From: Marek Behun <marek.behun@nic.cz>
To: u-boot@lists.denx.de
Subject: [RFC PATCH u-boot 01/12] build: use thin archives instead of incremental linking
Date: Fri, 5 Mar 2021 16:39:27 +0100 [thread overview]
Message-ID: <20210305163927.46685590@nic.cz> (raw)
In-Reply-To: <CAEUhbmUjW9HnBiUaaz8XyGbWuvUKo04dKZZ8ZC-oJMd_q8BChA@mail.gmail.com>
On Fri, 5 Mar 2021 21:34:42 +0800
Bin Meng <bmeng.cn@gmail.com> wrote:
> Hi Marek,
>
> On Fri, Mar 5, 2021 at 2:17 AM Marek Behun <marek.behun@nic.cz> wrote:
> >
> > On Thu, 4 Mar 2021 18:57:11 +0800
> > Bin Meng <bmeng.cn@gmail.com> wrote:
> >
> > > Hi Marek,
> > >
> > > On Wed, Mar 3, 2021 at 12:13 PM Marek Beh?n <marek.behun@nic.cz> wrote:
> > > >
> > > > Using thin archives instead of incremental linking
> > > > - saves disk space
> > > > - works better with dead code elimination
> > > > - prepares for potential LTO
> > >
> > > The commit message is a little bit confusing. This commit actually
> > > does 2 things: don't do incremental linking (using --whole-archive),
> > > and use thin archive (passing T to ar). I believe they are for
> > > different purposes, so we cannot say "using thin archives instead of
> > > incremental linking".
> > > > - -Wl,--start-group $(patsubst $(obj)/%,%,$(u-boot-spl-main)) \
> > > > - $(patsubst $(obj)/%,%,$(u-boot-spl-platdata)) -Wl,--end-group \
> > > > + -Wl,--whole-archive $(patsubst $(obj)/%,%,$(u-boot-spl-main)) -Wl,--no-whole-archive \
> > > > + -Wl,--start-group $(patsubst $(obj)/%,%,$(u-boot-spl-platdata)) -Wl,--end-group \
> > >
> > > u-boot-spl-platdata is still within --start-group, --end-group, is
> > > this intentional?
> >
> > I confess that I did not really study these options, I have made these
> > changes according to old LTO patches for Linux. But you are right that
> > it does not make sense. I have fixed this for the next version of this
> > patch.
> >
> > > Is P required to make everything work?
> >
> > It is not. Removed in next version.
>
> I did more investigation on this.
>
> The Linux kernel specially added P to ar, in below commit:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=9a6cfca4f4130444
>
> So it looks like we should keep P here?
>
> But I don't get the point of switching to thin archives. Based on my
> experiment, LTO does not rely on thin archives. The Linux kernel did
> not introduce thin archives for LTO.
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=a5967db9af51a84f
It does not matter whether we use thin archives or real archives. But
we did not use any of this before, instead we linked the object files
in a directory into one object file in that directory. And to do this
with LTO would cause unnecessary complications.
Marek
next prev parent reply other threads:[~2021-03-05 15:39 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-03 4:11 [RFC PATCH u-boot 00/12] U-Boot LTO (Sandbox + ARM Nokia RX-51) Marek Behún
2021-03-03 4:12 ` [RFC PATCH u-boot 01/12] build: use thin archives instead of incremental linking Marek Behún
2021-03-04 10:57 ` Bin Meng
2021-03-04 18:17 ` Marek Behun
2021-03-05 13:34 ` Bin Meng
2021-03-05 13:37 ` Tom Rini
2021-03-06 21:20 ` Marek Behun
2021-03-05 15:39 ` Marek Behun [this message]
2021-03-03 4:12 ` [RFC PATCH u-boot 02/12] sandbox: errno: avoid conflict with libc's errno Marek Behún
2021-03-05 3:00 ` Bin Meng
2021-03-05 15:37 ` Marek Behun
2021-03-05 16:39 ` Simon Glass
2021-03-05 16:50 ` Marek Behun
2021-03-05 16:58 ` Simon Glass
2021-03-05 17:24 ` Heinrich Schuchardt
2021-03-05 17:24 ` Marek Behun
2021-03-05 19:42 ` Simon Glass
2021-03-05 17:21 ` Heinrich Schuchardt
2021-03-07 3:14 ` Marek Behun
2021-03-12 4:45 ` Simon Glass
2021-03-03 4:12 ` [RFC PATCH u-boot 03/12] linker_lists: declare entries and lists externally visible Marek Behún
2021-03-04 13:47 ` Marek Behun
2021-03-05 3:04 ` Bin Meng
2021-03-05 15:49 ` Marek Behun
2021-03-03 4:12 ` [RFC PATCH u-boot 04/12] efi_loader: fix warning when linking with LTO Marek Behún
2021-03-05 5:50 ` Bin Meng
2021-03-03 4:12 ` [RFC PATCH u-boot 05/12] binman: declare symbols externally visible Marek Behún
2021-03-04 0:37 ` Tom Rini
2021-03-04 1:53 ` Marek Behun
2021-03-03 4:12 ` [RFC PATCH u-boot 06/12] build: support building with Link Time Optimizations Marek Behún
2021-03-05 6:34 ` Bin Meng
2021-03-03 4:12 ` [RFC PATCH u-boot 07/12] arch: sandbox: make LTO available Marek Behún
2021-03-03 4:12 ` [RFC PATCH u-boot 08/12] sandbox: build with LTO Marek Behún
2021-03-03 4:12 ` [RFC PATCH u-boot 09/12] ARM: make gd a function for LTO Marek Behún
2021-03-03 4:12 ` [RFC PATCH u-boot 10/12] string: make memcpy() visible to fix LTO linking errors Marek Behún
2021-03-03 4:12 ` [RFC PATCH u-boot 11/12] arch: ARM: make LTO available Marek Behún
2021-03-03 4:12 ` [RFC PATCH u-boot 12/12] Nokia RX-51: build with LTO Marek Behún
2021-03-03 16:11 ` [RFC PATCH u-boot 00/12] U-Boot LTO (Sandbox + ARM Nokia RX-51) Tom Rini
2021-03-03 16:41 ` Marek Behun
2021-03-03 16:47 ` Tom Rini
2021-03-03 21:36 ` Tom Rini
2021-03-04 0:40 ` Adam Ford
2021-03-04 7:07 ` [PATCH] arm: fix LTO build for some thumb-interwork cases Marek Behún
2021-03-04 10:43 ` [RFC PATCH u-boot 00/12] U-Boot LTO (Sandbox + ARM Nokia RX-51) Marek Behun
2021-03-04 13:46 ` Adam Ford
2021-03-04 13:50 ` Marek Behun
2021-03-04 13:59 ` Adam Ford
2021-03-04 14:58 ` Tom Rini
2021-03-04 15:07 ` Adam Ford
2021-03-04 15:37 ` Marek Behun
2021-03-04 16:18 ` Tom Rini
2021-03-04 15:59 ` Marek Behun
2021-03-04 22:18 ` Adam Ford
2021-03-04 22:33 ` Marek Behun
2021-03-05 11:25 ` Adam Ford
2021-03-05 12:31 ` Stefan Roese
2021-03-05 17:10 ` Adam Ford
2021-03-06 3:03 ` Adam Ford
2021-03-06 11:12 ` Adam Ford
2021-03-06 17:37 ` Marek Behun
2021-03-06 20:08 ` Tom Rini
2021-03-06 20:41 ` Pali Rohár
2021-03-06 20:54 ` Marek Behun
2021-03-06 21:00 ` Pali Rohár
2021-03-06 21:19 ` Marek Behun
2021-03-06 21:38 ` Pali Rohár
2021-03-06 21:49 ` Marek Behun
2021-03-07 3:45 ` Adam Ford
2021-03-07 4:06 ` Marek Behun
2021-05-09 14:14 ` Adam Ford
2021-05-09 18:44 ` Marek Behun
2021-05-10 16:28 ` Simon Glass
2021-05-13 9:22 ` Pali Rohár
2021-03-04 16:17 ` Tom Rini
2021-03-04 16:35 ` Pali Rohár
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=20210305163927.46685590@nic.cz \
--to=marek.behun@nic.cz \
--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