[GTALUG] Joy of hardware hacking - was - Help with Kernel: Linux version 3.4.42-gf151976 (hudsoncm at ilclbld86)
Russell Reiter
rreiter91 at gmail.com
Sun Feb 19 11:49:39 EST 2017
I'm top posting again, but with a differently named subject so I can say
this.
My first experience in seeing the joy of hacking in someone else's eyes,
was at a Tlug meeting. At this time I'm pretty much a newbie to the group
and Drew Sullivan turns to me and says, hey Russ, what does dump - blah
blah grep bla pipe more blah do? I said I haven't the foggiest idea. His
eyes lit up in pure joy as he explained that, it retrieved the data from
the tape device and wrote it back to the fixed disk.
So I learned that there is more than one way to do things in *nix and that
position, form and function all count and once the unifying concepts are
understood, the process seems to get easier from there. Not so much a how
do I do this? but why do I do it this way?
russ at HECTOR:~$ adb shell
shell at ghost:/ $ ls
acct
<snip the file system>
So for this project, it now seems I may able to blunt force UN-cruft some
of the problems. Likely there is an issue with the surface flinger (sort of
compositor and quite like a packer for TCL/TK, rolled into one) and HAL.
Disabling HW overlay forces everything to OpenGL, (GLES) It looks like the
VZW carrier hooks into the notification bar are abusing something.
My feelings at this point are, there was a reason only the notification bar
was showing up (fuzzy) at first and you couldn't swipe up a new screen with
out it blacking out. That is until you gave it the good old masons
handshake, some people call it percussive maintenance, I tend to call it
blunt force hacking. Whatever it is, I think the overlay's for any VZW
implemented animations and icons on the notice bar are bogus and I might be
able to pull those hooks out manually.
In any case I know know this device runs Android 5.1 and that I may be able
to regress to kit kat 4.4 which is what a MOTO XT1058 runs on.
Sorry can't resist the pun, looks like lots of diff - occult practices
needed now.
Russell
On Wed, Feb 15, 2017 at 4:51 PM, Russell Reiter <rreiter91 at gmail.com> wrote:
> Top posting this as it may be of some general interest.
>
> As I related in my earlier post, gaining access to apps on this phone was
> near impossible at first. In addition to finally enabling developer mode
> and those other changes I related below, I turned off all animations and
> chose simulated monochrome as color correction.
>
> I didn't really think a monochrome simulation would ease processor
> stress's, as simulating monochrome might be just like making brown sugar;
> you know, just adding the molasses back into the refined sugar, so to
> speak. - It worked tho.
>
> While waking the thing up still takes a lot of finger tapping and power
> button screen redrawing, it seems once NAND is fully energized and
> unsuspended, it behaves normally. No unexpected screen outs during use, the
> back of the unit is noticably cooler and no Row Access Strobe lines on the
> display.
>
> So while I still have to shake hands like a mason to get it going - it's
> going and I am, as always, having fun with a Linux kernel - even if it is
> cross compiled on a broken platform.
>
> So now I confess, I bought, for the first time in my life, a M$ Windows
> operating system. I've done some work on it and had copies donated to me by
> Dell for a volunteer project but I never bought it.
>
> Well I didn't specifically buy the OS, it came on the HP mini 110 I
> purchased, but I did activate it. I did that just to see Windows 10 before
> I check out Linux on Atom.
>
> As for this phone, not a screen flicker, blackout or other glitch in the
> entire time it took me to one finger compose this post.
> 20 min+-
>
> Russell
> Sent from a semi-resurrected Moto XT1060
> Almost ready to mod.
>
> On Feb 13, 2017 11:32 AM, "Russell Reiter" <rreiter91 at gmail.com> wrote:
>
> I like to tinker with stuff and this is also a kind of semi-long, maybe
> even funny story.
>
> I am asked quite often to help people with Android problems so I fool
> around with phones, usually theirs not mine. I did get to like Motorola
> enough to buy a Moto G and use it. A couple of months ago I was looking
> through ebay and saw an ad for an Unlocked Moto X and assuming they meant
> carrier unlocked, I ordered it.
>
> When it arrived after the estimated delivery date, I charged it and
> inserted my sim and no joy, it was locked to Verizon. I opened a trouble
> ticket and tried to get the seller to provide me with an unlock code but
> they couldn't. I guess because there isn't one. Ebay has refunded me and I
> guess the seller settled with them for false advertising because I am now
> in possession of a semi-borked Moto XT1060. (Looking at CRdroid at this
> point)
>
> I say semi-borked because after I rebooted without the sim, all sorts of
> gremlins showed up. Most especially what I believe is a Verizon "ghost"
> carrier induced BSOD. After a couple of intermittent hours of booting and
> fuzzing, I got enough of a screen to be able to navigate to settings,
> enable developer mode and force GPU rendering and disable HW overlays. Now
> I can actually use the thing on wifi. It still takes a lot of keypresses
> and screen taps to fuzz the thing on, but once I got the screen on, I was
> able to generate a bug report and mail it to myself.
>
> It also sent me a screen-shot of the screen I couldn't see at the time. At
> first foray I was operating in the very narrow space at the top of the
> screen, which I kept alive by dragging my finger down from the top and
> pressing the power button repeatedly to force a redraw of the screen.
>
> I now know, after getting the report, my phone thinks it is 10.24PM,
> Monday February 13th. Actually it looks like the bloatware deliberatly
> overheats the cpu and so far, as long as I keep the phone in the cold space
> between my windows, it seems to function almost normally, except it won't
> accept the sim card.
>
> So it seems the boot loader is locked at NAND: 0x30BE. Has anyone on the
> list had success in flashing an open source boot loader to a Motorola
> device?
>
> I've looked around a bit and the ghost rom is available but I can't seem
> to find a comparable stock motorola rom for this model.
>
> Any advice is appreciated and here's a couple hundred lines of debug code
> for anyone who is interested.
>
> Thanks in advance
> Russell
>
> ========================================================
> == dumpstate: 2017-02-13 22:24:05
> ========================================================
>
> Build: LPAS23.12-39.7-1
> Build fingerprint: 'motorola/ghost_verizon/ghost:
> 5.1/LPAS23.12-39.7-1/1:user/release-keys'
> Bootloader: 0x30BE
> Radio: msm
> Network: (unknown)
> Kernel: Linux version 3.4.42-gf151976 (hudsoncm at ilclbld86) (gcc version
> 4.8 (GCC) ) #1 SMP PREEMPT Thu Mar 24 13:07:11 CDT 2016
> Command line: console=null androidboot.hardware=qcom user_debug=31
> msm_rtb.filter=0x3F ehci-hcd.park=3 maxcpus=2 vmalloc=400M
> androidboot.emmc=true androidboot.serialno=T062808WJT
> androidboot.baseband=msm androidboot.mode=normal
> androidboot.bootloader=0x30BE androidboot.cid=0x2 androidboot.hwrev=0x8300
> androidboot.radio=0x2 battery= mot_sst=1 androidboot.modelno="Moto X"
> androidboot.carrier=vzw product=ghost androidboot.device=ghost
> androidboot.secure_hardware=1 androidboot.write_protect=1
> restart.download_mode=1
>
> ------ UPTIME (uptime) ------
> up time: 05:22:57, idle time: 00:23:57, sleep time: 04:51:47
> [uptime: 0.100s elapsed]
>
> ------ MMC PERF (/sys/block/mmcblk0/stat) ------
> 28364 8048 2870460 156850 15886 4451 783348
> 94420 0 90690 251070
>
> ------ MEMORY INFO (/proc/meminfo) ------
> MemTotal: 1836564 kB
> MemFree: 332876 kB
> Buffers: 6452 kB
> Cached: 853692 kB
> SwapCached: 0 kB
> Active: 870116 kB
> Inactive: 468376 kB
> Active(anon): 479560 kB
> Inactive(anon): 1692 kB
> Active(file): 390556 kB
> Inactive(file): 466684 kB
> Unevictable: 1092 kB
> Mlocked: 0 kB
> HighTotal: 1316860 kB
> HighFree: 101020 kB
> LowTotal: 519704 kB
> LowFree: 231856 kB
> SwapTotal: 0 kB
> SwapFree: 0 kB
> Dirty: 76 kB
> Writeback: 20 kB
> AnonPages: 479596 kB
> Mapped: 263488 kB
> Shmem: 1772 kB
> Slab: 41212 kB
> SReclaimable: 14572 kB
> SUnreclaim: 26640 kB
> KernelStack: 10296 kB
> PageTables: 24060 kB
> NFS_Unstable: 0 kB
> Bounce: 0 kB
> WritebackTmp: 0 kB
> CommitLimit: 918280 kB
> Committed_AS: 36379564 kB
> VmallocTotal: 401408 kB
> VmallocUsed: 91272 kB
> VmallocChunk: 150532 kB
>
> ------ CPU INFO (top -n 1 -d 1 -m 30 -t) ------
>
>
>
> User 66%, System 26%, IOW 1%, IRQ 0%
> User 115 + Nice 51 + Sys 66 + Idle 14 + IOW 4 + IRQ 0 + SIRQ 0 = 250
>
> PID TID PR CPU% S VSS RSS PCY UID Thread Proc
> 3263 3526 1 21% R 962764K 59040K fg u0_a46 MotoCareBgThrea
> com.motorola.motocare
> 12895 12895 0 8% R 19124K 1608K shell top top
> 1195 2116 1 8% S 1078640K 106124K fg u0_a12 RenderThread
> com.android.systemui
> 282 282 1 6% S 86956K 11388K fg system surfaceflinger
> /system/bin/surfaceflinger
> 1195 1195 1 4% R 1078640K 106124K fg u0_a12 ndroid.systemui
> com.android.systemui
> 981 1011 1 2% S 1161192K 115372K fg system PowerManagerSer
> system_server
> 981 998 1 1% S 1161192K 115372K fg system Binder_2
> system_server
> 981 1129 0 1% S 1161192K 115372K fg system WifiStateMachin
> system_server
> 981 1566 1 1% S 1161192K 115372K fg system Binder_E
> system_server
> 981 981 1 1% S 1161192K 115372K fg system system_server
> system_server
> 981 1363 1 0% S 1161192K 115372K fg system Binder_5
> system_server
> 981 1008 1 0% S 1161192K 115372K fg system android.display
> system_server
> 282 607 1 0% S 86956K 11388K fg system EventThread
> /system/bin/surfaceflinger
> 3263 3525 1 0% S 962764K 59040K fg u0_a46 MotoCareSvcThre
> com.motorola.motocare
> 16 16 0 0% S 0K 0K root kworker/u:1
> 1452 1477 0 0% S 960692K 48180K fg system GCDaemon
> com.motorola.process.system
> 1352 1373 1 0% S 961936K 50780K fg u0_a39 GCDaemon
> com.motorola.motodisplay
> 1579 1950 1 0% R 1009952K 66604K fg u0_a58 RenderThread
> com.android.launcher
> 1352 3814 1 0% S 961936K 50780K fg u0_a39 breathing drawe
> com.motorola.motodisplay
> 282 1587 1 0% S 86956K 11388K fg system Binder_4
> /system/bin/surfaceflinger
> 282 2545 1 0% S 86956K 11388K fg system Binder_5
> /system/bin/surfaceflinger
> 149 149 0 0% S 0K 0K root mmcqd/0
> 1195 1615 1 0% S 1078640K 106124K fg u0_a12 Binder_3
> com.android.systemui
> 981 1092 1 0% S 1161192K 115372K fg system Binder_3
> system_server
> 1579 1579 1 0% R 1009952K 66604K fg u0_a58 ndroid.launcher
> com.android.launcher
> 1263 1284 1 0% S 945312K 34852K fg u0_a61 Binder_1
> com.motorola.slpc
> 1195 12432 1 0% S 1078640K 106124K fg u0_a12 hwuiTask1
> com.android.systemui
> 981 1465 1 0% S 1161192K 115372K fg system Binder_8
> system_server
> 2429 2602 1 0% R 1113684K 100784K fg u0_a4 GoogleLocationS
> com.google.android.gms.persistent
> 2429 2603 1 0% S 1113684K 100784K fg u0_a4 Binder_5
> com.google.android.gms.persistent
> [top: 1.502s elapsed]
>
> ------ PROCRANK (procrank) ------
> *** exec(procrank): Permission denied
> *** procrank: Exit code 255
> [procrank: 0.100s elapsed]
>
> ------ VIRTUAL MEMORY STATS (/proc/vmstat) ------
> nr_free_pages 83681
> nr_inactive_anon 424
> nr_active_anon 119775
> nr_inactive_file 116690
> nr_active_file 97650
> nr_unevictable 273
> nr_mlock 0
> nr_anon_pages 119749
> nr_mapped 65894
> nr_file_pages 215057
> nr_dirty 30
> nr_writeback 0
> nr_slab_reclaimable 4067
> nr_slab_unreclaimable 6700
> nr_page_table_pages 6014
> nr_kernel_stack 1290
> nr_unstable 0
> nr_bounce 0
> nr_vmscan_write 0
> nr_vmscan_immediate_reclaim 7
> nr_writeback_temp 0
> nr_isolated_anon 0
> nr_isolated_file 0
> nr_shmem 444
> nr_dirtied 78103
> nr_written 66652
> nr_anon_transparent_hugepages 0
> nr_free_cma 0
> nr_dirty_threshold 15874
> nr_dirty_background_threshold 3968
> pgpgin 1435230
> pgpgout 266750
> pswpin 0
> pswpout 0
> pgalloc_normal 2409175
> pgalloc_high 2727696
> pgalloc_movable 0
> pgfree 5348019
> pgactivate 155050
> pgdeactivate 46
> pgfault 7022719
> pgmajfault 9302
> pgrefill_normal 0
> pgrefill_high 0
> pgrefill_movable 0
> pgsteal_kswapd_normal 27457
> pgsteal_kswapd_high 127101
> pgsteal_kswapd_movable 0
> pgsteal_direct_normal 290
> pgsteal_direct_high 1393
> pgsteal_direct_movable 0
> pgscan_kswapd_normal 31501
> pgscan_kswapd_high 163346
> pgscan_kswapd_movable 0
> pgscan_direct_normal 290
> pgscan_direct_high 2118
> pgscan_direct_movable 0
> pginodesteal 0
> slabs_scanned 48256
> kswapd_inodesteal 11
> kswapd_low_wmark_hit_quickly 74
> kswapd_high_wmark_hit_quickly 142
> kswapd_skip_congestion_wait 3
> pageoutrun 2831
> allocstall 31
> pgrotated 263
> compact_blocks_moved 5547
> compact_pages_moved 113444
> compact_pagemigrate_failed 1389
> compact_stall 0
> compact_fail 0
> compact_success 0
> compact_retry_success 0
> unevictable_pgs_culled 273
> unevictable_pgs_scanned 0
> unevictable_pgs_rescued 0
> unevictable_pgs_mlocked 0
> unevictable_pgs_munlocked 0
> unevictable_pgs_cleared 0
> unevictable_pgs_stranded 0
> unevictable_pgs_mlockfreed 0
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://gtalug.org/pipermail/talk/attachments/20170219/027c8cd8/attachment.html>
More information about the talk
mailing list