Security Updates, Improved Instruction Performance and AVX-512 Updates

With every new microarchitecture update, there are goals on several fronts: add new instructions, decrease the latency of current instructions, increase the throughput of current instructions, and remove bugs. The big headline addition for Sunny Cove and Ice Lake is AVX-512, which hasn’t yet appeared on a mainstream widely distributed consumer processor – technically we saw it in Cannon Lake, but that was a limited run CPU. Nonetheless, a lot of what went into Cannon Lake also shows up in the Sunny Cove design. To complicate matters, AVX-512 comes in plenty of different flavors. But on top of that, Intel also made a significant number of improvements to a number of instructions throughout the design.

Big thanks to InstLatX64 for his help in analyzing the benchmark results.


On security, almost all the documented hardware security fixes are in place with Sunny Cove. Through the CPUID results, we can determine that SSBD is enabled, as is IA32_ARCH_CAPABILITIES, L1D_FLUSH, STIBP, IBPB/IBRS and MD_CLEAR.

This aligns with Intel’s list of Sunny Cove security improvements:

Sunny Cove Security
AnandTech Description Name Solution
BCB Bound Check Bypass Spectre V1 Software
BTI Branch Target Injection Spectre V2 Hardware+OS
RDCL Rogue Data Cache Load V3 Hardware
RSSR Rogue System Register Read V3a Hardware
SSB Speculative Store Bypass V4 Hardware+OS
L1TF Level 1 Terminal Fault Foreshadow Hardware
MFBDS uArch Fill Buffer Data Sampling RIDL Hardware
MSBDS uArch Store Buffer Data Sampling Fallout Hardware
MLPDS uArch Load Port Data Sampling - Hardware
MDSUM uArch Data Sampling Uncachable Memory - Hardware

Aside from Spectre V1, which has no suitable hardware solution, almost all of the rest have been solved through hardware/firmware (Intel won’t distinguish which, but to a certain extent it doesn’t matter for new hardware). This is a step in the right direction, but of course it may have a knock-on effect, plus for anything that gets performance improvements being moved from firmware to hardware will be rolled into any advertised IPC increase.

Also on the security side is SGX, or Intel’s Software Guard Instructions. Sunny Cove now becomes Intel’s first public processor to enable both AVX-512 and SGX in the same design. Technically the first chip with both SGX and AVX-512 should have been Skylake-X, however that feature was ultimately disabled due to failing some test validation cases. But it now comes together for Sunny Cove in Ice Lake-U, which is also a consumer processor.

Instruction Improvements and AVX-512

As mentioned, Sunny Cove pulls a number of key improvements from the Cannon Lake design, despite the Cannon Lake chip having the same cache configuration as Skylake. One of the key points here is the 64-bit division throughput, which goes from a 97-cycle latency to an 18-cycle latency, blowing past AMD’s 45-cycle latency. As an ex-researcher with no idea about instruction latency or compiler options, working on high-precision math code, this speedup would have been critical.

  • IDIV -> 97-cycle to 18-cycle

For the general purpose registers, we see a lot of changes, and most of them quite sizable.

Sunny Cove GPR Changes
AnandTech Instruction Skylake Sunny Cove
Complex LEA Complex Load Effective Address 3 cycle latency
1 per cycle
1 cycle latency
2 per cycle
SHL/SHR Shift Left/Right 2 cycle latency
0.5 per cycle
1 cycle latency
1 per cycle
ROL/ROR Rotate Left/Right 2 cycle latency
0.5 per cycle
1 cycle latency
1 per cycle
SHLD/SHRD Double Precision Shift Left/Right 4 cycle latency
0.5 per cycle
4 cycle latency
1 per cycle
4*MOV Four repated string MOVS Limited instructions 104 bits/clock
All MOVS* Instructions

In the past we’ve seen x87 instructions being regressed, made slower, as they become obsolete. For whatever reason, Sunny Cove decreases the FMUL latency from 5 cycles to 4 cycles.

The SIMD units also go through some changes:

Sunny Cove SIMD
AnandTech Instruction Skylake Sunny Cove
SIMD Packing SIMD Packing now slower 1 cycle latency
1 per cycle
3 cycle latency
1 per cycle
AES* AES Crypto Instructions
(for 128-bit / 256-bit)
4 cycle latency
2 per cycle
3 cycle latency
2 per cycle
CLMUL Carry-Less Multiplication 7 cycle latency
1 per cycle
6 cycle latency
1 per cycle
PHADD/PHSUB Packed Horizontal Add/Subtract
and Saturate
3 cycle latency
0.5 per cycle
2 cycle latency
1 per cycle
VPMOV* xmm Vector Packed Move 2 cycle latency
0.5 per cycle
2 cycle latency
1 per cycle
VPMOV* ymm Vector Packed Move 4 cycle latency
0.5 per cycle
2 cycle latency
1 per cycle
VPMOVZX/SX* xmm Vector Packed Move 1 cycle latency
1 per cycle
1 cycle latency
2 per cycle
POPCNT Microcode 50% faster than SW (under L1-D size)
REP STOS* Repeated Store String 62 bits/cycle 54 bits/cycle
VPCONFLICT Still Microcode Only

We’ve already gone through all of the new AVX-512 instructions in our Sunny Cove microarchitecture disclosure. These include the following families:

  • AVX-512_VNNI (Vector Neural Network Instructions)
  • AVX-512_VBMI (Vector Byte Manipulation Instructions)
  • AVX-512_VBMI2 (second level VBMI)
  • AVX-512_ BITALG (bit algorithms)
  • AVX-512_IFMA (Integer Fused Multiply Add)
  • AVX-512_VAES (Vector AES)
  • AVX-512_VPCLMULQDQ (Carry-Less Multiplacation of Long Quad Words)
  • AVX-512+GFNI (Galois Field New Instructions)
  • SHA (not AVX-512, but still new)
  • GNA (Gaussian Neural Accelerator)

(Intel also has the GMM (Gaussian Mixture Model) inside the core since Skylake, but I’ve yet to see any information on this outside a single line in the coding manual.)

For all these new AVX-512 instructions, it’s worth noting that they can be run in 128-bit, 256-bit, or 512-bit mode, depending on the data types passed to it. Each of these can have corresponding latencies and throughputs, which often get worse when going for the 512-bit mode, but overall assuming you can fill the register with a 512-bit data type, then the overall raw processing will be faster, even with the frequency differential. This doesn’t take into account any additional overhead for entering the 512-bit power state, it should be noted.

Most of these new instructions are relatively fast, with most of them only 1-3 cycles of latency. We observed the following:

Sunny Cove Vector Instructions
AnandTech Instruction XMM YMM ZMM
VNNI Latency Vector Neural Network Instructions 5-cycle 5-cycle 5-cycle
Throughput 2/cycle 2/cycle 1/cycle
VPOPCNT* Latency Return the number of bits set to 1 3-cycle 3-cycle 3-cycle
Throughput 1/cycle 1/cycle 1/cycle
VPCOMPRESS* Latency Store Packed Data 3-cycle 3-cycle 3-cycle
Throughput 0.5/cycle 0.5/cycle 0.5/cycle
VPEXPAND* Latency Load Packed Data 5-cycle 5-cycle 5-cycle
Throughput 0.5/cycle 0.5/cycle 0.5/cycle
VPSHLD* Latency Vector Shift 1-cycle 1-cycle 1-cycle
Throughput 2/cycle 2/cycle 1/cycle
VAES* Latency Vector AES Instructions 3-cycle 3-cycle 3-cycle
Throughput 2/cycle 2/cycle 1/cycle
VPCLMUL Latency Vector Carry-Less Multiply 6-cycle 8-cycle 8-cycle
Throughput 1/cycle 0.5/cycle 0.5/cycle
GFNI Latency Galois Field New Instructions 3-cycle 3-cycle 3-cycle
Throughput 2/cycle 2/cycle 1/cycle

For all of the common AVX2 instructions, xmm/ymm latencies and throughputs are identical to Skylake, however zmm is often a few cycles slower for DIV/SQRT variants.

Other Noticeable Observations

From our testing, we were also able to prove some of the other parts of the core, such as the added store ports and shuffle units.

Our data shows that the second store port is not identical to the first, which explains the imbalance when it comes to writes: rather than supporting 2x64-bit with loads, it only supports either 1x64-bit write, or 1x32-bit write, or 2x16-bit writes. This means we mainly see speed ups with GPR/XMM data, and the result is only a small improvement for 512-bit SCATTER instructions. Otherwise, it seems not to work with any 256-bit or 512-bit operand (you can however use it with 64-bit AVX-512 mask registers). This is going to cause a slight headache for anyone currently limited by SCATTER stores.

The new shuffle unit is only 256-bit wide. It will handle a number of integer instructions (UNPCK, PSLLDQ, SHUF*, MOVSHDUP, but not PALIGNR or PACK), but only a couple of floating point instructions (SHUFPD, SHUFPS).

Cache and TLB Updates SPEC2017 and SPEC2006 Results (15W)


View All Comments

  • rangerdavid - Friday, August 2, 2019 - link

    What Moizy said. Ian, you are quite fair and diplomatic in your responses. And if you are in some kind of Intel marketing conspiracy, for heaven's sake, I hope you are getting a good cut! Buy yourself something pretty, fella... (grin) Reply
  • 0ldman79 - Friday, August 2, 2019 - link

    Rock on man.

    Well said.
  • Sailor23M - Monday, August 5, 2019 - link

    +1 Moizy thanks. I really do not have the time to go searching and collecting all the leaks out there, so this article was well timed for me and at a high level lets me know what to expect from these chips. Reply
  • close - Monday, August 5, 2019 - link

    @Moizy: I guess props to AT if this turns out to be the real situation.

    Curious what happens if we have another "oh our puff piece didn't notice they were using a sub 0 chiller under the table" kind of situation. Last time there was an anemic "oh, yeah, Intel could have been more straight forward" type reaction. So you can understand why people are skeptical about results that can't be independently validated, even (especially? ...given past experiences) if they come from AT.
  • tijag - Thursday, August 1, 2019 - link

    You sir are a first rate obtuse troll. Reply
  • chowmanga - Thursday, August 1, 2019 - link

    Would you say there was a difference between getting a product in advance and getting one when the rest of the press gets one? Reply
  • Moizy - Thursday, August 1, 2019 - link

    To me, the only meaningful difference is timing. The unsound logic of the argument is a) Intel invites a few select press to a preview event months before availability, b) the select few press are flattered by the privilege and develop positively biased feelings toward Intel, therefore c) they report positively biased, flawed findings due to the privilege.

    It's true that inviting a select few to the event could psychologically influence those few, making them slightly less objective than they may have otherwise been. But Ian's reporting isn't built on subjective opinions on Icy Lake, 10nm, and Intel. His reporting is built on his objective testing, using his standardized benchmarks. So even if Ian's view of Intel and Icy Lake were a little swayed by this privilege (which I doubt they were, he's not a new kid on the block, he deals with PR and Marketing and the like all the time), for this to translate into flawed, biased reporting, Ian would need to purposefully alter his standardized benchmarks in order to produced positively biased results. There is zero evidence that he did that here, and there is zero precedence to him doing that in the past, so the original logic is extremely flawed.
  • uberDoward - Friday, August 2, 2019 - link

    Based on the fact (as Ian mentioned at the beginning of the article) it DOES sound like others that were invited came unprepared, and willing to just spout Intel's rhetoric without objective measurements. Kudos to Ian for arriving prepared! Reply
  • Santoval - Friday, August 2, 2019 - link

    You are ranting but saying nothing substantial or even anything that makes sense. "Conflict of interest"? Really? In what way exactly, do you think they gifted Ian and the other AnandTech editors a few hundreds of Intel shares each? Maybe they gave them free vacations to Bali just to "thank" them? Or free top-end Intel based laptops and Intel SSDs for the next 5 years?

    These examples *would* be conflicts of interest. Previewing an Ice Lake reference design while pointing out an entire list of caveats and limitations is not an example of a conflict of interest. "Professional distance" does not mean declining a product preview (why on Earth should they?), it means keeping an equal distance from all the companies the products of which you preview.
  • bcronce - Friday, August 2, 2019 - link

    I learned a lot. Like trade-offs between latency and throughput in several different ways. Or increase in cache-hits vs latency. Reply

Log in

Don't have an account? Sign up now