Fix BeagleBone Black panic on system start

[kjopek_gmail.com]

Summary

There is now assertion which requires all memory allocations of positive size.
Negative and zero-sized allocations lead to panic, so plug them off.

Test Plan

Tested on my BBB.

[kjopek_gmail.com]

When used with latest DTB we still face panic, however related to event timer:

gpioled0: <GPIO LEDs> on ofwbus0
gpioled0: <beaglebone:green:heartbeat> failed to map pin
gpioled0: <beaglebone:green:mmc0> failed to map pin
gpioled0: <beaglebone:green:usr2> failed to map pin
gpioled0: <beaglebone:green:usr3> failed to map pin
panic: No usable event timer found!
cpuid = 0
time = 1
KDB: stack backtrace:
db_trace_self() at db_trace_self
         pc = 0xc05f124c  lr = 0xc007aa34 (db_trace_self_wrapper+0x30)
         sp = 0xc0f14cb8  fp = 0xc0f14dd0
db_trace_self_wrapper() at db_trace_self_wrapper+0x30
         pc = 0xc007aa34  lr = 0xc02ec554 (vpanic+0x140)
         sp = 0xc0f14dd8  fp = 0xc0f14df8
         r4 = 0x00000100  r5 = 0x00000000
         r6 = 0xc07dae31  r7 = 0xc0b22ea8
vpanic() at vpanic+0x140
         pc = 0xc02ec554  lr = 0xc02ec334 (doadump)
         sp = 0xc0f14e00  fp = 0xc0f14e04
         r4 = 0xc0b12f54  r5 = 0x00000001
         r6 = 0xc0b12f78  r7 = 0xc098e808
         r8 = 0x7fffffff  r9 = 0xffffffff
        r10 = 0xc0f05008
doadump() at doadump
         pc = 0xc02ec334  lr = 0xc027a368 (cpu_initclocks_bsp+0x8e8)
         sp = 0xc0f14e0c  fp = 0xc0f14e50
         r4 = 0xc0f05008  r5 = 0xc0f14e04
         r6 = 0xc02ec334 r10 = 0xc0f14e0c
cpu_initclocks_bsp() at cpu_initclocks_bsp+0x8e8
         pc = 0xc027a368  lr = 0xc0279210 (initclocks+0x2c)
         sp = 0xc0f14e58  fp = 0xc0f14e58
         r4 = 0xc0b12558  r5 = 0x0fffffff
         r6 = 0xc278c450  r7 = 0xc08e605c
         r8 = 0x00000000  r9 = 0x04000000
        r10 = 0xc0b12554
initclocks() at initclocks+0x2c
         pc = 0xc0279210  lr = 0xc0274b50 (mi_startup+0x2b0)
         sp = 0xc0f14e60  fp = 0xc0f14e88
mi_startup() at mi_startup+0x2b0
         pc = 0xc0274b50  lr = 0xc0000344 (btext+0x144)
         sp = 0xc0f14e90  fp = 0x00000000
         r4 = 0xc0000480  r5 = 0xc0bd0000
         r6 = 0x0000000a  r7 = 0x00c52078
         r8 = 0xc0e52000  r9 = 0x9cf03dd8
        r10 = 0x00000000
btext() at btext+0x144
         pc = 0xc0000344  lr = 0xc0000344 (btext+0x144)
         sp = 0xc0f14e90  fp = 0x00000000
KDB: enter: panic
[ thread pid 0 tid 100000 ]
Stopped at      kdb_enter+0x54: ldrb    r15, [r15, r15, ror r15]!
db>

[imp]

bummed to know there's still a panic in booting (I have issues like this when I updated my pandaboard too).
However, other panics shouldn't stop us from fixing this one.
TI DTS/FDT has always been a tough gig to follow, since it moves so fast (eg at all, it's not supposed to in theory), so I wish you luck and offer encouragement.