Jump to content


  • Content Count

  • Joined

  • Last visited

Community Reputation

26 Excellent

About GuyOverThere

  • Rank

Profile Information

  • Location
    Caracas, Venezuela
  • Interests
    Gaming, Linux, Tech and other stuff
  • Your Current Device(s)
    ZTE Racer x850
  1. Hi, my input here from test I was doing before my phone got stolen. Regarding CM9 and beyond and following development of 2.6.35 ICS kernel branch, that kernel uses different implementation of the framebuffer than 2.6.35 GB branch. In the ICS branch several yamato kgsl files (from zte if I'm not mistaken) were dropped in favor of adreno kgsl (opensource implementation) which means that using ICS kernels from blade will require gralloc opensource no matter what. Also, ICS and beyond requires the use of zte's adreno libs ICS+ compatible which also means that to get any kind of support from those libs 2.1-2.2 prebuilt gralloc can't be used. During my tests, I was able to get a 4.0.4 rom that boot but with the screen distorted, I wasn't able to find the origin of the issue, this is a log from that rom (relevant part): I/sysproc ( 184): Entered system_init() I/sysproc ( 184): ServiceManager: 0x15c238 I/SurfaceFlinger( 184): SurfaceFlinger is starting I/SurfaceFlinger( 184): use dithering I/SurfaceFlinger( 184): SurfaceFlinger's main thread ready to run. Initializing graphics H/W... E/msm7x27.gralloc( 184): We support 2 buffers I/msm7x27.gralloc( 184): using (fd=35) I/msm7x27.gralloc( 184): id = msmfb30_30001 I/msm7x27.gralloc( 184): xres = 240 px I/msm7x27.gralloc( 184): yres = 320 px I/msm7x27.gralloc( 184): xres_virtual = 240 px I/msm7x27.gralloc( 184): yres_virtual = 640 px I/msm7x27.gralloc( 184): bpp = 16 I/msm7x27.gralloc( 184): r = 11:5 I/msm7x27.gralloc( 184): g = 5:6 I/msm7x27.gralloc( 184): b = 0:5 I/msm7x27.gralloc( 184): width = 38 mm (160.421051 dpi) I/msm7x27.gralloc( 184): height = 51 mm (159.372543 dpi) I/msm7x27.gralloc( 184): refresh rate = 74.00 Hz D/CALCFPS ( 184): DEBUG_CALC_FPS: 0 D/CALCFPS ( 184): period: 10 D/CALCFPS ( 184): ignorethresh_us: 500000 D/CALCFPS ( 184): DEBUG_CALC_FPS: 0 D/CALCFPS ( 184): period: 10 D/CALCFPS ( 184): ignorethresh_us: 500000 D/FramebufferNativeWindow( 184): mNumBuffers = 2 D/libEGL ( 184): loaded /system/lib/egl/libGLES_android.so D/libEGL ( 184): loaded /system/lib/egl/libEGL_adreno200.so D/libEGL ( 184): loaded /system/lib/egl/libGLESv1_CM_adreno200.so D/libEGL ( 184): loaded /system/lib/egl/libGLESv2_adreno200.so I/SurfaceFlinger( 184): EGL informations: I/SurfaceFlinger( 184): # of configs : 44 I/SurfaceFlinger( 184): vendor : Android I/SurfaceFlinger( 184): version : 1.4 Android META-EGL I/SurfaceFlinger( 184): extensions: EGL_KHR_image EGL_KHR_image_base EGL_KHR_image_base EGL_KHR_fence_sync EGL_ANDROID_image_native_buffer EGL_ANDROID_image_native_buffer EGL_ANDROID_get_render_buffer I/SurfaceFlinger( 184): Client API: OpenGL ES I/SurfaceFlinger( 184): EGLSurface: 5-6-5-0, config=0x0 I/SurfaceFlinger( 184): OpenGL informations: I/SurfaceFlinger( 184): vendor : Qualcomm I/SurfaceFlinger( 184): renderer : Adreno (TM) 200 I/SurfaceFlinger( 184): version : OpenGL ES-CM 1.1 I/SurfaceFlinger( 184): extensions: GL_AMD_compressed_ATC_texture GL_AMD_performance_monitor GL_APPLE_texture_2D_limited_npot GL_ARB_vertex_buffer_object GL_EXT_texture_filter_anisotropic GL_EXT_texture_format_BGRA8888 GL_EXT_texture_type_2_10_10_10_REV GL_OES_blend_equation_separate GL_OES_blend_func_separate GL_OES_blend_subtract GL_OES_compressed_ETC1_RGB8_texture GL_OES_compressed_paletted_texture GL_OES_depth_texture GL_OES_depth24 GL_OES_draw_texture GL_OES_EGL_image GL_OES_EGL_image_external GL_OES_framebuffer_object GL_OES_matrix_palette GL_OES_packed_depth_stencil GL_OES_point_size_array GL_OES_point_sprite GL_OES_read_format GL_OES_rgb8_rgba8 GL_OES_stencil_wrap GL_OES_texture_cube_map GL_OES_texture_env_crossbar GL_OES_texture_float GL_OES_texture_half_float GL_OES_texture_half_float_linear GL_OES_texture_npot GL_OES_texture_mirrored_repeat GL_QCOM_binning_control GL_QCOM_extended_get GL_QCOM_tiled_rendering GL_AMD_compressed_3DC_texture I/SurfaceFlinger( 184): GL_MAX_TEXTURE_SIZE = 4096 I/SurfaceFlinger( 184): GL_MAX_VIEWPORT_DIMS = 4096 x 4096 I/SurfaceFlinger( 184): flags = 00000000 D/CALCFPS ( 184): DEBUG_CALC_FPS: 0 D/CALCFPS ( 184): period: 10 D/CALCFPS ( 184): ignorethresh_us: 500000 After that is mostly regular rom-loading info, nothing apparent of why the screen isn't working. This test was done using adreno libs ICS compatible and gralloc from source (blade ICS branch), and this is what happened at kernel level: 2>[11-28 00:41:12.120000] kgsl kgsl-3d0: |kgsl_gpummu_pagefault| mmu page fault: page=0x662a0000 pt=0 op=WRITE axi=4 <2>[11-28 00:41:12.290000] kgsl kgsl-3d0: |kgsl_gpummu_pagefault| mmu page fault: page=0x662a0000 pt=0 op=WRITE axi=4 <2>[11-28 00:41:12.460000] kgsl kgsl-3d0: |kgsl_gpummu_pagefault| mmu page fault: page=0x662a0000 pt=0 op=WRITE axi=4 <3>[11-28 00:41:12.490000] [62: synaptics_swq]msm_i2c msm_i2c.0: (0044) Error during data xfer (-5) <3>[11-28 00:41:12.490000] [62: synaptics_swq]synaptics-rmi-ts 0-0022: synaptics_i2c_read FAILED: read of register 20 <3>[11-28 00:41:12.490000] [62: synaptics_swq]synaptics_ts_work_func: synaptics_i2c_write failed, go to poweroff. <2>[11-28 00:41:12.630000] kgsl kgsl-3d0: |kgsl_gpummu_pagefault| mmu page fault: page=0x662a0000 pt=0 op=WRITE axi=4 <2>[11-28 00:41:12.800000] kgsl kgsl-3d0: |kgsl_gpummu_pagefault| mmu page fault: page=0x662a0000 pt=0 op=WRITE axi=4 <3>[11-28 00:41:12.910000] [62: synaptics_swq]msm_i2c msm_i2c.0: (0044) Error during data xfer (-5) ...this lines keep repeating over and over.... This is from a log using software rendering only (only libGLES_android.so): I/sysproc ( 174): Entered system_init() I/sysproc ( 174): ServiceManager: 0x1314e0 I/SurfaceFlinger( 174): SurfaceFlinger is starting I/SurfaceFlinger( 174): use dithering I/SurfaceFlinger( 174): SurfaceFlinger's main thread ready to run. Initializing graphics H/W... E/msm7x27.gralloc( 174): We support 2 buffers I/msm7x27.gralloc( 174): using (fd=35) I/msm7x27.gralloc( 174): id = msmfb30_30001 I/msm7x27.gralloc( 174): xres = 240 px I/msm7x27.gralloc( 174): yres = 320 px I/msm7x27.gralloc( 174): xres_virtual = 240 px I/msm7x27.gralloc( 174): yres_virtual = 640 px I/msm7x27.gralloc( 174): bpp = 16 I/msm7x27.gralloc( 174): r = 11:5 I/msm7x27.gralloc( 174): g = 5:6 I/msm7x27.gralloc( 174): b = 0:5 I/msm7x27.gralloc( 174): width = 38 mm (160.421051 dpi) I/msm7x27.gralloc( 174): height = 51 mm (159.372543 dpi) I/msm7x27.gralloc( 174): refresh rate = 74.00 Hz D/CALCFPS ( 174): DEBUG_CALC_FPS: 0 D/CALCFPS ( 174): period: 10 D/CALCFPS ( 174): ignorethresh_us: 500000 D/CALCFPS ( 174): DEBUG_CALC_FPS: 0 D/CALCFPS ( 174): period: 10 D/CALCFPS ( 174): ignorethresh_us: 500000 D/FramebufferNativeWindow( 174): mNumBuffers = 2 D/libEGL ( 174): loaded /system/lib/egl/libGLES_android.so I/SurfaceFlinger( 174): EGL informations: I/SurfaceFlinger( 174): # of configs : 8 I/SurfaceFlinger( 174): vendor : Android I/SurfaceFlinger( 174): version : 1.4 Android META-EGL I/SurfaceFlinger( 174): extensions: EGL_KHR_image_base EGL_ANDROID_image_native_buffer EGL_ANDROID_get_render_buffer I/SurfaceFlinger( 174): Client API: OpenGL ES I/SurfaceFlinger( 174): EGLSurface: 5-6-5-0, config=0x0 I/SurfaceFlinger( 174): OpenGL informations: I/SurfaceFlinger( 174): vendor : Android I/SurfaceFlinger( 174): renderer : Android PixelFlinger 1.4 I/SurfaceFlinger( 174): version : OpenGL ES-CM 1.0 I/SurfaceFlinger( 174): extensions: GL_OES_byte_coordinates GL_OES_fixed_point GL_OES_single_precision GL_OES_read_format GL_OES_compressed_paletted_texture GL_OES_draw_texture GL_OES_matrix_get GL_OES_query_matrix GL_OES_EGL_image GL_OES_compressed_ETC1_RGB8_texture GL_ARB_texture_compression GL_ARB_texture_non_power_of_two GL_ANDROID_user_clip_plane GL_ANDROID_vertex_buffer_object GL_ANDROID_generate_mipmap I/SurfaceFlinger( 174): GL_MAX_TEXTURE_SIZE = 4096 I/SurfaceFlinger( 174): GL_MAX_VIEWPORT_DIMS = 4096 x 4096 I/SurfaceFlinger( 174): flags = 00040000 D/CALCFPS ( 174): DEBUG_CALC_FPS: 0 D/CALCFPS ( 174): period: 10 D/CALCFPS ( 174): ignorethresh_us: 500000 ..... F/libc ( 174): Fatal signal 11 (SIGSEGV) at 0x4838f000 (code=2) D/libEGL ( 190): loaded /system/lib/egl/libGLES_android.so W/UsageStats( 174): Usage stats version changed; dropping I/ARMAssembler( 190): generated scanline__00000077:03010102_00000A01_00000000 [ 9 ipp] (30 ins) at [0x4082c1e8:0x4082c260] in 675002 ns I/ARMAssembler( 190): generated scanline__00000077:03545402_00000A01_00000000 [ 31 ipp] (52 ins) at [0x4082c268:0x4082c338] in 140000 ns I/ActivityManager( 174): Enabled StrictMode logging for AThread's Looper I/SystemServer( 174): Telephony Registry I/SystemServer( 174): Package Manager I/DEBUG ( 100): *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** I/DEBUG ( 100): Build fingerprint: 'ZTE/P772A10D/atlas40:4.0.4/IMM76D/20120608.113547:user/release-keys' I/DEBUG ( 100): pid: 174, tid: 187 >>> system_server <<< I/DEBUG ( 100): signal 11 (SIGSEGV), code 2 (SEGV_ACCERR), fault addr 4838f000 I/DEBUG ( 100): r0 4838f000 r1 00000000 r2 000001c0 r3 00000000 I/DEBUG ( 100): r4 00000013 r5 4838f000 r6 000001e0 r7 00000200 I/DEBUG ( 100): r8 4803c000 r9 400b6f3d 10 00100000 fp 001e0248 I/DEBUG ( 100): ip 00000000 sp 4813bd44 lr 00000000 pc 40078c28 cpsr 20000010 I/DEBUG ( 100): d0 7149f2ca7149f2ca d1 43a040007149f2ca I/DEBUG ( 100): d2 3f8000003f800000 d3 3f8000003f800000 I/DEBUG ( 100): d4 0000000000000000 d5 bf80000000000000 I/DEBUG ( 100): d6 bf800000bf800000 d7 7149f2ca3f000000 I/DEBUG ( 100): d8 0000000000000000 d9 0000000000000000 I/DEBUG ( 100): d10 0000000000000000 d11 0000000000000000 I/DEBUG ( 100): d12 0000000000000000 d13 0000000000000000 I/DEBUG ( 100): d14 0000000000000000 d15 0000000000000000 I/DEBUG ( 100): scr 20000010 I/DEBUG ( 100): I/DEBUG ( 100): #00 pc 00003c28 /system/lib/libcutils.so I/DEBUG ( 100): #01 pc 00016d4a /system/lib/libpixelflinger.so I/DEBUG ( 100): #02 pc 00016e3c /system/lib/libpixelflinger.so I/DEBUG ( 100): #03 pc 0000eea8 /system/lib/egl/libGLES_android.so (glClear) I/DEBUG ( 100): #04 pc 0001e90a /system/lib/libsurfaceflinger.so (_ZN7android14SurfaceFlinger21setupHardwareComposerERNS_6RegionE) I/DEBUG ( 100): #05 pc 0001e9e6 /system/lib/libsurfaceflinger.so (_ZN7android14SurfaceFlinger13handleRepaintEv) I/DEBUG ( 100): #06 pc 0001fd7e /system/lib/libsurfaceflinger.so (_ZN7android14SurfaceFlinger10threadLoopEv) I/DEBUG ( 100): #07 pc 00022994 /system/lib/libutils.so (_ZN7android6Thread11_threadLoopEPv) I/DEBUG ( 100): #08 pc 00022fce /system/lib/libutils.so I/DEBUG ( 100): #09 pc 00012fa8 /system/lib/libc.so (__thread_entry) I/DEBUG ( 100): #10 pc 00012acc /system/lib/libc.so (pthread_create) I/DEBUG ( 100): I/DEBUG ( 100): code around pc: I/DEBUG ( 100): 40078c08 e1b03e03 28a04002 28a04002 48a04002 .>[email protected]([email protected]([email protected] I/DEBUG ( 100): 40078c18 e1b03103 24801004 e1a03001 e2522020 .1.....$.0.. R. I/DEBUG ( 100): 40078c28 28a0500a 28a0500a 2afffffb e2822020 .P.(.P.(...* .. I/DEBUG ( 100): 40078c38 e1b02e02 28a0500a 48a04002 e1b02102 .....P.([email protected]!.. I/DEBUG ( 100): 40078c48 24801004 40c0e0b2 e49de004 e12fff1e [email protected]/. I/DEBUG ( 100): I/DEBUG ( 100): code around lr: I/DEBUG ( 100): 00000000 ffffffff ffffffff ffffffff ffffffff ................ I/DEBUG ( 100): 00000010 ffffffff ffffffff ffffffff ffffffff ................ I/DEBUG ( 100): 00000020 ffffffff ffffffff ffffffff ffffffff ................ I/DEBUG ( 100): 00000030 ffffffff ffffffff ffffffff ffffffff ................ I/DEBUG ( 100): 00000040 ffffffff ffffffff ffffffff ffffffff ................ I/DEBUG ( 100): I/DEBUG ( 100): memory map around addr 4838f000: I/DEBUG ( 100): 48344000-4838f000 /dev/graphics/fb0 I/DEBUG ( 100): 4838f000-483a3000 /system/lib/egl/libGLES_android.so I/DEBUG ( 100): 483a3000-483a4000 /system/lib/egl/libGLES_android.so I/DEBUG ( 100): I/DEBUG ( 100): stack: I/DEBUG ( 100): 4813bd04 40019ae7 /system/lib/libc.so I/DEBUG ( 100): 4813bd08 4004c510 I/DEBUG ( 100): 4813bd0c 00000003 I/DEBUG ( 100): 4813bd10 ffffffff I/DEBUG ( 100): 4813bd14 4813bdb0 I/DEBUG ( 100): 4813bd18 00133850 [heap] I/DEBUG ( 100): 4813bd1c 00000140 I/DEBUG ( 100): 4813bd20 00000000 I/DEBUG ( 100): 4813bd24 4001a037 /system/lib/libc.so I/DEBUG ( 100): 4813bd28 4813bdb0 I/DEBUG ( 100): 4813bd2c 405d20a4 /system/lib/libpixelflinger.so I/DEBUG ( 100): 4813bd30 ffffffff I/DEBUG ( 100): 4813bd34 00000001 I/DEBUG ( 100): 4813bd38 df0027ad I/DEBUG ( 100): 4813bd3c 00000000 I/DEBUG ( 100): 4813bd40 00000000 I/DEBUG ( 100): #00 4813bd44 405cfd4d /system/lib/libpixelflinger.so I/DEBUG ( 100): #01 4813bd48 00000001 I/DEBUG ( 100): 4813bd4c 00000000 I/DEBUG ( 100): 4813bd50 4813bd9c I/DEBUG ( 100): 4813bd54 0017ebe0 [heap] I/DEBUG ( 100): 4813bd58 00004000 I/DEBUG ( 100): 4813bd5c 00000000 I/DEBUG ( 100): 4813bd60 000000f0 I/DEBUG ( 100): 4813bd64 405cfe41 /system/lib/libpixelflinger.so Here the error is more apparent (somewhat), there's a page fault somewhere, a chunk of memory is trying to be read but that chunk is unavailable, sadly I wasn't able to trace it (I was in that process but the phone was stolen so...). Using software rendering only there was nothing but a black screen and interestingly nothing obviously wrong at kernel level, a puzzle =/ From input I got from blade dev guys, the software rendering failing was unheard of apparently, all that came to mind is that the problem should be in the kernel. Given the similarities of the boards, I concour with mike that the LCD panels are one possible cause, the other would be the framebuffer code itself, but I couldn't figure it out in time :c On gingerbread using opensource gralloc have a similar problem, disorted screen, errors with several pixelfinger related libraries and such. 4.0.4 logs attached at the end. Regards cm9-himax-test1.txt cm9-himax-dmesg1.txt cm9-soft-test3.txt cm9-soft-dmesg3.txt
  2. kudos mike! :D On my racer at least the wifi does work, from what I've seen from the other guys logcats, the wifi issue is the wpa_supplicant daemon not running
  3. If it's the same before and after calibration is possible that there's a second pointercal file in /data/local/pointercal but tslib in eclair roms by default uses /etc/pointercal, check anyway if you have that second copy (or somewhere else) in the rom. I can't use the values of the pointercal you posted since those values are the same used on mike's racermod kernel and deadlink's kernel which for whatever odd reason mess your calibration. You could take a logcat while doing the calibration, maybe that would give some other hints.
  4. Not alone :) So I don't know if I follow you correctly, you installed racermod, (let's keep cwm out for the moment), and then you installed your 2.1 rom over racermod without data wipe and your calibration works, is that correct? Also, what racer is this again? I'm guessing gen1 since it uses eclair, but where it was sold/bought? Again, a thing about eclair roms, the calibration on those roms is done by an app (TSCalibration.apk iirc) which means, tslib is integrated into InputDevice.java which also means it modifies /etc/pointercal, this hasen't being done by any 2.3 rom beyond rexchun BO2 and no one knows the modifications they did, so my best offer if you want a 2.3 rom would be this: use your 2.1 rom, do the calibration, take a copy of the /etc/pointercal file and post it here, that way I might be able to give you a kernel with those calibration values hardcoded in the kernel for you to test.
  5. Given the experimental nature of this rom, I think it's more helpful if it's tested in a clean state, by this I mean full-wipe and install, also, nothing extra installed beyond what's already in the rom except perhaps gapps. If there are leftover files (specially apk files), settings, blobs, whatever on the system before installing the rom it's quite possible to have bugs that are not really part of the original rom/kernel. The same applies for changing settings like CPU clock, compcache, governor, etc... make tests on the rom as it comes by default, if everything is good then setting A and B can be changed and if a problem arose from such change then is easier to backtrace the bug since only those A and B settings were applied before a particular error. In my case this is what I do for every test I try: Reboot into recovery. Backup my current working rom (nandroid). Factory reset/Data wipe. Mounts and storage -> format /boot and format /system Advanced -> wipe dalvik Install from zip -> chose zip from sdcard -> install rom. From here on is all tests... first touchscreen (pretty obvious xD), verify scrolling functionallity going from desk 3 to desk 1 to desk 5, then activate wifi and connect to my router, navigate web pages and close, then activate bluetooth and search for devices I know it should report, deactivate gps, reboot, install gapps, install google maps, activate gps, test apps like location, maps, etc... to verify gps status, open up dial, type numbers from 1 to 0, * and # type in and call to the number of other phone I have, check speaker sound, deactivate wifi and verify ril operation by connecting to google maps and web pages with edge and 3g, open up camera, take foto, return to camera and change to video recording mode, record video and close, open up android terminal and type every key, number and symbol, first in portrait mode and then in landscape, change screen brightness, verify accelerometer sensor and rotation animations, change wallpapers from cyanogenmod to live wallpapers, open up music app and play a song, verify sound up and down buttons, close long press power and reboot recovery. full wipe again, and restore my previous rom. done. From latest mike's rom the only error I had was magicsmoke which fc everytime. Also, in case of hardware errors (wifi can't be enable, which is different to wifi not connecting, accelerometer not working, speaker issues, camera, etc...) it'll be useful if dmesg is provided, similar to what mike said about logcat, from terminal emulator just run: dmesg > /sdcard/dmesg.txt and send that when you're able.
  6. I don't know where other racers are getting the wifi errors but in my phone wifi is working fine with the latest build, no fc and connects without issues, the only thing related to wifi I didn't try was tethering since I don't have what to test it with. Can't comment on the keypad/keyboard bug thing (if it's calibration for instance, since I had to use my kernel when testing it and keypad was working fine, I pressed secuentially 1, 2, 3,... till 0 and all was good) It's better for those who have issues with wifi or something else to provide at least logcat files, it will (or should) provide more info that just this isn't working or this fc.
  7. If it helps with anything, I use icecream sandwich keyboard from google play, more accurate than gingerbread's default. I've also though about dual-touch being at least part responsible of ghosts clicks however, like mike I haven't been able to remove it completely, while I've been able to change MT_* reporting to simple ABS_X/Y things while still keeping the little calibration info we have every time it ends the same way: points stop being reported only when a second point is marked, meaning, say you marked point x:129,y:130, and then you marked x:200,y:190, well this will be translated as a line drawn between the two points, like you never upped the finger/tool form the screen after the first point you marked. Very annoying, it has to be related to IRQ handling but I don't know how to change that :-\ or perhaps a check is needed like if (distance between first point and second > some number) { report new touch event; return IRQ_HANDLED; } I've a sightly different msm_ts.c than mike in my github sources but still have ghost clicks and not a thing I can do about it :( so yeah, I concur with mike that for the momment there's nothing much to do than to deal with the msm driver we have =/
  8. What phone do you have? If possible, can you specify the model and signature as reported by the stock rom? (if not just say "this is a zte carl gen2 phone bought in thailand" or something like that) I have no problems with those keys, works just fine in my phone (zte racer gen1, just for reference)
  9. The rom doesn't care about the gen, during cm building you can had either gen1 or gen2 address in boardconfig.mk (or whatever that's put) but that only affects kernel installation, thus, if you built with gen2 addresses, gen1 people will only need a gen1 kernel and done, the rom will still work as intended Edit:// Crap, forgot the libs xD libcm.so and libril-qc-1.so need to be changed, sending to you, sorry, work affects my mind
  10. As I said, ndk gcc complained about mooncake_keypad driver compilation so I had to cheat for it to be compiled. Buttons not working it is something I expected in this case, now if I get home, re-compile there with qvga samsung and still buttons aren't working then I'll be worried (there's no need to cheat keypad driver with my development box). Since mikeioannina is going to upload soon, let's hope he still had samsung qvga enabled in his defconfig :P Why zte...why.... why require samsung code for your devices...why!!!? :'(
  11. try this (same as before, change copybit and flash): http://www.mediafire...5dlsq2d0bkxt9ww The only change I made was to add samsung qvga, let's hope this is what your phone is needing :| Be warned, I had to ssh'd the only linux server we have, had to use ndk, made last minute install of several i686 libs, ndk's gcc was complaining during mooncake_keypad compilation about array index boundaries... so yeah, this kernel might be more problematic xD
  12. Found this on xda-developers: debug.sf.hw=1 debug.composition.type=gpu windowsmgr.max_events_per_sec=150 video.accelerate.hw=1 debug.performance.tuning=1 Gonna see if I can try it in my phone and report Edit:// Well this is rich... it seems to work, boot ok, screen working ok... and I haven't changed copybit yet (unless I've forgotten to revert and I'm still using copybit from blade lol) o: !!!!! mikeioannina, can you test? Edit2: // Forgot to add, I used rom toolbox which reported video.accelerate.hw=1 already, couldn't add windowsmgr.max_events_per_sec, after setting the property name the activity grew beyond my LCD size and the "add" button wasn't touchable, by default is 35, 150 was used by some atrix phone, so I was thinking more about using a number between 45 and 60 just to test. Is possible that debug.* are just for testing and not needed for hw accel and animations to work (with the exception of debug.sf.hw=0 that needs to exist in build.prop to avoid screen bugs)
  13. Good, then tonight I'm going to check a diff between deadlink's and .35 kernel, and since your device is expecting something from fb area I'll concentrate on that and see if I can pinpoint what's happening, why would you need something I don't is beyond me but it's the only starting point I have right now =( Edit:// Checked the configs, one thing that might be responsable is the lack of samsung qvga fb, deadlink has it, vamshi I'm not sure (never got the config of his #68 kernel), so by default, known .32 sources had 3 lcd framebuffers: lead qvga, himax qvga and samsung qvga, himax is the one that Racer "needs" as in it seems to be the principal chip support, lead qvga has to be compiled because it's a dependency of himax (setting himax without setting lead will stop himax compilation) and samsung ... no idea =/ I'm at work now but I'm gonna try to compile a gen1 kernel with only the samsung fb added to see if it works with your phone.
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.