This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
RFC: Should we add SHF_GNU_COMPRESSED?
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Cary Coutant <ccoutant at google dot com>
- Cc: Binutils <binutils at sourceware dot org>
- Date: Tue, 20 Jul 2010 11:50:20 -0700
- Subject: RFC: Should we add SHF_GNU_COMPRESSED?
On Tue, Jul 20, 2010 at 11:12 AM, Cary Coutant <ccoutant@google.com> wrote:
>> I noticed that --compress-debug-sections might generate bigger sections,
>> depending on input. Shouldn't we use uncompressed data instead if
>> the compressed one is bigger?
>
> I thought about that, and it would be a reasonable thing to do once
> we've rolled over to using a section type or flag. As it is, I was
> reluctant to set things up where we might have, say, a .zdebug_info
> and a .debug_abbrev section, as it might confuse some tools that
> aren't yet aware of the compressed sections.
>
My patch is almost ready. I am adding testcases. linker
--compress-debug-sections
will be a separate patch later
I don't like SHT_XXX and we don't have to use SHF_GNU_COMPRESSED.
You can peek .debug sections to see if there is a ZLIB header like what gdb
does. On the other hand, SHF_GNU_COMPRESSED can avoid the peek. I
can go with or without SHF_GNU_COMPRESSED. Any comments on that?
Thanks.
--
H.J.