1 # Resources and Specifications
3 This page aims to collect all the resources and specifications we need
4 in one place for quick access. We will try our best to keep links here
5 up-to-date. Feel free to add more links here.
11 This section is primarily a series of useful links found online
13 * [FSiC2019](https://wiki.f-si.org/index.php/FSiC2019)
14 * Fundamentals to learn to get started [[3d_gpu/tutorial]]
16 ## Is Open Source Hardware Profitable?
17 [RaptorCS on FOSS Hardware Interview](https://www.youtube.com/watch?v=o5Ihqg72T3c&feature=youtu.be)
21 * [3.0 PDF](https://openpowerfoundation.org/?resource_lib=power-isa-version-3-0)
22 * [2.07 PDF](https://openpowerfoundation.org/?resource_lib=ibm-power-isa-version-2-07-b)
24 ## Overview of the user ISA:
26 [Raymond Chen's PowerPC series](https://devblogs.microsoft.com/oldnewthing/20180806-00/?p=99425)
28 # RISC-V Instruction Set Architecture
30 **PLEASE UPDATE** - we are no longer implementing full RISCV, only user-space
33 The Libre RISC-V Project is building a hybrid CPU/GPU SoC. As the name
34 of the project implies, we will be following the RISC-V ISA I due to it
35 being open-source and also because of the huge software and hardware
36 ecosystem building around it. There are other open-source ISAs but none
37 of them have the same momentum and energy behind it as RISC-V.
39 To fully take advantage of the RISC-V ecosystem, it is important to be
40 compliant with the RISC-V standards. Doing so will allow us to to reuse
41 most software as-is and avoid major forks.
43 * [Official compiled PDFs of RISC-V ISA Manual]
44 (https://github.com/riscv/riscv-isa-manual/releases/latest)
45 * [Working draft of the proposed RISC-V Bitmanipulation extension](https://github.com/riscv/riscv-bitmanip/blob/master/bitmanip-draft.pdf)
46 * [RISC-V "V" Vector Extension](https://riscv.github.io/documents/riscv-v-spec/)
47 * [RISC-V Supervisor Binary Interface Specification](https://github.com/riscv/riscv-sbi-doc/blob/master/riscv-sbi.md)
49 Note: As far as I know, we aren't using the RISC-V V Extension directly
50 at the moment. However, there are many wiki pages that make a reference
51 to the V extension so it would be good to include it here as a reference
52 for comparative/informative purposes with regard to Simple-V.
55 - [Qemu emulation](https://github.com/qemu/qemu/commit/d5fee0bbe68d5e61e2d2beb5ff6de0b9c1cfd182)
58 # RTL Arithmetic SQRT, FPU etc.
61 * [Fast Floating Point Square Root](https://pdfs.semanticscholar.org/5060/4e9aff0e37089c4ab9a376c3f35761ffe28b.pdf)
62 * [Reciprocal Square Root Algorithm](http://www.acsel-lab.com/arithmetic/arith15/papers/ARITH15_Takagi.pdf)
64 ## CORDIC and related algorithms
66 * <https://bugs.libre-soc.org/show_bug.cgi?id=127> research into CORDIC
67 * <https://bugs.libre-soc.org/show_bug.cgi?id=208>
68 * [BKM (log(x) and e^x)](https://en.wikipedia.org/wiki/BKM_algorithm)
69 * [CORDIC](http://www.andraka.com/files/crdcsrvy.pdf)
70 - Does not have an easy way of computing tan(x)
71 * [zipcpu CORDIC](https://zipcpu.com/dsp/2017/08/30/cordic.html)
72 * [Low latency and Low error floating point TCORDIC](https://ieeexplore.ieee.org/document/7784797) (email Michael or Cole if you don't have IEEE access)
73 * <http://www.myhdl.org/docs/examples/sinecomp/> MyHDL version of CORDIC
75 ## IEEE Standard for Floating-Point Arithmetic (IEEE 754)
77 Almost all modern computers follow the IEEE Floating-Point Standard. Of
78 course, we will follow it as well for interoperability.
80 * IEEE 754-2019: <https://standards.ieee.org/standard/754-2019.html>
82 Note: Even though this is such an important standard used by everyone,
83 it is unfortunately not freely available and requires a payment to
84 access. However, each of the Libre RISC-V members already have access
87 * [Lecture notes - Floating Point Appreciation](http://pages.cs.wisc.edu/~markhill/cs354/Fall2008/notes/flpt.apprec.html)
89 Among other things, has a nice explanation on arithmetic, rounding modes and the sticky bit.
91 * [What Every Computer Scientist Should Know About Floating-Point Arithmetic](https://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html)
93 Nice resource on rounding errors (ulps and epsilon) and the "table maker's dilemma".
95 ## Past FPU Mistakes to learn from
97 * [Intel Underestimates Error Bounds by 1.3 quintillion on
98 Random ASCII – tech blog of Bruce Dawson ](https://randomascii.wordpress.com/2014/10/09/intel-underestimates-error-bounds-by-1-3-quintillion/)
99 * [Intel overstates FPU accuracy 06/01/2013](http://notabs.org/fpuaccuracy)
103 The Khronos Group creates open standards for authoring and acceleration
104 of graphics, media, and computation. It is a requirement for our hybrid
105 CPU/GPU to be compliant with these standards *as well* as with IEEE754,
106 in order to be commercially-competitive in both areas: especially Vulkan
107 and OpenCL being the most important. SPIR-V is also important for the
110 Thus the [[zfpacc_proposal]] has been created which permits runtime dynamic
111 switching between different accuracy levels, in userspace applications.
113 [**SPIR-V Main Page Link**](https://www.khronos.org/registry/spir-v/)
115 * [SPIR-V 1.5 Specification Revision 1](https://www.khronos.org/registry/spir-v/specs/unified1/SPIRV.html)
116 * [SPIR-V OpenCL Extended Instruction Set](https://www.khronos.org/registry/spir-v/specs/unified1/OpenCL.ExtendedInstructionSet.100.html)
117 * [SPIR-V GLSL Extended Instruction Set](https://www.khronos.org/registry/spir-v/specs/unified1/GLSL.std.450.html)
119 [**Vulkan Main Page Link**](https://www.khronos.org/registry/vulkan/)
121 * [Vulkan 1.1.122](https://www.khronos.org/registry/vulkan/specs/1.1-extensions/html/index.html)
123 [**OpenCL Main Page**](https://www.khronos.org/registry/OpenCL/)
125 * [OpenCL 2.2 API Specification](https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_API.html)
126 * [OpenCL 2.2 Extension Specification](https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_Ext.html)
127 * [OpenCL 2.2 SPIR-V Environment Specification](https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_Env.html)
129 * OpenCL released the proposed OpenCL 3.0 spec for comments in april 2020
131 * [Announcement video](https://youtu.be/h0_syTg6TtY)
132 * [Announcement video slides (PDF)](https://www.khronos.org/assets/uploads/apis/OpenCL-3.0-Launch-Apr20.pdf)
134 Note: We are implementing hardware accelerated Vulkan and
135 OpenCL while relying on other software projects to translate APIs to
136 Vulkan. E.g. Zink allows for OpenGL-to-Vulkan in software.
138 # Graphics and Compute API Stack
140 I found this informative post that mentions Kazan and a whole bunch of
141 other stuff. It looks like *many* APIs can be emulated on top of Vulkan,
142 although performance is not evaluated.
144 <https://synappsis.wordpress.com/2017/06/03/opengl-over-vulkan-dev/>
146 * Pixilica is heading up an initiative to create a RISC-V graphical ISA
148 * [Pixilica 3D Graphical ISA Slides](https://b5792ddd-543e-4dd4-9b97-fe259caf375d.filesusr.com/ugd/841f2a_c8685ced353b4c3ea20dbb993c4d4d18.pdf)
151 # Various POWER Communities
152 - [An effort to make a 100% Libre POWER Laptop](https://www.powerpc-notebook.org/en/)
153 The T2080 is a POWER8 chip.
154 - [Power Progress Community](https://www.powerprogress.org/campaigns/donations-to-all-the-power-progress-community-projects/)
155 Supporting/Raising awareness of various POWER related open projects on the FOSS
157 - [OpenPOWER](https://openpowerfoundation.org)
158 Promotes and ensure compliance with the Power ISA amongst members.
159 - [OpenCapi](https://opencapi.org)
160 High performance interconnect for POWER machines. One of the big advantages
161 of the POWER architecture. Notably more performant than PCIE Gen4, and is
162 designed to be layered on top of the physical PCIE link.
163 - [OpenPOWER “Virtual Coffee” Calls](https://openpowerfoundation.org/openpower-virtual-coffee-calls/)
164 Truly open bi-weekly teleconference lines for anybody interested in helping
165 advance or adopting the POWER architecture.
169 ## Free Silicon Conference
171 The conference brought together experts and enthusiasts who want to build
172 a complete Free and Open Source CAD ecosystem for designing analog and
173 digital integrated circuits. The conference covered the full spectrum of
174 the design process, from system architecture, to layout and verification.
176 * <https://wiki.f-si.org/index.php/FSiC2019#Foundries.2C_PDKs_and_cell_libraries>
178 * LIP6's Coriolis - a set of backend design tools:
179 <https://www-soc.lip6.fr/equipe-cian/logiciels/coriolis/>
181 Note: The rest of LIP6's website is in French, but there is a UK flag
182 in the corner that gives the English version.
184 * KLayout - Layout viewer and editor: <https://www.klayout.de/>
186 # The OpenROAD Project
188 OpenROAD seeks to develop and foster an autonomous, 24-hour, open-source
189 layout generation flow (RTL-to-GDS).
191 * <https://theopenroadproject.org/>
193 # Other RISC-V GPU attempts
195 * <https://fossi-foundation.org/2019/09/03/gsoc-64b-pointers-in-rv32>
197 * <http://bjump.org/manycore/>
199 * <https://resharma.github.io/RISCV32-GPU/>
201 TODO: Get in touch and discuss collaboration
203 # Tests, Benchmarks, Conformance, Compliance, Verification, etc.
207 RISC-V Foundation is in the process of creating an official conformance
208 test. It's still in development as far as I can tell.
210 * //TODO LINK TO RISC-V CONFORMANCE TEST
212 ## IEEE 754 Testing/Emulation
214 IEEE 754 has no official tests for floating-point but there are
215 well-known third party tools to check such as John Hauser's TestFloat.
217 There is also his SoftFloat library, which is a software emulation
218 library for IEEE 754.
220 * <http://www.jhauser.us/arithmetic/>
222 Jacob is also working on an IEEE 754 software emulation library written
223 in Rust which also has Python bindings:
225 * Source: <https://salsa.debian.org/Kazan-team/simple-soft-float>
226 * Crate: <https://crates.io/crates/simple-soft-float>
227 * Autogenerated Docs: <https://docs.rs/simple-soft-float/>
229 A cool paper I came across in my research is "IeeeCC754++ : An Advanced
230 Set of Tools to Check IEEE 754-2008 Conformity" by Dr. Matthias Hüsken.
232 * Direct link to PDF:
233 <http://elpub.bib.uni-wuppertal.de/servlets/DerivateServlet/Derivate-7505/dc1735.pdf>
237 OpenCL Conformance Tests
239 * <https://github.com/KhronosGroup/OpenCL-CTS>
241 Vulkan Conformance Tests
243 * <https://github.com/KhronosGroup/VK-GL-CTS>
245 MAJOR NOTE: We are **not** allowed to say we are compliant with any of
246 the Khronos standards until we actually make an official submission,
247 do the paperwork, and pay the relevant fees.
249 ## Formal Verification
251 Formal verification of Libre RISC-V ensures that it is bug-free in
252 regards to what we specify. Of course, it is important to do the formal
253 verification as a final step in the development process before we produce
254 thousands or millions of silicon.
256 Some learning resources I found in the community:
258 * ZipCPU: <http://zipcpu.com/> ZipCPU provides a comprehensive
259 tutorial for beginners and many exercises/quizzes/slides:
260 <http://zipcpu.com/tutorial/>
261 * Western Digital's SweRV CPU blog (I recommend looking at all their
262 posts): <https://tomverbeure.github.io/>
263 * <https://tomverbeure.github.io/risc-v/2018/11/19/A-Bug-Free-RISC-V-Core-without-Simulation.html>
264 * <https://tomverbeure.github.io/rtl/2019/01/04/Under-the-Hood-of-Formal-Verification.html>
268 * <https://www.ohwr.org/project/wishbone-gen>
272 ## Adding new instructions:
274 * <https://archive.fosdem.org/2015/schedule/event/llvm_internal_asm/>
278 * <https://danluu.com/branch-prediction/>
282 * [Migen - a Python RTL](https://jeffrey.co.in/blog/2014/01/d-flip-flop-using-migen/)
283 * [LiTeX](https://github.com/timvideos/litex-buildenv/wiki/LiteX-for-Hardware-Engineers)
284 An SOC builder written in Python Migen DSL. Allows you to generate functional
285 RTL for a SOC configured with cache, a RISCV core, ethernet, DRAM support,
286 and parameterizeable CSRs.
287 * [Migen Tutorial](http://blog.lambdaconcept.com/doku.php?id=migen:tutorial>)
288 * There is a great guy, Robert Baruch, who has a good
289 [tutorial](https://github.com/RobertBaruch/nmigen-tutorial) on nMigen.
290 He also build an FPGA-proven Motorola 6800 CPU clone with nMigen and put
291 [the code](https://github.com/RobertBaruch/n6800) and
292 [instructional videos](https://www.youtube.com/playlist?list=PLEeZWGE3PwbbjxV7_XnPSR7ouLR2zjktw)
294 * [Minerva](https://github.com/lambdaconcept/minerva)
295 An SOC written in Python nMigen DSL
296 * [Using our Python Unit Tests(old)](http://lists.libre-riscv.org/pipermail/libre-riscv-dev/2019-March/000705.html)
297 * <https://chisel.eecs.berkeley.edu/api/latest/chisel3/util/DecoupledIO.html>
298 * <http://www.clifford.at/papers/2016/yosys-synth-formal/slides.pdf>
302 * <https://wiki.f-si.org/index.php/FSiC2019>
303 * <https://fusesoc.net>
304 * <https://www.lowrisc.org/open-silicon/>
305 * <http://fpgacpu.ca/fpga/Pipeline_Skid_Buffer.html> pipeline skid buffer
306 * <https://pyvcd.readthedocs.io/en/latest/vcd.gtkw.html> GTKwave
307 * <http://www.sunburst-design.com/papers/CummingsSNUG2002SJ_Resets.pdf>
308 Synchronous Resets? Asynchronous Resets? I am so confused! How will I
309 ever know which to use? by Clifford E. Cummings
310 * <http://www.sunburst-design.com/papers/CummingsSNUG2008Boston_CDC.pdf>
311 Clock Domain Crossing (CDC) Design & Verification Techniques Using
312 SystemVerilog, by Clifford E. Cummings
313 In particular, see section 5.8.2: Multi-bit CDC signal passing using
314 1-deep / 2-register FIFO synchronizer.
315 * <http://www2.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-143.pdf>
316 Understanding Latency Hiding on GPUs, by Vasily Volkov
319 # Real/Physical Projects
321 * [Samuel's KC5 code](http://chiselapp.com/user/kc5tja/repository/kestrel-3/dir?ci=6c559135a301f321&name=cores/cpu)
322 * <https://chips4makers.io/blog/>
323 * <https://hackaday.io/project/7817-zynqberry>
324 * <https://github.com/efabless/raven-picorv32>
325 * <https://efabless.com>
326 * <https://efabless.com/design_catalog/default>
327 * <https://wiki.f-si.org/index.php/The_Raven_chip:_First-time_silicon_success_with_qflow_and_efabless>
328 * <https://mshahrad.github.io/openpiton-asplos16.html>
330 # ASIC tape-out pricing
332 * <https://europractice-ic.com/wp-content/uploads/2020/05/General-MPW-EUROPRACTICE-200505-v8.pdf>
336 * <https://toyota-ai.ventures/>
337 * [NLNet Applications](http://bugs.libre-riscv.org/buglist.cgi?columnlist=assigned_to%2Cbug_status%2Cresolution%2Cshort_desc%2Ccf_budget&f1=cf_nlnet_milestone&o1=equals&query_format=advanced&resolution=---&v1=NLnet.2019.02)
339 # Good Programming/Design Practices
341 * [Liskov Substitution Principle](https://en.wikipedia.org/wiki/Liskov_substitution_principle)
342 * [Principle of Least Astonishment](https://en.wikipedia.org/wiki/Principle_of_least_astonishment)
343 * <https://peertube.f-si.org/videos/watch/379ef007-40b7-4a51-ba1a-0db4f48e8b16>
344 * [Rust-Lang Philosophy and Consensus](http://smallcultfollowing.com/babysteps/blog/2019/04/19/aic-adventures-in-consensus/)
346 * <https://youtu.be/o5Ihqg72T3c>
347 * <http://flopoco.gforge.inria.fr/>
348 * Fundamentals of Modern VLSI Devices
349 <https://groups.google.com/a/groups.riscv.org/d/msg/hw-dev/b4pPvlzBzu0/7hDfxArEAgAJ>
353 * <https://www.crnhq.org/cr-kit/>
357 * <https://github.com/Isotel/mixedsim>
358 * <http://www.vlsiacademy.org/open-source-cad-tools.html>
359 * <http://ngspice.sourceforge.net/adms.html>
360 * <https://en.wikipedia.org/wiki/Verilog-AMS#Open_Source_Implementations>
362 # Libre-SOC Standards
364 This list auto-generated from a page tag "standards":
366 [[!inline pages="tagged(standards)" actions="no" archive="yes" quick="yes"]]
370 * [[resources/server-setup/web-server]]
371 * [[resources/server-setup/git-mirroring]]
372 * [[resources/server-setup/nagios-monitoring]]
376 * <https://www.fed4fire.eu/testbeds/>
378 # Really Useful Stuff
380 * <https://github.com/im-tomu/fomu-workshop/blob/master/docs/requirements.txt>
381 * <https://github.com/im-tomu/fomu-workshop/blob/master/docs/conf.py#L39-L47>