Editing nvidia/tegra/xavier

{{nvidia title|Tegra Xavier}}
{{chip
|name=Xavier
|image=xavier soc chip.png
|designer=Nvidia
|manufacturer=TSMC
|model number=Tegra194
|market=Artificial Intelligence
|market 2=Embedded
|first announced=January 8, 2018
|first launched=June, 2018
|family=Tegra
|isa=ARMv8
|isa family=ARM
|microarch=Carmel
|microarch 2=Volta
|process=12 nm
|transistors=9,000,000,000
|technology=CMOS
|die area=350 mm²
|word size=64 bit
|core count=8
|thread count=8
|tdp=30 W
|tdp typical=20 W
}}
'''Tegra Xavier''' is a {{arch|64}} [[ARM]] high-performance system on a chip for autonomous machines designed by [[Nvidia]] and introduced in [[2018]]. Xavier is incorporated into a number of [[Nvidia]]'s computers including the Jetson Xavier, Drive Xavier, and the Drive Pegasus.

== Overview ==
[[File:xavier overview.png|right|thumb|Overview (HC 30)]]
Xavier is an autonomous machine process designed by [[Nvidia]] and introduced at CES 2018. Silicon came back in the last week of December 2017 with sampling started in the first quarter of 2018. NVIDIA plans on mass production by the end of the year. NVIDIA reported that the product is a result of $2 billion R&D and 8,000 engineering years. The chip is said to have full redundancy and diversity in its functional blocks.

:[[File:xavier block.svg|500px]]

The design targets and architecture started back in [[2014]]. Fabricated on [[TSMC]] [[12 nm process]], the chip itself comprises an eight-core CPU cluster, GPU with additional inference optimizations, [[neural processor|deep learning accelerator]], vision accelerator, and a set of multimedia accelerators providing additional support for machine learning (stereo, LDC, optical flow). The ISP has been enhanced to provide native HDR support, higher precision math without offloading work to the GPU. Xavier features a large set of I/O and has been designed for safety and reliability supporting various standards such as Functional safety [[ISO-26262]] and [[ASIL]] level C. The CPU cluster is fully [[cache coherent]] and the coherency is extended to all the other [[accelerators]] on-chip.

== Architecture ==
=== CPU ===
{{main|nvidia/microarchitectures/carmel|l1=Carmel core}}
The chip features eight control/management {{nvidia|Carmel|l=arch}} cores, Nvidia's own custom {{arch|64}} [[ARM]] cores. Those cores implement [[ARMv8.2]] with [[RAS]] support and safety built-in, including dual-execution mode. The cluster consists of 4 duplexes, each sharing 2 MiB of L2 cache. All cores are fully [[cache coherent]] which is extended to {{nvidia|Volta|the GPU|l=arch}} and all the other accelerators in the chip. Compared to {{\\|Parker}} which was based on {{nvidia|Denver 2|l=arch}}, Nividia reports around 2x the multithreaded performance.

=== GPU ===
{{main|nvidia/microarchitectures/volta|l1=Volta}}
Xavier implements a derivative of their {{nvidia|Volta|l=arch}} GPU with a set of finer changes to address the machine earning market, paticlarly adding inference performance overtraining. It has eight Volta stream multiprocessors along with their standard 128 KiB of L1 cache and a 512 KiB of shared L2. Compared to Parker, Nvidia claims this GPU has 2.1x the graphics performance. Whereas their desktop parts (e.g., GV100) are a very powerful GPU that is used for training, the GPU here is optimized for inference. The most obvious change is that they added int8 support for lower precision to the CUDA tensor cores and those operate at the full 2x throughput of the FP16 FLOPS. There is also 512 CUDA tensor cores, a number that's comparable to Nvidia's top-end models for machine learning (e.g., the GV100 has 672). All of this yields 22.6 tera-operations (int8) per second.

{| class="wikitable"
|-
! Throughput
|-
| 22.6 DL TOPS 8-bit
|-
| 2.8 CUDA TFLOPS FP16
|-
| 1.4 CUDA TFLOPS FP32
|}

=== Accelerators ===
Xavier incorporates a set of accelerators designed to augment the functionality offered by the GPU and CPU in order to provide added flexibility and perhaps offer a way to implement some of the more common set of algorithms slightly more efficiently.
==== Programmable Vision Accelerator ====
Xavier incorporates a Programmable Vision Accelerator (PVA) for processing computer vision. There are actually two exact instances of the PVA on-chip, each can be used in lock-step or independently and are capable of implementing some of the common filter loop and other detection algorithms (e.g. [[Harris corner]], [[fast Fourier transform|FFTs]]). For each of the PVAs, there is a {{armh|Cortex-R5|l=arch}} core along with two dedicated vector processing units, each with its own memory and DMA. The DMA on the PVA is designed to operate on tiles of memory. To that end, the DMA performs the address calculation and can perform prefetching while the processing pipes operate. This is 7-slot VLIW architecture made of 2 scalar slots, 2 vector slots, and 3 memory operations. The pipe is 256 bit wide (slightly wider because of the [[guard bits]] keeping the precision for the operation) and all types can operate at full throughput (32x8b, 16x16b, and 8x32b vector math). The pipe supports additional operations beyond vector such as custom logic for table lookup and hardware looping.

{| class="wikitable"
|-
! Throughput
|-
| 1.7 CV TOPS
|}

It's worth noting that since the chip is expected to be connected to a network of cameras (e.g., side, front, inside), the PVA is capable of doing real-time [[encoding]] for all camera in [[high dynamic range]]. 

{| class="wikitable"
|-
! colspan="2" | Video Processing
|-
! Encode !! Decode
|-
| 1.2 GPIX/s || 1.8 GPIX/s
|}

The chip has an [[ISP]] with native full-range [[HDR]] support and [[tile rendering]] capable of processing at 1.5 GPIX/s.

==== Deep Learning Accelerator ====
The other accelerators on-die is the deep learning accelerator (DLA) which is actually a physical implementation of the open source Nvidia NVDLA architecture. Xavier has two instances of NVDLA which can offer a peak theoretical performance of 5.7 teraFLOPS (half precision FP) or twice the throughput at 11.4 TOPS for int8.

{| class="wikitable"
|-
! Throughput
|-
| 11.4 TOPS (int8)
|-
| 5.7 FLOPS (FP16)
|}

==== Stereo & Optical Flow Engine ====
Xavier has dedicated engines for stereo and optical flow

{| class="wikitable"
|-
! Throughput
|-
| 12.4 TOPS 8-bit
|-
| 6.2 TOPS 16-bit
|}

== Memory controller ==
{{memory controller
|type=LPDDR4X-4266
|ecc=Yes
|channels=8
|width=32 bit
|max bandwidth=127.1 GiB/s
}}

== I/O ==
* [[NVLINK]]
** NVLINK 1.0 (20 GB/s)
* [[PCIE]]
** Multiple 16GT/s gen4 controllers
** x8, x4, x2, x1 configurations
** Root port + Endpoint
* USB
** 3x USB3.1 (10 GT/s) ports
** 4x USB2.0 ports
* DISPLAY
** 4x DP/HDMI/eDP
** 4K @ 60 Hz
** DP HBR3
** HDMI 2.0
* CAMERA
** 16 CSI lanes
** 40 Gbps in DPHY 1.2 Mode
** 109 Gbps in CPHY 1.1 Mode
* OTHER
** Ethernet
** CAN
** UART
** UFS
** SDMMC
** SPIO
** I2C
** I2S
** GPIO

== Die ==
=== SoC ===
* 9,000,000,000 transistors
* 350 mm² die size
* [[TSMC]]'s [[12FFN]]

[[File:nvidia xavier die shot.png|900px]]


[[File:nvidia xavier die shot (annotated).png|900px]]

=== GPU ===
* ~89.2 mm² silicon area

:[[File:xavier die volta gpu.png|500px]]


:[[File:xavier die volta gpu (annotated).png|500px]]

=== CPU ===
See {{nvidia|carmel#CPU_Complex_2|Carmel § CPU Complex|l=arch}}.

=== PVA ===
:[[File:xavier die pva.png|650px]]

:[[File:xavier die pva (annotated).png|650px]]

=== MM Engine / DLA ===
* ~21.75 mm² silicon area

:[[File:xavier die mm-dl accel.png|650px]]

== Borads ==
<gallery mode=packed-hover heights="300px" widths="300px">
jetson_xavier_(front).png|Jetson Xavier, front
jetson_xavier_(back).png|Jetson Xavier, back
</gallery>

== Documents ==
* [[:File:ces2018 - nvidia drive xavier.pdf|CES 2018: Nvidia Drive Xavier]]

== Bibliography ==
* IEEE Hot Chips 30 Symposium (HCS) 2018.
core count	8 +
core name	Carmel +
designer	Nvidia +
die area	350 mm² (0.543 in², 3.5 cm², 350,000,000 µm²) +
family	Tegra +
first announced	January 8, 2018 +
first launched	June 2018 +
full page name	nvidia/tegra/xavier +
has ecc memory support	true +
instance of	microprocessor +
isa	ARMv8 +
isa family	ARM +
ldate	June 2018 +
main image	+
manufacturer	TSMC +
market segment	Artificial Intelligence + and Embedded +
max cpu count	4 +
max memory bandwidth	127.1 GiB/s (130,150.4 MiB/s, 136.473 GB/s, 136,472.586 MB/s, 0.124 TiB/s, 0.136 TB/s) +
max memory channels	8 +
microarchitecture	Carmel + and Volta +
model number	Tegra194 +
name	Xavier +
process	12 nm (0.012 μm, 1.2e-5 mm) +
smp max ways	4 +
supported memory type	LPDDR4X-4266 +
tdp	30 W (30,000 mW, 0.0402 hp, 0.03 kW) +
tdp (typical)	20 W (20,000 mW, 0.0268 hp, 0.02 kW) +
technology	CMOS +
thread count	8 +
transistor count	9,000,000,000 +
word size	64 bit (8 octets, 16 nibbles) +