMAND, &pci_command); if (pci_command & PCI_COMMAND_MASTER) { pci_command &= ~PCI_COMMAND_MASTER; pci_write_config_word(dev, PCI_COMMAND, pci_command); } pcibios_disable_device(dev); }. And dev->is_busmaster is set to 0 in pci_disable_device. Signed-off-by:Cai Huoqing <cai.huoqing@linux.dev> Reviewed-by:
Simon Horman <simon.horman@corigine.com> Acked-by:
Ping-Ke Shih <pkshih@realtek.com> Signed-off-by:
Kalle Valo <kvalo@kernel.org> Link: https://lore.kernel.org/r/20230323112613.7550-5-cai.huoqing@linux.dev
-
Ping-Ke Shih authored
A race condition can happen if netdev is registered, but NAPI isn't initialized yet, and meanwhile user space starts the netdev that will enable NAPI. Then, it hits BUG_ON(): kernel BUG at net/core/dev.c:6423! invalid opcode: 0000 [#1] PREEMPT SMP NOPTI CPU: 0 PID: 417 Comm: iwd Not tainted 6.2.7-slab-dirty #3 eb0f5a8a9d91 Hardware name: LENOVO 21DL/LNVNB161216, BIOS JPCN20WW(V1.06) 09/20/2022 RIP: 0010:napi_enable+0x3f/0x50 Code: 48 89 c2 48 83 e2 f6 f6 81 89 08 00 00 02 74 0d 48 83 ... RSP: 0018:ffffada1414f3548 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffffa01425802080 RCX: 0000000000000000 RDX: 00000000000002ff RSI: ffffada14e50c614 RDI: ffffa01425808dc0 RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000100 R12: ffffa01425808f58 R13: 0000000000000000 R14: ffffa01423498940 R15: 0000000000000001 FS: 00007f5577c0a740(0000) GS:ffffa0169fc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f5577a19972 CR3: 0000000125a7a000 CR4: 0000000000750ef0 PKRU: 55555554 Call Trace: <TASK> rtw89_pci_ops_start+0x1c/0x70 [rtw89_pci 6cbc75429515c181cbc386478d5cfb32ffc5a0f8] rtw89_core_start+0xbe/0x160 [rtw89_core fe07ecb874820b6d778370d4acb6ef8a37847f22] rtw89_ops_start+0x26/0x40 [rtw89_core fe07ecb874820b6d778370d4acb6ef8a37847f22] drv_start+0x42/0x100 [mac80211 c07fa22af8c3cf3f7d7ab3884ca990784d72e2d2] ieee80211_do_open+0x311/0x7d0 [mac80211 c07fa22af8c3cf3f7d7ab3884ca990784d72e2d2] ieee80211_open+0x6a/0x90 [mac80211 c07fa22af8c3cf3f7d7ab3884ca990784d72e2d2] __dev_open+0xe0/0x180 __dev_change_flags+0x1da/0x250 dev_change_flags+0x26/0x70 do_setlink+0x37c/0x12c0 ? ep_poll_callback+0x246/0x290 ? __nla_validate_parse+0x61/0xd00 ? __wake_up_common_lock+0x8f/0xd0 To fix this, follow Jonas' suggestion to switch the order of these functions and move register netdev to be the last step of PCI probe. Also, correct the error handling of rtw89_core_register_hw(). Fixes: e3ec7017 ("rtw89: add Realtek 802.11ax driver") Cc: stable@vger.kernel.org Reported-by:
Hyeonggon Yoo <42.hyeyoo@gmail.com> Link: https://lore.kernel.org/linux-wireless/CAOiHx=n7EwK2B9CnBR07FVA=sEzFagb8TkS4XC_qBNq8OwcYUg@mail.gmail.com/T/#tSuggested-by:
Jonas Gorski <jonas.gorski@gmail.com> Tested-by: Larry Finger<Larry.Finger@lwfinger.net> Reviewed-by: Larry Finger<Larry.Finger@lwfinger.net> Signed-off-by:
-
Chih-Kang Chang authored
Need to configure EDCCA threshold to default value before scan, and recall original value after scan to prevent probe request can't be sent out. Signed-off-by:
Chih-Kang Chang <gary.chang@realtek.com> Signed-off-by:
-
Chih-Kang Chang authored
We use ppdu_sts to obtain channel information from hardware, to ensure that the scan results have correct channel information. However, some of ppdu_sts that is filtered by the to_self check is also needed for the scan results. Therefore, we modify the filter part in front of CFO count. Signed-off-by:
-
Chin-Yen Lee authored
We find that when starting WoWLAN, the second join_info H2C is unnecessary and leads WoWLAN not enter power save mode if using new firmware, so remove it. Signed-off-by:
Chin-Yen Lee <timlee@realtek.com> Signed-off-by:
-
Chih-Kang Chang authored
By default the driver uses the 1M and 6M rate (0x0) for data frames in 2 GHz and 5/6 GHz bands respectively. But the rate that might not AP supported. Therefore, We modify the data lowest rate according to the lowest of AP supported rate. Signed-off-by:
-
Ping-Ke Shih authored
The register-based H2C/C2H are used to exchange information between driver and firmware, but only apply to narrow area because its data size is smaller than regular packet-based H2C/C2H. This kind of H2C/C2H must be paired. To identify if any H2C/C2H is missing, update counters to help diagnose this kind of problems. Signed-off-by:
-
Ching-Te Ku authored
Update driver with the supported firmware version of the below item. Bluetooth firmware BT_Coex_Ver: 0x07 Wi-Fi firmware version: RTL8852C->v0.27.56.10 RTL8852A->v0.24.36 RTL8852B->v0.29.29 Signed-off-by:
Ching-Te Ku <ku920601@realtek.com> Signed-off-by:
-
Ching-Te Ku authored
In order to reduce firmware code size cost, remove some counter value from the structure. But firmware didn't update version code. To parse the correct report, add another variation version v105 to parse it. Signed-off-by:
-
Ching-Te Ku authored
The LNA gain didn't set before, it will lead some WiFi RX issue. And the setting can increase both of WiFi & BT performance while they are both RX. Signed-off-by:
-
Ching-Te Ku authored
If driver enable firmware report during WiFi power saving, the firmware timer will lead to some power saving issue like, fail to enter LPS, can not leave LPS or some unexpected issue. Signed-off-by:
-
Ching-Te Ku authored
This LPS state will not turn off RF, and it can still do some basic traffic, only RTL8852B has the state. Coexistence need let Bluetooth know WiFi is still alive to prevent some Bluetooth performance issue. Signed-off-by:
-
- 31 Mar, 2023 17 commits
-
-
Tom Rix authored
clang with W=1 reports drivers/bcma/driver_mips.c:49:19: error: unused function 'mips_read32' [-Werror,-Wunused-function] static inline u32 mips_read32(struct bcma_drv_mips *mcore, ^ This function is not used so remove it. Signed-off-by:Tom Rix <trix@redhat.com> Reviewed-by:
-
Rob Herring authored
Replace of_get_address() and of_translate_address() calls with single call to of_address_to_resource(). Signed-off-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
Tom Rix authored
clang with W=1 reports drivers/net/wireless/marvell/mwifiex/11h.c:198:6: error: variable 'evt_buf' set but not used [-Werror,-Wunused-but-set-variable] u8 *evt_buf; ^ This variable is not used so remove it. Signed-off-by: -
Tom Rix authored
clang with W=1 reports drivers/net/wireless/broadcom/brcm80211/brcmsmac/ampdu.c:848:5: error: variable 'suc_mpdu' set but not used [-Werror,-Wunused-but-set-variable] u8 suc_mpdu = 0, tot_mpdu = 0; ^ This variable is not used so remove it. Signed-off-by: -
Wei Chen authored
If there is a failure during copy_from_user or user-provided data buffer is invalid, rtl_debugfs_set_write_reg should return negative error code instead of a positive value count. Fix this bug by returning correct error code. Moreover, the check of buffer against null is removed since it will be handled by copy_from_user. Fixes: 610247f4 ("rtlwifi: Improve debugging by using debugfs") Signed-off-by:
Wei Chen <harperchen1110@gmail.com> Reviewed-by:
-
Wei Chen authored
If there is a failure during copy_from_user or user-provided data buffer is invalid, rtl_debugfs_set_write_rfreg should return negative error code instead of a positive value count. Fix this bug by returning correct error code. Moreover, the check of buffer against null is removed since it will be handled by copy_from_user. Fixes: 610247f4 ("rtlwifi: Improve debugging by using debugfs") Signed-off-by:
-
Tom Rix authored
clang with W=1 reports drivers/net/wireless/broadcom/brcm80211/brcmsmac/mac80211_if.c:1051:6: error: variable 'has_5g' set but not used [-Werror,-Wunused-but-set-variable] int has_5g = 0; ^ This variable is not used so remove it. Signed-off-by: -
Tom Rix authored
clang with W=1 reports drivers/net/wireless/broadcom/b43legacy/radio.c:1713:5: error: unused function 'freq_r3A_value' [-Werror,-Wunused-function] u16 freq_r3A_value(u16 frequency) ^ This function is not used so remove it. Signed-off-by:Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
-
Gustavo A. R. Silva authored
Zero-length arrays as fake flexible arrays are deprecated and we are moving towards adopting C99 flexible-array members instead. Address the following warning found with GCC-13 and -fstrict-flex-arrays=3 enabled: In function ‘fortify_memset_chk’, inlined from ‘rtl_usb_probe’ at drivers/net/wireless/realtek/rtlwifi/usb.c:1044:2: ./include/linux/fortify-string.h:430:25: warning: call to ‘__write_overflow_field’ declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 430 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This helps with the ongoing efforts to tighten the FORTIFY_SOURCE routines on memcpy() and help us make progress towards globally enabling -fstrict-flex-arrays=3 [1]. Link: https://github.com/KSPP/linux/issues/21 Link: https://github.com/KSPP/linux/issues/277 Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1] Signed-off-by:Gustavo A. R. Silva <gustavoars@kernel.org> Reviewed-by:
-
Cai Huoqing authored
Remove pci_clear_master to simplify the code, the bus-mastering is also cleared in do_pci_disable_device, like this: ./drivers/pci/pci.c:2197 static void do_pci_disable_device(struct pci_dev *dev) { u16 pci_command; pci_read_config_word(dev, PCI_COMMAND, &pci_command); if (pci_command & PCI_COMMAND_MASTER) { pci_command &= ~PCI_COMMAND_MASTER; pci_write_config_word(dev, PCI_COMMAND, pci_command); } pcibios_disable_device(dev); }. And dev->is_busmaster is set to 0 in pci_disable_device. Signed-off-by: -
Gustavo A. R. Silva authored
Zero-length arrays as fake flexible arrays are deprecated and we are moving towards adopting C99 flexible-array members instead. Address the following warning found with GCC-13 and -fstrict-flex-array=3 enabled: drivers/net/wireless/rndis_wlan.c:2902:23: warning: array subscript 0 is outside array bounds of ‘struct ndis_80211_auth_request[0]’ [-Warray-bounds=] This helps with the ongoing efforts to tighten the FORTIFY_SOURCE routines on memcpy() and help us make progress towards globally enabling -fstrict-flex-arrays=3 [1]. Link: https://github.com/KSPP/linux/issues/21 Link: https://github.com/KSPP/linux/issues/274 Link: https://gcc.gnu.org/pipermail/gcc-patches/2022-October/602902.html [1] Signed-off-by:
-
Dan Carpenter authored
The rndis_bss_info_update() returns type bool. Change the NULL returns to false. Signed-off-by:
Dan Carpenter <error27@gmail.com> Signed-off-by:
-
Tom Rix authored
clang with W=1 reports drivers/net/wireless/realtek/rtw88/pci.c:92:21: error: unused function 'rtw_pci_get_tx_desc' [-Werror,-Wunused-function] static inline void *rtw_pci_get_tx_desc(struct rtw_pci_tx_ring *tx_ring, u8 idx) ^ This function is not used, so remove it. Signed-off-by: -
Christophe JAILLET authored
There is no point in allocating 'skb' and then freeing it if !channel. Make the sanity check first to slightly simplify the code. Signed-off-by:
Christophe JAILLET <christophe.jaillet@wanadoo.fr> Reviewed-by:
-
Tom Rix authored
clang with W=1 reports drivers/net/wireless/intel/ipw2x00/ipw2200.c:381:19: error: unused function '_ipw_read16' [-Werror,-Wunused-function] static inline u16 _ipw_read16(struct ipw_priv *ipw, unsigned long ofs) ^ This function and its wrapping marco are not used, so remove them. Signed-off-by: -
Ryder Lee authored
Similar to AP beacon, this enables the basic mesh EHT mode, including EHT operation IE and the fixed field of EHT operation information IE. As for the optional part (i.e. preamble puncturing bitmap) will be added in future patch. Tested-by:
Lian Chen <lian.chen@mediatek.com> Signed-off-by:
Ryder Lee <ryder.lee@mediatek.com> Link: https://lore.kernel.org/r/1e0ddb9001312451c3e99c4eed2072caf8075f61.1679935259.git.ryder.lee@mediatek.comSigned-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Johannes Berg authored
There are a few merge conflicts due to overlapping fixes and changes, merge wireless/main to fix them. Signed-off-by:
-
- 30 Mar, 2023 11 commits
-
-
Johannes Berg authored
Evidently Gregory sent this file but I (apparently every else) missed it entirely, remove that. Fixes: cf85123a ("wifi: iwlwifi: mvm: support enabling and disabling HW timestamping") Signed-off-by:
-
Johannes Berg authored
This function can be invoked for both MLO and non-MLO, so it must deal with multi-link correctly. Notable, on auth timeout, we'd otherwise get a warning due to the erroneous deflink usage in MLO cases. Signed-off-by:
Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20230329100040.b85f6052d51a.Iedfef4b4c4f3ca557aebc0093fdc3f5cfb49b507@changeidSigned-off-by:
-
Johannes Berg authored
The link management queues associated with the broadcast stations were forgotten and so the same queue was used with both broadcast stations. This leads to lost frames and warnings on cleanup and HW restart. Signed-off-by:
-
Johannes Berg authored
In the MLD code, we free probe_resp_data before we remove the MAC from the firmware, so we might receive another one from the device after freeing, and thus might leak it. Fix that by moving the free later. Signed-off-by:
-
Gregory Greenman authored
Start supporting API version 75 for AX devices. Signed-off-by:
-
Gregory Greenman authored
This field is used only for host TLC, so it can reside inside the corresponding lq_sta struct. Also, TLC lq_sta is cleared in iwl_mvm_rs_rate_init() upon association, but max_agg_bufsize is set earlier in iwl_mvm_sta_init(). Thus, place this field in the persistent part of lq_sta to retain its value. Signed-off-by:
-
Johannes Berg authored
By using the internal station add the station is installed in firmware with zeroed MAC addresses, which is wrong. Use the full installation function instead, to fill all data. Signed-off-by:
-
Gregory Greenman authored
Remove fw_active_links_num counter since we now have a bitmap of active links in vif. Also, update link activation status only when LINK_CONTEXT_MODIFY_ACTIVE bit set in changes parameter. Signed-off-by:
-
Gregory Greenman authored
Some mac parameters, such as HE support, can change at this stage. Update mac config before updating links. Signed-off-by:
-
Shaul Triebitz authored
For bcase/mcast tx frames use the link queue. Signed-off-by:
Shaul Triebitz <shaul.triebitz@intel.com> Signed-off-by:
-
Johannes Berg authored
Here we always have a link ID, not MAC ID, so clean up the naming. Signed-off-by:
-
