Semiconductor & Computer Engineering
From WikiChip
Difference between revisions of "intel/microarchitectures/cooper lake"
(Correcting Successor and contemporary architectures, fixed cache amounts, added core counts, and removed random designer labeled) |
|||
| (One intermediate revision by one other user not shown) | |||
| Line 6: | Line 6: | ||
|manufacturer=Intel | |manufacturer=Intel | ||
|introduction=June 18, 2020 | |introduction=June 18, 2020 | ||
| − | |process=14 nm | + | |process=14 nm++ |
| + | |cores=28 | ||
| + | |cores 2=24 | ||
| + | |cores 3=20 | ||
| + | |cores 4=18 | ||
| + | |cores 5=16 | ||
| + | |cores 6=8 | ||
|type=Superscalar | |type=Superscalar | ||
|oooe=Yes | |oooe=Yes | ||
| Line 51: | Line 57: | ||
|l1d per=core | |l1d per=core | ||
|l1d desc=8-way set associative | |l1d desc=8-way set associative | ||
| + | |l1=64 KiB | ||
| + | |l1 per=core | ||
|l2=1 MiB | |l2=1 MiB | ||
| − | |l2 per= | + | |l2 per=Core |
|l2 desc=16-way set associative | |l2 desc=16-way set associative | ||
|l3=1.375 MiB | |l3=1.375 MiB | ||
| Line 62: | Line 70: | ||
|predecessor=Cascade Lake | |predecessor=Cascade Lake | ||
|predecessor link=intel/microarchitectures/cascade lake | |predecessor link=intel/microarchitectures/cascade lake | ||
| − | |successor=Ice Lake (Server) | + | |successor=Sapphire Rapids |
| − | | | + | |successor link=intel/microarchitectures/sapphire rapids |
| − | |contemporary=Coffee Lake | + | |contemporary=Ice Lake (Server) |
| − | |contemporary link=intel/microarchitectures/coffee lake | + | |contemporary link=intel/microarchitectures/ice lake (server) |
| + | |contemporary 2=Coffee Lake | ||
| + | |contemporary 2 link=intel/microarchitectures/coffee lake | ||
}} | }} | ||
'''Cooper Lake''' ('''CPL''' / '''CPX''') is [[Intel]]'s successor to {{\\|Cascade Lake}}, a [[14 nm]] [[microarchitecture]] for the multiprocessing server market only. | '''Cooper Lake''' ('''CPL''' / '''CPX''') is [[Intel]]'s successor to {{\\|Cascade Lake}}, a [[14 nm]] [[microarchitecture]] for the multiprocessing server market only. | ||
Latest revision as of 17:40, 26 March 2024
| Edit Values | |
| Cooper Lake µarch | |
| General Info | |
| Arch Type | CPU |
| Designer | Intel |
| Manufacturer | Intel |
| Introduction | June 18, 2020 |
| Process | 14 nm++ |
| Core Configs | 28, 24, 20, 18, 16, 8 |
| Pipeline | |
| Type | Superscalar |
| OoOE | Yes |
| Speculative | Yes |
| Reg Renaming | Yes |
| Stages | 14-19 |
| Instructions | |
| ISA | x86-64 |
| Extensions | MOVBE, MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, POPCNT, AVX, AVX2, AES, PCLMUL, FSGSBASE, RDRND, FMA3, F16C, BMI, BMI2, VT-x, VT-d, TXT, TSX, RDSEED, ADCX, PREFETCHW, CLFLUSHOPT, XSAVE, SGX, MPX, AVX-512 |
| Cache | |
| L1I Cache | 32 KiB/core 8-way set associative |
| L1D Cache | 32 KiB/core 8-way set associative |
| L1 Cache | 64 KiB/core |
| L2 Cache | 1 MiB/Core 16-way set associative |
| L3 Cache | 1.375 MiB/core 11-way set associative |
| Cores | |
| Core Names | Cooper Lake X, Cooper Lake SP, Cooper Lake AP |
| Succession | |
| Contemporary | |
| Ice Lake (Server) Coffee Lake | |
Cooper Lake (CPL / CPX) is Intel's successor to Cascade Lake, a 14 nm microarchitecture for the multiprocessing server market only.
Launched in mid-2020, Cooper Lake covers the 4-way and 8-way multiprocessing segments while Ice Lake serves the single and dual-socket segments.
For scalable server class processors, Intel branded it as Xeon Gold and Xeon Platinum.
Contents
Codenames[edit]
Single and dual-socket Cooper Lake parts were scrapped before ever making it to market.
| Core | Abbrev | Target |
|---|---|---|
| Cooper Lake X | CPL-X | High-end desktops & enthusiasts market |
| Cooper Lake W | CPL-W | Enterprise/Business workstations |
| Cooper Lake SP | CPL-SP | Server Scalable Processors |
| Cooper Lake AP | CPL-AP | Server Advanced Processors |
Brands[edit]
 |
This section is empty; you can help add the missing info by editing this page. |
Release Dates[edit]
Cooper Lake and Ice Lake roadmap.
Cooper was first publicly disclosed in early 2019. Cooper Lake launched on June 18, 2020.
Process Technology[edit]
Cooper Lake is fabricated on Intel's 3rd generation enhanced 14nm++ process.
Architecture[edit]
Key changes from Cascade Lake[edit]
- SoC
- 2x UPI links (6, up from 3)
- Memory
- Higher data rate (3200 MT/s, up from 2933 MT/s)
- Optane DC DIMMs
- Apache Pass → Barlow Pass
- Platform
- Packaging
- Socket-P+
- 4189-contact flip-chip LGA (up from 3647 contacts)
- Socket-P+
This list is incomplete; you can help by expanding it.
New instructions[edit]
Cooper Lake introduced a number of new instructions:
- BFLOAT16 - A new data type for acceleration of AI workloads.
- AVX512 BF16 - AVX-512 Brain Float 16 extension
See also[edit]
Facts about "Cooper Lake - Microarchitectures - Intel"
| codename | Cooper Lake + |
| core count | 28 +, 24 +, 20 +, 18 +, 16 + and 8 + |
| designer | Intel + |
| first launched | June 18, 2020 + |
| full page name | intel/microarchitectures/cooper lake + |
| instance of | microarchitecture + |
| instruction set architecture | x86-64 + |
| manufacturer | Intel + |
| microarchitecture type | CPU + |
| name | Cooper Lake + |
| pipeline stages (max) | 19 + |
| pipeline stages (min) | 14 + |