12:28 <marekvrbka> Maybe this is fixed in regular OpenOCD, but I found an odd behavior. When using load_image or verify_image with file of size 4 or 8 bytes. It does not respond and nothing happens.

12:29 <marekvrbka> This was on the RISC-V fork of OpenOCD, maybe its fixed in the upstream

13:07 <marekvrbka> Ok, so I digged into the source code and found out that OpenOCD requires the program binary to be at least 9 bytes

13:07 <marekvrbka> In image.c:52

13:08 <marekvrbka> I can submit a patch which instead sets the image type as binary if not enough bytes could be read.

13:08 <marekvrbka> Should I do that?

13:25 <karlp> for my own curiousity, what are you loading that's that size?

13:33 <marekvrbka> Small testing programs. This was basically just a nop and a jump to itself. Need them for testing

13:37 <phr3ak> is there any command to put a string to memory or do you know a script that would utilize mwb command to do this?

13:46 <marekvrbka> That is very easy with GDB, but with tcl, no idea

13:52 <phr3ak> I could not do this easily with GDB

13:52 <phr3ak> I used array like {'H', 'e', 'l', 'l', 'o' } and it is painful

13:53 <phr3ak> and would be better to do this in openocd

13:57 <marekvrbka> Try:

13:57 <marekvrbka> p value="Hello"

14:01 <phr3ak> set (char[6])*0x20000000="hello"

14:57 <m__b> Hi all, I'm trying to debug the second core of a dual-core chip (stm32h7x5), but I'm having some trouble getting it to work, and I'm not sure where to look next.

14:58 <m__b> cpu0 debugging works great, so openocd is working, I think I'm just missing some config somewhere, but I'm not sure what's missing.

15:18 <borneoa___> m__b: are you using stlink?

15:22 <m__b> Yep, stlink v3

15:23 <m__b> I'm attempting to debug Zephyr on the M4 core if it makes a difference :)

15:40 <merethan> High-five!

15:40 <merethan> I'm still writing up my own build system though. No running yet.

16:12 <borneoa___> m__b: stlink can work with hla transport that only access the first AP ( thus only first core) and dap_direct transport that can access both. Check your config file

16:14 <borneoa___> m__b: the other possible issue is that that first core is in charge of starting the second. Maybe the second core is still not accessible. I don't know what should be done in this case as I never used these devices

16:28 <m__b> That's fair. I've found the dap_direct setting and have been messing with my config to see if I can get something up and running. I have firmware on core 0 that should set up the clocks, but maybe there's more there to figure out. Thanks for the suggestions!