- Instruction addresses are now checked in the IF stage, after the cache
and after the prefetch buffer.
- To deal with unaligned instructions, the PMP logic checks the current
address and the next in parallel.
- The spec_branch timing hack has been removed as it's no longer
relevant with the PMP logic moved.
- Various updates made to the icache testbench to account for the
changes.
- Relates to #1471
Signed-off-by: Tom Roberts <tomroberts@lowrisc.org>
This parameter forces a reset of all registers inside the core. This is
required to guarantee a common starting point for lockstep and thus
prevent spurious lockstep failure alerts.
Another minor change in this commit rearranges the writeback stage
multiplexing to gate incoming lsu write data when not valid. This stops
any X values from the data bus propagating to the register file
signalling (and thus to the lockstep comparison) which would cause the
lockstep alert to be X. It has the side effect of possibly reducing
power consumption in the register file.
Signed-off-by: Tom Roberts <tomroberts@lowrisc.org>
These changes correspond to similar changes in the prefetch buffer to
support branch prediction. A registered version of fill_ext_done was
required to prevent a combinational loop from branch_i in to valid_o
out.
Multiplexing priorities for fifo_addr have been swapped to match
fetch_addr_d in the same module and all similar multiplexing in the
icache (prioritize incoming branch_i over branch_mispredict_i). Note
however that it is not expected that these conditions will actually
occur together, and an assertion has been added to check that.
Signed-off-by: Tom Roberts <tomroberts@lowrisc.org>
- The prefetch buffer needs to know when space is available in the fetch
FIFO to accept a new external request.
- This change updates that logic to look at what is in the FIFO and what
is outstanding on the bus to decide when space is available rather
than always assuming the maximum number of requests are outstanding.
- This improves the usage efficiency of the FIFO and fixes#574
Signed-off-by: Tom Roberts <tomroberts@lowrisc.org>
- Drive a speculative version of the branch signal into the IF stage to
drive address muxing
- The speculative signal is the same as the regular branch signal but
assumes all conditional branches are taken
- This breaks the timing path from branch condition calculation into
address muxing (and therefore PMP error calculation)
- When the branch is not taken, any external request we might otherwise
have made is suppressed
- This has a minor performance cost (0.8% without I$, ~0% with I$)
Signed-off-by: Tom Roberts <tomroberts@lowrisc.org>
mtval should record which half of the instruction caused the error
rather than just recording the PC.
An extra signal is added in the IF stage to indicate when an error is
caused by the second half of an unaligned instruction. This signal is
then used to increment the PC by 2 for mtval capture on an error.
Fixes#709
- Fixes#288
- Add missing grant qualification to stop incorrect address updates
- Make RTL robust to spurious rvalid signalling
- Make sure a request is held until granted
- Remove incorrect assertion
- See issue #265 (partially fixes)
- Remove path from instr_rvalid_i to instr_req_o
- Prefetch buffer unit can now issue up to two outstanding requests
- Structure moved from state machine to request queue
- Change fetch fifo to use an unaligned flag rather than updating
the address each time
- Instantiate generic PMP module
- Wire up I-side and D-side PMP faults
- The output of the PMP check is used to gate external bus requests from the
I-side and LSU
- Each of those units progresses with their request as-if it was granted
externally and registers the PMP error
- The error is then sent to the controller at the appropriate time to trigger
an exception
We currently have a documentation block at the beginning of each file,
containing author credits and module-level documentation. The
module-level documentation is retained for historic reasons and
duplicated with the newer comments below it.
For the authors, maintaining author credits in the file is error-prone,
as this information gets outdated very soon. A more reliable way to see
who modified a file is to use the history information in git.
Additionally, we now have the CREDITS.md file, which lists all
contributors, even the ones which don't appear in the git history (e.g.
because the code was copied and commited by someone else).
- Add required signals to top-level
- Propagate error through fetch stages
- Add new exception type
- Update documentation for new exception type
- Resolves issue #109
The core handles unaligned instruction fetches by doing two separate
word-aligned instruction fetches. Without this commit, the core can
still output addresses which are not word aligned and relies on the
memory to ignore the LSBs of the address. This is not safe.
This commit simplifies the assignment of literals to enum types in
default cases by:
- defining or using existing enum values for all-zero values,
- feeding a single `1'bX` into the type cast instead of exact width
(the tools are fine with that).
This commit adds a `default` to all `unique case` statements. Also, in case
FSMs reach an undefined state, the `'X` is propagated to ease detection
in simulation. Both these changes are required by our coding guidelines.
The code base made extensive use of ASCII art headings/subheadings in
comments to delineate code. Switch to a more space efficient and easier
to edit format:
/////////
// Foo //
/////////
This change has been informed by advice from the lowRISC legal
committee.
The Solderpad 0.51 license states "the Licensor permits any Work
licensed under this License, at the option of the Licensee, to be
treated as licensed under the Apache License Version 2.0". We use this
freedom to convert license markings to Apache 2.0. This commit ensures
that we retain all authorship and copyright attribution information.
