updates/023_2020mar26_decoder_emulator_started.mdwn

   1 So many things happened since the last update they actually need to go
   2 in the main update, even in summary form.  One big thing: Raptor Engineering
   3 sponsored us with remote access to a TALOS II Workstation!
   4
   5 # Introduction
   6
   7 Here's the summary (if it can be called a summary):
   8
   9 * [An announcement](http://lists.libre-riscv.org/pipermail/libre-riscv-dev/2020-March/004995.html)
  10   that we got the funding (which is open to anyone - hint, hint) resulted in
  11   at least three people reaching out to join the team.  "We don't need
  12   permission to own our own hardware" got a *really* positive reaction.
  13 * New team member, Jock (hello Jock!) starts on the coriolis2 layout,
  14   after Jean-Paul from LIP6.fr helped to dramatically improve how coriolis2
  15   can be used.  This resulted in a
  16   [tutorial](https://libre-riscv.org/HDL_workflow/coriolis2/) and a
  17   [huge bug report discussion](http://bugs.libre-riscv.org/show_bug.cgi?id=178)
  18 * Work has started on the
  19   [POWER ISA decoder](http://bugs.libre-riscv.org/show_bug.cgi?id=186),
  20   verified through
  21   [calling GNU AS](https://git.libre-riscv.org/?p=soc.git;a=blob;f=src/soc/decoder/test/test_decoder_gas.py;h=9238d3878d964907c5569a3468d6895effb7dc02;hb=56d145e42ac75626423915af22d1493f1e7bb143) (yes, really!)
  22   and on a mini-simulator
  23   [calling QEMU](https://git.libre-riscv.org/?p=soc.git;a=blob;f=src/soc/simulator/qemu.py;h=9eb103bae227e00a2a1d2ec4f43d7e39e4f44960;hb=56d145e42ac75626423915af22d1493f1e7bb143)
  24   for verification.
  25 * Jacob's simple-soft-float library growing
  26   [Power FP compatibility](http://bugs.libre-riscv.org/show_bug.cgi?id=258)
  27   and python bindings.
  28 * A Conference call with OpenPOWER Foundation Director, Hugh, and Timothy
  29   Pearson from RaptorCS has been established every two weeks.
  30 * The OpenPOWER Foundation is also running some open
  31   ["Virtual Coffee"](https://openpowerfoundation.org/openpower-virtual-coffee-calls/)
  32   weekly round-table calls for anyone interested, generally, in OpenPOWER
  33   development.
  34 * Tim sponsors our team with access to a Monster Talos II system with a
  35   whopping 128 GB RAM.  htop lists a staggering 72 cores (18 real
  36   with 4-way hyperthreading).
  37 * [Epic MegaGrants](http://lists.libre-riscv.org/pipermail/libre-riscv-dev/2020-March/005262.html)
  38   reached out (hello!) to say they're still considering our
  39   request.
  40 * A marathon 3-hour session with [NLNet](http://nlnet.nl) resulted
  41   in the completion of the
  42   [Milestone tasks list(s)](http://bugs.libre-riscv.org/buglist.cgi?component=Milestones&list_id=567&resolution=---)
  43   and a
  44   [boat-load](http://lists.libre-riscv.org/pipermail/libre-riscv-dev/2020-March/thread.html)
  45   of bug reports to the list.
  46 * Immanuel Yehowshua is participating in the Georgia Tech
  47   [Create-X](https://create-x.gatech.edu/) Programme, and is establishing
  48   a Public Benefit Corporation in Atlanta, as an ethical vehicle for VC
  49   Funding.
  50 * A [Load/Store Buffer](http://bugs.libre-riscv.org/show_bug.cgi?id=216)
  51   design and
  52   [further discussion](http://bugs.libre-riscv.org/show_bug.cgi?id=257)
  53   including on
  54   [comp.arch](https://groups.google.com/forum/#!topic/comp.arch/cbGAlcCjiZE)
  55   inspired additional writeup
  56   on the
  57   [6600 scoreboard](https://libre-riscv.org/3d_gpu/architecture/6600scoreboard/)
  58   page.
  59
  60 Well dang, as you can see, suddenly it just went ballistic.  There's
  61 almost certainly things left off the list.  For such a small team there's
  62 a heck of a lot going on.  We have an awful lot to do, in a short amount
  63 of time: the 180nm tape-out is in October 2020 - only 7 months away.
  64
  65 # NLNet Funding announcement
  66
  67 TODO
  68
  69 # Coriolis2 experimentation started
  70
  71 TODO
  72
  73 # POWER ISA decoder and Simulator
  74
  75 TODO
  76
  77 # simple-soft-float Library and POWER FP emulation
  78
  79 The [simple-soft-float](https://salsa.debian.org/Kazan-team/simple-soft-float) library is a floating-point library Jacob wrote with the intention
  80 of being a reference implementation of IEEE 754 for hardware testing purposes. It's
  81 specifically designed to be written to be easier to understand instead of having the
  82 code obscured in pursuit of speed:
  83
  84 * Being easier to understand helps prevent bugs where the code does not
  85   match the IEEE spec.
  86 * It uses the [algebraics](https://salsa.debian.org/Kazan-team/algebraics) library that Jacob wrote since that allows using numbers
  87   that behave like exact real numbers, making reasoning about the code
  88   simpler.
  89 * It is written in Rust rather than highly-macro-ified C, since that helps with
  90   readability since operations aren't obscured, as well as safety, since Rust
  91   proves at compile time that the code won't seg-fault unless you specifically
  92   opt-out of those guarantees by using `unsafe`.
  93
  94 It currently supports 16, 32, 64, 128-bit FP for RISC-V, along with having a
  95 `DynamicFloat` type which allows dynamically specifying all aspects of how a
  96 particular floating-point type behaves -- if one wanted, they could configure it as a
  97 2048-bit floating-point type.
  98
  99 It also has Python bindings, thanks to the awesome [PyO3](https://pyo3.rs/) library for writing Python
 100 bindings in Rust.
 101
 102 We decided to write simple-soft-float instead of extending the industry-standard
 103 [Berkeley softfloat](http://www.jhauser.us/arithmetic/SoftFloat.html) library because of a range of issues, including not supporting
 104 Power FP, requiring recompilation to switch which ISA is being emulated, not
 105 supporting all the required operations, architectural issues such as depending on
 106 global variables, etc. We are still testing simple-soft-float against Berkeley softfloat
 107 where we can, however, since Berkeley softfloat is widely used and highly likely to
 108 be correct.
 109
 110 simple-soft-float is [gaining support for Power FP](http://bugs.libre-riscv.org/show_bug.cgi?id=258), which requires rewriting a lot of
 111 the status-flag handling code since Power supports a much larger set of
 112 floating-point status flags and exceptions than most other ISAs.
 113
 114 Thanks to RaptorCS for giving us remote access to a Power9 system, since that
 115 makes it much easier verifying that the test cases are correct.
 116
 117 API Docs for stable releases of both [simple-soft-float](https://docs.rs/simple-soft-float) and [algebraics](https://docs.rs/algebraics) are available
 118 on docs.rs.
 119
 120 # OpenPOWER Conference calls
 121
 122 TODO
 123
 124 # OpenPower Virtual Coffee Meetings
 125
 126 TODO
 127
 128 # Sponsorship by RaptorCS with a TALOS II Workstation
 129
 130 TODO
 131
 132 # Epic Megagrants
 133
 134 TODO
 135
 136 # NLNet Milestone tasks
 137
 138 TODO
 139
 140 # Georgia Tech CREATE-X
 141
 142 TODO
 143
 144 # LOAD/STORE Buffer and 6600 design documentation
 145
 146 TODO
 147
 148 # Conclusion
 149
 150 TODO
 151