Fix unit_base kb_base mixup in thread option conversion functions

[fio.git] / HOWTO

diff --git a/HOWTO b/HOWTO
index bfc60be50572a03074344d359719e886460a42d4..181e94ca99a68c39c9f082916c89e0034fbfba62 100644 (file)
--- a/HOWTO
+++ b/HOWTO
@@ -398,12 +398,13 @@ rw=str            Type of io pattern. Accepted values are:
 
                        read            Sequential reads
                        write           Sequential writes
+                       trim            Sequential trims
                        randwrite       Random writes
                        randread        Random reads
+                       randtrim        Random trims
                        rw,readwrite    Sequential mixed reads and writes
                        randrw          Random mixed reads and writes
-                       trimwrite       Mixed trims and writes. Blocks will be
-                                       trimmed first, then written to.
+                       trimwrite       Sequential trim+write sequences
 
                Fio defaults to read if the option is not specified.
                For the mixed io types, the default is to split them 50/50.
@@ -444,7 +445,7 @@ kb_base=int The base unit for a kilobyte. The defacto base is 2^10, 1024.
                1024 or 1000, with 1024 being the default.
 
 unified_rw_reporting=bool      Fio normally reports statistics on a per
-               data direction basis, meaning that read, write, and trim are
+               data direction basis, meaning that reads, writes, and trims are
                accounted and reported separately. If this option is set,
                the fio will sum the results and report them as "mixed"
                instead.
@@ -476,8 +477,17 @@ fadvise_hint=bool By default, fio will use fadvise() to advise the kernel
                on what IO patterns it is likely to issue. Sometimes you
                want to test specific IO patterns without telling the
                kernel about it, in which case you can disable this option.
-               If set, fio will use POSIX_FADV_SEQUENTIAL for sequential
-               IO and POSIX_FADV_RANDOM for random IO.
+               The following options are supported:
+
+                       sequential      Use FADV_SEQUENTIAL
+                       random          Use FADV_RANDOM
+                       1               Backwards-compatible hint for basing
+                                       the hint on the fio workload. Will use
+                                       FADV_SEQUENTIAL for a sequential
+                                       workload, and FADV_RANDOM for a random
+                                       workload.
+                       0               Backwards-compatible setting for not
+                                       issing a fadvise hint.
 
 fadvise_stream=int Notify the kernel what write stream ID to place these
                writes under. Only supported on Linux. Note, this option
@@ -819,6 +829,9 @@ ioengine=str        Defines how the job issues io to the file. The following
                        pmemblk Read and write through the NVML libpmemblk
                                interface.
 
+                       dev-dax Read and write through a DAX device exposed
+                               from persistent memory.
+
                        external Prefix to specify loading an external
                                IO engine object file. Append the engine
                                filename, eg ioengine=external:/tmp/foo.o