From WikiChip
Editing arm holdings/microarchitectures/cortex-a57

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 6: Line 6:
 
|manufacturer=TSMC
 
|manufacturer=TSMC
 
|introduction=Oct 30, 2012
 
|introduction=Oct 30, 2012
|isa=ARMv8
 
 
|predecessor=Cortex-A15
 
|predecessor=Cortex-A15
 
|predecessor link=arm_holdings/microarchitectures/cortex-a15
 
|predecessor link=arm_holdings/microarchitectures/cortex-a15
Line 13: Line 12:
 
}}
 
}}
 
'''Cortex-A57''' (codename '''Atlas''') is the successor to the {{armh|Cortex-A15|l=arch}}, a low-power high-performance [[ARM]] [[microarchitecture]] designed by [[ARM Holdings]] for the mobile market. This microarchitecture is designed as a synthesizable [[IP core]] and is sold to other semiconductor companies to be implemented in their own chips. The Cortex-A57, which implemented the {{arm|ARMv8}} ISA, is the a performant core which is often combined with a number of lower power cores (e.g. {{\\|Cortex-A53}}) in a {{armh|big.LITTLE}} configuration to achieve better energy/performance.
 
'''Cortex-A57''' (codename '''Atlas''') is the successor to the {{armh|Cortex-A15|l=arch}}, a low-power high-performance [[ARM]] [[microarchitecture]] designed by [[ARM Holdings]] for the mobile market. This microarchitecture is designed as a synthesizable [[IP core]] and is sold to other semiconductor companies to be implemented in their own chips. The Cortex-A57, which implemented the {{arm|ARMv8}} ISA, is the a performant core which is often combined with a number of lower power cores (e.g. {{\\|Cortex-A53}}) in a {{armh|big.LITTLE}} configuration to achieve better energy/performance.
 
== Compiler support ==
 
{| class="wikitable"
 
|-
 
! Compiler !! Arch-Specific || Arch-Favorable
 
|-
 
| [[Arm Compiler]] || <code>-mcpu=cortex-a57</code> || <code>-mtune=cortex-a57</code>
 
|-
 
| [[GCC]] || <code>-mcpu=cortex-a57</code> || <code>-mtune=cortex-a57</code>
 
|-
 
| [[LLVM]] || <code>-mcpu=cortex-a57</code> || <code>-mtune=cortex-a57</code>
 
|}
 
 
If the Cortex-A57 is coupled with the {{\\|Cortex-A53}} in a [[big.LITTLE]] system, GCC also supports the following option:
 
 
{| class="wikitable"
 
|-
 
! Compiler !! Tune
 
|-
 
| [[GCC]] || <code>-mtune=cortex-a57.cortex-a53</code>
 
|}
 
  
 
== Architecture ==
 
== Architecture ==
Line 44: Line 22:
  
 
== Die ==
 
== Die ==
<table class="wikitable">
+
=== Samsung [[Exynos 5433]] ===
<tr><th colspan="3">Cortex-A57 Clusters Silicon Areas (Estimated)</th></tr>
 
<tr><th>Company</th><td>[[Samsung]]</td><td>[[Renesas]]</td></tr>
 
<tr><th>Chip</th><td>[[Exynos 5433]]</td><td>[[R-Car H3]]</td></tr>
 
<tr><th>Process</th><td>[[20 nm]]</td><td>[[16 nm]]</td></tr>
 
<tr><th>Configuration</th><td>4x Cortex-A57<br>+ 2 MiB L2</td><td>4x Cortex-A57<br>+ 2 MiB L2</td></tr>
 
<tr><th>Cluster Size</th><td>~15.85 mm²</td><td>~10.21 mm² cluster</td></tr>
 
<tr><th>Core Size</th><td>~3 mm²</td><td>~1.66 mm² cluster</td></tr>
 
<tr><th>Cache Size</th><td>~3.87 mm²</td><td>~3.28 mm² cluster</td></tr>
 
</table>
 
 
 
=== 20 nm ===
 
==== Samsung [[Exynos 5433]] ====
 
 
* Samsung [[20 nm process]]
 
* Samsung [[20 nm process]]
 
* 113 mm² die size
 
* 113 mm² die size
 
* Mali-T760 (6 EU)
 
* Mali-T760 (6 EU)
* Quad-core {Primary Cortex-A57 [Quad] + Secondary A53 [Quad]}
+
* Quad-core {{\\|Cortex-A53}}
** 32 KiB L1I$ and 32 KiB L1D$ per core, and a shared 256 KiB L2
+
* Quad-core Cortex-A57
** 4.4 mm² per cluster
 
*** ~1 mm² per core
 
*** ~0.55 mm² for 256 KiB L2 cache
 
* Quad-core Cortex-A57 ([[big cores]])
 
 
** 48KB L1I$ and 32KB L1D$ per core, and a shared 2 MiB L2
 
** 48KB L1I$ and 32KB L1D$ per core, and a shared 2 MiB L2
 
** 15.85 mm² per cluster
 
** 15.85 mm² per cluster
Line 73: Line 35:
  
 
:[[File:exynos 5433 die.png|600px]]
 
:[[File:exynos 5433 die.png|600px]]
 
=== 16 nm ===
 
==== Renesas [[R-Car H3]] ====
 
* TSMC [[16 nm process]]
 
* 12.94 mm × 8.61 mm
 
* 111.36 mm² die size
 
* Quad-core {{\\|Cortex-A53}}
 
** ~3.27 mm² cluster
 
** ~0.60 mm² core
 
** ~0.7 mm² 512 KiB L2 cache
 
* Quad-core Cortex-A57
 
** ~10.21 mm² cluster
 
** ~1.66 mm² core
 
** ~3.28 mm² 2 MiB L2 cache
 
* {{\\|Cortex-R7}} (dual-core [[lock-step]])
 
** ~1.04 mm² cluster
 
* GX6650 GPU
 
** ~28.12 mm²
 
 
 
: [[File:r-car h3 die shot.png|650px]]
 
 
 
A57 Cluster:
 
 
:[[File:h3 a57 cluster.png|400px]]
 
 
== Bibliography ==
 
* Pyo, Jungyul, et al. "23.1 20nm high-K metal-gate heterogeneous 64b quad-core CPUs and hexa-core GPU for high-performance and energy-efficient mobile application processor." Solid-State Circuits Conference-(ISSCC), 2015 IEEE International. IEEE, 2015
 

Please note that all contributions to WikiChip may be edited, altered, or removed by other contributors. If you do not want your writing to be edited mercilessly, then do not submit it here.
You are also promising us that you wrote this yourself, or copied it from a public domain or similar free resource (see WikiChip:Copyrights for details). Do not submit copyrighted work without permission!

Cancel | Editing help (opens in new window)

This page is a member of 1 hidden category:

codenameCortex-A57 +
designerARM Holdings +
first launchedOctober 30, 2012 +
full page namearm holdings/microarchitectures/cortex-a57 +
instance ofmicroarchitecture +
instruction set architectureARMv8 +
manufacturerTSMC +
microarchitecture typeCPU +
nameCortex-A57 +