All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Cédric Le Goater" <clg@kaod.org>
To: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Gerome Burlats <gerome.burlats@smile.fr>,
	Romain Naour <romain.naour@gmail.com>,
	buildroot@buildroot.org
Subject: Re: [Buildroot] [PATCH 1/3] configs/qemu: Add a ppc64le-powernv8 configuration
Date: Wed, 12 Jan 2022 08:07:40 +0100	[thread overview]
Message-ID: <09709dad-59ac-23b6-e3f0-3ec3457c7a14@kaod.org> (raw)
In-Reply-To: <20220111214509.79c5c88e@windsurf>

Hello Thomas,

> Thanks, entire patch series applied. One small change was done: add
> appropriate entries to the DEVELOPERS file, so that we know who to
> contact for those defconfigs :-) Especially if our Gitlab CI
> infrastructure detects build issues, you will be notified.

sure. I have a generated set here :

   https://github.com/legoater/qemu-ppc-boot/tree/main/buildroot

that I use for non regression on QEMU PPC.

> Since you are a PowerPC expert, I'd like to take this opportunity to
> ask you about the toolchains we're providing at
> https://toolchains.bootlin.com/. For the moment for PowerPC, we are
> providing the following toolchains:
> 
> powerpc-440fp, not tested

The bamboo machine uses a 440EP CPU implementation, which has a
floating point unit.
     
> powerpc64-e5500, tested with qemu_ppc64_e5500_defconfig
> powerpc64-e6500, not tested

e6500 is nearly the same CPU as e5500. AFAIK, the main differences
are altivec instructions and more HV support for guest translations.

More info here :

   https://www.linux-kvm.org/page/E500_virtual_CPU_specification

> powerpc64le-power8, tested with qemu_ppc64le_pseries_defconfig
> powerpc64-power8, tested with qemu_ppc64_pseries_defconfig
> powerpc-e300c3, not tested
> powerpc-e500mc, not tested

you could use the 32bit ppce500 machine for that.

> I suppose I could now test the powerpc-440fp toolchain with the
> qemu_ppc_bamboo_defconfig you have added. Is that correct?

yes.
  
> What about the other toolchain configurations? Is there some matching
> Qemu emulation?>
> And generally speaking, does this selection of PowerPC architecture
> variants make sense to offer pre-compiled toolchains? For quite a

Here is a quick status of machine/CPUs that QEMU implements :

* prep

   prep: 604 CPU (we might remove 601-603 CPUs)

* embedded

   ref405ep: 405EP (user space segfaults but same image works on 44x CPUs
                    I hope to get that one fixed by QEMU 7.0 )
   bambo:    440FP
   sam460ex: 460EP is very much like the 440fp

* Mac

   g3beige:  G3 (or 75x CPU) family
   mac99:    G4 (7400 7410 7441 7447 7448 7450 7451 7455 7457) family

* book3e 32bit:

   ppce500:   e500mc
   mpc8544ds: mpc8544

* book3e 64bit:

   ppce500 (KVM): e5500, e6500
   
* book3s 64bit (para virtualized and baremetal):

   mac99: PPC970
   pseries (BE only): PPC970, PPC970mp, POWER5+
   pseries (LE, BE and KVM) : POWER7, POWER8, POWER9, POWER10
   powernv: POWER8, POWER9, POWER10

Most of the attention is on book3s 64bit CPUs. MAC emulation is well
maintained. Embedded world is not.

> while, we've had requests for other PowerPC variants:
> https://github.com/bootlin/toolchains-builder/issues/2. Your input
> would be very welcome :-)

OK. I will take a look.

Thanks,

C.
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

  reply	other threads:[~2022-01-12  7:23 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-10 21:02 [Buildroot] [PATCH 1/3] configs/qemu: Add a ppc64le-powernv8 configuration Cédric Le Goater
2022-01-10 21:02 ` [Buildroot] [PATCH 2/3] configs/qemu: Add a ppc-bamboo configuration Cédric Le Goater
2022-01-10 21:04 ` [Buildroot] [PATCH 1/3] configs/qemu: Add a ppc64le-powernv8 configuration Cédric Le Goater
2022-01-11 20:45 ` Thomas Petazzoni
2022-01-12  7:07   ` Cédric Le Goater [this message]
  -- strict thread matches above, loose matches on Subject: below --
2022-01-10 20:58 Cédric Le Goater

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=09709dad-59ac-23b6-e3f0-3ec3457c7a14@kaod.org \
    --to=clg@kaod.org \
    --cc=buildroot@buildroot.org \
    --cc=gerome.burlats@smile.fr \
    --cc=romain.naour@gmail.com \
    --cc=thomas.petazzoni@bootlin.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.