linux-kernel - Question about cacheline aligned memory for DMA transfers

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-next>] [day] [month] [year] [list]

Message-ID: <530602563.487644.1430912396946.JavaMail.open-xchange@oxbaltgw03.schlund.de>
Date:	Wed, 6 May 2015 13:39:56 +0200 (CEST)
From:	Johannes Thoma <johannes@...annesthoma.com>
To:	linux-kernel@...r.kernel.org
Subject: Question about cacheline aligned memory for DMA transfers

Dear list,

I am working on a memory corruption problem with a patched 2.6.21 kernel in the
USB driver. In order to better understand the problem I would like to ask how
current kernels handle the case where the same cacheline is used for a DMA-able
object and another object.

Does kmalloc return only memory that is cache line aligned? If not, do all
architectures handle cache line misalign
ed dma accesses correctly like the powerpc (although not yet in 2.6.21)
architecture does?

Thanks for any insights.

- Johannes

PS: Please CC me I am not on the list.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/