[PATCH] Adding a timing option for IR parsing

[Andrew Trick]

On Mar 22, 2013, at 11:29 AM, Eli Bendersky <eliben at google.com> wrote:

> 
> 
> 
> On Fri, Mar 22, 2013 at 11:25 AM, Chandler Carruth <chandlerc at google.com> wrote:
> On Fri, Mar 22, 2013 at 10:47 AM, Andrew Trick <atrick at apple.com> wrote:
> 
> On Mar 22, 2013, at 9:14 AM, Eli Bendersky <eliben at google.com> wrote:
> 
>> 
>> On Thu, Mar 21, 2013 at 7:24 PM, Chandler Carruth <chandlerc at google.com> wrote:
>> FYI, I reverted the patch in r177695 to fix our builds. We check that layering violations don't get any worse in our build system.
>> 
>> Can you explain what kind of build this breaks?
>> 
>> I understand the layering problem, but how do we still get to measure the time IR parsing took? Do you have plans to fix the IR Reader violation any time soon?
> 
> To answer your middle question. Just reuse -time-passes and expose the flag in LLVMContext.
> 
> On the last question, I would like to see an answer to Chandler's question about whether anyone cares if they now automatically link the IR parser. Maybe Chandler should re-ask on llvm-dev?
> 
> Chatted with Chris on IRC and he seemed happier with a new library. I'm adding that now.
> 
> It's called "IRReader" so we have IRReader/IRReader.h... if anyone really wants a different name, shout...
> 
> So should the flag go back to IRReader or to LLVMContext? I still think it makes sense to have it separate because we have a bunch of LLVM tools that read IR but don't run any "passes".

Back now to Nadav's original question. Why do we need a new command line flag? I don't see a good reason. If someone want to -time-passes, I think they'll be happy to get IR parsing time too. If the tool doesn't run formal passes, then it just shows parsing time and any other times the tool chooses to add under the TimePassesIsEnabled flag.

-Andy

> Chandler - are you going to add the timing back in when you create a new lib? If not let me know when it's good time to put it back in after you're done.
> 
> Eli
> 
>  

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20130322/5dabf09d/attachment.html>