Pre-Grant Publication Number: 20080046686
Please help the USPTO examine the application by evaluating the relevance of the publicly submitted prior art to the patent application.
Peer To Patent forwards the Top 10 most relevant prior art submissions and their annotations to the USPTO.
Review this prior art and click on the thumbs up (or down) to indicate whether this submission should be forwarded to the USPTO.
If you login then you can add an annotation by typing in the box at the bottom of the screen to comment on the relevance of the prior art to the claims of the patent application.
Review this prior art and click on the thumbs up (or down) to indicate whether this submission should be forwarded to the USPTO.
If you login then you can add an annotation by typing in the box at the bottom of the screen to comment on the relevance of the prior art to the claims of the patent application.
Prior Art Detail
Summary / Description
| Summary / Description | According to embodiments of the invention, there is disclosed a data processing unit, a method of operating the same, computer program product and an instruction. In one embodiment according to the invention, there is provided a data processing unit for a computer processor, the data processing unit comprising a deep register access mechanism capable of performing a permutation operation on at least one data operand accessed from a register file of the computer processor, the permutation operation being performed in series with (i) register access for the data operand and (ii) execution of a data processing operation on the operand. |
Basic Information
| Type of Prior Art | Issued Patents - US |
| Country | United States of America |
| Patent/Application # | US20060227966A1 |
| Kind Code | United States (US) - Patent Appl. Publ. within the ... - A1 |
| Patentee Name | Knowles, Simon; Bath, United Kingdom |
| Relevant Pages, Columns, or Lines | fig2, claims 1... |
| URL | |
| Filing Date | April 5, 2005 |
| Additional Information | |
Notes / To Do
| Notes | |
Excerpt
Excerpt Abstract
According to embodiments of the invention, there is disclosed a data processing unit, a method of operating the same, computer program product and an instruction. In one embodiment according to the invention, there is provided a data processing unit for a computer processor, the data processing unit comprising a deep register access mechanism capable of performing a permutation operation on at least one data operand accessed from a register file of the computer processor, the permutation operation being performed in series with (i) register access for the data operand and (ii) execution of a data processing operation on the operand. |
Relevance
Claims
1
Relevance
Claim 1 of disclosure essentially describes a system where the decode unit initiate a register fetch (fig 22, box 260) followed by formatting (in box 240). The formatting mentioned in disclosure is essentially handling half size values. For the “high”, one simply take the upper bits and shift them to the lower positions; for the “low”, one simply nix (and with bits 0) the upper bits.
This formatting functionality can be performed with a “permute logic” (see IBM VMX instruction set architecture, referred in the disclosure). By selecting the desired bits in the input logic of a permute (i.e. the high or low bits) and placing them in the desired position (i.e. the low bit), we essentially perform the disclosed processing for high/low with a permute instead of a shift or and-with-bit-zero. Thus there is no functional difference between using a permute logic or the shift/and approach. In fact, it is trivial to come up with the permute patterns that are needed to implement a specific handling of high or low disclosed here. Note that permute logic such as present in the Cell BE/SPE architecture [Power Efficient Processor Architecture and The Cell Processor, Hofstee, HPCA-11 2005] has a permute logic that can insert specific constants such as zero needed here.
The formatted data is then shipped to the execution unit (box 270 of fig 22). Knowles proposes exactly this scheme (see Knowles’ fig 2, box 202). The fact that this disclosure ships the operands 225 (in fig 22) first to box 240 instead of directly to 245 is meaningless, as no processing is done with the operand descriptor 225 prior to be shipped by wire 245 to box 260. Thus the scheme proposed here is entirely covered by Knowles.
While most permute logic cannot handle bit-level permute granularity, it is a trivial extension to extend a logic that permute individual byte quantities (such as the VMX permute logic) to bit quantities.
Claim 1 of disclosure essentially describes a system where the decode unit initiate a register fetch (fig 22, box 260) followed by formatting (in box 240). The formatting mentioned in disclosure is essentially handling half size values. For the “high”, one simply take the upper bits and shift them to the lower positions; for the “low”, one simply nix (and with bits 0) the upper bits.
This formatting functionality can be performed with a “permute logic” (see IBM VMX instruction set architecture, referred in the disclosure). By selecting the desired bits in the input logic of a permute (i.e. the high or low bits) and placing them in the desired position (i.e. the low bit), we essentially perform the disclosed processing for high/low with a permute instead of a shift or and-with-bit-zero. Thus there is no functional difference between using a permute logic or the shift/and approach. In fact, it is trivial to come up with the permute patterns that are needed to implement a specific handling of high or low disclosed here. Note that permute logic such as present in the Cell BE/SPE architecture [Power Efficient Processor Architecture and The Cell Processor, Hofstee, HPCA-11 2005] has a permute logic that can insert specific constants such as zero needed here.
The formatted data is then shipped to the execution unit (box 270 of fig 22). Knowles proposes exactly this scheme (see Knowles’ fig 2, box 202). The fact that this disclosure ships the operands 225 (in fig 22) first to box 240 instead of directly to 245 is meaningless, as no processing is done with the operand descriptor 225 prior to be shipped by wire 245 to box 260. Thus the scheme proposed here is entirely covered by Knowles.
While most permute logic cannot handle bit-level permute granularity, it is a trivial extension to extend a logic that permute individual byte quantities (such as the VMX permute logic) to bit quantities.
Claim Chart
All
2
Relevance
Claims 2 & 3 in this disclosure describe how the formatting is performed; this would correspond to signal to the permute logic in Knowles, and thus is prior art.
Claims 2 & 3 in this disclosure describe how the formatting is performed; this would correspond to signal to the permute logic in Knowles, and thus is prior art.
Claim Chart
All
3
Relevance
Claims 2 & 3 in this disclosure describe how the formatting is performed; this would correspond to signal to the permute logic in Knowles, and thus is prior art.
Claims 2 & 3 in this disclosure describe how the formatting is performed; this would correspond to signal to the permute logic in Knowles, and thus is prior art.
Claim Chart
All
5
Relevance
Claim 5 in this disclosure describe how to implement formatting. Knowles performs precisely this operation using its permute logic.
Claim 5 in this disclosure describe how to implement formatting. Knowles performs precisely this operation using its permute logic.
Claim Chart
All
