IP Library Granted Patent US 10,606,785
Granted Patent B2
US 10,606,785 · App. 16/171,342 · Granted Mar 31, 2020

Flex bus protocol negotiation and enabling sequence

Inventors: Debendra Das Sharma (Saratoga, CA); Michelle C. Jen (Mountain View, CA); Prahladachar Jayaprakash Bharadwaj (Bangalore, IN); Bruce A. Tennant (Hillsboro, OR); Mahesh Wagh (Portland, OR)
Assignee: Intel Corporation
G06F13/4027G06F13/4282G06F2213/0026
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,606,785
App. No.
16/171,342
Granted
Mar 31, 2020
Kind
B2
Abstract

Systems, methods, and devices can involve a host device that includes a root complex, a link, and an interconnect protocol stack coupled to a bus link. The interconnect protocol stack can include multiplexing logic to select one of a Peripheral Component Interconnect Express (PCIe) upper layer mode, or an accelerator link protocol upper layer mode, the PCIe upper layer mode or the accelerator link protocol upper layer mode to communicate over the link, and physical layer logic to determine one or more low latency features associated with one or both of the PCIe upper layer mode or the accelerator link protocol upper layer mode.

Claims (48)

1. An apparatus comprising:

a fabric that supports multiple interconnect protocols;

a link coupled to the fabric; and

interconnect protocol logic comprising:

multiplexing logic to select one of a Peripheral Component Interconnect Express (PCIe) upper layer mode, or an accelerator link protocol upper layer mode, the PCIe upper layer mode or the accelerator link protocol upper layer mode to communicate over the link, and

physical layer logic to determine one or more low latency features associated with one or both of the PCIe upper layer mode or the accelerator link protocol upper layer mode;

wherein the physical layer logic receives modified Training Set 1/Training Set 2 (TS1/TS2) ordered sets to enable one or more low latency features or one or more low latency features settings.

2. The apparatus of claim 1 , wherein the multiplexing logic is to select one of the PCIe upper layer mode or the accelerator link protocol upper layer mode during link initialization.

3. The apparatus of claim 1 , wherein if the accelerator link protocol upper layer mode is selected, the multiplexing logic is further to enable accelerator link protocols, the accelerator link protocols comprising either a single protocol or multiple protocols.

4. The apparatus of claim 3 , wherein the accelerator link protocols comprise one or more of an accelerator link input/output protocol, an accelerator link device-attached memory protocol, or an accelerator link cache protocol.

5. The apparatus of claim 1 , wherein if the accelerator link protocol upper layer mode is selected, the physical layer logic is to activate low latency features settings associated with the accelerator link mode.

6. The apparatus of claim 1 , wherein the physical layer logic is determine one or more low latency features associated with the accelerator link upper layer mode to activate based on the received modified PCIe TS1/TS2 ordered sets.

7. The apparatus of claim 1 , wherein the low latency features comprise one or more of sync header suppression, skip (SKP) ordered set suppression, and drift buffer mode.

8. The apparatus of claim 1 , wherein the apparatus comprises a root complex, compliant with a PCIe-based interconnect protocol, the root complex comprising hardware and software to

determine a protocol to be used on the link by a two phase process to occur while in PCIe 5.0 Configuration.Lanenum.Wait, Configuration.Lanenum.Accept, and Configuration.Complete stages, and before entering L0 at Gen1 Speed.

9. The apparatus of claim 8 , wherein the two phase process comprises a first phase, during the first phase, the physical layer logic to:

receive, from the root complex, a set of modified TS1 Ordered Sets providing low latency features, and

send, by the physical layer logic, to the root complex, a set of modified TS1 Ordered Sets indicating which low latency features to enable.

10. The apparatus of claim 9 , wherein the first phase is to occur during PCIe 5.0 Configuration.Lanenum.Wait or Configuration.Lanenum.Accept stages.

11. The apparatus of claim 9 , wherein the two phase process comprises a second phase, during the second phase, the physical layer logic to:

receive, from the root complex, an enable request including a set of modified TS2 Ordered Sets indicating whether the link is to operate in PCIe upper layer mode or an accelerator link protocol upper layer mode,

send, by the physical layer logic, to the root complex, a set of modified TS2 Ordered Sets with an enable response to acknowledge the enable request.

12. The apparatus of claim 11 , wherein the second phase is to occur during PCIe 5.0 Configuration.Complete.

13. A non-transitory computer-readable storage media comprising a plurality of instructions that in response to being executed cause multiplexing logic to select either Peripheral Component Interconnect Express (PCIe)-based upper layer mode, or an accelerator link protocol upper layer mode; and

cause physical layer logic coupled to a link to determine one or more low latency features associated with one or both of the PCIe upper layer mode or the accelerator link protocol upper layer mode;

wherein the instructions use modified PCIe 5.0 Training Set 1/Training Set 2 (TS1/TS2) ordered sets (OS) to negotiate which accelerator link protocols are to be enabled.

14. The non-transitory computer-readable storage media of claim 13 , further comprising instructions that in response to being executed cause multiplexing logic is to select one of the PCIe upper layer mode or the accelerator link protocol upper layer mode during link initialization.

15. The non-transitory computer-readable storage media of claim 13 , further comprising instructions that in response to being executed cause multiplexing logic, if the accelerator link mode was selected, to enable either a single protocol, or multiple protocols.

16. The non-transitory computer-readable storage media of claim 13 , further comprising instructions that in response to being executed cause the physical layer logic, if the accelerator link mode was selected, to enable low latency features associated with the accelerator link mode.

17. The non-transitory computer-readable storage media of claim 13 , wherein the instructions use modified PCIe TS1/TS2 ordered sets to negotiate which accelerator link low latency features are to be activated.

18. The non-transitory computer-readable storage media of claim 13 , wherein the accelerator link protocol upper layer mode comprises one or more protocols, the one or more protocols comprising an accelerator link input/output protocol, an accelerator link device-attached memory protocol, or an accelerator link cache protocol.

19. The non-transitory computer-readable storage media of claim 13 , wherein the low latency features include one or more of sync header suppression, skip (SKP) ordered set suppression, and drift buffer mode.

20. A system comprising:

a host device, the host device comprising:

a root complex;

a fabric that supports multiple interconnect protocols; and

a link coupled to the fabric;

an interconnect protocol stack that supports multiple interconnect protocols, the interconnect protocol stack coupled to the link, the interconnect protocol stack comprising:

multiplexing logic to select one of a Peripheral Component Interconnect Express (PCIe) upper layer mode, or an accelerator link protocol upper layer mode, the PCIe upper layer mode or the accelerator link protocol upper layer mode to communicate over the link, and

physical layer logic to determine one or more low latency features associated with one or both of the PCIe upper layer mode or the accelerator link protocol upper layer mode; and

an endpoint device, the endpoint device coupled to the host device across the link;

wherein the root complex is to:

transmit to the endpoint device a first set of modified training set (TS) 1 ordered sets, the first set of modified TS1 ordered sets comprising an advertisement of one or more low latency features to activate; and

receive, from the endpoint device, a second set of modified TS1 ordered sets, the second set of modified TS1 ordered sets identifying one or more low latency features to activate.

21. The system of claim 20 , wherein the root complex is to:

transmit a first set of modified TS2 ordered sets to the endpoint device, the first set of modified TS2 ordered sets indicating whether the endpoint device should operate in PCIe upper layer mode or accelerator link upper layer mode; and

receive a second set of modified TS2 ordered sets from the endpoint device, the second set of modified TS2 ordered sets indicating a PCIe upper layer mode or an accelerator link upper layer mode.

22. The system of claim 21 , wherein the received second set of modified TS2 ordered sets indicates the accelerator link mode is selected by the endpoint device, and wherein the physical layer logic is to activate low latency features settings associated with the accelerator link mode, the low latency features settings indicating one or more low latency features comprising one or more of sync header suppression, skip (SKP) ordered set suppression, and drift buffer mode.

Assignments (2)
CORRECTIVE ASSIGNMENT TO CORRECT THE ASSIGNMENT DOCUMENT TO INCLUDE THE APPLICATION SERIAL NUMBER AND FILING DATE FILING DATE IS OCTOBER 25, 2018 AND THE U.S. PATENT APPLICATION SERIAL NUMBER IS 16/171,342.. PREVIOUSLY RECORDED AT REEL: 048322 FRAME: 0455. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT. Recorded Mar 2, 2022
From: DAS SHARMA, DEBENDRA; JEN, MICHELLE C.; BHARADWAJ, PRAHLADACHAR JAYAPRAKASH; TENNANT, BRUCE; WAGH, MAHESH
To: INTEL CORPORATION
Reel/Frame 059354/0182 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 13, 2019
From: DAS SHARMA, DEBENDRA; JEN, MICHELLE C.; BHARADWAJ, PRAHLADACHAR JAYAPRAKASH; TENNANT, BRUCE; WAGH, MAHESH
To: INTEL CORPORATION
Reel/Frame 048322/0455 →
Continuity (2)
Provisional Application 62667324 · May 4, 2018
Related Publication 20190065426A1 · Feb 28, 2019
Cited By (7)
US 12,197,352 US 12,321,305 US 12,353,305 US 12,405,912 US 12,481,614 US 12,591,727 US 12,625,827