Dave Thaler
Dave Thaler
> > So endiannes needs to be a parameter, no? > > Agreed. I may or may not get to this for a couple of days though, busy at a...
> @dthaler this needs to be reopened I'm still working on changes for big-endian systems. Just slow because I have a bunch of other things going on besides this but...
llvm-objdump shows these instructions as ld_pseudo: ``` 18 00 00 00 ff ff ff ff 00 00 00 00 ff ff ff ff r0 = -0x1 ll 18 11 00...
The IETF [specification](https://www.ietf.org/archive/id/draft-ietf-bpf-isa-00.html#name-64-bit-immediate-instructio) does cover these now.
@elazarg I believe I've now addressed your comments. Please re-review.
This appears to be related to the TODO comment in https://github.com/vbpf/ebpf-verifier/blob/main/src/crab/split_dbm.hpp#L337
`SplitDBM::widen()` gives the crab log: ``` Before widening: DBM 1 [ meta_offset=[-4098, 0], packet_size=[0, 65534], r1.type=number, r10.stack_offset=512, r10.type=stack, r10.value=[512, 2147418112], r2.type=number, r2.value=[0, 1], r3.type=number, r4.type=number, r4.value=1, s[480...487].value=0, s[488...495].value=0, s[496...503].value=0, s[504...511].value=0]...
Can you provide either the .o file or the disassembly? Specifically, does the disassembly contain anything that PREVAIL thinks is a reference to a map?
@elazarg can the title of this issue be renamed to "Prevail does not support function calls" perhaps?
> NOTE that relref is a deprecated element in xml2rfc v3. Where is that stated? RFC 7991 doesn't say that anywhere I can find, and in fact says relref is...