[Bug gold/22914] Gold doesn't support .note.gnu.property section (NT_GNU_PROPERTY_TYPE_0)

[hjl.tools at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=22914

--- Comment #6 from H.J. Lu <hjl.tools at gmail dot com> ---
(In reply to Cary Coutant from comment #1)
> HJ,
> 
> The linux-abi document leaves a few things unspecified:
> 
> - Is the .note.gnu.property section SHF_ALLOC or not?

It is no different from other SHT_NOTE sections.

> - Does it go in the PT_NOTE segment? Should it be placed at the
> beginning of the segment?

If it has SHF_ALLOC, it should go into the PT_NOTE segment.

> - It says the .note.gnu.property section may be combined with other
> note sections. (Is that true, or is it missing a "not"?) If the linker
> combines it with other SHT_NOTE sections, does the name matter at all?

It isn't strictly required.  It is just another special section.

> - Does the linker key off the name of the section, or does it need to
> parse all SHT_NOTE sections looking for NT_GNU_PROPERTY_TYPE_0 and
> "GNU"? (Neither option is attractive -- I'd much prefer using a custom
> section type and program header type, like ARM, MIPS, and Itanium do
> for their attributes sections. The linker and loader aren't meant to
> be parsing note sections.)

Linker and loader need to look for NT_GNU_PROPERTY_TYPE_0 and "GNU",
similar to other "GNU" note sections.

> - Where did we end up with note section alignment for 64-bit objects?
> Is it still 4-byte aligned?

No, as documented in linux gABI:

2.1.7 Alignment of Note Sections
All entries in a PT_NOTE segment have the same alignment which equals to the
p_align field in program header.
According to gABI, each note entry should be aligned to 4 bytes in 32-bit
objects or 8 bytes in 64-bit objects. But .note.ABI-tag section (see Section
2.1.6) and .note.gnu.build-id section (see Section 2.1.4) are aligned
to 4 bytes in both 32-bit and 64-bit objects. Note parser should use p_align
for
note alignment, instead of assuming alignment based on ELF file class.

-- 
You are receiving this mail because:
You are on the CC list for the bug.