ARM Cortex-X1

Microprocessor core model by ARM

ARMv8.1-A, ARMv8.2-A, cryptography, RAS, ARMv8.3-A LDAPR instructions, ARMv8.4-A dot product

Physical specificationsCores

1–4 per cluster

Products, models, variantsProduct code name(s)

Hera

Variant(s)

ARM Cortex-A78, ARM Neoverse V1

HistorySuccessor(s)ARM Cortex-X2

The ARM Cortex-X1 is a central processing unit implementing the ARMv8.2-A 64-bit instruction set designed by ARM Holdings' Austin design centre as part of ARM's Cortex-X Custom (CXC) program.^[1]^[2]

Design

The Cortex-X1 design is based on the ARM Cortex-A78, but redesigned for purely performance instead of a balance of performance, power, and area (PPA).^[1]

The Cortex-X1 is a 5-wide decode out-of-order superscalar design with a 3K macro-OP (MOPs) cache. It can fetch 5 instructions and 8 MOPs per cycle, and rename and dispatch 8 MOPs, and 16 μOPs per cycle. The out-of-order window size has been increased to 224 entries. The backend has 15 execution ports with a pipeline depth of 13 stages and the execution latencies consists of 10 stages. It also features 4x128b SIMD units.^[3]^[4]^[5]^[6]

ARM claims the Cortex-X1 offers 30% faster integer and 100% faster machine learning performance than the ARM Cortex-A77.^[3]^[4]^[5]^[6]

The Cortex-X1 supports ARM's DynamIQ technology, expected to be used as high-performance cores when used in combination with the ARM Cortex-A78 mid and ARM Cortex-A55 little cores.^[1]^[2]

Architecture changes in comparison with ARM Cortex-A78

Around 20% performance improvement (+30% from A77)^[7]
- 30% faster integer
- 100% faster machine learning performance
Out-of-order window size has been increased to 224 entries (from 160 entries)
Up to 4x128b SIMD units (from 2x128b)
15% more silicon area
5-way decode (from 4-way)
8 MOPs/cycle decoded cache bandwidth (from 6 MOPs/cycle)
64 KB L1D + 64 KB L1I (from 32/64 KB L1)
Up to 1 MB/core L2 cache (from 512 KB/core max)
Up to 8 MB L3 cache (from 4 MB max)

Licensing

The Cortex-X1 is available as SIP core to partners of their Cortex-X Custom (CXC) program, and its design makes it suitable for integration with other SIP cores (e.g. GPU, display controller, DSP, image processor, etc.) into one die constituting a system on a chip (SoC).^[1]^[2]

Usage

Samsung Exynos 2100^[8]
Qualcomm Snapdragon 888(+)^[9]
Google Tensor^[10]

References

^ ^a ^b ^c ^d "Introducing the Arm Cortex-X Custom program". community.arm.com. Retrieved 2020-06-18.
^ ^a ^b ^c Ltd, Arm. "Cortex-X Custom CPU program". Arm | The Architecture for the Digital World. Retrieved 2020-06-18.
^ ^a ^b Frumusanu, Andrei. "Arm's New Cortex-A78 and Cortex-X1 Microarchitectures: An Efficiency and Performance Divergence". www.anandtech.com. Retrieved 2020-06-18.
^ ^a ^b "Arm Cortex-X1: The First From The Cortex-X Custom Program". WikiChip Fuse. 2020-05-26. Retrieved 2020-06-18.
^ ^a ^b McGregor, Jim. "Arm Unleashes CPU Performance With Cortex-X1". Forbes. Retrieved 2020-06-18.
^ ^a ^b "Arm Cortex-X1 and Cortex-A78 CPUs: Big cores with big differences". Android Authority. 2020-05-26. Retrieved 2020-06-18.
^ "Cortex-X1 – Microarchitectures – ARM – WikiChip". en.wikichip.org. Retrieved 2021-02-13.
^ "Exynos 2100 5G Mobile Processor: Specs, Features | Samsung". Samsung Semiconductor. Retrieved 2021-01-13.
^ "Qualcomm Snapdragon 888 5G Mobile Platform | Latest 5G Snapdragon Processor | Qualcomm". www.qualcomm.com. Retrieved 2021-01-13.
^ Amadeo, Ron (2021-10-19). "The "Google Silicon" team gives us a tour of the Pixel 6's Tensor SoC". Ars Technica.

Application ARM-based chips

Application
processors
(32-bit)

ARMv7-A

Cortex-A5	Actions ATM702x Amlogic M805/S805, T82x Atmel SAMA5D3 InfoTM iMAPx820, iMAPx15 Qualcomm Snapdragon S4 Play, 200 RDA RDA8810PL Telechips TCC892x
Cortex-A7	Allwinner A2x, A3x, A83T, H3, H8 NXP i.MX7, QorIQ LS10xx, NXP i.MX6UL Broadcom VideoCore BCM2836, BCM23550 Leadcore LC1813, LC1860/C, LC1913, LC1960 Marvell Armada PXA1920, 1500 mini plus MediaTek MT65xx Qualcomm Snapdragon 200, 400
Cortex-A8	Allwinner A1x Apple A4 Freescale i.MX5 Rockchip RK291x Samsung Exynos 3110(S5PC110), S5PV210 Texas Instruments OMAP 3 Texas Instruments Sitara AM3xxx Texas Instruments DM38x ZiiLABS ZMS-08
Cortex-A9	Actions ATM702x, ATM703x Altera Cyclone V, Arria V/10 Amlogic AML8726, MX, M6x, M801, M802/S802, S812, T86x Apple A5, A5X Broadcom VideoCore BCM21xxx, BCM28xxx Freescale i.MX6 HiSilicon K3V2, 910's InfoTM iMAPx912 Leadcore LC1810, LC1811 Marvell Armada 1500 mini MediaTek MT65xx Nvidia Tegra, 2, 3, 4i Nufront NuSmart 2816M, NS115, NS115M Renesas EMMA EV2, R-Car H1, RZ/A Rockchip RK292x, RK30xx, RK31xx Samsung Exynos 4 421x, 441x ST-Ericsson NovaThor Telechips TCC8803 Texas Instruments OMAP 4 Texas Instruments Sitara AM4xxx VIA WonderMedia WM88x0, 89x0 Xilinx Zynq-7000 ZiiLABS ZMS-20, ZMS-40
Cortex-A15	Allwinner A80 HiSilicon K3V3 MediaTek MT8135/V Nvidia Tegra 4, K1 Renesas R-Car H2 Samsung Exynos 5 52xx, 54xx Texas Instruments OMAP 5, DRA7xx, AM57xx Texas Instruments Sitara AM5xxx
Cortex-A17	MediaTek MT6595, MT5595 MStar 6A928 Rockchip RK3288
Others	Cortex-A12
ARMv7-A compatible	Apple A6, A6X, S1, S1P, S2, S3 Broadcom Brahma-B15 Marvell P4J Qualcomm Snapdragon S1, S2, S3, S4 Plus, S4 Pro, 600, 800 (Scorpion, Krait)

ARMv8-A

Others	Cortex-A32

Application
processors
(64-bit)

ARMv8-A

Cortex-A35	NXP i.MX8X MediaTek MT6799, MT8516 Rockchip RK3308
Cortex-A53	Actions GT7, S900, V700 Allwinner A64, H5, H64, R18 Altera Stratix 10 Amlogic S9 Family, T96x Broadcom BCM2837 EZchip TILE-Mx100 HiSilicon Kirin 620, 650, 655, 658, 659, 930, 935 Marvell Armada PXA1928, Mobile PXA1908/PXA1936 MediaTek MT673x, MT675x, MT6761V, MT6762/V, MT6763T, MT6765/G/H, MT6795, MT8161, MT8163, MT8165, MT8732, MT8735, MT8752 NXP ARM S32, QorIQ LS1088, LS1043, i.MX8M Qualcomm Snapdragon 215, 410, 412, 415, 425, 427, 430, 435, 429, 439, 450, 610, 615, 616, 617, 625, 626, 630 Renesas RZ/V2M Rockchip RK3328, RK3368 Samsung Exynos 7570, 7578, 7580, 7870, 7880 Texas Instruments Sitara AM6xxx UNISOC SC9820E, SC9832E, SC9860/GV Xilinx ZynqMP
Cortex-A57	AMD Opteron A1100-series NXP QorIQ LS20xx Nvidia Tegra X1 and Tegra X2 Qualcomm Snapdragon 808 and 810 Samsung Exynos 7 5433, 7420 HiSilicon Kirin Hi1610 and Hi1612
Cortex-A72	AWS Graviton Broadcom BCM2711 HiSilicon Kirin 950, 955, Kunpeng 916 MediaTek MT6797/D/T/X, MT8173, MT8176, MT8693 MStar 6A938 Qualcomm Snapdragon 650, 652, 653 Rockchip RK3399 NXP QorIQ LS2088, QorIQ LS1046A, QorIQ LX2160A, QorIQ LS1028A, i.MX8
Cortex-A73	Qualcomm Snapdragon 460, 636, 660, 632, 662, 665, 680, 685, 835 Samsung Exynos 7872, 7884, 7885, 7904, 9609, 9610, 9611 HiSilicon Kirin 710, 960, 970 MediaTek MT6771/V, MT6799, MT8183, MT8788 Amlogic S922X
Others	Cortex-A34
ARMv8-A compatible	Ampere eMAG Apple A7, A8, A8X, A9, A9X, A10, A10X Applied Micro X-Gene Cavium ThunderX Nvidia Tegra K1 (Denver), Tegra X2 (Denver2) Qualcomm Kryo, Falkor Samsung Exynos M1 (Mongoose), M2 (Mongoose)

ARMv8.1-A

ARMv8.1-A compatible	Cavium ThunderX2

ARMv8.2-A

Cortex-A55	Samsung Exynos 850 UNISOC SC9863/A, T603
Cortex-A75	Qualcomm Snapdragon 670, 710, 712, 845, 850 Samsung Exynos 9820, 9825 MediaTek MT6769H/T/V/Z, MT6768, MT6779V UNISOC T310, T606, T610, T616, T618, T619, T700, T710, T740
Cortex-A76	Google Tensor HiSilicon Kirin 810, 820, 980, 985, 990 Qualcomm Snapdragon 480(+), 675, 678, 720G, 730(G), 732G, 765(G), 768G, 855(+) and 860, 7c, 7c Gen 2, 8c, 8cx and 8cx Gen 2 Microsoft SQ1 and SQ2 MediaTek MT6781, MT6785V, MT6789, MT6833V/P, MT6853V/T, MT6873, MT6875, Dimensity 6020, 6080, 6100+, MT8192 Samsung Exynos 990 UNISOC T760, T770, T820
Cortex-A77	MediaTek Dimensity 1000(+) Qualcomm Snapdragon 690, 750G, 865(+) and 870 HiSilicon Kirin 9000 Samsung Exynos 880, 980
Cortex-A78	Google Tensor G2 MediaTek MT6877, MT6878, MT6879, MT6891, MT6893, Dimensity 7020, 7030, 7050, 8000, 8020, 8050, 8100, 8200, Kompanio 900T, 1200, 1380, 1300T Qualcomm Snapdragon 4 Gen 1, 4 Gen 2, 695, 6 Gen 1, 778G(+), 780G, 782G, 888(+) Samsung Exynos 1080, 1280, 1330, 1380, 2100
Cortex-X1	Google Tensor, Tensor G2 Qualcomm Snapdragon 888(+) Samsung Exynos 2100
Neoverse N1	Ampere Altra, Altra Max AWS Graviton2
Others	Cortex-A65, Cortex-A65AE, Cortex-A76AE, Cortex-A78C, Cortex-X1C, Neoverse E1
ARMv8.2-A compatible	Apple A11 Fujitsu A64FX HiSilicon TaiShan v110 Nvidia Tegra Xavier (Carmel) Samsung Exynos M3 (Meerkat), M4 (Cheetah), M5 (Lion)

ARMv8.3-A

ARMv8.3-A compatible	Apple A12, A12X/A12Z, S4, S5 Marvell ThunderX3

ARMv8.4-A

Neoverse V1	AWS Graviton3
ARMv8.4-A compatible	Apple A13, S6, S7, S8

ARMv8.5-A

ARMv8.5-A compatible	Apple A14, M1

ARMv8.6-A

ARMv8.6-A compatible	Apple A15, A16, M2, M3

ARMv9.0-A

Cortex-A510	MediaTek Dimensity 7200 (Pro/Ultra), 8300 (Ultra), 9000/9000+, 9200 Qualcomm Snapdragon 7 Gen 1, 7s/7+ Gen 2, 7 Gen 3, 8(+) Gen 1, 8 Gen 2 Samsung Exynos 2200
Cortex-A710	MediaTek Dimensity 9000/9000+ Qualcomm Snapdragon 7 Gen 1, 7s/7+ Gen 2, 8(+) Gen 1, 8 Gen 2 Samsung Exynos 2200
Cortex-A715	MediaTek Dimensity 7200 (Pro/Ultra), 8300 (Ultra), 9200 Qualcomm Snapdragon 7 Gen 3, 8 Gen 2
Cortex-X2	MediaTek Dimensity 9000/9000+ Qualcomm Snapdragon 7s/7+ Gen 2, 8(+) Gen 1 Samsung Exynos 2200
Cortex-X3	Google Tensor G3 MediaTek Dimensity 9200/9200+ Qualcomm Snapdragon 8 Gen 2
Neoverse N2	Alibaba YiTian 710 Microsoft Azure Cobalt 100
Neoverse V2	AWS Graviton4 Nvidia Grace

ARMv9.2-A

Cortex-A520	Qualcomm Snapdragon 7+ Gen 3, Snapdragon 8(s) Gen 3 Samsung Exynos 2400
Cortex-A720	MediaTek Dimensity 9300 Qualcomm Snapdragon 7+ Gen 3, Snapdragon 8(s) Gen 3 Samsung Exynos 2400
Cortex-X4	MediaTek Dimensity 9300 Qualcomm Snapdragon 7+ Gen 3, Snapdragon 8(s) Gen 3 Samsung Exynos 2400
Neoverse N3	-
Neoverse V3	-