ARM big.LITTLE: Difference between revisions
Content deleted Content added
Ghostofakina (talk | contribs) |
Centauri24 (talk | contribs) m added hyperlink |
||
| (41 intermediate revisions by 30 users not shown) | |||
| Line 2: | Line 2: | ||
{{pp-move-dispute|small=yes}} |
{{pp-move-dispute|small=yes}} |
||
{{Use British English|date=September 2013}} |
{{Use British English|date=September 2013}} |
||
{{Use dmy dates|date= |
{{Use dmy dates|date=December 2021}} |
||
[[File:ARMCortexA57A53.jpg|thumb|Cortex A57/A53 MPCore big.LITTLE CPU chip]] |
[[File:ARMCortexA57A53.jpg|thumb|Cortex A57/A53 MPCore big.LITTLE CPU chip]] |
||
'''ARM big.LITTLE''' is a [[heterogeneous computing]] architecture developed by [[ |
'''ARM big.LITTLE''' is a [[heterogeneous computing]] [[Computer architecture|architecture]] developed by [[Arm Holdings]], coupling relatively battery-saving and slower processor cores (''LITTLE'') with relatively more powerful and power-hungry ones (''big''). The intention is to create a [[multi-core processor]] that can adjust better to dynamic computing needs and use less power than [[clock scaling]] alone. ARM's marketing material promises up to a 75% savings in power usage for some activities.<ref name="arm-mkt">{{Cite web |url=http://www.arm.com/products/processors/technologies/biglittleprocessing.php |title=big.LITTLE technology |publisher=ARM.com |access-date=17 October 2012 |archive-url=https://web.archive.org/web/20121022055646/http://www.arm.com/products/processors/technologies/bigLITTLEprocessing.php |archive-date=22 October 2012 |url-status=dead }}</ref> Most commonly, ARM big.LITTLE architectures are used to create a [[multi-processor system-on-chip]] (MPSoC). |
||
In October 2011, big.LITTLE was announced along with the [[ARM Cortex-A7 MPCore|Cortex-A7]], which was designed to be [[Instruction set architecture|architecturally]] compatible with the [[ARM Cortex-A15 MPCore|Cortex-A15]].<ref name="Auto4J-1"/> |
In October 2011, big.LITTLE was announced along with the [[ARM Cortex-A7 MPCore|Cortex-A7]], which was designed to be [[Instruction set architecture|architecturally]] compatible with the [[ARM Cortex-A15 MPCore|Cortex-A15]].<ref name="Auto4J-1"/> In October 2012 ARM announced the [[ARM Cortex-A53|Cortex-A53]] and [[ARM Cortex-A57|Cortex-A57]] ([[ARMv8|ARMv8-A]]) cores, which are also intercompatible to allow their use in a big.LITTLE chip.<ref name="cortex-a50 announce"/> ARM later announced the [[ARM Cortex-A12|Cortex-A12]] at [[Computex Taipei|Computex 2013]] followed by the [[ARM Cortex-A17|Cortex-A17]] in February 2014. Both the Cortex-A12 and the Cortex-A17 can also be paired in a big.LITTLE configuration with the Cortex-A7.<ref>{{cite web |url=https://www.theverge.com/2013/6/2/4390076/arm-cortex-a12-mali-t622-v500 |title=ARM's new Cortex-A12 is ready to power 2014's $200 midrange smartphones |date=April 2014 |work=The Verge}}</ref><ref>{{cite web |url=http://www.anandtech.com/show/7739/arm-cortex-a17 |title=ARM Cortex A17: An Evolved Cortex A12 for the Mainstream in 2015 |date=April 2014 |publisher=AnandTech}}</ref> |
||
== The |
== The problem that big.LITTLE solves == |
||
For a given library of [[CMOS]] logic, active power increases as the logic switches more per second, while leakage increases with the number of transistors. So, CPUs designed to run fast are different from CPUs designed to save power. When a very fast [[Out-of-order execution|out-of-order]] CPU is |
For a given library of [[CMOS]] logic, active power increases as the logic switches more per second, while leakage increases with the number of transistors. So, CPUs designed to run fast are different from CPUs designed to save power. When a very fast [[Out-of-order execution|out-of-order]] CPU is idling at very low speeds, a CPU with much less leakage (fewer transistors) could do the same work. For example, it might use a smaller (fewer transistors) [[Cache (computing)|memory cache]], or a simpler microarchitecture such as removing [[out-of-order execution]]. big.LITTLE is a way to optimize for both cases: Power and speed, in the same system. |
||
In practice, a big.LITTLE system can be surprisingly inflexible. One issue is the number and types of power and clock domains that the IC provides. These may not match the standard power management features offered by an operating system. Another is that the CPUs no longer have equivalent abilities, and matching the right software task to the right CPU becomes more difficult. Most of these problems are being solved by making the electronics and software more flexible. |
In practice, a big.LITTLE system can be surprisingly inflexible. One issue is the number and types of power and clock domains that the IC provides. These may not match the standard power management features offered by an operating system. Another is that the CPUs no longer have equivalent abilities, and matching the right software task to the right CPU becomes more difficult. Most of these problems are being solved by making the electronics and software more flexible. |
||
== Run-state migration== |
== Run-state migration== |
||
There are three ways<ref>{{cite web | url=http://blogs.arm.com/soc-design/1009-ten-things-to-know-about-biglittle/ | title=Ten Things to Know About big.LITTLE | author=Brian Jeff | date=18 June 2013 | publisher=[[ARM Holdings]] | |
There are three ways<ref>{{cite web | url=http://blogs.arm.com/soc-design/1009-ten-things-to-know-about-biglittle/ | title=Ten Things to Know About big.LITTLE | author=Brian Jeff | date=18 June 2013 | publisher=[[ARM Holdings]] | access-date=2013-09-17 | archive-url=https://web.archive.org/web/20130910163539/http://blogs.arm.com/soc-design/1009-ten-things-to-know-about-biglittle/ | archive-date=10 September 2013 | url-status=dead }}</ref> for the different processor cores to be arranged in a big.LITTLE design, depending on the [[Scheduler (computing)|scheduler]] implemented in the [[Kernel (operating system)|kernel]].<ref>{{cite web | url=http://www.linaro.org/linaro-blog/2013/07/10/big-little-software-update/ | title=big.LITTLE Software Update | author=George Grey | publisher=[[Linaro]] | date=10 July 2013 | access-date=2013-09-17 | archive-url=https://web.archive.org/web/20131004230806/http://www.linaro.org/linaro-blog/2013/07/10/big-little-software-update/ | archive-date=4 October 2013 | url-status=dead }}</ref> |
||
=== Clustered switching === |
=== Clustered switching === |
||
[[File:Big.Little Cluster Switching.png|thumb|upright=1.7|Big.Little clustered switching]] |
[[File:Big.Little Cluster Switching.png|thumb|upright=1.7|Big.Little clustered switching]] |
||
The clustered model approach is the first and simplest implementation, arranging the processor into identically |
The clustered model approach is the first and simplest implementation, arranging the processor into identically sized clusters of "big" or "LITTLE" cores. The operating system scheduler can only see one cluster at a time; when the [[Load (computing)|load]] on the whole processor changes between low and high, the system transitions to the other cluster. All relevant data are then passed through the common [[L2 cache]], the active core cluster is powered off and the other one is activated. A Cache Coherent Interconnect (CCI) is used. This model has been implemented in the [[Samsung]] [[Exynos]] 5 Octa (5410).<ref>{{cite web | url=http://www.embedded.com/electronics-news/4419448/Benchmarking-ARM-s-big-little-architecture | title=Benchmarking ARM's big-little architecture | author=Peter Clarke | date=6 August 2013 | access-date=2013-09-17}}</ref> |
||
{{Clear}} |
{{Clear}} |
||
| Line 35: | Line 35: | ||
[[File:Global Task Scheduling.jpg|thumb|upright=1.7|Big.Little heterogeneous multi-processing]] |
[[File:Global Task Scheduling.jpg|thumb|upright=1.7|Big.Little heterogeneous multi-processing]] |
||
The most powerful use model of |
The most powerful use model of big.LITTLE architecture is [[Heterogeneous computing|heterogeneous]] [[Multiprocessing|multi-processing]] (HMP), which enables the use of all physical cores at the same time. [[Thread (computing)|Threads]] with [[Scheduling (computing)#Priority scheduling|high priority]] or computational intensity can in this case be allocated to the "big" cores while threads with less priority or less computational intensity, such as background tasks, can be performed by the "LITTLE" cores.<ref>{{citation|url=http://www.arm.com/files/downloads/big.LITTLE_Final.pdf |title=Big.LITTLE Processing with ARM Cortex-A15 & Cortex-A7 |date=September 2013 |publisher=[[ARM Holdings]] |access-date=2013-09-17 |url-status=dead |archive-url=https://web.archive.org/web/20120417183714/http://www.arm.com/files/downloads/big.LITTLE_Final.pdf |archive-date=17 April 2012 }}</ref> |
||
This model has been implemented in the [[Samsung]] [[Exynos]] starting with the Exynos 5 Octa series (5420, 5422, 5430),<ref name="anand-5420"/><ref name="Samsung Tomorrow">{{cite web | |
This model has been implemented in the [[Samsung]] [[Exynos]] starting with the Exynos 5 Octa series (5420, 5422, 5430),<ref name="anand-5420"/><ref name="Samsung Tomorrow">{{cite web |title =Samsung Unveils New Products from its System LSI Business at Mobile World Congress |publisher =Samsung Tomorrow |url =http://global.samsungtomorrow.com/?p=34630 |access-date =26 February 2013 |archive-date =16 March 2014 |archive-url =https://web.archive.org/web/20140316044700/http://global.samsungtomorrow.com/?p=34630 |url-status =dead }}</ref> and [[Apple A series]] processors starting with the [[Apple A11]].<ref>{{Cite news|url=https://www.apple.com/newsroom/2017/09/the-future-is-here-iphone-x/|title=The future is here: iPhone X|work=Apple Newsroom|access-date=2018-02-25|language=en-US}}</ref> |
||
{{Clear}} |
{{Clear}} |
||
==Scheduling== |
==Scheduling== |
||
The paired arrangement allows for switching to be done transparently to the [[operating system]] using the existing [[Dynamic voltage scaling|dynamic voltage]] and [[Dynamic frequency scaling|frequency scaling]] (DVFS) facility. The existing DVFS support in the kernel (e.g. <code>cpufreq</code> in Linux) will simply see a list of frequencies/voltages and will switch between them as it sees fit, just like it does on the existing hardware. However, the low-end slots will activate the 'Little' core and the high-end slots will activate the 'Big' core. |
The paired arrangement allows for switching to be done transparently to the [[operating system]] using the existing [[Dynamic voltage scaling|dynamic voltage]] and [[Dynamic frequency scaling|frequency scaling]] (DVFS) facility. The existing DVFS support in the kernel (e.g. <code>cpufreq</code> in Linux) will simply see a list of frequencies/voltages and will switch between them as it sees fit, just like it does on the existing hardware. However, the low-end slots will activate the 'Little' core and the high-end slots will activate the 'Big' core. This is the early solution provided by Linux's "deadline" CPU scheduler (not to be confused with the I/O scheduler with the same name) since 2012.<ref>{{cite web |last1=McKenney |first1=Paul |title=A big.LITTLE scheduler update |url=https://lwn.net/Articles/501501/ |website=LWN.net |date=12 June 2012}}</ref> |
||
Alternatively, all the cores may be exposed to the [[Scheduler (computing)|kernel scheduler]], which will decide where each process/thread is executed. This will be required for the non-paired arrangement but could possibly also be used on the paired cores. It poses unique problems for the kernel scheduler, which, at least with modern commodity hardware, has been able to assume all cores in a [[Symmetric multiprocessing|SMP]] system are equal rather than |
Alternatively, all the cores may be exposed to the [[Scheduler (computing)|kernel scheduler]], which will decide where each process/thread is executed. This will be required for the non-paired arrangement but could possibly also be used on the paired cores. It poses unique problems for the kernel scheduler, which, at least with modern commodity hardware, has been able to assume all cores in a [[Symmetric multiprocessing|SMP]] system are equal rather than heterogeneous. A 2019 addition to Linux 5.0 called ''Energy Aware Scheduling'' is an example of a scheduler that considers cores differently.<ref>{{cite web |last1=Perret |first1=Quentin |title=Energy Aware Scheduling merged in Linux 5.0 |url=https://community.arm.com/developer/ip-products/processors/b/processors-ip-blog/posts/energy-aware-scheduling-in-linux |website=community.arm.com |language=en |date=25 February 2019}}</ref><ref>{{cite web |title=Energy Aware Scheduling |url=https://www.kernel.org/doc/html/latest/scheduler/sched-energy.html |website=The Linux Kernel documentation}}</ref> |
||
==Advantages of global task scheduling== |
==Advantages of global task scheduling== |
||
| Line 51: | Line 51: | ||
* The ability to easily support non-symmetrical clusters (e.g. with 2 Cortex-A15 cores and 4 Cortex-A7 cores). |
* The ability to easily support non-symmetrical clusters (e.g. with 2 Cortex-A15 cores and 4 Cortex-A7 cores). |
||
* The ability to use all cores simultaneously to provide improved peak performance throughput of the SoC compared to IKS. |
* The ability to use all cores simultaneously to provide improved peak performance throughput of the SoC compared to IKS. |
||
==Implementations== |
|||
{| class="wikitable" |
|||
|- |
|||
! SoC !! [[Semiconductor device fabrication|Fabrication]] !! Big cores !! Medium cores !! Little cores !! GPU !! Memory interface !! Wireless radio technologies !! Availability !! Devices |
|||
|- |
|||
| [[HiSilicon#K3V3|HiSilicon K3V3]] |
|||
| 28 nm |
|||
| 1.8 GHz [[Multi-core processor|dual-core]] [[ARM Cortex-A15 MPCore|Cortex-A15]] |
|||
|{{N/A}} |
|||
| 1.2 GHz dual-core [[ARM Cortex-A7 MPCore|Cortex-A7]] |
|||
| [[Mali (GPU)|Mali]]-T658 |
|||
| |
|||
| |
|||
| H2 2013 |
|||
| |
|||
|- |
|||
| HiSilicon Kirin 710 |
|||
| 12 nm FinFET |
|||
| 2.2 GHz quad-core Cortex-A73 |
|||
|{{N/A}} |
|||
| 1.7 GHz quad-core Cortex-A53 |
|||
| Mali-G51 MP4 |
|||
| |
|||
| rowspan="2"|LTE Cat.12 (600 Mbit/s) |
|||
| Q3 2018 |
|||
| {{Collapsible list| Huawei Nova 3i| Huawei Nova 4e| Huawei Nova 5i| Huawei P20 Lite| Huawei P30 Lite| Huawei Mate 20 Lite| Honor 8x| Honor 10i| Honor 10 Lite| Honor 20i| Honor 20 Lite}} |
|||
|- |
|||
| HiSilicon Kirin 810 |
|||
| 7 nm FinFET |
|||
| 2.2 GHz dual-core Cortex-A76 |
|||
|{{N/A}} |
|||
| 1.9 GHz hexa-core Cortex-A55 |
|||
| Mali-G52 MP6 |
|||
| LPDDR4X |
|||
| Q2 2019 |
|||
| {{Collapsible list| Huawei Nova 5| Huawei Nova 5z| Huawei Nova 5i Pro| Huawei Nova 6 SE| Honor 9x| Honor 9x Pro| Honor 20S}} |
|||
|- |
|||
| HiSilicon Kirin 920 |
|||
| 28 nm |
|||
| 1.7-2.0 GHz quad-core Cortex-A15 |
|||
|{{N/A}} |
|||
| 1.3-1.6 GHz quad-core Cortex-A7 |
|||
| Mali-T628 MP4 |
|||
| LPDDR3 |
|||
| LTE Cat 6 |
|||
| Q3 2014 |
|||
| Huawei Honor 6 |
|||
|- |
|||
| HiSilicon Kirin 950/955 |
|||
| 16 nm FinFET+ |
|||
| 2.3-2.5 GHz quad-core ARM Cortex-A72 |
|||
|{{N/A}} |
|||
| 1.8 GHz quad-core ARM Cortex-A53 |
|||
| Mali-T880 MP4 |
|||
| LPDDR4 |
|||
| LTE Cat 6 |
|||
| Q4 2015 (Kirin 950) |
|||
Q2 2016 (Kirin 955) |
|||
| Huawei Mate 8, Huawei P9, Huawei Honor 8 |
|||
|- |
|||
| HiSilicon Kirin 960 |
|||
| 16 nm FinFET Compact |
|||
| 2.36 GHz quad-core ARM Cortex-A73 |
|||
|{{N/A}} |
|||
| 1.84 GHz quad-core ARM Cortex-A53 |
|||
| Mali-G71 MP8 |
|||
| LPDDR4-1600 Dual-Channel 64-Bit (28.8GB/s) |
|||
| LTE Cat 12/13 |
|||
| Q4 2016 |
|||
| {{Collapsible list| Huawei Mate 9| Huawei Mate 9 Pro| Huawei Mate 9 Porsche Design| Huawei P10| Huawei P10 Plus| Honor 8 Pro| Honor V9| Honor 9| Huawei MediaPad M5| Huawei Mediapad M5 Pro}} |
|||
|- |
|||
| Hisilicon Kirin 970 |
|||
| 10 nm FinFET+ |
|||
| 2.36 GHz Quad-Core ARM Cortex-A73 |
|||
|{{N/A}} |
|||
| 1.84 GHz Quad-Core ARM Cortex-A53 |
|||
| Mali-G72 MP12 @746MHz |
|||
|LPDDR4X-1866 Quad-Channel 64-Bit (29.8GB/s) |
|||
|LTE Cat.18/13 |
|||
|Q4 2017 |
|||
| {{Collapsible list|Huawei Mate 10| Huawei Mate 10 Pro| Huawei Mate 10 Porsche Design| Huawei P20| Huawei P20 Pro| Huawei Mate RS| Huawei Nova 3| Huawei Nova 4| Honor 10| Honor V10/View 10| Honor Note 10| Honor Play}} |
|||
|- |
|||
|Hisilicon Kirin 980 |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|7nm FinFET |
|||
|2.6 GHz Dual-Core ARM Cortex-A76 |
|||
|1.92 GHz Dual-Core ARM Cortex-A76 |
|||
|1.8 GHz Quad-Core ARM Cortex-A55 |
|||
|Mali-G76 MP10 @720MHz |
|||
|LPDDR4X-2133 Quad-Channel 64-Bit (34.1GB/s) |
|||
|LTE Cat.21/13 |
|||
|Q4 2018 |
|||
| {{Collapsible list|[[Huawei Mate 20]]| Huawei Mate 20 Pro| Huawei Mate 20 X| Huawei Mate 20 RS Porsche Design| [[Huawei P30]]| Huawei P30 Pro| Huawei Mate 20 X (5G)| [[Huawei Mate X]]| Huawei Nova 5 Pro| Huawei Nova 5T| [[Honor 20]]| Honor 20 Pro| Honor View 20/V20| Honor Magic 2/3D| Huawei MediaPad M6 (8.4")| Huawei MediaPad M6 (10.8")}} |
|||
|- |
|||
|Hisilicon Kirin 990 |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|7nm FinFET |
|||
|2.86 GHz Dual-Core Cortex-A76 |
|||
|2.09 GHz Dual-Core Cortex-A76 |
|||
|1.86 GHz Dual-Core Cortex-A55 |
|||
|Mali-G76 MP16 @600MHz |
|||
|LPDDR4X-2133 Quad-Channel 64-Bit (34.1GB/s) |
|||
|Balong 765 (Cat.19/13) |
|||
|Q4 2019 |
|||
| {{Collapsible list|[[Huawei Mate 30]]| Huawei Mate 30 Pro| Huawei Nova 6| Huawei Nova 6 5G| Honor V30| Honor V30 Pro}} |
|||
|- |
|||
|Hisilicon Kirin 990 5G |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|7nm+ FinFET EUV |
|||
|2.86 GHz Dual-Core Cortex-A76 |
|||
|2.36 GHz Dual-Core Cortex-A76 |
|||
|1.95 GHz Dual-Core Cortex-A55 |
|||
|Mali-G76 MP16 @600MHz |
|||
|LPDDR4X-2133 Quad-Channel 64-Bit (34.1GB/s) |
|||
|Balong 5000 (Sub-6GHz Only) |
|||
|Q4 2019 |
|||
| {{Collapsible list|Huawei Mate 30 5G| Huawei Mate 30 Pro 5G| Huawei Nova 6 5G| Honor V30 5G| Honor V30 Pro 5G| Huawei MatePad Pro}} |
|||
|- |
|||
| Samsung [[Exynos (system on chip)|Exynos]] 5 Octa (5410 model)<ref>{{cite web | url=https://arstechnica.com/gadgets/2013/01/samsungs-new-eight-core-exynos-5-octa-soc-promises-efficiency/ | title=Samsung’s new eight-core Exynos 5 Octa SoC promises not to hog battery | publisher=[[Ars Technica]] | date=10 January 2013 | author=Andrew Cunningham | accessdate=2013-01-10}}</ref><ref>{{cite web | url=https://www.engadget.com/2013/01/09/samsung-announces-exynos-5-octa-chip-at-ces/ | title=Samsung announces eight-core Exynos 5 'Octa' chip at CES | publisher=[[Engadget]] | date=9 January 2013 | author=James Trew | accessdate=2013-01-10}}</ref> |
|||
| 28 nm |
|||
| 1.6-1.8 GHz [[Multi-core processor|quad-core]] Cortex-A15 |
|||
|{{N/A}} |
|||
| 1.2 GHz quad-core Cortex-A7 |
|||
| PowerVR SGX544MP3 |
|||
| 32-bit dual-channel 800 MHz LPDDR3 (12.8 GB/sec) |
|||
| |
|||
| Q2 2013 |
|||
| Exynos 5-based [[Samsung Galaxy S4]] |
|||
|- |
|||
| Samsung Exynos 5 Octa (5420 model)<ref>{{cite press | url=http://www.samsung.com/global/business/semiconductor/minisite/Exynos/news_Samsung_Primes_Exynos5Octa_for_ARM_bigLITTLE_Technology_with_Heterogeneous_Multi_Processing_Capability.html?ism=SASep0913Twitter2 | title=Samsung Primes Exynos 5 Octa for ARM big.LITTLE Technology with Heterogeneous Multi-Processing Capability | publisher=[[Samsung Electronics]] | date=10 September 2013 | accessdate=2013-09-17}}</ref> |
|||
| 28 nm |
|||
| 1.8-2.0 GHz quad-core Cortex-A15 |
|||
|{{N/A}} |
|||
| 1.3 GHz quad-core Cortex-A7 |
|||
| Mali-T628 MP6 |
|||
| 32-bit dual-channel 933 MHz LPDDR3e (14.9 GB/sec) |
|||
| |
|||
| Q4 2013 |
|||
| Exynos 5-based [[Samsung Galaxy Note 3]] |
|||
|- |
|||
| Samsung Exynos 5 Octa (5422 model)<ref name="Samsung Tomorrow"/> |
|||
| 28 nm |
|||
| 2.1 GHz quad-core Cortex-A15 |
|||
|{{N/A}} |
|||
| 1.5 GHz quad-core Cortex-A7 |
|||
| Mali-T628 MP6 |
|||
| 32-bit dual-channel 933 MHz LPDDR3e (14.9 GB/sec) |
|||
| |
|||
| Q2 2014 |
|||
| Exynos 5-based [[Samsung Galaxy S5]], [[Odroid]]-XU3, Odroid-XU4 |
|||
|- |
|||
| Samsung Exynos 5 Hexa (5260 model)<ref name="Samsung Tomorrow"/> |
|||
| 28 nm |
|||
| 1.7 GHz dual-core Cortex-A15 |
|||
|{{N/A}} |
|||
| 1.3 GHz quad-core Cortex-A7 |
|||
| Mali-T624 |
|||
| 32-bit dual-channel 800 MHz LPDDR3e (12.8 GB/sec) |
|||
| |
|||
| Q2 2014 |
|||
| Samsung Galaxy Note 3 Neo |
|||
|- |
|||
| Samsung Exynos 5 Octa (5430 model)<ref>{{cite web | title = |
|||
Samsung Announces Exynos 5430: First 20nm Samsung SoC |publisher =AnandTech |url =http://anandtech.com/show/8382/samsung-announces-exynos-5430-first-20nm-samsung-soc | accessdate = 14 August 2014}}</ref> |
|||
| 20 nm |
|||
| 1.8 GHz quad-core Cortex-A15 |
|||
|{{N/A}} |
|||
| 1.3 GHz quad-core Cortex-A7 |
|||
| Mali-T628 MP6 |
|||
| 32-bit dual-channel 1066 MHz LPDDR3e (17.0 GB/sec) |
|||
| LTE Cat 6 |
|||
| Q3 2014 |
|||
| Samsung Galaxy Alpha<ref>{{cite web | title =Samsung Introduces Galaxy Alpha, the evolution of Galaxy Design |publisher =Samsung Tomorrow |url =http://global.samsungtomorrow.com/?p=39563 | accessdate = 14 August 2014}}</ref> |
|||
|- |
|||
| Samsung Exynos 7 Octa (5433 model)<ref>{{cite web | title = Samsung's Exynos 5433 is an A57/A53 ARM SoC |publisher =AnandTech |url =http://www.anandtech.com/show/8537/samsungs-exynos-5433-is-an-a57a53-arm-soc | accessdate = 17 September 2014}}</ref> |
|||
| 20 nm |
|||
| 1.9 GHz quad-core [[ARM Cortex-A57|Cortex-A57]] |
|||
|{{N/A}} |
|||
| 1.3 GHz quad-core Cortex-A53 |
|||
| Mali-T760 MP6 |
|||
| 32-bit dual-channel 825 MHz LPDDR3e (13.2 GB/sec) |
|||
| LTE Cat 6 |
|||
| Q4 2014 |
|||
| [[Samsung Galaxy Note 4]] (SM-N910C) |
|||
|- |
|||
| Samsung Exynos 7 Octa (7420 model)<ref>{{cite web | url =http://www.anandtech.com/show/8999/samsung-announces-the-galaxy-s6 |title=Samsung Announces the Galaxy S 6 and S 6 Edge | date=1 March 2015 | accessdate = 1 March 2015}}</ref> |
|||
| 14 nm |
|||
| 2.1 GHz quad-core Cortex-A57 |
|||
|{{N/A}} |
|||
| 1.5 GHz quad-core Cortex-A53 |
|||
| Mali-T760 MP8 |
|||
| LPDDR4 |
|||
| LTE Cat 9 |
|||
| Q2 2015 |
|||
| [[Samsung Galaxy S6]], [[Samsung Galaxy S6 Edge]], [[Samsung Galaxy Note 5]], Meizu PRO 5 |
|||
|- |
|||
| Samsung Exynos 7 Octa (7580 model) |
|||
| 28 nm HKMG |
|||
| 1.5 GHz quad-core Cortex-A53 |
|||
|{{N/A}} |
|||
| 1.5 GHz quad-core Cortex-A53 |
|||
| Mali-T720 MP2 |
|||
| LPDDR3 |
|||
| LTE Cat 6 |
|||
| Q2 2015 |
|||
| Samsung Galaxy J7, Samsung Galaxy S5 Neo, Samsung Galaxy A5/A7 (2016) |
|||
|- |
|||
| Samsung Exynos 7 Hexa (7650 model) |
|||
| 28 nm HKMG |
|||
| 1.7 GHz dual-core Cortex-A72 |
|||
|{{N/A}} |
|||
| 1.3 GHz quad-core Cortex-A53 |
|||
| Mali-T820 MP3 |
|||
| LPDDR3 |
|||
| LTE Cat 6 |
|||
| Q1 2016 |
|||
| |
|||
|- |
|||
| Samsung Exynos 7 Octa (7870 model) |
|||
| 14 nm LPP |
|||
| 1.7 GHz quad-core Cortex-A53 |
|||
|{{N/A}} |
|||
| 1.7 GHz quad-core Cortex-A53 |
|||
| Mali-T830 MP2 |
|||
| LPDDR3 |
|||
| LTE Cat 6 |
|||
| Q2 2016 |
|||
| {{Collapsible list|Samsung Galaxy Tab A 10.1 (2016)| Samsung Galaxy J7 (2016)| Samsung Galaxy J7 Prime| Samsung Galaxy J5 (2017)| Samsung Galaxy J7 (2017)| Samsung Galaxy J6 (2018)| [[Samsung Galaxy M10]]}} |
|||
|- |
|||
| Samsung Exynos 7 Octa (7880 model) |
|||
|14 nm LPP |
|||
| 1.9 GHz quad-core Cortex-A53 |
|||
|{{N/A}} |
|||
| 1.9 GHz quad-core Cortex-A53 |
|||
| Mali-T830 MP3 |
|||
| LPDDR3 |
|||
| LTE Cat 7 |
|||
| Q2 2016 |
|||
| Samsung Galaxy A5 (2017), Samsung Galaxy A7 (2017) |
|||
|- |
|||
| Samsung Exynos 7 Octa (7884 model) |
|||
|14 nm LPP |
|||
| 1.6 GHz dual-core Cortex-A73 |
|||
|{{N/A}} |
|||
| 1.35 GHz hexa-core Cortex-A53 |
|||
| Mali-G71 MP2 |
|||
| LPDDR4 |
|||
|Downlink: LTE Cat 12, Uplink: LTE Cat 13 |
|||
| Q2 2018 |
|||
| {{Collapsible list|Samsung Galaxy A10e| [[Samsung Galaxy A10]]| Samsung Galaxy A20e| [[Samsung Galaxy A20]]| Samsung Galaxy M10s}} |
|||
|- |
|||
| Samsung Exynos 7 Octa (7885 model) |
|||
|14 nm LPP |
|||
|2.2 GHz dual-core Cortex-A73 |
|||
|{{N/A}} |
|||
|1.6 GHz hexa-core Cortex-A53 |
|||
|Mali-G71 MP2 |
|||
|LPDDR4 |
|||
|Downlink: LTE Cat 12, Uplink: LTE Cat 13 |
|||
|Q1 2018 |
|||
|Samsung Galaxy A8 (2018), Samsung Galaxy A8+ (2018) |
|||
|- |
|||
| Samsung Exynos 7 Octa (7904 model) |
|||
| 14 nm LPP |
|||
|1.8 GHz dual-core Cortex-A73 |
|||
|{{N/A}} |
|||
|1.6 GHz hexa-core Cortex-A53 |
|||
|Mali-G71 MP2 |
|||
|LPDDR4 |
|||
|Downlink: LTE Cat 12, Uplink: LTE Cat 13 |
|||
| Q1 2019 |
|||
| {{Collapsible list|[[Samsung Galaxy A30]]| Samsung Galaxy A30s| [[Samsung Galaxy A40]]| [[Samsung Galaxy M20]]| [[Samsung Galaxy M30]]/A40s| Samsung Galaxy Tab A 10.1 (2019)}} |
|||
|- |
|||
| Samsung Exynos 8 Octa (8890 model) |
|||
| 14 nm LPP |
|||
| 2.6 GHz quad-core M1 "Mongoose" |
|||
|{{N/A}} |
|||
| 1.6 GHz quad-core Cortex-A53 |
|||
| Mali-T880 MP12 |
|||
| LPDDR4 |
|||
| Downlink: LTE Cat 12, Uplink: LTE Cat 13 |
|||
| Q1 2016 |
|||
| [[Samsung Galaxy S7]] (930F/FD), [[Samsung Galaxy S7 Edge]] (935F/FD), [[Samsung Galaxy Note 7]] (N930F/FD/G) |
|||
|- |
|||
| Samsung Exynos 9 Octa (8895 model) |
|||
| 10 nm FinFET LPE |
|||
| 2.3 GHz Quad-Core Exynos M2 "Mongoose" |
|||
|{{N/A}} |
|||
| 1.7 GHz Quad-Core Cortex-A53 |
|||
| Mali-G71 MP20 @546MHz |
|||
| LPDDR4X-1794 Dual-Channel 64-Bit (28.7GB/s) |
|||
| Shannon 355 LTE Downlink: LTE Cat 16, |
|||
Uplink: LTE Cat 13 |
|||
| Q1 2017 |
|||
| [[Samsung Galaxy S8]] (950F/FD), [[Samsung Galaxy S8+]] (955F/FD), [[Samsung Galaxy Note 8]] (N950F/FD) |
|||
|- |
|||
|Samsung Exynos 9 Octa (9609 Model) |
|||
| rowspan="3"|10 nm FinFET LPE |
|||
| 2.2 GHz Quad-core Cortex-A73 |
|||
|{{N/A}} |
|||
| 1.6 GHz Quad-core Cortex-A53 |
|||
| rowspan="3"|Mali-G72 MP3 |
|||
| rowspan="3"|LPDDR4X 64-bit (2×32-bit) Dual-channel |
|||
| rowspan="3"|Cat.12 3CA 600Mbit/s (DL) / |
|||
Cat.13 2CA 150Mbit/s (UL) |
|||
| Q2 2019 |
|||
|[[Motorola One]] Vision, Motorola One Action |
|||
|- |
|||
|Samsung Exynos 9 Octa (9610 Model) |
|||
| rowspan="2"|2.3 GHz Quad-core Cortex-A73 |
|||
|{{N/A}} |
|||
| rowspan="2"|1.7 GHz Quad-core Cortex-A53 |
|||
| Q4 2018 |
|||
|[[Samsung Galaxy A50]] |
|||
|- |
|||
|Samsung Exynos 9 Octa (9611 Model) |
|||
|{{N/A}} |
|||
| Q3 2019 |
|||
|Samsung Galaxy A50s, Samsung Galaxy A51, [[Samsung Galaxy M30s]], Samsung Galaxy Xcover Pro |
|||
|- |
|||
|Samsung Exynos 9 Octa (9810 Model) |
|||
|10nm FinFET LPP |
|||
|2.9 GHz Quad-Core Exynos M3 "Meerkat" |
|||
|{{N/A}} |
|||
|1.9 GHz Quad-Core Cortex-A55 |
|||
|Mali-G72 MP18 @572MHz |
|||
|LPDDR4X-1794 Quad-Channel 64-Bit (28.7GB/s) |
|||
|Shannon 360 LTE Downlink: LTE Cat.18, |
|||
Uplink: LTE Cat.13 |
|||
|Q1 2018 |
|||
|[[Samsung Galaxy S9]] (SM-960F/DS), [[Samsung Galaxy S9+]] (SM-965F/DS), [[Samsung Galaxy Note 9]] (SM-N960F/DS), Samsung Galaxy Note 10 Lite (SM-N770F/DS) |
|||
|- |
|||
|Samsung Exynos 9 Octa (9820 Model) |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|8nm LPP |
|||
|2.73 GHz Dual-Core Exynos M4 "Cheetah" |
|||
|2.31 GHz Dual-Core Cortex-A75 |
|||
|1.95 GHz Quad-Core Cortex-A55 |
|||
|Mali-G76 MP12 @702MHz |
|||
|LPDDR4X-2093 Quad-Channel 64-Bit (33.488GB/s) |
|||
|Shannon 5000 LTE Downlink: LTE Cat.20, |
|||
Uplink: LTE Cat.13 |
|||
|Q1 2019 |
|||
|[[Samsung Galaxy S10]] (SM-G973F/DS), [[Samsung Galaxy S10+]] (SM-G975F/DS), [[Samsung Galaxy S10e]] (SM-G970F/DS) |
|||
|- |
|||
|Samsung Exynos 9 Octa (9825 Model) |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|7nm+ FinFET EUV |
|||
|2.73 GHz Dual-Core Exynos M4 "Cheetah" |
|||
|2.4 GHz Dual-Core Cortex-A75 |
|||
|1.95 Ghz Quad-Core Cortex-A55 |
|||
|Mali-G76 MP12 |
|||
|LPDDR4X-2093 Quad-Channel 64-Bit (33.488GB/s) |
|||
|Shannon 5000 LTE Downlink: LTE Cat.20, |
|||
Uplink: LTE Cat.13 |
|||
|Q3 2019 |
|||
|[[Samsung Galaxy Note 10]] (SM-N970F/DS), [[Samsung Galaxy Note 10 5G]] (SM-N971F/DS), [[Samsung Galaxy Note 10+]] (SM-N975F/DS), [[Samsung Galaxy Note 10+ 5G]] (SM-N976F/DS) |
|||
|- |
|||
|Exynos 9 Octa (980 Model) |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|8nm FinFET LPP |
|||
|2.2 GHz Dual-Core Cortex-A77 |
|||
|{{N/A}} |
|||
|1.8 GHz Hexa-Core Cortex-A55 |
|||
|Mali-G76 MP5 |
|||
|LPDDR4X-2093 Quad-Channel 64-Bit (33.488GB/s) |
|||
|Shannon 5G |
|||
LTE Downlink: LTE Cat.16 |
|||
LTE Uplink: LTE Cat.18 |
|||
NR: Sub-6GHz |
|||
|Q3 2019 |
|||
|Vivo X30/X30 Pro |
|||
|- |
|||
|Exynos 9 Octa (990 Model) |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|7nm+ FinFET EUV |
|||
|Dual-Core "Exynos M5" |
|||
|Dual-Core Cortex-A76 |
|||
|Quad-Core Cortex-A55 |
|||
|Mali-G77 MP11 |
|||
|LPDDR5-2750 Quad-Channel 64-Bit (44GB/s) |
|||
|External: Exynos 5123 Modem NSA/SA |
|||
LTE Downlink: LTE Cat.24 |
|||
LTE Uplink: LTE Cat.22 |
|||
NR: Sub-6GHz, mmWave |
|||
|Q4 2019 |
|||
|[[Samsung Galaxy S20]] |
|||
|- |
|||
| [[Renesas Electronics|Renesas]] Mobile MP6530<ref>{{citation | url=http://renesasmobile.com/share/products/datasheets/renesasmobile-datasheet-mp6530.pdf | title=MP6530 | publisher=[[Renesas Mobile]] | date=December 2012 | accessdate=2013-09-17}}</ref> |
|||
| 28 nm |
|||
| 2.0 GHz dual-core Cortex-A15 |
|||
|{{N/A}} |
|||
| 1.0 GHz dual-core Cortex-A7 |
|||
| PowerVR SGX544 |
|||
| Dual-channel LPDDR3 |
|||
| LTE Cat 4 |
|||
| |
|||
| |
|||
|- |
|||
| [[Allwinner]] A80 Octa<ref>{{cite web |url=http://withimagination.imgtec.com/powervr/allwinner-ultraocta-a80-processor-packs-powervr-series6-gpu-64-cores |title=Allwinner UltraOcta A80 processor packs a PowerVR Series6 GPU with 64 cores |date=March 2014 |publisher=Imagination |access-date=19 March 2014 |archive-url=https://web.archive.org/web/20140903052616/http://withimagination.imgtec.com/powervr/allwinner-ultraocta-a80-processor-packs-powervr-series6-gpu-64-cores |archive-date=3 September 2014 |url-status=dead |df=dmy-all }}</ref> |
|||
| 28 nm |
|||
| Quad-core Cortex-A15 |
|||
|{{N/A}} |
|||
| Quad-core Cortex-A7 |
|||
| [[PowerVR]] G6230 |
|||
| Dual-channel DDR3/DDR3L/LPDDR3 or LPDDR2<ref>{{cite web |url=http://allwinnertech.com/en/clq/processora/2014/0223/261.html |title=A80 |date=May 2014 |publisher=Allwinner |access-date=1 May 2014 |archive-url=https://web.archive.org/web/20140502043230/http://allwinnertech.com/en/clq/processora/2014/0223/261.html |archive-date=2 May 2014 |url-status=dead |df=dmy-all }}</ref> |
|||
| |
|||
| |
|||
| |
|||
|- |
|||
| [[MediaTek]] MT6595<ref>{{cite web|url=http://www.mediatek.com/en/products/mobile-communications/mobile-chipsets/smartphone/mt6595 |title=MT6595 Octa-core LTE platform |date=April 2014 |publisher=MediaTek |url-status=dead |archiveurl=https://web.archive.org/web/20140415214822/http://www.mediatek.com/en/products/mobile-communications/mobile-chipsets/smartphone/mt6595/ |archivedate=15 April 2014 }}</ref> |
|||
| 28 nm |
|||
| 2.2 GHz quad-core Cortex-A17 |
|||
|{{N/A}} |
|||
| 1.7 GHz quad-core Cortex-A7 |
|||
| PowerVR G6200 (600 MHz) |
|||
| 32-bit dual-channel 933 MHz LPDDR3 (14.9 GB/sec) |
|||
| LTE Cat 4 |
|||
| Q2 2014 |
|||
| |
|||
|- |
|||
| MediaTek MT6595M |
|||
| 28 nm |
|||
| 2.0 GHz quad-core Cortex-A17 |
|||
|{{N/A}} |
|||
| 1.5 GHz quad-core Cortex-A7 |
|||
| PowerVR G6200 (450 MHz) |
|||
| 32-bit dual-channel 933 MHz LPDDR3 (14.9 GB/sec) |
|||
| LTE Cat 4 |
|||
| Q2 2014 |
|||
| |
|||
|- |
|||
| MediaTek MT6595 Turbo |
|||
| 28 nm |
|||
| 2.5 GHz quad-core Cortex-A17 |
|||
|{{N/A}} |
|||
| 1.7 GHz quad-core Cortex-A7 |
|||
| PowerVR G6200 (600 MHz) |
|||
| 32-bit dual-channel 933 MHz LPDDR3 (14.9 GB/sec) |
|||
| LTE Cat 4 |
|||
| Q3 2014 |
|||
| |
|||
|- |
|||
|MediaTek Helio P20 (MT6757) |
|||
|rowspan="4"|16nm FFC |
|||
| 2.3 GHz quad-core Cortex-A53 |
|||
|{{N/A}} |
|||
|rowspan="2"|1.6 GHz quad-core Cortex-A53 |
|||
|Mali-T880 MP2 @ 900 MHz |
|||
|rowspan="4"|16-bit dual-channel 1600 MHz LPDDR4x (12.8 GB/sec) |
|||
|rowspan="2"|Cat-6 |
|||
|Q3 2016 |
|||
| {{Collapsible list|Sony Xperia XA1/XA1 Plus/XA1 Ultra|Samsung Galaxy C7 (2017)|Samsung Galaxy J7 Max|Meizu E2|Meizu X|Elephone Z1|Alcatel Idol 5S|Infinix Hot 7 Pro}} |
|||
|- |
|||
|MediaTek Helio P25 (MT6757CD) |
|||
| 2.6 GHz quad-core Cortex-A53 |
|||
|{{N/A}} |
|||
|Mali-T880 MP2 @ 1 GHz |
|||
|Q2 2017 |
|||
| {{Collapsible list|Meizu M10|Meizu Pro 7|Blackview A80 Pro|Gionee S10|Doogee S60|Ulefone F1}} |
|||
|- |
|||
|MediaTek Helio P23 |
|||
| 2.3/2.5 GHz quad-core Cortex-A53 |
|||
|{{N/A}} |
|||
| rowspan="2"|1.65 GHz quad-core Cortex-A53 |
|||
|Mali-G71 MP2 @ 770 MHz |
|||
|Cat-6, Cat-7 DL / Cat-13 UL |
|||
|rowspan="2"|Q3 2017 |
|||
| {{Collapsible list|Oppo A83|Gionee S11|Blackview Max 1|Doogee S80|Ulefone Armor 3T|Vivo Y75|Elephone U|Alcatel 7}} |
|||
|- |
|||
|MediaTek Helio P30 |
|||
| 2.3 GHz quad-core Cortex-A53 |
|||
|{{N/A}} |
|||
|Mali-G71 MP2 @ 950 MHz + VPU |
|||
|Cat-7 DL / Cat-13 UL |
|||
|Gionee M7 |
|||
|- |
|||
|MediaTek Helio P60 (MT6771) |
|||
| rowspan="3"|12nm HPM |
|||
| 2.0 GHz quad-core Cortex-A73 |
|||
|{{N/A}} |
|||
| 2.0 GHz quad-core Cortex-A53 |
|||
| Mali-G72 MP3 @ 800 MHz |
|||
| Dual-channel LPDDR4x @ 1800 MHz |
|||
| Cat-7 (DL) / Cat-13 (UL) |
|||
| Q1 2018 |
|||
| {{Collapsible list|Realme 1|Realme 3|ZTE Blade A7|Oppo A3|Oppo F7|Nokia 5.1 Plus|Nokia X5|Vivo V11i|Vivo Z3i}} |
|||
|- |
|||
|MediaTek Helio P65 (MT6768) |
|||
| 2.0 GHz dual-core Cortex-A75 |
|||
|{{N/A}} |
|||
| 2.0 GHz hexa-core Cortex-A55 |
|||
| Mali-G52 MC2 @ 820 MHz |
|||
| Up to 8GB, dual-channel LPDDR4x @ 1866 MHz |
|||
| Cat-4, Cat-7 DL / Cat-13 UL |
|||
| Q3 2019 |
|||
| {{Collapsible list|Vivo Y19| Vivo Y5s| Vivo V17 Neo| Vivo S1 (SEA)| Vivo Y7s}} |
|||
|- |
|||
| MediaTek Helio P70 |
|||
| 2.1 GHz quad-core Cortex-A73 |
|||
|{{N/A}} |
|||
| 2.0 GHz quad-core Cortex-A53 |
|||
| Mali-G72 MP3 @ 900 MHz |
|||
| Up to 8GB, dual-channel LPDDR4x @ 1800 MHz |
|||
| Cat-7 (DL) / Cat-13 (UL) |
|||
| Q4 2018 |
|||
| {{Collapsible list|[[Moto G8]] Play| [[Motorola One]] Macro| Vivo V15| Vivo S1| Oppo F11/F11 Pro| Oppo F15| Oppo Reno2 F| ZTE Blade V10| Realme U1}} |
|||
|- |
|||
| MediaTek Helio P90 |
|||
| 12nm FinFET+ |
|||
| 2.2 GHz dual-core Cortex-A75 |
|||
| {{N/A}} |
|||
| 2.0 GHz hexa-core Cortex-A55 |
|||
| PowerVR GM9446 @ 970 MHz |
|||
| rowspan="3"|Up to 8GB, Dual-channel LPDDR4x @ 1866 MHz |
|||
| Cat-12 (DL) / Cat-13 (UL) |
|||
| Q1 2019 |
|||
| Oppo Reno Z, Oppo Reno2 Z |
|||
|- |
|||
| MediaTek Helio G70 |
|||
| rowspan="4"|12nm FFC |
|||
| rowspan="2"|2.0 GHz dual-core Cortex-A75 |
|||
|{{N/A}} |
|||
| rowspan="2"|1.7 GHz hexa-core Cortex-A55 |
|||
| Mali-G52 MC2 @ 820 MHz |
|||
| rowspan="2"|Cat-7 (DL) / Cat-13 (UL) |
|||
| rowspan="2"|Q2 2020 |
|||
| Realme C3 |
|||
|- |
|||
| MediaTek Helio G80 |
|||
|{{N/A}} |
|||
| Mali-G52 MC2 @ 950 MHz |
|||
| |
|||
|- |
|||
| MediaTek Helio G90 |
|||
| rowspan="2"|2.05 GHz dual-core Cortex-A76 |
|||
|{{N/A}} |
|||
| rowspan="2"|2.0 GHz hexa-core Cortex-A55 |
|||
| Mali-G76 MC4 @ 720 MHz |
|||
| rowspan="2"|Up to 10GB, Dual-channel LPDDR4x @ 2133 MHz |
|||
| rowspan="2"|Cat-12 (DL) / Cat-13 (UL) |
|||
| rowspan="2"|Q3 2019 |
|||
| |
|||
|- |
|||
| MediaTek Helio G90T (MT5785V) |
|||
| {{N/A}} |
|||
| Mali-G76 MC4 @ 800 MHz |
|||
| [[Redmi Note 8|Redmi Note 8 Pro]] |
|||
|- |
|||
|MediaTek Dimensity 800 |
|||
|rowspan="2"|7nm N7 |
|||
|2 GHz Quad-Core Cortex-A76 |
|||
|{{N/A}} |
|||
|rowspan="2"|2 GHz Quad-Core Cortex-A55 |
|||
|Mali-G57 MC4 |
|||
|LPDDR4X-1866 Dual-Channel 32-Bit (17.1GB/s) |
|||
|rowspan="2"|2G/3G/4G/5G Multi Mode NSA/SA |
|||
NR : Sub-6GHz 2CC CA, mmWave |
|||
|Q2 2020 |
|||
| |
|||
|- |
|||
|MediaTek Dimensity 1000 |
|||
|2.6 GHz Quad-Core Cortex-A77 |
|||
|{{N/A}} |
|||
|Mali-G77 MP9 |
|||
|LPDDR4X-1866 Quad-Channel 64-Bit (29.8GB/s) |
|||
|Q1 2020 |
|||
|Oppo Reno3 |
|||
|- |
|||
|[[Qualcomm]] Snapdragon 439 (SDM439) |
|||
|12 nm FinFET |
|||
|1.9 GHz Quad-core ARM Cortex-A53 |
|||
|{{N/A}} |
|||
|1.45 GHz Quad-core ARM Cortex-A53 |
|||
|[[Adreno]] 505 |
|||
|LPDDR3 Single-channel 933 MHz |
|||
|Download: Cat 7, up to 300 Mbit/s; Upload: Cat 13, up to 150 Mbit/s |
|||
|Q2 2018 |
|||
|{{Collapsible list|Samsung Galaxy A01|Vivo Y3| Vivo U1| Nokia 4.2| Redmi 7A|Redmi 8| Redmi 8A| Alcatel 3 (2019)}} |
|||
|- |
|||
|[[Qualcomm]] Snapdragon 415 (MSM8929)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-processors-415|title=Snapdragon 415 Processor|date=2018-10-02|website=Qualcomm|language=en|access-date=2019-12-31}}</ref> |
|||
| rowspan="6" |28 nm |
|||
|1.4 GHz Quad-core ARM Cortex-A53 |
|||
|{{N/A}} |
|||
|998 MHz Quad-core ARM Cortex-A53 |
|||
| rowspan="3" |[[Adreno]] 405 |
|||
| rowspan="2" |32-bit single-channel 667 MHz LPDDR3 |
|||
| rowspan="2" |LTE Cat 4 |
|||
|Q1 2015 |
|||
|Lenovo Vibe K5 |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 615/616 (MSM8939/v2/v3)<ref>{{cite web |url=http://www.qualcomm.com/products/snapdragon/processors/615 |title=Snapdragon 615 Processor Specs and Details |date=April 2014 |publisher=Qualcomm}}</ref> |
|||
| 1.5-1.7 GHz Quad-core ARM Cortex-A53 |
|||
|{{N/A}} |
|||
| 1.0-1.2 GHz Quad-core ARM Cortex-A53 |
|||
| Q3 2014 |
|||
|{{Collapsible list |
|||
| title = |Moto X Play|Asus ZenFone 2 Laser|ZTE Blade A711|ZTE Blade X9|Lenovo Vibe Shot Z90|Lenovo Vibe X2 Pro|Samsung Galaxy J7 ([[Samsung Galaxy J7|2015]]/[[Samsung Galaxy J7 (2016)|2016]])|Samsung Galaxy A7 ([[Samsung Galaxy A7 (2015)|2015]]/[[Samsung Galaxy A7 (2016)|2016]])|[[Samsung Galaxy A8 (2015)]]|Nubia Z9 mini|LG G4s Beat|Vivo Y37A|Vivo X6A|Oppo R5|Oppo R7s|HTC Desire 820|HTC One M8s|Xiaomi Mi4i|Redmi 3|Honor 5X|Huawei GX8|Lenovo Vibe K5 Plus |
|||
}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 617 (MSM8952)<ref>{{cite web |url=http://www.qualcomm.com/products/snapdragon/processors/617 |title=Snapdragon 617 processor |date=September 2015 |publisher=Qualcomm}}</ref> |
|||
| 1.5 GHz Quad-core ARM Cortex-A53 |
|||
|{{N/A}} |
|||
| 1.2 GHz Quad-core ARM Cortex-A53 |
|||
|32-bit single-channel 933 MHz LPDDR3 |
|||
| LTE Cat 7 |
|||
| Q4 2015 |
|||
|{{Collapsible list|[[HTC One A9]]|Alcatel Idol 4|Moto G4|Samsung Galaxy C5|ZTE Axon 7 mini|ZTE Axon Max|Nubia Z11 mini|Huawei G9 Lite|Honor 5A|BlackBerry DTEK50 |
|||
}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 650 (MSM8956)<ref>{{cite web |url=http://www.qualcomm.com/products/snapdragon/processors/650 |title=Snapdragon 650 Processor Specs and Details |date=November 2015 |publisher=Qualcomm}}</ref> |
|||
| 1.8 GHz Dual-core ARM Cortex-A72 |
|||
|{{N/A}} |
|||
| 1.4 GHz Quad-core ARM Cortex-A53 |
|||
| rowspan="3" |[[Adreno]] 510 |
|||
| rowspan="3" | 32-bit dual-channel 933 MHz LPDDR3 |
|||
| LTE Cat 7 |
|||
| Q4 2015 |
|||
|{{Collapsible list|Xiaomi Redmi Note 3 Pro|Xiaomi Mi Max|[[Sony Xperia X]]|[[Sony Xperia X Compact]] |
|||
}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 652 (MSM8976)<ref>{{cite web |url=http://www.qualcomm.com/products/snapdragon/processors/652 |title=Snapdragon 652 Processor Specs and Details |date=November 2015 |publisher=Qualcomm}}</ref> |
|||
| 1.8 GHz Quad-core ARM Cortex-A72 |
|||
|{{N/A}} |
|||
| 1.4 GHz Quad-core ARM Cortex-A53 |
|||
| LTE Cat 7 |
|||
| Q4 2015 |
|||
|{{Collapsible list|Xiaomi Mi Max|Samsung Galaxy A9/A9 Pro|Samsung Galaxy C9|HTC 10 Lifestyle|HTC U11 EYEs|Alcatel Idol 4S|BQ Aquaris X5 Plus|Nubia Z11 Max|Nubia Z17 Mini|Sharp Z3|Vivo X6/X6 Plus|Vivo X7/X7 Plus|Vivo X9s|LG G5 SE|Asus ZenFone 3 Ultra|Xiaomi Mi Max |
|||
}} |
|||
|- |
|||
|[[Qualcomm]] Snapdragon 653 (MS8976 Pro)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-653-mobile-platform|title=Snapdragon 653 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| 1.95 GHz Quad-core ARM Cortex-A72 |
|||
|{{N/A}} |
|||
| 1.44 GHz Quad-core ARM Cortex-A53 |
|||
| LTE Cat 7 |
|||
| Q4 2016 |
|||
|{{Collapsible list|Oppo F3 Plus| Nubia Z17 Mini/MiniS| Nubia Z17 Lite| Vivo X9s Plus| Oppo R9s Plus}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 630 (SDM630)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-630-mobile-platform|title=Snapdragon 630 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| rowspan="4"|14 nm |
|||
| 2.2 GHz Quad-core ARM Cortex-A53 |
|||
| {{N/A}} |
|||
| 1.8 GHz Quad-core ARM Cortex-A53 |
|||
| [[Adreno]] 508 |
|||
| rowspan="2"|LPDDR4 Dual-channel 1333 MHz |
|||
| rowspan="3"|Download: Cat 12, up to 600 Mbit/s; Upload: Cat 13, up to 150 Mbit/s |
|||
| Q2 2017 |
|||
| {{Collapsible list|[[Moto X4]]| [[Nokia 6.1]]| [[Nokia 7]]| [[Moto G6]] Plus| Asus ZenFone 4| Asus ZenFone 5 Lite/5Q| [[Sony Xperia XA2]]| Sony Xperia XA2 Plus| Sony Xperia XA2 Ultra| [[Sony Xperia 8]]| [[Sony Xperia 10]]| HTC U11 Life}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 636 (SDM636)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-636-mobile-platform|title=Snapdragon 636 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| 1.8 GHz Quad-core Kryo 260 Gold (Cortex-A73) |
|||
| {{N/A}} |
|||
| 1.6 GHz Quad-core Kryo 260 Silver (Cortex-A53) |
|||
| [[Adreno]] 509 |
|||
| Q3 2017 |
|||
| {{Collapsible list|[[Moto Z3]] Play| [[Moto G7]] Plus| [[Nokia 6.1 Plus]]| [[Nokia 6.2]]| [[Nokia 7.1]]| Lenovo K5 Pro| Lenovo S5 Pro| Lenovo Z5| [[Redmi Note 5]]| Redmi Note 5 Pro| [[Redmi Note 6 Pro]]| Xiaomi Mi Max 3| Meizu E3| HTC U12 Life| Asus ZenFone 5| Vivo Z1i| [[BlackBerry Key2]] LE| [[Sony Xperia 10 Plus]]}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 660 (SDM660)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-660-mobile-platform|title=Snapdragon 660 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| 2.2 GHz Quad-core Kryo 260 Gold (Cortex-A73) |
|||
| {{N/A}} |
|||
| 1.84 GHz Quad-core Kryo 260 Silver (Cortex-A53) |
|||
| [[Adreno]] 512 |
|||
| LPDDR4 Dual-channel 1866 MHz |
|||
| Q2 2017 |
|||
| {{Collapsible list|[[Nokia 7 Plus]]| [[Nokia 7.2]]| Nokia X71| Vivo Z1| Vivo V11| Redmi 7 Pro| [[Redmi Note 7]]/Note 7S| Xiaomi Mi A2| Xiaomi Mi 6x| Xiaomi Mi 8 Lite| Xiaomi Mi Note 3| Asus ZenFone 4| Oppo AX7 Pro| Oppo K1| [[Samsung Galaxy A6s]]| [[Samsung Galaxy A8 Star]]/A9 Star| [[Samsung Galaxy A9 (2018)]]| Samsung Galaxy S Luxury Edition| Realme 2 Pro| [[BlackBerry Key2]]| Lenovo S5 Pro GT| Meizu Z15| Nubia Z18 Mini| Sharp Aquos S2| Sharp Aquos S3}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 632 (SDM632)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-632-mobile-platform|title=Snapdragon 632 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| 1.8 GHz Quad-core Kryo 250 Gold (Cortex-A73) |
|||
| {{N/A}} |
|||
| 1.8 GHz Quad-core Kryo 250 Silver (Cortex-A53) |
|||
| [[Adreno]] 506 |
|||
| LPDDR3 |
|||
| Download: Cat 7, up to 300 Mbit/s; Upload: Cat 13, up to 150 Mbit/s |
|||
| Q2 2018 |
|||
| {{Collapsible list|[[Moto G7]]| Moto G7 Play| Moto G7 Power| Honor 8C| LG W30 Pro| Redmi 7| Redmi Y3| Meizu Note 8}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 670 (SDM670)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-670-mobile-platform|title=Snapdragon 670 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| 10 nm |
|||
| 2.0 GHz Dual-core Kryo 360 Gold (Cortex-A75) |
|||
| {{N/A}} |
|||
| 1.7 GHz Hexa-core Kryo 360 Silver (Cortex-A55) |
|||
| [[Adreno]] 615 |
|||
| rowspan="2"|LPDDR4X Dual-channel 1866 MHz |
|||
| rowspan="3"|Download: Cat 12, up to 600 Mbit/s; Upload: Cat 13, up to 150 Mbit/s |
|||
| Q3 2018 |
|||
| {{Collapsible list|[[Pixel 3a]]/Pixel 3a XL| Oppo R17| Vivo Z3| Vivo X23}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 675 (SM6150)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-675-mobile-platform|title=Snapdragon 675 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| rowspan="2"|11 nm |
|||
| 2.0 GHz Dual-core Kryo 460 Gold (Cortex-A76) |
|||
| {{N/A}} |
|||
| 1.7 GHz Hexa-core Kryo 460 Silver (Cortex-A55) |
|||
| [[Adreno]] 612 |
|||
| Q1 2019 |
|||
| {{Collapsible list|[[Motorola One]] Zoom| Motorola One Hyper| [[Moto Z4]]| Meizu Note 9| Meizu 16Xs| Vivo V15 Pro| Vivo U3| [[Redmi Note 7]] Pro| [[Samsung Galaxy M40]]/A60| [[Samsung Galaxy A70]]| Samsung Galaxy A70s}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 665 (SM6125)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-665-mobile-platform|title=Snapdragon 665 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| 2.0 GHz Quad-core Kryo 260 Gold (Cortex-A73) |
|||
| {{N/A}} |
|||
| 1.8 GHz Quad-core Kryo 260 Silver (Cortex-A53) |
|||
| [[Adreno]] 610 |
|||
| LPDDR3/LPDDR4X Dual‑channel up to 1866 MHz |
|||
| Q2 2019 |
|||
| {{Collapsible list|[[Realme 5]]| Oppo A11| Oppo A11x| Vivo U10| Vivo U3x| [[Redmi Note 8]]/Note 8T| [[Xiaomi Mi A3]]/Mi CC9e| [[Moto G8]] Plus|Moto G Power|Moto G Stylus}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 710 (SDM710)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-710-mobile-platform|title=Snapdragon 710 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
|rowspan="2"|10 nm |
|||
| 2.2 GHz Dual-core Kryo 360 Gold (Cortex-A75) |
|||
|{{N/A}} |
|||
|rowspan="2"|1.7 GHz Hexa-core Kryo 360 Silver (Cortex-A55) |
|||
|rowspan="2"|[[Adreno]] 616 |
|||
|rowspan="2"|LPDDR4X up to 8 GB, Dual-channel 16-bit (32-bit), 1866 MHz (14.9 GB/s) |
|||
|rowspan="2"|Download: Cat 15, up to 800 Mbit/s; Upload: Cat 13, up to 150 Mbit/s |
|||
| Q2 2018 |
|||
| {{Collapsible list|Realme 3 Pro| Realme X| Xiaomi Mi CC9/Mi 9 Lite| Meizu 16| Lenovo Z5 Pro| Vivo Z5x| Vivo NEX A| Oppo K3| [[Oppo Reno]]| Nokia 8.1| [[Samsung Galaxy A8s]]| HTC U19e| [[Motorola Razr (2020)]]}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 712 (SDM712)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-712-mobile-platform|title=Snapdragon 712 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| 2.3 GHz Dual-core Kryo 360 Gold (Cortex-A75) |
|||
|{{N/A}} |
|||
| Q1 2019 |
|||
| {{Collapsible list|Vivo Z1x| Vivo Z1 Pro| Vivo V17| Vivo S5| Vivo Z5| Realme 5 Pro| Realme XT| Xiaomi Mi 9 SE}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 730 (SM7150-AA/AB)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-730-mobile-platform|title=Snapdragon 730 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
|8 nm |
|||
| 2.2 GHz Dual-core Kryo 470 Gold (Cortex-A76) |
|||
|{{N/A}} |
|||
| 1.8 GHz Hexa-core Kryo 470 Silver (Cortex-A55) |
|||
|[[Adreno]] 618 |
|||
|LPDDR4X up to 8 GB, Dual-channel 16-bit (32-bit), 1866 MHz (14.9 GB/s) |
|||
|Download: Cat 15, up to 800 Mbit/s; Upload: Cat 13, up to 150 Mbit/s |
|||
| Q2 2019 |
|||
| {{Collapsible list|Oppo K5| [[Oppo Reno2]]| Realme X2| [[Redmi K20]]/Xiaomi Mi 9T| [[Redmi K30]]| Lenovo Z6| Samsung Galaxy A71| [[Samsung Galaxy A80]]| [[Xiaomi Mi CC9 Pro]]/Mi Note 10 & Note 10 Pro}} |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 765 (SM720-AA/AB)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-765-5g-mobile-platform|title=Snapdragon 765 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|7 nm |
|||
| 2.3 or 2.4 GHz Single-core Kryo 475 Prime |
|||
| 2.2 GHz Single-core Kryo 475 Gold |
|||
| 1.8 GHz Hexa-core Kryo 465 Silver |
|||
|[[Adreno]] 620 |
|||
|LPDDR4X, Dual-channel 16-bit (32-bit), 2133 MHz (17 GB/s) |
|||
|Qualcomm X52 5G/LTE |
|||
5G: download up to 3,7 Gbit/s, upload: up to 1,6 Gbit/s; LTE: download Cat 24, up to 1200 Mbit/s, upload Cat 22, up to 210 Mbit/s |
|||
| Q4 2019 |
|||
| [[Redmi K30|Redmi K30 5G]], Oppo Reno3 Vitality, Oppo Reno3 Pro, Realme X50 5G |
|||
|- |
|||
| [[Qualcomm]] Snapdragon 808 (MSM8992)<ref>{{cite web |url=http://www.qualcomm.com/products/snapdragon/processors/808 |title=Snapdragon 808 Processor Specs and Details |date=April 2014 |publisher=Qualcomm}}</ref> |
|||
| 20 nm |
|||
| 1.8 GHz Dual-core Cortex-A57 |
|||
|{{N/A}} |
|||
| 1.5 GHz Quad-core ARM Cortex-A53 |
|||
| [[Adreno]] 418 |
|||
| 32-bit 933 MHz LPDDR3 (14.9 GB/s) |
|||
| LTE Cat 6/7 |
|||
| H1 2015 |
|||
| {{Collapsible list|[[LG G4]]| [[Microsoft Lumia 950]]| [[Nexus 5X]]| [[BlackBerry Priv]]| [[LG V10]], Xiaomi Mi4c/Mi4S| Lenovo Vibe X3| Moto X Style| Acer Liquid Jade Primo}} |
|||
|- |
|||
| Qualcomm Snapdragon 810 (MSM8994)<ref>{{cite web |url=http://www.qualcomm.com/products/snapdragon/processors/810 |title=Snapdragon 810 Processor Specs and Details |date=April 2014 |publisher=Qualcomm}}</ref> |
|||
| 20 nm |
|||
| 2.0 GHz Quad-core Cortex-A57 |
|||
|{{N/A}} |
|||
| 1.5 GHz Quad-core ARM Cortex-A53 |
|||
| Adreno 430 |
|||
| 32-bit dual-channel 1600 MHz LPDDR4 (25.6 GB/s) |
|||
| LTE Cat 6/7 |
|||
| H1 2015 |
|||
|{{Collapsible list|[[Sony Xperia Z5]]/Z5 Premium| [[LG G Flex 2]]| [[OnePlus 2]]| [[Microsoft Lumia 950 XL]]| [[Nexus 6P]]| [[HTC One M9]]| Moto X Force| Nubia Z9/Max/Max Elite| ZTE Axon Lux| ZTE Axon Elite| HTC J Butterfly}} |
|||
|- |
|||
| Qualcomm Snapdragon 820/821 (MSM8996/MSM8996 Pro)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-820-mobile-platform|title=Snapdragon 820 Processor Specs and Details|publisher=Qualcomm}}</ref><ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-821-mobile-platform|title=Snapdragon 821 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| 14 nm LPP |
|||
| 1.8–2.34 GHz Dual-core [[Kryo_(microarchitecture)|Kryo]] |
|||
|{{N/A}} |
|||
| 1.36–2.19 GHz Dual-core Kryo |
|||
| Adreno 530 |
|||
| LPDDR4-1866 Dual-Channel 64-Bit (29.8GB/s) |
|||
| Downlink: LTE Cat 12, |
|||
Uplink: LTE Cat 13 |
|||
| Q4 2015 |
|||
|{{Collapsible list|[[LG G5]]| [[LG G6]]| LG G7 Fit| LG G Pad 5 10.1| [[OnePlus 3]]/3T| [[LG V20]]| [[Samsung Galaxy S7]]/[[S7 Edge]] (US)| [[Samsung Galaxy Note 7]] (US)| Xiaomi Mi5| Xiaomi Mi5s/Mi5s Plus| [[Pixel (smartphone)|Google Pixel]]/Pixel XL| [[Moto Z]]}} |
|||
|- |
|||
| Qualcomm [[List_of_Qualcomm_Snapdragon_systems-on-chip#Snapdragon_835,_845_and_850_(2017/18)|Snapdragon 835]] (MSM8998)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-835-mobile-platform|title=Snapdragon 835 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
| 10 nm FinFET |
|||
| 2.35–2.45 GHz Quad-core Kryo |
|||
|{{N/A}} |
|||
| 1.8–1.9 GHz Quad-core Kryo |
|||
| Adreno 540 @710MHz |
|||
| LPDDR4X-1866 Dual-Channel 64-Bit (29.8GB/s) |
|||
|Qualcomm X16 LTE Downlink: LTE Cat 16, |
|||
Uplink: LTE Cat 13 |
|||
| Q4 2016 |
|||
|{{Collapsible list|[[Samsung Galaxy S8]] (US/China)| Xiaomi Mi 6| [[HTC U11]]| Sony Xperia XZ Premium| ZTE Nubia Z17| [[OnePlus 5]]| Moto Z2 Force| [[Samsung Galaxy Note 8]] (US/China)| [[Nokia 8]]| [[LG V30]]| Asus ZenFone 4 Pro| Xperia XZ1| Xperia XZ1 Compact| [[Pixel 2|Pixel 2/Pixel XL 2]]| LG G7 One| [[Essential Phone]]| [[Razer Phone]]| Sharp Aquos V| [[Oculus_VR#Oculus_Quest|Oculus Quest]]}} |
|||
|- |
|||
|Qualcomm [[Snapdragon 845]] (SDM845)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-845-mobile-platform|title=Snapdragon 845 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
|10nm FinFET LPP |
|||
|2.8 GHz Quad-Core "Kryo 385 Gold" |
|||
|{{N/A}} |
|||
|1.8 GHz Quad-Core "Kryo 385 Silver" |
|||
|Adreno 630 @710MHz |
|||
|LPDDR4X-1866 Quad-Channel 64-Bit (29.9GB/s) |
|||
|Qualcomm X20 LTE Downlink: LTE Cat.18 |
|||
Uplink: LTE Cat.13 |
|||
|Q1 2018 |
|||
|{{Collapsible list|[[OnePlus 6]]| [[OnePlus 6T]]| [[Samsung Galaxy S9]]| [[Samsung Galaxy Note 9]]| [[Xiaomi Mi 8]]| [[Xiaomi Pocophone F1]]| [[Pixel 3]]/Pixel 3 XL| [[Nokia 9 PureView]]| [[Sony Xperia XZ2]]| [[Sony Xperia XZ2 Compact]]| [[Sony Xperia XZ2 Premium]]| [[Sony Xperia XZ3]]| [[LG G7 ThinQ]]| [[LG V40 ThinQ]]| Moto Z3| [[HTC U12+]]| [[Oppo Find X]]| [[Vivo NEX Dual Display]]| Meizu 16/16 Plus| Razer Phone 2| [[ROG Phone]]| Asus Zenfone 5z| [[Nubia X]]| Nubia Z18| ZTE Axon 9 Pro| Nubia Red Magic Mars| Sharp Aquos R2/R2 Compact| Sharp Aquos Zero}} |
|||
|- |
|||
|Qualcomm [[List_of_Qualcomm_Snapdragon_systems-on-chip#Snapdragon_835,_845_and_850_(2017/18)|Snapdragon 850]] (SDM850) |
|||
|10nm FinFET LPP |
|||
|2.95 GHz Quad-Core "Kryo 385 Gold" |
|||
|{{N/A}} |
|||
|1.8GHz Quad-Core "Kryo 385 Silver" |
|||
|Adreno 630 @710MHz |
|||
|LPDDR4X-1866 Quad-Channel 64-Bit (29.9GB/s) |
|||
|Qualcomm X20 LTE Downlink: LTE Cat.18 |
|||
Uplink: LTE Cat.13 |
|||
|Q3 2018 |
|||
| |
|||
|- |
|||
|Qualcomm [[Snapdragon 855]] (SM8150)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-855-mobile-platform|title=Snapdragon 855 Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|7nm FinFET N7 |
|||
|2.84 GHz Single-Core "Kryo 485 Gold Prime" |
|||
|2.42 GHz Tri-Core "Kryo 485 Gold" |
|||
|1.8 GHz Quad-Core "Kryo 485 Silver" |
|||
|Adreno 640 @585MHz |
|||
|LPDDR4X-2133 Quad-Channel 64-Bit (34.13GB/s) |
|||
|Qualcomm X24 LTE Downlink: LTE Cat.20 |
|||
Uplink: LTE Cat.13 |
|||
|Q1 2019 |
|||
|{{Collapsible list|[[OnePlus 7]]| [[Samsung Galaxy A90 5G]]| [[Samsung Galaxy S10]]| [[Samsung Galaxy Note 10]]| [[Samsung Galaxy Fold]]| [[Xiaomi Mi 9]]| [[Pixel 4]]/Pixel 4 XL| [[Sony Xperia 1]]| [[Sony Xperia 5]]| [[LG G8 ThinQ]]| [[LG V50 ThinQ]]| Lenovo Z5 Pro GT| Lenovo Z6 Pro| [[Redmi K20]] Pro| [[Oppo Reno]] 10x Zoom/5G| Vivo iQOO| Meizu 16s| Meizu 16T| [[Asus ZenFone#Sixth generation (2019)|Asus ZenFone 6]]| ZTE Axon 10 Pro| Xiaomi Black Shark 2| [[Xiaomi Mi MIX 3]]}} |
|||
|- |
|||
|Qualcomm [[List_of_Qualcomm_Snapdragon_systems-on-chip#Snapdragon_855,_855+,_8cx_and_SQ1_(2019)|Snapdragon 855+]] (SM8150-AC)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-855-plus-mobile-platform|title=Snapdragon 855+ Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|7nm FinFET N7 |
|||
|2.96 GHz Single-Core "Kryo 485 Gold Prime" |
|||
|2.42 GHz Tri-Core "Kryo 485 Gold" |
|||
|1.8 GHz Quad-Core "Kryo 485 Silver" |
|||
|Adreno 640 @672MHz |
|||
|LPDDR4X-2133 Quad-Channel 64-Bit (34.13GB/s) |
|||
|Qualcomm X24 LTE Downlink: LTE Cat.20 |
|||
Uplink: LTE Cat.13 |
|||
|Q3 2019 |
|||
|{{Collapsible list|[[OnePlus 7T]]| [[Samsung Galaxy Z Flip]], Asus [[ROG Phone II]]| Oppo Reno Ace| [[Vivo NEX 3]]| Vivo iQOO Pro| [[Nubia Z20]]| Nubia Red Magic 3s| Meizu 16s Pro| [[Xiaomi Mi 9 Pro]]| Xiaomi Black Shark 2 Pro| [[Xiaomi Mi MIX Alpha]]| [[Redmi K20]] Pro Premium| Realme X2 Pro}} |
|||
|- |
|||
|Qualcomm [[List of Qualcomm Snapdragon systems-on-chip#Snapdragon 865|Snapdragon 865]] (SM8250)<ref>{{Cite web|url=https://www.qualcomm.com/products/snapdragon-865-5g-mobile-platform|title=Snapdragon 865 5G Processor Specs and Details|publisher=Qualcomm}}</ref> |
|||
([[ARM DynamIQ|DynamIQ]]) |
|||
|7nm+ FinFET EUV (N7P) |
|||
|2.84 GHz Single-Core "Kryo 585 Gold Prime" |
|||
|2.42 GHz Tri-Core "Kryo 585 Gold" |
|||
|1.8 GHz Quad-Core "Kryo 585 Silver" |
|||
|Adreno 650 |
|||
|LPDDR5-2750 Quad-Channel 64-Bit (44GB/s) |
|||
or |
|||
LPDDR4X-2133 Quad-Channel 64-Bit (33.4GB/s) |
|||
|External: Qualcomm X55 5G NSA/SA |
|||
LTE Downlink: LTE Cat.22 |
|||
LTE Uplink: LTE Cat.13 |
|||
5G: Sub-6GHz, mmWave |
|||
|Q4 2019 |
|||
|ZTE Axon 10s Pro 5G, [[Samsung Galaxy S20]] |
|||
|- |
|||
|Qualcomm [[List of Qualcomm Snapdragon systems-on-chip#Snapdragon_855,_855+,_8cx_and_SQ1_(2019)|Snapdragon 8cx]] (8cx) |
|||
|7nm FinFET N7 |
|||
|2.84 GHz Quad-Core "Kryo 495" |
|||
|{{N/A}} |
|||
|1.8 GHz Quad-Core "Kryo 495" |
|||
|Adreno 680 |
|||
|LPDDR4X-2133 Octa-Channel 128-Bit (68.26GB/s) |
|||
|Qualcomm X24 LTE Downlink: LTE Cat.20 |
|||
Uplink: LTE Cat.13 |
|||
|Q3 2019 |
|||
| |
|||
|- |
|||
|Microsoft [[List of Qualcomm Snapdragon systems-on-chip#Snapdragon_855,_855+,_8cx_and_SQ1_(2019)|SQ1]] (SQ1) |
|||
|7nm FinFET N7 |
|||
|3 GHz Quad-Core "Kryo 495" |
|||
|{{N/A}} |
|||
|1.8 GHz Quad-Core "Kryo 495" |
|||
|Adreno 685 |
|||
|LPDDR4X-2133 Octa-Channel 128-Bit (68.26GB/s) |
|||
|Qualcomm X24 LTE Downlink: LTE Cat.20 |
|||
Uplink: LTE Cat.13 |
|||
|Q3 2019 |
|||
|[[Surface Pro X|Microsoft Surface Pro X]] |
|||
|} |
|||
=={{Anchor|DynamIQ}}Successor== |
=={{Anchor|DynamIQ}}Successor== |
||
In May 2017, ARM announced DynamIQ as the successor to big.LITTLE.<ref>{{cite news|last1=Humrick|first1=Matt|title=Exploring Dynamiq and ARM's New CPUs|url=http://www.anandtech.com/show/11441/dynamiq-and-arms-new-cpus-cortex-a75-a55| |
In May 2017, ARM announced DynamIQ as the successor to big.LITTLE.<ref>{{cite news|last1=Humrick|first1=Matt|title=Exploring Dynamiq and ARM's New CPUs|url=http://www.anandtech.com/show/11441/dynamiq-and-arms-new-cpus-cortex-a75-a55|access-date=10 July 2017|publisher=Anandtech|date=29 May 2017}}</ref> DynamIQ is expected to allow for more flexibility and scalability when designing multi-core processors. In contrast to big.LITTLE, it increases the maximum number of cores in a cluster to 8 for Armv8.2 CPUs, 12 for Armv9 and 14 for Armv9.2<ref>{{Cite web |last=Ltd |first=Arm |title=DynamIQ – Arm® |url=https://www.arm.com/technologies/dynamiq |access-date=2023-10-18 |website=Arm {{!}} The Architecture for the Digital World |language=en}}</ref> and allows for varying core designs within a single cluster, and up to 32 total clusters. The technology also offers more fine grained per core voltage control and faster L2 cache speeds. However, DynamIQ is incompatible with previous ARM designs and is initially only supported by the [[ARM Cortex-A75|Cortex-A75]] and [[ARM Cortex-A55|Cortex-A55]] CPU cores and their successors. |
||
==References== |
==References== |
||
{{Reflist|30em|refs= |
{{Reflist|30em|refs= |
||
<ref name="Auto4J-1">{{cite press release | url=http://www.arm.com/about/newsroom/arm-unveils-its-most-energy-efficient-application-processor-ever-with-biglittle-processing.php | title=ARM Unveils its Most Energy Efficient Application Processor Ever; Redefines Traditional Power And Performance Relationship With big.LITTLE Processing | publisher=[[ARM Holdings]] | date=19 October 2011 | |
<ref name="Auto4J-1">{{cite press release | url=http://www.arm.com/about/newsroom/arm-unveils-its-most-energy-efficient-application-processor-ever-with-biglittle-processing.php | title=ARM Unveils its Most Energy Efficient Application Processor Ever; Redefines Traditional Power And Performance Relationship With big.LITTLE Processing | publisher=[[ARM Holdings]] | date=19 October 2011 | access-date=2012-10-31}}</ref> |
||
<ref name="cortex-a50 announce">{{cite press release | url=http://www.arm.com/about/newsroom/arm-launches-cortex-a50-series-the-worlds-most-energy-efficient-64-bit-processors.php | title=ARM Launches Cortex-A50 Series, the |
<ref name="cortex-a50 announce">{{cite press release | url=http://www.arm.com/about/newsroom/arm-launches-cortex-a50-series-the-worlds-most-energy-efficient-64-bit-processors.php | title=ARM Launches Cortex-A50 Series, the World's Most Energy-Efficient 64-bit Processors | publisher=[[ARM Holdings]] | access-date=2012-10-31}}</ref> |
||
<ref name="anand-5420">{{cite web | url=http://www.anandtech.com/show/7313/samsung-announces-biglittle-mp-support-in-exynos-5420 | title=Samsung Announces big.LITTLE MP Support in Exynos 5420 | publisher=[[AnandTech]] | date=2013-09-11 | author=Brian Klug | |
<ref name="anand-5420">{{cite web | url=http://www.anandtech.com/show/7313/samsung-announces-biglittle-mp-support-in-exynos-5420 | title=Samsung Announces big.LITTLE MP Support in Exynos 5420 | publisher=[[AnandTech]] | date=2013-09-11 | author=Brian Klug | access-date=2013-09-16}}</ref> |
||
}} |
}} |
||
== Further reading == |
== Further reading == |
||
* {{cite web | url=https://lwn.net/Articles/534646/ | title=big.LITTLE MP status Jan 25, 2013 | publisher=[[LWN.net]] | date=25 January 2013 | author=David Zinman | |
* {{cite web | url=https://lwn.net/Articles/534646/ | title=big.LITTLE MP status Jan 25, 2013 | publisher=[[LWN.net]] | date=25 January 2013 | author=David Zinman | access-date=2013-01-25}} |
||
* {{cite web | url=https://lwn.net/Articles/481055/ | title=Linux support for ARM big.LITTLE | publisher=[[LWN.net]] | date=15 February 2012 | author=Nicolas Pitre | |
* {{cite web | url=https://lwn.net/Articles/481055/ | title=Linux support for ARM big.LITTLE | publisher=[[LWN.net]] | date=15 February 2012 | author=Nicolas Pitre | access-date=2012-10-18}} |
||
* {{cite web | url=https://lwn.net/Articles/501501/ | title=A big.LITTLE scheduler update | publisher=[[LWN.net]] | date=12 June 2012 | author=Paul McKenney | |
* {{cite web | url=https://lwn.net/Articles/501501/ | title=A big.LITTLE scheduler update | publisher=[[LWN.net]] | date=12 June 2012 | author=Paul McKenney | access-date=2012-10-18}} |
||
* {{cite web | url=https://lwn.net/Articles/514063/ | title=KS2012: ARM: A big.LITTLE update | publisher=[[LWN.net]] | date=5 September 2012 | author=Jake Edge | |
* {{cite web | url=https://lwn.net/Articles/514063/ | title=KS2012: ARM: A big.LITTLE update | publisher=[[LWN.net]] | date=5 September 2012 | author=Jake Edge | access-date=2012-10-18}} |
||
* {{cite web | url=https://arstechnica.com/gadgets/2011/10/arms-new-cortex-a7-is-tailor-made-for-android-superphones/ | title=ARM's new Cortex A7 is tailor-made for Android superphones | publisher=[[Ars Technica]] | date=20 October 2011 | author = Jon Stokes | |
* {{cite web | url=https://arstechnica.com/gadgets/2011/10/arms-new-cortex-a7-is-tailor-made-for-android-superphones/ | title=ARM's new Cortex A7 is tailor-made for Android superphones | publisher=[[Ars Technica]] | date=20 October 2011 | author = Jon Stokes | access-date=2012-10-31 }} |
||
* {{cite web | url=https://arstechnica.com/information-technology/2012/10/arm-goes-64-bit-with-new-cortex-a53-and-cortex-a57-designs/ | title=ARM goes 64-bit with new Cortex-A53 and Cortex-A57 designs | publisher=[[Ars Technica]] | date=30 October 2012 | author=Andrew Cunningham | |
* {{cite web | url=https://arstechnica.com/information-technology/2012/10/arm-goes-64-bit-with-new-cortex-a53-and-cortex-a57-designs/ | title=ARM goes 64-bit with new Cortex-A53 and Cortex-A57 designs | publisher=[[Ars Technica]] | date=30 October 2012 | author=Andrew Cunningham | access-date=2012-10-31}} |
||
==External links== |
==External links== |
||
Latest revision as of 22:01, 30 August 2024
ARM big.LITTLE is a heterogeneous computing architecture developed by Arm Holdings, coupling relatively battery-saving and slower processor cores (LITTLE) with relatively more powerful and power-hungry ones (big). The intention is to create a multi-core processor that can adjust better to dynamic computing needs and use less power than clock scaling alone. ARM's marketing material promises up to a 75% savings in power usage for some activities.[1] Most commonly, ARM big.LITTLE architectures are used to create a multi-processor system-on-chip (MPSoC).
In October 2011, big.LITTLE was announced along with the Cortex-A7, which was designed to be architecturally compatible with the Cortex-A15.[2] In October 2012 ARM announced the Cortex-A53 and Cortex-A57 (ARMv8-A) cores, which are also intercompatible to allow their use in a big.LITTLE chip.[3] ARM later announced the Cortex-A12 at Computex 2013 followed by the Cortex-A17 in February 2014. Both the Cortex-A12 and the Cortex-A17 can also be paired in a big.LITTLE configuration with the Cortex-A7.[4][5]
The problem that big.LITTLE solves
[edit]For a given library of CMOS logic, active power increases as the logic switches more per second, while leakage increases with the number of transistors. So, CPUs designed to run fast are different from CPUs designed to save power. When a very fast out-of-order CPU is idling at very low speeds, a CPU with much less leakage (fewer transistors) could do the same work. For example, it might use a smaller (fewer transistors) memory cache, or a simpler microarchitecture such as removing out-of-order execution. big.LITTLE is a way to optimize for both cases: Power and speed, in the same system.
In practice, a big.LITTLE system can be surprisingly inflexible. One issue is the number and types of power and clock domains that the IC provides. These may not match the standard power management features offered by an operating system. Another is that the CPUs no longer have equivalent abilities, and matching the right software task to the right CPU becomes more difficult. Most of these problems are being solved by making the electronics and software more flexible.
Run-state migration
[edit]There are three ways[6] for the different processor cores to be arranged in a big.LITTLE design, depending on the scheduler implemented in the kernel.[7]
Clustered switching
[edit]
The clustered model approach is the first and simplest implementation, arranging the processor into identically sized clusters of "big" or "LITTLE" cores. The operating system scheduler can only see one cluster at a time; when the load on the whole processor changes between low and high, the system transitions to the other cluster. All relevant data are then passed through the common L2 cache, the active core cluster is powered off and the other one is activated. A Cache Coherent Interconnect (CCI) is used. This model has been implemented in the Samsung Exynos 5 Octa (5410).[8]
In-kernel switcher (CPU migration)
[edit]
CPU migration via the in-kernel switcher (IKS) involves pairing up a 'big' core with a 'LITTLE' core, with possibly many identical pairs in one chip. Each pair operates as one so-termed virtual core, and only one real core is (fully) powered up and running at a time. The 'big' core is used when the demand is high and the 'LITTLE' core is employed when demand is low. When demand on the virtual core changes (between high and low), the incoming core is powered up, running state is transferred, the outgoing is shut down, and processing continues on the new core. Switching is done via the cpufreq framework. A complete big.LITTLE IKS implementation was added in Linux 3.11. big.LITTLE IKS is an improvement of cluster migration (§ Clustered switching), the main difference being that each pair is visible to the scheduler.
A more complex arrangement involves a non-symmetric grouping of 'big' and 'LITTLE' cores. A single chip could have one or two 'big' cores and many more 'LITTLE' cores, or vice versa. Nvidia created something similar to this with the low-power 'companion core' in their Tegra 3 System-on-Chip.
Heterogeneous multi-processing (global task scheduling)
[edit]
The most powerful use model of big.LITTLE architecture is heterogeneous multi-processing (HMP), which enables the use of all physical cores at the same time. Threads with high priority or computational intensity can in this case be allocated to the "big" cores while threads with less priority or less computational intensity, such as background tasks, can be performed by the "LITTLE" cores.[9]
This model has been implemented in the Samsung Exynos starting with the Exynos 5 Octa series (5420, 5422, 5430),[10][11] and Apple A series processors starting with the Apple A11.[12]
Scheduling
[edit]The paired arrangement allows for switching to be done transparently to the operating system using the existing dynamic voltage and frequency scaling (DVFS) facility. The existing DVFS support in the kernel (e.g. cpufreq in Linux) will simply see a list of frequencies/voltages and will switch between them as it sees fit, just like it does on the existing hardware. However, the low-end slots will activate the 'Little' core and the high-end slots will activate the 'Big' core. This is the early solution provided by Linux's "deadline" CPU scheduler (not to be confused with the I/O scheduler with the same name) since 2012.[13]
Alternatively, all the cores may be exposed to the kernel scheduler, which will decide where each process/thread is executed. This will be required for the non-paired arrangement but could possibly also be used on the paired cores. It poses unique problems for the kernel scheduler, which, at least with modern commodity hardware, has been able to assume all cores in a SMP system are equal rather than heterogeneous. A 2019 addition to Linux 5.0 called Energy Aware Scheduling is an example of a scheduler that considers cores differently.[14][15]
Advantages of global task scheduling
[edit]- Finer-grained control of workloads that are migrated between cores. Because the scheduler is directly migrating tasks between cores, kernel overhead is reduced and power savings can be correspondingly increased.
- Implementation in the scheduler also makes switching decisions faster than in the cpufreq framework implemented in IKS.
- The ability to easily support non-symmetrical clusters (e.g. with 2 Cortex-A15 cores and 4 Cortex-A7 cores).
- The ability to use all cores simultaneously to provide improved peak performance throughput of the SoC compared to IKS.
Successor
[edit]In May 2017, ARM announced DynamIQ as the successor to big.LITTLE.[16] DynamIQ is expected to allow for more flexibility and scalability when designing multi-core processors. In contrast to big.LITTLE, it increases the maximum number of cores in a cluster to 8 for Armv8.2 CPUs, 12 for Armv9 and 14 for Armv9.2[17] and allows for varying core designs within a single cluster, and up to 32 total clusters. The technology also offers more fine grained per core voltage control and faster L2 cache speeds. However, DynamIQ is incompatible with previous ARM designs and is initially only supported by the Cortex-A75 and Cortex-A55 CPU cores and their successors.
References
[edit]- ^ "big.LITTLE technology". ARM.com. Archived from the original on 22 October 2012. Retrieved 17 October 2012.
- ^ "ARM Unveils its Most Energy Efficient Application Processor Ever; Redefines Traditional Power And Performance Relationship With big.LITTLE Processing" (Press release). ARM Holdings. 19 October 2011. Retrieved 31 October 2012.
- ^ "ARM Launches Cortex-A50 Series, the World's Most Energy-Efficient 64-bit Processors" (Press release). ARM Holdings. Retrieved 31 October 2012.
- ^ "ARM's new Cortex-A12 is ready to power 2014's $200 midrange smartphones". The Verge. April 2014.
- ^ "ARM Cortex A17: An Evolved Cortex A12 for the Mainstream in 2015". AnandTech. April 2014.
- ^ Brian Jeff (18 June 2013). "Ten Things to Know About big.LITTLE". ARM Holdings. Archived from the original on 10 September 2013. Retrieved 17 September 2013.
- ^ George Grey (10 July 2013). "big.LITTLE Software Update". Linaro. Archived from the original on 4 October 2013. Retrieved 17 September 2013.
- ^ Peter Clarke (6 August 2013). "Benchmarking ARM's big-little architecture". Retrieved 17 September 2013.
- ^ Big.LITTLE Processing with ARM Cortex-A15 & Cortex-A7 (PDF), ARM Holdings, September 2013, archived from the original (PDF) on 17 April 2012, retrieved 17 September 2013
- ^ Brian Klug (11 September 2013). "Samsung Announces big.LITTLE MP Support in Exynos 5420". AnandTech. Retrieved 16 September 2013.
- ^ "Samsung Unveils New Products from its System LSI Business at Mobile World Congress". Samsung Tomorrow. Archived from the original on 16 March 2014. Retrieved 26 February 2013.
- ^ "The future is here: iPhone X". Apple Newsroom. Retrieved 25 February 2018.
- ^ McKenney, Paul (12 June 2012). "A big.LITTLE scheduler update". LWN.net.
- ^ Perret, Quentin (25 February 2019). "Energy Aware Scheduling merged in Linux 5.0". community.arm.com.
- ^ "Energy Aware Scheduling". The Linux Kernel documentation.
- ^ Humrick, Matt (29 May 2017). "Exploring Dynamiq and ARM's New CPUs". Anandtech. Retrieved 10 July 2017.
- ^ Ltd, Arm. "DynamIQ – Arm®". Arm | The Architecture for the Digital World. Retrieved 18 October 2023.
Further reading
[edit]- David Zinman (25 January 2013). "big.LITTLE MP status Jan 25, 2013". LWN.net. Retrieved 25 January 2013.
- Nicolas Pitre (15 February 2012). "Linux support for ARM big.LITTLE". LWN.net. Retrieved 18 October 2012.
- Paul McKenney (12 June 2012). "A big.LITTLE scheduler update". LWN.net. Retrieved 18 October 2012.
- Jake Edge (5 September 2012). "KS2012: ARM: A big.LITTLE update". LWN.net. Retrieved 18 October 2012.
- Jon Stokes (20 October 2011). "ARM's new Cortex A7 is tailor-made for Android superphones". Ars Technica. Retrieved 31 October 2012.
- Andrew Cunningham (30 October 2012). "ARM goes 64-bit with new Cortex-A53 and Cortex-A57 designs". Ars Technica. Retrieved 31 October 2012.
External links
[edit]- big.LITTLE Processing
- big.LITTLE Processing with ARM CortexTM-A15 & Cortex-A7 (PDF) (full technical explanation)