linux-kernel - RE: [PATCH] hpsa: fix boot on ia64 (atomic

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]

Date:   Wed, 17 Mar 2021 13:19:23 +0000
From:   David Laight <David.Laight@...LAB.COM>
To:     "'Martin K. Petersen'" <martin.petersen@...cle.com>,
        Arnd Bergmann <arnd@...nel.org>
CC:     "Don.Brace@...rochip.com" <Don.Brace@...rochip.com>,
        Sergei Trofimovich <slyfox@...too.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-ia64@...r.kernel.org" <linux-ia64@...r.kernel.org>,
        "storagedev@...rochip.com" <storagedev@...rochip.com>,
        linux-scsi <linux-scsi@...r.kernel.org>,
        "jszczype@...hat.com" <jszczype@...hat.com>,
        "Scott.Benesh@...rochip.com" <Scott.Benesh@...rochip.com>,
        "Scott.Teel@...rochip.com" <Scott.Teel@...rochip.com>,
        "thenzl@...hat.com" <thenzl@...hat.com>,
        "John Paul Adrian Glaubitz" <glaubitz@...sik.fu-berlin.de>
Subject: RE: [PATCH] hpsa: fix boot on ia64 (atomic_t alignment)

From: Martin K. Petersen 
> Sent: 17 March 2021 02:26
> 
> Arnd,
> 
> > Actually that still feels wrong: the annotation of the struct is to
> > pack every member, which causes the access to be done in byte units on
> > architectures that do not have hardware unaligned load/store
> > instructions, at least for things like atomic_read() that does not go
> > through a cmpxchg() or ll/sc cycle.
> 
> > This change may fix itanium, but it's still not correct. Other
> > architectures would have already been broken before the recent change,
> > but that's not a reason against fixing them now.
> 
> I agree. I understand why there are restrictions on fields consumed by
> the hardware. But for fields internal to the driver the packing doesn't
> make sense to me.

Jeepers -- that global #pragma pack(1) is bollocks.

I think there are a couple of __u64 that are 32bit aligned.
Just marking those field __packed __aligned(4) should have
the desired effect.
Or use a typedef for '__u64 with 32bit alignment'.
(There probably ought to be one in types.h)

Then add compile-time asserts that any non-trivial structures
the hardware accesses are the right size.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)