Related
**New fun from Samsung**
I'll be editing my posts as scripts finish but I want to get a few things up now.
In an attempt to see how difficult it will be to port custom ROMs between the different Galaxy devises, I've started looking for unique files and diffs in the sources Samgung has released.
As far as unique files, the Vibrant is the only one
*The Vibrant Epic and I9000 sources still have these files*
http://www.simtec.co.uk/products/EB2410ITX/
T959 Only /linux-2.6.29/include/asm-arm/plat-s3c24xx/regs-iis.h
T959 Only /linux-2.6.29/include/asm-arm/plat-s3c24xx/regs-s3c2412-iis.h
*The I897 source now includes the A1026 and WM8994 files. They no longer need to be copied over. Yay?*
http://www.audience.com/about/1026.php
T959 Only /linux-2.6.29/sound/soc/codecs/A1026_dev.h
T959 Only /linux-2.6.29/sound/soc/codecs/A1026_i2c_drv.c
T959 Only /linux-2.6.29/sound/soc/codecs/A1026_i2c_drv.h
T959 Only /linux-2.6.29/sound/soc/codecs/A1026_regs.h
T959 Only /linux-2.6.29/sound/soc/codecs/audience.c
T959 Only /linux-2.6.29/sound/soc/codecs/audience.h
http://www.wolfsonmicro.com/products/audio_hubs/WM8994/
T959 Only /linux-2.6.29/sound/soc/codecs/ttymisc.c
T959 Only /linux-2.6.29/sound/soc/codecs/ttymisc.h
T959 Only /linux-2.6.29/sound/soc/codecs/wm8994_def_sw.h
T959 Only /linux-2.6.29/sound/soc/codecs/wm8994_kepler.c
T959 Only /linux-2.6.29/sound/soc/codecs/wm8994_kepler_sw.c
T959 Only /linux-2.6.29/sound/soc/codecs/wm8994_sw.c
which the Captivate should have as well (at least the codecs files, I think the test s3c24 files are just left over junk)
The Vibrant Kernel.tar looks like it now includes the files from 3d_module_GPL.tar.gz but the SGH-T959_OpenSource.tar stll includes it so I will have to look into this.
New Captivate Source only
NEW I897 ./linux-2.6.29/arch/arm/configs/kepler_rev02_defconfig
NEW I897 ./linux-2.6.29/drivers/input/keyboard/MMH_ST959_R01_V07_bin.c
NEW I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V14_bin.c
NEW I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V12_bin.c
NEW I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V08_bin.c
NEW I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V10_bin.c
NEW I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V15_bin.c
NEW I897 ./linux-2.6.29/include/config/aries/ver/b2.h
NEW I897 ./linux-2.6.29/include/config/kepler/audio
NEW I897 ./linux-2.6.29/include/config/kepler/audio/a1026.h
NEW I897 ./linux-2.6.29/include/config/kepler/ver/b2.h
NEW I897 ./linux-2.6.29/include/config/audio
NEW I897 ./linux-2.6.29/include/config/audio/hw
NEW I897 ./linux-2.6.29/include/config/audio/hw/rev02.h
NEW I897 ./linux-2.6.29/include/config/keypad/melfas
NEW I897 ./linux-2.6.29/include/config/keypad/melfas/touch.h
New Vibrant Source only
NEW T959 ./linux-2.6.29/arch/arm/configs/T959_rev05_defconfig
NEW T959 ./linux-2.6.29/include/config/onenand.h
NEW T959 ./linux-2.6.29/include/config/jupiter/ver/b5.h
NEW T959 ./linux-2.6.29/include/config/t959
NEW T959 ./linux-2.6.29/include/config/t959/ver
NEW T959 ./linux-2.6.29/include/config/t959/ver/b5.h
NEW T959 ./linux-2.6.29/include/config/s5pc110/kepler
NEW T959 ./linux-2.6.29/include/config/s5pc110/kepler/board.h
NEW T959 ./linux-2.6.29/include/config/s5pc110/t959
NEW T959 ./linux-2.6.29/include/config/s5pc110/t959/board.h
The only files common to both that have differences in them are
*this is still true the VIbrant received the update to the FSR files we did*
kernel/linux-2.6.29/.config
kernel/linux-2.6.29/sound/soc/codecs/wm8994_aries.c
kernel/modules/Makefile
kernel/linux-2.6.29/sound/soc/codecs/wm8994_aries.c
148c148
I897 #define TUNING_RECOGNITION_MAIN_AIF1ADCL_VOL 0xCB // 400h
---
T959 #define TUNING_RECOGNITION_MAIN_AIF1ADCL_VOL 0xD0 // 400h
I'm sure the volume change is critical to the well being of our phones
What makes the Captivate a Captivate and the Vibrant a Vibrant?
Captivate (and I9000) CONFIG_ARIES_VER_B3
Vibrant CONFIG_JUPITER_VER_B5
along with a bunch of networking, crypto... ok there are a lot of differences between the Captivate and Vibrant. Much more than between the I897 and the Galaxy S
Brain...hurts.....
The Captivate is now a CONFIG_ARIES_VER_B2 CONFIG_S5PC110_KEPLER_BOARD and CONFIG_KEPLER_VER_B2
and the VIbrant is now CONFIG_ARIES_VER_B3 CONFIG_S5PC110_T959_BOARD and CONFIG_T959_VER_B5
The I9000 is still a CONFIG_ARIES_VER_B3 right now because as of Sep 6th 2010 the source has not changed from, oh a few days ago when I posted this the first time.
The other parts of the .config files are now very close to each other. The VIbrant now has all the extra crypto and networking flags the Captivate had.
See post 11 for everything you ever wanted to know about the Captivate vs Vibrant .config
*I'm not sure how the qt602240 stuff works, I'll re add some stuff once I've had a chance to look at some of the changes between the old files and the new ones*
qt602240.c controls the touch screen and contains fun things
Taken from qt602240_ts.h
/* Orient */
#define QT602240_NORMAL 0x0
#define QT602240_DIAGONAL 0x1
#define QT602240_HORIZONTAL_FLIP 0x2
#define QT602240_ROTATED_90_COUNTER 0x3
#define QT602240_VERTICAL_FLIP 0x4
#define QT602240_ROTATED_90 0x5
#define QT602240_ROTATED_180 0x6
#define QT602240_DIAGONAL_COUNTER 0x7
unhelpful sets the orientation to 9(!) in his kernel which I think is what it is in the default android source.
*Old ramble about orientation deleted, I may re add it or something similar when I have more information*
OK wall 'o text over. Once I finish with the I9000 I'll add that stuff too.
Sherpa
I really appreciate you looking into this and hope you get a handle on it so that we could be able to indifferently use any firmware/kernel with just a few modifications, enable HSUPA and use different 3G bands without losing GPS functionality or auto-rotate.
Cheers.
This would be helpful for Eugene since he's having some issues with GPS on his rom.
I wonder, can you use this method to look inside the modem.bin for both? See what is different.
Or did I miss it in your wall of text?
madjsp said:
This would be helpful for Eugene since he's having some issues with GPS on his rom.
I wonder, can you use this method to look inside the modem.bin for both? See what is different.
Or did I miss it in your wall of text?
Click to expand...
Click to collapse
modem.bin isn't in there, it's not part of the source
But there *is* something in the kernel compilation that "checks" your modem.bin to see if it's an ATT one, if it fails that test you still get HSUPA but you lose audio on your voice calls (You just hear hissing). There's GOT to be something in there, I've been completely unsuccessful so far figuring out what though =/
I was hoping for someone to this, thanks alot mate.
I897 vs I9000
OK here are the unique files between the Captivate and the Galaxy
*fixed for Sep 6th source drop*
I897 ./linux-2.6.29/arch/arm/configs/kepler_rev02_defconfig
I897 ./linux-2.6.29/drivers/input/keyboard/MMH_ST959_R01_V07_bin.c
I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V08_bin.c
I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V10_bin.c
I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V12_bin.c
I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V14_bin.c
I897 ./linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V15_bin.c
I897 ./linux-2.6.29/include/config/audio
I897 ./linux-2.6.29/include/config/kepler
I897 ./linux-2.6.29/include/config/aries/ver/b2.h
I897 ./linux-2.6.29/include/config/keypad/melfas
I897 ./linux-2.6.29/include/media/isx005_platform.h
I897 ./linux-2.6.29/sound/soc/codecs/A1026_dev.h
I897 ./linux-2.6.29/sound/soc/codecs/A1026_i2c_drv.c
I897 ./linux-2.6.29/sound/soc/codecs/A1026_i2c_drv.h
I897 ./linux-2.6.29/sound/soc/codecs/A1026_regs.h
I897 ./linux-2.6.29/sound/soc/codecs/audience.c
I897 ./linux-2.6.29/sound/soc/codecs/audience.h
I897 ./linux-2.6.29/sound/soc/codecs/HAC.c
I897 ./linux-2.6.29/sound/soc/codecs/HAC.h
I897 ./linux-2.6.29/sound/soc/codecs/ttymisc.c
I897 ./linux-2.6.29/sound/soc/codecs/ttymisc.h
I897 ./linux-2.6.29/sound/soc/codecs/wm8994_def_behold3.h
I897 ./linux-2.6.29/sound/soc/codecs/wm8994_def_kepler.h
I897 ./linux-2.6.29/sound/soc/codecs/wm8994_def_sw.h
I897 ./linux-2.6.29/sound/soc/codecs/wm8994_kepler_sw.c
I897 ./linux-2.6.29/sound/soc/codecs/wm8994_kepler.c
I897 ./linux-2.6.29/sound/soc/codecs/wm8994_sw.c
I9000 ./linux-2.6.29/include/asm-arm/plat-s3c24xx/regs-s3c2412-iis.h
I9000 ./linux-2.6.29/include/asm-arm/plat-s3c24xx/regs-iis.h
The I9000 s3c24xx files are the same as the Vibrant and I'm still not sure what they are doing there. They are #include all over the place in arch/arm and sound/soc
There are a lot of files that are different between the USGSM phones (there are multiple comments using that as a reference) which is weird because there are fewer differences between the I900 .config and the I897 than the Vibrant.
These look like they just add new #ifdefs so I think they should work with the I9000 source
linux-2.6.29/arch/arm/Kconfig
linux-2.6.29/arch/arm/mach-s5pc110/include/mach/gpio-jupiter.h
linux-2.6.29/arch/arm/mach-s5pc110/include/mach/sec_jack.h
linux-2.6.29/arch/arm/mach-s5pc110/include/mach/sec_jack.h
linux-2.6.29/arch/arm/mach-s5pc110/Kconfig
linux-2.6.29/arch/arm/plat-s5pc11x/pm.c
linux-2.6.29/drivers/input/touchscreen/Kconfig
linux-2.6.29/drivers/media/video/Kconfig
linux-2.6.29/drivers/media/video/Makefile
linux-2.6.29/drivers/media/video/samsung/tv20_ap/s5p_tv_base.c
linux-2.6.29/drivers/mmc/core/core.c
linux-2.6.29/drivers/mmc/core/host.c
linux-2.6.29/drivers/mmc/host/sdhci.h
linux-2.6.29/drivers/mmc/host/sdhci-s3c.c
linux-2.6.29/drivers/sensor/compass/yamaha.c
linux-2.6.29/include/linux/mmc/host.h
linux-2.6.29/kernel/power/earlysuspend.c
linux-2.6.29/sound/soc/codecs/Makefile
linux-2.6.29/sound/soc/codecs/wm8994.h
modules/bcm4329/src/dhd/sys/dhd.h
modules/bcm4329/src/dhd/sys/dhd_linux.c
modules/bcm4329/src/dhd/sys/dhd_sdio.c
these files are different enough that I'm not sure they can be swapped with out some work
linux-2.6.29/arch/arm/mach-s5pc110/mach-common.c
linux-2.6.29/arch/arm/mach-s5pc110/mach-jupiter.c
linux-2.6.29/arch/arm/mach-s5pc110/sec_jack.c
linux-2.6.29/.config
linux-2.6.29/drivers/input/keyboard/cypress/cypress-touchkey.c
linux-2.6.29/drivers/input/keyboard/melfas-touchkey.c
linux-2.6.29/drivers/input/keyboard/s3c-keypad.c
linux-2.6.29/drivers/input/keyboard/s3c-keypad.h
linux-2.6.29/drivers/input/touchscreen/qt602240.c
linux-2.6.29/drivers/media/video/ce147.c
linux-2.6.29/drivers/media/video/samsung/mfc50/mfc_opr.c
linux-2.6.29/drivers/power/s5pc110_battery.c
linux-2.6.29/drivers/regulator/max8998.c
linux-2.6.29/drivers/rtc/alarm.c
linux-2.6.29/drivers/sensor/optical/gp2a.c
linux-2.6.29/drivers/staging/android/lowmemorykiller.c
linux-2.6.29/drivers/video/samsung/s3cfb_tl2796.c
linux-2.6.29/sound/soc/codecs/wm8994_aries.c
linux-2.6.29/sound/soc/codecs/wm8994.c
modules/bcm4329/src/include/epivers.h
modules/bcm4329/src/wl/sys/wl_iw.c
modules/dpram/dpram.c
modules/dpram/fsr/FSR_BML.h
modules/dpram/fsr/FSR_DBG.h
modules/dpram/fsr/FSR.h
modules/dpram/fsr/FSR_LLD_4K_OneNAND.h
modules/dpram/fsr/FSR_LLD_FlexOND.h
modules/dpram/fsr/FSR_LLD.h
modules/dpram/fsr/FSR_LLD_OneNAND.h
modules/dpram/fsr/FSR_OAM.h
modules/dpram/fsr/FSR_PAM.h
modules/dpram/fsr/FSR_PartitionID.h
modules/dpram/fsr/FSR_STL.h
it looks like we have an older version of FSR for some reason? All the headers in modules/dpram/fsr list
I897 Flex Sector Remapper : RFS_1.3.1_b060-LinuStoreIII_1.1.0_b022-FSR_1.1.1_b112_RC
I9000 Flex Sector Remapper : LinuStoreIII_1.2.0_b032-FSR_1.2.1p1_b129_RTM
modules/dpram/fsr/FSR_LLD_PureNAND.h is 1.2 in both. *shrug*
*We now have the same version as the I9000*
linux-2.6.29/drivers/sensor/optical/gp2a.c has only one change
I897 wake_lock_timeout(&prx_wake_lock, 5*HZ);
---
I9000 wake_lock_timeout(&prx_wake_lock, 3*HZ);
linux-2.6.29/drivers/media/video/ce147.c doesn't have a #ifdefined for the I9000 but there is a section to check if its a I897 or a T959
#if defined(CONFIG_S5PC110_KEPLER_BOARD)
unsigned char ce147_str_model[9] = "SGH-I897";
#endif
#if defined (CONFIG_S5PC110_T959_BOARD)
unsigned char ce147_str_model[9] = "SGH-T959";
#endif
---
#ifdefined for unsigned char ce147_str_model[9] = "GT-I9000";
Looks like the Captivate is not only a CONFIG_ARIES_VER_B3 but also CONFIG_S5PC110_KEPLER_BOARD and a Vibrant is CONFIG_JUPITER_VER_B5 and CONFIG_S5PC110_T959_BOARD which is interesting.
CONFIG_S5PC110_FLEMING_BOARD is also floating around in there but I'm not sure which phone it refers to.
The obvious next question is what did Samsung change vs the stock kernel I think I have a 2.1 around here somewhere.
For laughs I did a diff vs cm6 and there are 9138 files that are different and 3946 that are unique to the I897 source. Considering cm6 has 36260 files total (in the version I have) that shouldn't be too hard for those guys right? Right?
Ah well I'll be back once I have the 2.1 stuff.
Sherpa
Doh
I should have mentioned that quite a few of the differences between the USGSM source and the Galaxy have to do with the A1026 codecs
God lord hahaha the CM differences. This does give me a bit of a starting point for pruning through some of the files in search of the HSUPA fail though. Still... that's a lot of crap to go through!
I build my kernels from the T959 source. If compatibility with a different modem.bin differs it is either due to one of the board config options or something in the initramfs - remember that initramfs contains files that set properties and control startup.
Sent from my SAMSUNG-SGH-I897 using XDA App
Lol you guy seem close to all the info needed to build the ultimate rom for these phones maybe just a matter of weeks, just in time for froyo to get you guys started on something else.
Sent from my SAMSUNG-SGH-I897 using Tapatalk
Samsung just released new versions of the I897, T959 the source for the D700(Epic) and I'm checking the I9000 now. Ignore the previous information until I have a few passes at the new stuff.
I really need to reorganize my posts.
Sherpa
Just so this isn't a totally useless post, here are the differences between the new Captivate old Captivate .config
NEW I897 CONFIG_KEPLER_AUDIO_A1026=y
NEW I897 CONFIG_BLK_DEV_INITRD=y
NEW I897 CONFIG_INITRAMFS_SOURCE=""
NEW I897 CONFIG_ARIES_VER_B2=y
OLD I897 CONFIG_ARIES_VER_B3=y
NEW I897 CONFIG_KEPLER_VER_B2=y
OLD I897 CONFIG_KEPLER_VER_B0=y
NEW I897 CONFIG_KEYPAD_MELFAS_TOUCH=y
OLD I897 CONFIG_KEYPAD_CYPRESS_TOUCH=y
OLD I897 CONFIG_VIDEO_S5KA3DFX=y
NEW I897 CONFIG_AUDIO_HW_Rev02=y
The changes between the old version and the Sep 6th Version
Files ./I897/kernel/build_kernel.sh and ./OldI897/kernel/build_kernel.sh differ
Only in ./I897/kernel/linux-2.6.29/.config
Only in ./I897/kernel/linux-2.6.29/arch/arm/configs/kepler_rev02_defconfig
Only in ./I897/kernel/linux-2.6.29/drivers/input/keyboard/MMH_ST959_R01_V07_bin.c
Only in ./I897/kernel/linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V08_bin.c
Only in ./I897/kernel/linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V10_bin.c
Only in ./I897/kernel/linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V12_bin.c
Only in ./I897/kernel/linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V14_bin.c
Only in ./I897/kernel/linux-2.6.29/drivers/input/keyboard/VTW0014_R00_V15_bin.c
Only in ./I897/kernel/linux-2.6.29/include/linux/utsrelease.h
Only in ./I897/kernel/linux-2.6.29/include/linux/version.h
Only in ./I897/kernel/linux-2.6.29/scripts/basic/docproc
Only in ./I897/kernel/linux-2.6.29/scripts/basic/fixdep
Only in ./I897/kernel/linux-2.6.29/scripts/basic/hash
Only in ./I897/kernel/linux-2.6.29/scripts/conmakehash
Only in ./I897/kernel/linux-2.6.29/scripts/ihex2fw
Only in ./I897/kernel/linux-2.6.29/scripts/kallsyms
Only in ./I897/kernel/linux-2.6.29/scripts/kconfig/conf
Only in ./I897/kernel/linux-2.6.29/scripts/kconfig/lex.zconf.c
Only in ./I897/kernel/linux-2.6.29/scripts/kconfig/zconf.hash.c
Only in ./I897/kernel/linux-2.6.29/scripts/kconfig/zconf.tab.c
Only in ./I897/kernel/linux-2.6.29/scripts/mod/elfconfig.h
Only in ./I897/kernel/linux-2.6.29/scripts/mod/mk_elfconfig
Only in ./I897/kernel/linux-2.6.29/scripts/mod/modpost
Captivate vs Vibrant .config 9/6/10
There is just too much information to put all this stuff in one post so rather than try and mash it into the first one I'll start breaking things down by topic.
.config diff
I897 CONFIG_KEPLER_AUDIO_A1026=y
I897 CONFIG_ARIES_VER_B2=y
T959 CONFIG_ARIES_VER_B3=y
I897 CONFIG_S5PC110_KEPLER_BOARD=y
T959 CONFIG_S5PC110_T959_BOARD=y
I897 CONFIG_KEPLER_VER_B2=y
T959 CONFIG_T959_VER_B5=y
I897 CONFIG_KEYPAD_MELFAS_TOUCH=y
T959 CONFIG_KEYPAD_CYPRESS_TOUCH=y
I897 CONFIG_AUDIO_HW_Rev02=y
Thats all thats left, everything else is the same now.
*CONFIG_T959_VER_B5 is in the .config file but NOT any of the defconfigs. make a copy before running make clean/mrproper with the Vibrant source*
And the meat to the previous potatoes, all the files that reference the CONFIG
CONFIG_ARIES_VER_B2 (I879)
linux-2.6.29/arch/arm/configs/aries_rev02_defconfig
CONFIG_ARIES_VER_B2 (I879) & T959 CONFIG_ARIES_VER_B3 (T959)
linux-2.6.29/sound/soc/codecs/wm8994_aries.c
linux-2.6.29/sound/soc/codecs/wm8994_kepler.c
linux-2.6.29/sound/soc/codecs/wm8994_universal.c
linux-2.6.29/sound/soc/codecs/wm8994_kepler_sw.c
linux-2.6.29/arch/arm/mach-s5pc110/mach-jupiter.c
linux-2.6.29/arch/arm/mach-s5pc110/include/mach/gpio-jupiter.h
linux-2.6.29/arch/arm/configs/kepler_rev02_defconfig
linux-2.6.29/arch/arm/configs/aries_rev03_defconfig
linux-2.6.29/drivers/sensor/optical/gp2a.c
linux-2.6.29/drivers/media/video/s5ka3dfx.c
linux-2.6.29/drivers/input/keyboard/s3c-keypad.h
linux-2.6.29/drivers/input/keyboard/s3c-keypad.h
linux-2.6.29/drivers/input/touchscreen/qt602240.c
CONFIG_KEPLER_AUDIO_A1026 (I897)
linux-2.6.29/sound/soc/codecs/wm8994.c
linux-2.6.29/sound/soc/codecs/wm8994.h
linux-2.6.29/arch/arm/mach-s5pc110/mach-jupiter.c
linux-2.6.29/arch/arm/configs/kepler_rev02_defconfig
linux-2.6.29/arch/arm/plat-s5pc11x/pm.c
CONFIG_KEPLER_AUDIO_A1026 (I897) & CONFIG_AUDIO_HW_Rev02 (I897)
linux-2.6.29/arch/arm/configs/kepler_rev02_defconfig
*There are no files that have #ifdefs for AUDIO_HW_Rev02*
I897 CONFIG_S5PC110_KEPLER_BOARD (I897) & CONFIG_S5PC110_T959_BOARD (T959)
linux-2.6.29/arch/arm/mach-s5pc110/mach-jupiter.c
linux-2.6.29/arch/arm/mach-s5pc110/sec_jack.c
linux-2.6.29/arch/arm/configs/kepler_rev02_defconfig
linux-2.6.29/drivers/usb/gadget/fsa9480_i2c.c
linux-2.6.29/drivers/usb/gadget/adb_ums_acm_mtp_rndis.c
linux-2.6.29/drivers/usb/gadget/f_mass_storage.c
linux-2.6.29/drivers/regulator/max8998_function.c
linux-2.6.29/drivers/video/samsung/logo_rgb24_wvga_portrait.h
linux-2.6.29/drivers/sensor/compass/yamaha.c
linux-2.6.29/drivers/media/video/ce147.c
linux-2.6.29/drivers/media/video/samsung/tv20_ap/s5p_tv_base.c
linux-2.6.29/drivers/input/keyboard/melfas_download.c
linux-2.6.29/drivers/input/keyboard/cypress/cypress-touchkey.c
linux-2.6.29/drivers/input/keyboard/melfas-touchkey.c
linux-2.6.29/drivers/input/keyboard/s3c-keypad.h
linux-2.6.29/drivers/input/keyboard/s3c-keypad.c
linux-2.6.29/drivers/input/touchscreen/qt602240.c
linux-2.6.29/drivers/power/s5pc110_battery.h
linux-2.6.29/drivers/power/fuel_gauge.c
linux-2.6.29/drivers/power/s5pc110_battery.c
CONFIG_S5PC110_T959_BOARD (T959)
linux-2.6.29/arch/arm/mach-s5pc110/mach-common.c
linux-2.6.29/drivers/mmc/host/sdhci.h
linux-2.6.29/drivers/mmc/host/sdhci-s3c.c
linux-2.6.29/drivers/input/keyboard/cypress/issp_main.c
CONFIG_KEPLER_VER_B2 (I897)
linux-2.6.29/arch/arm/configs/kepler_rev02_defconfig
CONFIG_T959_VER_B5 (T959)
linux-2.6.29/drivers/input/keyboard/s3c-keypad.h
*CONFIG_T959_VER_B5 is in the .config file but NOT any of the defconfigs. make a copy before running make clean/mrproper with the Vibrant source*
CONFIG_KEPLER_VER_B2 (I897) & CONFIG_T959_VER_B5 (T959)
linux-2.6.29/arch/arm/mach-s5pc110/include/mach/max8998_function.h
linux-2.6.29/arch/arm/plat-s5pc11x/pm.c
linux-2.6.29/drivers/regulator/max8998_function.c
linux-2.6.29/drivers/input/keyboard/s3c-keypad.c
linux-2.6.29/drivers/power/s5pc110_battery.h
linux-2.6.29/drivers/power/fuel_gauge.c
linux-2.6.29/drivers/power/s5pc110_battery.c
*CONFIG_T959_VER_B5 is in the .config file but NOT any of the defconfigs. make a copy before running make clean/mrproper with the Vibrant source*
CONFIG_KEYPAD_MELFAS_TOUCH (I897)
linux-2.6.29/arch/arm/configs/aries_rev04_defconfig
linux-2.6.29/arch/arm/configs/aries_rev00_defconfig
linux-2.6.29/arch/arm/configs/aries_rev05_defconfig
linux-2.6.29/arch/arm/configs/aries_rev02_defconfig
linux-2.6.29/arch/arm/configs/aries_rev01_defconfig
*No files use this CONFIG yet*
CONFIG_KEYPAD_MELFAS_TOUCH (I897) & CONFIG_KEYPAD_CYPRESS_TOUCH IT959)
linux-2.6.29/arch/arm/configs/kepler_rev02_defconfig
linux-2.6.29/arch/arm/configs/aries_rev03_defconfig
*No files use this CONFIG yet*
Nap time.
Sherpa
I'll take a wild guess: one of those AUDIO .config differences must be the culprit for the voice not working with a non-ATT modem.bin.
Wouldn't it be relatively easy to just replace the files and .config references with the vibrant/i9000 ones and end up with a Captivate kernel with working audio along with HSUPA support from an i9000 modem.bin?
The again, would you be so kind to build a few kernels, each with a different configuration for the various audio related differences, for people to test and conclude what causes the incompatibility?
SetiroN said:
I'll take a wild guess: one of those AUDIO .config differences must be the culprit for the voice not working with a non-ATT modem.bin.
Wouldn't it be relatively easy to just replace the files and .config references with the vibrant/i9000 ones and end up with a Captivate kernel with working audio along with HSUPA support from an i9000 modem.bin?
The again, would you be so kind to build a few kernels, each with a different configuration for the various audio related differences, for people to test and conclude what causes the incompatibility?
Click to expand...
Click to collapse
I'm just an idea guy trying to put out as much information as I can although I'm pretty sure Zilch would try just about anything I or anyone else compiles I have a new theory about screen rotation but I need the new sources at work to check the CONFIG tags in the /drivers/sensor/compass/yamaha.c file. I figure the download should be done in a few days or so.
Sherpa
Or I could scroll up and see them from the list I already made.....
Guilty Just don't mess with the power regulation or anything else terrifying and we've got a deal!
TMC_Sherpa said:
I need the new sources at work to check the CONFIG tags in the /drivers/sensor/compass/yamaha.c file. I figure the download should be done in a few days or so.
...
Or I could scroll up and see them from the list I already made.....
Edit Reason: I'm an idiot
Click to expand...
Click to collapse
thanks man!
Zilch25 said:
Guilty Just don't mess with the power regulation or anything else terrifying and we've got a deal!
Click to expand...
Click to collapse
Well at least the rest of us will know how good a job Samsung is doing with their wear leveling. I'm surprised designer posted the Odin one-click software and not you
TMC_Sherpa said:
Well at least the rest of us will know how good a job Samsung is doing with their wear leveling. I'm surprised designer posted the Odin one-click software and not you
Click to expand...
Click to collapse
I'll have you know I haven't flashed my phone once today... as for how many times I've done so this month... eh I'd rather not say
Fun things about the Yamaha compass. This would be much easier if Samsung called the file something other than yamaha.c or put anything in the header other than Samsung 2006-2015 but Yamaha does make compass/accelerometer chips and this is the only file in the /drivers/sensors directory that includes our CONFIG information to build it differently depending on the phone.
This is only a guess at this point although I would love to see someone mess with these values and check the orientation.
Hey hey, what do you know, it does work. Thanks to xcaliburinhand for building a I9000 kernel with our rotation in it.
I9000 garray #define YAMAHA_GSENSOR_TRANSFORMATION \
{ { -1, 0, 0}, \
{ 0, -1, 0}, \
{ 0, 0, -1} }
I897 garray #define YAMAHA_GSENSOR_TRANSFORMATION_KEPLER \
{ { 0, 1, 0}, \
{ -1, 0, 0}, \
{ 0, 0, -1} }
T959 garray #define YAMAHA_GSENSOR_TRANSFORMATION_KEPLER \
{ { -1, 0, 0}, \
{ 0, -1, 0}, \
{ 0, 0, -1} }
I9000 marray #define YAMAHA_MSENSOR_TRANSFORMATION \
{ { -1, 0, 0}, \
{ 0, 1, 0}, \
{ 0 , 0 , -1} }
I897 marray #define YAMAHA_MSENSOR_TRANSFORMATION_KEPLER \
{ { 0, -1, 0}, \
{ -1, 0, 0}, \
{ 0 , 0 , -1} }
T959 marray #define YAMAHA_MSENSOR_TRANSFORMATION_KEPLER \
{ { -1, 0, 0}, \
{ 0, 1, 0}, \
{ 0 , 0 , -1} }
*No idea what these are for or why they need to be there*
I9000 marray_00 #define YAMAHA_MSENSOR_TRANSFORMATION_00 \
{ { 0, -1, 0}, \
{ -1, 0, 0}, \
{ 0 , 0 , -1} }
I897 marray_B5 #define YAMAHA_MSENSOR_TRANSFORMATION_KEPLER_B5 \
{ { 0, 1, 0}, \
{ -1, 0, 0}, \
{ 0 , 0 , 1} }
I9000 garray_emul #define YAMAHA_GSENSOR_TRANSFORMATION_EMUL \
{ { -1, 0, 0}, \
{ 0, 1, 0}, \
{ 0, 0, 1} }
I9000 marray_emul #define YAMAHA_MSENSOR_TRANSFORMATION_EMUL \
{ { 1, 0, 0}, \
{ 0, 1, 0}, \
{ 0 , 0 , 1} }
For some reason our marray and garray are different to the Vibrant and I9000 and I'm guessing that the chip is oriented differently on our boards for some reason.
This all seems kinda ridiculous, there are files for AK8973b which is a 3-axis compass and an accel directory for a SMB380 which is a 3-axis accelerometer. Leftovers or junk for future phones? *shrug*
Someone (not Zilch, we need his phone alive and well...or alive anyway) needs to crack their phone open and take some better pics. We need to look for the triple tuning fork or any other chippies that look interesting.
As long as I'm doing slightly annoying Captivate vs Vibrant stuff
melfas-touchkey.c & cypress-touchkey.c
#define DEVICE_NAME "melfas-touchkey"
#if defined(CONFIG_S5PC110_T959_BOARD)
static int touchkey_keycode[] = {NULL, KEY_BACK, KEY_ENTER, KEY_MENU, KEY_END}; // BEHOLD3
#else
static int touchkey_keycode[5] = {NULL, KEY_BACK, KEY_MENU, KEY_ENTER, KEY_END};
#endif
is why the keys are backwards
If only I could figure out how
Code:
static DEVICE_ATTR(brightness, S_IRUGO | S_IWUSR | S_IWOTH | S_IXOTH, NULL, touch_led_control);
static ssize_t touch_led_control(struct device *dev, struct device_attribute *attr, const char *buf, size_t size)
{
unsigned char data;
if(sscanf(buf, "%d\n", &data) == 1) {
printk("touch_led_control: %d \n", data);
i2c_touchkey_write(&data, 1);
touchkey_led_status = data;
}
else
printk("touch_led_control Error\n");
return size;
}
if (device_create_file(touchkey_update_device.this_device, &dev_attr_brightness) < 0)
{
printk("%s device_create_file fail dev_attr_touch_update\n",__FUNCTION__);
pr_err("Failed to create device file(%s)!\n", dev_attr_brightness.attr.name);
}
works someone could get the message notification project going again.
TMC_Sherpa said:
I9000 garray #define YAMAHA_GSENSOR_TRANSFORMATION \
{ { -1, 0, 0}, \
{ 0, -1, 0}, \
{ 0, 0, -1} }
I897 garray #define YAMAHA_GSENSOR_TRANSFORMATION_KEPLER \
{ { 0, 1, 0}, \
{ -1, 0, 0}, \
{ 0, 0, -1} }
T959 garray #define YAMAHA_GSENSOR_TRANSFORMATION_KEPLER \
{ { -1, 0, 0}, \
{ 0, -1, 0}, \
{ 0, 0, -1} }
I9000 marray #define YAMAHA_MSENSOR_TRANSFORMATION \
{ { -1, 0, 0}, \
{ 0, 1, 0}, \
{ 0 , 0 , -1} }
I897 marray #define YAMAHA_MSENSOR_TRANSFORMATION_KEPLER \
{ { 0, -1, 0}, \
{ -1, 0, 0}, \
{ 0 , 0 , -1} }
T959 marray #define YAMAHA_MSENSOR_TRANSFORMATION_KEPLER \
{ { -1, 0, 0}, \
{ 0, 1, 0}, \
{ 0 , 0 , -1} }
*No idea what these are for or why they need to be there*
I9000 marray_00 #define YAMAHA_MSENSOR_TRANSFORMATION_00 \
{ { 0, -1, 0}, \
{ -1, 0, 0}, \
{ 0 , 0 , -1} }
I897 marray_B5 #define YAMAHA_MSENSOR_TRANSFORMATION_KEPLER_B5 \
{ { 0, 1, 0}, \
{ -1, 0, 0}, \
{ 0 , 0 , 1} }
I9000 garray_emul #define YAMAHA_GSENSOR_TRANSFORMATION_EMUL \
{ { -1, 0, 0}, \
{ 0, 1, 0}, \
{ 0, 0, 1} }
I9000 marray_emul #define YAMAHA_MSENSOR_TRANSFORMATION_EMUL \
{ { 1, 0, 0}, \
{ 0, 1, 0}, \
{ 0 , 0 , 1} }
Click to expand...
Click to collapse
This stuff here is linear algebra stuff. Looks like matrix transformations having to do with x, y, z orientation. If these matrices are different between the I-9000 and Captivate kernels, they are most definitely responsible for the orientation problems.
Hello all, so I have finally made my own kernel from source
Haven't added anything to it, but I wanting to OC it a bit.
I have been looking through the .C files of the source but can not come up with anything, the closest I have gotten is this located in the freq_table.c
Code:
/*********************************************************************
* FREQUENCY TABLE HELPERS *
*********************************************************************/
int cpufreq_frequency_table_cpuinfo(struct cpufreq_policy *policy,
struct cpufreq_frequency_table *table)
{
unsigned int min_freq = ~0;
unsigned int max_freq = 0;
unsigned int i;
But not sure where exactly this leads to set the parameters, any help will be greatly appreciated, thanks!
Ask Damian
Sent from my GT-I9000 using XDA App
check attachements (patches and whole files) to my post with insane kernel. (sig) This might give you a hint where to look, you are welcome to implement it differently (maybe will be more stable). The files are so-called raspdeep's oc code.
Can somebody point me on kernel for GB 2.3.4 supporting Voodoo Color? Thanks.
Hello XDA-Developers... that's my first post by the way but I think that we can do something very important to the ATRIX users.
I'm from Cuba, I got an Atrix about 2 weeks (I have a Samsung Focus too) but I decided take that Android phone thinking in the Atrix-Lapdock combination. Here I have the really bad issue of wifi problems with my country code operator, I tryed every fix out there in the internet and here like
[Solved] Fix for nonworking WiFi on Simunlocked ATT Atrix 4G http://forum.xda-developers.com/showthread.php?t=1058260
[FIX] Final fix for nonworking WIFI on unlocked ATT Atrix in some countries http://forum.xda-developers.com/showthread.php?t=1077329
but I did'nt have a success
Reading about that issue I think that the problem is just inside the code of ATRIX wifi drivers and I made my own decision of search in the source looking for country codes segment, do a modification and re-compile the drivers.
After a week digging in every files of wifi drivers source code I think that I finally found a solution but I need help. I'm a noob in Adroid world but I think that mybe with help of some developers from XDA we can do that job that we will appreciated for many users around the world.
Well, now the important thing. I donwloaded a atrix-wifi-module from Atrix Dev Team on github, after a few days looking the source I found that in \open-src\src\dhd\sys\dhd_custom_gpio.c file...
"dhd_custom_gpio.c" that C file can be our solution
and here a piece of code, the function with the list of countries
Code:
/* Customized Locale table : OPTIONAL feature */
const struct cntry_locales_custom translate_custom_table[] = {
/* Table should be filled out based on custom platform regulatory requirement */
#ifdef COUNTRY_REGIONS
{"", "XY", 4}, /* universal if Counry code is unknow or empty */
{"US", "US", 69}, /* input ISO "US" to : US regrev 69 */
{"CA", "US", 69}, /* input ISO "CA" to : US regrev 69 */
{"EU", "EU", 5}, /* European union countries */
{"AT", "EU", 5},
{"BE", "EU", 5},
{"BG", "EU", 5},
{"CY", "EU", 5},
{"CZ", "EU", 5},
{"DK", "EU", 5},
{"EE", "EU", 5},
{"FI", "EU", 5},
{"FR", "EU", 5},
{"DE", "EU", 5},
{"GR", "EU", 5},
{"HU", "EU", 5},
{"IE", "EU", 5},
{"IT", "EU", 5},
{"LV", "EU", 5},
{"LI", "EU", 5},
{"LT", "EU", 5},
{"LU", "EU", 5},
{"MT", "EU", 5},
{"NL", "EU", 5},
{"PL", "EU", 5},
{"PT", "EU", 5},
{"RO", "EU", 5},
{"SK", "EU", 5},
{"SI", "EU", 5},
{"ES", "EU", 5},
{"SE", "EU", 5},
{"GB", "EU", 5}, /* input ISO "GB" to : EU regrev 05 */
{"KR", "XY", 3},
{"AU", "XY", 3},
{"CN", "XY", 3}, /* input ISO "CN" to : XY regrev 03 */
{"TW", "XY", 3},
{"AR", "XY", 3},
{"MX", "XY", 3}
#endif /* COUNTRY_REGIONS */
};
At the end of the file are another important function
Code:
/* Customized Locale convertor
* input : ISO 3166-1 country abbreviation
* output: customized cspec
*/
void get_customized_country_code(char *country_iso_code, wl_country_t *cspec)
{
int size, i;
size = ARRAYSIZE(translate_custom_table);
if (cspec == 0)
return;
if (size == 0)
return;
for (i = 0; i < size; i++) {
if (strcmp(country_iso_code, translate_custom_table[i].iso_abbrev) == 0) {
memcpy(cspec->ccode, translate_custom_table[i].custom_locale, WLC_CNTRY_BUF_SZ);
cspec->rev = translate_custom_table[i].custom_locale_rev;
return;
}
}
#ifdef COUNTRY_REGIONS
/* if no country code matched return first universal code from translate_custom_table */
memcpy(cspec->ccode, translate_custom_table[0].custom_locale, WLC_CNTRY_BUF_SZ);
cspec->rev = translate_custom_table[0].custom_locale_rev;
#endif /* COUNTRY_REGIONS */
return;
}
And finally we can see that comment
/* if no country code matched return first universal code from translate_custom_table */
I hope we can recompile it changing universalcode {"", "XY", 4} and put {"", "US", 69}
then if the country code of our provider is not in the list the wifi driver will act like we are in US and that can solve the wifi problems for every users...
I need help to do that work please... Developers from XDA help me!!!!
are you running 2.2 or 2.3 or even 4.0? The lg drivers only work with 2.2,, if you are on 2.2 i suggest you unlock your bootloader and flash cm7.2 and try if wifi works than. Another thing is the consideration that the wifi chip is broken (i doubt it though), can you do a dmesg and a logcat and post them here while you try connecting to wifi?
Have you tried the fix where you edit the database to US and make it read only than, if so did it work?
With editing and recompiling those drivers/modules i cant really help you since its not my domain of knowledge, but your idea seems reasonable and probably doable. I dont have access to my atrix now, so i cant test or anything for you, possibly i could compile/crosscompile that module for you, i would need the full sources for the driver/kernels with the makefile and the properly edited file as you suggested.
ScorpionXII said:
Hello XDA-Developers... that's my first post by the way but I think that we can do something very important to the ATRIX users.
I'm from Cuba, I got an Atrix about 2 weeks (I have a Samsung Focus too) but I decided take that Android phone thinking in the Atrix-Lapdock combination. Here I have the really bad issue of wifi problems with my country code operator, I tryed every fix out there in the internet and here like
[Solved] Fix for nonworking WiFi on Simunlocked ATT Atrix 4G http://forum.xda-developers.com/showthread.php?t=1058260
[FIX] Final fix for nonworking WIFI on unlocked ATT Atrix in some countries http://forum.xda-developers.com/showthread.php?t=1077329
but I did'nt have a success
Reading about that issue I think that the problem is just inside the code of ATRIX wifi drivers and I made my own decision of search in the source looking for country codes segment, do a modification and re-compile the drivers.
After a week digging in every files of wifi drivers source code I think that I finally found a solution but I need help. I'm a noob in Adroid world but I think that mybe with help of some developers from XDA we can do that job that we will appreciated for many users around the world.
Well, now the important thing. I donwloaded a atrix-wifi-module from Atrix Dev Team on github, after a few days looking the source I found that in \open-src\src\dhd\sys\dhd_custom_gpio.c file...
"dhd_custom_gpio.c" that C file can be our solution
and here a piece of code, the function with the list of countries
At the end of the file are another important function
And finally we can see that comment
/* if no country code matched return first universal code from translate_custom_table */
I hope we can recompile it changing universalcode {"", "XY", 4} and put {"", "US", 69}
then if the country code of our provider is not in the list the wifi driver will act like we are in US and that can solve the wifi problems for every users...
I need help to do that work please... Developers from XDA help me!!!!
Click to expand...
Click to collapse
@ripee, list of codes
CyanogenMod is a free, community built, aftermarket firmware distribution of Android 4.3 (Jelly Bean), which is designed to increase performance and reliability over stock Android for your device.
Code:
#include <std_disclaimer.h>
/*
* Your warranty is now void.
*
* I am not responsible for bricked devices, dead SD cards,
* thermonuclear war, or you getting fired because the alarm app failed. Please
* do some research if you have any concerns about features included in this ROM
* before flashing it! YOU are choosing to make these modifications, and if
* you point the finger at me for messing up your device, I will laugh at you. Hard. A lot.
*/
CyanogenMod is based on the Android Open Source Project with extra contributions from many people within the Android community. It can be used without any need to have any Google application installed. Linked below is a package that has come from another Android project that restore the Google parts. CyanogenMod does still include various hardware-specific code, which is also slowly being open-sourced anyway.
All the source code for CyanogenMod is available in the CyanogenMod Github repo. And if you would like to contribute to CyanogenMod, please visit out Gerrit Code Review. You can also view the Changelog for a full list of changes & features.
IMPORTANT INFORMATION
Click to expand...
Click to collapse
If this thread becomes a mess, I'll lock it immediately
We'll not support users and answer questions from users which:
- are running a custom kernel
- have flashed mods
- modified system files
- didn't follow our intructions word by word
- are unfriendly
Even if you tell us that your problem is not related to your custom kernel /mod / magic => WE DON'T CARE!
Your mod => your problem!
WIKI
Click to expand...
Click to collapse
Official CyanogenMod Wiki: http://wiki.cyanogenmod.org/w/N7000_Info
HOW TO INSTALL
Click to expand...
Click to collapse
First time installing CyanogenMod 10.2 to your Galaxy Note, or coming from another ROM:
- Read known issues ans FAQs
- Make sure you're running a proper working ClockworkMod-Recovery
- Copy GApps and CM10.2 ZIPs to your SDCard
- Boot into Recovery
- Flash CM10.2 zip from SDCard
- Flash GApps zip from SDCard
- DO A DATA WIPE / FACTORY RESET (otherwise your device will be stuck at boot)
- Reboot
- Don't restore system data using Titanium Backup!
- Restoring Apps + Data might cause problems and is not recommended, avoid it if possible!
Upgrading from earlier version of CyanogenMod 10.2:
- Copy CM10.2 ZIP to your SDCard
- Boot into Recovery
- Flash CM10.2 zip from SDCard
- Reboot
Upgrading from CyanogenMod 10/10.1:
- Read known issues ans FAQs
- Copy GApps and CM10.2 ZIPs to your SDCard
- Boot into Recovery
- Flash CM10.2 zip from SDCard
- Flash GApps zip from SDCard
- Wipe data, we're not sure if everything will work yet.
- Reboot
SCREENSHOTS
Click to expand...
Click to collapse
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
KNOWN ISSUES
Click to expand...
Click to collapse
Known bugs:
- some minor graphics glitches
- no camera fix has been merged
Remember we just merged yesterday (26th july), expect bugs in apps
DOWNLOADS
Click to expand...
Click to collapse
http://forum.xda-developers.com/showthread.php?t=2381499
CHANGELOG
Click to expand...
Click to collapse
http://changelog.bbqdroid.org/#/n7000/next
SUPPORT
Click to expand...
Click to collapse
Chat:
http://webchat.freenode.net/?channels=teamhacksung-support
irc.freenode.net #teamhacksung-support
Don't expect any support if you:
- are not running stock cm-kernel
- have installed any mods
- have modified system files
DEAR KANGERS
Click to expand...
Click to collapse
If you're going to reuse our work, which we're doing for free, be fair and give proper credits.
This is the only payment we're really demanding and we deserve it to be mentioned because of the countless hours we've put into this project.
Open-Source doesn't mean Out-of-Respect!!!
SUPPORT US
Click to expand...
Click to collapse
If you want to donate a few bucks for the work we're doing in our freetime:
I will get new battery today and test this 4.3
Wysłane z mojego GT-N7000 za pomocą Tapatalk 4
So dev thread got locked within just three days.
I see, Note N7000 people are the craziest and notorious trolls here in XDA. :laugh:
Thanks for your support @XpLoDWilD.
Thanks and sorry at the same time...
M3TALLICA said:
So dev thread got locked within just three days.
I see, Note N7000 people are the craziest and notorious trolls here in XDA. :laugh:
Thanks for your support @XpLoDWilD.
Click to expand...
Click to collapse
Cause Note1 got the worst official JB update from all devices out there...therefore everybody is praying for a hero that makes a rom that works as it should....
As long as I keep the N1, cm ROMs are far better then any other tw ROM out there...
Running latest build cm10.2...some small glitches already mentioned but looks promising so far...
Thanks a lot for devs that try to bring this to work...
Sent from my GT-N7000 using xda app-developers app
Building your own CM10.2
After building and using a CM10.1 for a few days now I started to try to build the experimental CM10.2 myself.
I used the How To in the CyanogenMod wiki and changed some settings related to 10.2 (or maybe better: I tried to change).
The history:
In the morning my first build succeeded and could be installed on the device. Testing it without factory reset and only cleaning the caches. It booted in the usual way. Then I got the offer to choose between Nova and Trebuchet launcher (I'm using Nova in 10.1). The selection of the launcher resulted in a restart. So not really successful.
After a factory reset I get the restart when choosing the language in the setup. My build configuration is definitely incomplete/wrong.
Now I switched to the provided built and with this one I could boot into CM10.2 without making a factory reset. Using it right now to connect with a ssh client to my build computer at home and trying a new build.
The idea:
Maybe there are other members here who want to make their own CM10.2 builds as early as possible. Or maybe some did already.
So I want to write down the necessary changes to the normal CM10.1 procedure and I hope that this might help to solve some problems.
With the "migration" of this thread from Original Development to General I assume it is ok to do so (if not I will delete it asap).
Changes in the original Tutorial:
How To Build CyanogenMod Android for Samsung Galaxy Note (International) ("n7000")
I followed the tutorial which is a really good one. I will list the changes I made for CM 10.2.
2.3 Initialize the CyanogenMod source repository
Switched from cm-10.1 to cm-10.2
Code:
$ cd ~/android/system/
$ repo init -u git://github.com/CyanogenMod/android.git -b cm-10.2
2.8 Prepare the device-specific code
removed this. So if my new build will work this thread might get useless because there isn't much to change in the 10.1 tutorial
With these changes I can start to build CM 10.2 and get an installable zip.
Right now I'm building it again and will try later if "my" restart will disappear or some things more need to be changed.
And of course I hope some members might help me or add necessary parts.
Thanks in advance.
Thanks for the thread and sorry for the OT xplodwild
Sent from my GT-N7000 using xda premium
Yeah thank you XpLoDWilD and i am sorry too
it's a semi off-topic: i almost have one rule when i change/buy my phone: to be CM supported.
By far the best rom out there, not all the fancy things producer does (although NEMESIS project, it is one as the name states)(although i would like a center clock and a ram bar at the recent apps ), CM doesn't care about the age of the phone, or if one company must force end updating a rom, for the people to buy the next model (S thingy).
Thank you CM (i refer the the people, project and everything related to)!
Long Live CM
now you are spamming this thread too
In my oppinion, asking if the possibility of bricking one's device due to a EMMC bug and the fact that the new Android version uses a command that can trigger said bug IS development related. But heck, whatever. Then let me ask the same question here: could this ROM possibly brick my phone or was that taken care of during developement?
DeKubus said:
In my oppinion, asking if the possibility of bricking one's device due to a EMMC bug and the fact that the new Android version uses a command that can trigger said bug IS development related. But heck, whatever. Then let me ask the same question here: could this ROM possibly brick my phone or was that taken care of during developement?
Click to expand...
Click to collapse
i want to think that the developers already thought about that, and if there was any danger it was posted already...
why do we have the experimental builds, just to screw up our phones, see how many ppl flash 10.2 and screw their phones, statistical emmc chips fried?
I asked this question in the closed thread and because of that I took the liberty to repeat it in a PM.
fstrim (it is a part of the CM10.2 sources) command works on eMMC and the developers will check if this is a real threat or only a wrong assumption.
So we'll have to wait a little - the answers will come.
So this is the fun thread?
Sent from my GT-N7000 using Tapatalk 4 Beta
I hope it won't be (ab)used as such. As xplodwild made it clear by closing the other one without further warning.
I love CYANOGEN MODE.
I can't live with cyanamode...
That's why I bought gt n7000
Thank tlyou cm
Long live our CM
Sent from my GT-N7000 using xda premium
To answer the FSTRIM / EMMC brick bug: You should be safe. Right now CM kernels block all erase, and all Sammy kernels translate secure into nonsecure.
I found an info in other kernel development thread - and of course I do not know it this statement is true, but it seems logical:
fstrim won't work with MMC_CAP_ERASE disabled.
So with the safe CM kernels (here it is disabled) we still can't experience the brickbug.
But we won't get the trim benefits with 4.3 (see the XDA Portal today with a report about it).
The source of fstrim.c (in system/vold) is quite simple to read.
The magic is done in this line (part of an if-statement):
Code:
ioctl(fd, FITRIM, &range)
If the fstrim <-> MMC_CAP_ERASE relation is true, we are safe but will still suffer from the increasing lags.
Let's wait for a more reliable analysis by Entropy512 or other eMMC-Gurus.
The complete function here:
Code:
static void *do_fstrim_filesystems(void *ignored)
{
int i;
int fd;
int ret = 0;
struct fstrim_range range = { 0 };
struct stat sb;
extern struct fstab *fstab;
SLOGI("Starting fstrim work...\n");
/* Log the start time in the event log */
LOG_EVENT_LONG(LOG_FSTRIM_START, get_boot_time_ms());
for (i = 0; i < fstab->num_entries; i++) {
/* Skip raw partitions */
if (!strcmp(fstab->recs[i].fs_type, "emmc") ||
!strcmp(fstab->recs[i].fs_type, "mtd")) {
continue;
}
/* Skip read-only filesystems */
if (fstab->recs[i].flags & MS_RDONLY) {
continue;
}
if (fs_mgr_is_voldmanaged(&fstab->recs[i])) {
continue; /* Should we trim fat32 filesystems? */
}
if (stat(fstab->recs[i].mount_point, &sb) == -1) {
SLOGE("Cannot stat mount point %s\n", fstab->recs[i].mount_point);
ret = -1;
continue;
}
if (!S_ISDIR(sb.st_mode)) {
SLOGE("%s is not a directory\n", fstab->recs[i].mount_point);
ret = -1;
continue;
}
fd = open(fstab->recs[i].mount_point, O_RDONLY);
if (fd < 0) {
SLOGE("Cannot open %s for FITRIM\n", fstab->recs[i].mount_point);
ret = -1;
continue;
}
memset(&range, 0, sizeof(range));
range.len = ULLONG_MAX;
SLOGI("Invoking FITRIM ioctl on %s", fstab->recs[i].mount_point);
if (ioctl(fd, FITRIM, &range)) {
SLOGE("FITRIM ioctl failed on %s", fstab->recs[i].mount_point);
ret = -1;
}
SLOGI("Trimmed %llu bytes on %s\n", range.len, fstab->recs[i].mount_point);
close(fd);
}
/* Log the finish time in the event log */
LOG_EVENT_LONG(LOG_FSTRIM_FINISH, get_boot_time_ms());
SLOGI("Finished fstrim work.\n");
/* Release the wakelock that let us work */
release_wake_lock(FSTRIM_WAKELOCK);
return (void *)ret;
}
---------- Post added at 05:47 PM ---------- Previous post was at 05:44 PM ----------
XpLoDWilD said:
To answer the FSTRIM / EMMC brick bug: You should be safe. Right now CM kernels block all erase, and all Sammy kernels translate secure into nonsecure.
Click to expand...
Click to collapse
I just made my post and then saw this statement.
So no brick bug danger because of the blocked commands but we won't get the trim benefits in 4.3 for the N7000
On this ROM you cannot record in full HD but when you install lg camera it is working ok
Sent from my GT-N7000 using XDA Premium HD app
re TRIM, see also @Entropy512 comment here
all is nice with this rom.. cept for some graphical stuff in Feedly and during screen rotation..
it will run hot initially after that this rom runs as per normal
To get root working properly, need to disable root in developer settings, reboot the phone. After the phone reboots, enable root again. Root should work properly. Well, at least thats how i get my phone;s root working!
thanks!
A question, how do i see the list of notifications?
As a game addict i was bored to find out only 2 touch max is supported on my j5.anyone kindly tell me if its possible to increase this as some games are tough with just 2 touch.
Khanmodx said:
As a game addict i was bored to find out only 2 touch max is supported on my j5.anyone kindly tell me if its possible to increase this as some games are tough with just 2 touch.
Click to expand...
Click to collapse
I have just asked @Nick Verse yesterday about this.
It is possible to increase it, but we don't know yet what modification is needed.
If you or any other people wanna help, then go here: https://github.com/Galaxy-J5/kernel_samsung_j5lte/tree/cm-12.1/drivers/input/touchscreen .
We found some files like:
synaptics_i2c_rmi4.h
synaptics_dsx/synaptics_dsx_core.h
imagis_30xxc/ist30xxc.h
Also, in synaptics_i2c_rmi4.c there is following code:
PHP:
#define F12_FINGERS_TO_SUPPORT 10
.......
struct synaptics_rmi4_f12_extra_data {
unsigned char data1_offset;
unsigned char data15_offset;
unsigned char data15_size;
unsigned char data15_data[(F12_FINGERS_TO_SUPPORT + 7) / 8];
};
.......
/* Maximum number of fingers supported */
if ((query[1] & MASK_3BIT) <= 4)
fhandler->num_of_data_points = (query[1] & MASK_3BIT) + 1;
else if ((query[1] & MASK_3BIT) == 5)
fhandler->num_of_data_points = 10;
.......
/* Maximum number of fingers supported */
fhandler->num_of_data_points = min(ctrl_23.max_reported_objects,
(unsigned char)F12_FINGERS_TO_SUPPORT);
unsigned char data15_data[(F12_FINGERS_TO_SUPPORT + 7) / 8]; ---> [(10+7) / 8 ] = 2.
Saw a thread in xda where 10 multitouch points are possible in cm firmware onlyhttp://forum.xda-developers.com/showthread.php?t=1384927)link. But was not able to implement it coz of limited info about roms. Then a pal told me it hardware that i hve to use a hardware simulation but seems to tiring to do all that just to boost multitouch.