02:54
_whitelogger has joined #rust-embedded
03:48
MayaRodriguez[m4 has joined #rust-embedded
03:48
<
MayaRodriguez[m4 >
Get in touch with this platform for greatness you’ll definitely thank me later
ℹ️❤️
👇🏻👇🏻👇🏻
03:49
MayaRodriguez[m4 has left #rust-embedded [#rust-embedded]
03:53
EthanWu[m] has joined #rust-embedded
03:54
<
i509vcb[m] >
no message deletions of spam?
03:56
<
EthanWu[m] >
two weeks ago I said I made it embassy on stm32mp257f-dk with defmt, promised to open-source later.
03:57
<
EthanWu[m] >
hardware validated
03:59
<
EthanWu[m] >
not a cliche: dont anticipate all llm code to be slop. last mile uio worked helped by llm refactor.
04:28
cr1901 has quit [Ping timeout: 250 seconds]
04:28
cr1901 has joined #rust-embedded
04:37
sroemer has joined #rust-embedded
04:37
sroemer has joined #rust-embedded
06:44
bandini has joined #rust-embedded
06:56
sroemer has quit [Quit: WeeChat 4.7.2]
07:47
rom4ik has quit [Quit: bye]
07:48
rom4ik has joined #rust-embedded
07:49
rom4ik has quit [Client Quit]
07:50
rom4ik has joined #rust-embedded
09:05
rainbyte_ has joined #rust-embedded
09:06
rainbyte has quit [Ping timeout: 256 seconds]
09:09
sroemer has joined #rust-embedded
09:24
rainbyte_ has quit [Ping timeout: 264 seconds]
09:32
rainbyte has joined #rust-embedded
11:33
sroemer has quit [Quit: WeeChat 4.7.2]
13:25
gkoebel has quit [Read error: Connection reset by peer]
13:27
gkoebel has joined #rust-embedded
13:28
gkoebel has quit [Read error: Connection reset by peer]
14:04
gkoebel has joined #rust-embedded
14:04
gkoebel has joined #rust-embedded
14:04
gkoebel has quit [Changing host]
14:33
gkoebel has quit [Ping timeout: 276 seconds]
14:33
gkoebel has joined #rust-embedded
14:33
gkoebel has joined #rust-embedded
14:33
gkoebel has quit [Changing host]
14:49
<
thejpster[m] >
has anyone ever done SEGGER RTT out of secure state on a TrustZone enabled chip?
14:50
<
thejpster[m] >
because I'm trying to and seeing nothing
14:50
diondokter[m] has joined #rust-embedded
14:50
<
diondokter[m] >
Well, secure state is the default. So yes, I have
14:50
<
diondokter[m] >
Which chip?
14:50
<
thejpster[m] >
it's only the default if you set some sticky bit to enable it. Otherwise they boot in nonsecure state by default.
14:50
<
thejpster[m] >
STM32U5A5
14:51
<
diondokter[m] >
Oh? Then STM is different from all other chips I've used
14:51
<
thejpster[m] >
if SRAM is stil at 0x2000_0000 on STM32, then it's definitely nonsecure state. Secure SRAM lives at 0x3000_0000.
14:51
<
diondokter[m] >
The nRFs and NXPs all start in secure state
14:52
<
diondokter[m] >
What's the address the debugger uses to access the RTT buffer?
14:53
<
thejpster[m] >
my RTT buffer is at 300001a8, in secure SRAM
14:53
<
diondokter[m] >
Sure. The chips boots secure into whatever software is there. In that case the SPM
14:53
<
diondokter[m] >
thejpster[m]: Maybe try the nonsecure alias?
14:53
<
thejpster[m] >
but probe-rs gets the address from the symbol table?
14:55
<
diondokter[m] >
thejpster[m]: Just to try, set your memory.x to use the non-secure alias
14:55
<
diondokter[m] >
Maybe that doesn't work on STM though
14:56
<
thejpster[m] >
I didn't expect that to work, but it did
14:57
<
diondokter[m] >
So, often (chip settings dependent), the secure side can access anything in the non-secure alias just fine.
14:57
<
diondokter[m] >
But debugger access is different again and so 🤷♂️
14:57
<
diondokter[m] >
It's all cursed
14:57
<
thejpster[m] >
it begs the question why there are two aliases
14:58
<
thejpster[m] >
I bet I couldn't execute a ram func from 0x2000_0000
14:58
<
diondokter[m] >
Well, the non-secure can't access anything with a secure alias. And regions can be turned off in the non-secure alias
14:58
<
diondokter[m] >
Not sure!
14:59
<
thejpster[m] >
nope, it ran a ram func just fine
15:01
<
diondokter[m] >
Secure accessing non-secure data is a feature. How otherwise would a secure enclave encrypt a byte slice?
15:01
<
diondokter[m] >
It's the other way around that's protected
15:06
sroemer has joined #rust-embedded
15:06
sroemer has quit [Changing host]
15:06
sroemer has joined #rust-embedded
15:14
<
thejpster[m] >
cool, well ThreadX is running in Secure State without issue
15:14
<
thejpster[m] >
but now I see
15:14
<
thejpster[m] >
> warning: linker stderr: rust-lld: address (0xc000258) of section .text is not a multiple of alignment (16)
15:14
<
thejpster[m] >
Is this new cargo now printing linker warnings by default?
15:14
<
diondokter[m] >
Yes
15:15
<
diondokter[m] >
Was changed in the most recent update I think
15:17
<
thejpster[m] >
I'm using a fairly normal cortex-m-rt linker script
15:18
<
thejpster[m] >
0c000040 g O .vector_table 00000218 __INTERRUPTS
15:18
<
thejpster[m] >
0c000258 g .text 00000000 __stext
15:38
bandini has quit [Quit: WeeChat 4.9.4]
16:04
ouilemur has quit [Quit: WeeChat 4.9.4]
16:08
<
az1[m] >
Yeah RA is (mostly) secure / ns callable by default.
16:14
<
az1[m] >
<diondokter[m]> Secure accessing non-secure data is a feature. How otherwise would a secure enclave encrypt a byte slice?...
16:14
<
az1[m] >
Maybe it's chip dependent? I've run into errors trying to get secure access of non-secure data (and some stuff here is NS only like ethernet DMA).
16:14
ouilemur has joined #rust-embedded
16:16
<
diondokter[m] >
az1[m]: It is! IIRC the NXP chips I've used do allow it except when you set some extra vendor bits
16:17
sroemer has quit [Quit: WeeChat 4.7.2]
16:17
<
diondokter[m] >
The biggest issue with trustzone is how it left so many things up to the vendor
16:17
<
az1[m] >
diondokter[m]: Yeah… I think on the RA chips "without" an SAU it doesn't matter but on the chips with an SAU (what Renesas calls their v2 implementation) it does.
16:18
<
az1[m] >
Yep, plenty of opportunity for creativity.
16:20
<
az1[m] >
Plus there's also trustzone privileges if you're feeling extra feisty.
16:24
<
az1[m] >
Actually, scratch that. There's seven different ways on this MCU that a register can respond to security context lol.
16:26
gkoebel has quit [Quit: gkoebel]
16:28
glitchy has quit [Remote host closed the connection]
16:31
<
az1[m] >
Welp the Baochip boards shipped.
16:33
glitchy has joined #rust-embedded
16:33
glitchy has quit [Max SendQ exceeded]
16:36
glitchy has joined #rust-embedded
17:15
reto[m] has joined #rust-embedded
17:15
<
reto[m] >
And arrived, got mine today:)
17:17
glitchy has quit [Remote host closed the connection]
17:19
glitchy has joined #rust-embedded
17:19
glitchy has quit [Max SendQ exceeded]
17:22
glitchy has joined #rust-embedded
17:57
arrubin has joined #rust-embedded
18:49
rainingmessages1 has quit [Quit: bye]
18:49
rainingmessages1 has joined #rust-embedded
19:16
Noah[m]1 has joined #rust-embedded
19:16
<
Noah[m]1 >
<dirbaio[m]> and "homies" and "bro" is an interesting choice of words to talk to a group of 3.8k people you don't know personally
19:16
<
Noah[m]1 >
yo bro you not down with the homies?
19:17
<
Noah[m]1 >
sorry, couldn't resist, somehow this popped up on my screen now and I was chuckling
19:17
<
dirbaio[m] >
u not have enough aura
19:17
<
DnielBuga[m] >
sadge mate, real L attitude
19:17
<
Noah[m]1 >
rare W boiiis
19:18
<
Noah[m]1 >
I really have to resist to not keep this going :D
20:05
dinkelhacker has quit [Server closed connection]
20:05
dinkelhacker has joined #rust-embedded
20:56
adamgreig[m] has quit [Quit: Idle timeout reached: 172800s]
21:45
gkoebel has joined #rust-embedded
21:45
gkoebel has quit [Client Quit]
21:45
gkoebel has joined #rust-embedded
21:45
gkoebel has quit [Client Quit]
23:19
gkoebel has joined #rust-embedded
23:28
gkoebel has quit [Read error: Connection reset by peer]
23:39
majors has quit [Server closed connection]
23:40
majors has joined #rust-embedded