From: Josh Poimboeuf <jpoimboe@kernel.org>
To: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>,
live-patching@vger.kernel.org, linux-kernel@vger.kernel.org,
x86@kernel.org, Miroslav Benes <mbenes@suse.cz>,
Petr Mladek <pmladek@suse.com>,
Joe Lawrence <joe.lawrence@redhat.com>,
Jiri Kosina <jikos@kernel.org>,
Marcos Paulo de Souza <mpdesouza@suse.com>,
Song Liu <song@kernel.org>
Subject: Re: [RFC 28/31] x86/alternative: Create symbols for special section entries
Date: Wed, 4 Sep 2024 09:44:29 -0700 [thread overview]
Message-ID: <20240904164429.hstbg5beejt32mlu@treble> (raw)
In-Reply-To: <20240904123918.GCZthU9rOJLWUKBbsv@fat_crate.local>
On Wed, Sep 04, 2024 at 02:39:18PM +0200, Borislav Petkov wrote:
> On Tue, Sep 03, 2024 at 09:28:29PM -0700, Josh Poimboeuf wrote:
> > Take a more generic approach: for the "array of structs" style sections,
> > annotate each struct entry with a symbol containing the entry. This
> > makes it easy for tooling to parse the data and avoids the fragility of
> > hardcoding section details.
>
> This is more a question for my own understanding: so you want a generic
> approach where objtool doesn't need to know each special section and what it
> needs to copy there from but simply copy those new, special symbols over?
>
> Which you use as markers of sorts: "this is a relevant symbol, pls copy it"...
>
> I think you mean that but lemme confirm anyway.
Yes, that's exactly it. I should clarify a bit more.
> Also, have you checked whether this can be done with some compiler switch
> already?
Not that I know of, since the compiler usually doesn't have visibility
to these sections.
It might be possible to specify "entsize" in the .pushsection flags,
which is an ELF section header attribute which objtool could read.
But that wouldn't work for .altinstr_replacement and other sections with
variable-size entries. Also, "objtool klp diff" works by copying
symbols, so the current solution is definitely simpler as it fits in
nicely with the rest of the implementation.
--
Josh
next prev parent reply other threads:[~2024-09-04 16:44 UTC|newest]
Thread overview: 94+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-03 3:59 [RFC 00/31] objtool, livepatch: Livepatch module generation Josh Poimboeuf
2024-09-03 3:59 ` [RFC 01/31] x86/alternative: Refactor INT3 call emulation selftest Josh Poimboeuf
2024-09-03 3:59 ` [RFC 02/31] x86/module: Improve relocation error messages Josh Poimboeuf
2024-09-03 3:59 ` [RFC 03/31] x86/kprobes: Remove STACK_FRAME_NON_STANDARD annotation Josh Poimboeuf
2024-09-03 3:59 ` [RFC 04/31] kernel/sys: Don't reference UTS_RELEASE directly Josh Poimboeuf
2024-09-03 3:59 ` [RFC 05/31] x86/compiler: Tweak __UNIQUE_ID naming Josh Poimboeuf
2024-09-03 7:56 ` Peter Zijlstra
2024-09-04 2:01 ` Josh Poimboeuf
2024-09-08 19:43 ` David Laight
2024-09-03 3:59 ` [RFC 06/31] elfnote: Use __UNIQUE_ID() for note symbols Josh Poimboeuf
2024-09-03 3:59 ` [RFC 07/31] kbuild: Remove "kmod" prefix from __KBUILD_MODNAME Josh Poimboeuf
2024-09-03 7:58 ` Peter Zijlstra
2024-09-04 2:11 ` Josh Poimboeuf
2024-09-04 7:53 ` Peter Zijlstra
2024-09-03 3:59 ` [RFC 08/31] objtool: Remove .parainstructions reference Josh Poimboeuf
2024-09-03 3:59 ` [RFC 09/31] objtool: Const string cleanup Josh Poimboeuf
2024-09-03 3:59 ` [RFC 10/31] objtool: Use 'struct elf' in elf macros Josh Poimboeuf
2024-09-03 3:59 ` [RFC 11/31] objtool: Add section/symbol type helpers Josh Poimboeuf
2024-09-03 3:59 ` [RFC 12/31] objtool: 'objname' refactoring Josh Poimboeuf
2024-09-03 3:59 ` [RFC 13/31] objtool: Support references to all symbol types in special sections Josh Poimboeuf
2024-09-03 3:59 ` [RFC 14/31] objtool: Refactor add_jump_destinations() Josh Poimboeuf
2024-09-03 3:59 ` [RFC 15/31] objtool: Interval tree cleanups Josh Poimboeuf
2024-09-03 3:59 ` [RFC 16/31] objtool: Simplify fatal error handling Josh Poimboeuf
2024-09-03 4:00 ` [RFC 17/31] objtool: Open up the elf API Josh Poimboeuf
2024-09-03 4:00 ` [RFC 18/31] objtool: Disallow duplicate prefix symbols Josh Poimboeuf
2024-09-03 4:00 ` [RFC 19/31] objtool: Add elf_create_file() Josh Poimboeuf
2024-09-03 4:00 ` [RFC 20/31] objtool: Add UD1 detection Josh Poimboeuf
2024-09-03 8:17 ` Peter Zijlstra
2024-09-04 2:25 ` Josh Poimboeuf
2024-09-03 4:00 ` [RFC 21/31] objtool: Fix x86 addend calcuation Josh Poimboeuf
2024-09-04 9:24 ` laokz
2024-09-04 16:15 ` Josh Poimboeuf
2024-09-03 4:00 ` [RFC 22/31] objtool: Make find_symbol_containing() less arbitrary Josh Poimboeuf
2024-09-03 4:00 ` [RFC 23/31] objtool: Handle __pa_symbol() relocations Josh Poimboeuf
2024-09-03 4:00 ` [RFC 24/31] objtool: Make STACK_FRAME_NON_STANDARD consistent Josh Poimboeuf
2024-09-03 4:00 ` [RFC 25/31] objtool: Fix interval tree insertion for zero-length symbols Josh Poimboeuf
2024-09-03 4:00 ` [RFC 26/31] objtool: Make interval tree functions "static inline" Josh Poimboeuf
2024-09-03 4:00 ` [RFC 27/31] objtool: Fix weak symbol detection Josh Poimboeuf
2024-09-03 8:26 ` Peter Zijlstra
2024-09-04 3:55 ` Josh Poimboeuf
2024-09-04 7:42 ` Peter Zijlstra
2024-09-04 16:03 ` Josh Poimboeuf
2024-09-03 4:00 ` [RFC 28/31] x86/alternative: Create symbols for special section entries Josh Poimboeuf
2024-09-03 8:29 ` Peter Zijlstra
2024-09-04 4:28 ` Josh Poimboeuf
2024-09-04 8:08 ` Peter Zijlstra
2024-09-04 16:13 ` Josh Poimboeuf
2024-09-04 12:39 ` Borislav Petkov
2024-09-04 16:44 ` Josh Poimboeuf [this message]
2024-09-06 10:19 ` Borislav Petkov
2024-09-06 16:53 ` Josh Poimboeuf
2024-09-06 6:51 ` [RFC 28/31] x86/alternative: Create symbols for special section entrie Weinan Liu
2024-09-07 6:28 ` Josh Poimboeuf
2024-09-03 4:00 ` [RFC 29/31] objtool: Calculate function checksums Josh Poimboeuf
2024-09-04 7:54 ` Peter Zijlstra
2024-09-04 16:11 ` Josh Poimboeuf
2024-09-03 4:00 ` [RFC 30/31] livepatch: Enable -ffunction-sections -fdata-sections Josh Poimboeuf
2024-09-03 4:00 ` [RFC 31/31] objtool, livepatch: Livepatch module generation Josh Poimboeuf
2024-09-04 21:38 ` Jeff Johnson
2024-09-05 4:15 ` Josh Poimboeuf
2024-09-12 2:39 ` laokz
2024-09-03 17:32 ` [RFC 00/31] " Song Liu
2024-09-04 4:30 ` Josh Poimboeuf
2024-09-04 5:26 ` Song Liu
2024-09-04 6:37 ` Josh Poimboeuf
2024-09-04 7:09 ` Josh Poimboeuf
2024-09-04 20:23 ` Song Liu
2024-09-04 20:59 ` Josh Poimboeuf
2024-09-04 21:32 ` Song Liu
2024-09-05 4:13 ` Josh Poimboeuf
2024-09-05 7:13 ` Josh Poimboeuf
2024-09-05 21:34 ` Song Liu
2024-09-07 6:46 ` Josh Poimboeuf
2024-09-07 17:43 ` Song Liu
2024-09-07 20:14 ` Josh Poimboeuf
2024-09-08 5:04 ` Song Liu
2024-09-09 21:19 ` Josh Poimboeuf
2024-09-09 21:43 ` Song Liu
2024-09-06 13:56 ` Joe Lawrence
2024-09-06 17:00 ` Josh Poimboeuf
2024-09-06 21:01 ` Joe Lawrence
2024-09-06 22:45 ` Josh Poimboeuf
2024-09-07 1:47 ` Josh Poimboeuf
2024-09-07 14:17 ` Joe Lawrence
2024-09-11 7:39 ` Josh Poimboeuf
2024-09-12 13:44 ` Joe Lawrence
2024-09-13 14:39 ` Joe Lawrence
2024-09-13 23:09 ` Josh Poimboeuf
2024-09-11 13:27 ` Petr Mladek
2024-09-11 16:20 ` Josh Poimboeuf
2024-09-12 16:05 ` Song Liu
2024-09-13 18:16 ` [External] " A K M Fazla Mehrab .
2024-09-17 7:12 ` Petr Mladek
2024-09-23 2:29 ` Chen Zhongjin
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=20240904164429.hstbg5beejt32mlu@treble \
--to=jpoimboe@kernel.org \
--cc=bp@alien8.de \
--cc=jikos@kernel.org \
--cc=joe.lawrence@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=live-patching@vger.kernel.org \
--cc=mbenes@suse.cz \
--cc=mpdesouza@suse.com \
--cc=peterz@infradead.org \
--cc=pmladek@suse.com \
--cc=song@kernel.org \
--cc=x86@kernel.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