We’re running a custom yocto build so I understand support will be very limited, but we’re having issues with the WiFi dying about once a week (range from daily to less often) on a particular board.
This is the console output when it happens
[ 4513.844028] mmc0: Timeout waiting for hardware interrupt. retries left=0 opcode=0
[ 4513.851512] mmc0: sdhci: ============ SDHCI REGISTER DUMP ===========
[ 4513.857952] mmc0: sdhci: Sys addr: 0xe64f2000 | Version: 0x00000002
[ 4513.864389] mmc0: sdhci: Blk size: 0x000000c4 | Blk cnt: 0x00000001
[ 4513.870829] mmc0: sdhci: Argument: 0x100000c4 | Trn mode: 0x00000013
[ 4513.877268] mmc0: sdhci: Present: 0x01f8820e | Host ctl: 0x00000013
[ 4513.883706] mmc0: sdhci: Power: 0x00000002 | Blk gap: 0x00000080
[ 4513.890145] mmc0: sdhci: Wake-up: 0x00000008 | Clock: 0x0000003f
[ 4513.896582] mmc0: sdhci: Timeout: 0x0000008f | Int stat: 0x00000000
[ 4513.903019] mmc0: sdhci: Int enab: 0x117f100b | Sig enab: 0x117f100b
[ 4513.909456] mmc0: sdhci: ACmd stat: 0x00000000 | Slot int: 0x00000502
[ 4513.915894] mmc0: sdhci: Caps: 0x07eb0000 | Caps_1: 0x8000b407
[ 4513.922331] mmc0: sdhci: Cmd: 0x0000353a | Max curr: 0x00ffffff
[ 4513.928768] mmc0: sdhci: Resp[0]: 0x00002000 | Resp[1]: 0x00000000
[ 4513.935205] mmc0: sdhci: Resp[2]: 0x00000000 | Resp[3]: 0x00000000
[ 4513.941642] mmc0: sdhci: Host ctl2: 0x00000088
[ 4513.946083] mmc0: sdhci: ADMA Err: 0x00000003 | ADMA Ptr: 0xa4447204
[ 4513.952520] mmc0: sdhci-esdhc-imx: ========= ESDHC IMX DEBUG STATUS DUMP =========
[ 4513.960086] mmc0: sdhci-esdhc-imx: cmd debug status: 0x3100
[ 4513.965742] mmc0: sdhci-esdhc-imx: data debug status: 0x32a0
[ 4513.971484] mmc0: sdhci-esdhc-imx: trans debug status: 0x33a1
[ 4513.977314] mmc0: sdhci-esdhc-imx: dma debug status: 0x3400
[ 4513.982970] mmc0: sdhci-esdhc-imx: adma debug status: 0x35b4
[ 4513.988712] mmc0: sdhci-esdhc-imx: fifo debug status: 0x3680
[ 4513.994455] mmc0: sdhci-esdhc-imx: async fifo debug status: 0x3750
[ 4514.000718] mmc0: sdhci: ============================================
The dmesg buffer gets filled up with errors after that so I’m not able to get any messages from right before it happens (I’ve never caught it in time).
Here’s the output from a newly restarted system relating to mmc0/mwifiex:
[ 0.633892] mmc0: SDHCI controller on 30b40000.mmc [30b40000.mmc] using ADMA
[ 0.692678] mmc0: new ultra high speed SDR104 SDIO card at address 0001
[ 3.161700] mwifiex_sdio mmc0:0001:1: info: FW download over, size 623240 bytes
[ 4.100473] mwifiex_sdio mmc0:0001:1: WLAN FW is active
[ 4.100879] btmrvl_sdio mmc0:0001:2: sdio device tree data not available
[ 4.141294] mwifiex_sdio mmc0:0001:1: Unknown api_id: 3
[ 4.141303] mwifiex_sdio mmc0:0001:1: Unknown api_id: 4
[ 4.141307] mwifiex_sdio mmc0:0001:1: Unknown GET_HW_SPEC TLV type: 0x217
[ 4.171792] mwifiex_sdio mmc0:0001:1: info: MWIFIEX VERSION: mwifiex 1.0 (16.68.1.p197)
[ 4.171800] mwifiex_sdio mmc0:0001:1: driver_version = mwifiex 1.0 (16.68.1.p197)
[ 5.090946] mwifiex_sdio mmc0:0001:1: info: trying to associate to ‘XXXXXX’ bssid d0:21:f9:6d:e5:23
[ 5.117179] mwifiex_sdio mmc0:0001:1: info: associated to bssid d0:21:f9:6d:e5:23 successfully
uname:
Linux XXXXXX 5.4.193-5.3.0-devel+git.5a24da287b86 #1 SMP PREEMPT Wed Jun 29 16:15:02 UTC 2022 aarch64 aarch64 aarch64 GNU/Linux
Connected with connman, and turned of powersave (/usr/sbin/iw mlan0 get power_save)
Anyone else getting this?