From WikiChip
Editing intel/microarchitectures/broadwell (client)
Warning: You are not logged in. Your IP address will be publicly visible if you make any edits. If you log in or create an account, your edits will be attributed to your username, along with other benefits.
The edit can be undone.
Please check the comparison below to verify that this is what you want to do, and then save the changes below to finish undoing the edit.
This page supports semantic in-text annotations (e.g. "[[Is specified as::World Heritage Site]]") to build structured and queryable content provided by Semantic MediaWiki. For a comprehensive description on how to use annotations or the #ask parser function, please have a look at the getting started, in-text annotation, or inline queries help pages.
Latest revision | Your text | ||
Line 1: | Line 1: | ||
{{intel title|Broadwell|arch}} | {{intel title|Broadwell|arch}} | ||
{{microarchitecture | {{microarchitecture | ||
− | |atype=CPU | + | | atype = CPU |
− | |name=Broadwell | + | | name = Broadwell |
− | |designer=Intel | + | | designer = Intel |
− | |manufacturer=Intel | + | | manufacturer = Intel |
− | |introduction=October, 2014 | + | | introduction = October, 2014 |
− | |process=14 nm | + | | phase-out = |
− | |cores=2 | + | | process = 14 nm |
− | |cores 2=4 | + | | cores = 2 |
− | |cores 3=6 | + | | cores 2 = 4 |
− | |cores 4=8 | + | | cores 3 = 6 |
− | |cores 5=10 | + | | cores 4 = 8 |
− | |cores 6=12 | + | | cores 5 = 10 |
− | |cores 7=14 | + | | cores 6 = 12 |
− | |cores 8=16 | + | | cores 7 = 14 |
− | |cores 9=18 | + | | cores 8 = 16 |
− | |cores 10=20 | + | | cores 9 = 18 |
− | |cores 11=22 | + | | cores 10 = 20 |
− | |type=Superscalar | + | | cores 11 = 22 |
− | |speculative=Yes | + | |
− | |renaming=Yes | + | | pipeline = Yes |
− | |stages min=14 | + | | type = Superscalar |
− | |stages max=19 | + | | OoOE = Yes |
− | | | + | | speculative = Yes |
− | |extension=MOVBE | + | | renaming = Yes |
− | |extension 2=MMX | + | | isa = IA-32 |
− | |extension 3=SSE | + | | isa 2 = x86-64 |
− | |extension 4=SSE2 | + | | stages min = 14 |
− | |extension 5=SSE3 | + | | stages max = 19 |
− | |extension 6=SSSE3 | + | | issues = 4 |
− | |extension 7=SSE4.1 | + | |
− | |extension 8=SSE4.2 | + | | inst = Yes |
− | |extension 9=POPCNT | + | | feature = |
− | |extension 10=AVX | + | | extension = MOVBE |
− | |extension 11=AVX2 | + | | extension 2 = MMX |
− | |extension 12=AES | + | | extension 3 = SSE |
− | |extension 13=PCLMUL | + | | extension 4 = SSE2 |
− | |extension 14=FSGSBASE | + | | extension 5 = SSE3 |
− | |extension 15=RDRND | + | | extension 6 = SSSE3 |
− | |extension 16=FMA3 | + | | extension 7 = SSE4.1 |
− | |extension 17=F16C | + | | extension 8 = SSE4.2 |
− | |extension 18=BMI | + | | extension 9 = POPCNT |
− | |extension 19=BMI2 | + | | extension 10 = AVX |
− | |extension 20=VT-x | + | | extension 11 = AVX2 |
− | |extension 21=VT-d | + | | extension 12 = AES |
− | |extension 22=TXT | + | | extension 13 = PCLMUL |
− | |extension 23=TSX | + | | extension 14 = FSGSBASE |
− | |extension 24=RDSEED | + | | extension 15 = RDRND |
− | |extension 25=ADCX | + | | extension 16 = FMA3 |
− | |extension 26=PREFETCHW | + | | extension 17 = F16C |
− | |l1i=32 KiB | + | | extension 18 = BMI |
− | |l1i per=core | + | | extension 19 = BMI2 |
− | |l1i desc=8-way set associative | + | | extension 20 = VT-x |
− | |l1d=32 KiB | + | | extension 21 = VT-d |
− | |l1d per=core | + | | extension 22 = TXT |
− | |l1d desc=8-way set associative | + | | extension 23 = TSX |
− | |l2=256 KiB | + | | extension 24 = RDSEED |
− | |l2 per=core | + | | extension 25 = ADCX |
− | |l2 desc=8-way set associative | + | | extension 26 = PREFETCHW |
− | |l3=1.5 MiB | + | |
− | |l3 per=core | + | | cache = Yes |
− | |l4=128 MiB | + | | l1i = 32 KiB |
− | |l4 per=package | + | | l1i per = core |
− | |l4 desc=on Iris Pro GPUs only | + | | l1i desc = 8-way set associative |
− | |core name=Broadwell Y | + | | l1d = 32 KiB |
− | |core name 2=Broadwell U | + | | l1d per = core |
− | |core name 3=Broadwell H | + | | l1d desc = 8-way set associative |
− | |core name 4=Broadwell DT | + | | l2 = 256 KiB |
− | |core name 5=Broadwell EP | + | | l2 per = core |
− | |core name 6=Broadwell EX | + | | l2 desc = 8-way set associative |
− | |core name 7=Broadwell E | + | | l3 = 1.5 MiB |
− | |predecessor=Haswell | + | | l3 per = core |
− | |predecessor link=intel/microarchitectures/haswell | + | | l3 desc = |
− | |successor=Skylake | + | | l4 = 128 MiB |
− | + | | l4 per = package | |
− | + | | l4 desc = on Iris Pro GPUs only | |
− | |successor | + | |
− | + | | core names = Yes | |
− | + | | core name = Broadwell Y | |
− | + | | core name 2 = Broadwell U | |
− | + | | core name 3 = Broadwell H | |
− | + | | core name 4 = Broadwell DT | |
− | + | | core name 5 = Broadwell EP | |
− | + | | core name 6 = Broadwell EX | |
+ | | core name 7 = Broadwell E | ||
+ | |||
+ | | succession = Yes | ||
+ | | predecessor = Haswell | ||
+ | | predecessor link = intel/microarchitectures/haswell | ||
+ | | successor = Skylake | ||
+ | | successor link = intel/microarchitectures/skylake | ||
}} | }} | ||
'''Broadwell''' ('''BDW''') is [[Intel]]'s [[microarchitecture]] based on the [[14 nm process]] for mobile, desktops, and servers. Introduced in early 2015, Broadwell is a [[process shrink]] of {{\\|Haswell}} which introduced several enhancements. Broadwell is named after [[wikipedia:Broadwell, Illinois|Broadwell, Illinois]]. | '''Broadwell''' ('''BDW''') is [[Intel]]'s [[microarchitecture]] based on the [[14 nm process]] for mobile, desktops, and servers. Introduced in early 2015, Broadwell is a [[process shrink]] of {{\\|Haswell}} which introduced several enhancements. Broadwell is named after [[wikipedia:Broadwell, Illinois|Broadwell, Illinois]]. | ||
Line 109: | Line 116: | ||
== Process Technology == | == Process Technology == | ||
− | + | Broadwell is designed to be manufactured using [[14 nm]] Tri-gate [[FinFET]] transistors. This correlates to 8 nm Fin width and a 42 nm Fin pitch (shown below). SRAM cell is at .0588 µm². | |
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | Broadwell is designed to be manufactured using [[14 nm]] Tri-gate [[FinFET]] transistors. This correlates to 8 nm Fin width and a 42 nm Fin pitch (shown below). SRAM cell is at | ||
− | |||
− | + | Scaling: | |
+ | [[File:intel 14nm gate.png|215px|left]] | ||
{| class="wikitable" | {| class="wikitable" | ||
|- | |- | ||
− | ! !! Haswell !! Broadwell !! Δ | + | ! !! Haswell !! Broadwell !! Δ |
|- | |- | ||
| || [[22 nm]] || [[14 nm]] || | | || [[22 nm]] || [[14 nm]] || | ||
|- | |- | ||
− | | Fin Pitch || 60 nm || 42 || 0.70x | + | | Fin Pitch || 60 nm || 42 nm || 0.70x |
− | |||
− | |||
− | |||
− | |||
|- | |- | ||
| Gate Pitch || 90 nm || 70 nm || 0.78x | | Gate Pitch || 90 nm || 70 nm || 0.78x | ||
Line 147: | Line 135: | ||
| Cell Height || 840 nm || 399 nm || 0.48x | | Cell Height || 840 nm || 399 nm || 0.48x | ||
|} | |} | ||
+ | {{clear}} | ||
== Architecture== | == Architecture== | ||
+ | [[File:Intel-Xeon-processor-D-1500-wafer.jpg|right|thumb|350px|Broadwell {{intel|Xeon D}} wafer]] | ||
Broadwell is for the most part identical to {{\\|Haswell}} with several enhancements, including new instruction set extensions. | Broadwell is for the most part identical to {{\\|Haswell}} with several enhancements, including new instruction set extensions. | ||
=== Key changes from {{\\|Haswell}} === | === Key changes from {{\\|Haswell}} === | ||
− | |||
* ~5% IPC improvement | * ~5% IPC improvement | ||
* FP multiplication instructions has reduced latency (3 cycles, down from 5) | * FP multiplication instructions has reduced latency (3 cycles, down from 5) | ||
** Affects AVX, SSE, and FP instructions | ** Affects AVX, SSE, and FP instructions | ||
− | * {{x86|CLMUL}} instructions are now a single [[ | + | * {{x86|CLMUL}} instructions are now a single [[μop]], improving latency and throughput |
* The second-level TLB (STLB) | * The second-level TLB (STLB) | ||
** Table was enlarged (1,536 entries, up from 1024) | ** Table was enlarged (1,536 entries, up from 1024) | ||
Line 177: | Line 166: | ||
==== New instructions ==== | ==== New instructions ==== | ||
+ | {{main|#Added instructions|l1=See #Added_instructions for the complete list}} | ||
Broadwell introduced a number of new instructions: | Broadwell introduced a number of new instructions: | ||
* {{x86|RDSEED|<code>RDSEED</code>}} - Generates 16, 32 or 64 bit random numbers seeds ([[NIST SP 800-90B]] & [[NIST SP 800-90C]]) | * {{x86|RDSEED|<code>RDSEED</code>}} - Generates 16, 32 or 64 bit random numbers seeds ([[NIST SP 800-90B]] & [[NIST SP 800-90C]]) | ||
* {{x86|ADCX|<code>ADCX</code>}} - Arbitrary precision integer operations | * {{x86|ADCX|<code>ADCX</code>}} - Arbitrary precision integer operations | ||
* {{x86|PREFETCHW|<code>PREFETCHW</code>}} - Prefetch data into caches, hinting a write is expected in the future | * {{x86|PREFETCHW|<code>PREFETCHW</code>}} - Prefetch data into caches, hinting a write is expected in the future | ||
− | |||
=== Block Diagram === | === Block Diagram === | ||
Line 187: | Line 176: | ||
=== Memory Hierarchy === | === Memory Hierarchy === | ||
− | |||
* Cache | * Cache | ||
** L1 Cache: | ** L1 Cache: | ||
Line 199: | Line 187: | ||
*** Per core | *** Per core | ||
** L3 Cache: | ** L3 Cache: | ||
− | *** 1.5 | + | *** 1.5 - 3 MiB per core, 64 B line size |
− | *** | + | *** 16-20 -way set associative |
*** Write-back policy | *** Write-back policy | ||
** L4 Cache: | ** L4 Cache: | ||
Line 238: | Line 226: | ||
== High Core count (EP) == | == High Core count (EP) == | ||
* Key Changes from {{\\|Haswell}}: | * Key Changes from {{\\|Haswell}}: | ||
− | ** Up to | + | ** Up to 22 cores (up from 18) |
− | ** Up to | + | ** Up to 44 threads (up from 36) |
− | ** Up to | + | ** Up to 55 MiB [[LLC]] (up from 45 MiB) |
** Up to 2400 DDR (from 2133) | ** Up to 2400 DDR (from 2133) | ||
Line 280: | Line 268: | ||
! Low Core Count (LCC) !! Medium Core Count (MCC) !! High Core Count (HCC) | ! Low Core Count (LCC) !! Medium Core Count (MCC) !! High Core Count (HCC) | ||
|- | |- | ||
− | | Up to 10 Cores || 12-14 Cores || 16 | + | | Up to 10 Cores || 12-14 Cores || 16+ Cores |
|- | |- | ||
| 246.24 mm² || 306.18 mm² || 456.12 mm² | | 246.24 mm² || 306.18 mm² || 456.12 mm² | ||
Line 288: | Line 276: | ||
|[[File:E5 v4 LCC.png|300px]] || [[File:E5 v4 MCC.png|300px]] || [[File:E5 v4 HCC.png|300px]] | |[[File:E5 v4 LCC.png|300px]] || [[File:E5 v4 MCC.png|300px]] || [[File:E5 v4 HCC.png|300px]] | ||
|} | |} | ||
+ | |||
+ | == Die == | ||
+ | ===Dual-core Broadwell die=== | ||
− | + | : [[File:broadwell die (dual-core).jpg|850px]] | |
− | |||
− | |||
− | |||
* 1,300,000,000 transistors | * 1,300,000,000 transistors | ||
− | * 82 | + | * 82 mm<sup>2</sup> |
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
===Dual-core Broadwell with {{intel|Iris Pro}} die=== | ===Dual-core Broadwell with {{intel|Iris Pro}} die=== | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
: [[File:broadwell with iris pro die (dual-core).png|850px]] | : [[File:broadwell with iris pro die (dual-core).png|850px]] | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
===Deca-core Broadwell === | ===Deca-core Broadwell === | ||
− | + | * {{intel|Core i7-6950X}} | |
− | + | * Deca-core microprocessor | |
− | * | ||
− | |||
* 3,400,000,000 transistors | * 3,400,000,000 transistors | ||
− | * 246 mm<sup>2</sup> | + | * 246 mm<sup>2</sup> |
− | |||
:[[File:broadwell (deca-core) die shot.png|650px]] | :[[File:broadwell (deca-core) die shot.png|650px]] | ||
Line 379: | Line 339: | ||
created and tagged accordingly. | created and tagged accordingly. | ||
− | Missing a chip? please dump its name here: | + | Missing a chip? please dump its name here: http://en.wikichip.org/wiki/WikiChip:wanted_chips |
--> | --> | ||
− | + | <table class="wikitable sortable" style="overflow-x: scroll; min-width: 1350px;"> | |
− | <table class=" | + | <tr><th colspan="15" style="background:#D6D6FF;">Broadwell Chips</th></tr> |
− | <tr | + | <tr><th colspan="12">Main processor</th><th colspan="3">IGP</th></tr> |
− | <tr | + | <tr><th>Model</th><th>Family</th><th>Platform</th><th>Core</th><th>Launched</th><th>SDP</th><th>TDP</th><th>C</th><th>T</th><th>Freq</th><th>TBT</th><th>Max Mem</th><th>Name</th><th>Freq</th><th>Max Freq</th></tr> |
− | + | {{table sep|col=15|[[Uniprocessors]]}} | |
− | < | + | {{#ask: [[Category:microprocessor models by intel]][[instance of::microprocessor]][[microarchitecture::Broadwell]][[max cpu count::1]] |
− | {{#ask: [[Category:microprocessor models by intel]] [[instance of::microprocessor]] [[microarchitecture::Broadwell]] [[max cpu count::1]] | ||
|?full page name | |?full page name | ||
|?model number | |?model number | ||
− | |||
− | |||
|?microprocessor family | |?microprocessor family | ||
+ | |?platform | ||
|?core name | |?core name | ||
+ | |?first launched | ||
+ | |?sdp | ||
+ | |?tdp | ||
|?core count | |?core count | ||
|?thread count | |?thread count | ||
− | |||
− | |||
− | |||
|?base frequency#GHz | |?base frequency#GHz | ||
|?turbo frequency (1 core)#GHz | |?turbo frequency (1 core)#GHz | ||
− | + | |?max memory | |
− | |||
− | |||
− | |?max memory | ||
|?integrated gpu | |?integrated gpu | ||
|?integrated gpu base frequency | |?integrated gpu base frequency | ||
|?integrated gpu max frequency | |?integrated gpu max frequency | ||
|format=template | |format=template | ||
− | |template=proc table | + | |template=proc table 2 |
− | + | |userparam=16 | |
− | |||
− | |||
− | |userparam= | ||
|mainlabel=- | |mainlabel=- | ||
− | |limit= | + | |limit=150 |
}} | }} | ||
− | + | {{table sep|col=15|[[Multiprocessors]]}} | |
− | {{#ask: [[Category:microprocessor models by intel]] [[instance of::microprocessor]] [[microarchitecture::Broadwell]] [[max cpu count:: | + | {{#ask: [[Category:microprocessor models by intel]][[instance of::microprocessor]][[microarchitecture::Broadwell]][[max cpu count::!1]] |
|?full page name | |?full page name | ||
|?model number | |?model number | ||
− | |||
− | |||
|?microprocessor family | |?microprocessor family | ||
+ | |?platform | ||
|?core name | |?core name | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
|?first launched | |?first launched | ||
− | |? | + | |?sdp |
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
|?tdp | |?tdp | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
|?core count | |?core count | ||
|?thread count | |?thread count | ||
− | |||
− | |||
− | |||
|?base frequency#GHz | |?base frequency#GHz | ||
|?turbo frequency (1 core)#GHz | |?turbo frequency (1 core)#GHz | ||
− | + | |?max memory | |
− | |||
− | |||
− | |?max memory | ||
|?integrated gpu | |?integrated gpu | ||
|?integrated gpu base frequency | |?integrated gpu base frequency | ||
|?integrated gpu max frequency | |?integrated gpu max frequency | ||
|format=template | |format=template | ||
− | |template=proc table | + | |template=proc table 2 |
− | + | |userparam=16 | |
− | |||
− | |||
− | |userparam= | ||
|mainlabel=- | |mainlabel=- | ||
− | |limit= | + | |limit=150 |
}} | }} | ||
− | {{ | + | {{table count|col=15|ask=[[Category:microprocessor models by intel]][[instance of::microprocessor]][[microarchitecture::Broadwell]]}} |
</table> | </table> | ||
− |
Facts about "Broadwell - Microarchitectures - Intel"
codename | Broadwell + |
core count | 2 +, 4 +, 6 +, 8 +, 10 +, 12 +, 14 +, 16 +, 18 +, 20 + and 22 + |
designer | Intel + |
first launched | October 2014 + |
full page name | intel/microarchitectures/broadwell (client) + |
instance of | microarchitecture + |
instruction set architecture | x86-64 + |
manufacturer | Intel + |
microarchitecture type | CPU + |
name | Broadwell + |
pipeline stages (max) | 19 + |
pipeline stages (min) | 14 + |
process | 14 nm (0.014 μm, 1.4e-5 mm) + |