|
|
Line 1: |
Line 1: |
− | {{armh title|Cortex-A53|arch}}
| |
− | {{microarchitecture
| |
− | |atype=CPU
| |
− | |name=Cortex-A53
| |
− | |designer=ARM Holdings
| |
− | |manufacturer=TSMC
| |
− | |manufacturer 2=Samsung
| |
− | |manufacturer 3=GlobalFoundries
| |
− | |manufacturer 4=SMIC
| |
− | |introduction=October 30, 2012
| |
− | |process=40 nm
| |
− | |process 2=28 nm
| |
− | |process 3=20 nm
| |
− | |process 4=16 nm
| |
− | |process 5=14 nm
| |
− | |process 6=10 nm
| |
− | |cores=1
| |
− | |cores 2=2
| |
− | |cores 3=3
| |
− | |cores 4=4
| |
− | |type=In-order
| |
− | |oooe=No
| |
− | |speculative=Yes
| |
− | |renaming=No
| |
− | |stages=8
| |
− | |decode=2-way
| |
− | |isa=ARMv8
| |
− | |feature=Hardware virtualization
| |
− | |extension=FPU
| |
− | |extension 2=NEON
| |
− | |extension 3=TrustZone
| |
− | |l1i=8-64 KiB
| |
− | |l1i per=core
| |
− | |l1i desc=2-way set associative
| |
− | |l1d=8-64 KiB
| |
− | |l1d per=core
| |
− | |l1d desc=4-way set associative
| |
− | |l2=128 KiB - 2 MiB
| |
− | |l2 per=cluster
| |
− | |l2 desc=16-way set associative
| |
− | |predecessor=Cortex-A7
| |
− | |predecessor link=arm_holdings/microarchitectures/cortex-a7
| |
− | |successor=Cortex-A55
| |
− | |successor link=arm_holdings/microarchitectures/cortex-a55
| |
− | |pipeline=Yes
| |
− | |issues=2
| |
− | |core names=<!-- Yes if specify -->
| |
− | }}
| |
− | '''Cortex-A53''' (formerly '''Apollo''') is an ultra-high efficiency [[microarchitecture]] designed by [[ARM Holdings]] as a successor to the {{armh|Cortex-A7|l=arch}}. The Cortex-A53, which implemented the {{arm|ARMv8}} ISA, is typically found in entry-level smartphone and other embedded devices. Often A53 cores are combined with higher performance processors (e.g. based on {{armh|Cortex-A57|l=arch}} or {{armh|Cortex-A72|l=arch}}) in {{armh|big.LITTLE}} configuration to achieve better energy/performance.
| |
| | | |
− | Note that this microarchitecture is designed as a synthesizable [[IP core]] and is sold to other semiconductor companies to be implemented in their own chips.
| |
− |
| |
− | == Process Technology ==
| |
− | {{empty section}}
| |
− |
| |
− | == Compiler support ==
| |
− | {| class="wikitable"
| |
− | |-
| |
− | ! Compiler !! Arch-Specific || Arch-Favorable || Arch-Target
| |
− | |-
| |
− | | [[GCC]] || <code>-march=armv8-a</code> || <code>-mtune=cortex-a53</code> || <code>-mcpu=cortex-a53</code>
| |
− | |-
| |
− | | [[LLVM]] || <code>-march=armv8-a</code> || <code>-mtune=cortex-a53</code> || <code>-mcpu=cortex-a53</code>
| |
− | |}
| |
− |
| |
− | Note that for big.LITTLE systems it's possible to specify more specific performance tunes:
| |
− |
| |
− | * <code>-mtune=cortex-a57.cortex-a53</code>
| |
− | * <code>-mtune=cortex-a72.cortex-a53</code>
| |
− | * <code>-mtune=cortex-a73.cortex-a53</code>
| |
− |
| |
− |
| |
− |
| |
− | == Architecture ==
| |
− | === Key changes from {{\\|Cortex-A7}} ===
| |
− | {{empty section}}
| |
− | === Block Diagram ===
| |
− | {{empty section}}
| |
− | === Memory Hierarchy ===
| |
− | {{empty section}}
| |
− |
| |
− | == Licensees ==
| |
− | Arm named the following companies as licensees.
| |
− |
| |
− | {{collist
| |
− | |count = 3
| |
− | |
| |
− | * [[AMD]]
| |
− | * [[Broadcom]]
| |
− | * [[Samsung]]
| |
− | * [[Altera]]
| |
− | * [[STmicroelectronics]]
| |
− | * [[MediaTek]]
| |
− | * [[Qualcomm]]
| |
− | * [[Xilinx]]
| |
− | }}
| |
− |
| |
− | == Die ==
| |
− | === 20 nm ===
| |
− | ==== Samsung [[Exynos 5433]] ====
| |
− | * Samsung [[20 nm process]]
| |
− | * 113 mm² die size
| |
− | * Mali-T760 (6 EU)
| |
− | * Quad-core Cortex-A53 ([[small cores]])
| |
− | ** 32 KiB L1I$ and 32 KiB L1D$ per core, and a shared 256 KiB L2
| |
− | ** 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
| |
− | ** 15.85 mm² per cluster
| |
− | *** ~3 mm² per core
| |
− | *** ~3.87 mm² for 2 MiB L2 cache
| |
− |
| |
− |
| |
− | :[[File:exynos 5433 die.png|600px]]
| |
− |
| |
− | ==== MediaTek [[Helio X20]] ====
| |
− |
| |
− | * TSMC [[20 nm process]]
| |
− | * 100 mm² die size
| |
− | * Quad-core ULP Cortex-A53
| |
− | ** ~21.81 mm² per cluster
| |
− | *** ~4.23 mm² per core
| |
− | * Quad-core efficient Cortex-A53
| |
− | ** ~29.73 mm² per cluster
| |
− | *** ~5.41 mm² per core
| |
− | * Dual-core High-performance {{\\|Cortex-A72}} + 1 MiB L2
| |
− | ** ~27.36 mm² per cluster
| |
− | *** ~ 9.60 mm² per core
| |
− | *** ~ 7.50 mm² for 1 MiB L2
| |
− |
| |
− |
| |
− | :[[File:mt6797 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 L2 cache
| |
− | * Quad-core {{\\|Cortex-A57}}
| |
− | ** ~10.21 mm² cluster
| |
− | ** ~1.66 mm² core
| |
− | ** ~3.28 mm² 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]]
| |
− |
| |
− | == All Cortex-A53 Chips ==
| |
− | <!-- NOTE:
| |
− | This table is generated automatically from the data in the actual articles.
| |
− | If a microprocessor is missing from the list, an appropriate article for it needs to be
| |
− | created and tagged accordingly.
| |
− |
| |
− | Missing a chip? please dump its name here: http://en.wikichip.org/wiki/WikiChip:wanted_chips
| |
− | -->
| |
− | {{comp table start}}
| |
− | <table class="comptable sortable tc18 tc19 tc20 tc21 tc22 tc23">
| |
− | <tr class="comptable-header"><th> </th><th colspan="25">List of all Cortex-A53 Chips</th></tr>
| |
− | <tr class="comptable-header"><th> </th><th colspan="10">Main processor</th><th colspan="3">IGP</th></tr>
| |
− | {{comp table header 1|cols=Launched, Designer, Family, Process, Core, C, T, L2$, L3$, Frequency, Max Mem, Designer, Name, Frequency}}
| |
− | {{#ask: [[Category:all microprocessor models]] [[microarchitecture::Cortex-A53]]
| |
− | |?full page name
| |
− | |?model number
| |
− | |?first launched
| |
− | |?designer
| |
− | |?microprocessor family
| |
− | |?process
| |
− | |?core name
| |
− | |?core count
| |
− | |?thread count
| |
− | |?l2$ size
| |
− | |?l3$ size
| |
− | |?base frequency#GHz
| |
− | |?max memory#GiB
| |
− | |?integrated gpu designer
| |
− | |?integrated gpu
| |
− | |?integrated gpu base frequency
| |
− | |format=template
| |
− | |template=proc table 3
| |
− | |searchlabel=
| |
− | |sort=microprocessor family, model number
| |
− | |order=asc,asc
| |
− | |userparam=15
| |
− | |mainlabel=-
| |
− | |limit=100
| |
− | |valuesep=,
| |
− | }}
| |
− | {{comp table count|ask=[[Category:all microprocessor models]] [[microarchitecture::Cortex-A53]]}}
| |
− | </table>
| |
− | {{comp table end}}
| |
− |
| |
− | == 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
| |
− | * Mair, Hugh T., et al. "4.3 A 20nm 2.5 GHz ultra-low-power tri-cluster CPU subsystem with adaptive power allocation for optimal mobile SoC performance." Solid-State Circuits Conference (ISSCC), 2016 IEEE International. IEEE, 2016.
| |