Spookcity / ROMS-G800F

10 stars 0 forks source link

Android P development 2 #27

Closed Spookcity closed 7 months ago

Spookcity commented 5 years ago

A continuation of #26

polaurent commented 5 years ago

@Spookcity , you're last release works quite correctly , but i encounter a problem with MMS and emoji in SMS have you the same ?

Mikahlo commented 5 years ago

The last LineageOS 16 is fluid and works pretty well. Calling, sms, data etc. works. One thing i noticed is that there is something going on with wifi. If wifi is on and you put the device into sleep mode, wifi gets randomly disconnected after some time for no obvious reason. It connects as soon as you turn the screen on. wifi.txt

I also spotted some weird symbols in the log: 02-11 14:52:07.458 E/KernelWakelockReader( 2297): Failed to parse proc line: ???(??????@???$ 02-11 14:52:07.459 E/KernelWakelockReader( 2297): Failed to parse proc line: ?????$

From dmesg is this one: [ 23.799976] [2] 2360 WARNING: at /mnt/shit/AndroidSources/los/kernel/samsung/kminilte/drivers/regulator/core.c:1556 _regulator_disable+0x124/0x23c() [ 23.800195] [2] 2360 unbalanced disables for VLED_3.3V I guess this makes the phone charge faster? xD

Thanks again for doing this stuff! Android 9 on a device that's 5 years old and was only supported till Android 6 is great!

Spookcity commented 5 years ago

@polaurent Interesting,I honestly don't use SMS or MMS at all really,so I'm not sure. What exactly is the issue?

@Mikahlo Thanks a lot for the log and info. Strangely enough I do not have the issue of WiFi disconnecting. I will look into my commits and see if perhaps I did something that may have caused this. I'm not too sure what the KernelWakelockReader error is about either. It seems weird indeed. I will look into it. And you are most welcome. I am just glad that I have learned enough over the last year plus that I am able to do this.

Side note: I have made some progress with our storage issues I believe. Downloading album cover art etc via certain music apps now works as well as an issue I had with copying items from extSD or lan/samba folder to internal storage in MiXplorer. Hopefully other issues will be solved as well. I want to look over my commit later,but will push to my kernel repo soon after.

polaurent commented 5 years ago

@Spookcity : MMS are unable to be downloaded ( photos ) , in received SMS all emoji are "?" .

Spookcity commented 5 years ago

@polaurent Yes,that is def a problem. Is this something that had worked for you in previous builds or had you not tried? I looked over my commits and saw none that should have had any effect on this.

Also,This commit has fixed a number of storage problems for me. So simple,yet so easily overlooked. I am going to upload the new boot image (for LOS 16 only) if anyone would like to try and perhaps give some feedback.Thanks! Download - BootImage

Just make sure to select "install image" in TWRP and be SURE to select boot partition!!

polaurent commented 5 years ago

@Spookcity :Concerning your Bootimage, it as not make any change but by the way it's my fault, i forget to check the restore default APN, the previous auto-created was buggy ( but looks correct, greyed and not editable), not sure that it's not a post boot error.

Another thing, as i get an SD card installed as Internal storage, each time after reboot and before unlocking the phone, i got a message concerning the SD card telling it's not installed ( seems to be a feature cause the card is scrambled but curious). As there is no option to decide if the sd as to be user or system scrambled , no reason to have this message.

polaurent commented 5 years ago

@Spookcity : i think it as something to be in reference of the multi-user schema and an existing "guest" profile. Concerning SD, as each profile as it's scrambling schema, it's obvious that the System could not reach the SD card before login but it as to know the case and should not give any warning.

last report, i loose sometimes the data connection activity ( think following roaming ), i've to reactivate it manually.

Mikahlo commented 5 years ago

@Spookcity : tested the boot image, copied stuff with MiX, internal and external sd, FTP'd stuff around, made backups, edited files and saved my game in Andor's Trail xD Congrats! This commit should solve issues in some apps.

Spookcity commented 5 years ago

@polaurent I'm not fully sure I understand what is going on. Any chance you can provide a log?

@Mikahlo Great! Many thanks for testing and providing feedback. I will try to do some looking into the WiFi issue as well as the strange KernelWakelockReader error soon.

Mikahlo commented 5 years ago

@Spookcity : no worries, I take logs frequently to figure out if the wifi issue is a problem with android or my router. The disconnect happens quiet often so I should be able to make a better guess on what's going on. I really do not know anything about the wakelock stuff. It could be connected with the higher battery drain especially when the modem is used. My experience so far is that the battery life is the same as the android 7 builds when the phone is in flight mode. As soon as the flight mode is turned off, the battery drain is huge. Nonetheless, many thanks again for the great work you do!

Spookcity commented 5 years ago

Yeah,it is certainly possible that the wakelock error is causing some unnecessary battery drain. Strangely,my battery life is pretty great on Pie builds. EIther way, I stumbled upon This by accident really and it appears that this is all that was needed to fix our WiFi hotspot. So I am going to disable some things in our kernels default config that were enable trying to sort this out. We shall see if it changes anything.

Oleggluk commented 5 years ago

Hello.)

The creators of LineageOS updated to Android Pie devices five years ago .. Samsung Galaxy S5 LTE (G900AZ / F / M / R4 / R7 / T / V) ... https://lineageos.org/Changelog-22/

Perhaps there are ideas for correcting errors on s 5mini? (Google Translate)

Spookcity commented 5 years ago

Hey @Oleggluk Good to see you around stranger! Thanks for the link I will check it out. I will say though,that LOS 16 is already very stable imo. Much more so than any Oreo was. The only real issues that I am aware of is NFC and HW composition. NFC may never be fixed and the HW composition is seeming more and more like our libdisplay.so or hwcomposer.exynos3.so needs to be hex edited to fix.

Mikahlo commented 5 years ago

wifi issues

These lines are from the latest logcat I took. Maybe one of them leads to a hint of what is going on. I also changed the modem version which changed nothing.

03-26 06:41:56.306 E/android.hardware.wifi@1.0-service( 2151): Failed to register radio mode change callback 03-26 06:41:56.308 E/android.hardware.wifi@1.0-service( 2151): Failed to set DFS flag; DFS channels may be unavailable. 03-26 06:41:56.842 E/android.hardware.wifi@1.0-service( 2151): Failed to get driver version: SUCCESS

03-26 19:01:41.527 E/WifiScanner( 2337): listener cannot be found 03-26 19:06:14.666 E/WifiVendorHal( 2337): getApfCapabilities(l.1737) failed {.code = ERROR_NOT_AVAILABLE, .description = } 03-26 19:06:14.676 E/WifiVendorHal( 2337): getWifiLinkLayerStats(l.937) failed {.code = ERROR_NOT_AVAILABLE, .description = } 03-26 19:06:15.005 E/WifiVendorHal( 2337): stopRssiMonitoring(l.2404) failed {.code = ERROR_NOT_AVAILABLE, .description = }

Spookcity commented 5 years ago

@Mikahlo Thanks a lot for the log. Such a weird issue. You are not alone as another user reported WiFi disconnects on XDA. It's strange because my WiFi is very stable. I will try to track this down. I think I may move the android.hardware.wifi HIDL back to version 1.0. I think the newer versions have addition features that our device doesn't support anyway. Perhaps with some luck this will solve it.

Mikahlo commented 5 years ago

wifi insanity

Quick update: nothing. The issue could be anything. Drivers, kernel, network services or applications, ipv4 or ipv6. I have some parts of a log and dmesg which show symptoms but no reasons.

04-14 10:17:43.073 I/wpa_supplicant( 2952): wlan0: WPA: Group rekeying completed with ::::73:84 [GTK=CCMP] 04-14 10:17:43.088 W/IpClient.wlan0( 2294): [IpReachabilityMonitor] WARN ALERT neighbor went from: NeighborEvent{@58206670,RTM_NEWNEIGH,if=9,192.168.178.1,NUD_REACHABLE,[::::73:82]} to: NeighborEvent{@58208608,RTM_NEWNEIGH,if=9,192.168.178.1,NUD_FAILED,[null]} 04-14 10:17:43.088 W/IpReachabilityMonitor( 2294): FAILURE: LOST_PROVISIONING, NeighborEvent{@58208608,RTM_NEWNEIGH,if=9,192.168.178.1,NUD_FAILED,[null]} 04-14 10:17:43.105 I/wpa_supplicant( 2952): wlan0: CTRL-EVENT-DISCONNECTED bssid=::::73:84 reason=3 locally_generated=1 04-14 10:17:43.107 E/wpa_supplicant( 2952): nl80211: Failed to open /proc/sys/net/ipv4/conf/wlan0/drop_unicast_in_l2_multicast: No such file or directory 04-14 10:17:43.107 E/wpa_supplicant( 2952): nl80211: Failed to set IPv4 unicast in multicast filter 04-14 10:17:43.111 E/wpa_supplicant( 2952): nl80211: Failed to open /proc/sys/net/ipv4/conf/wlan0/drop_unicast_in_l2_multicast: No such file or directory 04-14 10:17:43.111 E/wpa_supplicant( 2952): nl80211: Failed to set IPv4 unicast in multicast filter 04-14 10:17:43.112 D/WifiStateMachine( 2294): WifiStateMachine: Leaving Connected state 04-14 10:17:43.113 I/EthernetTracker( 2294): interfaceLinkStateChanged, iface: wlan0, up: true 04-14 10:17:43.117 E/WifiVendorHal( 2294): stopRssiMonitoring(l.2404) failed {.code = ERROR_NOT_AVAILABLE, .description = } 04-14 10:17:43.119 I/EthernetTracker( 2294): interfaceLinkStateChanged, iface: wlan0, up: true 04-14 10:17:43.119 D/DhcpClient( 2294): doQuit 04-14 10:17:43.122 D/ConnectivityService( 2294): NetworkAgentInfo [WIFI () - 132] EVENT_NETWORK_INFO_CHANGED, going from CONNECTED to DISCONNECTED 04-14 10:17:43.122 D/ConnectivityService( 2294): NetworkAgentInfo [WIFI () - 132] got DISCONNECTED, was satisfying 8 04-14 10:17:43.142 D/WifiStateMachine( 2294): Start Disconnecting Watchdog 31 04-14 10:17:43.159 E/WifiVendorHal( 2294): stopRssiMonitoring(l.2404) failed {.code = ERROR_NOT_AVAILABLE, .description = } 04-14 10:17:43.160 D/DhcpClient( 2294): onQuitting 04-14 10:17:43.166 D/CommandListener( 2071): Setting iface cfg 04-14 10:17:43.170 I/WifiStateMachine( 2294): disconnectedstate enter 04-14 10:17:43.187 W/wificond( 2144): Copy constructor is only used for unit tests 04-14 10:17:43.187 D/WifiNetworkAgent( 2294): NetworkAgent: NetworkAgent channel lost 04-14 10:17:43.187 W/wificond( 2144): Failed to get NL80211_ATTR_EXT_FEATURES 04-14 10:17:43.189 W/wificond( 2144): Copy constructor is only used for unit tests 04-14 10:17:43.192 E/WifiVendorHal( 2294): getSupportedFeatureSet(l.1131) failed {.code = ERROR_NOT_AVAILABLE, .description = } 04-14 10:17:43.194 D/DhcpClient( 2294): Receive thread stopped 04-14 10:17:43.194 W/wificond( 2144): Failed to get NL80211_ATTR_EXT_FEATURES 04-14 10:17:43.195 D/CommandListener( 2071): Clearing all IP addresses on wlan0 04-14 10:17:43.197 W/wificond( 2144): Copy constructor is only used for unit tests 04-14 10:17:43.197 W/wificond( 2144): Failed to get NL80211_ATTR_EXT_FEATURES 04-14 10:17:43.201 E/WifiVendorHal( 2294): getSupportedFeatureSet(l.1138) failed {.code = ERROR_NOT_AVAILABLE, .description = } 04-14 10:17:43.216 E/BatteryExternalStatsWorker( 2294): no controller energy info supplied for wifi

[20518.277154] [0] 5785 CFG80211-INFO2) wl_cfg80211_disconnect : Reason 3 [20518.295634] [0] 5786 CFG80211-INFO2) wl_is_linkdown : Link down Reason : WLC_E_LINK [20518.296052] [0] 5786 link down if wlan0 may call cfg80211_disconnected. event : 16, reason=2 from e8:73:84 [20518.297747] [0] 32333 regulatory is not upadted via restore_regulatory_settings. [20518.375553] [2] 2829 cyttsp5_i2c_adapter 3-0024: TSP DVFS Off 2 [20518.406321] [2] 959 cyttsp5_i2c_adapter 3-0024: P [0] [20518.406632] [2] 959 cyttsp5_i2c_adapter 3-0024: TSP DVFS On 2 0 1 [20518.427204] [3] 2144 wl_android_wifi_off in [20518.432382] [0] 2144 wake disabled for irq 304 [20518.433056] [0] 2144 wifi_platform_set_power = 0 [20518.433196] [0] 2144 ------------------------------------------------ [20518.433393] [0] 2144 ------------------------------------------------ [20518.433587] [0] 2144 brcm_wlan_power Enter: power off [20518.435841] [0] 2144 gpio_table = [6] [20518.491480] [2] 959 cyttsp5_i2c_adapter 3-0024: R [0] [20518.506670] [2] 959 cyttsp5_i2c_adapter 3-0024: P [0] [20518.506967] [2] 959 cyttsp5_i2c_adapter 3-0024: TSP DVFS On 2 1 1 [20518.555034] [2] 959 cyttsp5_i2c_adapter 3-0024: R [0] [20518.555678] [0] 1 init: Received control message 'stop' for 'wpa_supplicant' from pid: 2144 (/system/bin/wificond) [20518.556011] [0] 1 init: Sending signal 9 to service 'wpa_supplicant' (pid 5785) process group... [20518.563292] [0] 1 libprocessgroup: Successfully killed process cgroup uid 0 pid 5785 in 6ms [20518.567911] [2] 959 cyttsp5_i2c_adapter 3-0024: cyttsp5_mt_lift_all: [20518.602253] [0] 1 init: Service 'wpa_supplicant' (pid 5785) received signal 9 [20518.709521] [2] 2829 cyttsp5_i2c_adapter 3-0024: change_dvfs_lock called! [20518.856327] [2] 2829 cyttsp5_i2c_adapter 3-0024: TSP DVFS Off 2 [20518.868828] [3] 2094 dhd_prot_ioctl : bus is down. we have nothing to do [20518.869084] [3] 2094 dhd_wl_ioctl: WLC_GET_VAR: arp_hostip, ret = -1 [20518.869232] [3] 2094 dhd_prot_ioctl : bus is down. we have nothing to do [20518.869457] [3] 2094 dhd_wl_ioctl: WLC_SET_VAR: arp_hostip_clear, ret = -1 [20518.869606] [3] 2094 dhd_aoe_hostip_clr failed code -1 [20518.869822] [3] 2094 aoe_update_host_ipv4_table failed

[20525.613099] [3] 2111 Dongle Host Driver, version 1.141.64.27 (security_patch) [20525.613151] [3] 2111 Compiled in drivers/net/wireless/bcmdhd on Apr 11 2019 at 12:06:05 [20525.613551] [3] 2111 wl_android_wifi_on in [20525.613746] [3] 2111 wifi_platform_set_power = 1 [20525.613869] [3] 2111 ------------------------------------------------ [20525.614081] [3] 2111 ------------------------------------------------ [20525.614297] [3] 2111 brcm_wlan_power Enter: power on [20525.652048] [2] 959 cyttsp5_i2c_adapter 3-0024: R [0] [20525.664961] [2] 959 cyttsp5_i2c_adapter 3-0024: cyttsp5_mt_lift_all: [20525.790553] [2] 2829 cyttsp5_i2c_adapter 3-0024: change_dvfs_lock called! [20525.855406] [0] 2111 gpio_table = [6] [20525.955832] [2] 2829 cyttsp5_i2c_adapter 3-0024: TSP DVFS Off 2 [20526.060525] [0] 2111 sdio_reset_comm(): [20526.061088] [0] 2111 mmc_host mmc1: Bus speed (slot 0) = 25000000Hz (slot req 400000Hz, actual 390625HZ div = 32) [20526.081234] [0] 2111 mmc_host mmc1: Bus speed (slot 0) = 50000000Hz (slot req 50000000Hz, actual 50000000HZ div = 0) [20526.096321] [0] 2111 F1 signature OK, socitype:0x1 chip:0x4334 rev:0x3 pkg:0x2 [20526.098911] [0] 2111 DHD: dongle ram size is set to 524288(orig 524288) at 0x0 [20526.103373] [0] 2111 CHIP VER = [0x3] [20526.103578] [0] 2111 ----- CHIP bcm4334_B2 ----- [20526.103921] [0] 2111 dhdsdio_download_firmware: firmware path=/vendor/etc/wifi/bcmdhd_sta.bin_b2, nvram path=/system/etc/wifi/nvram_net.txt [20526.249508] [0] 2111 dhdsdio_write_vars: Download, Upload and compare of NVRAM succeeded. [20526.598435] [0] 2111 dhd_bus_init: enable 0x06, ready 0x06 (waited 0us) [20526.599133] [0] 2111 wake enabled for irq 304 [20526.599886] [0] 2010 type=1302 audit(1555229490.817:8392): item=0 name="/vendor/bin/hw/wpa_supplicant" inode=3161 dev=103:0a mode=0100755 ouid=0 ogid=2000 rdev=00:00 obj=u:object_r:hal_wifi_supplicant_default_exec:s0 [20526.600225] [0] 2010 type=1400 audit(1555229491.117:8395): avc: denied { search } for pid=2111 comm="e" name="wifi" dev="mmcblk0p3" ino=20 scontext=u:r:hal_wifi_default:s0 tcontext=u:object_r:wifi_efs_file:s0 tclass=dir permissive=1 [20526.601683] [0] 2010 type=1400 audit(1555229491.117:8395): avc: denied { search } for pid=2111 comm="e" name="wifi" dev="mmcblk0p3" ino=20 scontext=u:r:hal_wifi_default:s0 tcontext=u:object_r:wifi_efs_file:s0 tclass=dir permissive=1 [20526.602108] [0] 2010 type=1400 audit(1555229491.117:8395): avc: denied { read } for pid=2111 comm="e" name=".mac.info" dev="mmcblk0p3" ino=35 scontext=u:r:hal_wifi_default:s0 tcontext=u:object_r:wifi_efs_file:s0 tclass=file permissive=1 [20526.602709] [0] 2010 type=1400 audit(1555229491.117:8395): avc: denied { read } for pid=2111 comm="e" name=".mac.info" dev="mmcblk0p3" ino=35 scontext=u:r:hal_wifi_default:s0 tcontext=u:object_r:wifi_efs_file:s0 tclass=file permissive=1 [20526.603075] [0] 2010 type=1400 audit(1555229491.117:8395): avc: denied { open } for pid=2111 comm="e" name=".mac.info" dev="mmcblk0p3" ino=35 scontext=u:r:hal_wifi_default:s0 tcontext=u:object_r:wifi_efs_file:s0 tclass=file permissive=1 [20526.616969] [1] 2111 Firmware up: op_mode=0x0015, MAC=3c:61:19 [20526.623293] [1] 2111 dhdcdc_set_ioctl: SET PM to 2 [20526.624314] [1] 2111 [WIFI_SEC] sec_control_pm: /data/.psm.info open failed, so set PM to 2 [20526.640352] [1] 2111 Firmware version = wl0: Jan 17 2018 19:36:42 version 6.10.58.116.5 (B2 Station/P2P) FWID 01-aa54582f [20526.641507] [0] 2010 type=1400 audit(1555229491.117:8395): avc: denied { open } for pid=2111 comm="e" name=".mac.info" dev="mmcblk0p3" ino=35 scontext=u:r:hal_wifi_default:s0 tcontext=u:object_r:wifi_efs_file:s0 tclass=file permissive=1 [20526.642130] [0] 2010 type=1400 audit(1555229491.117:8395): avc: denied { read } for pid=2111 comm="e" name=".wifiver.info" dev="mmcblk0p21" ino=11 scontext=u:r:hal_wifi_default:s0 tcontext=u:object_r:wifi_data_file:s0 tclass=file permissive=1 [20526.643407] [1] 2111 [WIFI_SEC] .wifiver.info already saved. [20526.644533] [1] 2111 dhd_wl_ioctl: WLC_SET_VAR: ampdu_hostreorder, ret = -23 [20526.644709] [1] 2111 dhd_preinit_ioctls wl ampdu_hostreorder failed -23 [20526.646555] [1] 2111 dhd_wl_ioctl: WLC_GET_VAR: pfnlbest, ret = -23 [20526.647647] [1] 2111 dhd_wl_ioctl: WLC_SET_VAR: interworking, ret = -23 [20526.647818] [1] 2111 dhd_interworking_enable: enableing interworking failed, ret=-23 [20526.649271] [1] 2111 CFG80211-INFO2) wl_cfg80211_attach_post : p2p0: p2p_dev_addr=3e:61:19 [20526.711112] [0] 2010 type=1400 audit(1555229491.117:8395): avc: denied { read } for pid=2111 comm="e" name=".wifiver.info" dev="mmcblk0p21" ino=11 scontext=u:r:hal_wifi_default:s0 tcontext=u:object_r:wifi_data_file:s0 tclass=file permissive=1 [20526.711655] [0] 2010 type=1400 audit(1555229491.117:8395): avc: denied { open } for pid=2111 comm="e" name=".wifiver.info" dev="mmcblk0p21" ino=11 scontext=u:r:hal_wifi_default:s0 tcontext=u:object_r:wifi_data_file:s0 tclass=file permissive=1 [20526.733713] [1] 2111 wl_create_event_handler(): thread:wl_event_handler:b8a started [20526.733781] [3] 2954 CFG80211-INFO2) wl_event_handler : tsk Enter, tsk = 0xee067634 [20526.741811] [1] 2111 audit: audit_lost=43341 audit_rate_limit=5 audit_backlog_limit=64 [20526.742118] [1] 2111 audit: rate limit exceeded [20526.784274] [3] 1 init: Received control message 'interface_start' for 'android.hardware.wifi.offload@1.0::IOffload/default' from pid: 1967 (/system/bin/hwservicemanager) [20526.784741] [3] 1 init: Could not find service hosting interface android.hardware.wifi.offload@1.0::IOffload/default [20526.976600] [1] 2952 dhd_set_suspend: Remove extra suspend setting [20527.017839] [3] 2144 CFG80211-TRACE) __wl_cfg80211_scan : START SCAN [20529.776698] [1] 2954 CFG80211-TRACE) wl_escan_handler : SCAN COMPLETED: scanned AP count=11 [20529.927571] [1] 2952 dhd_wl_ioctl: WLC_IOCTL: cmd: 23, ret = -17 [20529.964565] [1] 2952 CFG80211-INFO2) wl_cfg80211_connect : Connectting withe8:73:84 channel (11) ssid "SOMENETWORK", len (6) [20529.964877] [1] 2952 [20530.012715] [1] 2954 wl_bss_connect_done succeeded with e8:73:84 [20530.021366] [1] 2954 dhdcdc_set_ioctl: SET PM to 0 [20530.022452] [1] 2206 ETHER_TYPE_802_1X: ver 2, type 3, replay 1 [20530.025172] [0] 18337 regulatory is not upadted via queue_regulatory_request. [20530.045061] [2] 2952 ETHER_TYPE_802_1X [TX]: ver 1, type 3, replay 1 [20530.062121] [1] 2206 ETHER_TYPE_802_1X: ver 2, type 3, replay 2 [20530.063152] [2] 2952 ETHER_TYPE_802_1X [TX]: ver 1, type 3, replay 2 [20530.102347] [2] 2952 CFG80211-TRACE) wl_cfg80211_set_btcoex_dhcp : DHCP is complete [20530.114721] [1] 2954 wl_bss_connect_done succeeded with e8:73:84 [20530.176880] [1] 2952 CFG80211-TRACE) wl_cfg80211_set_btcoex_dhcp : DHCP session starts [20530.266703] [1] 2952 dhd_set_suspend: Remove extra suspend setting [20530.282894] [1] 2952 dhdcdc_set_ioctl: SET PM to 0 [20530.327642] [1] 2952 dhdcdc_set_ioctl: SET PM to 2 [20530.332190] [1] 2952 CFG80211-TRACE) wl_cfg80211_set_btcoex_dhcp : DHCP is complete

Maybe the issues will get solved with future repo updates on the lineage side.

Mikahlo commented 5 years ago

@Spookcity another small update: since yesterday I'm using your AEX ROM from mid april and the wifi issue seems to be gone. No disconnections till now. I also updated TWRP to 3.3.0-0. Thanks again for your great work!

Spookcity commented 5 years ago

@Mikahlo Very interesting. That's good to know. If that's the case then it's obviously something within the Lineage sources. Please let me know if you have any disconnects while using AEX,because if not I'll try to look into what the differences might be within the 2 sources. I made 25-ish commits to the kernel related to wireless,but if it's something in the sources then it won't matter. I cherry picked the commits from other 3.4 kernels (klte,exynos5420) though,so I don't see why it will hurt to have them either.

Mikahlo commented 5 years ago

@Spookcity After testing some more the issue reoccurred. The difference is that wifi sometimes reconnects itself in standby and this connection is than stable. It's really a weird issue. You should be enable to reproduce this by connecting to a network, put the phone in standby and look after around 3 to 5 minutes. It should than be disconnected. The baseband/modem doesn't seem to influence this.

Spookcity commented 5 years ago

@Mikahlo Thanks a lot for the update. A bit of a shame as it may have been easier to figure out if it was purely related to LOS sources. I have tried to reproduce this over the last day or two (paying closer attention when I wake my phone) I am not able to. My WiFi never seems to disconnect. I can't think of any reason why I don't have this issue and you do. I use a static IP for my phone though. Although this is rather recent (last month or so) and I highly doubt this has anything to do with it. My WiFi was stable before this as well. I will continue to try to investigate when time permits.

skulko commented 5 years ago

Hello I noticed this in device/samsung/kminilte/overlay/frameworks/base/core/res/res/values/config.xml:

<bool name="config_batterySdCardAccessibility">true</bool>

(committed a long time ago...) But the SD Ccard can't be removed without removing the battery (and power off first). So shouldn't it be 'false'?

Spookcity commented 5 years ago

Hey @skulko Seems like you are correct. I am honestly not really sure what effect this has,but I will change this to false and commit later. Thanks! Any and all help is greatly appreciated!

skulko commented 5 years ago

@Spookcity Great thanks! though I agree, it probably won't change much.. Small question: Is the binary_patcher.sh to be applied? It is not mentioned in the build how-to, however it is in the patch (device/samsung/smdk3470-common/patch/) folder and quite recent...

Spookcity commented 5 years ago

@skulko Nope, no need to apply binary_patcher.sh as it's already applied.

skulko commented 5 years ago

Hi, There is this in smdk3470-common/Boardconfig.mk (at the very end of the file): WIFI_BAND := 802_11_ABG

But since the s5 mini also supports 802.11 N shouldn't it be WIFI_BAND := 802_11_ABGN ? There are only a few google results with abgn however.... I don't know if it's even a valid option...

Spookcity commented 5 years ago

@skulko I actually added the WiFI band for Pie builds. It was never there in the past. If I remember I will add "N" and do a build and see what happens,but in all honesty I am nearly positive this has no effect on anything.

skulko commented 4 years ago

Hey Spook thanks for your reply. I only had to add a small piece of the puzzle...

what I experienced so far is there is a little catch with NFC fix, it will only work if it is turned on during boot phase. Therefore if you power off with nfc turned off it won't work the next time. Maybe you have an idea how this can be resolved?

Spookcity commented 4 years ago

@skulko Ahh yes, this was a similar issue that I had if I recall correctly. I really dont know to be honest. I saw This commit and this is when I gave up as it is my understanding that it is just broken and beyond my knowledge to fix at this point. Wish I could be more helpful,sorry,but unless someone has worked on this I am afraid it is beyond my skill level.

skulko commented 4 years ago

I looked over it and I think the problem with this commit is our outdated extract-files.sh: on line 29 it would only look proprietary-files.txt but we would most likely need to add another (proprietary-files-pn547.txt). However doing this would (most likely, I don't know at which stage these files are executed.?) require further changes to setup-makefiles.sh, line 32 as this includes only proprietary-files.txt too. All this is probably required because in proprietary-files-pn547.txt there is a checksum: (I guess) vendor/lib/hw/nfc_nci.msm8974.so|7dcb79a385dd1155cb9b6310a3e7b85b7dc8db13

we could maybe try adding nci.msm8974.so directly to proprietary-files.txt and make changes accordingly.. but this leaves me with where to get this file...does s5mini share the same nfc fw with s5 (libpn547_fw)? I dont know if I got anything right or if it helps..

skulko commented 4 years ago

I wonder whether adding the firmware as per this commit would be enough.

Spookcity commented 4 years ago

@skulko I'll be honest, it's been such a long time since I've messed with this that I'm not really sure. I am pretty sure that we don't share the same NFC fw with the S5 (klte)
I did try using a prebuilt from our build of Lineage 15, obviously without any luck.I'm pretty sure what I tried was very similar to that commit, but again it's been awhile.

If I am able to find the time I will try to look into this some more and see if I'm able to help out in some way.

skulko commented 4 years ago

Uh found it. BoardConfig.mk says the chipset is s3fwrn5. It seems that lineageos dropped the support for device using these nfc chips because of it. If they can't fix it I most certainly cannot. But I wonder, what exactly was dropped in aosp(?) that we cannot use it anymore...

Spookcity commented 4 years ago

@skulko Yeah, I'm not sure what changed or happened that this was dropped. I was only able to find that it was "broken" So not super helpful.

I do know that when I was researching this I was only able to find I think 2 other devices that used the s3fwrn5 chipset. I believe the Note 4 and maybe the Galaxy S6 used a variant. Not sure. I think it was only ever used in a very small number of devices (possibly < 5) This makes it difficult and highly unlikely that someone more knowledgeable than myself is every going to fix this sadly.

skulko commented 4 years ago

yes sadly. I am sure you know a lot more than I do about this stuff...

this is the most recent patch I found implementing our chipset.

Spookcity commented 4 years ago

@skulko Yeah, idk. I only know this because I spent a good deal of time trying to get this to work, so I did a lot of searching. Gave myself a headache over it 😀

I think I may have stumbled upon that patch you posted as well. Not positive, but I tried anything I could possibly think of (or found) to get this working with no luck at all.

skulko commented 4 years ago

I believe you, I hope I am not straining your nerves with these (old) issues again...

Maybe the last thing I can come up with right now is that the file in libnfc-sec-hal.conf (last line): RFREG_FILE="/etc/sec_s3fwrn5_rfreg.bin" is missing on the device. I think we could get it from here

skulko commented 4 years ago

uh the file is not missing but rather the location in libnfc-sec-hal.conf should be

RFREG_FILE="/system/vendor/etc/sec_s3fwrn5_rfreg.bin"

Spookcity commented 4 years ago

@skulko No, not straining my nerves at all. Enough time has passed for me to be able to revisit this without annoyance 😀

I just wish I could be of more helpful, but I really have no new ideas on this. Perhaps you're onto something with the location? This is something I'm pretty sure I never noticed.

skulko commented 4 years ago

about the camera: I set TARGET_USES_GRALLOC1 := true in BoardConfigCommon.mk, and reverted the patch for switching graphics allocator to passthrough.

It 's a bit more stable, I guess but still no video rec'... I can do a pull request if you like...

Spookcity commented 4 years ago

Yeah, sounds good to me. I saw the NFC pull require also. I'll take care of that when I'm at my PC. The NFC issue is just a pain. I'm glad I'm not the only one who found this out haha Thanks for working on this!

skulko commented 4 years ago

hehe yes no problem.I know the feeling now :dagger: :) I also tried adding a hal for NFC but with no positive result.

Also for the camera i tried setting (both): media.stagefright.legacyencoder=true media.stagefright.less-secure=true , as described [here] (https://github.com/LiquidRemix/android_frameworks_av/commit/fcf3571951d7b0ee5d0bd43a2903e3f5f5b16c5f) But it makes matters worse; maybe because of the patcher?! I'll probably try with only 'media.stagefright.less-secure=true' set; but it seems my virtualbox is broken again :(

Spookcity commented 4 years ago

Hehe,yes. Seems like maybe I tried one or both of these flags at some point, but it's been so long I really don't recall. Maybe I didn't. Either way, it's been enough time that perhaps things have changed within the ROM sources to yield better results. It could be a patcher issue indeed. I'd have to have a look. Good luck with your virtualbox. My results with virtual machines have always been rather poor. Admittedly though I've not used them much.

skulko commented 4 years ago

I could probably do without virtualbox, now that no special java/jre is required. Sometimes it works but on other occasions its just a black screen when I boot up virtualbox.

Anyway, I believe this might fix the video recording issue. It looks like it wasn't ported to P/Lineage-16. Did you try it maybe at some point? however I don't know how to create a patch; my last attempts with git diff and using the output for the patch failed ...

Spookcity commented 4 years ago

I can't honestly say whether I tried this or not. Maybe it will fix it. That would be great! I created a patch. Hopefully it will apply OK. If not I will work on it tomorrow if I have more time. I will also try a build with it when time permits. I'll just go ahead push the patch and if it doesn't work or apply properly I'll fix or remove later. Thanks again for working on this!

https://github.com/Spookcity/android_device_samsung_smdk3470-common/commit/045ae8f3dd898cb69dc8a74e0144ab367d15abc4

skulko commented 4 years ago

darn,no luck. The patch didn't work in the first place so i changed static void passMetadataBuffer(MediaBuffer to static void passMetadataBuffer(MediaBufferBase so it would compile but, as kinda foreseeable, no working video rec. afterwards.

What I didn't try but might be wort it is: TARGET_PROCESS_SDK_VERSION_OVERRIDE := \ /system/bin/mediaserver=22

Spookcity commented 4 years ago

Hmm,well that's a shame. Guess we just can't be that lucky. Sorta like the NFC issue,I tried many things,but it's been so long I kinda forget exactly what I did. Just lots of searching on here and gerrit for similar devices to see if it was resolved. Obviously without luck. If time permits I will try to do a bit more searching into this.

skulko commented 4 years ago

I wonder what the purpose of hardware/samsung_slsi/exynos3470/Android.mk is, since the if clauses seem to prevent that the content/variables there are used. Because I think BOARD_USE_STOREMETADATA and maybe more could be useful.

regards

Spookcity commented 4 years ago

I'll be honest, I'm not really sure off hand. I'd have to have a closer look. Very interestingly, I once did a build with the whole samsung_slsi...directory excluded by accident. I don't remember noticing any issues. I found that very strange, but forgot about it and never looked into it. I suppose there had to be some audio issues at the very least without it, but I don't remember.

skulko commented 4 years ago

Hmm this is odd but interesting indeed. Because what I see is that in smdk3470-common/BoardConfigCommon.mk TARGET_PRODUCT_PLATFORM is 'exynos3' but Android.mk in hardware/samsung_slsi/exynos3470 requires exynos5: ifeq($(TARGET_BOARD_PLATFORM),exynos5) . What I think is that this was just fine when lineage was still cm, because the paths in smdk3470-common/libshims/libExynosOMX/Android.mk refer to non-existing folders like hardware/samsung_slsi-cm. So I don't see how -- likely -- required files for exynos3 get included, e.g. hardware/samsung_slsi/exynos3470/exynos_omx/codecs/{audio,video}/exynos3/

Spookcity commented 4 years ago

Interesting indeed. I think you are probably onto something there. This would also explain why I noticed no ill effects, at least in a short time testing, with this directory excluded. I should have investigated this further when this happened, but got distracted I guess. I'll have to try and take a look when I have some time.