we don't have a curve25519 accelerator on the vexii-test device
due to a lack of space to fit it. so for vexii-test targets
we swap in the u32 (vs u32e) backend.
This requires re-writing the config.toml file based on the
presence (or absence) of a vexii-test flag.
this is a minimal target for building test images that need
none of the xous goodies, but may rely upon e.g. APIs and
libraries. Closer to coding for a baremetal embedded system.
this allows us to build a simulation-focused xous image
this was previously called cramium-fpga but that precise
configuration is less accurate as the simulation environment
makes more shortcuts (and can include more peripherals)
than the fpga environment.
we don't have pio block on nto; let's avoid maintaining two
versions of the code base for a non-essential feature (the
downside of not having this is that the system could end
up locking up if no processes call yield_slice(), but in
practice this happens often enough that we can get away with
it -- at least at this stage).
the `cramium-soc` target is now the simulation target (just
the bare bones chip in a simulation harness, using the `dabao`
board configuration which is basically just a breakout board).
baosec continues to be the prototype for the first USB dongle
coming out.
right now our patch set assumes a single HW configuration for
cryptography accelerators.
This isn't going to be true going forward. A more scalable approach
to patching accelerators is going to be needed.
For now, we're just falling back to software acceleration for
the cramium target.
There's a hardware bug that makes concurrent camera and SPIM
traffic unreliable. The work-around is to place the camera
code inside the XIP area so we don't have to hit SPIM to swap
in code while the camera is running.
However, we do need to figure out a more reliable fix for this.
this is a slimmed down target designed for a smaller memory
footprint. there is some loss of process isolation security but
also a boost in performance due to more items sharing a memory
space.
at the moment an SVD pointer hangs out that messes builds up
when changing targets (sometimes) -- this is the pointer that
includes extra files in the build for targets that have disjoint
RTL sources. add a flag for cleaning this up.
We had a "coincidence" before that allowed things to work,
where we had some data line up that we happened to think
was some other header data, so we "fixed" it by
subtracting the offset of the header data.
Actually, it wasn't.
Clarify that decrypt_src_page_at() offsets are relative
to the beginning of encrypted data, and not relative
to the absolute beginning of the disk.
Image creation is now fixed so that stuff lines up to what
this routine expects, and some asserts added to try and
save ourselves in the future as we try to fix other bugs
we discover over time.
loaders are hard.
- swap works on hardware (at least reading in the encrypted
images works)
- renode version is deliberately configured to trigger an edge
case in the swapper, and is broken -- to be debugged
Haven't dug into it, but the AES instruction is not working
correctly in MPW cramium target silicon. There isn't a great way
to selectively disable hardware acceleration for sub-targets of
patched crates. So, as a work-around, we use xtask to patch
the default arguments of the AES crate, just for cramium-soc
targets, to use software emulated AES. This allows us to move
on with bring-up. To wit, there is another AES accelerator
on-chip that could be much faster than RV32 AES acceleration
(and with better DPA characteristics), and to use that option,
we might have to use a patch like this anyways, because Rust
doesn't have a good way to pass feature flags through patched
crates like this.