00:01
leah has joined #linux-amlogic
00:09
vagrantc has quit [Quit: leaving]
00:23
djrscally has quit [Ping timeout: 244 seconds]
00:51
leah has quit [Ping timeout: 260 seconds]
00:52
leah has joined #linux-amlogic
04:02
hexdump0815 has quit [Ping timeout: 268 seconds]
04:17
chewitt has joined #linux-amlogic
07:28
djrscally has joined #linux-amlogic
08:12
_whitelogger has quit [Ping timeout: 260 seconds]
08:13
_whitelogger_ has joined #linux-amlogic
09:04
jacobk has quit [Ping timeout: 276 seconds]
09:05
jacobk has joined #linux-amlogic
10:13
naoki has quit [Quit: naoki]
10:13
naoki1 has joined #linux-amlogic
10:14
naoki1 has quit [Client Quit]
10:14
<
f_ >
Soo another thing :p
10:14
naoki has joined #linux-amlogic
10:14
<
f_ >
This is what gets printed in aquaman UART logs:
10:14
<
f_ >
> Load bl33 from eMMC, src: 0x00094200, des: 0x01700000, size: 0x00085200, part: 0
10:14
<
f_ >
aml log : R1024 check pass!
10:15
<
f_ >
And this is what gets printed in lepotato logs:
10:15
<
f_ >
> Load bl33 from eMMC, src: 0x00050200, des: 0x01000000, size: 0x000eca00
10:15
<
f_ >
Do you notice something odd?
10:17
<
f_ >
"0x01700000 != 0x01000000"
10:18
<
f_ >
I didn't really take a look at bl2_load_images much in bl2, I merely just thought "must be it" and moved on before.
10:19
<
f_ >
But there
*is* a check for secureboot. Specifically, if secureboot is enabled, storage_load always loads images to 0x01700000. Else, the usual load address is honoured
10:19
<
f_ >
storage_load is the one that takes care of printing what it loads, too..
10:20
<
f_ >
I bet they probably overlooked that one
10:20
<
f_ >
or something else
10:28
<
f_ >
And like.. they only do this for the BL* images
10:28
<
f_ >
The FIP header is loaded at the usual address but they then memcpy to 0x1700000 :p
10:28
<
f_ >
storage_load(0xc000,0x1400000,0x4000,s_fip_header_d900981e);
10:28
<
f_ >
memcpy(0x1700000,0x1400000,0x4000)
11:05
<
f_ >
BTW, has anyone tried using gxl FIP format on gxbb?
11:05
<
f_ >
I seem to remember that gxbb BL2 seemed to have some handling of gxl's FIP format
11:06
<
f_ >
well, at least what I remember is under some circumstances it would print "New fip structure!" over UART, just like what gxl usually does.
11:08
<
f_ >
> if (something_fip == L'\x87654321') {serial_puts("New fip structure!\n");} < is present in both gxbb and gxl bl2
12:12
naoki has quit [Quit: naoki]
13:14
paulk has quit [Ping timeout: 268 seconds]
13:15
paulk has joined #linux-amlogic
16:08
jacobk has quit [Ping timeout: 260 seconds]
17:24
jacobk has joined #linux-amlogic
17:30
buzzmarshall has joined #linux-amlogic
17:54
ldevulder has quit [Quit: Leaving]
17:54
ldevulder has joined #linux-amlogic
18:04
ldevulder has quit [Quit: Leaving]
18:32
jacobk has quit [Ping timeout: 276 seconds]
19:19
jacobk has joined #linux-amlogic
21:04
jacobk has quit [Ping timeout: 260 seconds]
21:53
jacobk has joined #linux-amlogic
22:03
jacobk has quit [Ping timeout: 260 seconds]
23:10
konsgn has quit [Ping timeout: 252 seconds]
23:33
naoki has joined #linux-amlogic
23:39
djrscally has quit [Ping timeout: 272 seconds]