Re: svn commit: r285270 - head/sys/sys

2015-07-08 Thread Zbigniew Bodek
Hello, You are absolutely right. Especially semicolons... I must have missed this somehow. Thanks. Fixed in r285293. Best regards zbb 2015-07-08 21:47 GMT+02:00 John-Mark Gurney : > Zbigniew Bodek wrote this message on Wed, Jul 08, 2015 at 13:53 +: >> +#if defined(__arm__) >> + #define _

Re: svn commit: r285270 - head/sys/sys

2015-07-08 Thread John-Mark Gurney
Zbigniew Bodek wrote this message on Wed, Jul 08, 2015 at 13:53 +: > +#if defined(__arm__) > + #define __BUS_DMAMAP_SYNC_DEFAULT mb(); > +#elif defined(__aarch64__) > + #define __BUS_DMAMAP_SYNC_DEFAULT dmb(sy); These shouldn't have ; after them... > +#else

Re: svn commit: r285270 - head/sys/sys

2015-07-08 Thread Jason Harmening
+2 This has always bugged me. I'll do it if no one else gets to it first. If the new pmap KPI makes it in, I'll be messing with a bunch of the MD busdma code anyway. On Wed, Jul 8, 2015 at 9:47 AM, Adrian Chadd wrote: > On 8 July 2015 at 07:42, Ian Lepore wrote: > > On Wed, 2015-07-08 at 07:

Re: svn commit: r285270 - head/sys/sys

2015-07-08 Thread Adrian Chadd
On 8 July 2015 at 07:42, Ian Lepore wrote: > On Wed, 2015-07-08 at 07:30 -0700, Adrian Chadd wrote: >> Why is this implemented in sys/sys/bus_dma.h, rather than in a machdep >> header? >> >> > > Indeed, this stuff is a mess that really needs to be cleaned up. The > pre-existing assumption that M

Re: svn commit: r285270 - head/sys/sys

2015-07-08 Thread Ian Lepore
On Wed, 2015-07-08 at 07:30 -0700, Adrian Chadd wrote: > Why is this implemented in sys/sys/bus_dma.h, rather than in a machdep header? > > Indeed, this stuff is a mess that really needs to be cleaned up. The pre-existing assumption that MI functions don't need to be called, just because one of

Re: svn commit: r285270 - head/sys/sys

2015-07-08 Thread Zbigniew Bodek
Because machdep header includes this file (not the other way). Best regards zbb 2015-07-08 16:30 GMT+02:00 Adrian Chadd : > Why is this implemented in sys/sys/bus_dma.h, rather than in a machdep header? > > > -a > > > On 8 July 2015 at 06:53, Zbigniew Bodek wrote: >> Author: zbb >> Date: Wed Jul

Re: svn commit: r285270 - head/sys/sys

2015-07-08 Thread Adrian Chadd
Why is this implemented in sys/sys/bus_dma.h, rather than in a machdep header? -a On 8 July 2015 at 06:53, Zbigniew Bodek wrote: > Author: zbb > Date: Wed Jul 8 13:52:59 2015 > New Revision: 285270 > URL: https://svnweb.freebsd.org/changeset/base/285270 > > Log: > Add memory barrier to bus_

Re: svn commit: r285270 - head/sys/sys

2015-07-08 Thread Andrew Turner
On Wed, 8 Jul 2015 13:53:00 + (UTC) Zbigniew Bodek wrote: > Author: zbb > Date: Wed Jul 8 13:52:59 2015 > New Revision: 285270 > URL: https://svnweb.freebsd.org/changeset/base/285270 > > Log: > Add memory barrier to bus_dmamap_sync() > > On platforms which are fully IO-coherent, the

svn commit: r285270 - head/sys/sys

2015-07-08 Thread Zbigniew Bodek
Author: zbb Date: Wed Jul 8 13:52:59 2015 New Revision: 285270 URL: https://svnweb.freebsd.org/changeset/base/285270 Log: Add memory barrier to bus_dmamap_sync() On platforms which are fully IO-coherent, the map might be null. We need to guarantee that all data is observable after the