If the first one is killed for some unrelated reason (e.g. write port
hazard), the second one will still issue to the div-sqrt unit. While
it will itself later be killed, the fact that the later instruction
acquires a resource needed by the former instruction leads to deadlock.
This reverts commit b64b87ad07.
The crossings already have buffering in those places where it was
appropriate. Adding more does not help flow through paths.
* coreplex collapse: peripherals now in coreplex
* coreplex: better factoring of TLBusWrapper attachement points
* diplomacy: allow monitorless :*= and :=*
* rocket: don't connect monitors to tile tim slave ports
* rename chip package to system
* coreplex: only sbus has a splitter
* TLFragmenter: Continuing my spot battles on requires without explanatory strings
* pbus: toFixedWidthSingleBeatSlave
* tilelink: more verbose requires
* use the new system package for regression
* sbus: add more explicit FIFO attachment points
* delete leftover top-level utils
* cleanup ResetVector and RTC
I was examining a WB-stage control signal instead of a MEM-stage control
signal. I refactored the code to group the signals together, so that this
sort of bug is less likely going forward.
* interrupts: Less pessimistic synchronization for the different interrupt types. There are some issues with the interrupt number assignments.
* interrupts: Allow an option to NOT synchronize all the external interrupts coming into PLIC
* interrupts: ExampleRocketChipTop uses PeripheryAsyncExtInterrupts. Realized 'abstract' doesn't do what I thought in Scala.
* interrupts: use consistent async/periph/core ordering
* interrupts: Properly condition on 0 External interrupts
* interrupts: CLINT is also synchronous to periph clock
Single-precision values are stored in the regfile as double-precision,
so that FSD on a single-precision value stores a proper double and
FLD restores it as either a double or a single.
This is simpler, reduces what would have become a critical path in
the commit stage, and will make it easier to support the mbadinst
CSR if it is implemented.
We were zero-extending it, which is a double-precision zero in the recoded
format. So, when spilled and reloaded with fsd/fld, the original value
was destroyed. Instead, set the MSBs so that it represents sNaN. When
spilled, the single-precision number will be preserved as the NaN payload.
