David Hildenbrand <da...@redhat.com> writes: > On 22.02.22 16:22, Markus Armbruster wrote: >> The error message added in commit 3ff333effa "pc-dimm: fix error >> messages if no slots were defined" is misleading: >> >> $ qemu-system-x86_64 -object >> memory-backend-file,id=mem1,size=1M,mem-path=1G.img -device >> pc-dimm,id=dimm1,memdev=mem1 >> qemu-system-x86_64: -device pc-dimm,id=dimm1,memdev=mem1: no slots where >> allocated, please specify the 'slots' option >> $ qemu-system-x86_64 -object >> memory-backend-file,id=mem1,size=1M,mem-path=1G.img -device >> pc-dimm,id=dimm1,memdev=mem1,slots=0 >> qemu-system-x86_64: -device pc-dimm,id=dimm1,memdev=mem1,slots=0: >> Property 'pc-dimm.slots' not found >> >> The property it called 'slot', not 'slots'. With that fixed, we get >> another bad error message: >> >> $ qemu-system-x86_64 -object >> memory-backend-file,id=mem1,size=1M,mem-path=1G.img -device >> pc-dimm,id=dimm1,memdev=mem1,slot=0 >> qemu-system-x86_64: -device pc-dimm,id=dimm1,memdev=mem1,slot=0: invalid >> slot number 0, valid range is [0-18446744073709551615] >> >> Left for another day. >> > > We're referring to the "-m 2g,maxmem=8g,slots=5" slots parameter. And I > agree that we can make that clearer somehow :)
Aha! So this patch actually moves us sideways rather than forward. Please ignore it. A patch that moves us forward would be nice :)
