Enable vectorization of call instructions in the loop vectorizer
James.Molloy at arm.com
Tue Dec 17 11:18:07 PST 2013
Fully in agreement, but you can't put metadata on function declarations :)
Sent from my Windows Phone
From: Philip Reames<mailto:listmail at philipreames.com>
Sent: 17/12/2013 19:15
To: James Molloy<mailto:James.Molloy at arm.com>; llvm-commits at cs.uiuc.edu<mailto:llvm-commits at cs.uiuc.edu>
Subject: Re: RFC: Enable vectorization of call instructions in the loop vectorizer
On 12/16/13 9:08 AM, James Molloy wrote:
So the version attached reads metadata attached to CallInsts. The schema for the metadata is detailed in the proposed LangRef addition, but basically it describes a list of potential vectorization candidates. Each candidate has a vector width, a llvm::Function* (or MDString) giving the target function and a string describing how the function arguments need to be handled.
Specifically with regards to metadata, why is the metadata on the call site not the function declaration? I would expect that a call to "cos" would always vectorize to the same "cos4". Is supporting different vectorizations at different points of the program a key goal?
Ignoring most of the discussion since it seems like a good conclusion has been reached.
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590
ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the llvm-commits