Bug 133

Summary: adaptable / collapsible pipelines
Product: Libre-SOC's first SoC Reporter: Luke Kenneth Casson Leighton <lkcl>
Component: ALU (including IEEE754 16/32/64-bit FPU)Assignee: Luke Kenneth Casson Leighton <lkcl>
Status: CONFIRMED ---    
Severity: enhancement CC: libre-soc-bugs, programmerjake
Priority: ---    
Version: unspecified   
Hardware: PC   
OS: Linux   
NLnet milestone: --- total budget (EUR) for completion of task and all subtasks: 0
budget (EUR) for this task, excluding subtasks' budget: 0 parent task for budget allocation:
child tasks for budget allocation: The table of payments (in EUR) for this task; TOML format:

Description Luke Kenneth Casson Leighton 2019-08-17 04:12:18 BST
an adaptable pipeline would be really nice to have, trading latency for
depth and allowing increased clock rate.
Comment 1 Luke Kenneth Casson Leighton 2019-08-17 04:13:03 BST
https://groups.google.com/forum/#!topic/comp.arch/fcq-GLQqvas

---

This is by IBM but for a pipelined MPEG decoder:

Fine-Grain Real-Time Reconfigurable Pipelining 2003
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.6.1015

Other search terms are "transparent pipeline" "collapsed pipeline"
"Adaptive pipeline"

Transparent mode flip-flops for collapsible pipelines 2007
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.182.2056

Adaptive pipeline structures for speculation control 2003
http://apt.cs.manchester.ac.uk/ftp/pub/amulet/papers/efthym_async03.pdf
Comment 2 Jacob Lifshay 2019-08-17 04:14:54 BST
(In reply to Luke Kenneth Casson Leighton from comment #0)
> an adaptable pipeline would be really nice to have, trading latency for
> depth and allowing increased clock rate.

let's just not make another NetBurst. :)
Comment 3 Luke Kenneth Casson Leighton 2019-08-17 09:47:08 BST
(In reply to Jacob Lifshay from comment #2)

> let's just not make another NetBurst. :)

https://pcper.com/2011/08/yes-netburst-really-was-that-bad-cpu-architectures-tested/

"One way that Pentium Netburst kept high clock rates was by having a ridiculously huge pipeline, 2-4x larger than the first generation of Core 2 parts which replaced it; unfortunately the Pentium 4 branch prediction was terrible keeping the processor stuck needing to dump its pipeline perpetually."

hey it's good that like, y'know... we're doing this processor with so many
historical mistakes to learn from...