Mali (processor)

Series of graphics processing units produced by ARM Holdings From Wikipedia, the free encyclopedia

Mali (processor)

The Mali and Immortalis series of graphics processing units (GPUs) and multimedia processors are semiconductor intellectual property cores produced by Arm Holdings for licensing in various ASIC designs by Arm partners.

Quick Facts Release date, Architecture ...
Mali
Thumb
ARM Cortex A57 A53 big.LITTLE SoC with a Mali-T624 GPU
Release date2005
Architecture
  • Utgard
  • Midgard
  • Bifrost
  • Valhall
ModelsSee Variants
Cores1 to 32 cores
Fabrication process4 to 40 nm
API support
DirectX9 to 12
OpenCL1.1 to 3.0
OpenGL2.0 to 3.0
Vulkan1.0 to 1.3
Close

Mali GPUs were developed by Falanx Microsystems A/S, which was a spin-off of a research project from the Norwegian University of Science and Technology.[1] Arm Holdings acquired Falanx Microsystems A/S on June 23, 2006 and renamed the company to Arm Norway.[2]

It was originally named Malaik, but the team shortened the name to Mali, Serbo-Croatian for "small", which was thought to be fitting for a mobile GPU.[3]

On June 28, 2022, Arm announced their Immortalis series of GPUs with hardware-based Ray Tracing support.[4]

Graphics processors

Summarize
Perspective

Utgard

In 2005, Falanx announced their Utgard GPU Architecture, the Mali-200 GPU.[5] Arm followed up with the Mali-300, Mali-400, Mali-450, and Mali-470. Utgard was a non-unified GPU (discrete pixel and vertex shaders).[1]

More information Model, Launch date ...
Comparison of Mali Utgard graphics processing units
Model Launch date Type EUs/Shader core count Core clock rate

(MHz)

L2 cache size Fillrate GFLOPS
(per core)
OpenGL ES
M△/s GT/s (GP/s)
Mali-55/110 2005 Fixed function pipeline[6] 1 2.8 0.1  ? 1.1
Mali-200 2007[7] Programmable pipeline[6] 1 5  ? 0.2 2.0
Mali-300 2010[8] 1 500 8 KiB 55 0.5 5
Mali-400 MP 2008 1–4 200–600 8–256 KiB 55 0.5 1.2–5.4
Mali-450 MP 2012 1–8 300–750 8–512 KiB 142 2.6 4.5–11.9
Mali-470 MP 2015 1–4 250–650 8–256 KiB 71 0.65 8–20.8
Close

Midgard

1st generation

On November 10, 2010, Arm announced their Midgard 1st gen GPU Architecture, including the Mali-T604 and later the Mali-T658 GPU in 2011.[9][10][11][12] Midgard uses a Hierarchical Tiling system.[1]

2nd generation

On August 6, 2012, Arm announced their Midgard 2nd gen GPU Architecture, including the Mali-T678 GPU.[13] Midgard 2nd gen introduced Forward Pixel Kill.[1][14]

3rd generation

On October 29, 2013, Arm announced their Midgard 3rd gen GPU Architecture, including the Mali-T760 GPU.[15][1][16][17][18]

4rd generation

On October 27, 2014, Arm announced their Midgard 4th gen GPU Architecture, including the Mali-T860, Mali-T830, Mali-T820. Their flagship Mali-T880 GPU was announced on February 3, 2015. New microarchitectural features include:[19]

  • Up to 16 cores for the Mali-T880, with 256KB – 2MB L2 cache

Bifrost

1st generation

On May 27, 2016, Arm announced their Bifrost GPU Architecture, including the Mali-G71 GPU. New microarchitectural features include:[20][21]

  • Unified shaders with quad vectorization
  • Scalar ISA
  • Clauses execution
  • Full cache coherency
  • Up to 32 cores for the Mali-G71, with 128KB – 2MB L2 cache
  • Arm claims the Mali-G71 has 40% more performance density and 20% better energy efficiency than the Mali-T880

2nd generation

On May 29, 2017, Arm announced their Bifrost 2nd gen GPU Architecture, including the Mali-G72 GPU. New microarchitectural features include:[22][23]

  • Arithmetic optimizations and increased caches
  • Up to 32 cores for the Mali-G72, with 128KB – 2MB L2 cache
  • Arm claims the Mali-G72 has 20% more performance density and 25% better energy efficiency than the Mali-G71

3rd generation

On May 31, 2018, Arm announced their Bifrost 3rd gen GPU Architecture, including the Mali-G76 GPU. New microarchitectural features include:[24][25]

  • 8 execution lanes per engine (up from 4). Doubled pixel and texel throughput
  • Up to 20 cores for the Mali-G76, with 512KB – 4MB L2 cache
  • Arm claims the Mali-G76 has 30% more performance density and 30% better energy efficiency than the Mali-G72

Valhall

1st generation

On May 27, 2019, Arm announced their Valhall GPU Architecture, including the Mali-G77 GPU, and in October Mali-G57 GPUs. New microarchitectural features include:[26][27][28]

  • New superscalar engine
  • Simplified scalar ISA
  • New dynamic scheduling
  • Up to 16 cores for the Mali-G77, with 512KB – 2MB L2 cache
  • Arm claims the Mali-G77 has 30% more performance density and 30% better energy efficiency than the Mali-G76

2nd generation

On May 26, 2020, Arm announced their Valhall 2nd Gen GPU Architecture, including the Mali-G78. New microarchitectural features include:[29][30][31]

  • Asynchronous clock domains
  • New FMA units and increase Tiler throughput
  • Up to 24 cores for the Mali-G78, with 512KB – 2MB L2 cache
  • Arm Frame Buffer Compression (AFBC)
  • Arm claims the Mali-G78 has 15% more performance density and 10% better energy efficiency than the Mali-G77

3rd generation

On May 25, 2021, Arm announced their Valhall 3rd Gen GPU Architecture (as part of TCS21), including the Mali-G710, Mali-G510, and Mali-G310 GPUs. New microarchitectural features include:[32][33][34]

  • Larger shader cores (2x compared to Valhall 2nd Gen)
  • New GPU frontend, Command Stream Frontend (CSF) replaces the Job Manager
  • Up to 16 cores for the Mali-G710, with 512KB – 2MB L2 cache
  • Arm claims the Mali-G710 has 20% more performance density and 20% better energy efficiency than the Mali-G78

4th generation

On June 28, 2022, Arm announced their Valhall 4th Gen GPU Architecture (as part of TCS22), including the Immortalis-G715, Mali-G715, and Mali-G615 GPUs. New microarchitectural features include:[4][35]

  • Ray Tracing support (hardware-based)
  • Variable Rate Shading[36]
  • New Execution Engine, with doubled the FMA block, Matrix Multiply instruction support, and PPA improvements
  • Arm Fixed Rate Compression (AFRC)
  • Arm claims the Immortalis-G715 has 15% more performance & 15% better energy efficiency than the Mali-G710[37]

5th generation

On May 29, 2023, Arm announced their 5th Gen Arm GPU Architecture (as part of TCS23), including the Immortalis-G720, Mali-G720 and Mali-G620 GPUs.[38][39][40] New microarchitectural features include:[41]

  • Deferred vertex shading (DVS) pipeline
  • Arm claims the Immortalis-G720 has 15% more performance and uses up to 40% less memory bandwidth than the Immortalis-G715

Technical details

Like other embedded IP cores for 3D rendering acceleration, the Mali GPU does not include display controllers driving monitors, in contrast to common desktop video cards. Instead, the Mali ARM core is a pure 3D engine that renders graphics into memory and passes the rendered image over to another core to handle display.

ARM does, however, license display controller SIP cores independently of the Mali 3D accelerator SIP block, e.g. Mali DP500, DP550 and DP650.[42]

ARM also supplies tools to help in authoring OpenGL ES shaders named Mali GPU Shader Development Studio and Mali GPU User Interface Engine.

Display controllers such as the ARM HDLCD display controller are available separately.[43]

Variants

The Mali core grew out of the cores previously produced by Falanx and currently constitute:[44]

More information Model, Micro-archi-tecture ...
Model Micro-
archi-
tecture
Type Launch date EUs/Shader core count Shading Units Total Shaders Fab (nm) Die size (mm2) Core clock rate (MHz) L2 cache size Fillrate GFLOPS
(per core)
GFLOPS
(total)
API (version)
M△/s GT/s (GP/s) Vulkan OpenGL ES OpenCL
Mali-T604[45] Midgard 1st gen Unified shader model +

SIMD ISA

Nov 2010[46] 1–4 32
28
 ? 533 32–256 KiB 90 0.533 17 3.1 Full Profile 1.1
Mali-T658[45] Nov 2011[47] 1–8  ?  ?  ?  ?  ?
Mali-T622 Midgard 2nd gen Jun 2013[48] 1–2 32
28
 ? 533  ?  ? 8.5
Mali-T624 2012-08 1–4  ? 533–600  ?  ? 17–19.2
Mali-T628 1–8  ? 533–695  ?  ? 17–23.7
Mali-T678[49] 1–8 28  ?  ?  ?  ?
Mali-T720 Midgard 3rd gen 2013-10 1–8  ? 400–700 650 (MP8@
650 MHz)
5.2 (MP8
@650 MHz)
6.8–11.9
Mali-T760 1–16 28
14
1.75 mm2 per shader core at 14 nm[50] 600–772 256–2048 KiB[51] 1300 0.6–12.6 GTexel/s (bilinear)[52] 10.4 17–26.2 1.0[53] 3.2[54] Full Profile 1.2
Mali-T820 Midgard 4th gen Q4 2015 1–4 28  ? 600 32–256 KiB[51] 400 ? 2.6 10.2
Mali-T830  ? 600–950 400 ? 2.6 20.4–32.3
Mali-T860 1–16  ? 350–700 256–2048 KiB[51] 1300 ? 10.4 11.9–23.8
Mali-T880 Q2 2016 1–16 16  ? 650–1000 1700 ? 13.6 22.1–34
Mali-G31 Bifrost 1st gen Unified shader model + Unified memory +

scalar, clause-based ISA

Q1 2018 1–6[55] 4 or 8 per core 4–48 28
12
 ? 650 32kB–512kB ? 1.3 8–16 @ 1000 MHz 48–576 @ 1000 MHz 1.3[56] Full Profile 2.0
Mali-G51[57] Q4 2016 1–6[58] 8 or 12 per core 8–72 28
16
14
12
10
 ? 1000 ? 3.9 16–24 @ 1000 MHz 16–144 @ 1000 MHz
Mali-G71[59] Q2 2016 1–32 12 per core 12–384 16
14
10
 ? 546–1037 128–2048 KiB 1850 0.7–24.7

GTexel/s

(bilinear)[60]

27.2 24 @ 1000 MHz 24–768 @ 1000 MHz
Mali-G52 Bifrost 2nd gen Q1 2018 1–6 16 or 24 per core 16–144 16
12
8
7
 ? 850 32-512 KiB ? 6.8 32–48 @ 1000 MHz 32–288 @ 1000 MHz
Mali-G72 Q2 2017 1–32 12 per core 12–384 16
12
10
1.36 mm2 per shader core at 10 nm[61] 572–1050 128–2048 KiB 27.2 24 @ 1000 MHz 24–768 @ 1000 MHz
Mali-G76 Bifrost 3rd gen Q2 2018 4–20 24 per core 96–480 12
8
7
 ? 600–800 512–4096 KiB  ?  ? 48 @ 1000 MHz 192–960 @ 1000 MHz
Mali-G57 Valhall 1st gen Superscalar engine + Unified memory +

simplified scalar ISA

Q2 2019 1–6 32 per core 32–192 12
7
6
? 950[62] 64–512 KiB ? ? ? 64 @ 1000 MHz 64–384 @ 1000 MHz
Mali-G77 7–16 224–512 7
6
? 695–850 512–2048 KiB ? ? ? 448–1024 @ 1000 MHz
Mali-G68 Valhall 2nd gen Q2 2020 1–6 32–192 6
5
3
64–384 @ 1000 MHz
Mali-G78 7–24 224–768 5 759-848 448–1536 @ 1000 MHz
Mali-G310 Valhall 3rd gen Q2 2021 1 16 or 32 or 64 16–64 6
5
4
256–1024 KiB 32–128 @ 1000 MHz
Mali-G510 2–6 48 or 64 per core 96–384 96–128 @ 1000 MHz 192–768 @ 1000 MHz
Mali-G610 1–6 64 per core 64–384 512–2048 KiB 128 @ 1000 MHz 128–768 @ 1000 MHz
Mali-G710 7–16 448–1024 650,850
900
2648 92 896–2048 @ 1000 MHz
Mali-G615 Valhall 4th gen Q2 2022 1–6 128 per core 128–768 4 256 @ 1000 MHz 256–1536 @ 1000 MHz
Mali-G715 7–9 896–1152 1792–2304 @ 1000 MHz
Immortalis-G715 10–16 1280–2048 2560–4096 @ 1000 MHz
Mali-G620 5th Gen[63] Deferred Vertex Shading (DVS) Q2 2023 1–5 128–640 256–1024 KiB 332.8 @ 1300 MHz 332.8–1664 @ 1300 MHz Full Profile 3.0
Mali-G720 6–9 768–1152 512–2048 KiB 1996.8–2995.2 @ 1300 MHz
Immortalis-G720 Q4 2023 10–16 1280–2048 3328–5324.8 @ 1300 
MHz
Mali-G625 Q2 2024 1–5 128–640 4
3
256–1024 KiB 332.8–1664 @ 1300 MHz
Mali-G725 6–9 768–1152 512–4096 KiB 1996.8–2995.2 @ 1300 MHz
Immortalis-G925 10–24 1280–3072 3328–7987.2 @ 1300 
MHz
Model Micro-
archi-
tecture
Type Launch date EUs/Shader core count Shading Units Total Shaders Fab

(nm)

Die size (mm2) Core clock rate (MHz) Max L2 cache size Fillrate (Max core count) FP32 GFLOPS
(per core)
GFLOPS
(total)
Vulkan Open
GL/ES
Open
CL
Close

Some microarchitectures (or just some chips?) support cache coherency for the L2 cache with the CPU.[64][65]

Adaptive Scalable Texture Compression (ASTC) is supported by Mali-T620, T720/T760, T820/T830/T860/T880[66] and Mali-G series.

Implementations

The Mali GPU variants can be found in the following systems on chips (SoCs):

More information Vendor, SoC name ...
Close

Video processors

Summarize
Perspective

Mali Video is the name given to ARM Holdings' dedicated video decoding and video encoding ASIC. There are multiple versions implementing a number of video codecs, such as HEVC, VP9, H.264 and VP8. As with all ARM products, the Mali video processor is a semiconductor intellectual property core licensed to third parties for inclusion in their chips. Real time encode-decode capability is central to videotelephony. An interface to ARM's TrustZone technology is also built-in to enable digital rights management of copyrighted material.

Mali-V500

The first version of a Mali Video processor was the V500, released in 2013 with the Mali-T622 GPU.[119] The V500 is a multicore design, sporting 1–8 cores, with support for H.264 and a protected video path using ARM TrustZone. The 8 core version is sufficient for 4K video decode at 120 frames per second (fps). The V500 can encode VP8 and H.264, and decode H.264, H.263, MPEG4, MPEG2, VC-1/WMV, Real, VP8.

Mali-V550

Released with the Mali-T800 GPU, ARM V550 video processors added both encode and decode HEVC support, 10-bit color depth, and technologies to further reduced power consumption.[120] The V550 also included technology improvements to better handle latency and save bandwidth.[121] Again built around the idea of a scalable number of cores (1–8) the V550 could support between 1080p60 (1 core) to 4K120 (8 cores). The V550 supported HEVC Main, H.264, VP8, JPEG encode, and HEVC Main 10, HEVC Main, H.264, H.263, MPEG4, MPEG2, VC-1/WMV, Real, VP8, JPEG decode.

Mali-V61

The Mali V61 video processor (formerly named Egil) was released with the Mali Bifrost GPU in 2016.[122][123] V61 has been designed to improve video encoding, in particular HEVC and VP9, and to allow for encoding either a single or multiple streams simultaneously.[124] The design continues the 1–8 variable core number design, with a single core supporting 1080p60 while 8 cores can drive 4Kp120. It can decode and encode VP9 10-bit, VP9 8-bit, HEVC Main 10, HEVC Main, H.264, VP8, JPEG and decode only MPEG4, MPEG2, VC-1/WMV, Real, H.263.[125]

Mali-V52

The Mali V52 video processor was released with the Mali G52 and G31 GPUs in March 2018.[126] The processor is intended to support 4K (including HDR) video on mainstream devices.[127]

The platform is scalable from 1 to 4 cores and doubles the decode performance relative to V61. It also adds High 10 H.264 encode (Level 5.0) and decode (Level 5.1) capabilities, as well as AVS Part 2 (Jizhun) and Part 16 (AVS+, Guangdian) decode capability for YUV420.[128]

Mali-V76

The Mali V76 video processor was released with the Mali G76 GPU and Cortex-A76 CPU in 2018.[129] The V76 was designed to improve video encoding and decoding performance. The design continues the 2–8 variable core number design, with 8 cores capable of 8Kp60 decoding and 8Kp30 encoding. It claims improves HEVC encode quality by 25% relative to Mali-V61 at launch. The AV1 codec is not supported.

Mali-V77

The Mali V77 video processor was released with the Mali G77 GPU and Cortex-A77 CPU in 2019.

Comparison

More information V500, V550 ...
Close

Display processors

Mali-D71

The Mali-D71 added Arm Framebuffer Compression (AFBC) 1.2 encoder, support for ARM CoreLink MMU-600 and Assertive Display 5. Assertive Display 5 has support for HDR10 and hybrid log–gamma (HLG).

Mali-D77

The Mali-D77 added features including asynchronous timewarp (ATW), lens distortion correction (LDC), and chromatic aberration correction (CAC)[broken anchor]. The Mali-D77 is also capable of 3K (2880x1440) @ 120 Hz and 4K @ 90 Hz.[134]

More information DP500, DP550 ...
Close

Image signal processors

Summarize
Perspective

Mali-C71

On April 25, 2017 the Mali-C71 was announced, ARM's first image signal processor (ISP).[146][147][148]

Mali-C52 and Mali-C32

On January 3, 2019 the Mali-C52 and C32 were announced, aimed at everyday devices including drones, smart home assistants and security, and internet protocol (IP) camera.[149]

Mali-C71AE

On September 29, 2020 the Mali-C71AE image signal processor was introduced, alongside the Cortex-A78AE CPU and Mali-G78AE GPU.[150] It supports up to 4 real-time cameras or up to 16 virtual cameras with a maximum resolution of 4096 x 4096 each.[151]

Mali-C55

On June 8, 2022 the Mali-C55 ISP was introduced as successor to the C52.[152][153] It is the smallest and most configurable image signal processor from Arm, and support up to 8 camera with a max resolution of 48 megapixel each. Arm claims improved tone mapping and spatial noise reduction compared to the C52. Multiple C55 ISPs can be combined to support higher than 48 megapixel resolutions.

Comparison

More information C32, C52 ...
Close

The Lima, Panfrost and Panthor FOSS drivers

Summarize
Perspective

On January 21, 2012, Phoronix reported that Luc Verhaegen was driving a reverse-engineering attempt aimed at the Mali series of GPUs, specifically the Mali 200 and Mali 400 versions. The project was known as Lima and targeted support for OpenGL ES 2.0.[155] The reverse-engineering project was presented at FOSDEM, February 4, 2012,[156][157] followed by the opening of a website[158] demonstrating some renders. On February 2, 2013, Verhaegen demonstrated Quake III Arena in timedemo mode, running on top of the Lima driver.[159] In May 2018, a Lima developer posted the driver for inclusion in the Linux kernel.[160] In May 2019, the Lima driver became part of the mainline Linux kernel.[161] The Mesa userspace counterpart was merged at the same time. It currently supports OpenGL ES 1.1, 2.0 and parts of Desktop OpenGL 2.1, and the fallback emulation in MESA provides full support for graphical desktop environments.[162]

Panfrost is a reverse-engineered driver effort for Mali Txxx (Midgard) and Gxx (Bifrost) GPUs. Introducing Panfrost[163] talk was presented at X.Org Developer's Conference 2018. As of May 2019, the Panfrost driver is part of the mainline Linux kernel.[164] and MESA. Panfrost supports OpenGL ES 2.0, 3.0 and 3.1, as well as OpenGL 3.1.[165]

Later Collabora has developed[166] panthor driver for G310, G510, G710 GPUs.

See also

  • Adreno – GPU developed by Qualcomm (formerly AMD, then Freescale)
  • Atom family of SoCs – with Intel graphics core, not licensed to third parties
  • AMD mobile APUs – with AMD graphics core, licensed to Samsung[167]
  • PowerVR – by Imagination Technologies
  • Tegra – family of SoCs by Nvidia with the graphics core available as a SIP block to third parties
  • VideoCore – family of SoCs by Broadcom with the graphics core available as a SIP block to third parties
  • Vivante – available as SIP block to third parties
  • Imageon – old AMD mobile GPU

References

Loading related searches...

Wikiwand - on

Seamless Wikipedia browsing. On steroids.