[PATCH] Fully disable DWARF accelerator tables

[Eric Christopher]

Yep. Something like that.

-eric

On Fri, Jan 31, 2014 at 11:01 AM, Robinson, Paul
<Paul_Robinson at playstation.sony.com> wrote:
> Okay, a bit less intrusive, like this?  I agree it's easier to
> see what's going on this way, and less likely to accidentally
> screw up pubtypes like the previous patch did (missed that test,
> sorry, I'm officially doing too many things at once).
> --paulr
>
>
>> -----Original Message-----
>> From: Eric Christopher [mailto:echristo at gmail.com]
>> Sent: Thursday, January 30, 2014 2:11 PM
>> To: Robinson, Paul
>> Cc: llvm-commits at cs.uiuc.edu commits (llvm-commits at cs.uiuc.edu)
>> Subject: Re: [PATCH] Fully disable DWARF accelerator tables
>>
>> Why not just check in the functions rather than at all the callsites?
>>
>> if (!DD->...) return;
>>
>> -eric
>>
>> On Thu, Jan 30, 2014 at 1:55 PM, Robinson, Paul
>> <Paul_Robinson at playstation.sony.com> wrote:
>> > The DWARF accelerator tables are or are not supported/emitted
>> depending on
>> > the target.  However the work to collect the info to put into the
>> > accelerator tables is always done regardless of target.  The attached
>> patch
>> > keeps that extra work (and memory consumption) from happening.
>> >
>> > No functional change really, and therefore no test, because the same
>> > (mangled) names appear in .debug_str because of DW_AT_linkage_name.  I
>> found
>> > this because we have a local patch to suppress DW_AT_linkage_name but
>> the
>> > mangled names were still showing up in .debug_str.
>> >
>> > Thanks,
>> >
>> > --paulr
>> >
>> >
>> >
>> >
>> >
>> >
>> > _______________________________________________
>> > llvm-commits mailing list
>> > llvm-commits at cs.uiuc.edu
>> > http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>> >
>