Re: [PATCH, v1] Add code for processing version 5 DWP files (for use with DWARF v5)

[Simon Marchi <simark@simark.ca>]

On 2020-07-21 12:06 a.m., Caroline Tice wrote:
> The DWARF v5 Spec describes a (slightly) new format for V5 .dwp files.
> This patch updates GDB to allow it to read/process .dwp files in the
> new DWARF v5 format, while continuing to be able to read/process .dwp
> files in the older V1 & V2 formats.

Can you please describe in the commit message what those differences are?

> The one thing I felt a little odd about in this patch:  I couldn't
> re-use the enum dwarf_sect
> definitions, because  in version 5 several of the sections have the
> same name as in the previous versions, but have a different ordering,
> with different numbers attached.  So I had to create a new enum,
> dwarf_sect_v5 for this purpose.

That part would need to be cross-posted to the binutils mailing list.  binutils
does use the DW_SECT_* enumerators, presumably to read dwp files too, so they
would likely use those new DWARF 5 enumerators eventually.

> Is this patch ok to commit?

It would be useful to precise somewhere, perhaps in the comment on `struct dwp_sections`.
that versions 1 and 2 are pre-standard versions, and that version 5 was introduced in
DWARF5.  And that versions 3 and 4 don't exist.

I don't have time to do an in-depth review right now, but one question that came to mind
is: is an advantage of having virtual_v2_or_v5_dwo_sections over having separate
virtual_v2_dwo_sections and virtual_v5_dwo_sections?  Now when using v2 or v5, there are
fields you don't use (because they are either v2-specific or v5-specific), so I imagine
it's just more error prone.  Does it avoid a lot of code duplication?

Simon