svn commit: r299683 - in head/sys/arm64: arm64 include

Andrew Turner andrew at fubar.geek.nz
Fri May 20 12:29:09 UTC 2016


On Fri, 20 May 2016 11:44:44 +0200
Wojciech Macek <wma at semihalf.com> wrote:

> Where can I find any details about how was this change tested or
> reviewed? Apparently it breaks AHCI on armv8. Log below.
> 
> Wojtek
> 
> 
> mountroot> ufs:/dev/ada0s2
> > Trying to mount root from ufs:/dev/ada0s2 []...
> > warning: no time-of-day clock registered, system time will not be
> > set accurately
> > Setting hostuuid: 701a2acc-d0a0-11e5-8550-001517169365.
> > Setting hostid: 0x2abffeca.
> > No suitable dump device was found.
> > Starting file system checks:
> > /dev/ada0s2: FILE SYSTEM CLEAN; SKIPPING CHECKS
> > /dev/ada0s2: clean, 16575173 free (25325 frags, 2068731 blocks, 0.1%
> > fragmentation)
> > panic: unsupported combination of sync operations: 0x00000008  

How were you getting into dma_dcache_sync? That should only happen when
map->sync_count != 0, this should only be possible when the BF_COHERENT
flag is set. This flag should never get set.

Andrew


More information about the svn-src-head mailing list