add youtube video first boot on ECP5
[libreriscv.git] / resources.mdwn
index 248b960de6007391b31e9b9762062e9d973c12a4..95f2c319f145fd410754ba4c0e3728ebcdcf409a 100644 (file)
@@ -4,17 +4,40 @@ This page aims to collect all the resources and specifications we need
 in one place for quick access. We will try our best to keep links here
 up-to-date. Feel free to add more links here.
 
-# Libre-RISC-V Standards
+[[!toc  ]]
 
-This list auto-generated from a page tag "standards":
+# Getting Started
+
+This section is primarily a series of useful links found online
+
+* [FSiC2019](https://wiki.f-si.org/index.php/FSiC2019)
+* Fundamentals to learn to get started [[3d_gpu/tutorial]]
+
+## Is Open Source Hardware Profitable?
+[RaptorCS on FOSS Hardware Interview](https://www.youtube.com/watch?v=o5Ihqg72T3c&feature=youtu.be)
+
+# OpenPOWER ISA
+
+* [3.0 PDF](https://openpowerfoundation.org/?resource_lib=power-isa-version-3-0)
+* [2.07 PDF](https://openpowerfoundation.org/?resource_lib=ibm-power-isa-version-2-07-b)
+
+## Overview of the user ISA:
+
+[Raymond Chen's PowerPC series](https://devblogs.microsoft.com/oldnewthing/20180806-00/?p=99425)
+
+## OpenPOWER OpenFSI Spec (2016)
 
-[[!inline pages="tagged(standards)" actions="no" archive="yes" feedonly="yes" quick="yes"]]
+* [OpenPOWER OpenFSI Spec](http://openpowerfoundation.org/wp-content/uploads/resources/OpenFSI-spec-100/OpenFSI-spec-20161212.pdf)
 
+* [OpenPOWER OpenFSI Compliance Spec](http://openpowerfoundation.org/wp-content/uploads/resources/openpower-fsi-thts-1.0/openpower-fsi-thts-20180130.pdf)
 
 # RISC-V Instruction Set Architecture
 
+**PLEASE UPDATE** - we are no longer implementing full RISCV, only user-space
+RISCV
+
 The Libre RISC-V Project is building a hybrid CPU/GPU SoC. As the name
-of the project implies, we will be following the RISC-V ISA due to it
+of the project implies, we will be following the RISC-V ISA due to it
 being open-source and also because of the huge software and hardware
 ecosystem building around it. There are other open-source ISAs but none
 of them have the same momentum and energy behind it as RISC-V.
@@ -23,19 +46,44 @@ To fully take advantage of the RISC-V ecosystem, it is important to be
 compliant with the RISC-V standards. Doing so will allow us to to reuse
 most software as-is and avoid major forks.
 
-* Official compiled PDFs of RISC-V ISA Manual:
-  <https://github.com/riscv/riscv-isa-manual/releases/latest>
-* Working draft of the proposed RISC-V Bitmanipulation extension:
-  <https://github.com/riscv/riscv-bitmanip/blob/master/bitmanip-draft.pdf>
-* RISC-V "V" Vector Extension:
-  <https://riscv.github.io/documents/riscv-v-spec/>
+* [Official compiled PDFs of RISC-V ISA Manual]
+ (https://github.com/riscv/riscv-isa-manual/releases/latest)
+* [Working draft of the proposed RISC-V Bitmanipulation extension](https://github.com/riscv/riscv-bitmanip/blob/master/bitmanip-draft.pdf)
+* [RISC-V "V" Vector Extension](https://riscv.github.io/documents/riscv-v-spec/)
+* [RISC-V Supervisor Binary Interface Specification](https://github.com/riscv/riscv-sbi-doc/blob/master/riscv-sbi.md)
 
 Note: As far as I know, we aren't using the RISC-V V Extension directly
 at the moment. However, there are many wiki pages that make a reference
 to the V extension so it would be good to include it here as a reference
 for comparative/informative purposes with regard to Simple-V.
 
-# IEEE Standard for Floating-Point Arithmetic (IEEE 754)
+# Radix MMU
+ - [Qemu emulation](https://github.com/qemu/qemu/commit/d5fee0bbe68d5e61e2d2beb5ff6de0b9c1cfd182)
+
+# D-Cache
+
+## D-Cache Possible Optimizations papers and links
+- [ACDC: Small, Predictable and High-Performance Data Cache](https://dl.acm.org/doi/10.1145/2677093)
+
+
+# RTL Arithmetic SQRT, FPU etc.
+
+## Sqrt
+* [Fast Floating Point Square Root](https://pdfs.semanticscholar.org/5060/4e9aff0e37089c4ab9a376c3f35761ffe28b.pdf)
+* [Reciprocal Square Root Algorithm](http://www.acsel-lab.com/arithmetic/arith15/papers/ARITH15_Takagi.pdf)
+
+## CORDIC and related algorithms
+
+* <https://bugs.libre-soc.org/show_bug.cgi?id=127> research into CORDIC
+* <https://bugs.libre-soc.org/show_bug.cgi?id=208>
+* [BKM (log(x) and e^x)](https://en.wikipedia.org/wiki/BKM_algorithm)
+* [CORDIC](http://www.andraka.com/files/crdcsrvy.pdf)
+ - Does not have an easy way of computing tan(x)
+* [zipcpu CORDIC](https://zipcpu.com/dsp/2017/08/30/cordic.html)
+* [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)
+* <http://www.myhdl.org/docs/examples/sinecomp/> MyHDL version of CORDIC
+
+## IEEE Standard for Floating-Point Arithmetic (IEEE 754)
 
 Almost all modern computers follow the IEEE Floating-Point Standard. Of
 course, we will follow it as well for interoperability.
@@ -47,6 +95,20 @@ it is unfortunately not freely available and requires a payment to
 access. However, each of the Libre RISC-V members already have access
 to the document.
 
+* [Lecture notes - Floating Point Appreciation](http://pages.cs.wisc.edu/~markhill/cs354/Fall2008/notes/flpt.apprec.html)
+
+Among other things, has a nice explanation on arithmetic, rounding modes and the sticky bit.
+
+* [What Every Computer Scientist Should Know About Floating-Point Arithmetic](https://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html)
+
+Nice resource on rounding errors (ulps and epsilon) and the "table maker's dilemma".
+
+## Past FPU Mistakes to learn from
+
+* [Intel Underestimates Error Bounds by 1.3 quintillion on 
+Random ASCII – tech blog of Bruce Dawson ](https://randomascii.wordpress.com/2014/10/09/intel-underestimates-error-bounds-by-1-3-quintillion/)
+* [Intel overstates FPU accuracy 06/01/2013](http://notabs.org/fpuaccuracy)
+
 # Khronos Standards
 
 The Khronos Group creates open standards for authoring and acceleration
@@ -59,28 +121,26 @@ Kazan driver.
 Thus the [[zfpacc_proposal]] has been created which permits runtime dynamic
 switching between different accuracy levels, in userspace applications.
 
-**SPIR-V Main Page <https://www.khronos.org/registry/spir-v/>**
+[**SPIR-V Main Page Link**](https://www.khronos.org/registry/spir-v/)
+
+* [SPIR-V 1.5 Specification Revision 1](https://www.khronos.org/registry/spir-v/specs/unified1/SPIRV.html)
+* [SPIR-V OpenCL Extended Instruction Set](https://www.khronos.org/registry/spir-v/specs/unified1/OpenCL.ExtendedInstructionSet.100.html)
+* [SPIR-V GLSL Extended Instruction Set](https://www.khronos.org/registry/spir-v/specs/unified1/GLSL.std.450.html)
+
+[**Vulkan Main Page Link**](https://www.khronos.org/registry/vulkan/)
 
-* SPIR-V 1.5 Specification Revision 1:
-  <https://www.khronos.org/registry/spir-v/specs/unified1/SPIRV.html>
-* SPIR-V OpenCL Extended Instruction Set:
-  <https://www.khronos.org/registry/spir-v/specs/unified1/OpenCL.ExtendedInstructionSet.100.html>
-* SPIR-V GLSL Extended Instruction Set:
-  <https://www.khronos.org/registry/spir-v/specs/unified1/GLSL.std.450.html>
+* [Vulkan 1.1.122](https://www.khronos.org/registry/vulkan/specs/1.1-extensions/html/index.html)
 
-**Vulkan Main Page <https://www.khronos.org/registry/vulkan/>**
+[**OpenCL Main Page**](https://www.khronos.org/registry/OpenCL/)
 
-* Vulkan 1.1.122:
-  <https://www.khronos.org/registry/vulkan/specs/1.1-extensions/html/index.html>
+* [OpenCL 2.2 API Specification](https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_API.html)
+* [OpenCL 2.2 Extension Specification](https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_Ext.html)
+* [OpenCL 2.2 SPIR-V Environment Specification](https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_Env.html)
 
-**OpenCL Main Page <https://www.khronos.org/registry/OpenCL/>**
+* OpenCL released the proposed OpenCL 3.0 spec for comments in april 2020
 
-* OpenCL 2.2 API Specification:
-  <https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_API.html>
-* OpenCL 2.2 Extension Specification:
-  <https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_Ext.html>
-* OpenCL 2.2 SPIR-V Environment Specification:
-  <https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_Env.html>
+* [Announcement video](https://youtu.be/h0_syTg6TtY)
+* [Announcement video slides (PDF)](https://www.khronos.org/assets/uploads/apis/OpenCL-3.0-Launch-Apr20.pdf)
 
 Note: We are implementing hardware accelerated Vulkan and
 OpenCL while relying on other software projects to translate APIs to
@@ -94,7 +154,30 @@ although performance is not evaluated.
 
 <https://synappsis.wordpress.com/2017/06/03/opengl-over-vulkan-dev/>
 
-# Free Silicon Conference
+* Pixilica is heading up an initiative to create a RISC-V graphical ISA
+
+* [Pixilica 3D Graphical ISA Slides](https://b5792ddd-543e-4dd4-9b97-fe259caf375d.filesusr.com/ugd/841f2a_c8685ced353b4c3ea20dbb993c4d4d18.pdf)
+
+
+# Various POWER Communities
+ - [An effort to make a 100% Libre POWER Laptop](https://www.powerpc-notebook.org/en/)
+   The T2080 is a POWER8 chip.
+ - [Power Progress Community](https://www.powerprogress.org/campaigns/donations-to-all-the-power-progress-community-projects/)
+   Supporting/Raising awareness of various POWER related open projects on the FOSS
+   community
+ - [OpenPOWER](https://openpowerfoundation.org)
+   Promotes and ensure compliance with the Power ISA amongst members.
+ - [OpenCapi](https://opencapi.org)
+   High performance interconnect for POWER machines. One of the big advantages
+   of the POWER architecture. Notably more performant than PCIE Gen4, and is
+   designed to be layered on top of the physical PCIE link.
+ - [OpenPOWER “Virtual Coffee” Calls](https://openpowerfoundation.org/openpower-virtual-coffee-calls/)
+   Truly open bi-weekly teleconference lines for anybody interested in helping
+   advance or adopting the POWER architecture.
+
+# Conferences
+
+## Free Silicon Conference
 
 The conference brought together experts and enthusiasts who want to build
 a complete Free and Open Source CAD ecosystem for designing analog and
@@ -118,6 +201,16 @@ layout generation flow (RTL-to-GDS).
 
 * <https://theopenroadproject.org/>
 
+# Other RISC-V GPU attempts
+
+* <https://fossi-foundation.org/2019/09/03/gsoc-64b-pointers-in-rv32>
+
+* <http://bjump.org/manycore/>
+
+* <https://resharma.github.io/RISCV32-GPU/>
+
+TODO: Get in touch and discuss collaboration
+
 # Tests, Benchmarks, Conformance, Compliance, Verification, etc.
 
 ## RISC-V Tests
@@ -127,17 +220,22 @@ test. It's still in development as far as I can tell.
 
 * //TODO LINK TO RISC-V CONFORMANCE TEST
 
-## IEEE 754 Tests
+## IEEE 754 Testing/Emulation
+
+IEEE 754 has no official tests for floating-point but there are
+well-known third party tools to check such as John Hauser's TestFloat.
 
-IEEE 754 has no official tests for floating-point but there are several
-well-known third party tools to check such as John Hauser's SoftFloat
-and TestFloat.
+There is also his SoftFloat library, which is a software emulation
+library for IEEE 754.
 
 * <http://www.jhauser.us/arithmetic/>
 
-Jacob is also making a Rust library to check IEEE 754 operations.
+Jacob is also working on an IEEE 754 software emulation library written
+in Rust which also has Python bindings:
 
-* <http://lists.libre-riscv.org/pipermail/libre-riscv-dev/2019-September/002737.html>
+* Source: <https://salsa.debian.org/Kazan-team/simple-soft-float>
+* Crate: <https://crates.io/crates/simple-soft-float>
+* Autogenerated Docs: <https://docs.rs/simple-soft-float/>
 
 A cool paper I came across in my research is "IeeeCC754++ : An Advanced
 Set of Tools to Check IEEE 754-2008 Conformity" by Dr. Matthias Hüsken.
@@ -160,20 +258,138 @@ the Khronos standards until we actually make an official submission,
 do the paperwork, and pay the relevant fees.
 
 ## Formal Verification
-Formal verification of Libre RISC-V ensures that it is bug-free in regards to what we specify.
-Of course, it is important to do the formal verification as a final step in the development process before
-we produce thousands or millions of silicon.
+
+Formal verification of Libre RISC-V ensures that it is bug-free in
+regards to what we specify.  Of course, it is important to do the formal
+verification as a final step in the development process before we produce
+thousands or millions of silicon.
 
 Some learning resources I found in the community:
 
-* ZipCPU: <http://zipcpu.com/>
+* ZipCPU: <http://zipcpu.com/> ZipCPU provides a comprehensive
+  tutorial for beginners and many exercises/quizzes/slides:
+  <http://zipcpu.com/tutorial/>
+* Western Digital's SweRV CPU blog (I recommend looking at all their
+  posts): <https://tomverbeure.github.io/>
+* <https://tomverbeure.github.io/risc-v/2018/11/19/A-Bug-Free-RISC-V-Core-without-Simulation.html>
+* <https://tomverbeure.github.io/rtl/2019/01/04/Under-the-Hood-of-Formal-Verification.html>
+
+## Automation
+
+* <https://www.ohwr.org/project/wishbone-gen>
+
+# LLVM
+
+## Adding new instructions:
+
+* <https://archive.fosdem.org/2015/schedule/event/llvm_internal_asm/>
+
+# Branch Prediction
+
+* <https://danluu.com/branch-prediction/>
+
+# Python RTL Tools
+
+* [Migen - a Python RTL](https://jeffrey.co.in/blog/2014/01/d-flip-flop-using-migen/)
+* [LiTeX](https://github.com/timvideos/litex-buildenv/wiki/LiteX-for-Hardware-Engineers)
+  An SOC builder written in Python Migen DSL. Allows you to generate functional
+  RTL for a SOC configured with cache, a RISCV core, ethernet, DRAM support,
+  and parameterizeable CSRs.
+* [Migen Tutorial](http://blog.lambdaconcept.com/doku.php?id=migen:tutorial>)
+* There is a great guy, Robert Baruch, who has a good
+  [tutorial](https://github.com/RobertBaruch/nmigen-tutorial) on nMigen.
+  He also build an FPGA-proven Motorola 6800 CPU clone with nMigen and put
+  [the code](https://github.com/RobertBaruch/n6800) and
+  [instructional videos](https://www.youtube.com/playlist?list=PLEeZWGE3PwbbjxV7_XnPSR7ouLR2zjktw)
+  online.
+* [Minerva](https://github.com/lambdaconcept/minerva)
+  An SOC written in Python nMigen DSL
+* [Using our Python Unit Tests(old)](http://lists.libre-riscv.org/pipermail/libre-riscv-dev/2019-March/000705.html)
+* <https://chisel.eecs.berkeley.edu/api/latest/chisel3/util/DecoupledIO.html>
+* <http://www.clifford.at/papers/2016/yosys-synth-formal/slides.pdf>
+
+# Other
+
+* <https://wiki.f-si.org/index.php/FSiC2019>
+* <https://fusesoc.net>
+* <https://www.lowrisc.org/open-silicon/>
+* <http://fpgacpu.ca/fpga/Pipeline_Skid_Buffer.html> pipeline skid buffer
+* <https://pyvcd.readthedocs.io/en/latest/vcd.gtkw.html> GTKwave
+* <http://www.sunburst-design.com/papers/CummingsSNUG2002SJ_Resets.pdf>
+  Synchronous Resets? Asynchronous Resets? I am so confused! How will I
+  ever know which to use? by Clifford E. Cummings
+* <http://www.sunburst-design.com/papers/CummingsSNUG2008Boston_CDC.pdf>
+  Clock Domain Crossing (CDC) Design & Verification Techniques Using
+  SystemVerilog, by Clifford E. Cummings
+  In particular, see section 5.8.2: Multi-bit CDC signal passing using
+  1-deep / 2-register FIFO synchronizer.
+* <http://www2.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-143.pdf>
+  Understanding Latency Hiding on GPUs, by Vasily Volkov
+* Efabless "Openlane" <https://github.com/efabless/openlane>
+* Co-simulation plugin for verilator, transferring to ECP5
+  <https://github.com/vmware/cascade>
+
+
+# Real/Physical Projects
+
+* [Samuel's KC5 code](http://chiselapp.com/user/kc5tja/repository/kestrel-3/dir?ci=6c559135a301f321&name=cores/cpu)
+* <https://chips4makers.io/blog/>
+* <https://hackaday.io/project/7817-zynqberry>
+* <https://github.com/efabless/raven-picorv32> 
+* <https://efabless.com>
+* <https://efabless.com/design_catalog/default>
+* <https://wiki.f-si.org/index.php/The_Raven_chip:_First-time_silicon_success_with_qflow_and_efabless>
+* <https://mshahrad.github.io/openpiton-asplos16.html>
+
+# ASIC tape-out pricing
+
+* <https://europractice-ic.com/wp-content/uploads/2020/05/General-MPW-EUROPRACTICE-200505-v8.pdf>
+
+# Funding
+
+* <https://toyota-ai.ventures/>
+* [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)
+
+# Good Programming/Design Practices
+
+* [Liskov Substitution Principle](https://en.wikipedia.org/wiki/Liskov_substitution_principle)
+* [Principle of Least Astonishment](https://en.wikipedia.org/wiki/Principle_of_least_astonishment)
+* <https://peertube.f-si.org/videos/watch/379ef007-40b7-4a51-ba1a-0db4f48e8b16>
+* [Rust-Lang Philosophy and Consensus](http://smallcultfollowing.com/babysteps/blog/2019/04/19/aic-adventures-in-consensus/)
+
+* <https://youtu.be/o5Ihqg72T3c>
+* <http://flopoco.gforge.inria.fr/>
+* Fundamentals of Modern VLSI Devices
+  <https://groups.google.com/a/groups.riscv.org/d/msg/hw-dev/b4pPvlzBzu0/7hDfxArEAgAJ>
+
+# 12 skills summary
+
+* <https://www.crnhq.org/cr-kit/>
+
+# Analog Simulation
+
+* <https://github.com/Isotel/mixedsim>
+* <http://www.vlsiacademy.org/open-source-cad-tools.html>
+* <http://ngspice.sourceforge.net/adms.html>
+* <https://en.wikipedia.org/wiki/Verilog-AMS#Open_Source_Implementations>
+
+# Libre-SOC Standards
+
+This list auto-generated from a page tag "standards":
+
+[[!inline pages="tagged(standards)" actions="no" archive="yes" quick="yes"]]
 
-ZipCPU provides a comprehensive tutorial for beginners and many exercises/quizzes/slides: <http://zipcpu.com/tutorial/>
+# Server setup
 
+* [[resources/server-setup/web-server]]
+* [[resources/server-setup/git-mirroring]]
+* [[resources/server-setup/nagios-monitoring]]
 
-* Western Digital's SweRV CPU blog (I recommend looking at all their posts): <https://tomverbeure.github.io/>
+# Testbeds
 
-<https://tomverbeure.github.io/risc-v/2018/11/19/A-Bug-Free-RISC-V-Core-without-Simulation.html>
+* <https://www.fed4fire.eu/testbeds/>
 
-<https://tomverbeure.github.io/rtl/2019/01/04/Under-the-Hood-of-Formal-Verification.html>
+# Really Useful Stuff
 
+* <https://github.com/im-tomu/fomu-workshop/blob/master/docs/requirements.txt>
+* <https://github.com/im-tomu/fomu-workshop/blob/master/docs/conf.py#L39-L47>