US20010054121A1 - Method and circuit for controlling a first-in-first-out (fifo) buffer using a bank of fifo address registers capturing and saving beginning and ending write-pointer addresses - Google Patents
Method and circuit for controlling a first-in-first-out (fifo) buffer using a bank of fifo address registers capturing and saving beginning and ending write-pointer addresses Download PDFInfo
- Publication number
- US20010054121A1 US20010054121A1 US09/232,764 US23276499A US2001054121A1 US 20010054121 A1 US20010054121 A1 US 20010054121A1 US 23276499 A US23276499 A US 23276499A US 2001054121 A1 US2001054121 A1 US 2001054121A1
- Authority
- US
- United States
- Prior art keywords
- write
- fifo buffer
- data
- register
- fifo
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000012546 transfer Methods 0.000 claims description 16
- 238000004886 process control Methods 0.000 claims description 15
- 238000001514 detection method Methods 0.000 claims description 9
- 238000013500 data storage Methods 0.000 claims description 7
- 230000003139 buffering effect Effects 0.000 claims description 6
- 230000003068 static effect Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 description 27
- 230000001360 synchronised effect Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 102100029272 5-demethoxyubiquinone hydroxylase, mitochondrial Human genes 0.000 description 5
- 101000770593 Homo sapiens 5-demethoxyubiquinone hydroxylase, mitochondrial Proteins 0.000 description 5
- 101000738400 Homo sapiens Cyclin-dependent kinase 11B Proteins 0.000 description 5
- 101100113692 Caenorhabditis elegans clk-2 gene Proteins 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 2
- 230000009977 dual effect Effects 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 101100328957 Caenorhabditis elegans clk-1 gene Proteins 0.000 description 1
- 241000549194 Euonymus europaeus Species 0.000 description 1
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F5/00—Methods or arrangements for data conversion without changing the order or content of the data handled
- G06F5/06—Methods or arrangements for data conversion without changing the order or content of the data handled for changing the speed of data flow, i.e. speed regularising or timing, e.g. delay lines, FIFO buffers; over- or underrun control therefor
- G06F5/10—Methods or arrangements for data conversion without changing the order or content of the data handled for changing the speed of data flow, i.e. speed regularising or timing, e.g. delay lines, FIFO buffers; over- or underrun control therefor having a sequence of storage locations each being individually accessible for both enqueue and dequeue operations, e.g. using random access memory
- G06F5/12—Means for monitoring the fill level; Means for resolving contention, i.e. conflicts between simultaneous enqueue and dequeue operations
- G06F5/14—Means for monitoring the fill level; Means for resolving contention, i.e. conflicts between simultaneous enqueue and dequeue operations for overflow or underflow handling, e.g. full or empty flags
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0614—Improving the reliability of storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
- G06F3/0656—Data buffering arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0674—Disk device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F5/00—Methods or arrangements for data conversion without changing the order or content of the data handled
- G06F5/06—Methods or arrangements for data conversion without changing the order or content of the data handled for changing the speed of data flow, i.e. speed regularising or timing, e.g. delay lines, FIFO buffers; over- or underrun control therefor
- G06F5/065—Partitioned buffers, e.g. allowing multiple independent queues, bidirectional FIFO's
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F5/00—Methods or arrangements for data conversion without changing the order or content of the data handled
- G06F5/06—Methods or arrangements for data conversion without changing the order or content of the data handled for changing the speed of data flow, i.e. speed regularising or timing, e.g. delay lines, FIFO buffers; over- or underrun control therefor
- G06F5/10—Methods or arrangements for data conversion without changing the order or content of the data handled for changing the speed of data flow, i.e. speed regularising or timing, e.g. delay lines, FIFO buffers; over- or underrun control therefor having a sequence of storage locations each being individually accessible for both enqueue and dequeue operations, e.g. using random access memory
- G06F5/12—Means for monitoring the fill level; Means for resolving contention, i.e. conflicts between simultaneous enqueue and dequeue operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2205/00—Indexing scheme relating to group G06F5/00; Methods or arrangements for data conversion without changing the order or content of the data handled
- G06F2205/10—Indexing scheme relating to groups G06F5/10 - G06F5/14
- G06F2205/108—Reading or writing the data blockwise, e.g. using an extra end-of-block pointer
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2205/00—Indexing scheme relating to group G06F5/00; Methods or arrangements for data conversion without changing the order or content of the data handled
- G06F2205/12—Indexing scheme relating to groups G06F5/12 - G06F5/14
- G06F2205/126—Monitoring of intermediate fill level, i.e. with additional means for monitoring the fill level, e.g. half full flag, almost empty flag
Definitions
- the invention relates generally to controlling a first-in-first-out (FIFO) buffer and more precisely to a FIFO controller for buffering a plurality of disparate data blocks or entities flowing between asynchronous data processes in a manner preventing overlap or corruption of data.
- FIFO first-in-first-out
- FIFO structures are well known for buffering data passing between two data handling processes which operate asynchronously with respect to each other.
- a FIFO memory is a structure wherein data words, objects, blocks, or entities are taken out of the structure in the order of receipt.
- a sender places data objects into the FIFO structure, while a receiver collects the objects from the structure.
- a first-in-first-out (FIFO) buffer 44 is used to provide temporary storage for data that is being transferred between two or more asynchronous data processing systems. These data processing systems typically have specialized clocking circuits, e.g. Clk- 1 and Clk- 2 operating at predetermined frequencies (phases) which are asynchronous with respect to each other.
- the FIFO buffer 44 shown in the FIG. 1 example is, for example, a dual port memory array having a preferred row dimension of thirty two bits wide with thirty two rows in height or depth (i.e. a 32 by 32 memory array).
- One typical data processing system is a disk-type magnetic recording system or disk drive 30 .
- the disk drive 30 uses a magnetic transducer element, or head 31 to record information onto (i.e., write) and to retrieve information from (i.e., read) a magnetic medium formed on a surface of a rotating disk 32 .
- Each storage disk 32 comprises an annular substrate onto which is deposited a magnetic recording medium.
- FIG. 1 shows a single head 31 and disk 32 , in practice multiple storage surfaces and heads are frequently employed. In the FIG. 1 example the storage surface of disk 32 is divided into thousands of concentric, annular bands, or “tracks” 33 each having a predetermined radial extent.
- Each head 31 is supported in close proximity to an associated disk 32 surface by a head positioning assembly, or actuator 34 , that supports the head 31 near the disk 32 surface and moves it from one radial position to another, thereby permitting use of a single head 31 for reading and writing along each of the multiple concentric tracks 33 .
- the positioner assembly 34 for each head 31 or group of heads 31 includes an actuator arm 35 and an actuator motor 36 .
- the actuator motor 36 moves the actuator arm 35 to change the position of the head 31 with relation to the tracks 33 on the disk 32 .
- a disk drive 30 having a plurality of disks 32 stacked on a common spindle will typically employ a single actuator motor 36 to move a corresponding number of ganged-together actuator arms 35 and heads 31 in unison.
- a disk controller (not shown) controls operations of the disk spindle and the actuator motor 36 .
- the disk drive 30 further includes a disk channel 37 , such as a partial-response maximum-likelihood (PRML) synchronous sampling data detection channel, for encoding, detecting, decoding and controlling flow of data read from or written to the disk 32 at a data rate in accordance with a disk clock CLK- 1 .
- the disk clock CLK- 1 may remain fixed, or it may be varied as a function of radius of a particular data track 33 (since relative rotational velocity varies with track radius).
- the disk drive 30 typically transmits data back and forth between itself and a random-access-memory (RAM) buffer 25 .
- the RAM buffer 25 includes one or a plurality of electronic data storage integrated circuit memory chips for storing data therein.
- the RAM buffer 25 is coupled with a RAM Port 26 .
- the RAM Port 26 is coupled to the disk drive 30 via a disk-direct-memory-access (DDMA) 40 , which is defined within the disk drive 30 .
- the RAM Port 26 circuit is further connected to several other clients including: a RAM buffer manager 24 ; a standard bus interface 28 , such as a Small-Computer-System Interface (SCSI), which connects to a host computer 22 and also may be connected as well as to several other clients (n).
- SCSI Small-Computer-System Interface
- the RAM Port 26 circuit provides access arbitration among the clients seeking access to the RAM buffer 25 , including the DDMA 40 , host interface 28 and other clients (n) such as a disk drive controller.
- the DDMA 40 generally includes a RAM Port Interface circuit 42 , a FIFO buffer 44 , and a disk interface circuit 45 .
- the RAM Port interface 42 communicates data as well as control information to the RAM buffer 25 via the RAM Port 26 and RAM buffer manager 24 .
- the FIFO buffer 44 is connected to the RAM Port Interface 42 circuit and acts as a temporary storage space for data that is in the process of being transferred to/from the disk drive 30 and the RAM 25 .
- the disk interface circuit 45 is connected to the disk channel 37 , via a bus 27 , as well as to the FIFO buffer 44 .
- the disk interface 45 operates to communicate data to/from the disk channel 37 and storage surface of disk 32 .
- the DDMA 40 further includes a disk read/write process control 41 and a RAM read/write process control 43 .
- the disk read/write process control circuit 41 is connected to the disk interface 45 and is synchronous with CLK- 1 of the disk channel 37 .
- the RAM read/write process control circuit 43 is connected to the RAM Port interface 42 and is synchronous with a second clock CLK- 2 used to write data to and from the RAM buffer 25 .
- the disk and RAM read/write process control circuits 41 , 43 enable the RAM Port interface 42 and the Disk Drive interface 45 to either read data from or write data to the FIFO buffer 44 depending on the direction of data flow between the RAM buffer 25 and the disk drive 30 .
- a typical data transfer from the disk drive 30 to the RAM buffer 25 a plurality of data blocks written in tracks 33 are sensed as magnetic flux transitions by the head 31 and recovered as binary data by the disk channel 37 .
- the recovered data is transmitted to the FIFO buffer 44 via the disk interface 45 .
- a write-pointer register 48 Prior to writing data to the FIFO buffer 44 , a write-pointer register 48 is preset to point to the first row of the FIFO 44 .
- the disk read/write process control circuit 41 is enabled for writing data to the FIFO buffer 44
- the RAM read/write process control circuit 43 is enabled for reading data from the FIFO buffer 44 .
- the disk interface 45 begins writing 32 -bit data segments to the FIFO buffer 44 synchronously with clock CLK- 1 . Additionally, the write-pointer address register 48 is incremented as each data segment is written into the FIFO buffer 44 . Once the FIFO buffer 44 accumulates a predetermined number of data segments, a control signal is issued to the RAM Port Interface 42 indicating that the FIFO buffer 44 has data which needs to be read out.
- the RAM Port Interface 42 arbitrates with the RAM Port 26 for direct access to the RAM buffer 25 via the RAM buffer manager 24 .
- the RAM Port Interface 26 reads data out of the FIFO buffer 44 and transmits the data to the RAM buffer 25 via RAM buffer manager 24 .
- a read-pointer address register 46 Prior to the RAM Port Interface 42 reading data out of the FIFO buffer 44 , a read-pointer address register 46 is preset to point to the beginning of data written into the FIFO buffer 44 , i.e. the first row in this example. Thereafter, the RAM Port Interface 42 begins reading data segments from the FIFO buffer 44 in synchronism with clock CLK- 2 .
- the read-pointer address register 46 is incremented as each data segment is read from the FIFO buffer 44 . Once the read-pointer address register 45 is incremented to point to a next buffer row, the present buffer row is free to be written with a next data segment of the data block being transferred.
- the FIFO buffer 44 holds only 32 4-byte segments (128 bytes or 1024 bits) the buffer 44 typically cannot store an entire disk data block (e.g. 512 bytes or 4096 bits) at once. Accordingly, the FIFO buffer 44 is cycled (rolls over) several times during the transfer of a single block of data. After the data block is both written to and read from the FIFO buffer 44 , the write-pointer address register 48 and the read-pointer address register 46 are reset and the above process is repeated for transferring a next data block from the disk drive 30 to the host 22 .
- a typical data transfer from the RAM buffer 25 to the disk drive 30 a plurality of data blocks are transferred from the host 22 to the RAM buffer 25 and then the blocks of data are subsequently transferred to the FIFO buffer 44 via the RAM Port Interface 42 . Similar to that of transferring data from the drive 30 to the RAM buffer 25 , the FIFO buffer 44 presets the write-pointer register 48 to point to the beginning of the FIFO 44 . Next, the RAM read/write process control circuit 43 is enabled for writing data to the FIFO buffer 44 while the disk read/write process control circuit 41 is enabled for reading data from the FIFO buffer 44 .
- the RAM Port Interface 42 begins preemptively writing data segments to the FIFO buffer 44 in synchronism with clock CLK- 2 . This is to ensure that the data is available for transfer to the disk 32 when the head 31 arrives at the correct location over the disk 32 . Additionally, the write-pointer address register 48 is incremented as each data segment is written into the FIFO buffer 44 .
- the RAM buffer 25 write process pauses and waits for space to become available in the FIFO buffer 44 . Space becomes available in the FIFO buffer 44 as a result of the disk interface 45 read process, which reads data from the FIFO buffer 44 and sends the data to the disk channel 37 for subsequent storage on the disk 32 . When sufficient space becomes available in the FIFO buffer 44 , the process of transferring data from the RAM buffer 25 to the FIFO buffer 44 is resumed. This process is repeated until all data is transferred from the RAM buffer 25 to the disk 32 .
- the read-pointer address register 46 Prior to the disk drive interface 45 reading data out of the FIFO 44 , the read-pointer address register 46 is preset to the beginning of the FIFO buffer 44 . Thereafter, the disk drive interface 45 begins reading data segments from the FIFO buffer 44 synchronously with clock CLK- 1 . Additionally, the read-pointer address register 46 is incremented synchronously with each data segment read from the FIFO buffer 44 . After the data block is both written to and read from the FIFO buffer 44 , the write-pointer address register 48 and the read-pointer address 44 are reset and the above process is repeated for transferring the next data block from the host 22 to the disk drive 30 .
- this delay may result in forcing the disk drive 30 to wait until the FIFO buffer 44 is again receptive to receiving data.
- Such buffer access waiting period may cause the disk drive 30 incur another complete revolution of the storage disk 32 to arrive at the same data in a track 33 and then read the data.
- the wait associated with a second disk revolution to read a track is known in the art as “rotational latency”, and excessive rotational latencies degrade disk drive system performance.
- Another problem occurs when writing a sequence of contiguous data blocks from the disk drive 30 to the FIFO buffer 44 , via the disk interface 45 , if there is insufficient space in the buffer 44 to accommodate the data.
- This full condition of the FIFO buffer 44 occurs when the RAM Port Interface 42 cannot read data out of the FIFO 44 fast enough to make room for data being written into the FIFO 44 . If the FIFO buffer 44 does not have space for data being written thereto from the disk drive 30 , an overflow condition occurs, which can result in a data transfer error.
- a hitherto unsolved need has remained for a FIFO buffer control method and circuit that facilitates simultaneous multiple block writing to and reading from a FIFO buffer without data overlap. Additionally, a FIFO buffer control circuit and method is needed that does not require a write-pointer register and a read-pointer register to be reset after each data block transfer so that a plurality of contiguous data blocks, which are much larger than the FIFO storage capacity, can be seamlessly transferred back and forth between the disk drive and the host computer.
- An object of the present invention is to provide a FIFO buffer control that enables an end of a first data block to occupy a FIFO buffer register simultaneously with a beginning of a second data block without data overlap, in a manner overcoming limitations and drawbacks of the prior art.
- Another object of the present invention is to provide a FIFO buffer control that does not require the write-pointer or the read-pointer to be reset after each data block transfer.
- a FIFO buffer control circuit enables a FIFO buffer to simultaneously contain portions of plural adjacent data blocks without data overlap.
- the FIFO control captures and saves a FIFO buffer write-pointer address that is associated with beginning of a first data block to be written to the FIFO buffer.
- the FIFO control captures and saves a buffer write-pointer address associated with the end of the first data block.
- the FIFO control captures and saves FIFO buffer write-pointer addresses associated with a beginning of a second data block to be written to the FIFO buffer.
- the FIFO buffer control captures and saves a buffer write-pointer address associated with the end of the second data block. This process is repeated for capturing and saving the write-pointer addresses associated with each successive data block written into the FIFO buffer.
- a read pointer address register of the FIFO buffer is loaded with the previously saved write pointer address that corresponds to the beginning of the first data block to be read. Thereafter, data is read from the FIFO buffer until the end of the first data block is detected. At this instant, the read pointer address register is loaded with the saved write-pointer address associated with the beginning of the second data block written to the FIFO buffer. Then, the second data block is read from the FIFO buffer until the end of the second data block is detected. This read process is repeated until all data blocks written into the FIFO buffer by the sending unit have been read out to the receiving unit.
- the write-pointer address associated with the beginning and ending of each data block written to the FIFO buffer is captured and saved, successive data blocks can be sequentially written thereto without resetting the write-pointer address register. Additionally, since the read-pointer address register is loaded with the write-pointer address associated with the beginning of each data block, the read-pointer address does not have to be reset after each data block is read from the FIFO buffer. Most importantly, the FIFO control circuit enables segments of a plurality of data blocks to occupy the FIFO buffer register simultaneously without data overlap between such blocks, because the beginning and ending pointers of each data block have been previously captured and saved.
- FIG. 1 is a high level block diagram of a conventional first-in-first-out (FIFO) buffer circuit within a disk drive system for buffering data between a disk process synchronized to a disk read-write clock and a RAM buffer process synchronized to a second clock. Connection of the disk drive system via an interface to a host computer is also shown.
- FIFO first-in-first-out
- FIG. 2 is similar to FIG. 1 and shows a high level block diagram of a first-in-first-out (FIFO) buffer control circuit in accordance with principles of the present invention for buffering multi-block data flow between a disk process synchronized to a disk read-write clock and a RAM buffer process synchronized to a second clock, without data overlap.
- FIFO first-in-first-out
- FIG. 3 exemplifies a circuit implementation of the FIG. 2 FIFO buffer control circuit in accordance with principles of the present invention.
- FIG. 4 is a flow chart illustrating a method of writing multi-block data to the FIG. 2 FIFO buffer according to principles of the present invention.
- FIG. 5 is a flow chart illustrating a method of reading multi-block data from the FIG. 2 FIFO buffer according to principles of the present invention.
- a disk drive 130 including a DDMA 140 having one preferred embodiment of the present invention is set forth.
- the DDMA 140 controls the bi-directional transfer of data between first and second asynchronous data processing systems, such as a random-access-memory (RAM) buffer 25 and a disk channel 37 of a disk drive system 30 .
- RAM random-access-memory
- the RAM buffer 25 includes one or a plurality of electronic data storage dynamic random access (DRAM) or static random access (SRAM) integrated circuit memory chips for storing data that may subsequently be accessed by the host computer processor 22 via the interface 28 .
- DRAM electronic data storage dynamic random access
- SRAM static random access
- the RAM Port 26 circuit provides arbitration among the number of clients, including the RAM buffer manager 24 , for direct communication with the RAM buffer 25 and/or the DDMA 140 .
- the RAM Port 26 circuit is coupled to the disk drive via disk-direct-memory-access (DDMA) 140 circuit which embodies principles of the present invention.
- DDMA disk-direct-memory-access
- the FIFO buffer 144 includes a data storage space or FIFO buffer space 149 .
- the dimensions of the FIFO buffer 149 is a row width of thirty-two bits by thirty-two rows (32 ⁇ 32 memory array).
- the FIFO buffer 144 further includes a read-pointer address register 146 that stores a read-pointer address associated with a location in the FIFO buffer 144 where data is read.
- the FIFO buffer 144 includes a write-pointer address register 148 that stores a write-pointer address associated with a location in the FIFO buffer 144 where data is written. So far, the description of the FIFO buffer 144 is essentially the same as the FIFO buffer 44 of FIG. 1.
- a bank of FIFO write address registers 150 is coupled to a FIFO buffer 144 as well as to a RAM Port Interface circuit 142 and a disk drive interface circuit 145 .
- the bank of FIFO address registers 150 includes four FIFO buffer write address registers, a first write-start register 152 a , a first write-end address register 152 b , a second write-start register 154 a , and a second write-end register 154 b .
- the registers 152 a , 152 b , 154 a and 154 b have parallel inputs connected to receive addresses from a write-pointer address register 148 .
- Write-start address registers 152 a and 154 a have outputs connected in parallel leading to a read-pointer address register.
- Write-end address registers 152 b and 154 b have outputs connected in parallel leading to a FIFO status detection circuit 147 .
- the first write-start address register 152 a is controlled to capture and save the beginning write-pointer address associated with the beginning of a first data block to be written to the FIFO buffer 44 .
- the first write-end address register 152 b is controlled to pass the active write pointer through the FIFO status logic, or capture and save an ending write-pointer address associated with the ending of the first data block written to the FIFO buffer 144 .
- the second write-start address register 154 a is controlled to capture and save the beginning write-pointer address associated with beginning of a second data block to be written to the FIFO buffer 144 . Furthermore, the second write-end address register 154 b is also controlled to pass the active write pointer through the FIFO status logic, or capture and save an ending write-pointer address associated with the ending of the second data block transferred to the FIFO buffer 44 . In this manner, both the first pair of FIFO address registers 152 a , 152 b , and the second pair of FIFO address registers 154 a and 154 b , capture and save the beginning and ending write-pointer addresses associated with first and second data blocks written to the FIFO buffer 144 . The captured address pointers enable the FIFO buffer 144 to store segments of multiple data blocks without overlapping data between the data blocks and without data corruption in the event of an aborted data transfer operation.
- control of the address registers is by the RAM Port interface 142 under the direction of a RAM read/write process control 143 .
- control of the address registers is by the Disk interface 145 under the direction of a disk read/write process control 141 .
- Suitable multiplexing enables dual access to the address registers 152 a , 152 b , 154 a , and 154 b via both the RAM port interface 142 and the disk interface 145 .
- the FIFO buffer 144 further includes a FIFO status detection circuit 147 .
- the FIFO status detection circuit 147 includes a comparator for comparing an address held in the read-pointer address register 146 with an address held in the write-pointer address register 148 in order to determine the number of data segments currently held in the FIFO buffer 144 . More precisely, the FIFO status detection circuit 147 compares the address held in the read-pointer address register 146 with an incrementing write-pointer address value held in either register 152 b or 154 b .
- This incrementing write-pointer address held in either register 152 b or 154 b which is generated by the write-pointer address register 148 , increments until an end of a data segment is reached, whereupon this address or write-end address held in either register 152 b or 154 b is captured and stored therein.
- the FIFO status detection circuit 147 compares the address held in the read-pointer address register 146 with the captured and stored write-end address held in either register 152 b or 154 b . In comparing the write-pointer address with the read-pointer address in such a manner, information such as whether the FIFO buffer 144 is full, half-full, or empty is readily determined.
- FIG. 3 exemplifies one preferred circuit implementation of the FIG. 2 FIFO DDMA circuit 140 .
- the bank of address registers 150 includes four multiplexer flip-flops (MUX-flops) 52 a , 52 b , 54 a and 54 b which implement FIFO buffer address registers 152 a , 152 b , 154 a and 154 b respectively.
- Each of the MUX-flops receives an enable control signal over lines 55 - 58 from a write-state control circuit 60 .
- the write-state control circuit 60 includes combinatorial logic coupled with the RAM Port interface 142 and the Disk Drive interface 145 .
- the controlling interface 142 or 145 provides several control signals to the write-state control circuit 60 including: save write-start address over line 63 , and a write block sequence number over line 62 ; and save write-end address over line 61 .
- a first block A is assigned a write sequence number of 0, while a second block B is assigned a write sequence number of 1.
- a third block C is assigned a write sequence number of 0, etc.
- Activation of the save write-start address over line 63 , and its associated write sequence number control signals over line 62 causes one of the write-start flip-flops 52 a or 54 a to capture and save the start-write pointer address value for the identified block at a particular instant in time.
- This captured and saved address value is associated with the beginning of a data block that is imminently to be written to the FIFO buffer 44 .
- the bank of registers 150 receives the FIFO buffer write-pointer address values from the write-pointer address register circuit 148 over a line 70 .
- the write-pointer address register circuitry 148 includes a single MUX-flop 48 a connected to receive an increment-write-address control signal over a line 72 from either the disk drive interface 142 or the RAM Port interface 145 , depending on which one is enabled for writing data to the FIFO buffer 144 .
- the increment-write-address control signal on line 72 enables the write-pointer address register 148 to increment the write pointer address value stored therein as each data segment 12 is written to a row of the FIFO buffer 44 .
- An adder 48 b adds a binary one to the output of the write MUX-flop 48 a on a write-address bus 49 and feeds that value back to a data input of the MUX-flop 48 a .
- the write-address bus 49 is connected to address the FIFO buffer array 149 as well as to provide starting and ending block addresses to the address register bank 150 .
- the read-pointer address circuit 146 is also coupled to the outputs of the write-start address registers 52 a and 54 a .
- the read-pointer address circuit 146 receives address values from the address register 52 a over a line 74 and receives address values from the address register 54 a over a line 75 .
- the values received over lines 74 , 75 are indicative of the beginning row address of a data block to be read from FIFO buffer 144 .
- Line 74 provides a value indicative of the write-start address associated with a first data block written to the FIFO buffer 144 .
- line 75 provides another value indicative of the write-start address value associated with a second data block written to the FIFO buffer 144 .
- the read-pointer address circuit 146 includes a single MUX-flop 46 a , which accepts the saved write-start address values received from a MUX-flop 52 a or 54 a via line 74 or 75 . Additionally, the read-pointer address circuit 146 receives control signals from a read-state control circuit 80 over lines 81 - 83 .
- the read-state control circuit 80 includes combinatorial logic that receives a control signal from either the disk drive interface 145 or the RAM Port interface 142 , depending on data flow direction and which of the two are enabled for reading data from the FIFO buffer 144 .
- the control signals received by the read-state control circuit 80 include: load read address over line 84 , a block read sequence number, i.e., zero or one, over a line 85 , and an increment read address over a line 83 .
- the load read address control signal on line 84 enables the write-start address of a data block held in a write-start MUX-flop 52 a or 54 a to be loaded into the read-pointer address register circuit 146 .
- the read sequence control signal on line 85 enables the read pointer address register circuit 146 to select either MUX-flop 52 a or 54 a for reading either the first or the second data block written to the FIFO buffer 144 .
- the increment read address pointer control signal on line 83 enables the read-pointer address register circuit 146 to increment the MUX-flop 46 a as each data segment (row) is read from the FIFO buffer 44 .
- An adder 46 b adds a binary one to an output of the read address MUX-flop 46 a on a read address bus 47 and feeds the sum to a data input of the MUX-flop 46 a.
- the FIFO buffer status circuit 147 includes a first comparator 90 , a second comparator 91 , and a third comparator 92 .
- the first comparator circuit 90 compares the next FIFO buffer write address value input to the write-pointer address circuit 148 via a path 93 with a FIFO buffer read address value input to the read-pointer address circuit 146 via a path 94 .
- the first comparator circuit 90 subtracts the read address value from the write address value and determines from the difference whether the FIFO buffer 144 is either less than or equal to half full or whether the FIFO buffer 144 is full.
- the second comparator circuit 91 receives a registered current write or write-end address value from one of the write-end address registers 52 b or 54 b via a path 95 from a multiplexer circuit 96 .
- This address value may be actively following the current write address pointer 48 a , or the captured and stored write-end address if all data has been written to the FIFO buffer 144 .
- the multiplexer circuit 96 selects between the saved write-end address of a first data block or the saved write-end address of a second data block previously captured and saved in the bank 150 .
- the multiplexer circuit 96 selects between these saved addresses based on the read sequence control signal transmitted to the multiplexer over line 85 .
- the second comparator circuit 91 receives the current FIFO read address value from the read-pointer address circuit 146 over FIFO read address path 47 . Thereafter, the second comparator circuit 91 subtracts the current FIFO read address value from the selected block write-end address value received from one of the write-end registers 52 b and 54 b via multiplexer 96 in order to determine if the FIFO buffer 144 is greater than half full.
- the third comparator circuit 92 subtracts a next row read address taken from the input of the read address MUX-flop 46 a from the registered write-end address value from one of the write-end registers 52 b , or 54 b as selected by multiplexer 96 and provided over path 95 to determine if the FIFO buffer 44 is empty.
- a write-address validation circuit 86 for each block A and B is coupled with the read-pointer address circuit 146 and read-state control circuit 80 for verifying that a data block associated with the saved write-start address previously captured and saved in the bank of registers 150 was actually written.
- the write-address validation circuit 86 includes a first MUX-flop 87 coupled with a second MUX-flop 88 whereby the first MUX-flop 87 is enabled if either sequence of the read-pointer address register circuit 146 is loaded with a read address value. Thereafter, the second MUX-flop 88 puts out a valid signal on line 89 when the write-pointer address register 148 is incremented via line 72 , thereby writing data to the first FIFO buffer 144 location.
- the RAM Port Interface circuit 142 includes a state machine which implements two control processes illustrated in the flow outlines in FIGS. 4 and 5.
- the disk interface circuit 145 also includes a state machine which implements the two control processes illustrated in the flow outlines in FIGS. 4 and 5.
- the RAM Port Interface circuit 142 is executing the FIG. 4 flow diagram
- the disk interface circuit 145 will be executing the FIG. 5 flow diagram. This enables the RAM port interface 142 to write data to the FIFO buffer 144 and enables the disk interface circuit 145 to read the data from the FIFO buffer 144 .
- the disk interface circuit 145 will be executing the FIG. 4 flow diagram. This enables the RAM port interface 142 to read data from the FIFO buffer 144 and enables the disk interface circuit 145 to write data to the FIFO buffer 144 .
- the RAM Port Interface circuit 142 receives a command to start writing data blocks to the FIFO buffer 144 at step 400 .
- the disk read process control 141 circuit Prior to writing the first data block to the FIFO buffer 144 , the disk read process control 141 circuit is enabled to permit the disk drive interface to read data from the FIFO buffer 144 and the RAM read process control 143 circuit is enabled to permit the RAM Port interface 142 to write data to the FIFO buffer 144 at step 405 .
- the RAM Port Interface 142 sends a control signal to the FIFO buffer 144 to determine if there is space in the FIFO to accommodate data to be written therein at step 410 .
- the data transfer process pauses at step 410 until there is space available for data in the FIFO buffer 144 .
- a start-write address associated with the beginning of the data block to be written to the FIFO buffer 144 is captured and saved at step 420 . This start-write address is saved in the first start-write register 52 a defined within the first pair 152 a , 152 b of registers incorporated within the bank of registers 150 .
- data segments (FIFO rows) of the first data block are sequentially written to the FIFO buffer 144 at step 430 .
- the RAM Port Interface 142 sends a control signal to the write-pointer address register 148 to increment the write-pointer address stored therein at step 440 as each data segment is written to the FIFO buffer 144 .
- the write-pointer address is incremented at step 440 until the first data block has been successfully written to the FIFO buffer 144 .
- the process of writing data segments to the FIFO buffer 144 may pause, if it is determined in step 455 that the FIFO buffer 144 does not have space for data to be written.
- the write pointer address is captured and saved in the first write-end register 152 b at a step 460 .
- step 470 After writing the first data block to the FIFO buffer 144 , it is determined at a step 470 whether another data block needs to be written to the FIFO buffer 144 , and if not, the process ends at step 480 . On the other hand, if it is determined that another data block needs to be written to the FIFO buffer 144 , the write-pointer address register 148 is permitted to continue incrementing concomitant with data segments defined within the next successive data block that is written to the FIFO buffer 144 . However, the beginning and ending addresses of this next data block are captured and saved in the second pair 154 a , 154 b , of FIFO buffer address registers.
- a third successive data block is written thereto.
- the beginning and ending write-pointer addresses of this data block are captured and stored in the first pair 152 a , 152 b , of FIFO buffer address registers defined within the bank 150 .
- the first pair 152 a , 152 b , and the second pair 154 a , 154 b , of FIFO buffer address registers alternate in capturing and storing the beginning and ending write-pointer addresses for the plurality of data blocks written to the FIFO buffer 44 .
- the bank 150 of FIFO address registers may be expanded by adding additional pairs of address registers (not shown) connected in parallel with registers 152 a , 152 b , and 154 a , 154 b . In such an arrangement, segments of more than two data blocks can simultaneously occupy the FIFO buffer 144 at the same time.
- the disk drive interface 145 receives a command to begin reading data from the FIFO buffer 44 at a step 500 . Thereafter, the disk drive interface circuit 145 determines if the FIFO buffer 144 has accumulated a predetermined number of data segments at step 510 defined within the first data block. When it is determined in step 510 that the FIFO buffer 144 contains data to be read, the disk drive interface 145 loads the read-pointer address register 146 with the FIFO buffer 144 write-pointer address previously saved in the first start-write register 152 a at step 520 .
- the disk drive interface 145 When data is available for reading from the FIFO buffer 144 , the disk drive interface 145 reads data from the FIFO buffer 144 at step 530 . At the same time, the disk drive interface 145 sends a control signal to the read-pointer address register 146 to increment the read-pointer address stored therein concomitant with each data segment read from the FIFO buffer 144 at step 540 . The read-pointer address register 146 is incremented until the first data block has been successfully read from the FIFO buffer 144 at step 530 .
- the FIFO status detection circuit 147 compares the active read-pointer address register 146 (incrementing) with the active write-pointer address register 148 (incrementing) to determine when the first data block has been completely written to and read from the FIFO buffer 144 .
- the FIFO buffer status register 147 switches from comparing the active read-pointer address register 146 with the active write-pointer address register 148 to comparing the active read-pointer address register 146 with a static write-pointer address saved in the first write-end register 152 b . In this manner the FIFO status detection circuit 147 compares pointer addresses associated with the same block of data and provides a delineation control between adjacent data blocks written to the FIFO buffer 144 .
- the above described method and circuit for controlling a FIFO buffer 144 has many advantages over the prior art, such as, enabling more than one data block to occupy the FIFO buffer 144 simultaneously without data overlap between adjacently successive data blocks. Additionally, the method and circuit for controlling a FIFO buffer 144 does not require the write-pointer address register 148 or the read-pointer address register 146 to be reset after each data block transfer is completed. Rather, the read-pointer address register 146 and the write-pointer address register 148 can continue to increment until reaching a maximum counter count value, whereby the read/write address pointers wrap around to zero and counting is continued without latency.
Abstract
Description
- The invention relates generally to controlling a first-in-first-out (FIFO) buffer and more precisely to a FIFO controller for buffering a plurality of disparate data blocks or entities flowing between asynchronous data processes in a manner preventing overlap or corruption of data.
- FIFO structures are well known for buffering data passing between two data handling processes which operate asynchronously with respect to each other. A FIFO memory is a structure wherein data words, objects, blocks, or entities are taken out of the structure in the order of receipt. A sender places data objects into the FIFO structure, while a receiver collects the objects from the structure.
- Referring to FIG. 1, generally, a first-in-first-out (FIFO)
buffer 44 is used to provide temporary storage for data that is being transferred between two or more asynchronous data processing systems. These data processing systems typically have specialized clocking circuits, e.g. Clk-1 and Clk-2 operating at predetermined frequencies (phases) which are asynchronous with respect to each other. TheFIFO buffer 44 shown in the FIG. 1 example is, for example, a dual port memory array having a preferred row dimension of thirty two bits wide with thirty two rows in height or depth (i.e. a 32 by 32 memory array). - One typical data processing system is a disk-type magnetic recording system or
disk drive 30. Thedisk drive 30 uses a magnetic transducer element, orhead 31 to record information onto (i.e., write) and to retrieve information from (i.e., read) a magnetic medium formed on a surface of a rotatingdisk 32. Eachstorage disk 32 comprises an annular substrate onto which is deposited a magnetic recording medium. Although FIG. 1 shows asingle head 31 anddisk 32, in practice multiple storage surfaces and heads are frequently employed. In the FIG. 1 example the storage surface ofdisk 32 is divided into thousands of concentric, annular bands, or “tracks” 33 each having a predetermined radial extent. Eachhead 31 is supported in close proximity to an associateddisk 32 surface by a head positioning assembly, oractuator 34, that supports thehead 31 near thedisk 32 surface and moves it from one radial position to another, thereby permitting use of asingle head 31 for reading and writing along each of the multiple concentric tracks 33. Thepositioner assembly 34 for eachhead 31 or group ofheads 31 includes an actuator arm 35 and anactuator motor 36. Theactuator motor 36 moves the actuator arm 35 to change the position of thehead 31 with relation to the tracks 33 on thedisk 32. Adisk drive 30 having a plurality ofdisks 32 stacked on a common spindle will typically employ asingle actuator motor 36 to move a corresponding number of ganged-together actuator arms 35 andheads 31 in unison. A disk controller (not shown) controls operations of the disk spindle and theactuator motor 36. - The
disk drive 30, further includes adisk channel 37, such as a partial-response maximum-likelihood (PRML) synchronous sampling data detection channel, for encoding, detecting, decoding and controlling flow of data read from or written to thedisk 32 at a data rate in accordance with a disk clock CLK-1. The disk clock CLK-1 may remain fixed, or it may be varied as a function of radius of a particular data track 33 (since relative rotational velocity varies with track radius). Thedisk drive 30 typically transmits data back and forth between itself and a random-access-memory (RAM)buffer 25. TheRAM buffer 25 includes one or a plurality of electronic data storage integrated circuit memory chips for storing data therein. TheRAM buffer 25 is coupled with aRAM Port 26. TheRAM Port 26 is coupled to thedisk drive 30 via a disk-direct-memory-access (DDMA) 40, which is defined within thedisk drive 30. TheRAM Port 26 circuit is further connected to several other clients including: aRAM buffer manager 24; astandard bus interface 28, such as a Small-Computer-System Interface (SCSI), which connects to ahost computer 22 and also may be connected as well as to several other clients (n). TheRAM Port 26 circuit provides access arbitration among the clients seeking access to theRAM buffer 25, including theDDMA 40,host interface 28 and other clients (n) such as a disk drive controller. - The DDMA40 generally includes a RAM
Port Interface circuit 42, aFIFO buffer 44, and a disk interface circuit 45. TheRAM Port interface 42 communicates data as well as control information to theRAM buffer 25 via theRAM Port 26 andRAM buffer manager 24. TheFIFO buffer 44 is connected to theRAM Port Interface 42 circuit and acts as a temporary storage space for data that is in the process of being transferred to/from thedisk drive 30 and theRAM 25. The disk interface circuit 45 is connected to thedisk channel 37, via abus 27, as well as to theFIFO buffer 44. The disk interface 45 operates to communicate data to/from thedisk channel 37 and storage surface ofdisk 32. - The DDMA40 further includes a disk read/
write process control 41 and a RAM read/write process control 43. The disk read/writeprocess control circuit 41 is connected to the disk interface 45 and is synchronous with CLK-1 of thedisk channel 37. The RAM read/write process control circuit 43 is connected to theRAM Port interface 42 and is synchronous with a second clock CLK-2 used to write data to and from theRAM buffer 25. The disk and RAM read/writeprocess control circuits 41, 43 enable theRAM Port interface 42 and the Disk Drive interface 45 to either read data from or write data to theFIFO buffer 44 depending on the direction of data flow between theRAM buffer 25 and thedisk drive 30. - In a typical data transfer from the
disk drive 30 to theRAM buffer 25, a plurality of data blocks written in tracks 33 are sensed as magnetic flux transitions by thehead 31 and recovered as binary data by thedisk channel 37. The recovered data is transmitted to theFIFO buffer 44 via the disk interface 45. Prior to writing data to theFIFO buffer 44, a write-pointer register 48 is preset to point to the first row of the FIFO 44. Next, the disk read/writeprocess control circuit 41 is enabled for writing data to theFIFO buffer 44, and the RAM read/write process control circuit 43 is enabled for reading data from theFIFO buffer 44. Thereafter, the disk interface 45 begins writing 32-bit data segments to theFIFO buffer 44 synchronously with clock CLK-1. Additionally, the write-pointer address register 48 is incremented as each data segment is written into theFIFO buffer 44. Once theFIFO buffer 44 accumulates a predetermined number of data segments, a control signal is issued to theRAM Port Interface 42 indicating that theFIFO buffer 44 has data which needs to be read out. - At this point, the
RAM Port Interface 42 arbitrates with theRAM Port 26 for direct access to theRAM buffer 25 via theRAM buffer manager 24. When access is granted to theRAM buffer 25, theRAM Port Interface 26 reads data out of theFIFO buffer 44 and transmits the data to theRAM buffer 25 viaRAM buffer manager 24. Prior to theRAM Port Interface 42 reading data out of theFIFO buffer 44, a read-pointer address register 46 is preset to point to the beginning of data written into theFIFO buffer 44, i.e. the first row in this example. Thereafter, theRAM Port Interface 42 begins reading data segments from theFIFO buffer 44 in synchronism with clock CLK-2. The read-pointer address register 46 is incremented as each data segment is read from theFIFO buffer 44. Once the read-pointer address register 45 is incremented to point to a next buffer row, the present buffer row is free to be written with a next data segment of the data block being transferred. - Since the
FIFO buffer 44 holds only 32 4-byte segments (128 bytes or 1024 bits) thebuffer 44 typically cannot store an entire disk data block (e.g. 512 bytes or 4096 bits) at once. Accordingly, theFIFO buffer 44 is cycled (rolls over) several times during the transfer of a single block of data. After the data block is both written to and read from theFIFO buffer 44, the write-pointer address register 48 and the read-pointer address register 46 are reset and the above process is repeated for transferring a next data block from thedisk drive 30 to thehost 22. - In a typical data transfer from the
RAM buffer 25 to thedisk drive 30, a plurality of data blocks are transferred from thehost 22 to theRAM buffer 25 and then the blocks of data are subsequently transferred to theFIFO buffer 44 via theRAM Port Interface 42. Similar to that of transferring data from thedrive 30 to theRAM buffer 25, theFIFO buffer 44 presets the write-pointer register 48 to point to the beginning of the FIFO 44. Next, the RAM read/write process control circuit 43 is enabled for writing data to theFIFO buffer 44 while the disk read/writeprocess control circuit 41 is enabled for reading data from theFIFO buffer 44. Thereafter, theRAM Port Interface 42 begins preemptively writing data segments to theFIFO buffer 44 in synchronism with clock CLK-2. This is to ensure that the data is available for transfer to thedisk 32 when thehead 31 arrives at the correct location over thedisk 32. Additionally, the write-pointer address register 48 is incremented as each data segment is written into theFIFO buffer 44. Once theFIFO buffer 44 is filled, theRAM buffer 25 write process pauses and waits for space to become available in theFIFO buffer 44. Space becomes available in theFIFO buffer 44 as a result of the disk interface 45 read process, which reads data from theFIFO buffer 44 and sends the data to thedisk channel 37 for subsequent storage on thedisk 32. When sufficient space becomes available in theFIFO buffer 44, the process of transferring data from theRAM buffer 25 to theFIFO buffer 44 is resumed. This process is repeated until all data is transferred from theRAM buffer 25 to thedisk 32. - Prior to the disk drive interface45 reading data out of the
FIFO 44, the read-pointer address register 46 is preset to the beginning of theFIFO buffer 44. Thereafter, the disk drive interface 45 begins reading data segments from theFIFO buffer 44 synchronously with clock CLK-1. Additionally, the read-pointer address register 46 is incremented synchronously with each data segment read from theFIFO buffer 44. After the data block is both written to and read from theFIFO buffer 44, the write-pointer address register 48 and the read-pointer address 44 are reset and the above process is repeated for transferring the next data block from thehost 22 to thedisk drive 30. - Generally, the asynchronous operating characteristics of both the
RAM 25 and thedisk drive 30 with respect to one another cause problems in transferring data therebetween. One such problem occurs as a result of having to reset the write-pointer address register 48 and the read-pointer address register 46 after each data block transfer. Another problem occurs due to theRAM Port interface 42 having to arbitrate to gain access to theRAM buffer 25. Collectively, this arbitration time and the time required to reset the write-pointer register 48 and read-pointer register 46 can cause theFIFO buffer 44 to reject reception of a second data block until the former data block is completely read out of theFIFO buffer 44 and the pointers 46 and 48 have been reset. In transferring data from thedisk drive 30 to theRAM 25, this delay may result in forcing thedisk drive 30 to wait until theFIFO buffer 44 is again receptive to receiving data. Such buffer access waiting period may cause thedisk drive 30 incur another complete revolution of thestorage disk 32 to arrive at the same data in a track 33 and then read the data. The wait associated with a second disk revolution to read a track is known in the art as “rotational latency”, and excessive rotational latencies degrade disk drive system performance. - Conventional methods of solving this problem have been to simultaneously write to and read from the
FIFO buffer 44 without resetting the pointer registers 46 and 48. However, this solution introduces additional problems when a first data block being written to theFIFO buffer 44 is aborted somewhere between the beginning and ending of the data block. When the transfer of the first data block is aborted and the writing of a second data block immediately follows, the read-pointer address associated with the tail end of the aborted first data block can overlap with the write-pointer address associated with the beginning of the second data block being written to theFIFO buffer 44. This data overlap condition causes data from the second data block to be erroneously transferred as data of the first data block. - Another problem occurs when writing a sequence of contiguous data blocks from the
disk drive 30 to theFIFO buffer 44, via the disk interface 45, if there is insufficient space in thebuffer 44 to accommodate the data. This full condition of theFIFO buffer 44 occurs when theRAM Port Interface 42 cannot read data out of theFIFO 44 fast enough to make room for data being written into theFIFO 44. If theFIFO buffer 44 does not have space for data being written thereto from thedisk drive 30, an overflow condition occurs, which can result in a data transfer error. - A similar problem occurs when the
RAM Port Interface 42 does not transfer a steady stream of data to theFIFO 44 such that theFIFO 44 runs out of data or underflows. This FIFO underflow condition again results in reduced performance of thedisk drive 30, since thedisk drive 30 has to incur another rotational latency delay in order to bring the data block track destination beneath thetransducer head 31 for data writing during a second pass over the track 33. - Thus, a hitherto unsolved need has remained for a FIFO buffer control method and circuit that facilitates simultaneous multiple block writing to and reading from a FIFO buffer without data overlap. Additionally, a FIFO buffer control circuit and method is needed that does not require a write-pointer register and a read-pointer register to be reset after each data block transfer so that a plurality of contiguous data blocks, which are much larger than the FIFO storage capacity, can be seamlessly transferred back and forth between the disk drive and the host computer.
- An object of the present invention is to provide a FIFO buffer control that enables an end of a first data block to occupy a FIFO buffer register simultaneously with a beginning of a second data block without data overlap, in a manner overcoming limitations and drawbacks of the prior art.
- Another object of the present invention is to provide a FIFO buffer control that does not require the write-pointer or the read-pointer to be reset after each data block transfer.
- In accordance with principles of the present invention, a FIFO buffer control circuit enables a FIFO buffer to simultaneously contain portions of plural adjacent data blocks without data overlap. During a process of transferring data blocks between a sending unit and a receiving unit asynchronously clocked with the sending unit, the FIFO control captures and saves a FIFO buffer write-pointer address that is associated with beginning of a first data block to be written to the FIFO buffer. After writing the first data block to the FIFO buffer, the FIFO control captures and saves a buffer write-pointer address associated with the end of the first data block. Next, the FIFO control captures and saves FIFO buffer write-pointer addresses associated with a beginning of a second data block to be written to the FIFO buffer. After writing the second block to the FIFO buffer, the FIFO buffer control captures and saves a buffer write-pointer address associated with the end of the second data block. This process is repeated for capturing and saving the write-pointer addresses associated with each successive data block written into the FIFO buffer.
- Meanwhile, a read pointer address register of the FIFO buffer is loaded with the previously saved write pointer address that corresponds to the beginning of the first data block to be read. Thereafter, data is read from the FIFO buffer until the end of the first data block is detected. At this instant, the read pointer address register is loaded with the saved write-pointer address associated with the beginning of the second data block written to the FIFO buffer. Then, the second data block is read from the FIFO buffer until the end of the second data block is detected. This read process is repeated until all data blocks written into the FIFO buffer by the sending unit have been read out to the receiving unit.
- Since the write-pointer address associated with the beginning and ending of each data block written to the FIFO buffer is captured and saved, successive data blocks can be sequentially written thereto without resetting the write-pointer address register. Additionally, since the read-pointer address register is loaded with the write-pointer address associated with the beginning of each data block, the read-pointer address does not have to be reset after each data block is read from the FIFO buffer. Most importantly, the FIFO control circuit enables segments of a plurality of data blocks to occupy the FIFO buffer register simultaneously without data overlap between such blocks, because the beginning and ending pointers of each data block have been previously captured and saved.
- These and other objects, advantages, aspects and features of the present invention will be more fully understood and appreciated upon consideration of the following detailed description of a preferred embodiment, presented in conjunction with the accompanying drawings.
- In the drawings:
- FIG. 1 is a high level block diagram of a conventional first-in-first-out (FIFO) buffer circuit within a disk drive system for buffering data between a disk process synchronized to a disk read-write clock and a RAM buffer process synchronized to a second clock. Connection of the disk drive system via an interface to a host computer is also shown.
- FIG. 2 is similar to FIG. 1 and shows a high level block diagram of a first-in-first-out (FIFO) buffer control circuit in accordance with principles of the present invention for buffering multi-block data flow between a disk process synchronized to a disk read-write clock and a RAM buffer process synchronized to a second clock, without data overlap.
- FIG. 3 exemplifies a circuit implementation of the FIG. 2 FIFO buffer control circuit in accordance with principles of the present invention.
- FIG. 4 is a flow chart illustrating a method of writing multi-block data to the FIG. 2 FIFO buffer according to principles of the present invention.
- FIG. 5 is a flow chart illustrating a method of reading multi-block data from the FIG. 2 FIFO buffer according to principles of the present invention.
- Referring to FIG. 2, a
disk drive 130 including aDDMA 140 having one preferred embodiment of the present invention is set forth. TheDDMA 140 controls the bi-directional transfer of data between first and second asynchronous data processing systems, such as a random-access-memory (RAM)buffer 25 and adisk channel 37 of adisk drive system 30. In FIG. 2, elements which remain essentially unchanged from the previously discussed FIG. 1 example are assigned the same reference numeral. TheRAM buffer 25 includes one or a plurality of electronic data storage dynamic random access (DRAM) or static random access (SRAM) integrated circuit memory chips for storing data that may subsequently be accessed by thehost computer processor 22 via theinterface 28. TheRAM Port 26 circuit provides arbitration among the number of clients, including theRAM buffer manager 24, for direct communication with theRAM buffer 25 and/or theDDMA 140. TheRAM Port 26 circuit is coupled to the disk drive via disk-direct-memory-access (DDMA) 140 circuit which embodies principles of the present invention. - The
FIFO buffer 144 includes a data storage space or FIFO buffer space 149. In one preferred implementation, the dimensions of the FIFO buffer 149 is a row width of thirty-two bits by thirty-two rows (32×32 memory array). TheFIFO buffer 144 further includes a read-pointer address register 146 that stores a read-pointer address associated with a location in theFIFO buffer 144 where data is read. Additionally, theFIFO buffer 144 includes a write-pointer address register 148 that stores a write-pointer address associated with a location in theFIFO buffer 144 where data is written. So far, the description of theFIFO buffer 144 is essentially the same as theFIFO buffer 44 of FIG. 1. - A bank of FIFO write
address registers 150 is coupled to aFIFO buffer 144 as well as to a RAM Port Interface circuit 142 and a diskdrive interface circuit 145. The bank of FIFO address registers 150 includes four FIFO buffer write address registers, a first write-start register 152 a, a first write-end address register 152 b, a second write-start register 154 a, and a second write-end register 154 b. The registers 152 a, 152 b, 154 a and 154 b have parallel inputs connected to receive addresses from a write-pointer address register 148. Write-start address registers 152 a and 154 a have outputs connected in parallel leading to a read-pointer address register. Write-end address registers 152 b and 154 b have outputs connected in parallel leading to a FIFOstatus detection circuit 147. - The first write-start address register152 a is controlled to capture and save the beginning write-pointer address associated with the beginning of a first data block to be written to the
FIFO buffer 44. The first write-end address register 152 b is controlled to pass the active write pointer through the FIFO status logic, or capture and save an ending write-pointer address associated with the ending of the first data block written to theFIFO buffer 144. - The second write-start address register154 a is controlled to capture and save the beginning write-pointer address associated with beginning of a second data block to be written to the
FIFO buffer 144. Furthermore, the second write-end address register 154 b is also controlled to pass the active write pointer through the FIFO status logic, or capture and save an ending write-pointer address associated with the ending of the second data block transferred to theFIFO buffer 44. In this manner, both the first pair of FIFO address registers 152 a, 152 b, and the second pair of FIFO address registers 154 a and 154 b, capture and save the beginning and ending write-pointer addresses associated with first and second data blocks written to theFIFO buffer 144. The captured address pointers enable theFIFO buffer 144 to store segments of multiple data blocks without overlapping data between the data blocks and without data corruption in the event of an aborted data transfer operation. - When a data block is being transferred from the
RAM buffer 25 to thedisk channel 37, control of the address registers is by the RAM Port interface 142 under the direction of a RAM read/write process control 143. When a data block is being transferred from thedisk channel 37 to theRAM buffer 25, control of the address registers is by theDisk interface 145 under the direction of a disk read/write process control 141. Suitable multiplexing enables dual access to the address registers 152 a, 152 b, 154 a, and 154 b via both the RAM port interface 142 and thedisk interface 145. - The
FIFO buffer 144 further includes a FIFOstatus detection circuit 147. The FIFOstatus detection circuit 147 includes a comparator for comparing an address held in the read-pointer address register 146 with an address held in the write-pointer address register 148 in order to determine the number of data segments currently held in theFIFO buffer 144. More precisely, the FIFOstatus detection circuit 147 compares the address held in the read-pointer address register 146 with an incrementing write-pointer address value held in either register 152 b or 154 b. This incrementing write-pointer address held in either register 152 b or 154 b, which is generated by the write-pointer address register 148, increments until an end of a data segment is reached, whereupon this address or write-end address held in either register 152 b or 154 b is captured and stored therein. At this point, the FIFOstatus detection circuit 147 compares the address held in the read-pointer address register 146 with the captured and stored write-end address held in either register 152 b or 154 b. In comparing the write-pointer address with the read-pointer address in such a manner, information such as whether theFIFO buffer 144 is full, half-full, or empty is readily determined. - FIG. 3 exemplifies one preferred circuit implementation of the FIG. 2
FIFO DDMA circuit 140. In this embodiment of the present invention, the bank of address registers 150 includes four multiplexer flip-flops (MUX-flops) 52 a, 52 b, 54 a and 54 b which implement FIFO buffer address registers 152 a, 152 b, 154 a and 154 b respectively. Each of the MUX-flops receives an enable control signal over lines 55-58 from a write-state control circuit 60. The write-state control circuit 60 includes combinatorial logic coupled with the RAM Port interface 142 and theDisk Drive interface 145. - Depending on which interface is enabled for writing data to the
FIFO buffer 144, the controllinginterface 142 or 145 provides several control signals to the write-state control circuit 60 including: save write-start address over line 63, and a write block sequence number overline 62; and save write-end address over line 61. For example, a first block A is assigned a write sequence number of 0, while a second block B is assigned a write sequence number of 1. A third block C is assigned a write sequence number of 0, etc. Activation of the save write-start address over line 63, and its associated write sequence number control signals overline 62 causes one of the write-start flip-flops 52 a or 54 a to capture and save the start-write pointer address value for the identified block at a particular instant in time. This captured and saved address value is associated with the beginning of a data block that is imminently to be written to theFIFO buffer 44. - Similarly, activation of the save end address over line61, and its associated write sequence number over
line 62 again causes one of the write-end flip-flops 52 b or 54 b to capture and save the write-end pointer address value for the identified block at a particular instant in time. However, this captured and saved address pointer is associated with an ending of the identified data block written to theFIFO buffer 44. The write sequence control signals received online 62 enable the write-state control circuit 60 to capture and save the beginning and ending write-pointer addresses associated with the same block of data. - The bank of
registers 150 receives the FIFO buffer write-pointer address values from the write-pointeraddress register circuit 148 over a line 70. The write-pointeraddress register circuitry 148 includes a single MUX-flop 48 a connected to receive an increment-write-address control signal over aline 72 from either the disk drive interface 142 or theRAM Port interface 145, depending on which one is enabled for writing data to theFIFO buffer 144. The increment-write-address control signal online 72 enables the write-pointer address register 148 to increment the write pointer address value stored therein as each data segment 12 is written to a row of theFIFO buffer 44. An adder 48 b adds a binary one to the output of the write MUX-flop 48 a on a write-address bus 49 and feeds that value back to a data input of the MUX-flop 48 a. The write-address bus 49 is connected to address the FIFO buffer array 149 as well as to provide starting and ending block addresses to theaddress register bank 150. - The read-
pointer address circuit 146 is also coupled to the outputs of the write-start address registers 52 a and 54 a. The read-pointer address circuit 146 receives address values from the address register 52 a over a line 74 and receives address values from the address register 54 a over a line 75. The values received over lines 74, 75 are indicative of the beginning row address of a data block to be read fromFIFO buffer 144. Line 74 provides a value indicative of the write-start address associated with a first data block written to theFIFO buffer 144. Similarly, line 75 provides another value indicative of the write-start address value associated with a second data block written to theFIFO buffer 144. - The read-
pointer address circuit 146 includes a single MUX-flop 46 a, which accepts the saved write-start address values received from a MUX-flop 52 a or 54 a via line 74 or 75. Additionally, the read-pointer address circuit 146 receives control signals from a read-state control circuit 80 over lines 81-83. The read-state control circuit 80 includes combinatorial logic that receives a control signal from either thedisk drive interface 145 or the RAM Port interface 142, depending on data flow direction and which of the two are enabled for reading data from theFIFO buffer 144. The control signals received by the read-state control circuit 80 include: load read address over line 84, a block read sequence number, i.e., zero or one, over aline 85, and an increment read address over a line 83. The load read address control signal on line 84 enables the write-start address of a data block held in a write-start MUX-flop 52 a or 54 a to be loaded into the read-pointeraddress register circuit 146. The read sequence control signal online 85 enables the read pointeraddress register circuit 146 to select either MUX-flop 52 a or 54 a for reading either the first or the second data block written to theFIFO buffer 144. The increment read address pointer control signal on line 83 enables the read-pointeraddress register circuit 146 to increment the MUX-flop 46 a as each data segment (row) is read from theFIFO buffer 44. An adder 46 b adds a binary one to an output of the read address MUX-flop 46 a on aread address bus 47 and feeds the sum to a data input of the MUX-flop 46 a. - The FIFO
buffer status circuit 147 includes a first comparator 90, asecond comparator 91, and a third comparator 92. The first comparator circuit 90 compares the next FIFO buffer write address value input to the write-pointer address circuit 148 via apath 93 with a FIFO buffer read address value input to the read-pointer address circuit 146 via apath 94. The first comparator circuit 90 subtracts the read address value from the write address value and determines from the difference whether theFIFO buffer 144 is either less than or equal to half full or whether theFIFO buffer 144 is full. - The
second comparator circuit 91 receives a registered current write or write-end address value from one of the write-end address registers 52 b or 54 b via a path 95 from amultiplexer circuit 96. This address value may be actively following the current write address pointer 48 a, or the captured and stored write-end address if all data has been written to theFIFO buffer 144. Themultiplexer circuit 96 selects between the saved write-end address of a first data block or the saved write-end address of a second data block previously captured and saved in thebank 150. Themultiplexer circuit 96 selects between these saved addresses based on the read sequence control signal transmitted to the multiplexer overline 85. Additionally, thesecond comparator circuit 91 receives the current FIFO read address value from the read-pointer address circuit 146 over FIFO readaddress path 47. Thereafter, thesecond comparator circuit 91 subtracts the current FIFO read address value from the selected block write-end address value received from one of the write-end registers 52 b and 54 b viamultiplexer 96 in order to determine if theFIFO buffer 144 is greater than half full. - Finally, the third comparator circuit92 subtracts a next row read address taken from the input of the read address MUX-
flop 46 a from the registered write-end address value from one of the write-end registers 52 b, or 54 b as selected bymultiplexer 96 and provided over path 95 to determine if theFIFO buffer 44 is empty. - It is important to note that the saved write address on line95 is equal to, and changes with, the write-pointer address 48 a on
line 49 until a save is requested via line 55 or 56. - Optionally, a write-
address validation circuit 86 for each block A and B is coupled with the read-pointer address circuit 146 and read-state control circuit 80 for verifying that a data block associated with the saved write-start address previously captured and saved in the bank ofregisters 150 was actually written. The write-address validation circuit 86 includes a first MUX-flop 87 coupled with a second MUX-flop 88 whereby the first MUX-flop 87 is enabled if either sequence of the read-pointeraddress register circuit 146 is loaded with a read address value. Thereafter, the second MUX-flop 88 puts out a valid signal online 89 when the write-pointer address register 148 is incremented vialine 72, thereby writing data to thefirst FIFO buffer 144 location. - Referring to FIGS. 2, 4, and5, the RAM Port Interface circuit 142 includes a state machine which implements two control processes illustrated in the flow outlines in FIGS. 4 and 5. Similarly, the
disk interface circuit 145 also includes a state machine which implements the two control processes illustrated in the flow outlines in FIGS. 4 and 5. When the RAM Port Interface circuit 142 is executing the FIG. 4 flow diagram, thedisk interface circuit 145 will be executing the FIG. 5 flow diagram. This enables the RAM port interface 142 to write data to theFIFO buffer 144 and enables thedisk interface circuit 145 to read the data from theFIFO buffer 144. Alternatively, when the RAM Port Interface circuit 142 is executing the FIG. 5 flow diagram, thedisk interface circuit 145 will be executing the FIG. 4 flow diagram. This enables the RAM port interface 142 to read data from theFIFO buffer 144 and enables thedisk interface circuit 145 to write data to theFIFO buffer 144. - In transferring data from the RAM asynchronous system or
RAM buffer 25 to the second asynchronous system ordisk drive system 30, the RAM Port Interface circuit 142 receives a command to start writing data blocks to theFIFO buffer 144 atstep 400. Prior to writing the first data block to theFIFO buffer 144, the disk read process control 141 circuit is enabled to permit the disk drive interface to read data from theFIFO buffer 144 and the RAM readprocess control 143 circuit is enabled to permit the RAM Port interface 142 to write data to theFIFO buffer 144 atstep 405. - Next, the RAM Port Interface142 sends a control signal to the
FIFO buffer 144 to determine if there is space in the FIFO to accommodate data to be written therein atstep 410. In the event theFIFO buffer 144 is full, the data transfer process pauses atstep 410 until there is space available for data in theFIFO buffer 144. Once it is determined that there is space available in theFIFO buffer 144 for data atstep 410, then a start-write address associated with the beginning of the data block to be written to theFIFO buffer 144 is captured and saved atstep 420. This start-write address is saved in the first start-write register 52 a defined within the first pair 152 a, 152 b of registers incorporated within the bank ofregisters 150. - Thereafter, data segments (FIFO rows) of the first data block are sequentially written to the
FIFO buffer 144 atstep 430. At the same time, the RAM Port Interface 142 sends a control signal to the write-pointer address register 148 to increment the write-pointer address stored therein atstep 440 as each data segment is written to theFIFO buffer 144. The write-pointer address is incremented atstep 440 until the first data block has been successfully written to theFIFO buffer 144. However, the process of writing data segments to theFIFO buffer 144 may pause, if it is determined instep 455 that theFIFO buffer 144 does not have space for data to be written. Upon detecting the end of this first data block atstep 450, the write pointer address is captured and saved in the first write-end register 152 b at astep 460. - After writing the first data block to the
FIFO buffer 144, it is determined at astep 470 whether another data block needs to be written to theFIFO buffer 144, and if not, the process ends atstep 480. On the other hand, if it is determined that another data block needs to be written to theFIFO buffer 144, the write-pointer address register 148 is permitted to continue incrementing concomitant with data segments defined within the next successive data block that is written to theFIFO buffer 144. However, the beginning and ending addresses of this next data block are captured and saved in the second pair 154 a, 154 b, of FIFO buffer address registers. After the next successive data block has been written to theFIFO buffer 144, a third successive data block is written thereto. However, the beginning and ending write-pointer addresses of this data block are captured and stored in the first pair 152 a, 152 b, of FIFO buffer address registers defined within thebank 150. In accordance with the FIG. 4 flow process, the first pair 152 a, 152 b, and the second pair 154 a, 154 b, of FIFO buffer address registers alternate in capturing and storing the beginning and ending write-pointer addresses for the plurality of data blocks written to theFIFO buffer 44. Additionally, depending upon the size of theFIFO buffer 144, thebank 150 of FIFO address registers may be expanded by adding additional pairs of address registers (not shown) connected in parallel with registers 152 a, 152 b, and 154 a, 154 b. In such an arrangement, segments of more than two data blocks can simultaneously occupy theFIFO buffer 144 at the same time. - Referring to FIGS. 2 and 5, the
disk drive interface 145 receives a command to begin reading data from theFIFO buffer 44 at astep 500. Thereafter, the diskdrive interface circuit 145 determines if theFIFO buffer 144 has accumulated a predetermined number of data segments atstep 510 defined within the first data block. When it is determined instep 510 that theFIFO buffer 144 contains data to be read, thedisk drive interface 145 loads the read-pointer address register 146 with theFIFO buffer 144 write-pointer address previously saved in the first start-write register 152 a atstep 520. When data is available for reading from theFIFO buffer 144, thedisk drive interface 145 reads data from theFIFO buffer 144 atstep 530. At the same time, thedisk drive interface 145 sends a control signal to the read-pointer address register 146 to increment the read-pointer address stored therein concomitant with each data segment read from theFIFO buffer 144 atstep 540. The read-pointer address register 146 is incremented until the first data block has been successfully read from theFIFO buffer 144 atstep 530. Upon detecting the end of this first data block atstep 560, thedisk drive interface 145 determines if another data block needs to be read from theFIFO buffer 144 atstep 570, and if so, the read-pointer address register 146 is loaded with theFIFO buffer 144 write-pointer address previously saved in the second start-write register 154 a. Thereafter the above process is repeated atstep 580 for reading the second data block from theFIFO buffer 144. If thedisk drive interface 145 determines that there are no additional data blocks waiting in theFIFO buffer 144 to be read, then the read process ends atstep 590. - During the process of writing and reading the first data block to/from the
FIFO buffer 144, the FIFOstatus detection circuit 147 compares the active read-pointer address register 146 (incrementing) with the active write-pointer address register 148 (incrementing) to determine when the first data block has been completely written to and read from theFIFO buffer 144. After the first data block has been transferred, however, the FIFObuffer status register 147 switches from comparing the active read-pointer address register 146 with the active write-pointer address register 148 to comparing the active read-pointer address register 146 with a static write-pointer address saved in the first write-end register 152 b. In this manner the FIFOstatus detection circuit 147 compares pointer addresses associated with the same block of data and provides a delineation control between adjacent data blocks written to theFIFO buffer 144. - In the event of an interruption of reading a first data block (block abort process), the read-
pointer address register 146 is reloaded with the previously saved write-start address of the next successive data block that is being written to theFIFO buffer 144. In this manner, the read-pointer address register 146 is jumped to the beginning of the next successive data block for continuing the process of reading data from theFIFO buffer 44 and thereby skips data segments of the now-aborted first block which may still be present within theFIFO buffer 144. - The above described method and circuit for controlling a
FIFO buffer 144 has many advantages over the prior art, such as, enabling more than one data block to occupy theFIFO buffer 144 simultaneously without data overlap between adjacently successive data blocks. Additionally, the method and circuit for controlling aFIFO buffer 144 does not require the write-pointer address register 148 or the read-pointer address register 146 to be reset after each data block transfer is completed. Rather, the read-pointer address register 146 and the write-pointer address register 148 can continue to increment until reaching a maximum counter count value, whereby the read/write address pointers wrap around to zero and counting is continued without latency. - Having thus described an embodiment of the invention, it will now be appreciated that the objects of the invention have been fully achieved, and it will be understood by those skilled in the art that many changes in construction and widely differing embodiments and applications of the invention will suggest themselves without departing from the spirit and scope of the invention. The disclosure and the description herein are purely illustrative and are not intended to be in any sense limiting.
Claims (23)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/232,764 US6381659B2 (en) | 1999-01-19 | 1999-01-19 | Method and circuit for controlling a first-in-first-out (FIFO) buffer using a bank of FIFO address registers capturing and saving beginning and ending write-pointer addresses |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/232,764 US6381659B2 (en) | 1999-01-19 | 1999-01-19 | Method and circuit for controlling a first-in-first-out (FIFO) buffer using a bank of FIFO address registers capturing and saving beginning and ending write-pointer addresses |
Publications (2)
Publication Number | Publication Date |
---|---|
US20010054121A1 true US20010054121A1 (en) | 2001-12-20 |
US6381659B2 US6381659B2 (en) | 2002-04-30 |
Family
ID=22874481
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/232,764 Expired - Lifetime US6381659B2 (en) | 1999-01-19 | 1999-01-19 | Method and circuit for controlling a first-in-first-out (FIFO) buffer using a bank of FIFO address registers capturing and saving beginning and ending write-pointer addresses |
Country Status (1)
Country | Link |
---|---|
US (1) | US6381659B2 (en) |
Cited By (35)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6618390B1 (en) * | 1999-05-21 | 2003-09-09 | Advanced Micro Devices, Inc. | Method and apparatus for maintaining randomly accessible free buffer information for a network switch |
GB2386485A (en) * | 2002-03-12 | 2003-09-17 | Toshiba Res Europ Ltd | Modulo addressing apparatus and methods |
US20030206475A1 (en) * | 2001-08-23 | 2003-11-06 | Jiann-Jeng Duh | FIFO memory devices that support all combinations of DDR and SDR read and write modes |
US20040093476A1 (en) * | 2002-11-11 | 2004-05-13 | Thompson Ryan C. | System for preventing memory usage conflicts when generating and merging computer architecture test cases |
US20040193785A1 (en) * | 2003-03-11 | 2004-09-30 | Zayas Fernando A. | System for improving the performance of read/write testing in a hard drive |
US20040193812A1 (en) * | 2003-03-11 | 2004-09-30 | Zayas Fernando A. | Method for improving the performance of read/write testing in a hard drive |
EP1482402A2 (en) * | 2003-05-27 | 2004-12-01 | Micronas GmbH | Fill level capture in a buffer |
US20040250039A1 (en) * | 2003-06-05 | 2004-12-09 | Hanks D. Mitchel | System and method for using swappable storage for high data content multi-source data storage |
US20050018514A1 (en) * | 2001-08-23 | 2005-01-27 | Knaack Roland T. | Integrated DDR/SDR flow control managers that support multiple queues and mux, demux and broadcast operating modes |
US20050286534A1 (en) * | 2004-06-28 | 2005-12-29 | Inching Chen | Systems and methods for packet multicasting with a multi-read buffer |
US7120075B1 (en) | 2003-08-18 | 2006-10-10 | Integrated Device Technology, Inc. | Multi-FIFO integrated circuit devices that support multi-queue operating modes with enhanced write path and read path queue switching |
US20070008985A1 (en) * | 2005-06-30 | 2007-01-11 | Sridhar Lakshmanamurthy | Method and apparatus to support efficient check-point and role-back operations for flow-controlled queues in network devices |
US7287102B1 (en) * | 2003-01-31 | 2007-10-23 | Marvell International Ltd. | System and method for concatenating data |
US20080049240A1 (en) * | 2006-08-28 | 2008-02-28 | Fujitsu Limited | Image data buffer apparatus and data transfer system for efficient data transfer |
US20090043927A1 (en) * | 2005-08-15 | 2009-02-12 | Sony Computer Entertainment Inc. | Buffer managing method and buffer managing apparatus |
US20100005250A1 (en) * | 2008-07-02 | 2010-01-07 | Cradle Technologies, Inc. | Size and retry programmable multi-synchronous fifo |
US8032725B1 (en) * | 2007-01-30 | 2011-10-04 | Netapp, Inc. | Backup data management on a fractionally reserved storage system |
US20120079202A1 (en) * | 2010-09-28 | 2012-03-29 | Kai Chirca | Multistream prefetch buffer |
CN102402486A (en) * | 2010-09-17 | 2012-04-04 | 腾讯科技(深圳)有限公司 | Method and device for reading data |
US20120151243A1 (en) * | 2010-12-13 | 2012-06-14 | Oracle International Corporation | Synchronization circuit that facilitates multiple parallel reads and writes |
US20120226852A1 (en) * | 2011-03-03 | 2012-09-06 | Mstar Semiconductor, Inc. | Control method and controller for dram |
US20130166772A1 (en) * | 2010-06-07 | 2013-06-27 | Adode Systems Incorporated | Buffering Media Content |
CN104636087A (en) * | 2015-02-09 | 2015-05-20 | 华为技术有限公司 | Method and device for controlling reading data |
US20150324136A1 (en) * | 2014-05-07 | 2015-11-12 | Lsi Corporation | Storage system having fifo storage and reserved storage |
US20160011989A1 (en) * | 2014-07-08 | 2016-01-14 | Fujitsu Limited | Access control apparatus and access control method |
US20160092285A1 (en) * | 2014-09-25 | 2016-03-31 | Intel Corporation | Method and Apparatus for Approximating Detection of Overlaps Between Memory Ranges |
US20160139880A1 (en) * | 2014-11-14 | 2016-05-19 | Cavium, Inc. | Bypass FIFO for Multiple Virtual Channels |
CN108073602A (en) * | 2016-11-11 | 2018-05-25 | 北京京东尚科信息技术有限公司 | The method and device of processing business data segment |
US20190163362A1 (en) * | 2017-11-29 | 2019-05-30 | International Business Machines Corporation | Host Controlled Data Chip Address Sequencing for a Distributed Memory Buffer System |
US10489069B2 (en) | 2017-11-29 | 2019-11-26 | International Business Machines Corporation | Address/command chip synchronized autonomous data chip address sequencer for a distributed buffer memory system |
US10534555B2 (en) | 2017-11-29 | 2020-01-14 | International Business Machines Corporation | Host synchronized autonomous data chip address sequencer for a distributed buffer memory system |
CN112667165A (en) * | 2020-12-29 | 2021-04-16 | 湖南博匠信息科技有限公司 | Data storage system and data storage method thereof |
CN113900974A (en) * | 2021-12-07 | 2022-01-07 | 睿思芯科(深圳)技术有限公司 | Storage device, data storage method and related equipment |
US11385862B2 (en) * | 2016-05-06 | 2022-07-12 | Texas Instmments Incorporated | Data flow control for multi-chip select |
US11587600B2 (en) | 2017-11-29 | 2023-02-21 | International Business Machines Corporation | Address/command chip controlled data chip address sequencing for a distributed memory buffer system |
Families Citing this family (53)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6789144B1 (en) * | 1999-05-27 | 2004-09-07 | Advanced Micro Devices, Inc. | Apparatus and method in a network interface device for determining data availability in a random access memory |
US6859399B1 (en) * | 2000-05-17 | 2005-02-22 | Marvell International, Ltd. | Memory architecture and system and multiport interface protocol |
US6826650B1 (en) * | 2000-08-22 | 2004-11-30 | Qlogic Corporation | Disk controller configured to perform out of order execution of write operations |
US6862653B1 (en) * | 2000-09-18 | 2005-03-01 | Intel Corporation | System and method for controlling data flow direction in a memory system |
US6658525B1 (en) * | 2000-09-28 | 2003-12-02 | International Business Machines Corporation | Concurrent access of an unsegmented buffer by writers and readers of the buffer |
US7233998B2 (en) * | 2001-03-22 | 2007-06-19 | Sony Computer Entertainment Inc. | Computer architecture and software cells for broadband networks |
US6952756B1 (en) * | 2001-05-08 | 2005-10-04 | Lewiz Communications | Method and apparatus for speculative loading of a memory |
US6801991B2 (en) * | 2001-12-21 | 2004-10-05 | Agere Systems Inc. | Method and apparatus for buffer partitioning without loss of data |
US6941393B2 (en) * | 2002-03-05 | 2005-09-06 | Agilent Technologies, Inc. | Pushback FIFO |
US7007114B1 (en) * | 2003-01-31 | 2006-02-28 | Qlogic Corporation | System and method for padding data blocks and/or removing padding from data blocks in storage controllers |
US7080188B2 (en) * | 2003-03-10 | 2006-07-18 | Marvell International Ltd. | Method and system for embedded disk controllers |
US7039771B1 (en) | 2003-03-10 | 2006-05-02 | Marvell International Ltd. | Method and system for supporting multiple external serial port devices using a serial port controller in embedded disk controllers |
US7492545B1 (en) | 2003-03-10 | 2009-02-17 | Marvell International Ltd. | Method and system for automatic time base adjustment for disk drive servo controllers |
US7099963B2 (en) * | 2003-03-10 | 2006-08-29 | Qlogic Corporation | Method and system for monitoring embedded disk controller components |
US7870346B2 (en) * | 2003-03-10 | 2011-01-11 | Marvell International Ltd. | Servo controller interface module for embedded disk controllers |
US8447900B1 (en) | 2003-06-12 | 2013-05-21 | Marvell International Ltd. | Adaptive multiple FIFO scheme |
US7237074B2 (en) * | 2003-06-13 | 2007-06-26 | Sandisk Corporation | Tracking cells for a memory system |
DE10342255A1 (en) * | 2003-09-11 | 2005-04-07 | Bts Media Solutions Gmbh | Circuit for controlling a memory |
US20050071828A1 (en) * | 2003-09-25 | 2005-03-31 | International Business Machines Corporation | System and method for compiling source code for multi-processor environments |
US7444632B2 (en) * | 2003-09-25 | 2008-10-28 | International Business Machines Corporation | Balancing computational load across a plurality of processors |
US7496917B2 (en) * | 2003-09-25 | 2009-02-24 | International Business Machines Corporation | Virtual devices using a pluarlity of processors |
US7389508B2 (en) * | 2003-09-25 | 2008-06-17 | International Business Machines Corporation | System and method for grouping processors and assigning shared memory space to a group in heterogeneous computer environment |
US7318218B2 (en) * | 2003-09-25 | 2008-01-08 | International Business Machines Corporation | System and method for processor thread for software debugging |
US7146529B2 (en) * | 2003-09-25 | 2006-12-05 | International Business Machines Corporation | System and method for processor thread acting as a system service processor |
US7478390B2 (en) * | 2003-09-25 | 2009-01-13 | International Business Machines Corporation | Task queue management of virtual devices using a plurality of processors |
US7523157B2 (en) * | 2003-09-25 | 2009-04-21 | International Business Machines Corporation | Managing a plurality of processors as devices |
US20050071578A1 (en) * | 2003-09-25 | 2005-03-31 | International Business Machines Corporation | System and method for manipulating data with a plurality of processors |
US7549145B2 (en) * | 2003-09-25 | 2009-06-16 | International Business Machines Corporation | Processor dedicated code handling in a multi-processor environment |
US7236998B2 (en) * | 2003-09-25 | 2007-06-26 | International Business Machines Corporation | System and method for solving a large system of dense linear equations |
US7415703B2 (en) * | 2003-09-25 | 2008-08-19 | International Business Machines Corporation | Loading software on a plurality of processors |
US7475257B2 (en) | 2003-09-25 | 2009-01-06 | International Business Machines Corporation | System and method for selecting and using a signal processor in a multiprocessor system to operate as a security for encryption/decryption of data |
US7516456B2 (en) | 2003-09-25 | 2009-04-07 | International Business Machines Corporation | Asymmetric heterogeneous multi-threaded operating system |
US7139150B2 (en) * | 2004-02-10 | 2006-11-21 | Marvell International Ltd. | Method and system for head position control in embedded disk drive controllers |
US7120084B2 (en) * | 2004-06-14 | 2006-10-10 | Marvell International Ltd. | Integrated memory controller |
US8166217B2 (en) * | 2004-06-28 | 2012-04-24 | Marvell International Ltd. | System and method for reading and writing data using storage controllers |
US9201599B2 (en) * | 2004-07-19 | 2015-12-01 | Marvell International Ltd. | System and method for transmitting data in storage controllers |
US7757009B2 (en) | 2004-07-19 | 2010-07-13 | Marvell International Ltd. | Storage controllers with dynamic WWN storage modules and methods for managing data and connections between a host and a storage device |
US8032674B2 (en) * | 2004-07-19 | 2011-10-04 | Marvell International Ltd. | System and method for controlling buffer memory overflow and underflow conditions in storage controllers |
US8230174B2 (en) | 2004-07-26 | 2012-07-24 | Integrated Device Technology, Inc. | Multi-queue address generator for start and end addresses in a multi-queue first-in first-out memory system |
US7290112B2 (en) * | 2004-09-30 | 2007-10-30 | International Business Machines Corporation | System and method for virtualization of processor resources |
US7386661B2 (en) | 2004-10-13 | 2008-06-10 | Marvell International Ltd. | Power save module for storage controllers |
US7240267B2 (en) | 2004-11-08 | 2007-07-03 | Marvell International Ltd. | System and method for conducting BIST operations |
US7802026B2 (en) * | 2004-11-15 | 2010-09-21 | Marvell International Ltd. | Method and system for processing frames in storage controllers |
US20060155940A1 (en) | 2005-01-10 | 2006-07-13 | Mario Au | Multi-queue FIFO memory systems that utilize read chip select and device identification codes to control one-at-a-time bus access between selected FIFO memory chips |
US7609468B2 (en) * | 2005-04-06 | 2009-10-27 | Marvell International Ltd. | Method and system for read gate timing control for storage controllers |
US8094552B1 (en) * | 2005-11-03 | 2012-01-10 | Seagate Technology Llc | Adaptive buffer for frame based storage communications protocols |
US7613211B1 (en) * | 2006-02-18 | 2009-11-03 | Wideband Semiconductors, Inc. | Digital clock smoothing apparatus and method |
US7924521B1 (en) * | 2007-04-10 | 2011-04-12 | Marvell International Ltd. | Data wedge format table synchronization |
US8139072B2 (en) * | 2008-04-14 | 2012-03-20 | Mcgowan Scott James | Network hardware graphics adapter compression |
US8566515B2 (en) * | 2009-01-12 | 2013-10-22 | Maxim Integrated Products, Inc. | Memory subsystem |
TWI433163B (en) * | 2009-12-23 | 2014-04-01 | Mstar Semiconductor Inc | Memory controller and associated controlling method |
US8392689B1 (en) * | 2010-05-24 | 2013-03-05 | Western Digital Technologies, Inc. | Address optimized buffer transfer requests |
CN103119548A (en) | 2010-09-23 | 2013-05-22 | 马维尔以色列(M.I.S.L.)有限公司 | Low latency first-in-first-out (FIFO) buffer |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2181870B (en) * | 1985-10-14 | 1988-11-23 | Anamartic Ltd | Control circuit for chained circuit modules |
US4949301A (en) * | 1986-03-06 | 1990-08-14 | Advanced Micro Devices, Inc. | Improved pointer FIFO controller for converting a standard RAM into a simulated dual FIFO by controlling the RAM's address inputs |
US5133062A (en) * | 1986-03-06 | 1992-07-21 | Advanced Micro Devices, Inc. | RAM buffer controller for providing simulated first-in-first-out (FIFO) buffers in a random access memory |
JPS6375955A (en) * | 1986-09-19 | 1988-04-06 | Fujitsu Ltd | Program mode access control system |
US5267191A (en) * | 1989-04-03 | 1993-11-30 | Ncr Corporation | FIFO memory system |
US5305319A (en) * | 1991-01-31 | 1994-04-19 | Chips And Technologies, Inc. | FIFO for coupling asynchronous channels |
EP0522224B1 (en) * | 1991-07-10 | 1998-10-21 | International Business Machines Corporation | High speed buffer management |
US5365485A (en) * | 1993-11-22 | 1994-11-15 | Texas Instruments Incorporated | Fifo with fast retransmit mode |
US5781802A (en) | 1995-02-03 | 1998-07-14 | Vlsi Technology, Inc. | First-in-first-out (FIFO) controller for buffering data between systems which are asynchronous and free of false flags and internal metastability |
IL116984A (en) * | 1996-01-31 | 2000-07-26 | Galileo Technology Ltd | Multiple FIFO array and method of construction thereof |
KR0165515B1 (en) * | 1996-02-17 | 1999-01-15 | 김광호 | Fifo method and apparatus of graphic data |
-
1999
- 1999-01-19 US US09/232,764 patent/US6381659B2/en not_active Expired - Lifetime
Cited By (57)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6618390B1 (en) * | 1999-05-21 | 2003-09-09 | Advanced Micro Devices, Inc. | Method and apparatus for maintaining randomly accessible free buffer information for a network switch |
US20050018514A1 (en) * | 2001-08-23 | 2005-01-27 | Knaack Roland T. | Integrated DDR/SDR flow control managers that support multiple queues and mux, demux and broadcast operating modes |
US7082071B2 (en) | 2001-08-23 | 2006-07-25 | Integrated Device Technology, Inc. | Integrated DDR/SDR flow control managers that support multiple queues and MUX, DEMUX and broadcast operating modes |
US20030206475A1 (en) * | 2001-08-23 | 2003-11-06 | Jiann-Jeng Duh | FIFO memory devices that support all combinations of DDR and SDR read and write modes |
US7158440B2 (en) | 2001-08-23 | 2007-01-02 | Integrated Device Technology, Inc. | FIFO memory devices having write and read control circuits that support x4N, x2N and xN data widths during DDR and SDR modes of operation |
US6778454B2 (en) | 2001-08-23 | 2004-08-17 | Integrated Device Technology, Inc. | FIFO memory devices that support all combinations of DDR and SDR read and write modes |
US6795360B2 (en) | 2001-08-23 | 2004-09-21 | Integrated Device Technology, Inc. | Fifo memory devices that support all four combinations of DDR or SDR write modes with DDR or SDR read modes |
GB2386485B (en) * | 2002-03-12 | 2004-06-23 | Toshiba Res Europ Ltd | Modulo addressing apparatus and methods |
GB2386485A (en) * | 2002-03-12 | 2003-09-17 | Toshiba Res Europ Ltd | Modulo addressing apparatus and methods |
US6845440B2 (en) * | 2002-11-11 | 2005-01-18 | Hewlett-Packard Development Company, L.P. | System for preventing memory usage conflicts when generating and merging computer architecture test cases |
US20040093476A1 (en) * | 2002-11-11 | 2004-05-13 | Thompson Ryan C. | System for preventing memory usage conflicts when generating and merging computer architecture test cases |
US7287102B1 (en) * | 2003-01-31 | 2007-10-23 | Marvell International Ltd. | System and method for concatenating data |
US20040193812A1 (en) * | 2003-03-11 | 2004-09-30 | Zayas Fernando A. | Method for improving the performance of read/write testing in a hard drive |
US20040193785A1 (en) * | 2003-03-11 | 2004-09-30 | Zayas Fernando A. | System for improving the performance of read/write testing in a hard drive |
EP1482402A2 (en) * | 2003-05-27 | 2004-12-01 | Micronas GmbH | Fill level capture in a buffer |
EP1482402A3 (en) * | 2003-05-27 | 2006-03-29 | Micronas GmbH | Fill level capture in a buffer |
US20040250039A1 (en) * | 2003-06-05 | 2004-12-09 | Hanks D. Mitchel | System and method for using swappable storage for high data content multi-source data storage |
US7120075B1 (en) | 2003-08-18 | 2006-10-10 | Integrated Device Technology, Inc. | Multi-FIFO integrated circuit devices that support multi-queue operating modes with enhanced write path and read path queue switching |
US20050286534A1 (en) * | 2004-06-28 | 2005-12-29 | Inching Chen | Systems and methods for packet multicasting with a multi-read buffer |
US20070008985A1 (en) * | 2005-06-30 | 2007-01-11 | Sridhar Lakshmanamurthy | Method and apparatus to support efficient check-point and role-back operations for flow-controlled queues in network devices |
US7505410B2 (en) * | 2005-06-30 | 2009-03-17 | Intel Corporation | Method and apparatus to support efficient check-point and role-back operations for flow-controlled queues in network devices |
US20090043927A1 (en) * | 2005-08-15 | 2009-02-12 | Sony Computer Entertainment Inc. | Buffer managing method and buffer managing apparatus |
US7904618B2 (en) * | 2005-08-15 | 2011-03-08 | Sony Computer Entertainment Inc. | Buffer managing method and buffer managing apparatus |
US20080049240A1 (en) * | 2006-08-28 | 2008-02-28 | Fujitsu Limited | Image data buffer apparatus and data transfer system for efficient data transfer |
EP1895470A3 (en) * | 2006-08-28 | 2011-02-23 | Fujitsu Semiconductor Limited | Image data buffer apparatus and data transfer system for efficient data transfer |
US8687222B2 (en) * | 2006-08-28 | 2014-04-01 | Fujitsu Semiconductor Limited | Image data buffer apparatus and data transfer system for efficient data transfer |
US8032725B1 (en) * | 2007-01-30 | 2011-10-04 | Netapp, Inc. | Backup data management on a fractionally reserved storage system |
US20100005250A1 (en) * | 2008-07-02 | 2010-01-07 | Cradle Technologies, Inc. | Size and retry programmable multi-synchronous fifo |
US8681526B2 (en) * | 2008-07-02 | 2014-03-25 | Cradle Ip, Llc | Size and retry programmable multi-synchronous FIFO |
US8904033B2 (en) * | 2010-06-07 | 2014-12-02 | Adobe Systems Incorporated | Buffering media content |
US20130166772A1 (en) * | 2010-06-07 | 2013-06-27 | Adode Systems Incorporated | Buffering Media Content |
CN102402486A (en) * | 2010-09-17 | 2012-04-04 | 腾讯科技(深圳)有限公司 | Method and device for reading data |
US20120079202A1 (en) * | 2010-09-28 | 2012-03-29 | Kai Chirca | Multistream prefetch buffer |
US8631265B2 (en) * | 2010-12-13 | 2014-01-14 | Oracle International Corporation | Synchronization circuit that facilitates multiple parallel reads and writes |
US20120151243A1 (en) * | 2010-12-13 | 2012-06-14 | Oracle International Corporation | Synchronization circuit that facilitates multiple parallel reads and writes |
US20120226852A1 (en) * | 2011-03-03 | 2012-09-06 | Mstar Semiconductor, Inc. | Control method and controller for dram |
US8782332B2 (en) * | 2011-03-03 | 2014-07-15 | Mstar Semiconductor, Inc. | Control method and controller for DRAM |
TWI447728B (en) * | 2011-03-03 | 2014-08-01 | Mstar Semiconductor Inc | Controlling method and controller for dram |
US20150324136A1 (en) * | 2014-05-07 | 2015-11-12 | Lsi Corporation | Storage system having fifo storage and reserved storage |
US20160011989A1 (en) * | 2014-07-08 | 2016-01-14 | Fujitsu Limited | Access control apparatus and access control method |
US20160092285A1 (en) * | 2014-09-25 | 2016-03-31 | Intel Corporation | Method and Apparatus for Approximating Detection of Overlaps Between Memory Ranges |
US9910650B2 (en) * | 2014-09-25 | 2018-03-06 | Intel Corporation | Method and apparatus for approximating detection of overlaps between memory ranges |
US20160139880A1 (en) * | 2014-11-14 | 2016-05-19 | Cavium, Inc. | Bypass FIFO for Multiple Virtual Channels |
US9824058B2 (en) * | 2014-11-14 | 2017-11-21 | Cavium, Inc. | Bypass FIFO for multiple virtual channels |
CN104636087A (en) * | 2015-02-09 | 2015-05-20 | 华为技术有限公司 | Method and device for controlling reading data |
US11385862B2 (en) * | 2016-05-06 | 2022-07-12 | Texas Instmments Incorporated | Data flow control for multi-chip select |
CN108073602A (en) * | 2016-11-11 | 2018-05-25 | 北京京东尚科信息技术有限公司 | The method and device of processing business data segment |
US10489069B2 (en) | 2017-11-29 | 2019-11-26 | International Business Machines Corporation | Address/command chip synchronized autonomous data chip address sequencer for a distributed buffer memory system |
US10534555B2 (en) | 2017-11-29 | 2020-01-14 | International Business Machines Corporation | Host synchronized autonomous data chip address sequencer for a distributed buffer memory system |
US10747442B2 (en) * | 2017-11-29 | 2020-08-18 | International Business Machines Corporation | Host controlled data chip address sequencing for a distributed memory buffer system |
US10976939B2 (en) | 2017-11-29 | 2021-04-13 | International Business Machines Corporation | Address/command chip synchronized autonomous data chip address sequencer for a distributed buffer memory system |
US11379123B2 (en) | 2017-11-29 | 2022-07-05 | International Business Machines Corporation | Address/command chip synchronized autonomous data chip address sequencer for a distributed buffer memory system |
US20190163362A1 (en) * | 2017-11-29 | 2019-05-30 | International Business Machines Corporation | Host Controlled Data Chip Address Sequencing for a Distributed Memory Buffer System |
US11587600B2 (en) | 2017-11-29 | 2023-02-21 | International Business Machines Corporation | Address/command chip controlled data chip address sequencing for a distributed memory buffer system |
US11687254B2 (en) | 2017-11-29 | 2023-06-27 | International Business Machines Corporation | Host synchronized autonomous data chip address sequencer for a distributed buffer memory system |
CN112667165A (en) * | 2020-12-29 | 2021-04-16 | 湖南博匠信息科技有限公司 | Data storage system and data storage method thereof |
CN113900974A (en) * | 2021-12-07 | 2022-01-07 | 睿思芯科(深圳)技术有限公司 | Storage device, data storage method and related equipment |
Also Published As
Publication number | Publication date |
---|---|
US6381659B2 (en) | 2002-04-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6381659B2 (en) | Method and circuit for controlling a first-in-first-out (FIFO) buffer using a bank of FIFO address registers capturing and saving beginning and ending write-pointer addresses | |
US5301310A (en) | Parallel disk storage array system with independent drive operation mode | |
US8392689B1 (en) | Address optimized buffer transfer requests | |
US20100208385A1 (en) | Storage device with read-ahead function | |
US20020016884A1 (en) | Method and apparatus for synchronous data transfers in a memory device with selectable data or address paths | |
US4507693A (en) | Control apparatus for rotation type storage device | |
US6775744B2 (en) | Disk memory device | |
JPS581465B2 (en) | Data signal buffering method | |
US20070168626A1 (en) | Transforming flush queue command to memory barrier command in disk drive | |
US9026746B2 (en) | Signal control device and signal control method | |
US6735672B2 (en) | Data storage array device and data access method | |
US4903195A (en) | Method for controlling data transfer | |
KR19990028247A (en) | Disk drive data protection system | |
US4651277A (en) | Control system for a magnetic disk drive unit | |
US4584617A (en) | Method and apparatus for improving the transfer time in rotating data storage systems | |
US6924953B2 (en) | Dual communication port sharing apparatus | |
JP3520032B2 (en) | Data processing device | |
EP1163569B1 (en) | Method and circuit for receiving dual edge clocked data | |
US6751717B2 (en) | Method and apparatus for clock synchronization between a system clock and a burst data clock | |
CN101201725B (en) | Tape recording apparatus and data write-control method | |
US3911400A (en) | Drive condition detecting circuit for secondary storage facilities in data processing systems | |
US6792504B2 (en) | Read on arrival scheme for a disc drive | |
US6161155A (en) | Apparatus and method for storing retrievable boundary information into a buffer memory of a receiving device | |
US6745300B1 (en) | Magnetic disk device and disk access method therefor | |
US6088818A (en) | Data read apparatus and data read method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: QUANTUM CORPORATION, CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:PROCH, TIMOTHY;HORGAN, NICK;REEL/FRAME:009733/0059 Effective date: 19990114 |
|
AS | Assignment |
Owner name: MAXTOR CORPORATION, CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:QUANTUM CORPORATION;REEL/FRAME:012653/0726 Effective date: 20010724 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
FPAY | Fee payment |
Year of fee payment: 4 |
|
AS | Assignment |
Owner name: JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT Free format text: SECURITY AGREEMENT;ASSIGNORS:MAXTOR CORPORATION;SEAGATE TECHNOLOGY LLC;SEAGATE TECHNOLOGY INTERNATIONAL;REEL/FRAME:022757/0017 Effective date: 20090507 Owner name: WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATE Free format text: SECURITY AGREEMENT;ASSIGNORS:MAXTOR CORPORATION;SEAGATE TECHNOLOGY LLC;SEAGATE TECHNOLOGY INTERNATIONAL;REEL/FRAME:022757/0017 Effective date: 20090507 |
|
FPAY | Fee payment |
Year of fee payment: 8 |
|
AS | Assignment |
Owner name: MAXTOR CORPORATION, CALIFORNIA Free format text: RELEASE;ASSIGNOR:JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT;REEL/FRAME:025662/0001 Effective date: 20110114 Owner name: SEAGATE TECHNOLOGY LLC, CALIFORNIA Free format text: RELEASE;ASSIGNOR:JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT;REEL/FRAME:025662/0001 Effective date: 20110114 Owner name: SEAGATE TECHNOLOGY HDD HOLDINGS, CALIFORNIA Free format text: RELEASE;ASSIGNOR:JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT;REEL/FRAME:025662/0001 Effective date: 20110114 Owner name: SEAGATE TECHNOLOGY INTERNATIONAL, CALIFORNIA Free format text: RELEASE;ASSIGNOR:JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT;REEL/FRAME:025662/0001 Effective date: 20110114 |
|
AS | Assignment |
Owner name: THE BANK OF NOVA SCOTIA, AS ADMINISTRATIVE AGENT, Free format text: SECURITY AGREEMENT;ASSIGNOR:SEAGATE TECHNOLOGY LLC;REEL/FRAME:026010/0350 Effective date: 20110118 |
|
AS | Assignment |
Owner name: EVAULT INC. (F/K/A I365 INC.), CALIFORNIA Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENT RIGHTS;ASSIGNOR:WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATERAL AGENT AND SECOND PRIORITY REPRESENTATIVE;REEL/FRAME:030833/0001 Effective date: 20130312 Owner name: SEAGATE TECHNOLOGY LLC, CALIFORNIA Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENT RIGHTS;ASSIGNOR:WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATERAL AGENT AND SECOND PRIORITY REPRESENTATIVE;REEL/FRAME:030833/0001 Effective date: 20130312 Owner name: SEAGATE TECHNOLOGY US HOLDINGS, INC., CALIFORNIA Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENT RIGHTS;ASSIGNOR:WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATERAL AGENT AND SECOND PRIORITY REPRESENTATIVE;REEL/FRAME:030833/0001 Effective date: 20130312 Owner name: SEAGATE TECHNOLOGY INTERNATIONAL, CAYMAN ISLANDS Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENT RIGHTS;ASSIGNOR:WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATERAL AGENT AND SECOND PRIORITY REPRESENTATIVE;REEL/FRAME:030833/0001 Effective date: 20130312 |
|
FPAY | Fee payment |
Year of fee payment: 12 |