On 13 May 2019, Coly Li said: > Since there is no simple way to detect whether the backing divice is a > md raid device Is this really true? I find this very surprising. Lots of mkfs programs can detect exactly this (to determine the RAID stripe width), and it's obvious in sysfs as well... > this patch simply ignores I/O failures for read-ahead > bios on backing device, to avoid bogus backing device failure on a > degrading md raid array. ... but this feels like a good idea in any case. Readahead is a performance optimization: if it fails, it is not a user-visible sign of device failure in the same way that explicit I/O or metadata I/O is. -- NULL && (void)
