marcz Posted January 18, 2019 Posted January 18, 2019 I find quite strange that CONFIG_DM_MIRROR is not set in the kernel. It makes impossible under lvm to have logical volume mirror, and I just experienced an impossibility to make a pvmove to change one drive for another one. All my filesystems are built upon lvm, the only possibility would be to rebuild a kernel. But I find it quite heavy, since I have to platforms bananapi and espressobin; and it needs to be done for each release. But why no have built dm-mirror as module with the usual CONFIG_DM_MIRROR=m ?
Werner Posted January 18, 2019 Posted January 18, 2019 Maybe because no one had a need for it yet. You can build your own kernel and edit the config using the build script: https://github.com/armbian/build If you think this might be useful for others feel free to send a PR with the changes to the kernel config so all future builds include this.
Igor Posted September 24, 2020 Posted September 24, 2020 6 hours ago, Quardex said: We need this module https://docs.armbian.com/Process_Contribute/ https://docs.armbian.com/Developer-Guide_Build-Preparation/
Werner Posted September 24, 2020 Posted September 24, 2020 https://github.com/armbian/build/pull/2222 WIP. If you need it faster either donate or build yourself using the tools mentioned above. Either way give some feedback if it worked (the module or donation )
dolphs Posted September 24, 2020 Posted September 24, 2020 perhaps an idea using " compile.sh " and select particular option from kernel ( save .config in user space ) KERNEL_CONFIGURE=yes
Werner Posted September 24, 2020 Posted September 24, 2020 2 hours ago, Werner said: https://github.com/armbian/build/pull/2222 Done. Go checkout dm-mirror branch and build. 1
Quardex Posted September 24, 2020 Posted September 24, 2020 (edited) Thanks a lot all! Firstly will try compile.. Edited September 24, 2020 by Quardex
Quardex Posted September 28, 2020 Posted September 28, 2020 Tried build branch twice. Failed atm.. root@orangepione:~# modprobe dm-mirror modprobe: FATAL: Module dm-mirror not found in directory /lib/modules/5.8.11-sunxi Seems I need add that into my action. On 9/24/2020 at 11:51 AM, dolphs said: perhaps an idea using " compile.sh " and select particular option from kernel ( save .config in user space ) KERNEL_CONFIGURE=yes
Werner Posted September 29, 2020 Posted September 29, 2020 7 hours ago, Quardex said: Tried build branch twice. Failed atm.. https://github.com/armbian/build/blob/2b656aca508369e274f9e6573f020eca4296267b/config/kernel/linux-sunxi-current.config#L2179 modules is there. However there is not much what we can further do about that if - for example- there is an upstream issue with it.
dolphs Posted September 29, 2020 Posted September 29, 2020 @Quardex - OPiOne is a H3 board isn't it , well current branch should be fine indeed Nevertheless perhaps you can give kernel 5.9 a go ( currently rc7, production ready in about 2,5 weeks imho ) Try building with e.g. ( assuming Debian and no Desktop ) : ./compile.sh BRANCH=current RELEASE=buster BUILD_MINIMAL=yes BUILD_DESKTOP=no KERNEL_ONLY=no KERNEL_CONFIGURE=yes BOARD=orangepione Go to kernel configuration and select : " Device Drivers ", " <M> DAX: direct access to differentiated memory " Uhm DAX was in kernel 5.9 (rc5) an issue if I remember correctly, well your module should be here, also for 5.8 that is no difference : " Device Drivers " -> ", " Multiple devices driver support (RAID and LVM) ---> ", " <M> Mirror Target ". I just checked out and indeed it is not selected so enable <M> and give it a go. Perhaps more dependencies needed, I am not sure but you should be able to take your pick and build it according your needs. It should build and hopefully it helps you out
Werner Posted September 30, 2020 Posted September 30, 2020 dm harmonization has been pushed to master 1
Quardex Posted October 4, 2020 Posted October 4, 2020 (edited) Atm when I try compile master brunch I got that error: CC [M] drivers/net/wireless/rtl8189es/os_dep/linux/mlme_linux.o CC [M] drivers/net/wireless/rtl8189es/os_dep/linux/recv_linux.o CC [M] drivers/net/wireless/rtl8189es/os_dep/linux/ioctl_cfg80211.o [ error ] ERROR in function compile_kernel [ compilation.sh:411 ] [ error ] Kernel was not built [ @host ] [ o.k. ] Process terminated Tried twice on different computers. Edited October 4, 2020 by Quardex
Werner Posted October 4, 2020 Posted October 4, 2020 2 minutes ago, Quardex said: Atm when I try compile master brunch I got that error: CC [M] drivers/net/wireless/rtl8189es/os_dep/linux/mlme_linux.o CC [M] drivers/net/wireless/rtl8189es/os_dep/linux/recv_linux.o CC [M] drivers/net/wireless/rtl8189es/os_dep/linux/ioctl_cfg80211.o [ error ] ERROR in function compile_kernel [ compilation.sh:411 ] [ error ] Kernel was not built [ @host ] [ o.k. ] Process terminated Tried twice on different computers. Which board? Look at output/debug folder and check the compilation.txt for errors. Try with setting EXTRAWIFI=no
Quardex Posted October 4, 2020 Posted October 4, 2020 (edited) 4 hours ago, Werner said: Which board? Look at output/debug folder and check the compilation.txt for errors. Try with setting EXTRAWIFI=no orangepione With "EXTRAWIFI=no" compiled fine Log without EXTRAWIFI=no: output.log: Displaying message: ERROR in function compile_kernel compilation.sh:411 err Displaying message: Kernel was not built @host err Displaying message: Process terminated info compilation.log: In file included from ./include/net/cfg80211.h:21, from drivers/net/wireless/realtek/rtl8723cs/include/osdep_service_linux.h:92, from drivers/net/wireless/realtek/rtl8723cs/include/osdep_service.h:41, from drivers/net/wireless/realtek/rtl8723cs/include/drv_types.h:32, from drivers/net/wireless/realtek/rtl8723cs/hal/phydm/rtl8703b/../phydm_types.h:172, from drivers/net/wireless/realtek/rtl8723cs/hal/phydm/rtl8703b/../phydm_precomp.h:24, from drivers/net/wireless/realtek/rtl8723cs/hal/phydm/rtl8703b/halphyrf_8703b.c:22: ./include/linux/ieee80211.h:1600: note: this is the location of the previous definition 1600 | #define IEEE80211_MAX_AMPDU_BUF 0x100 | In file included from drivers/net/wireless/realtek/rtl8723cs/include/drv_types.h:35, from drivers/net/wireless/realtek/rtl8723cs/core/rtw_mp.c:22: drivers/net/wireless/realtek/rtl8723cs/include/wifi.h:1009: warning: "IEEE80211_MAX_AMPDU_BUF" redefined 1009 | #define IEEE80211_MAX_AMPDU_BUF 0x40 | In file included from ./include/net/cfg80211.h:21, from drivers/net/wireless/realtek/rtl8723cs/include/osdep_service_linux.h:92, from drivers/net/wireless/realtek/rtl8723cs/include/osdep_service.h:41, from drivers/net/wireless/realtek/rtl8723cs/include/drv_types.h:32, from drivers/net/wireless/realtek/rtl8723cs/core/rtw_mp.c:22: ./include/linux/ieee80211.h:1600: note: this is the location of the previous definition 1600 | #define IEEE80211_MAX_AMPDU_BUF 0x100 | In file included from drivers/net/wireless/realtek/rtl8723cs/include/drv_types.h:35, from drivers/net/wireless/realtek/rtl8723cs/core/rtw_mp_ioctl.c:22: drivers/net/wireless/realtek/rtl8723cs/include/wifi.h:1009: warning: "IEEE80211_MAX_AMPDU_BUF" redefined 1009 | #define IEEE80211_MAX_AMPDU_BUF 0x40 | In file included from ./include/net/cfg80211.h:21, from drivers/net/wireless/realtek/rtl8723cs/include/osdep_service_linux.h:92, from drivers/net/wireless/realtek/rtl8723cs/include/osdep_service.h:41, from drivers/net/wireless/realtek/rtl8723cs/include/drv_types.h:32, from drivers/net/wireless/realtek/rtl8723cs/core/rtw_mp_ioctl.c:22: ./include/linux/ieee80211.h:1600: note: this is the location of the previous definition 1600 | #define IEEE80211_MAX_AMPDU_BUF 0x100 | drivers/net/wireless/rtl8189es/core/rtw_mlme_ext.c: In function ‘mgt_dispatcher’: drivers/net/wireless/rtl8189es/core/rtw_mlme_ext.c:1093:6: warning: this statement may fall through [-Wimplicit-fallthrough=] 1093 | if(check_fwstate(pmlmepriv, WIFI_AP_STATE) == _TRUE) | ^ drivers/net/wireless/rtl8189es/core/rtw_mlme_ext.c:1098:3: note: here 1098 | case WIFI_ASSOCREQ: | ^~~~ drivers/net/wireless/rtl8189es/core/rtw_iol.c: In function ‘rtw_IOL_cmd_buf_dump’: drivers/net/wireless/rtl8189es/core/rtw_iol.c:261:3: warning: this ‘if’ clause does not guard... [-Wmisleading-indentation] 261 | if(j%32 ==0) printk("\n");j++; | ^~ drivers/net/wireless/rtl8189es/core/rtw_iol.c:261:29: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the ‘if’ 261 | if(j%32 ==0) printk("\n");j++; | ^ drivers/net/wireless/rtl8189es/os_dep/linux/ioctl_cfg80211.c:6549:37: error: ‘cfg80211_rtw_mgmt_frame_register’ undeclared here (not in a function) 6549 | .update_mgmt_frame_registrations = cfg80211_rtw_mgmt_frame_register, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ make[4]: *** [scripts/Makefile.build:281: drivers/net/wireless/rtl8189es/os_dep/linux/ioctl_cfg80211.o] Error 1 make[3]: *** [scripts/Makefile.build:497: drivers/net/wireless/rtl8189es] Error 2 make[2]: *** [scripts/Makefile.build:497: drivers/net/wireless] Error 2 make[1]: *** [scripts/Makefile.build:497: drivers/net] Error 2 make: *** [Makefile:1754: drivers] Error 2 Edited October 4, 2020 by Quardex
Werner Posted October 4, 2020 Posted October 4, 2020 Try to clear your cache directory. My builds from today seem fine: https://werner.armbian.de/Okt04/
Quardex Posted October 4, 2020 Posted October 4, 2020 (edited) Ok will try Edited October 4, 2020 by Quardex
Quardex Posted October 4, 2020 Posted October 4, 2020 (edited) After clearing the cache the same error: In file included from drivers/net/wireless/realtek/rtl8723cs/include/drv_types.h:35, from drivers/net/wireless/realtek/rtl8723cs/core/rtw_mp_ioctl.c:22: drivers/net/wireless/realtek/rtl8723cs/include/wifi.h:1009: warning: "IEEE80211_MAX_AMPDU_BUF" redefined 1009 | #define IEEE80211_MAX_AMPDU_BUF 0x40 | In file included from ./include/net/cfg80211.h:21, from drivers/net/wireless/realtek/rtl8723cs/include/osdep_service_linux.h:92, from drivers/net/wireless/realtek/rtl8723cs/include/osdep_service.h:41, from drivers/net/wireless/realtek/rtl8723cs/include/drv_types.h:32, from drivers/net/wireless/realtek/rtl8723cs/core/rtw_mp_ioctl.c:22: ./include/linux/ieee80211.h:1600: note: this is the location of the previous definition 1600 | #define IEEE80211_MAX_AMPDU_BUF 0x100 | drivers/net/wireless/rtl8189es/core/rtw_mlme_ext.c: In function ‘mgt_dispatcher’: drivers/net/wireless/rtl8189es/core/rtw_mlme_ext.c:1093:6: warning: this statement may fall through [-Wimplicit-fallthrough=] 1093 | if(check_fwstate(pmlmepriv, WIFI_AP_STATE) == _TRUE) | ^ drivers/net/wireless/rtl8189es/core/rtw_mlme_ext.c:1098:3: note: here 1098 | case WIFI_ASSOCREQ: | ^~~~ drivers/net/wireless/rtl8189es/core/rtw_iol.c: In function ‘rtw_IOL_cmd_buf_dump’: drivers/net/wireless/rtl8189es/core/rtw_iol.c:261:3: warning: this ‘if’ clause does not guard... [-Wmisleading-indentation] 261 | if(j%32 ==0) printk("\n");j++; | ^~ drivers/net/wireless/rtl8189es/core/rtw_iol.c:261:29: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the ‘if’ 261 | if(j%32 ==0) printk("\n");j++; | ^ drivers/net/wireless/rtl8189es/os_dep/linux/ioctl_cfg80211.c:6549:37: error: ‘cfg80211_rtw_mgmt_frame_register’ undeclared here (not in a function) 6549 | .update_mgmt_frame_registrations = cfg80211_rtw_mgmt_frame_register, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ make[4]: *** [scripts/Makefile.build:281: drivers/net/wireless/rtl8189es/os_dep/linux/ioctl_cfg80211.o] Error 1 make[3]: *** [scripts/Makefile.build:497: drivers/net/wireless/rtl8189es] Error 2 make[2]: *** [scripts/Makefile.build:497: drivers/net/wireless] Error 2 make[1]: *** [scripts/Makefile.build:497: drivers/net] Error 2 make: *** [Makefile:1754: drivers] Error 2 Need someone else to check it. Edited October 4, 2020 by Quardex
Werner Posted October 4, 2020 Posted October 4, 2020 No idea. Just built kernel package for OPi1 flawless including WiFi drivers.
Igor Posted October 4, 2020 Posted October 4, 2020 57 minutes ago, Quardex said: After clearing the cache the same error: - are you using official master branch? - is it up to date? git pull We have an automated job which runs whenever things are changed to detect errors and there are none atm.
Quardex Posted October 4, 2020 Posted October 4, 2020 (edited) Quote - is it up to date? git pull Nope, it wasn't. After update the compilation was successful and the module worked properly. Thank you all! Last question: https://www.armbian.com/donate/ Is this variant actual to make donation? Edited October 4, 2020 by Quardex
Werner Posted October 5, 2020 Posted October 5, 2020 You can also consider a subscription https://forum.armbian.com/subscriptions/
Recommended Posts