This is the mail archive of the binutils@sourceware.org mailing list for the binutils project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [PATCH] give .comment sections SHF_MERGE|SHF_STRINGS

From: Cary Coutant <ccoutant at google dot com>
To: "H.J. Lu" <hjl dot tools at gmail dot com>
Cc: Roland McGrath <roland at redhat dot com>, binutils at sources dot redhat dot com, Andreas Schwab <schwab at redhat dot com>, Nick Clifton <nickc at redhat dot com>
Date: Fri, 24 Jul 2009 15:26:58 -0700
Subject: Re: [PATCH] give .comment sections SHF_MERGE|SHF_STRINGS
References: <20090724190834.36DC23CCB@magilla.sf.frob.com> <6dc9ffc80907241515u159634f6kd2f7a6d5b854c248@mail.gmail.com>

>> The compiler-generated .ident strings linked together tend all to be the
>> same, resulting in giant .comment sections with a duplicate string for each
>> translation unit. ?We have the technology. ?We can make them smaller, faster,
>> better than before.
>>
>> This looks obviously trivial and safe. ?But I can't be bothered to actually
>> test it or anything. ?Anyone inspired?
>
> How about making it comdat?

That would then keep only one string, even if some happened to be
different from the rest -- unless you copy the string itself into the
comdat key, but that would be pretty wasteful. This is what merge
sections are for.

-cary

Follow-Ups:
- Re: [PATCH] give .comment sections SHF_MERGE|SHF_STRINGS
  - From: H.J. Lu

References:
- [PATCH] give .comment sections SHF_MERGE|SHF_STRINGS
  - From: Roland McGrath
- Re: [PATCH] give .comment sections SHF_MERGE|SHF_STRINGS
  - From: H.J. Lu

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]