From: Jules Maselbas <jmaselbas@zdiv.net>
To: Ahmad Fatoum <a.fatoum@pengutronix.de>
Cc: barebox@lists.infradead.org
Subject: Re: [PATCH 2/3] scripts: Select XZ --arm64 BCJ filter for 64-bit arm
Date: Thu, 7 Sep 2023 11:48:10 +0200 [thread overview]
Message-ID: <ZPmcWuI9wqCPfCpK@tour> (raw)
In-Reply-To: <fb906f30-4765-3c9e-b89f-92ae93ff59cc@pengutronix.de>
On Thu, Sep 07, 2023 at 10:56:21AM +0200, Ahmad Fatoum wrote:
> On 06.09.23 17:08, Jules Maselbas wrote:
> > This change will select the arm64 BCJ filter for aarch64 instead of
> > the arm filter which was always selected for arm based architectures.
> >
> > The selected filter is also tested, by looking if it is present in
> > the long help message of xz, this is because the arm64 BCJ filter is
> > a recent addition to xz.
> >
> > Signed-off-by: Jules Maselbas <jmaselbas@zdiv.net>
>
> How much of an improvement is BCJ=--arm over BCJ=""?
In my case using the --arm BCJ on a aarch64 barebox is slightly worst
than no BCJ at all.
$ xz -vkf --check=crc32 --lzma2=dict=32MiB barebox.bin
100 % 84.7 KiB / 236.9 KiB = 0.357
$ xz -vkf --check=crc32 --arm --lzma2=dict=32MiB barebox.bin
100 % 85.5 KiB / 236.9 KiB = 0.361
$ xz -vkf --check=crc32 --arm64 --lzma2=dict=32MiB barebox.bin
100 % 81.0 KiB / 236.9 KiB = 0.342
> I wonder if we should fallback from --arm64 to --arm.
>
> Cheers,
> Ahmad
>
> > ---
> > scripts/xz_wrap.sh | 5 ++++-
> > 1 file changed, 4 insertions(+), 1 deletion(-)
> >
> > diff --git a/scripts/xz_wrap.sh b/scripts/xz_wrap.sh
> > index 7a2d372f48..5b5f3adcff 100755
> > --- a/scripts/xz_wrap.sh
> > +++ b/scripts/xz_wrap.sh
> > @@ -16,8 +16,11 @@ case $SRCARCH in
> > x86) BCJ=--x86 ;;
> > powerpc) BCJ=--powerpc ;;
> > ia64) BCJ=--ia64; LZMA2OPTS=pb=4 ;;
> > - arm) BCJ=--arm ;;
> > + arm) BCJ=--arm$S64 ;;
> > sparc) BCJ=--sparc ;;
> > esac
> >
> > +# clear BCJ filter if unsupported
> > +xz -H | grep -q -- $BCJ || BCJ=
> > +
> > exec xz --check=crc32 $BCJ --lzma2=$LZMA2OPTS,dict=32MiB
>
> --
> Pengutronix e.K. | |
> Steuerwalder Str. 21 | http://www.pengutronix.de/ |
> 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
> Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
>
next prev parent reply other threads:[~2023-09-07 9:49 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-06 15:08 [PATCH 1/3] decompressors: Update xz to include ARM64 BCJ decoder Jules Maselbas
2023-09-06 15:08 ` [PATCH 2/3] scripts: Select XZ --arm64 BCJ filter for 64-bit arm Jules Maselbas
2023-09-07 8:56 ` Ahmad Fatoum
2023-09-07 9:48 ` Jules Maselbas [this message]
2023-09-07 9:51 ` Ahmad Fatoum
2023-09-06 15:08 ` [PATCH 3/3] kbuild: remove duplicated xz compressions target Jules Maselbas
2023-09-07 8:53 ` Ahmad Fatoum
2023-09-07 9:00 ` [PATCH 1/3] decompressors: Update xz to include ARM64 BCJ decoder Ahmad Fatoum
2023-09-07 9:57 ` Jules Maselbas
2023-09-07 10:08 ` Ahmad Fatoum
2023-09-07 10:42 ` Jules Maselbas
2023-09-07 10:46 ` Ahmad Fatoum
2023-09-07 10:51 ` [PATCH] fixup! " Jules Maselbas
2023-09-07 13:37 ` Jules Maselbas
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=ZPmcWuI9wqCPfCpK@tour \
--to=jmaselbas@zdiv.net \
--cc=a.fatoum@pengutronix.de \
--cc=barebox@lists.infradead.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