Hi, The eMMC spec defines the cache "barrier" capability of the eMMC device - in-order flushing of cached data. The journal (jbd2) code submitting the commit record, sends it as REQ_FLUSH | REQ_FUA in order to - flush the previously written journalling data and - write commit record bypassing the cache. It seems, that flushing and cache bypassing may be replaced by cache barriers, at least from crash consistency point of view. Is this assumption correct? Thanks, Alexey -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
