nlnet_2021_3mdeb_cavatools.mdwn

   1 # NLnet User-operated Grant Request for 3mdeb Power ISA Simulator
   2
   3 * 2021-08-071
   4 * <https://bugs.libre-soc.org/show_bug.cgi?id=939>
   5
   6 ## Project name
   7
   8 Libre-SOC 3mdeb Cavatools: Power ISA Simulator
   9
  10 ## Website / wiki
  11
  12 <https://libre-soc.org/nlnet_2021_3mdeb_cavatools>
  13
  14 Please be short and to the point in your answers; focus primarily on
  15 the what and how, not so much on the why. Add longer descriptions as
  16 attachments (see below). If English isn't your first language, don't
  17 worry - our reviewers don't care about spelling errors, only about
  18 great ideas. We apologise for the inconvenience of having to submit in
  19 English. On the up side, you can be as technical as you need to be (but
  20 you don't have to). Do stay concrete. Use plain text in your reply only,
  21 if you need any HTML to make your point please include this as attachment.
  22
  23 ## Abstract: Can you explain the whole project and its expected outcome(s).
  24
  25 Cavatools is currently a high performance user-operated simulator of
  26 the RISC-V ISA. The primary objective of the project is to extend it to
  27 implement the scalar Power ISA and the Libre-SOC Draft SVP64
  28 Extensions. This will allow rapid prototyping of Extensions to the
  29 Power ISA long before they reach silicon (which is very costly).
  30 In turn this helps Libre-SOC to deliver on its commitment to provide
  31 user-trustable processors for use in Internet routers, desktop,
  32 smartphone and other user-operated devices where security and transparency
  33 is expected.
  34
  35 # Have you been involved with projects or organisations relevant to this project before? And if so, can you tell us a bit about your contributions?
  36
  37 3mdeb is currently helping Libre-SOC with the (horribly slow,
  38 easy-to-read, easy-to-use) Libre-SOC Power ISA Simulator which is 20,000 times
  39 slower than cavatools. 3mdeb is also helping with ISA level unit tests in
  40 Libre-SOC's code base that will be used to cross-validate a huge range of
  41 Power ISA simulators and actual silicon implementations.
  42
  43 # Requested Amount
  44
  45 EUR $50,000.
  46
  47 # Explain what the requested budget will be used for?
  48
  49 * To create a compiler which takes Libre-SOC Machine-readable
  50   Power ISA specification files and generate c code
  51 * To extend cavatools to include support for the Scalar
  52   parts of the Power ISA
  53 * To then add support for Libre-SOC's Draft SVP64 Extensions
  54 * To enhance it to include gdb "remote" machine interface
  55   support
  56 * To add Power ISA RADIX MMU emulation
  57 * To extend cavatools to run a very basic linux
  58   initramfs in-memory with basic serial console access
  59 * To demonstrate running first a single core linux kernel
  60   and later a SMP one, with busybox
  61 * To use the exact same Specification c compiler to create
  62   an "illegal instruction trap" emulator, integrated
  63   into the linux kernel, for emulating Power ISA SIMD instructions
  64   (extending the existing trap-and-emulate code already present
  65    in ppc64 linux kernel source code)
  66
  67 # Does the project have other funding sources, both past and present?
  68
  69 Although there is NLnet funding for the Libre-SOC Simulator
  70 (written in python) and associayed unit tests, cavatools, which is
  71 written in c by Peter Hsu, does not have funding for the Power ISA
  72 aditions. cavatools itself is a very new project.
  73
  74 # Compare your own project with existing or historical efforts.
  75
  76 Although there are quite a few Power ISA simulators, none of them
  77 are up-to-date or are suited to high performance, like cavatools.
  78 cavatools is multi-process and extremely fast, using relatively little
  79 memory, where power-gem5, which has a different focus and has huge flexibility
  80 and usefulness for research, uses vast amounts
  81 of memory and is much slower. cavatools also has hardware-level cycle-accurate emulation which is extremely useful and important for analysing experimental
  82 instructions, which is a feature that no other Power ISA Simulator has.
  83 DolphinPC and pearpc are over 15 years old and were targetted at 32 bit
  84 emulation of much older Power ISA processors. Libre-SOC's python-based
  85 simulator only achieves aroubd 2,000 instructions per second on
  86 high performance hardware whereas
  87 cavatools achieves 200,000 instructions per second per processor
  88 on modest hardware.
  89 IBM's own Power ISA simulator is proprietary and, because it contains
  90 confidential experimentation internal to IBM, may not be made public.
  91
  92 ## What are significant technical challenges you expect to solve during the project, if any?
  93
  94 This is at its heart a compiler project, which can be a challenging
  95 area. However the language being implemented is quite small and limited,
  96 so the project is relatively straightforward.
  97
  98 ## Describe the ecosystem of the project, and how you will engage with relevant actors and promote the outcomes?
  99
 100 The project will be developed entirely publicly and transparently,
 101 using Libre-SOC Project Resources which are already set up for trustable
 102 auditability and transparency. The mailing lists therefore are always
 103 publicly available.
 104
 105 Online conferences and talks will be given as progress
 106 is made, as well as working with Libre-SOC to send out development
 107 reports and progress.
 108
 109 # Extra info to be submitted