00:08
<
marex >
xypron: they still sell those sispmctl usb-controlled outlets ?
00:12
<
marex >
oh wow, they still do
00:32
qschulz has quit [Quit: qschulz]
00:35
qschulz has joined #u-boot
01:34
umbramalison has quit [Quit: %So long and thanks for all the fish%]
01:35
umbramalison has joined #u-boot
01:44
thopiekar has quit [Ping timeout: 252 seconds]
01:45
thopiekar has joined #u-boot
01:46
pavelow has quit [Quit: No Ping reply in 180 seconds.]
01:46
pavelow has joined #u-boot
02:18
vagrantc has quit [Quit: leaving]
02:23
LeSpocky has quit [Ping timeout: 252 seconds]
02:24
LeSpocky has joined #u-boot
02:31
mmu_man has quit [Ping timeout: 272 seconds]
05:54
hanetzer has quit [Ping timeout: 272 seconds]
05:55
rvalue has quit [Read error: Connection reset by peer]
05:56
rvalue has joined #u-boot
05:56
hanetzer has joined #u-boot
06:00
akaWolf has quit [Ping timeout: 272 seconds]
06:02
akaWolf has joined #u-boot
06:11
hanetzer has quit [Ping timeout: 245 seconds]
06:13
hanetzer has joined #u-boot
06:15
GNUtoo has quit [Remote host closed the connection]
06:15
GNUtoo has joined #u-boot
06:28
rvalue has quit [Ping timeout: 245 seconds]
06:46
rvalue has joined #u-boot
08:04
alpernebbi has joined #u-boot
08:36
thopiekar has quit [Ping timeout: 252 seconds]
08:38
thopiekar has joined #u-boot
09:03
Xeroine has joined #u-boot
09:06
mmu_man has joined #u-boot
10:19
Xeroine has quit [Ping timeout: 260 seconds]
10:38
gsz has joined #u-boot
10:41
mmu_man has quit [Ping timeout: 252 seconds]
11:12
Xeroine has joined #u-boot
11:28
akaWolf has quit [Ping timeout: 252 seconds]
12:02
akaWolf has joined #u-boot
13:21
haritz has quit [Remote host closed the connection]
13:29
rvalue has quit [Read error: Connection reset by peer]
13:29
rvalue has joined #u-boot
14:20
BWhitten has joined #u-boot
14:58
vagrantc has joined #u-boot
15:05
gsz has quit [Ping timeout: 245 seconds]
15:45
gsz has joined #u-boot
15:58
haritz has joined #u-boot
15:58
haritz has joined #u-boot
16:02
___nick___ has joined #u-boot
16:03
haritz has quit [Remote host closed the connection]
16:42
___nick___ has joined #u-boot
16:43
___nick___ has quit [Client Quit]
16:45
___nick___ has joined #u-boot
16:54
akaWolf has quit [Ping timeout: 252 seconds]
17:00
haritz has joined #u-boot
17:00
haritz has joined #u-boot
17:01
akaWolf has joined #u-boot
17:13
mmu_man has joined #u-boot
17:40
gsz has quit [Quit: leaving]
17:57
thopiekar has quit [Ping timeout: 240 seconds]
17:59
thopiekar has joined #u-boot
18:02
prabhakarlad has quit [Quit: Client closed]
18:28
urja has quit [Quit: WeeChat 3.4]
18:30
urja has joined #u-boot
18:30
<
sjg1 >
Tartarus: I sent another patch for common.h - if we can turn it around quickly I could do more for this release. There are about 15 patches to land. I could send more than one at a time if that would help, but any other PRs can often invalidate it, which is why I ask about timing
18:44
persmule has quit [Ping timeout: 268 seconds]
18:46
persmule has joined #u-boot
18:59
<
marex >
sjg1: are you around for a short binman question ?
19:02
<
marex >
sjg1: it's essentially a post-processing of imx8m flash.bin generated using binman
19:03
<
marex >
sjg1: it calls external tool, cst, for which you need to generate custom input text file
19:03
<
marex >
sjg1: I wonder whether this can be wrapped into binman/bintool, I'm sure it can
19:03
<
marex >
sjg1: that would help trimming all the scripted savagery
19:06
<
sjg1 >
marex: Yes it looks like that could be done with a new entry type
19:06
<
marex >
new entry type ?
19:08
<
marex >
sjg1: so this would be a new entry in the fitImage ?
19:12
<
sjg1 >
marex: It could be. From what I can tell you need cst to sign things
19:12
<
marex >
sjg1: yes, I need to sign the SPL after its been augmented with mkimage header , i.e. adding anything to fitImage is useless at that point
19:13
<
marex >
and I need to sign fitImage components, which is the second half of it (the design there is demented)
19:13
<
marex >
sjg1: I think the vblock is the second half ?
19:14
<
marex >
the nxp approach signs the blobs within the fitImage and the tree, so it seems to partly duplicate what fitImage already supports
19:18
<
sjg1 >
marex: You can fairly easily sign FIT parts, if the FIT is built by binman. The contents of a FIT are just other entries, so one of those entries can be the signature stuff
19:19
<
sjg1 >
marex: gbb is another entry type (etype) supported by binman
19:21
<
marex >
sjg1: no, one of the entries cannot be the signature stuff
19:21
<
marex >
sjg1: the CST blob has to be exactly 0x1000 bytes past the fitImage tree and the DEK (encryption) blob must be 0x2000 bytes past the CST blob
19:22
xroumegue has joined #u-boot
19:22
<
marex >
oh, and a generated IVT has to be placed at fitImage tree + 0x1000 - 0x20
19:22
xroumegue has quit [Client Quit]
19:25
<
sjg1 >
marex: Then use offset = <(0x1000 - 0x20)> in the IVT entry
19:28
<
marex >
sjg1: I don't understand
19:43
<
marex >
sjg1: I need to place the generated blobs at specific offsets, not at random offsets where binman somehow decides to place them
19:45
<
marex >
sjg1: also, I am not looking for introduction to DT or how to describe things in DT, I am looking for prior art how to implement what I need
19:51
xroumegue has joined #u-boot
20:02
<
sjg1 >
marex: Yeah well that is the point of the 'offset' property. It is relative to its parent entry
20:03
<
sjg1 >
marex: I've given you the prior art I can think of
20:03
Xeroine has quit [Ping timeout: 252 seconds]
20:03
<
sjg1 >
marex: You might also find all the rockchip stuff happy as they have been moving things to binman too
20:04
___nick___ has quit [Ping timeout: 245 seconds]
20:18
jpox has joined #u-boot
20:18
<
Forty-Bot >
why does nxp name all of its secure boot stuff "cst"
20:18
<
Forty-Bot >
even when they're completely different tools...
20:39
sbach has quit [Read error: Connection reset by peer]
20:42
sbach has joined #u-boot
20:42
<
marex >
Forty-Bot: Code Signing Tool
20:42
<
marex >
Forty-Bot: they also have CSF, Code Signing File I think
20:42
<
Forty-Bot >
I know what it stands for
20:42
<
marex >
and they also have a lot of other TLAs
20:42
<
Forty-Bot >
but it bugs me that there's like 3 different tools with the same name
20:42
<
marex >
security and TLAs are cool
20:42
<
marex >
Forty-Bot: there is only one CST for HAB
20:43
<
Forty-Bot >
yeah, but there's another cst for qoriq
20:43
<
Forty-Bot >
named the same thing, but totally different tool
20:43
<
marex >
Forty-Bot: is that AHAB CST ?
20:43
<
Forty-Bot >
no, it's some other format
20:46
<
marex >
Forty-Bot: maybe they wanted it to be more convoluted and confusing, security by obscurity :)
20:46
<
marex >
it's so secure you cannot even get to the security, let alone mess with it
20:46
<
Forty-Bot >
the thing is
20:46
<
Forty-Bot >
they just call openssl
20:47
<
Forty-Bot >
and the whole tool is a wrapper for the binary format
20:47
<
marex >
Forty-Bot: but its binary, proprietary, seeeeekrit
20:47
<
marex >
(it did cross my mind to just rewrite it from scratch)
20:47
<
Forty-Bot >
if I had 6 months I'd do it
20:48
<
marex >
its probably not that complex
20:48
<
marex >
what is higher on my list of crap to fix is the DDR calibrator
20:48
<
Forty-Bot >
yeah, but I don't have self control
20:49
persmule has quit [Remote host closed the connection]
20:51
persmule has joined #u-boot
20:55
persmule has quit [Read error: Connection reset by peer]
20:55
GNUtoo has quit [Write error: Connection reset by peer]
20:56
persmule has joined #u-boot
21:00
GNUtoo has joined #u-boot
21:31
BWhitten has quit [Ping timeout: 245 seconds]
21:50
gsz has joined #u-boot
22:27
haritz has quit [Ping timeout: 240 seconds]
22:30
jpox has quit [Ping timeout: 240 seconds]
23:09
haritz has joined #u-boot
23:09
haritz has joined #u-boot
23:12
haritz has quit [Remote host closed the connection]
23:46
haritzondo has joined #u-boot
23:50
haritzondo has quit [Client Quit]
23:50
haritz has joined #u-boot
23:50
haritz has joined #u-boot