Hi Coly Thanks for your quick reply. I’ve already solved this problem by updating kernel version to 3.18.63. So I think there must be a bug in old bcache code of 3.10 kernel version. > 在 2017年8月9日,下午8:58,Coly Li <i@xxxxxxx> 写道: > > On 2017/8/9 下午3:09, cgxu wrote: >> HI >> >> I make bcache device using NVMe ssd device as cache and 4T SATA disk as backing device. >> >> I found write performance is extremely poor in writeback cache_mode but raw ssd has very high performance. >> >> I monitored io workload using iostat for a while, so I'm sure it really worked in writeback mode. >> >> can anyone explain this problem? Is there any compatible issue with special hardware? > > > Could you please to provide exact steps to reproduce the poor > performance situation ? When back ground writeback and garbage > collection is performing, front end regular I/O (especially write I/O) > performance is expected to downgrade. > > > -- > Coly Li > -- To unsubscribe from this list: send the line "unsubscribe linux-bcache" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html