linux-kernel - Re: [PATCH] m68k/q40: Replace q40ide with pata

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] [day] [month] [year] [list]

Message-ID: <3b789204-a959-a266-9fc3-1b2be09c981a@gmail.com>
Date:   Tue, 8 Jun 2021 15:01:30 +1200
From:   Michael Schmitz <schmitzmic@...il.com>
To:     Finn Thain <fthain@...ux-m68k.org>
Cc:     Geert Uytterhoeven <geert@...ux-m68k.org>,
        Richard Zidlicky <rz@...ux-m68k.org>,
        Jens Axboe <axboe@...nel.dk>,
        "David S. Miller" <davem@...emloft.net>,
        linux-m68k@...ts.linux-m68k.org, linux-kernel@...r.kernel.org,
        linux-ide@...r.kernel.org
Subject: Re: [PATCH] m68k/q40: Replace q40ide with pata_falcon and falconide

Hi Finn,



Am 07.06.2021 um 14:07 schrieb Finn Thain:
> On Mon, 7 Jun 2021, Michael Schmitz wrote:
>
>> Hi Finn,
>>
>> are there any substantial changes from what I'd tested before?
>
> The request_region() stuff is new, and I reworked the host->get_lock and
> host->release_lock changes. So you may want to review it again.

Done that - the weirdness around request_irq() that I remembered is not 
actually an issue, that's taken care of by setting host->get_lock. And 
your patch doesn't change anything for Falcon.

Tested on my Falcon - I can still use both IDE and SCSI (no stress test 
done but I doubt it would turn out much different).

Not setting host->get_lock on Q40 will correctly request the IRQ for IDE 
there, the same lowlevel IO routines can be shared by both, and there 
wasn't ever much more that these old drivers did.

Please add my Reviewed-by as well as Tested-by.

Cheers,

	Michael


>> Your patch conflicts with (I think) the Mac IDE patches now in Geert's
>> tree, and I'd rather avoid hand-patching the lot if at all possible:
>>
>> m68k_q40_Replace_q40ide_with_pata_falcon_and_falconide
>> Applying: m68k/q40: Replace q40ide with pata_falcon and falconide
>> error: patch failed: arch/m68k/configs/multi_defconfig:351
>> error: arch/m68k/configs/multi_defconfig: patch does not apply
>> error: patch failed: drivers/ide/Kconfig:731
>> error: drivers/ide/Kconfig: patch does not apply
>> error: patch failed: drivers/ide/Makefile:29
>> error: drivers/ide/Makefile: patch does not apply
>>
>
> I should have sent these as a two-part series. Sorry about that.
> You'll need to apply this patch first:
> https://lore.kernel.org/linux-ide/3de9a93d-2f7-a650-1fa-c2129a4a765b@nippy.intranet/T/
>
>> What's your baseline for this patch?
>>
>
> I've used v5.12 and v5.13-rc5, but mostly I develop on the mac68k branch
> in my github repository.
>
>> (In the alternative: just add my Tested-by if nothing changed aside from the
>> IO region reservation)
>>
>
> I have tested these patches in Aranym (including a multi-platform
> atari/q40 build which required your ISA fix) but I don't have any way to
> test the q40-only code.
>