ectrans benchmark with computing sp and dp in a sequence from a single binary #46
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… SP calls (instead of JPRB). Explicit type cast of JPRD to JPRB is added in gpnorm MIN/MAX calls to remove compilation warning of NVHPC
…tpm_pol commented out.
…rnal, introduce JPRC => JPRB type to annotate number constants in ./internal. This is because nvhpc preprocessor does not allow _JPRB macro.
…nd files with ectrans module and submodules, to allow for modern Fortran library imports. Currently modules only include legacy trans/external files for backward compatibility
…le-based, precision-dependent routines of trans
…s dp version after sp version
…CS are still inaccurate, as gstats is not prepared to separate SP and DP computations, possibly modification of FIAT is needed
|
Closed as no longer actively under development. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This draft PR still depends on ninja. As compared to the "legacydriver" PR, this adds an alternative benchmark program that does not include trans/external routines, but imports the new ectrans module explicitely via USE statement. The program itself is split into: lean, precision independent executable calling extracted components of the original benchmark, the precision-dependent driver module that executes the original benchmark components, and the data module that serves as a container for all precision-dependent matrices. Thanks to this structure, SP and DP namespaces and data storage is completely separated and can co-exist in a single binary. This is an upgrade as compared to the "legacydriver" PR, where only one: SP or DP data could be used, since they were allocated in the main program with JPRB.