linux-kernel - Re: [PATCH 06/17] pci: Add generic pcibios_{fixup_bus,align

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20170608081228.GA26932@infradead.org>
Date:   Thu, 8 Jun 2017 01:12:28 -0700
From:   Christoph Hellwig <hch@...radead.org>
To:     Geert Uytterhoeven <geert@...ux-m68k.org>
Cc:     Palmer Dabbelt <palmer@...belt.com>,
        Linux-Arch <linux-arch@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Arnd Bergmann <arnd@...db.de>, Olof Johansson <olof@...om.net>,
        albert@...ive.com, patches@...ups.riscv.org,
        Bjorn Helgaas <bhelgaas@...gle.com>,
        linux-pci <linux-pci@...r.kernel.org>
Subject: Re: [PATCH 06/17] pci: Add generic pcibios_{fixup_bus,align_resource}

On Wed, Jun 07, 2017 at 09:19:49AM +0200, Geert Uytterhoeven wrote:
> CC pci folks

Ok, replying with pci folks in Cc then :)

Weak symbols have (rightly) gotten a bad reputation, so maybe
we should approach this without them.

It seems we have a large number of emptry pcibios_fixup_bus calls
alreayd, so I think we should simply have the architectures
that do define it define a Kconfig or header symbol and not call
it at all otherwise.

For the ones that exist as lot just seem to call pci_read_bridge_bases
and/or pcibios_fixup_device_resources in one form or another,
and I wonder why we even need the arch indirection for that.

Similarly for pcibios_align_resource: a lot of architetures seem
to have a noop, and the once that don't mostly seem copy and
paste code, so we should again have a symbol for architectures
to opt into it, and we probably should have a generic helper
for the VGA window mirroring code instead of duplicating it multiple
times.