On Mon, May 25, 2015 at 12:25:19PM +0200, Marek Vasut wrote: > On Monday, May 25, 2015 at 11:51:04 AM, Josh Wu wrote: > > Hi, Marek > > Hi! > > > On 5/22/2015 9:30 AM, Marek Vasut wrote: > > > On Friday, May 22, 2015 at 02:34:25 AM, Tom Rini wrote: > > >> On Fri, May 22, 2015 at 01:54:55AM +0200, Marek Vasut wrote: > > >>> On Wednesday, May 20, 2015 at 01:24:06 PM, Tom Rini wrote: > > >>>> On Tue, May 19, 2015 at 07:58:26PM +0200, Marek Vasut wrote: > > >>>>> Updated PR, dropped the patch from Hans and picked the OHCI cache > > >>>>> patch instead: > > >>>>> > > >>>>> The following changes since commit > > > > > > 0e6b7a28243175ae0874d53b6e6e4eff8548d71f: > > >>>>> Merge git://git.denx.de/u-boot-samsung (2015-05-18 09:15:15 -0400) > > >>>>> > > >>>>> are available in the git repository at: > > >>>>> git://git.denx.de/u-boot-usb.git HEAD > > >>>>> > > >>>>> for you to fetch changes up to > > >>>>> d07e7c0b5bd9dc0044eafc730131d32539608303: > > >>>>> usb: ohci: enable cache support (2015-05-19 19:57:30 +0200) > > >>>> > > >>>> Sigh, NAK still. Many PowerPC are failing over flush_dcache_range > > >>>> (which should exist for them now..), > > >>> > > >>> Which ones please? > > >> > > >> M5485BFE,lp8x4x,vpac270_nor_128,colibri_pxa270,eb_cpux9k2,M5485AFE,at91r > > >> m92 > > >> 00ek_ram,M5475EFE,M5475CFE,M5485HFE,polaris,M5485DFE,M5475DFE,vpac270_o > > >> nd_2 > > >> 56,vpac270_nor_256,at91rm9200ek,M5485CFE,M5475FFE,M5475BFE,M5485GFE,M54 > > >> 85EF > > >> E,balloon3,M5485FFE,eb_cpux9k2_ram,M5475GFE,M5475AFE,trizepsiv,edb9315a > > >> > > >> Which now that I look at it is not a list of PowerPC boards (I have to > > >> admit my eyes glazed over at M... and I filed that as PowerPC) but m68k > > >> and some ARM. > > > > > > Yeah. Josh, these boards were broken by your patch > > > "usb: ohci: enable cache support" . I am dropping > > > it now, so can you please fix the patch and resubmit? > > > > Ok, After a check it seems those cpus doesn't define the > > flush_dcache_range()/invalidate_dcache_range() functions. I think I need > > to find a way to distinguish those cpus. > > Better yet, make those few CPUs define those functions as empty.
... if that in fact makes sense :) -- Tom
signature.asc
Description: Digital signature
_______________________________________________ U-Boot mailing list [email protected] http://lists.denx.de/mailman/listinfo/u-boot
