public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
From: Alexandru Gagniuc <mr.nuke.me@gmail.com>
To: u-boot@lists.denx.de
Subject: [PATCH RFC 00/10] image: Reduce the abuse of #ifdefs in image-sig.c
Date: Fri, 14 May 2021 14:45:52 -0500	[thread overview]
Message-ID: <20210514194602.598322-1-mr.nuke.me@gmail.com> (raw)

This series is motivated by Simon's bratwurst:
   [PATCH v2 00/50] image: Reduce #ifdefs and ad-hoc defines in image code


A big problem with current mkimage code, as well as the code it shares
with the targets is that it uses a lot of #ifdefs. Some of the #ifdefs
are defined based on other macros, or CONFIG_() options.

Simon's approach to fixing this is to extend Kconfig to the host-side
of mkimage, and replace ifdefs with "if (CONFIG_IS_ENABLED())"
wherever possible. This would resolve most cosmetic issues caused by
the ungodly abuse of #ifdefs.

I do not like the aforementioned approach, because I believe it is a
band-aid to a much deeper problem. I believe the fundamental problem
is the incorrect separation of target code and host code, which has
led to the accumulation of crud over the years. I believe in
refactoring the current code in order to reduce the need for decision
points and branch divergence between the host and target.

In this series I intend to demonstrate a proof-of-concept for achieving
this with respect to signing algorithms. I treat the three
image_get_*_algo() functions as an interface, and decouple the host and
target implementations. This enable a dramatic reduction of #ifdefs
decision points in image-sig.c

The existing implementation is mostly suited for the host-side, where
it is reused. On the target-side, I implement a linker-list based
array of crypto_algo structures, inspired by the DM driver lists.

Two macros are deleted, rsa.h, and ecdsa.h are completely cleaned of
#ifdefs, and the new host-side implementation of image-sig.c has
zero #ifdefs. This comes at a minimal increase in the noumber of source
lines of code.

Only image_get_crypto_algo() is implemented as a linker list in this
POC. image_get_checksum_algo() and image_get_padding_algo() would also
see healthy benefits.

Alexandru Gagniuc (10):
  common: Move host-only logic in image-sig.c to separate file
  common: image-sig.c: Remove host-specific logic and #ifdefs
  image: Add support for placing crypto_algo in linker lists
  [UNTESTED] image: Add support for relocating crypto_algos in linker
    lists
  image: rsa: Move verification algorithm to a linker list
  image: image-sig.c: Remove crypto_algos array
  lib: ecdsa: Remove #ifdefs from ecdsa.h
  lib: rsa: Remove #ifdefs from rsa.h
  image: Eliminate IMAGE_ENABLE_VERIFY macro
  image: Eliminate IMAGE_ENABLE_VERIFY_ECDSA macro

 common/image-sig-host.c | 134 ++++++++++++++++++++++++++++++++++++++++
 common/image-sig.c      |  71 +++++----------------
 include/image.h         |  13 ++--
 include/u-boot/ecdsa.h  |  25 --------
 include/u-boot/rsa.h    |  47 --------------
 lib/rsa/rsa-verify.c    |  16 +++++
 tools/Makefile          |   2 +-
 7 files changed, 172 insertions(+), 136 deletions(-)
 create mode 100644 common/image-sig-host.c

-- 
2.31.1

             reply	other threads:[~2021-05-14 19:45 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-14 19:45 Alexandru Gagniuc [this message]
2021-05-14 19:45 ` [PATCH RFC 01/10] common: Move host-only logic in image-sig.c to separate file Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass
2021-05-17 14:19     ` Alex G.
2021-05-14 19:45 ` [PATCH RFC 02/10] common: image-sig.c: Remove host-specific logic and #ifdefs Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass
2021-05-14 19:45 ` [PATCH RFC 03/10] image: Add support for placing crypto_algo in linker lists Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass
2021-05-14 19:45 ` [PATCH RFC 04/10] [UNTESTED] image: Add support for relocating crypto_algos " Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass
2021-05-19  7:07     ` Michal Simek
2021-05-14 19:45 ` [PATCH RFC 05/10] image: rsa: Move verification algorithm to a linker list Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass
2021-05-14 19:45 ` [PATCH RFC 06/10] image: image-sig.c: Remove crypto_algos array Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass
2021-05-14 19:45 ` [PATCH RFC 07/10] lib: ecdsa: Remove #ifdefs from ecdsa.h Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass
2021-05-14 19:46 ` [PATCH RFC 08/10] lib: rsa: Remove #ifdefs from rsa.h Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass
2021-05-14 19:46 ` [PATCH RFC 09/10] image: Eliminate IMAGE_ENABLE_VERIFY macro Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass
2021-05-14 19:46 ` [PATCH RFC 10/10] image: Eliminate IMAGE_ENABLE_VERIFY_ECDSA macro Alexandru Gagniuc
2021-05-15 15:20   ` Simon Glass

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=20210514194602.598322-1-mr.nuke.me@gmail.com \
    --to=mr.nuke.me@gmail.com \
    --cc=u-boot@lists.denx.de \
    /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