I am encountering issues during a memory probe with the following details

[Roma10boss]

Memory Map:
START END #PAGES
0000000000000000 - 000000000009ffff 000000a0
00000000000c0000 - 00000000dfffffff 000dff40
00000000fb000000 - 00000000fba68fff 00000a69

Current Action: Probing Memory
Access Mode: Normal
Progress: 33267 / 33267 (100%)
Speed: 67 MB/s
Address: 0x000000081F380000
Pages read: 920137 / 8516480 (10%)
Pages failed: 7596343 (89%)
Memory Probe: Completed.

I'm unsure about the significance of the high failure rate, I would appreciate any guidance on resolving this issue. If there are specific steps I can take or additional information needed, please let me know. Thank you for your help.

[ufrisk]

Try to avoid the probe command if possible. Probe is unfortunately super aggressive and if you're on AMD things are guaranteed to stop working until you reboot.

You can check if DMA is working with pcileech.exe display -min 0x1000 -device fpga -v

If wanting to dump memory on AMD use a memory map, either you can try -memmap auto or you can create your own memory map for your targeted system (best way to do it).

More info at: https://github.com/ufrisk/LeechCore/wiki/Device_FPGA_AMD_Thunderbolt

[serkonda7]

@ufrisk could you please add this disclaimer regarding aggressive probe to the readme.

[ufrisk]

Thanks for the suggestion. I'll add a warning about it since it's not the first time I get this question. I'll add it probably on Thursday since I'll update some cached offsets in info.db after this months patch Tuesday anyway then. It will look like the below image.

[serkonda7]

Awesome! Thank you very much

[FocuzJS]

This doesn't effect 3xxx and newer AMD CPUs, it's mainly the 2xxx generation I've seen with the bridge disabling bus mastering in firmware on a bad map call 😃