CN105681222A - Method and apparatus for data receiving and caching, and communication system - Google Patents
Method and apparatus for data receiving and caching, and communication system Download PDFInfo
- Publication number
- CN105681222A CN105681222A CN201610120423.6A CN201610120423A CN105681222A CN 105681222 A CN105681222 A CN 105681222A CN 201610120423 A CN201610120423 A CN 201610120423A CN 105681222 A CN105681222 A CN 105681222A
- Authority
- CN
- China
- Prior art keywords
- data
- read request
- data read
- spatial cache
- identification information
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/90—Buffering arrangements
- H04L49/9005—Buffering arrangements using dynamic buffer space allocation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a method and apparatus for data receiving and caching, and a communication system. The method comprises: after sending of a data reading request is detected, a caching space is distributed to the data reading request; and when return data from the data reading request are received, the return data are cached to a caching space of the data reading request. With implementation of the method, apparatus, and system, because the corresponding storage space is distributed in advance in the receiving and cashing process before the reading request is initiated, the caching management complexity is reduced. Meanwhile, the difficulty of credit updating is also reduced because of continuous storage; and a problem of high realization difficulty of a hardware circuit for processing a data packet returned in an out-of-order mode by using a chain table structure in the prior art can be solved.
Description
Technical field
The present invention relates to the communications field, particularly relate to a kind of data receiver caching method, device and communication system.
Background technology
In communication technology, after local terminal sends data read request to opposite end, obtain the return data that opposite end sends for data read request, go forward side by side row cache for follow-up, it is conventional communication process, but is as technical development, return the data length of data also growing, so, it is common to use the transmission of multiple packets returns data. If local terminal sends multiple data read request to opposite end simultaneously, these data read request all adopt multiple packet transmission return data mode, due to these data read request packet can out of order return, correspondence, when local terminal carries out buffer memory, these packets arise that chaotic buffer memory.
In order to process the packet of out of order return, practices well is the relation adopting list structure (LinkedList) to record between storage address and the packet of packet. But, owing to list structure itself is more complicated, adopt hardware circuit to realize difficulty bigger; For receiving terminal, after data export from reception buffer memory, it is necessary to the credit that upgrades in time is to opposite end, and the dispersion storage of list structure correspondence, this updates to credit and brings no small difficulty.
Therefore, those skilled in the art urgently provide a kind of data receiver caching method, and the hardware circuit to solve the packet existence that existing employing list structure processes out of order return realizes the technical problem that difficulty is big.
Summary of the invention
The invention provides a kind of data receiver caching method, device and communication system, the hardware circuit to solve the packet existence that existing employing list structure processes out of order return realizes the problem that difficulty is big.
The invention provides a kind of data receiver caching method, comprising:
After transmission data read request being detected, distribute spatial cache for data read request;
When receiving the return data of data read request, the data buffer storage spatial cache to data read request will be returned.
Further, include returning data buffer storage to the spatial cache of data read request: when returning data and including two and data above bag, by two and data above bag according to sending time-sequencing, and buffer memory.
Further, by return data buffer storage to data read request spatial cache by two and data above bag according to, after sending time-sequencing, also including: first packet after sequence is carried packet header buffer memory, and other packets remove buffer memory behind packet header.
Further, after by the spatial cache of return data buffer storage to data read request, also include: after return data data read request being detected all correctly return, the data of buffer memory are read from the spatial cache of data read request, carry out credit renewal the spatial cache of initialization data read request.
Further, distribute spatial cache for data read request to include: obtain the identification information of data read request, identification information according to data read request is data read request distribution spatial cache, the one_to_one corresponding of the identification information of spatial cache and the identification information of data read request; Include returning the data buffer storage spatial cache to data read request: determine the identification information of data read request according to the identification information returning data, identification information according to spatial cache and the corresponding relation of the identification information of data read request, will return the data buffer storage spatial cache to correspondence.
Further, when data read request is read request, distribute spatial cache for data read request also to include: obtain reading data length information corresponding to read request and identification information, the reading data length information corresponding according to read request and identification information, for the spatial cache of the corresponding length of read request distribution.
The invention provides a kind of data receiver buffer storage, comprising:
Distribution module, after being used for detecting transmission data read request, distributes spatial cache for data read request;
Cache module, for when receiving the return data of data read request, returning the data buffer storage spatial cache to data read request.
Further, cache module is for when returning data and including two and data above bag, by two and data above bag according to sending time-sequencing, and buffer memory.
Further, cache module by two and data above bag according to after sending time-sequencing, first packet after being additionally operable to sequence carries packet header buffer memory, and other packets remove buffer memory behind packet header.
Further, cache module by return data buffer storage to data read request spatial cache after, it is additionally operable to after return data data read request being detected all correctly return, the data of buffer memory are read from the spatial cache of data read request, carry out credit renewal the spatial cache of initialization data read request.
Further, distribution module, for obtaining the identification information of data read request, is data read request distribution spatial cache according to the identification information of data read request, the one_to_one corresponding of the identification information of spatial cache and the identification information of data read request; Cache module for determining the identification information of data read request according to the identification information returning data, and the corresponding relation according to the identification information of spatial cache Yu the identification information of data read request, by spatial cache extremely corresponding for return data buffer storage.
Further, when data read request is read request, distribution module is additionally operable to obtain reading data length information corresponding to read request and identification information, the reading data length information corresponding according to read request and identification information, for the spatial cache of the corresponding length of read request distribution.
The invention provides a kind of communication system, it includes the first equipment and the second equipment, the first equipment and the second equipment by high performance interconnection serial bus communication, and the first equipment and/or the second equipment include data receiver buffer storage provided by the invention.
Beneficial effects of the present invention:
The invention provides a kind of new data receiver caching method, after transmission data read request being detected, distribute spatial cache for data read request;When receiving the return data of data read request, the data buffer storage spatial cache to data read request will be returned, so method by adopting the memory space distributing correspondence before initiating read request premise in receiving buffer memory, simplify the complexity of cache management, simultaneously, Coutinuous store also reduces the difficulty that credit updates, and the hardware circuit solving the packet existence that existing employing list structure processes out of order return realizes the problem that difficulty is big.
Accompanying drawing explanation
The structural representation of the data receiver buffer storage that Fig. 1 provides for first embodiment of the invention;
The flow chart of the data receiver caching method that Fig. 2 provides for second embodiment of the invention;
The structural representation of the data receiver buffer storage that Fig. 3 provides for third embodiment of the invention;
Fig. 4 is the structural representation of TagRAM in third embodiment of the invention;
Fig. 5 receives the read-write pointer variation diagram being buffered in when receiving packet in third embodiment of the invention.
Detailed description of the invention
Now in conjunction with the mode of accompanying drawing the present invention made by detailed description of the invention and further annotate explanation.
First embodiment:
The structural representation of the data receiver buffer storage that Fig. 1 provides for first embodiment of the invention, as shown in Figure 1, in the present embodiment, data receiver buffer storage 1 provided by the invention includes:
Distribution module 11, after being used for detecting transmission data read request, distributes spatial cache for data read request;
Cache module 12, for when receiving the return data of data read request, returning the data buffer storage spatial cache to data read request.
In certain embodiments, cache module in above-described embodiment 12 is for when returning data and including two and data above bag, by two and data above bag according to sending time-sequencing, and buffer memory.
In certain embodiments, the cache module 12 in above-described embodiment by two and data above bag according to after sending time-sequencing, first packet after being additionally operable to sequence carries packet header buffer memory, and other packets remove buffer memory behind packet header.
In certain embodiments, cache module 12 in above-described embodiment by return data buffer storage to data read request spatial cache after, it is additionally operable to after return data data read request being detected all correctly return, the data of buffer memory are read from the spatial cache of data read request, carry out credit renewal the spatial cache of initialization data read request.
In certain embodiments, distribution module 11 in above-described embodiment is for obtaining the identification information of data read request, identification information according to data read request is data read request distribution spatial cache, the one_to_one corresponding of the identification information of spatial cache and the identification information of data read request; Cache module for determining the identification information of data read request according to the identification information returning data, and the corresponding relation according to the identification information of spatial cache Yu the identification information of data read request, by spatial cache extremely corresponding for return data buffer storage.
In certain embodiments, when data read request is read request, distribution module 11 in above-described embodiment is additionally operable to obtain reading data length information corresponding to read request and identification information, the reading data length information corresponding according to read request and identification information, for the spatial cache of the corresponding length of read request distribution.
Corresponding, the invention provides a kind of communication system, it includes the first equipment and the second equipment, the first equipment and the second equipment by high performance interconnection serial bus communication, and the first equipment and/or the second equipment include data receiver buffer storage 1 provided by the invention.
Second embodiment:
The flow chart of the data receiver caching method that Fig. 2 provides for second embodiment of the invention, as shown in Figure 2, in the present embodiment, data receiver caching method provided by the invention includes:
S201: after transmission data read request being detected, distribute spatial cache for data read request;
S202: when receiving the return data of data read request, will return the data buffer storage spatial cache to data read request.
In certain embodiments, including returning data buffer storage to the spatial cache of data read request in above-described embodiment: when returning data and including two and data above bag, by two and data above bag according to sending time-sequencing, and buffer memory.
In certain embodiments, in above-described embodiment by return data buffer storage to data read request spatial cache by two and data above bag according to after sending time-sequencing, also include: first packet after sequence is carried packet header buffer memory, and other packets remove buffer memory behind packet header.
In certain embodiments, method in above-described embodiment by return data buffer storage to data read request spatial cache after, also include: after return data data read request being detected all correctly return, the data of buffer memory are read from the spatial cache of data read request, carry out credit renewal the spatial cache of initialization data read request.
In certain embodiments, including for data read request distribution spatial cache in above-described embodiment: obtain the identification information of data read request, identification information according to data read request is data read request distribution spatial cache, the one_to_one corresponding of the identification information of spatial cache and the identification information of data read request; Include returning the data buffer storage spatial cache to data read request: determine the identification information of data read request according to the identification information returning data, identification information according to spatial cache and the corresponding relation of the identification information of data read request, will return the data buffer storage spatial cache to correspondence.
In certain embodiments, when data read request is read request, also including for data read request distribution spatial cache in above-described embodiment: obtain reading data length information corresponding to read request and identification information, the reading data length information corresponding according to read request and identification information, for the spatial cache of the corresponding length of read request distribution.
In conjunction with concrete application scenarios the present invention done and further annotate explanation.
3rd embodiment:
The present embodiment illustrates for high-speed serial communication field, it is provided that a kind of method and device being likely to the packet that out of order (OutofOrder) arrives by reception buffer memory (ReceiverBuffer) and attached control logical process.
PCIe (PeripheralComponentInterconnectExpress) is the high performance interconnection universal serial bus of a new generation, it has point-to-point topological structure, and high speed data transfer ability, it is widely used between the device in various computer and communication system and interconnects. When two devices communicate, request and response are all sent to the other side with the form of TLP (TransactionLayerPacket) packet. Such as, when local terminal initiates read request (MemoryReadRequest, MRd), opposite end returns the reading data of request with the TLP bag of CplD (CompletionwithData). To same read request, PCIe protocol allows to return back read data with the form of one or more CplD.When there is two and above read request in system simultaneously, all returning with the form of multiple CplD if reading data, the CplD of corresponding different read requests is likely to alternately arrive read request side. Now, the receiving terminal of read request side must process the CplD packet of out of order arrival, namely to each read request, and the Tag field according to MRd packet header respectively, from corresponding multiple CplD, recover reading data.
In order to process the packet of out of order return, practices well is the relation adopting list structure (LinkedList) to record between storage address and the packet of packet. Such as, US Patent No. 9021482B2 (time of disclosure: 2015-4-28), and the patent such as US Patent No. 20130246682A1 (time of disclosure: 2013-9-19) all adopts the scheme based on list structure. But, owing to list structure itself is more complicated, adopt hardware circuit to realize difficulty bigger; On the other hand, for PCIe receiving terminal, after data export from reception buffer memory, it is necessary to the credit that upgrades in time (Credit) is to opposite end, and the dispersion storage of list structure correspondence, this updates to credit and brings no small difficulty.
For the problems referred to above, the present embodiment proposes a kind of new reception buffer memory management method, to reduce the design complexities that receiving terminal processes the packet of out of order return.
Concrete, the reception buffer memory management method of the present embodiment offer, including:
When sending NP (Non-PostedRequest) to opposite end and asking, the TLP header packet informations such as Tag value corresponding for this request are passed to the reception caching management module of local terminal, when this request is MRd, in addition it is also necessary to reading data length information is passed to the reception caching management module of local terminal;
Receive the memory space that caching management module is the CplD distribution abundance that will return in advance according to information such as above-mentioned Tag value, reading data lengths in receiving buffer memory;
When the form of reading one or more CplD according to this returns from opposite end, Tag value according to CplD packet header, CplD is stored among the buffer address allocated in advance (to multiple CplD, except first corresponding for MRd CplD, the packet header of all the other CplD need not be stored in reception buffer memory);
After all correctly returning Deng the CplD that NP request is corresponding, from receiving the data reading correspondence buffer memory, it is the reading data that NP request is corresponding, credit renewal can be carried out to opposite end simultaneously.
The present embodiment adopts the method distributing memory space in advance, it is ensured that multiple CplD corresponding for same MRd are Coutinuous store in receiving buffer memory, simplifies the complexity of cache management; Meanwhile, Coutinuous store also reduces the difficulty that credit updates.
Below in conjunction with accompanying drawing, the present embodiment is described in detail.
Shown in Fig. 3 is the one reception cache management device of the embodiment of the present invention, and it specifically includes that
TLPDecoder31: be used for resolving the information such as TLP bag form, the Tag field in extraction packet header;
TagRAM32: send the information such as Tag value when NP asks, read request length for recording local terminal, and index relevant information when Completion returns according to Tag, whether legal detect this Completion bag;
ReceiverBuffer33: for the TLP bag after storing and resolving;
OrderingFIFO (FirstInFirstOut, first in first out) 34: for recording the sequencing that TLP bag arrives;
CompletionTimeoutTimer35: record local terminal sends the time of NP request, if Completion does not also return at the appointed time, then sends time-out information.
In the present embodiment, when local terminal sends NP request, the information such as time of this request being initiated, read request data length, Tag value is recorded and is stored in TagRAM, and the Requestvalid field in TagRAM is set to 1;Meanwhile, the write pointer rbuf_wptr of ReceiverBuffer is increased the numerical value that read request data length is corresponding, be namely that the CplD that will return distributes memory space in advance.
For tackling same NP read request, return the situation of multiple CplD, it is necessary to recording the write pointer Cpld_wptr of CplD corresponding to each read request respectively, storage is in TagRAM. TagRAM is with the Tag value of TLP for reading, writing address, and its logical storage structure is as shown in Figure 4. For solving reading and writing conflict, TagRAM can be split into multiple little RAM being reading, writing address with Tag value, it is also possible to partly or entirely adopts depositor to realize.
TLPDecoder module often receives a CplD, then read TagRAM with Tag value for address, if the Cpld_rcvd read is 1, then the Cpld_wptr of reading is write the initial address of this CplD as toward ReceiverBuffer, otherwise using rbuf_wptr as initial address. If this CplD is last CplD that not to be NP read request corresponding, after write packet tail, in addition it is also necessary to the Cpld_wptr after write packet tail is newly stored into TagRAM, as the initial write address of the follow-up CplD of this request. Meanwhile, the information such as the storage initial address of first CplD and Tag value are stored in OrderingFIFO, it is ensured that from ReceiverBuffer, sequentially read data, facilitate credit to update.
The Completion bag corresponding when Tag value the earliest in OrderingFIFO all receives, then can start reading data from ReceiverBuffer, export to rear class logic from the Decodeddata/cmd interface of Fig. 3 after the process such as decoding.
CompletionTimeoutTimer module comprises the enumerator of a free-running operation, and each clock cycle adds 1 automatically, automatically returns 0 after being added to maximum, and restarts to add up. After NP read request sends, write down current Counter Value, and be stored in NPRequestTime field corresponding in TagRAM; After corresponding Completion returns, then Cpld_rcvd corresponding for TagRAM is put 1. Meanwhile, CompletionTimeoutTimer module periodically surveys the NPRequestTime of the request of RequestValid=1 in TagRAM, if what the difference of present counter value and NPRequestTime exceeded time-out time arranges value, then Is_timeout field is put 1, namely this request occurs in that Completion time-out, it is necessary to will discharge in the ReceiverBuffer space distributed for this request in advance.
Shown in Fig. 5 is in receiving CplD process, receives the change procedure of buffer memory reading and writing pointer. In fig. 5, it is assumed that local terminal successively have issued two MRd, corresponding Tag respectively A, B. The T=0 moment, MRdA has been sent from, first CplD of corresponding TagA, namely CplDA0 has returned and has been stored in ReceiverBuffer, Cpld_wptr is updated over, but ReceiverBuffer does not also start output, read pointer Rbuf_rptr=0, owing to shifting to an earlier date allocation space write pointer Rbuf_wptr=A; In the T=1 moment, MRdB has been sent from, first CplD of corresponding TagB, and namely CplDB0 has returned and has been stored in ReceiverBuffer, and the Cpld_wptr of corresponding TagB is updated over, owing to shifting to an earlier date allocation space ReceiverBuffer write pointer Rbuf_wptr=B; The T=2 moment, second CplD of corresponding TagA, namely CplDA1 has returned and has been stored in ReceiverBuffer, Cpld_wptr and is updated over;The T=3 moment, last CplD of corresponding TagA, namely CplDA2 has returned and has been stored in ReceiverBuffer; The T=4 moment, last CplD of corresponding TagB, namely CplDB1 has returned and has been stored in ReceiverBuffer, and the CplD of corresponding TagA has all exported from ReceiverBuffer, now, read pointer Rbuf_rptr=A. Additionally, it is stored in ReceiverBuffer process at TLP bag, if in the end the stage finds that data are wrong, such as LCRC (LinkCyclicRedundancyCheck) verification makes mistakes, it is necessary to ReceiverBuffer write pointer return back to the initial address storing this data packet head.
In summary, by the enforcement of the present invention, at least there is following beneficial effect:
The invention provides a kind of new data receiver caching method, after transmission data read request being detected, distribute spatial cache for data read request; When receiving the return data of data read request, the data buffer storage spatial cache to data read request will be returned, so method by adopting the memory space distributing correspondence before initiating read request premise in receiving buffer memory, simplify the complexity of cache management, simultaneously, Coutinuous store also reduces the difficulty that credit updates, and the hardware circuit solving the packet existence that existing employing list structure processes out of order return realizes the problem that difficulty is big.
Below it is only the specific embodiment of the present invention; not the present invention is done any pro forma restriction; every any simple modification, equivalent variations, combination or modification embodiment of above done according to the technical spirit of the present invention, all still falls within the protection domain of technical solution of the present invention.
Claims (13)
1. a data receiver caching method, it is characterised in that including:
After transmission data read request being detected, distribute spatial cache for described data read request;
When receiving the return data of described data read request, by the spatial cache of described return data buffer storage to described data read request.
2. data receiver caching method as claimed in claim 1, it is characterized in that, the described spatial cache by described return data buffer storage to described data read request includes: when including two and data above bag when described return data, by said two and data above bag according to sending time-sequencing, and buffer memory.
3. data receiver caching method as claimed in claim 2, it is characterized in that, described by the spatial cache of described return data buffer storage to described data read request by said two and data above bag according to after sending time-sequencing, also include: first packet after sequence is carried packet header buffer memory, and other packets remove buffer memory behind packet header.
4. data receiver caching method as claimed in claim 1, it is characterized in that, after by the spatial cache of described return data buffer storage to described data read request, also include: after the return data described data read request being detected all correctly return, the data of buffer memory are read from the spatial cache of described data read request, carry out credit renewal, and initialize the spatial cache of described data read request.
5. the data receiver caching method as described in any one of Claims 1-4, it is characterised in that
Described for described data read request distribute spatial cache include: obtain the identification information of described data read request, identification information according to described data read request is described data read request distribution spatial cache, the one_to_one corresponding of the identification information of the identification information of described spatial cache and described data read request;
The spatial cache of described return data buffer storage to described data read request is included: determine the identification information of described data read request according to the identification information of described return data, the corresponding relation of the identification information of the identification information according to described spatial cache and described data read request, by described return data buffer storage to corresponding spatial cache.
6. data receiver caching method as claimed in claim 5, it is characterized in that, when described data read request is read request, described for described data read request distribute spatial cache also include: obtain reading data length information corresponding to described read request and identification information, the reading data length information corresponding according to described read request and identification information, for the spatial cache of the corresponding length of described read request distribution.
7. a data receiver buffer storage, it is characterised in that including:
Distribution module, after being used for detecting transmission data read request, distributes spatial cache for described data read request;
Cache module, for when receiving the return data of described data read request, by the spatial cache of described return data buffer storage to described data read request.
8. data receiver buffer storage as claimed in claim 7, it is characterised in that when described cache module is for including two and data above bag when described return data, by said two and data above bag according to sending time-sequencing, and buffer memory.
9. data receiver buffer storage as claimed in claim 8, it is characterized in that, described cache module by said two and data above bag according to after sending time-sequencing, first packet after being additionally operable to sequence carries packet header buffer memory, and other packets remove buffer memory behind packet header.
10. data receiver buffer storage as claimed in claim 7, it is characterized in that, described cache module by described return data buffer storage to described data read request spatial cache after, it is additionally operable to after the return data described data read request being detected all correctly return, the data of buffer memory are read from the spatial cache of described data read request, carry out credit renewal, and initialize the spatial cache of described data read request.
11. the data receiver buffer storage as described in any one of claim 7 to 10, it is characterised in that
Described distribution module is for obtaining the identification information of described data read request, identification information according to described data read request is described data read request distribution spatial cache, the one_to_one corresponding of the identification information of the identification information of described spatial cache and described data read request;
Described cache module for determining the identification information of described data read request according to the identification information of described return data, the corresponding relation of the identification information of the identification information according to described spatial cache and described data read request, by described return data buffer storage to corresponding spatial cache.
12. data receiver buffer storage as claimed in claim 11, it is characterized in that, when described data read request is read request, described distribution module is additionally operable to obtain reading data length information corresponding to described read request and identification information, the reading data length information corresponding according to described read request and identification information, for the spatial cache of the corresponding length of described read request distribution.
13. a communication system, it is characterized in that, including the first equipment and the second equipment, described first equipment and the second equipment by high performance interconnection serial bus communication, described first equipment and/or the second equipment include the data receiver buffer storage as described in any one of claim 7 to 112.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610120423.6A CN105681222A (en) | 2016-03-03 | 2016-03-03 | Method and apparatus for data receiving and caching, and communication system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610120423.6A CN105681222A (en) | 2016-03-03 | 2016-03-03 | Method and apparatus for data receiving and caching, and communication system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105681222A true CN105681222A (en) | 2016-06-15 |
Family
ID=56306547
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610120423.6A Pending CN105681222A (en) | 2016-03-03 | 2016-03-03 | Method and apparatus for data receiving and caching, and communication system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105681222A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111694777A (en) * | 2019-03-11 | 2020-09-22 | 苏州微影激光技术有限公司 | DMA transmission method based on PCIe interface |
| CN111724262A (en) * | 2020-06-24 | 2020-09-29 | 上海金仕达软件科技有限公司 | Follow-up packet query system of application server and working method thereof |
| CN114553776A (en) * | 2022-02-28 | 2022-05-27 | 深圳市风云实业有限公司 | Signal out-of-order control and rate self-adaptive transmission device and transmission method thereof |
| CN117724833A (en) * | 2023-12-06 | 2024-03-19 | 无锡众星微系统技术有限公司 | PCIe tag cache self-adaptive resource allocation method and device based on stream attribute |
| CN118113230A (en) * | 2024-04-23 | 2024-05-31 | 沐曦集成电路(上海)有限公司 | Method for reading memory data based on address remapping |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050223123A1 (en) * | 2004-03-31 | 2005-10-06 | Jen Michelle C | Out-of-order servicing of read requests with minimal additional storage |
| CN1797380A (en) * | 2004-12-28 | 2006-07-05 | 株式会社东芝 | Receiving apparatus, transmitting/receiving apparatus, receiving method and transmitting/receiving method |
| CN102571609A (en) * | 2012-03-01 | 2012-07-11 | 重庆中天重邮通信技术有限公司 | Recombination sequencing method of fast serial interface programmable communication interface-express (PCI-E) protocol completion with data (CplD) |
| CN103049240A (en) * | 2011-10-13 | 2013-04-17 | 北京同步科技有限公司 | Peripheral component interconnect express (PCI-E) device and data receiving reordering method thereof |
| CN105281832A (en) * | 2014-07-02 | 2016-01-27 | 北京邮电大学 | Neighbor discovery method in ultraviolet light network |
-
2016
- 2016-03-03 CN CN201610120423.6A patent/CN105681222A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050223123A1 (en) * | 2004-03-31 | 2005-10-06 | Jen Michelle C | Out-of-order servicing of read requests with minimal additional storage |
| CN1797380A (en) * | 2004-12-28 | 2006-07-05 | 株式会社东芝 | Receiving apparatus, transmitting/receiving apparatus, receiving method and transmitting/receiving method |
| CN103049240A (en) * | 2011-10-13 | 2013-04-17 | 北京同步科技有限公司 | Peripheral component interconnect express (PCI-E) device and data receiving reordering method thereof |
| CN102571609A (en) * | 2012-03-01 | 2012-07-11 | 重庆中天重邮通信技术有限公司 | Recombination sequencing method of fast serial interface programmable communication interface-express (PCI-E) protocol completion with data (CplD) |
| CN105281832A (en) * | 2014-07-02 | 2016-01-27 | 北京邮电大学 | Neighbor discovery method in ultraviolet light network |
Non-Patent Citations (2)
| Title |
|---|
| 程光，周爱平，吴桦: "《互联网大数据挖掘与分类》", 31 December 2015 * |
| 陈乃塘: "深入探索PCI Express数据交易层(下)", 《电子测试》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111694777A (en) * | 2019-03-11 | 2020-09-22 | 苏州微影激光技术有限公司 | DMA transmission method based on PCIe interface |
| CN111694777B (en) * | 2019-03-11 | 2022-03-29 | 苏州微影激光技术有限公司 | DMA transmission method based on PCIe interface |
| CN111724262A (en) * | 2020-06-24 | 2020-09-29 | 上海金仕达软件科技有限公司 | Follow-up packet query system of application server and working method thereof |
| CN111724262B (en) * | 2020-06-24 | 2024-03-22 | 上海金仕达软件科技股份有限公司 | Subsequent package query system of application server and working method thereof |
| CN114553776A (en) * | 2022-02-28 | 2022-05-27 | 深圳市风云实业有限公司 | Signal out-of-order control and rate self-adaptive transmission device and transmission method thereof |
| CN114553776B (en) * | 2022-02-28 | 2023-10-10 | 深圳市风云实业有限公司 | Signal disorder control and rate self-adaptive transmission device and transmission method thereof |
| CN117724833A (en) * | 2023-12-06 | 2024-03-19 | 无锡众星微系统技术有限公司 | PCIe tag cache self-adaptive resource allocation method and device based on stream attribute |
| CN117724833B (en) * | 2023-12-06 | 2024-05-28 | 无锡众星微系统技术有限公司 | PCIe tag cache self-adaptive resource allocation method and device based on stream attribute |
| CN118113230A (en) * | 2024-04-23 | 2024-05-31 | 沐曦集成电路(上海)有限公司 | Method for reading memory data based on address remapping |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106933775B (en) | Data processing system and method for data processing | |
| CN105681222A (en) | Method and apparatus for data receiving and caching, and communication system | |
| CN100481043C (en) | Method, system, and program for handling input/output commands | |
| US7069373B2 (en) | USB endpoint controller flexible memory management | |
| CN114780458A (en) | Data processing method and storage system | |
| CN102147778B (en) | Data transmission system based on half-duplex serial bus and transmission control method | |
| CN107783727B (en) | Access method, device and system of memory device | |
| CN103631534B (en) | Data storage system and managing method thereof | |
| WO2023093334A1 (en) | System for executing atomic operation, and atomic operation method and apparatus | |
| CN111651396B (en) | Optimized PCIE (peripheral component interface express) complete packet out-of-order management circuit implementation method | |
| CN105373484A (en) | Memory distribution, storage and management method in network communication chip | |
| US8589607B1 (en) | Methods and structure for hardware management of serial advanced technology attachment (SATA) DMA non-zero offsets in a serial attached SCSI (SAS) expander | |
| CN113852533B (en) | Multi-channel data communication system and method and electronic equipment | |
| CN104866432A (en) | Memory subsystem with wrapped-to-continuous read | |
| CN112256624B (en) | DMA communication device, chip, equipment and method for high-speed interconnection network interface chip | |
| CN107783927B (en) | Circuit conversion system and method for connecting AXI interface and DMA interface | |
| CN107783909B (en) | Memory address bus expansion method and device | |
| CN1705929B (en) | Method and system for returning data to read requests received over a bus | |
| US7752355B2 (en) | Asynchronous packet based dual port link list header and data credit management structure | |
| CN116719764A (en) | Data synchronization method, system and related device | |
| CN116009770A (en) | Read response circuit, method, data transmission system and related equipment | |
| CN111694777B (en) | DMA transmission method based on PCIe interface | |
| CN105022707B (en) | Interface unit device | |
| CN105721338A (en) | Method and device for processing received data | |
| CN116756078B (en) | Notification method and device of pcie data packet and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 518057 Guangdong city of Shenzhen province Nanshan District high tech Industrial Park Road eight South South technology Howare Technology Building 16 Applicant after: Shenzhen Pango Microsystems Co., Ltd. Address before: 518057 Guangdong city of Shenzhen province Nanshan District high tech Industrial Park Road eight South South technology Howare Technology Building 16 Applicant before: SHENZHEN PANGO MICROSYSTEMS CO., LTD. |
|
| COR | Change of bibliographic data | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160615 |