Latest findings...
With the fio2.12 /dev/dax & mmap(). I run trace before start test. We can see above open(), mmap()
5.630816 cpu=65 pid=45154 tgid=45154 open [2] entry *pathname=0x7f62b6fabed0 flags=NOATIME mode=0600 filename: /dev/dax0.0
5.630819 cpu=65 pid=45154 tgid=45154 open [2] ret=3 syscallbeg= 0.000003 *pathname=0x7f62b6fabed0 flags=NOATIME mode=0600
5.630833 cpu=65 pid=45154 tgid=45154 mmap [9] entry *addr=0x0 length=-2147483648 prot=READ|WRITE flags=SHARED fd=3 offset=0x0
5.630836 cpu=65 pid=45154 tgid=45154 mmap [9] ret=0x7f1a00200000 syscallbeg= 0.000003 *addr=0x0 length=-2147483648 prot=READ|WRITE flags=SHARED fd=3 offset=0x0
5.630836 cpu=65 pid=45154 tgid=45154 madvise [28] entry *addr=0x7f1a00200000 length=-2147483648 advise=0x1
5.630837 cpu=65 pid=45154 tgid=45154 madvise [28] ret=0x0 syscallbeg= 0.000001 *addr=0x7f1a00200000 length=-2147483648 advise=0x1
How it could be that mmap() length size -2Gb ??? (length=-2147483648). I pointed +290g,
dl560g10spmem01:/var/work/fio-3.2 # /usr/bin/fio --filename=/dev/dax0.0 --rw=randrw --refill_buffers --norandommap --randrepeat=0 --ioengine=mmap --bssplit=4k/4:8k/7:16k/7:32k/15:64k/65:128k/1:256k/1 --rwmixread=5 --iodepth=1 --numjobs=16 --runtime=1800 --group_reporting --name=4-rand-rw-3xx --size=290g
-----Original Message-----
From: Elliott, Robert (Persistent Memory)
Sent: Wednesday, November 29, 2017 10:23 AM
To: Gavriliuk, Anton (HPS Ukraine) <anton.gavriliuk@xxxxxxx>; Rebecca Cran <rebecca@xxxxxxxxxxxx>; Sitsofe Wheeler <sitsofe@xxxxxxxxx>; Tomohiro Kusumi <tkusumi@xxxxxxxxxx>
Cc: fio@xxxxxxxxxxxxxxx; Kani, Toshimitsu <toshi.kani@xxxxxxx>
Subject: RE: fio 3.2
> -----Original Message-----
> From: Gavriliuk, Anton (HPS Ukraine)
> Sent: Wednesday, November 29, 2017 1:40 AM
> To: Elliott, Robert (Persistent Memory) <elliott@xxxxxxx>; Rebecca
> Cran <rebecca@xxxxxxxxxxxx>; Sitsofe Wheeler <sitsofe@xxxxxxxxx>
> Cc: fio@xxxxxxxxxxxxxxx; Kani, Toshimitsu <toshi.kani@xxxxxxx>
> Subject: RE: fio 3.2
>
> Previously I used fio 2.12 and it works fine. Now I build fio 3.2
> with gcc 4.8.5 and it returns error,
>
> dl560g10spmem01:~ # /usr/local/bin/fio --name=4-rand-rw-3xx --
> ioengine=mmap --iodepth=1 --rw=randrw --
> bssplit=4k/4:8k/7:16k/7:32k/15:64k/65:128k/1:256k/1 --rwmixread=5 --
> size=100g --numjobs=16 --group_reporting --runtime=120 --
> filename=/dev/dax0.0 --allow_file_create=0
> 4-rand-rw-3xx: (g=0): rw=randrw, bs=(R) 4096B-256KiB, (W) 4096B-
> 256KiB, (T) 4096B-256KiB, ioengine=mmap, iodepth=1 ...
> fio-3.2
> Starting 16 processes
> 4-rand-rw-3xx: failed to get file size of /dev/dax0.0
I see the same thing.
Commit 0f34169a broke it, moving a check from a non-character device path up a level to where it now triggers on a character device (for which the size is left set to -1ULL).
This gets it working again:
git revert 0f34169a
make
Tomohiro, could you help prepare a better fix?
��.n��������+%������w��{.n�������^n�r������&��z�ޗ�zf���h���~����������_��+v���)ߣ�
