linux-kernel - Re: [PATCH 1/1] Userspace I/O (UIO): Add support for userspace DMA

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: <208aa0f00812031844u779584e8q633a5f1b92a97426@mail.gmail.com>
Date:	Wed, 3 Dec 2008 18:44:01 -0800
From:	"Edward Estabrook" <edward.estabrook.lkml@...il.com>
To:	"Leon Woestenberg" <leon.woestenberg@...il.com>
Cc:	edward_estabrook@...lent.com, linux-kernel@...r.kernel.org,
	hjk@...utronix.de, gregkh@...e.de, edward.estabrook@...il.com
Subject: Re: [PATCH 1/1] Userspace I/O (UIO): Add support for userspace DMA

> You are using a magic value to mmap(), just to keep options open:
>
> Would this in the future allow us to also add support for streaming
> non-coherent DMA directly into user space buffers?

Yes, but additional userspace control is also required.
Specifically:
 1) The specification of direction at the time of buffer creation,
    unless force DMA_BIDIRECTIONAL is assumed.
 2) An API to faciliate userspace control over dma_unmap_page() /
    dma_sync_single_for_cpu() calls (perhaps by further overloading
    of the write method).

The use of this magic value does not prevent the future addition of
streaming DMA support, but does not provide sufficient controls.
--
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/