Use u32 for Vsock related buffer sizes

[RiverPhillips]

Changes

Use u32 for all Vsock reletated buffers instead of usize, since the Virtio specification states that lengths of virtio-buffers fit into u32

Reason

Close #4627

License Acceptance

By submitting this pull request, I confirm that my contribution is made under
the terms of the Apache 2.0 license. For more information on following Developer
Certificate of Origin and signing off your commits, please check
CONTRIBUTING.md.

PR Checklist

• If a specific issue led to this PR, this PR closes the issue.
• The description of changes is clear and encompassing.
• Any required documentation changes (code and docs) are included in this
  PR.
• API changes follow the Runbook for Firecracker API changes.
• User-facing changes are mentioned in CHANGELOG.md.
• All added/changed functionality is tested.
• New TODOs link to an issue.
• Commits meet
  contribution quality standards.

---

• This functionality cannot be added in rust-vmm.

[roypat]

Hi, thanks for picking this up! After looking at this and #4637 again, I'm starting to think that changing everything in iovec.rs to u32 might not be that good an idea after all, with how much try_into() it introduces because of usize<->u32 conversations (the main goal for this exercise of using u32 was that I was hoping we'd eliminate try_into()s).

Could you drop all changes outside of the vsock module from this PR, and do the usize<->u32 thing inside {write,read}_at_offset_from in packet.rs, without any changes in iovec.rs? It should look something like

pub fn read_at_offset_from<T: ReadVolatile + Debug>(
    &mut self,
    src: &mut T,
    offset: u32,
    count: u32,
) -> Result<u32, VsockError> {
    match self.buffer {
        VsockPacketBuffer::Tx(_) => Err(VsockError::UnwritableDescriptor),
        VsockPacketBuffer::Rx(ref mut buffer) => {
            if count
                > buffer.len()
                    .saturating_sub(VSOCK_PKT_HDR_SIZE)
                    .saturating_sub(offset)
            {
                return Err(VsockError::GuestMemoryBounds);
            }

            buffer
                .write_volatile_at(src, (offset + VSOCK_PKT_HDR_SIZE) as usize, count as usize)
                .map_err(|err| VsockError::GuestMemoryMmap(GuestMemoryError::from(err)))
                .and_then(|read| read.try_into().map_err(|_| VsockError::DescChainOverflow))
        }
    }
}

As an aside (I can't leave a comment on it), in connection.rs in line 241, you can do

diff --git a/src/vmm/src/devices/virtio/vsock/csm/connection.rs b/src/vmm/src/devices/virtio/vsock/csm/connection.rs
index 9fe744058..c1ed4eeb0 100644
--- a/src/vmm/src/devices/virtio/vsock/csm/connection.rs
+++ b/src/vmm/src/devices/virtio/vsock/csm/connection.rs
@@ -237,8 +237,7 @@ where
                         // length of the read data.
                         // Safe to unwrap because read_cnt is no more than max_len, which is bounded
                         // by self.peer_avail_credit(), a u32 internally.
-                        pkt.set_op(uapi::VSOCK_OP_RW)
-                            .set_len(u32::try_from(read_cnt).unwrap());
+                        pkt.set_op(uapi::VSOCK_OP_RW).set_len(read_cnt);
                         METRICS.rx_bytes_count.add(read_cnt as u64);
                     }
                     self.rx_cnt += Wrapping(pkt.len());

:)

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 84.32%. Comparing base (3acf37d) to head (2983c30).
Report is 4 commits behind head on main.

❗ Current head 2983c30 differs from pull request most recent head 0de8b13

Please upload reports for the commit 0de8b13 to get more accurate results.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #4788   +/-   ##
=======================================
  Coverage   84.32%   84.32%           
=======================================
  Files         249      249           
  Lines       27512    27513    +1     
=======================================
+ Hits        23200    23201    +1     
  Misses       4312     4312           

Flag	Coverage Δ
5.10-c5n.metal	84.55% <100.00%> (+<0.01%)	⬆️
5.10-m5n.metal	84.53% <100.00%> (-0.01%)	⬇️
5.10-m6a.metal	83.82% <100.00%> (+<0.01%)	⬆️
5.10-m6g.metal	80.90% <100.00%> (+<0.01%)	⬆️
5.10-m6i.metal	84.52% <100.00%> (-0.01%)	⬇️
5.10-m7g.metal	80.90% <100.00%> (+<0.01%)	⬆️
6.1-c5n.metal	84.55% <100.00%> (+<0.01%)	⬆️
6.1-m5n.metal	84.53% <100.00%> (+<0.01%)	⬆️
6.1-m6a.metal	83.83% <100.00%> (+<0.01%)	⬆️
6.1-m6g.metal	80.90% <100.00%> (+<0.01%)	⬆️
6.1-m6i.metal	84.52% <100.00%> (-0.02%)	⬇️
6.1-m7g.metal	80.90% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[RiverPhillips]

Sorry, I'll try and fix this, looks like I forgot the DCO and the rebase changed a bunch more commits than mine

[roypat]

Thanks! Sorry for the late response, I was out most of last week. I've gone ahead and squashed your commits, everything else looks good to me!