[go: up one dir, main page]

CN105512071B - High speed interface host-side controller - Google Patents

High speed interface host-side controller Download PDF

Info

Publication number
CN105512071B
CN105512071B CN201510894510.2A CN201510894510A CN105512071B CN 105512071 B CN105512071 B CN 105512071B CN 201510894510 A CN201510894510 A CN 201510894510A CN 105512071 B CN105512071 B CN 105512071B
Authority
CN
China
Prior art keywords
physical layer
logical network
speed data
clock signal
electronic physical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201510894510.2A
Other languages
Chinese (zh)
Other versions
CN105512071A (en
Inventor
王万丰
冀晓亮
惠志强
侯慧瑛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Zhaoxin Semiconductor Co Ltd
Original Assignee
Shanghai Zhaoxin Integrated Circuit Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Zhaoxin Integrated Circuit Co Ltd filed Critical Shanghai Zhaoxin Integrated Circuit Co Ltd
Priority to CN201510894510.2A priority Critical patent/CN105512071B/en
Publication of CN105512071A publication Critical patent/CN105512071A/en
Application granted granted Critical
Publication of CN105512071B publication Critical patent/CN105512071B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/38Information transfer, e.g. on bus
    • G06F13/382Information transfer, e.g. on bus using universal interface adapter
    • G06F13/385Information transfer, e.g. on bus using universal interface adapter for adaptation of a particular data processing system to different peripheral devices
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2213/00Indexing scheme relating to interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F2213/38Universal adapter
    • G06F2213/3802Harddisk connected to a computer port

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Transfer Systems (AREA)

Abstract

The host-side controller of low data dithering, low speed data is provided with logical network layer, transfers to electronic physical layer to be converted to high speed data delivery to external device (ED) via across time domain data transport module.The clock signal of the electronic physical layer operation is also transferred to the logical network layer, the logical network layer is provided first low speed data according to this.It is the low speed data that the external device (ED) provides that across the time domain data transport module reads in the logical network layer according to logical network layer end clock, and exports first low speed data to the electronic physical layer according to electronic physical layer end clock.

Description

高速数据接口主机端控制器High-speed data interface host-side controller

技术领域technical field

本发明涉及高速数据接口主机端控制器,特别涉及与外部装置作高速传输的高速数据接口主机端控制器。The invention relates to a high-speed data interface host controller, in particular to a high-speed data interface host controller for high-speed transmission with external devices.

背景技术Background technique

高速数据接口,如:串行高级技术附件(SATA)、快捷外设互联标准(PCIE)、安全数字输入/输出卡(SDIO)、通用串行总线(USB)等,极容易因时钟信号延时,而发生数据抖动;明显影响高速数据传输。High-speed data interfaces, such as: Serial Advanced Technology Attachment (SATA), Peripheral Interconnect Express (PCIE), Secure Digital Input/Output Card (SDIO), Universal Serial Bus (USB), etc., are easily delayed by clock signals , and data jitter occurs; it obviously affects high-speed data transmission.

发明内容Contents of the invention

本发明提供一种低数据抖动的主机端控制器(host controller),也可以芯片组的南桥实现。The present invention provides a low data jitter host controller (host controller), which can also be realized by the south bridge of the chipset.

根据本发明一种实施方式实现的一种高速数据接口主机端控制器,包括逻辑物理层、电子物理层、以及跨时域数据传输模块。该逻辑物理层提供第一低速数据,再由该电子物理层转换为第一高速数据,并传递至第一外部装置。该电子物理层操作用的时钟信号还传递至该逻辑物理层,使该逻辑物理层据以提供所述第一低速数据。该跨时域数据传输模块耦接于所述逻辑物理层与所述电子物理层之间,根据逻辑物理层端时钟读入该逻辑物理层为该第一外部装置提供的所述第一低速数据,并根据电子物理层端时钟输出所述第一低速数据至该电子物理层。A high-speed data interface host controller implemented according to an embodiment of the present invention includes a logical physical layer, an electronic physical layer, and a cross-time domain data transmission module. The logical physical layer provides the first low-speed data, which is then converted into the first high-speed data by the electronic physical layer and transmitted to the first external device. The clock signal used for the operation of the electronic physical layer is also transmitted to the logical physical layer, so that the logical physical layer provides the first low-speed data accordingly. The cross-time domain data transmission module is coupled between the logical physical layer and the electronic physical layer, reads in the first low-speed data provided by the logical physical layer for the first external device according to the logical physical layer terminal clock , and output the first low-speed data to the electronic physical layer according to the electronic physical layer terminal clock.

本发明跨时域数据传输模块有效解决操作时钟在电子物理层端以及逻辑物理层端的异步问题。The cross-time domain data transmission module of the invention effectively solves the asynchronous problem of the operation clock at the electronic physical layer end and the logical physical layer end.

下文特举实施例,并配合所附图示,详细说明本发明内容。Hereinafter, specific embodiments are cited, and the contents of the present invention are described in detail in conjunction with the accompanying drawings.

附图说明Description of drawings

图1为方块图,描述根据本发明一种实施方式实现的高速数据接口主机端控制器100;FIG. 1 is a block diagram illustrating a high-speed data interface host controller 100 implemented according to an embodiment of the present invention;

图2A、图2B根据本发明一种实施方式图解跨时域数据传输模块TXCDC;2A and 2B illustrate the cross-time domain data transmission module TXCDC according to an embodiment of the present invention;

图3以波形图说明缓存R_A1操作,其中缓存R_A1编号0~7的缓存器分别命名为R_A1_0~R_A1_7;以及FIG. 3 illustrates the operation of the buffer R_A1 with a waveform diagram, wherein the buffers numbered 0-7 of the buffer R_A1 are respectively named R_A1_0-R_A1_7; and

图4为方块图,描述根据本发明一种实施方式实现的主机端控制器400,是以单一电子物理层EPHY连接至少一个外接装置。FIG. 4 is a block diagram illustrating a host-side controller 400 implemented according to an embodiment of the present invention, which connects at least one external device with a single electronic physical layer EPHY.

具体实施方式Detailed ways

以下叙述列举本发明的多种实施例。以下叙述介绍本发明的基本概念,且并非意图限制本发明内容。实际发明范围应依照权利要求书而界定。The following description lists various embodiments of the present invention. The following description introduces the basic concept of the present invention and is not intended to limit the content of the present invention. The actual scope of the invention should be defined in accordance with the claims.

图1为方块图,描述根据本发明一种实施方式实现的高速数据接口主机端控制器100。高速数据接口主机端控制器100包括逻辑物理层LPHY(编号同逻辑物理层(logicalphysical layer)缩写LPHY)、多个电子物理层(electrical physical layer,缩写EPHY)EPHYA以及EPHYB、多工器ECLKMUX、以及跨时域数据传输模块TXCDC。图1中仅示出两个电子物理层,但本发明并不以此为限。FIG. 1 is a block diagram illustrating a high-speed data interface host-side controller 100 implemented according to an embodiment of the present invention. The high-speed data interface host-side controller 100 includes a logical physical layer LPHY (the number is the same as the logical physical layer (logical physical layer) abbreviated LPHY), a plurality of electronic physical layers (electrical physical layer, abbreviated EPHY) EPHYA and EPHYB, a multiplexer ECLKMUX, and Cross-time domain data transmission module TXCDC. Only two electronic physical layers are shown in FIG. 1 , but the present invention is not limited thereto.

电子物理层EPHYA以及EPHYB连接外部装置;电子物理层EPHYA连接硬盘HDA1与HDA2,且电子物理层EPHYB连接硬盘HDB1与HDB2。电子物理层EPHYA与EPHYB分别根据时钟信号MPLLCLK_A与MPLLCLK_B操作;时钟信号MPLLCLK_A可由电子物理层EPHYA内部产生,且时钟信号MPLLCLK_B可由电子物理层EPHYB内部产生。值得注意的是,图1中的电子物理层EPHYA以及EPHYB均仅连接两个硬盘,但本发明在此并不限制每个电子物理层所连接的外部装置的类型和数量。The electronic physical layers EPHYA and EPHYB are connected to external devices; the electronic physical layer EPHYA is connected to the hard disks HDA1 and HDA2, and the electronic physical layer EPHYB is connected to the hard disks HDB1 and HDB2. The electronic physical layers EPHYA and EPHYB operate according to the clock signals MPLLCLK_A and MPLLCLK_B respectively; the clock signal MPLLCLK_A can be internally generated by the electronic physical layer EPHYA, and the clock signal MPLLCLK_B can be internally generated by the electronic physical layer EPHYB. It should be noted that the electronic physical layers EPHYA and EPHYB in FIG. 1 are only connected to two hard disks, but the present invention does not limit the type and quantity of external devices connected to each electronic physical layer.

多工器ECLKMUX接收电子物理层EPHYA与EPHYB所对应的时钟信号MPLLCLK_A与MPLLCLK_B,并输出共同时钟信号MPLLCLK_COM。共同时钟信号MPLLCLK_COM将引入该逻辑物理层LPHY以及该跨时域数据传输模块TXCDC。The multiplexer ECLKMUX receives the clock signals MPLLCLK_A and MPLLCLK_B corresponding to the electronic physical layers EPHYA and EPHYB, and outputs a common clock signal MPLLCLK_COM. The common clock signal MPLLCLK_COM will be introduced into the logical physical layer LPHY and the cross-time domain data transmission module TXCDC.

针对硬盘HDA1、HDA2、HDB1与HDB2,逻辑物理层LPHY内分别以电路模块PHYA1、PHYA2、PHYB1与PHYB2基于该共同时钟信号MPLLCLK_COM提供低速数据DA1_COM、DA2_COM、DB1_COM、DB2_COM传递至该跨时域数据传输模块TXCDC。For hard disks HDA1, HDA2, HDB1 and HDB2, circuit modules PHYA1, PHYA2, PHYB1 and PHYB2 in the logical physical layer LPHY provide low-speed data DA1_COM, DA2_COM, DB1_COM, DB2_COM based on the common clock signal MPLLCLK_COM and transmit them to the cross-time domain data transmission Module TXCDC.

跨时域数据传输模块TXCDC耦接于逻辑物理层LPHY和电子物理层EPHYA及EPHYB之间,其为基于跨时域技术(clock domain crossing)操作。跨时域数据传输模块TXCDC根据该共同时钟信号MPLLCLK_COM读入该逻辑物理层LPHY提供的上述低速数据DA1_COM、DA2_COM、DB1_COM、DB2_COM。在一实施例中,跨时域数据传输模块TXCDC为所述外部装置HDA1、HDA2、HDB1与HDB2各自提供一缓存(绘于图2A、图2B)以缓存对应不同外部装置的低速数据。跨时域数据传输模块TXCDC还根据对应的电子物理层EPHYA及EPHYB对应的时钟信号(电子物理层EPHYA对应时钟信号MPLLCLK_A、电子物理层EPHYB对应时钟信号MPLLCLK_B)分别将上述缓存的低速数据取出。参考图示,依据时钟信号MPLLCLK_A取出的低速数据DA1_A由电子物理层EPHYA转换为高速数据后发送至硬盘HDA1,依据时钟信号MPLLCLK_A取出的低速数据DA2_A由电子物理层EPHYA转换为高速数据后发送至硬盘HDA2,依据时钟信号MPLLCLK_B取出的数据DB1_B由电子物理层EPHYB转换为高速数据后发送至硬盘HDB1,依据时钟信号MPLLCLK_B取出的数据DB2_B由电子物理层EPHYB转换为高速数据后发送至硬盘HDB2。特别是,各缓存具有多层缓存深度,使得低速数据读入缓存以及数据读出缓存得以跨时域实现。The cross-time domain data transmission module TXCDC is coupled between the logical physical layer LPHY and the electronic physical layers EPHYA and EPHYB, and operates based on clock domain crossing. The cross-time domain data transmission module TXCDC reads in the above-mentioned low-speed data DA1_COM, DA2_COM, DB1_COM, DB2_COM provided by the logical physical layer LPHY according to the common clock signal MPLLCLK_COM. In one embodiment, the cross-time domain data transmission module TXCDC provides each of the external devices HDA1 , HDA2 , HDB1 and HDB2 with a buffer (shown in FIG. 2A , FIG. 2B ) to buffer low-speed data corresponding to different external devices. The cross-time domain data transmission module TXCDC also fetches the low-speed data from the above buffer according to the clock signals corresponding to the corresponding electronic physical layer EPHYA and EPHYB (the electronic physical layer EPHYA corresponds to the clock signal MPLLCLK_A, and the electronic physical layer EPHYB corresponds to the clock signal MPLLCLK_B). Referring to the figure, the low-speed data DA1_A extracted according to the clock signal MPLLCLK_A is converted into high-speed data by the electronic physical layer EPHYA and then sent to the hard disk HDA1, and the low-speed data DA2_A extracted according to the clock signal MPLLCLK_A is converted into high-speed data by the electronic physical layer EPHYA and sent to the hard disk HDA2, the data DB1_B taken out according to the clock signal MPLLCLK_B is converted into high-speed data by the electronic physical layer EPHYB and sent to the hard disk HDB1, and the data DB2_B taken out according to the clock signal MPLLCLK_B is converted into high-speed data by the electronic physical layer EPHYB and sent to the hard disk HDB2. In particular, each cache has a multi-layer cache depth, so that low-speed data read-in cache and data read-out cache can be implemented across time domains.

如图1所示,单纯根据共同时钟信号MPLLCLK_COM操作的逻辑物理层LPHY将降低设计门槛。传统技术的逻辑物理层LPHY的对应不同的电子物理层(例如EPHYA与EPHYB)的不同电路模块(例如PHYA1、PHYA2与PHYB1、PHYA2)根据不同电子物理层的时钟信号(例如MPLLCLK_A与MPLLCLK_B)操作,由于各电子物理层的时钟信号(例如MPLLCLK_A与MPLLCLK_B)为异步时钟信号,将导致时钟树(Clock Tree)复杂,而本发明单纯根据共同时钟信号MPLLCLK_COM操作的逻辑物理层LPHY将大大简化时钟树。此外,设置于逻辑物理层LPHY与电子物理层EPHYA及EPHYB之间的跨时域数据传输模块TXCDC将有效抑制走线延时问题。相较于传统技术将逻辑物理层直接耦接电子物理层的长走线,跨时域数据传输模块TXCDC将数据走线截半,及时修正走线延时。As shown in Figure 1, the logical physical layer LPHY operated solely according to the common clock signal MPLLCLK_COM will lower the design threshold. Different circuit modules (such as PHYA1, PHYA2 and PHYB1, PHYA2) corresponding to different electronic physical layers (such as EPHYA and EPHYB) of the logical physical layer LPHY of the conventional technology operate according to clock signals (such as MPLLCLK_A and MPLLCLK_B) of different electronic physical layers, Since the clock signals of each electronic physical layer (such as MPLLCLK_A and MPLLCLK_B) are asynchronous clock signals, the clock tree (Clock Tree) will be complicated, and the logical physical layer LPHY operated solely according to the common clock signal MPLLCLK_COM in the present invention will greatly simplify the clock tree. In addition, the cross-time domain data transmission module TXCDC arranged between the logical physical layer LPHY and the electronic physical layers EPHYA and EPHYB will effectively suppress the wiring delay problem. Compared with the traditional technology that directly couples the logical physical layer to the long trace of the electronic physical layer, the cross-time domain data transmission module TXCDC cuts the data trace in half and corrects the trace delay in time.

在一种实施方式中,时钟信号MPLLCLK_A和MPLLCLK_B走线距离将用来判断电子物理层EPHYA以及EPHYB哪个最靠近该逻辑物理层LPHY。图1是电子物理层EPHYA为最近电子物理层。多工器ECLKMUX以最近电子物理层EPHYA的时钟信号MPLLCLK_A作为该共同时钟信号MPLLCLK_COM,使较少走线延时的时钟信号MPLLCLK_A为逻辑物理层LPHY所用。值得注意的是,在一实施例中,这里的走线距离是指在专用集成电路(Application SpecificIntegrated Circuits,ASIC)中时钟信号MPLLCLK_A和MPLLCLK_B从电子物理层EPHYA以及EPHYB至多工器ECLKMUX的走线距离。在前述实施方式中,根据时钟信号的走线距离选择共同时钟信号MPLLCLK_COM是基于时钟信号MPLLCLK_A和MPLLCLK_B为频率相同的异步时钟的前提下,如果时钟信号MPLLCLK_A和MPLLCLK_B本身频率即不同,则会用其它方式选择共同时钟信号MPLLCLK_COM,详见后述。In one embodiment, the routing distance of the clock signals MPLLCLK_A and MPLLCLK_B is used to determine which of the electronic physical layers EPHYA and EPHYB is closest to the logical physical layer LPHY. Figure 1 is the electron physical layer EPHYA is the closest electron physical layer. The multiplexer ECLKMUX uses the clock signal MPLLCLK_A of the nearest electronic physical layer EPHYA as the common clock signal MPLLCLK_COM, so that the clock signal MPLLCLK_A with less wiring delay is used by the logical physical layer LPHY. It is worth noting that, in one embodiment, the wiring distance here refers to the wiring distance of the clock signals MPLLCLK_A and MPLLCLK_B from the electronic physical layer EPHYA and EPHYB to the multiplexer ECLKMUX in an ASIC (Application Specific Integrated Circuits, ASIC) . In the aforementioned embodiments, the selection of the common clock signal MPLLCLK_COM according to the routing distance of the clock signal is based on the premise that the clock signals MPLLCLK_A and MPLLCLK_B are asynchronous clocks with the same frequency. If the frequencies of the clock signals MPLLCLK_A and MPLLCLK_B are different, other The mode selects the common clock signal MPLLCLK_COM, which will be described later.

一种实施方式中,跨时域数据传输模块TXCDC至逻辑物理层LPHY的走线(传送DA1_COM、DA2_COM、DB1_COM、DB2_COM)距离设计为短于最近电子物理层EPHYA至逻辑物理层LPHY的距离,甚至该跨时域数据传输模块TXCDC至最近电子物理层EPHYA的走线距离(传送DA1_A、DA2_A、DB1_B、DB2_B)也设计为短于该最近电子物理层EPHYA至逻辑物理层LPHY的距离。如此设计要诀将使得跨时域数据传输模块TXCDC对数据走线延时的修正更为准确。In one embodiment, the distance from the cross-time domain data transmission module TXCDC to the logical physical layer LPHY (transmitting DA1_COM, DA2_COM, DB1_COM, DB2_COM) is designed to be shorter than the distance from the nearest electronic physical layer EPHYA to the logical physical layer LPHY, or even The routing distance (transmitting DA1_A, DA2_A, DB1_B, DB2_B) from the cross-time domain data transmission module TXCDC to the closest electronic physical layer EPHYA is also designed to be shorter than the distance from the closest electronic physical layer EPHYA to the logical physical layer LPHY. Such design tips will make the correction of the data routing delay by the cross-time domain data transmission module TXCDC more accurate.

一种实施方式中,逻辑物理层LPHY以并行方式提供低速数据DA1_COM、DA2_COM、DB1_COM、DB2_COM至跨时域数据传输模块TXCDC,跨时域数据传输模块TXCDC以并行方式发送低速数据DA1_A、DA2_A、DB1_B、DB2_B至电子物理层EPHYA以及EPHYB,且电子物理层EPHYA以及EPHYB包括将数据DA1_A、DA2_A、DB1_B、DB2_B自并行低速数据转换为串行高速数据(例如差分信号)后才传输至硬盘HDA1、HDA2、HDB1与HDB2。如此设计使得低速的逻辑物理层LPHY与高速的电子物理层EPHYA以及EPHYB结合,利于实现高速数据接口,如:串行高级技术附件(SATA)、快捷外设互联标准(PCIE)、安全数字输入/输出卡(SDIO)、通用串行总线(USB)等。In one embodiment, the logical physical layer LPHY provides low-speed data DA1_COM, DA2_COM, DB1_COM, DB2_COM to the cross-time domain data transmission module TXCDC in parallel, and the cross-time domain data transmission module TXCDC sends low-speed data DA1_A, DA2_A, DB1_B in parallel , DB2_B to the electronic physical layer EPHYA and EPHYB, and the electronic physical layer EPHYA and EPHYB include converting data DA1_A, DA2_A, DB1_B, DB2_B from parallel low-speed data to serial high-speed data (such as differential signals) before transferring to hard disk HDA1, HDA2 , HDB1 and HDB2. This design makes the low-speed logical physical layer LPHY combined with the high-speed electronic physical layer EPHYA and EPHYB, which is conducive to the realization of high-speed data interfaces, such as: Serial Advanced Technology Attachment (SATA), Express Peripheral Interconnect Standard (PCIE), secure digital input/ Output card (SDIO), Universal Serial Bus (USB), etc.

一种实施方式中,时钟信号MPLLCLK_A与MPLLCLK_B以及共同时钟信号MPLLCLK_COM的频率相同,均为300MHz。逻辑物理层LPHY以及跨时域数据传输模块TXCDC为20位并行传输,则电子物理层EPHYA可实现6Gbps的高速串行传输。In one embodiment, the clock signals MPLLCLK_A and MPLLCLK_B and the common clock signal MPLLCLK_COM have the same frequency, which is 300 MHz. The logical physical layer LPHY and the cross-time domain data transmission module TXCDC are 20-bit parallel transmission, and the electronic physical layer EPHYA can realize 6Gbps high-speed serial transmission.

图1实施方式并不意图限定电子物理层的数量、电子物理层连接的外接装置数量、以及电子物理层与逻辑物理层的相对布局。甚至,多个电子物理层的时钟信号允许是不同频率。假设电子物理层EPHYA的时钟信号MPLLCLK_A频率为300MHz,电子物理层EPHYB的时钟信号MPLLCLK_B频率为150MHz。在这里实施方式中,多工器ECLKMUX将不考虑时钟信号MPLLCLK_A和MPLLCLK_B的走线距离,而是以最高频的时钟信号MPLLCLK_A作为该共同时钟信号MPLLCLK_COM。逻辑物理层LPHY包括分频器分频300Mz的该共同时钟信号MPLLCLK_COM以得到多个分频共同时钟信号(未绘示),例如还获得150Mz的时钟信号。逻辑物理层LPHY中的电路模块PHYA1以及PHYA2依照300MHz的该共同时钟信号MPLLCLK_COM操作(例如提供低速数据DA1_COM和DA2_COM),而电路模块PHYB1以及PHYB2依照分频获得的150MHz的时钟信号操作(例如提供低速数据DB1_COM和DB2_COM)。其中逻辑物理层LPHY中的各电路模块依据哪个分频共同时钟信号操作视各电路模块对应的外部装置(HDA1和HDA2、HDB1和HDB2)所连接的电子物理层(EPHYA和EPHYB)对应的时钟信号(MPLLCLK_A和MPLLCLK_B)的时钟频率而定,即各电路模块的分频共同时钟信号的时钟频率与各电路模块对应的外部装置所连接的电子物理层对应的时钟信号的时钟频率相同。在另一实施例中,跨时域数据传输模块TXCDC还包括分频器分频300MHz的该共同时钟信号MPLLCLK_COM以得到多个分频共同时钟信号(未绘示),例如还获得150MHz的时钟信号;跨时域数据传输模块TXCDC根据对应硬盘HDA1与HDA2所连接的电子物理层EPHYA的时钟信号MPLLCLK_A的时钟频率300MHz的分频共同时钟信号(频率为300MHz)将数据DA1_COM、DA2_COM读入,并根据对应硬盘HDB1与HDB2所连接的电子物理层EPHYB的时钟信号MPLLCLK_B的时钟频率150MHz的分频共同时钟信号(频率为150MHz)将数据DB1_COM、DB2_COM读入对应的缓存。值得注意的是,跨时域数据传输模块TXCDC将数据DA1_A、DA2_A读出是根据对应硬盘HDA1与HDA2所连接的电子物理层EPHYA的时钟信号MPLLCLK_A,且将数据DB1_B、DB2_B读出是根据对应硬盘HDB1与HDB2所连接的电子物理层EPHYB的时钟信号MPLLCLK_B。The embodiment in FIG. 1 is not intended to limit the number of electronic physical layers, the number of external devices connected to the electronic physical layers, and the relative layout of the electronic physical layers and logical physical layers. Furthermore, the clock signals of multiple electronic physical layers are allowed to be of different frequencies. It is assumed that the frequency of the clock signal MPLLCLK_A of the electronic physical layer EPHYA is 300 MHz, and the frequency of the clock signal MPLLCLK_B of the electronic physical layer EPHYB is 150 MHz. In this embodiment, the multiplexer ECLKMUX will not consider the wiring distance of the clock signals MPLLCLK_A and MPLLCLK_B, but uses the clock signal MPLLCLK_A with the highest frequency as the common clock signal MPLLCLK_COM. The logical physical layer LPHY includes a frequency divider to divide the 300Mz common clock signal MPLLCLK_COM to obtain multiple frequency-divided common clock signals (not shown), for example, to obtain a 150Mz clock signal. The circuit modules PHYA1 and PHYA2 in the logical physical layer LPHY operate according to the common clock signal MPLLCLK_COM of 300MHz (such as providing low-speed data DA1_COM and DA2_COM), while the circuit modules PHYB1 and PHYB2 operate according to the 150MHz clock signal obtained by frequency division (such as providing low-speed data DB1_COM and DB2_COM). Among them, each circuit module in the logical physical layer LPHY operates the clock signal corresponding to the electronic physical layer (EPHYA and EPHYB) connected to the external device (HDA1 and HDA2, HDB1 and HDB2) corresponding to each circuit module according to which frequency-division common clock signal (MPLLCLK_A and MPLLCLK_B) depends on the clock frequency, that is, the clock frequency of the frequency-divided common clock signal of each circuit module is the same as the clock frequency of the clock signal corresponding to the electronic physical layer connected to the external device corresponding to each circuit module. In another embodiment, the cross-time-domain data transmission module TXCDC further includes a frequency divider to divide the common clock signal MPLLCLK_COM of 300MHz to obtain multiple frequency-divided common clock signals (not shown), for example, to obtain a 150MHz clock signal The data DA1_COM, DA2_COM are read in by the frequency division common clock signal (frequency is 300MHz) of the clock frequency 300MHz of the clock signal MPLLCLK_A of the clock signal MPLLCLK_A of the electronic physical layer EPHYA that the corresponding hard disk HDA1 and HDA2 are connected, and the data DA1_COM, DA2_COM are read in by the cross-time domain data transmission module TXCDC The data DB1_COM and DB2_COM are read into the corresponding cache by the frequency-divided common clock signal (frequency 150MHz) corresponding to the clock signal MPLLCLK_B of the electronic physical layer EPHYB connected to the hard disks HDB1 and HDB2 with a clock frequency of 150MHz. It is worth noting that the cross-time domain data transmission module TXCDC reads the data DA1_A and DA2_A according to the clock signal MPLLCLK_A of the electronic physical layer EPHYA connected to the corresponding hard disk HDA1 and HDA2, and reads the data DB1_B and DB2_B according to the corresponding hard disk The clock signal MPLLCLK_B of the electronic physical layer EPHYB connected to HDB1 and HDB2.

图2A、图2B根据本发明一种实施方式图解跨时域数据传输模块TXCDC。跨时域数据传输模块TXCDC对应硬盘HDA1、HDA2、HDB1与HDB2分别提供缓存R_A1、R_A2、R_B1与R_B2,各自例如包括8个编号0~7的缓存器(各缓存器尺寸同并行低速数据尺寸),即缓存深度包括0~7。逻辑物理层LPHY提供的低速数据DA1_COM、DA2_COM、DB1_COM、DB2_COM根据共同时钟信号MPLLCLK_COM先读入缓冲器W_Buf,然后写入指标产生器WPTR++根据共同时脉信号MPLLCLK_COM操作写入分配器W_DMUX,以将对应的低速数据DA1_COM、DA2_COM、DB1_COM、DB2_COM再推入缓存R_A1、R_A2、R_B1与R_B2中的缓存器。R_A1、R_A2、R_B1与R_B2中缓存器里面的内容还可在读取指标产生器RPTR++根据对应的时钟信号MPLLCLK_A和MPLLCLK_B操作读取选择器R_MUX,以将对应的低速数据读取到缓冲器R_Buf,再经由缓冲器R_Buf根据对应的时钟信号MPLLCLK_A/MPLLCLK_B输出为数据DA1_A、DA2_A、DB1_B、DB2_B。FIG. 2A and FIG. 2B illustrate a cross-time domain data transmission module TXCDC according to an embodiment of the present invention. The cross-time domain data transmission module TXCDC provides buffers R_A1, R_A2, R_B1 and R_B2 corresponding to the hard disks HDA1, HDA2, HDB1 and HDB2 respectively, and each includes, for example, 8 buffers numbered 0 to 7 (the size of each buffer is the same as the parallel low-speed data size) , that is, the cache depth includes 0 to 7. The low-speed data DA1_COM, DA2_COM, DB1_COM, and DB2_COM provided by the logical physical layer LPHY are first read into the buffer W_Buf according to the common clock signal MPLLCLK_COM, and then written into the index generator WPTR++ and written into the distributor W_DMUX according to the common clock signal MPLLCLK_COM to transfer the corresponding The low-speed data DA1_COM, DA2_COM, DB1_COM, DB2_COM are then pushed into the registers in the buffers R_A1, R_A2, R_B1 and R_B2. The contents of the buffers in R_A1, R_A2, R_B1 and R_B2 can also be read in the index generator RPTR++ according to the corresponding clock signals MPLLCLK_A and MPLLCLK_B to operate the read selector R_MUX to read the corresponding low-speed data to the buffer R_Buf, Then output as data DA1_A, DA2_A, DB1_B, DB2_B through the buffer R_Buf according to the corresponding clock signal MPLLCLK_A/MPLLCLK_B.

图3以波形图说明缓存R_A1操作,其中缓存R_A1编号0~7的缓存器分别命名为R_A1_0~R_A1_7。依照共同时钟信号MPLLCLK_COM推入缓存R_A1不同缓存深度的缓存器R_A1_0~R_A1_7的并行数据D0~D7将顺利经由时钟信号MPLLCLK_A取出,反映于数据DA1_A上。依照共同时钟信号MPLLCLK_COM推入缓存R_A1的数据D8~D15将逐个更新缓存器R_A1_0~R_A1_7。不同缓存深度的缓存器R_A1_0~R_A1_7的数据D8~D15也将顺利经由时钟信号MPLLCLK_A取出,反映于数据DA1_A上。FIG. 3 uses a waveform diagram to illustrate the operation of the buffer R_A1, wherein the registers numbered 0-7 of the buffer R_A1 are respectively named R_A1_0-R_A1_7. The parallel data D0-D7 pushed into the buffer R_A1 of the registers R_A1_0-R_A1_7 with different buffer depths according to the common clock signal MPLLCLK_COM will be successfully fetched through the clock signal MPLLCLK_A and reflected on the data DA1_A. The data D8-D15 pushed into the buffer R_A1 according to the common clock signal MPLLCLK_COM will update the registers R_A1_0-R_A1_7 one by one. The data D8-D15 of the registers R_A1_0-R_A1_7 with different buffer depths will also be fetched smoothly through the clock signal MPLLCLK_A, and reflected on the data DA1_A.

图4为方块图,描述根据本发明一种实施方式实现的高速数据接口主机端控制器400,为以单一电子物理层EPHY连接至少一个外接装置,图例包括硬盘HD1以及HD2,对应的,跨时域数据传输模块TXCDC包括两组缓存设计。逻辑物理层LPHY与跨时域数据传输模块TXCDC可以并行方式传输数据。电子物理层EPHY可包括并行至串行转换。FIG. 4 is a block diagram, describing a high-speed data interface host-side controller 400 implemented according to an embodiment of the present invention, for connecting at least one external device with a single electronic physical layer EPHY. The illustration includes hard disks HD1 and HD2, correspondingly, across time The domain data transmission module TXCDC includes two sets of cache designs. The logical physical layer LPHY and the cross-time domain data transmission module TXCDC can transmit data in parallel. The electronic physical layer EPHY may include parallel-to-serial conversion.

相较于图1的高速数据接口主机端控制器100还设计有多工器ECLKMUX,图4主机端控制器400上单一电子物理层EPHY的时钟信号可单纯以时钟信号走线CLK_trace传输至逻辑物理层LPHY,供电路模块PHY_1以及PHY_2操作参考以分别输出第一和第二低速数据。图4中,跨时域数据传输模块TXCDC自该时钟信号走线CLK_trace上的逻辑物理层端节点MPLLCLK_L接收逻辑物理层端时钟(以下同样称之MPLLCLK_L),并自该时钟信号走线CLK_trace的电子物理层端节点MPLLCLK_E接收电子物理层端时钟(以下同样称之MPLLCLK_E)。逻辑物理层端节点MPLLCLK_L为该时钟信号走线CLK_trace上与该逻辑物理层LPHY同侧的输入至跨时域数据传输模块TXCDC的节点,电子物理层端节点MPLLCLK_E为该时钟信号走线CLK_trace上与该电子物理层EPHY同侧的输入至跨时域数据传输模块TXCDC的节点。逻辑物理层端节点MPLLCLK_L较该电子物理层端节点MPLLCLK_E在该时钟信号走线CLK_trace上靠近该逻辑物理层LPHY。图4的跨时域数据传输模块TXCDC仍旧可有效解决时钟信号走线延时问题。在一种实施方式中,跨时域数据传输模块TXCDC至该逻辑物理层LPHY的走线距离设计为短于该时钟信号走线CLK_trace,且该跨时域数据传输模块TXCDC至该电子物理层EPHY的走线距离短于该时钟信号走线CLK_trace。相较于传统技术将逻辑物理层直接耦接电子物理层的长走线,跨时域数据传输模块TXCDC将数据走线截半,及时修正走线延时。Compared with the high-speed data interface host-side controller 100 in FIG. 1 is also designed with a multiplexer ECLKMUX, the clock signal of the single electronic physical layer EPHY on the host-side controller 400 in FIG. The layer LPHY is referenced by the circuit modules PHY_1 and PHY_2 to output the first and second low-speed data respectively. In Fig. 4, the cross-time domain data transmission module TXCDC receives the logical physical layer end clock (hereinafter also referred to as MPLLCLK_L) from the logical physical layer end node MPLLCLK_L on the clock signal line CLK_trace, and transmits the clock signal from the electronic circuit of the clock signal line CLK_trace. The physical layer end node MPLLCLK_E receives the electronic physical layer end clock (hereinafter also referred to as MPLLCLK_E). The logical physical layer end node MPLLCLK_L is the node that is input to the cross-time domain data transmission module TXCDC on the same side as the logical physical layer LPHY on the clock signal line CLK_trace, and the electronic physical layer end node MPLLCLK_E is the node connected to the clock signal line CLK_trace. The same side of the electronic physical layer EPHY is input to the node of the cross-time domain data transmission module TXCDC. The logical physical layer end node MPLLCLK_L is closer to the logical physical layer LPHY on the clock signal trace CLK_trace than the electronic physical layer end node MPLLCLK_E. The cross-time domain data transmission module TXCDC in Figure 4 can still effectively solve the problem of clock signal routing delay. In one embodiment, the wiring distance from the cross-time domain data transmission module TXCDC to the logical physical layer LPHY is designed to be shorter than the clock signal wiring CLK_trace, and the cross-time domain data transmission module TXCDC to the electronic physical layer EPHY The trace distance is shorter than the clock signal trace CLK_trace. Compared with the traditional technology that directly couples the logical physical layer to the long trace of the electronic physical layer, the cross-time domain data transmission module TXCDC cuts the data trace in half and corrects the trace delay in time.

虽然本发明已以较佳实施例揭露如上,然其并非用以限定本发明,任何熟悉此项技艺者,在不脱离本发明的精神和范围内,当可做些许更动与润饰,因此本发明的保护范围当视权利要求书所界定的为准。Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Anyone familiar with this art can make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, this The scope of protection of the invention should be defined by the claims.

Claims (9)

  1. A kind of 1. high speed interface host-side controller, it is characterised in that including:
    Logical network layer and electronic physical layer, wherein, the logical network layer provides the first low speed data, then by the electronics physics Layer is converted to the first high-speed data, and is transferred to the first external device (ED), and the clock signal of the electronic physical layer operation also passes The logical network layer is handed to, the logical network layer is provided first low speed data according to this;And
    Across time domain data transport module, it is coupled between the logical network layer and the electronic physical layer, according to logical physical It is first low speed data that first external device (ED) provides that layer end clock, which reads in the logical network layer, and according to electronics physics Layer end clock exports first low speed data to the electronic physical layer.
  2. 2. high speed interface host-side controller according to claim 1, it is characterised in that:
    It is somebody's turn to do across time domain data transport module and also provides the first caching for first external device (ED), caches according to the logical physical First low speed data that layer end clock is read in.
  3. 3. high speed interface host-side controller according to claim 1, it is characterised in that also include:
    Clock signal cabling, the clock signal is transferred to the logical network layer from the electronic physical layer,
    Wherein, across time domain data transport module is somebody's turn to do from the logic thing of the logical network layer end node reception on the clock signal cabling Manage layer end clock, and from the electronic physical layer end node reception of the clock signal cabling electronic physical layer end clock.
  4. 4. high speed interface host-side controller according to claim 3, it is characterised in that the logical network layer end segment Point close logical network layer on the clock signal cabling compared with the electronic physical layer end node.
  5. 5. high speed interface host-side controller according to claim 3, it is characterised in that:
    The cable run distance of across time domain data transport module to the logical network layer is shorter than the clock signal cabling;And
    Across the time domain data transport module to the cable run distance of the electronic physical layer is shorter than the clock signal cabling.
  6. 6. high speed interface host-side controller according to claim 1, it is characterised in that:
    The logical network layer provides first low speed data in a parallel fashion extremely should across time domain data transport module;
    It is somebody's turn to do across time domain data transport module and sends first low speed data in a parallel fashion to the electronic physical layer;And
    The electronic physical layer just transmitted after first low speed data is converted into serial high speed from parallel low speed data to First external device (ED).
  7. 7. high speed interface host-side controller according to claim 1, it is characterised in that:
    The logical network layer also provides the second low speed data and is converted to the second high-speed data by the electronic physical layer, and is transferred to Two external device (ED)s;
    Should across time domain data transport module always according to the logical network layer end clock read in the logical network layer for this outside second Second low speed data that device provides, and second low speed data is exported to the electricity according to the electronic physical layer end clock Muon physics layer.
  8. 8. high speed interface host-side controller according to claim 7, it is characterised in that:
    It is somebody's turn to do across time domain data transport module and also provides the second caching for the electronic physical layer, caches according to the logical network layer Second low speed data for holding clock to read in.
  9. 9. high speed interface host-side controller according to claim 1, it is characterised in that:
    The clock signal, the logical network layer end clock and the electronic physical layer end clock are same clock source.
CN201510894510.2A 2015-12-07 2015-12-07 High speed interface host-side controller Active CN105512071B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510894510.2A CN105512071B (en) 2015-12-07 2015-12-07 High speed interface host-side controller

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510894510.2A CN105512071B (en) 2015-12-07 2015-12-07 High speed interface host-side controller

Publications (2)

Publication Number Publication Date
CN105512071A CN105512071A (en) 2016-04-20
CN105512071B true CN105512071B (en) 2018-04-03

Family

ID=55720068

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510894510.2A Active CN105512071B (en) 2015-12-07 2015-12-07 High speed interface host-side controller

Country Status (1)

Country Link
CN (1) CN105512071B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101561794A (en) * 2009-06-05 2009-10-21 威盛电子股份有限公司 Universal serial bus device
CN101689209A (en) * 2007-05-15 2010-03-31 克罗诺洛吉克有限公司 Method and system for reducing triggering latency in universal serial bus data acquisition
CN102023945A (en) * 2009-09-22 2011-04-20 鸿富锦精密工业(深圳)有限公司 Serial peripheral interface bus-based equipment and data transmission method thereof

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2790110A1 (en) * 2009-05-20 2014-10-15 Chronologic Pty Limited Precision synchronisation architecture for superspeed Universal Serial BUS devices

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101689209A (en) * 2007-05-15 2010-03-31 克罗诺洛吉克有限公司 Method and system for reducing triggering latency in universal serial bus data acquisition
CN101561794A (en) * 2009-06-05 2009-10-21 威盛电子股份有限公司 Universal serial bus device
CN102023945A (en) * 2009-09-22 2011-04-20 鸿富锦精密工业(深圳)有限公司 Serial peripheral interface bus-based equipment and data transmission method thereof

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
SATA控制器的设计与FPGA验证;杨佳朋;《中国优秀硕士学位论文全文数据库信息科技辑》;20140315;全文 *

Also Published As

Publication number Publication date
CN105512071A (en) 2016-04-20

Similar Documents

Publication Publication Date Title
JP3820843B2 (en) Directional coupled memory module
US8352774B2 (en) Inter-clock domain data transfer FIFO circuit
US9959236B2 (en) Observing an internal link via an existing port for system on chip devices
CN105320490B (en) Method and apparatus for asynchronous FIFO circuit
CN104915303B (en) High speed digital I based on PXIe buses/O systems
WO2007106834A1 (en) A common analog interface for multiple processor cores
JP2015156645A (en) System on chip, bus interface circuit, and bus interface method
CN104022828B (en) A kind of optical fiber data transmission method based on asynchronous communication model
US8977882B2 (en) System for data transfer between asynchronous clock domains
CN104022775A (en) FIFO protocol based digital interface circuit for SerDes technology
CN105550134B (en) High speed interface host-side controller
CN105680871A (en) Parallel-serial/ serial-parallel conversion device for data serial transmission
CN115407692B (en) An 8-channel high-speed synchronous broadband signal acquisition and processing module
US8510485B2 (en) Low power digital interface
CN105512071B (en) High speed interface host-side controller
CN113806277B (en) Data Transmission System Between FPGA and DSP Based on SRIO Protocol
US20140297906A1 (en) Buffer circuit and semiconductor integrated circuit
KR20250096746A (en) Data lane deskew and speed adaptation in packages containing multiple circuit dies
CN110008157A (en) A kind of hardware structure of deserializer
CN117009276A (en) Method, device and system for signal compression transmission based on AXI bus
CN113220609B (en) Signal receiving and transmitting system and method
CN219958216U (en) Chip prototype verification debugging board
CN120123284B (en) A circuit architecture and implementation method for improving SPI communication baud rate
CN203689519U (en) High-speed collecting and storing system based on radar signals
CN219179825U (en) Time deskewing circuit, system and electronic equipment

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CP01 Change in the name or title of a patent holder
CP01 Change in the name or title of a patent holder

Address after: Room 301, 2537 Jinke Road, Zhangjiang High Tech Park, Pudong New Area, Shanghai 201203

Patentee after: Shanghai Zhaoxin Semiconductor Co.,Ltd.

Address before: Room 301, 2537 Jinke Road, Zhangjiang High Tech Park, Pudong New Area, Shanghai 201203

Patentee before: VIA ALLIANCE SEMICONDUCTOR Co.,Ltd.