<HTML dir=ltr><HEAD><TITLE>RE: AW: PowerPC PCI DMA issues (prefetch/coherency?)</TITLE>
<META http-equiv=Content-Type content="text/html; charset=unicode">
<META content="MSHTML 6.00.2900.3527" name=GENERATOR></HEAD>
<BODY>
<DIV id=idOWAReplyText94079 dir=ltr>
<DIV dir=ltr><FONT face=Arial color=#000000 size=2><FONT face="Times New Roman">Benjamin Herrenschmidt wrote:</FONT></FONT></DIV>
<DIV dir=ltr><FONT face=Arial color=#000000 size=2><FONT face="Times New Roman">> Do you know many drivers that do config space accesses without using<BR>> the config space accessors ?<BR><BR>> Such drivers should be banned to oblivion.<BR><BR>> Cheers,<BR>> Ben.</FONT><BR></FONT></DIV>
<DIV dir=ltr><FONT face=Arial color=#000000 size=2>I'm not aware of such drivers in the 2.6.30+ kernel.</FONT></DIV>
<DIV dir=ltr><FONT face=Arial size=2></FONT> </DIV>
<DIV dir=ltr><FONT face=Arial size=2>Thanks,</FONT></DIV>
<DIV dir=ltr><FONT face=Arial size=2>Pravin</FONT></DIV></DIV>
<DIV>
<P><FONT size=2>On Thu, 2009-09-10 at 22:35 -0700, Pravin Bathija wrote:<BR>> Thanks Stefan. The whole intention of the patch/hack (or whatever one<BR>> might call it :) ) was to avoid rogue drivers from setting<BR>> pci_cache_line_size to non-zero value even though the underlying<BR>> hardware doesn't support MRM calls. Nonetheless this approach works<BR>> only if the drivers use the kernel API for PCI config space access<BR>> provided by the powerpc platform driver.<BR>><BR><BR></FONT></P></DIV></BODY></HTML>