This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: [gold] Merging string literals with bigger alignment
- From: Alexander Ivchenko <aivchenk at gmail dot com>
- To: Cary Coutant <ccoutant at google dot com>
- Cc: Ian Lance Taylor <iant at google dot com>, binutils <binutils at sourceware dot org>, Sriraman Tallam <tmsriram at google dot com>
- Date: Fri, 22 Mar 2013 18:38:59 +0400
- Subject: Re: [gold] Merging string literals with bigger alignment
- References: <CACysShhXDPxUsqdUprwhyGcaNiS9X_ODbrW847wHHiwbUh4LhA at mail dot gmail dot com> <CAKOQZ8xROxuJMD19P7JoM084LDyDKk_anwiuoWRpS+2MLeThnA at mail dot gmail dot com> <CACysShhaCv68A39oTk5FfmsdRFjhoOjQEzEZys4U0-sZs-x+4A at mail dot gmail dot com> <CAKOQZ8ypF=yCMmrwn-8jSp_WBMdrNi6NqwJRqabG7ZUwV2yMhg at mail dot gmail dot com> <CACysShhyE2G_8fz1pLSjTdJ_f9OYdSHHB9zvydWPaJNHjFkQ8g at mail dot gmail dot com> <CAKOQZ8wd_7bfSfMsm_WmJnSdT_J4qF-JM+3Ex=tU0iDJCOWETQ at mail dot gmail dot com> <CACysShh22grCuoa7-b3QV7D_Q-RBYrtLH8ifr2yRcS93+GRk1g at mail dot gmail dot com> <CAHACq4pqmPp7JqwOJ_Ux0NN7yVZzq82t4uf9ZUKTa8eXL+Znrg at mail dot gmail dot com>
> Presumably the compiler aligns each string within the input section to
> the section's alignment, but your patch doesn't check for that. If you
> find a string in the input section that doesn't begin at an alignment
> boundary, you should either refuse to merge that section, or preserve
> that string's alignment modulo the alignment boundary (e.g., if a
> string in the .rodata.str1.8 section begins at offset 0x14, you should
> ensure that the string's new offset is at an offset that is 4 modulo
> 8).
But isn't that defined that if the string is placed in, say,
rodata.1.8 section, than
we can count that its alignment is 8? If the user doesn't want that string to
be 8-aligned or wants 4 modulo 8 then he\she should not put that string in
8-aligned section. As far as I understand BFD merge code is not checking that
(why should it?) and we are compatible
> One nit:
>
> +Stringpool_template<Stringpool_char>::new_key_offset(size_t length,
> uint64_t addralign)
>
> This line is too long.
>
> -cary
Missed that, thanks.