Conversation

"RVA23 without V extension" is a new one

1
0
1
@never_released I'm assuming that a quote from a device manufacturer, not a distro etc. What device or IP actually is it?
1
0
1
@never_released Hmm, it's using the cp100 core from ultrarisc that patches have started appearing for. I didn't look super hard, but what's on google etc claims "Support RV64GCBHX, X being high performance arithmetic and load/store instructions", which is short of the claimed rva23 minus V - although perhaps that just to avoid the gigantic string that would produce.
1
0
1

@conor what I see is rv64imafdcbh_zicntr_zicsr_zifencei_zihpm (from a Geekbench result post)

1
0
1
@never_released off the top of my head, I don't remember if everything after the first _ is stuff the kernel automatically adds or not, but most of them are because they were part of i when kernel port was merged. That would put it much shorter of rva23 compliance if accurate.
0
0
2