Guest Dazzozo Posted September 27, 2012 Report Share Posted September 27, 2012 Daz, Samsung Tuna uses bcm4330 and from what i can tell, they use the bcmdhd driver just fine. https://github.com/C.../BoardConfig.mk # Wifi related defines BOARD_WPA_SUPPLICANT_DRIVER := NL80211 WPA_SUPPLICANT_VERSION := VER_0_8_X BOARD_WPA_SUPPLICANT_PRIVATE_LIB := lib_driver_cmd_bcmdhd BOARD_HOSTAPD_DRIVER := NL80211 BOARD_HOSTAPD_PRIVATE_LIB := lib_driver_cmd_bcmdhd BOARD_WLAN_DEVICE := bcmdhd WIFI_DRIVER_FW_PATH_PARAM := "/sys/module/bcmdhd/parameters/firmware_path" #WIFI_DRIVER_MODULE_PATH := "/system/lib/modules/bcmdhd.ko" WIFI_DRIVER_FW_PATH_STA := "/vendor/firmware/fw_bcmdhd.bin" WIFI_DRIVER_FW_PATH_P2P := "/vendor/firmware/fw_bcmdhd_p2p.bin" WIFI_DRIVER_FW_PATH_AP := "/vendor/firmware/fw_bcmdhd_apsta.bin" [/CODE] https://github.com/C...ybean/device.mk [CODE]BOARD_WLAN_DEVICE_REV := bcm4330_b2 WIFI_BAND := 802_11_ABG $(call inherit-product-if-exists, hardware/broadcom/wlan/bcmdhd/firmware/bcm4330/device-bcm.mk) Kernel source is here: https://github.com/C...ng_tuna/tree/jb Would love to help but no device tree made public yet from you Worth a try, revving up a build now Link to comment Share on other sites More sharing options...
Guest timfimjim Posted September 27, 2012 Report Share Posted September 27, 2012 The longer we wait for the next release, the more things that are getting fixed and the better the ROM will be. Until a new release comes out (depending on how much time you have), either use another ROM, or look into development and maybe you could help Dazzozo? :) Link to comment Share on other sites More sharing options...
Guest Curlie Posted September 27, 2012 Report Share Posted September 27, 2012 thats true ;) Link to comment Share on other sites More sharing options...
Guest unaszplodrmann Posted September 27, 2012 Report Share Posted September 27, 2012 LOL Another unicorn just died. ... Good luck with the Broadcom stuff. At least they are providing source these days. Link to comment Share on other sites More sharing options...
Guest Simon O Posted September 27, 2012 Report Share Posted September 27, 2012 I've not had much luck with bcm4330 open source, case in point Blade2 :P Yeah I didn't have much look on the blade2 myself with that but at the time a few people suggested using the Samsung code. I've just seen somebody else suggest using it too. In the meantime hex editing will work although it's rather messy. I so need to pick up a G300 cheap. Guess who missed the Tesco ones.. yeah :( Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 27, 2012 Report Share Posted September 27, 2012 Yeah, I hex edited module_layout, but then that led to another 20 address differences, symbols in cfg80211, etc. I'll probably just suck it up and hex edit it all. I think all the addresses can be found in System.map, though I'm not sure how many are actually in dhd_4330.ko... I tried brcm80211, I ended up with two modules and didn't manage to get anywhere with it. I also tried the Tuna method (pretty much word for word) and couldn't get any interface online at all. There are no modules involved there, it all seems to build in to the kernel itself. What happened to the Tesco deal? Has it ended? Get oneeee, more devs :P Link to comment Share on other sites More sharing options...
Guest tcpaulh Posted September 27, 2012 Report Share Posted September 27, 2012 Can still be had cheap Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 27, 2012 Report Share Posted September 27, 2012 Well waddaya know. https://github.com/Dazzozo/android_kernel_huawei_u8815/commit/4b3a8d883b3bff7f7f269b929f5936612a613d49 We are version 5.90.125.120.2, and so is this. Link to comment Share on other sites More sharing options...
Guest mack_ Posted September 27, 2012 Report Share Posted September 27, 2012 Well waddaya know. https://github.com/Dazzozo/android_kernel_huawei_u8815/commit/4b3a8d883b3bff7f7f269b929f5936612a613d49 We are version 5.90.125.120.2, and so is this. Good thing?! Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 28, 2012 Report Share Posted September 28, 2012 Even with the exact same version: <4>[1204, insmod] [ 431.605674] [DHD] <4>[1204, insmod] [ 431.605684] Dongle Host Driver, version 5.90.125.120.2 <4>[1204, insmod] [ 431.605691] Compiled in drivers/net/wireless/bcmdhd on Sep27 2012 at 18:01:18 <4>[1204, insmod] [ 443.604564] [DHD] dhd_module_init: sdio_register_driver timeout Hex hacking, here we come. :P Link to comment Share on other sites More sharing options...
Guest tcpaulh Posted September 28, 2012 Report Share Posted September 28, 2012 Arse biscuits Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 28, 2012 Report Share Posted September 28, 2012 Right, I've taken a look, pretty trivial task, I might need to compile a dummy bcmdhd so I can get the addresses for the cfg80211 symbols. The rest of them are just staring me in the face in System.map. The System.map I'm working from: http://g300.thebronasium.com/tmp/System.map Link to comment Share on other sites More sharing options...
Guest popoyaya Posted September 28, 2012 Report Share Posted September 28, 2012 How's the wifi voodoo going Daz? Having tasted the buttery elixir of Jelly Bean I'm loathe to go back to sticky old ICS! Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 28, 2012 Report Share Posted September 28, 2012 I got distracted and found the vendor/qcom/msm7627a repo, and I'm building with some optimisations Qcom use for 7x27a. I also merged recent updates to the kernel which broke things, and then I fixed them. :P I'll work on the module once I see it through that all this builds still. Link to comment Share on other sites More sharing options...
Guest wrtease Posted September 28, 2012 Report Share Posted September 28, 2012 keep up the good work man your getting their.i have some stuff for the g330 ,even a kernal 3.021 .hope you get this up and running good luck Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 28, 2012 Report Share Posted September 28, 2012 Right, everything builds and boots, feels much nicer, time to work on the module. :P Link to comment Share on other sites More sharing options...
Guest b4da55 Posted September 28, 2012 Report Share Posted September 28, 2012 Great work daz :D Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 28, 2012 Report Share Posted September 28, 2012 I found a little endian big endian converter which is saving my ass here. :P bcmdhd helps too, I dumped modversions on it, and they're all in the same order as dhd_4330. Copy address -> paste it in to converter -> copy output -> replace it in dhd_4330. Long night. :D Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 28, 2012 Report Share Posted September 28, 2012 Done all but three addresses. <4>[1468, Thread-147] [ 5337.642574] dhd: disagrees about version of symbol kobject_uevent_env <4>[1468, Thread-147] [ 5337.642649] dhd: Unknown symbol kobject_uevent_env (err -22) <4>[1468, Thread-147] [ 5337.642759] dhd: disagrees about version of symbol init_net <4>[1468, Thread-147] [ 5337.642809] dhd: Unknown symbol init_net (err -22) <4>[1468, Thread-147] [ 5337.644151] dhd: disagrees about version of symbol dev_get_by_name <4>[1468, Thread-147] [ 5337.644204] dhd: Unknown symbol dev_get_by_name (err -22) Why these? These aren't in bcmdhd, so I'm going to have to find the addresses somewhere else. I can't manage to make System.map ones work. Edit: I managed to find init_net in cfg80211. The other two are really tricky, I can't find them in any module. Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 28, 2012 Report Share Posted September 28, 2012 I got them, but RIP. <4>[1505, Thread-154] [ 6385.288416] dhd_module_init: Enter <4>[1505, Thread-154] [ 6385.288597] ## wifi_probe <4>[1505, Thread-154] [ 6385.288609] wifi_set_power = 1 <3>[1505, Thread-154] [ 6385.441589] bcm_wifi_set_power: wifi power successed to pull up <4>[1505, Thread-154] [ 6385.441614] dhd_enable_mmchost_polling: /sys/devices/platform/msm_sdcc.2/polling <4>[1505, Thread-154] [ 6385.441672] store_polling : no need to enable polling for slot 2 (as host->pdev_id) <4>[1505, Thread-154] [ 6385.476377] <4>[1505, Thread-154] [ 6385.476389] Dongle Host Driver, version 5.90.125.120.2 <6>[1479, kworker/u:1] [ 6385.743191] mmc2: new high speed SDIO card at address0001 <4>[1479, kworker/u:1] [ 6385.746047] dhd_customer_oob_irq_map oob gpio is : 48;wifi slot is :2 <4>[1479, kworker/u:1] [ 6385.746162] alloc static buf at dc540000! <4>[1479, kworker/u:1] [ 6385.746516] dhd_common_init:fw_path = /data/misc/wifi/load/firmware.bin <4>[1479, kworker/u:1] [ 6385.751782] DHD: dongle ram size is set to 294912(orig 294912) <4>[1479, kworker/u:1] [ 6385.752067] ------------[ cut here ]------------ <4>[1479, kworker/u:1] [ 6385.752106] WARNING: at /home/Dazzozo/source/cm10/kernel/huawei/u8815/net/wireless/util.c:136 ieee80211_set_bitrate_flags+0x16c/0x180[cfg80211]() <4>[1479, kworker/u:1] [ 6385.752121] Modules linked in: dhd(+) cfg80211 [last unloaded: cfg80211] <4>[1479, kworker/u:1] [ 6385.752169] [<c0043344>] (unwind_backtrace+0x0/0x11c)from [<c00b2f7c>] (warn_slowpath_common+0x4c/0x64) <4>[1479, kworker/u:1] [ 6385.752192] [<c00b2f7c>] (warn_slowpath_common+0x4c/0x64) from [<c00b2fac>] (warn_slowpath_null+0x18/0x1c) <4>[1479, kworker/u:1] [ 6385.752221] [<c00b2fac>] (warn_slowpath_null+0x18/0x1c) from [<bf7a2740>] (ieee80211_set_bitrate_flags+0x16c/0x180 [cfg80211]) <4>[1479, kworker/u:1] [ 6385.752261] [<bf7a2740>] (ieee80211_set_bitrate_flags+0x16c/0x180 [cfg80211]) from [<bf7a0f54>] (wiphy_register+0x320/0x414 [cfg80211]) <4>[1479, kworker/u:1] [ 6385.752326] [<bf7a0f54>] (wiphy_register+0x320/0x414 [cfg80211]) from [<bf7ec434>] (wl_cfg80211_attach+0x208/0xae8 [dhd]) <4>[1479, kworker/u:1] [ 6385.752392] [<bf7ec434>] (wl_cfg80211_attach+0x208/0xae8 [dhd]) from [<bf7c82f4>] (dhd_attach+0x2b4/0x6e8 [dhd]) <4>[1479, kworker/u:1] [ 6385.752454] [<bf7c82f4>] (dhd_attach+0x2b4/0x6e8 [dhd]) from [<bf7d6bf0>] (dhdsdio_probe+0x300/0x60c [dhd]) <4>[1479, kworker/u:1] [ 6385.752521] [<bf7d6bf0>] (dhdsdio_probe+0x300/0x60c [dhd]) from [<bf7deaf0>] (bcmsdh_probe+0xe4/0x138 [dhd]) <4>[1479, kworker/u:1] [ 6385.752569] [<bf7deaf0>] (bcmsdh_probe+0xe4/0x138 [dhd]) from [<c03d2f9c>] (sdio_bus_probe+0x88/0xec) <4>[1479, kworker/u:1] [ 6385.752596] [<c03d2f9c>] (sdio_bus_probe+0x88/0xec) from [<c02b378c>] (driver_probe_device+0x144/0x28c) <4>[1479, kworker/u:1] [ 6385.752624] [<c02b378c>] (driver_probe_device+0x144/0x28c) from [<c02b2548>] (bus_for_each_drv+0x4c/0x84) <4>[1479, kworker/u:1] [ 6385.752646] [<c02b2548>] (bus_for_each_drv+0x4c/0x84)from [<c02b35cc>] (device_attach+0x74/0xa0) <4>[1479, kworker/u:1] [ 6385.752666] [<c02b35cc>] (device_attach+0x74/0xa0) from [<c02b2dc0>] (bus_probe_device+0x24/0x40) <4>[1479, kworker/u:1] [ 6385.752687] [<c02b2dc0>] (bus_probe_device+0x24/0x40)from [<c02b1394>] (device_add+0x3f8/0x574) <4>[1479, kworker/u:1] [ 6385.752709] [<c02b1394>] (device_add+0x3f8/0x574) from [<c03d318c>] (sdio_add_func+0x34/0x4c) <4>[1479, kworker/u:1] [ 6385.752732] [<c03d318c>] (sdio_add_func+0x34/0x4c) from [<c03d27e8>] (mmc_attach_sdio+0x278/0x324) <4>[1479, kworker/u:1] [ 6385.752754] [<c03d27e8>] (mmc_attach_sdio+0x278/0x324) from [<c03cc8d8>] (mmc_rescan+0x274/0x31c) <4>[1479, kworker/u:1] [ 6385.752781] [<c03cc8d8>] (mmc_rescan+0x274/0x31c) from [<c00c7340>] (process_one_work+0x1f8/0x34c) <4>[1479, kworker/u:1] [ 6385.752806] [<c00c7340>] (process_one_work+0x1f8/0x34c) from [<c00c7e2c>] (worker_thread+0x1fc/0x370) <4>[1479, kworker/u:1] [ 6385.752829] [<c00c7e2c>] (worker_thread+0x1fc/0x370) from [<c00cb480>] (kthread+0x7c/0x84) <4>[1479, kworker/u:1] [ 6385.752851] [<c00cb480>] (kthread+0x7c/0x84) from [<c003ec14>] (kernel_thread_exit+0x0/0x8) <4>[1479, kworker/u:1] [ 6385.752866] ---[ end trace f2c4dc69874a0e10 ]--- <4>[1479, kworker/u:1] [ 6385.765784] Broadcom Dongle Host Driver: register interface [eth0] MAC: 00:90:4c:11:22:33 <4>[1505, Thread-154] [ 6385.776051] wifi_set_power = 0 <3>[1505, Thread-154] [ 6385.777091] bcm_wifi_set_power: wifi power successed to pull down <4>[1505, Thread-154] [ 6385.777101] =========== WLAN placed in RESET ======== <4>[1511, wpa_supplicant] [ 6385.915737] <4>[1511, wpa_supplicant] [ 6385.915752] Dongle Host Driver, version 5.90.125.120.2 <4>[1511, wpa_supplicant] [ 6385.915769] wl_android_wifi_on in <4>[1511, wpa_supplicant] [ 6385.915781] wifi_set_power = 1 <3>[1511, wpa_supplicant] [ 6386.236074] bcm_wifi_set_power: wifi power successed to pull up <4>[1511, wpa_supplicant] [ 6386.236099] =========== WLAN going back to live ======== <4>[1511, wpa_supplicant] [ 6386.236111] sdio_reset_comm(): <4>[1511, wpa_supplicant] [ 6386.420549] dhd_preinit_ioctls: use MAC address innvram 00:90:4c:ce:43:30 <4>[1511, wpa_supplicant] [ 6386.420574] Firmware up: op_mode=1, Broadcom Dongle Host Driver mac=00:90:4c:ce:43:30 <4>[1511, wpa_supplicant] [ 6386.897467] wl_android_wifi_off in <3>[1511, wpa_supplicant] [ 6386.902556] bcm_wifi_set_power: wifi power successed to pull down <4>[1511, wpa_supplicant] [ 6386.902612] =========== WLAN placed in RESET ======== <4>[1511, wpa_supplicant] [ 6386.908156] do_exit: exit code=65280 <4>[1511, wpa_supplicant] [ 6386.908506] wpa_supplicant used greatest stack depth: 4768 bytes left [/CODE] Unless there's any other solution I'm probably going to have to revert parts of the net code in the kernel to 3.0.8. Link to comment Share on other sites More sharing options...
Guest tcpaulh Posted September 28, 2012 Report Share Posted September 28, 2012 So it should look more like ? What's good to search for in google? I can't read the output well. Where's the first failure point? Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 28, 2012 Report Share Posted September 28, 2012 It begins to start printing the error at ------------[ cut here ]------------ You can relate that to the other log on that page. Currently I'm just seeing if I can find any bcm4330 drivers that have identical print messages. Link to comment Share on other sites More sharing options...
Guest tcpaulh Posted September 29, 2012 Report Share Posted September 29, 2012 (edited) Wild stab in the dark https://aur.archlinu...ts=all&detail=1 I give myself 1% chance of that being useful Something about gzipped modules Edited September 29, 2012 by tcpaulh Link to comment Share on other sites More sharing options...
Guest Dazzozo Posted September 29, 2012 Report Share Posted September 29, 2012 Right, working on Bluetooth. Scanning ... 68:1A:B2:CC:BE:0E Crescent[/CODE] Looks like a good sign. Link to comment Share on other sites More sharing options...
Guest oldfella Posted September 29, 2012 Report Share Posted September 29, 2012 Right, working on Bluetooth. Scanning ... 68:1A:B2:CC:BE:0E Crescent Looks like a good sign. excitement approaches ! Link to comment Share on other sites More sharing options...
Recommended Posts