Compiler controls for program language constructs

Patent metadata

Field	Value
Publication number	US-10732930-B2
Application number	US-201715816384-A
Country	US
Kind code	B2
Filing date	Nov 17, 2017
Priority date	Jun 23, 2017
Publication date	Aug 4, 2020
Grant date	Aug 4, 2020

Abstract

Setting or updating of floating point controls is managed. Floating point controls include controls used for floating point operations, such as rounding mode and/or other controls. Further, floating point controls include status associated with floating point operations, such as floating point exceptions and/or others. The management of the floating point controls includes efficiently updating the controls, while reducing costs associated therewith.

First claim

What is claimed is:

1. A computer-implemented method of facilitating processing within a computing environment, the computer-implemented method comprising:
obtaining, by a compiler, a portion of code;
determining whether the portion of code has a user-specified floating point control associated therewith, the determining checking whether there is at least one delimiter included in the portion of code that indicates there is the user-specified floating point control, the at least one delimiter including an explicit begin floating point control and an explicit end floating point control that indicates for code between the explicit begin floating point control and the explicit end floating point control that there is the user-specified floating point control to be applied to the code between the explicit begin floating point control and the explicit end floating point control and wherein the explicit begin floating point control specifies the user-specified floating point control to be applied to the code between the explicit begin floating point control and the explicit end floating point control;
generating an internal representation of the portion of code that includes the user-specified floating point control, based on determining the portion of code has the user-specified floating point control;
obtaining the internal representation of the portion of code;
determining, based on obtaining and reading the internal representation of the portion of code, whether the internal representation has the user-specified floating point control associated therewith, the user-specified floating point control to modify a floating point operation; and
generating one or more machine instructions of an instruction stream based on determining the internal representation has the user-specified floating point control associated therewith, the one or more machine instructions generated facilitating executing the floating point operation using the user-specified floating point control without obtaining the user-specified floating point control from a global resource during the floating point operation, wherein the global resource includes, in part, a floating point control register, and wherein a machine instruction of the one or more machine instructions includes:
a control prefix specifying the user-specified floating point control to be used during the floating point operation, the user-specified floating point control modifying the floating point operation without referencing a floating point control in the global resource; and
an opcode, the opcode specifying the floating point operation that is modified by the user-specified floating point control, the opcode being preceded by the control prefix in the machine instruction, wherein execution of the floating point operation specified by the opcode is directly modified by the user-specified floating point control specified in the control prefix in lieu of what value exists for the floating point control in the global resource.

2. The computer-implemented method of claim 1 , further comprising performing one or more optimizations based on the user-specified floating point control.

3. The computer-implemented method of claim 1 , further including optimizing, based on the user-specified floating point control, a previous internal representation to generate the internal representation that is obtained.

4. The computer-implemented method of claim 3 , wherein the optimizing includes using constant folding based on the user-specified floating point control.

5. The computer-implemented method of claim 1 , wherein the user-specified floating point control includes a rounding mode.

6. The computer-implemented method of claim 1 , wherein the internal representation is based on an abstract syntax tree.

7. The computer-implemented method of claim 1 , wherein the user-specified floating point control includes a floating point exception.

8. The computer-implemented method of claim 1 , wherein the user-specified floating point control is selected from the group consisting of round to nearest, round towards nearest with ties to even, round towards zero, round towards plus infinity, round towards minus infinity, round up, round down, and round away_from_0.

Assignees

• IBM

Inventors

• Gschwind Michael K
• Salapura Valentina

Classifications

• G06F7/483 Primary

  Computations with numbers represented by a non-linear combination of denominational numbers, e.g. rational numbers, logarithmic number system or floating-point numbers {(G06F7/4806, G06F7/4824, G06F7/49, G06F7/491, G06F7/544 take precedence)} · CPC title

• G06F7/49947

  Rounding · CPC title

• G06F7/49973

  Rounding towards negative infinity, e.g. truncation of two's complement numbers (G06F7/49957 takes precedence) · CPC title

• G06F8/443

  Optimisation · CPC title

• G06F7/4991 Primary

  Overflow or underflow · CPC title