linux-kernel - Re: [RFC PATCH 0/3] ATA: Replace deprecated PCI functions

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 for Android: free password hash cracker in your pocket

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

Message-ID: <973b3b99-77eb-4a27-9a4a-8547d2a2cc76@kernel.org>
Date: Mon, 9 Dec 2024 10:14:47 +0900
From: Damien Le Moal <dlemoal@...nel.org>
To: Philipp Stanner <pstanner@...hat.com>, Niklas Cassel <cassel@...nel.org>,
 Mikael Pettersson <mikpelinux@...il.com>
Cc: linux-ide@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH 0/3] ATA: Replace deprecated PCI functions

On 12/5/24 02:10, Philipp Stanner wrote:
> Hi,
> 
> many of you probably know that I'm trying to remove pcim_iomap_regions()
> from the kernel. One of the more difficult users is ATA, because it's
> the only subsystem I've seen so far that accesses that table
> pcim_iomap_table() administrates.
> 
> This series only builds as a whole because of patch 1. That's why I
> submit it as an RFC.
> 
> I want to know whether you agree with the basic idea, and whether your
> subsystem wants this series to be squashed into a single commit that
> builds.

Overall, looks OK to me. But I think that at least patches 1 and 2 need to be
squashed together. For patch 3, if you can have a minimal "make it compile"
change in patch 2, you can keep the cleanup separate. Otherwise, squashing all
patches together seems fine too.

> Another solution would be to provide a struct ata_host.iomap2 or
> something like that, phase out the pcim_iomap_regions() users, and then
> remove iomap2 again.

That seems unnecessary.

> 
> Please tell me your preferred way.
> 
> (This is the revived version of an old series from August. In case
> someone is wondering)
> 
> Thx,
> P.
> 
> Philipp Stanner (3):
>   ata: Allocate PCI iomap table statically
>   ata: Replace deprecated PCI functions
>   libata-sff: Simplify request of PCI resources
> 
>  drivers/ata/ata_piix.c      |   7 +-
>  drivers/ata/libata-sff.c    | 130 +++++++++++++++++++++++-------------
>  drivers/ata/pata_atp867x.c  |  13 ++--
>  drivers/ata/pata_hpt3x3.c   |  10 +--
>  drivers/ata/pata_ninja32.c  |  11 +--
>  drivers/ata/pata_pdc2027x.c |  11 ++-
>  drivers/ata/pata_sil680.c   |  12 ++--
>  drivers/ata/pdc_adma.c      |   9 ++-
>  drivers/ata/sata_inic162x.c |  10 ++-
>  drivers/ata/sata_mv.c       |   9 +--
>  drivers/ata/sata_nv.c       |   8 +--
>  drivers/ata/sata_promise.c  |   8 ++-
>  drivers/ata/sata_qstor.c    |   7 +-
>  drivers/ata/sata_sil.c      |   8 ++-
>  drivers/ata/sata_sil24.c    |  20 +++---
>  drivers/ata/sata_sis.c      |   8 +--
>  drivers/ata/sata_svw.c      |  10 +--
>  drivers/ata/sata_sx4.c      |  19 +++++-
>  drivers/ata/sata_via.c      |  31 +++++----
>  drivers/ata/sata_vsc.c      |   8 ++-
>  include/linux/libata.h      |   7 +-
>  21 files changed, 216 insertions(+), 140 deletions(-)
> 


-- 
Damien Le Moal
Western Digital Research