From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754120Ab0DIPPq (ORCPT ); Fri, 9 Apr 2010 11:15:46 -0400 Received: from iolanthe.rowland.org ([192.131.102.54]:44956 "HELO iolanthe.rowland.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1752953Ab0DIPPp (ORCPT ); Fri, 9 Apr 2010 11:15:45 -0400 Date: Fri, 9 Apr 2010 11:15:43 -0400 (EDT) From: Alan Stern X-X-Sender: stern@iolanthe.rowland.org To: Oliver Neukum cc: Daniel Mack , , Pedro Ribeiro , , Greg KH , , Subject: Re: USB transfer_buffer allocations on 64bit systems In-Reply-To: <201004091650.31488.oliver@neukum.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 9 Apr 2010, Oliver Neukum wrote: > Am Freitag, 9. April 2010 16:41:48 schrieb Alan Stern: > > > It'll work on x86. On incoherent architectures this violates the cacheline > > > rules for DMA-mapping if you have to bounce. > > > > Not true. Consider: The driver allocates a 16-byte buffer (xbuf) > > divided up into four sets of four bytes, and sets > > > > urb[i].transfer_buffer_dma = xbuf_dma + 4*i; > > > > Then usb_submit_urb(urb[i]) will copy the appropriate four bytes to a > > bounce buffer and map the bounce buffer. Accesses to the other parts > > of xbuf won't violate the cacheline rules, because xbuf isn't mapped > > for DMA -- only the bounce buffer is. When urb[i] completes, the > > bounce buffer contents will be copied back to the original four bytes > > in xbuf. Again, there is no violation of cacheline rules. > > I think you are assuming that either every or no part of the buffer is mapped > for DMA in place. I don't think you can assume that. Yes I can, because the code that makes this decision is part of usbcore and it is under my control. Alan Stern From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alan Stern Subject: Re: USB transfer_buffer allocations on 64bit systems Date: Fri, 9 Apr 2010 11:15:43 -0400 (EDT) Message-ID: References: <201004091650.31488.oliver@neukum.org> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-path: In-Reply-To: <201004091650.31488.oliver-GvhC2dPhHPQdnm+yROfE0A@public.gmane.org> Sender: linux-usb-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Oliver Neukum Cc: Daniel Mack , linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Pedro Ribeiro , akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org, Greg KH , alsa-devel-K7yf7f+aM1XWsZ/bQMPhNw@public.gmane.org, linux-usb-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: alsa-devel@alsa-project.org On Fri, 9 Apr 2010, Oliver Neukum wrote: > Am Freitag, 9. April 2010 16:41:48 schrieb Alan Stern: > > > It'll work on x86. On incoherent architectures this violates the cacheline > > > rules for DMA-mapping if you have to bounce. > > > > Not true. Consider: The driver allocates a 16-byte buffer (xbuf) > > divided up into four sets of four bytes, and sets > > > > urb[i].transfer_buffer_dma = xbuf_dma + 4*i; > > > > Then usb_submit_urb(urb[i]) will copy the appropriate four bytes to a > > bounce buffer and map the bounce buffer. Accesses to the other parts > > of xbuf won't violate the cacheline rules, because xbuf isn't mapped > > for DMA -- only the bounce buffer is. When urb[i] completes, the > > bounce buffer contents will be copied back to the original four bytes > > in xbuf. Again, there is no violation of cacheline rules. > > I think you are assuming that either every or no part of the buffer is mapped > for DMA in place. I don't think you can assume that. Yes I can, because the code that makes this decision is part of usbcore and it is under my control. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html