On 15/03/18 17:08, Evgeniy Didin wrote: > Hi Adrian, > >> On 14/03/18 14:32, Evgeniy Didin wrote: >>> Hi Adrian, >>>>> >>>>> Was the performance affected? i.e. the results from bonnie++ >>>> >>>> I have run bonnie++ several times before and after mentioned commit. Here is output: >>>> --------------------------------------------------<8------------------------------------------------------------------------- >>>> Before commit: >>>> ARCLinux,512M,6442,50,7211,0,5333,0,12954,99,345638,100,+++++,+++,16,22322,100,+++++,+++,32735,99,22921,100,+++++,+++,32211,100 >>>> ARCLinux,512M,6611,51,7248,0,5162,0,12951,99,344763,99,+++++,+++,16,22649,99,+++++,+++,32723,99,22572,100,+++++,+++,31697,100 >>>> ARCLinux,512M,6597,52,7307,0,5368,0,12934,99,343987,99,+++++,+++,16,22775,100,+++++,+++,32713,99,22832,100,+++++,+++,31899,99 >>>> After commit: >>>> ARCLinux,512M,4453,36,6474,1,5852,0,12940,99,344329,100,+++++,+++,16,22168,98,+++++,+++,32760,99,22755,100,+++++,+++,32205,100 >>>> ARCLinux,512M,5159,42,6944,1,5658,0,12945,99,344290,100,+++++,+++,16,22295,100,+++++,+++,32715,99,22888,99,+++++,+++,32007,99 >>>> ARCLinux,512M,5170,42,6840,2,5315,0,12946,99,343889,100,+++++,+++,16,22445,100,+++++,+++,32681,99,22967,100,+++++,+++,32117,99 >>>> --------------------------------------------------<8------------------------------------------------------------------------ >>>> ^^^^ >>>> If i understand correctly, in third column is shown "sequential output" per character speed (K/sec), which decreased. >>>> >>>>> What mount options did you use? >>>> >>>> I didn't use any options. Simply: >>>> # mount /dev/mmcblk0p1 /mnt >>> >>> Note that the file system on SDcard was ext4. >> >> Please try: >> >> perf record -a -e mmc:* -- bonnie++ -u root -r 256 -s 512 -x 1 -d /mnt >> and share the resulting perf.data file when you are able >> to reproduce the hung task message. >> > > I have applied the patch below and collected perf.data using command above, hung message also appeared. Is this the information you needed? > (perf.data is in attachments to this > message). Yes. Unfortunately the clock used is not accurate enough to correctly order the events across different CPUs, which makes it very hard to see delays between requests. You could try a different clock - refer the --clockid option to perf record. Nevertheless it shows there are no I/O errors which means the error recovery can be ruled out as a problem. The issue could be caused by the I/O scheduler. Under blk-mq the default scheduler is the mq-deadline scheduler whereas without blk-mq you would probably have been using cfq by default. You could try the bfq scheduler: echo bfq > /sys/block/mmcblk0/queue/scheduler But you might need to add it to the kernel config i.e. CONFIG_IOSCHED_BFQ=y Alternatively you could fiddle with the scheduler parameters: With mq-deadline they are: # grep -H . /sys/block/mmcblk0/queue/iosched/* /sys/block/mmcblk0/queue/iosched/fifo_batch:16 /sys/block/mmcblk0/queue/iosched/front_merges:1 /sys/block/mmcblk0/queue/iosched/read_expire:500 /sys/block/mmcblk0/queue/iosched/write_expire:5000 /sys/block/mmcblk0/queue/iosched/writes_starved:2 You could try decreasing the write_expire and/or fifo_batch. > >> You may want to add this patch first: >> >> https://urldefense.proofpoint.com/v2/url?u=https-3A__marc.info_-3Fl-3Dlinux-2Dmmc-26m-3D152110580324200&d=DwIDaQ&c=DPL6_X_6JkXFx7AXWqB0tg&r=vQk-RIbjwN0zvlwiMSpq3LYUTNf7Gqc4 >> ujhosYITtAw&m=LeVFmgzU0ydw-qHaAVh-kZI-v4mldnCycYxIGxqDexU&s=SyGvsVJBOzegO2Mz3hQ5j6CY8KrjLWJERVHfosSmKSs&e= > > Best regards, > Evgeniy Didin > -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
