From: Frederick Virchanza Gotham <cauldwell.thomas@gmail.com>
To: binutils@sourceware.org
Subject: Linker : Make a map of typeinfo to vtable
Date: Fri, 19 Jan 2024 22:29:21 +0000 [thread overview]
Message-ID: <CALtZhhMUxmUzWrHLB8xDA_0z0yjgZDLp8DqaofJvVk6rZznQ=g@mail.gmail.com> (raw)
[-- Attachment #1: Type: text/plain, Size: 2236 bytes --]
Over on the mailing list for new proposals for the C++ programming
language, I am proposing that we be able to get a pointer to a class's
vtable from its typeinfo.
So I want to edit the source code for the GNU linker in order to add two
new symbols to every binary:
(1) __map_typeinfohash_vtable
(2) __map_typeinfohash_vtable_size
The first one is an array mapping the hashes of type_info's to vtable
pointers. The second one is the number of elements in the array. The array
would look something like the following in C++:
pair<size_t, void const*> const g_vtables[] = {
{ typeid(MyClass).hash_code(), &_ZTI7MyClass },
{ typeid(YourClass).hash_code(), &_ZTI9YourClass },
};
Next I can write a function called 'std::get_polymorphic_facilitator' which
would
perform a binary search through this array, in order to convert a typeinfo
to a vtable pointer. So if a massive program like Chromium has 6,000
classes, then a worst case binary search would be about
13 checks. On a desktop PC that's less than a microsecond.
Afterward I would fork libstdc++ to edit the <typeinfo> standard header
file to add the following inline function:
inline void const *get_polymorphic_facilitator(type_info const &ti)
{
extern pair<size_t,void const*> const *const
__map_typeinfohash_vtable;
extern size_t const __map_typeinfohash_vtable_size;
return lower_bound( p, p + __map_typeinfohash_vtable_size,
ti.hash_code(), OnlyCompareFirst()
)->second;
}
But first of all I need to edit the source code for 'ld' so that, after it
has generated all the vtables and typeinfo's for every class, I can then
create my aforementioned array which maps typeinfo's to vtables. Or,
instead of waiting until after all the vtables and typeinfo's have been
generated, I can populate my array as and when these things are generated
(I imagine this might be easier than waiting until they're all generated,
but whatever works).
I've never edited a compiler or linker before, so could I please ask for a
little guidance here? Could someone point me toward the right source file
for 'ld' where I should start making changes to implement this? I would
appreciate any tips you can give me please.
next reply other threads:[~2024-01-19 22:29 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-19 22:29 Frederick Virchanza Gotham [this message]
2024-01-19 22:52 ` Frederick Virchanza Gotham
2024-01-20 19:44 ` Frederick Virchanza Gotham
2024-01-21 3:27 ` Frederick Virchanza Gotham
2024-01-21 22:35 ` Frederick Virchanza Gotham
2024-01-22 16:50 ` Michael Matz
2024-01-25 10:17 ` Frederick Virchanza Gotham
2024-01-25 23:02 ` Frederick Virchanza Gotham
2024-01-27 10:56 ` Frederick Virchanza Gotham
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CALtZhhMUxmUzWrHLB8xDA_0z0yjgZDLp8DqaofJvVk6rZznQ=g@mail.gmail.com' \
--to=cauldwell.thomas@gmail.com \
--cc=binutils@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).