Commit | Line | Data |
---|---|---|
cbb5901b JA |
1 | Queue sysfs files |
2 | ================= | |
3 | ||
4 | This text file will detail the queue files that are located in the sysfs tree | |
5 | for each block device. Note that stacked devices typically do not export | |
6 | any settings, since their queue merely functions are a remapping target. | |
7 | These files are the ones found in the /sys/block/xxx/queue/ directory. | |
8 | ||
9 | Files denoted with a RO postfix are readonly and the RW postfix means | |
10 | read-write. | |
11 | ||
4004e90c NJ |
12 | add_random (RW) |
13 | ---------------- | |
db4ced14 | 14 | This file allows to turn off the disk entropy contribution. Default |
4004e90c NJ |
15 | value of this file is '1'(on). |
16 | ||
005411ea JL |
17 | dax (RO) |
18 | -------- | |
19 | This file indicates whether the device supports Direct Access (DAX), | |
20 | used by CPU-addressable storage to bypass the pagecache. It shows '1' | |
21 | if true, '0' if not. | |
22 | ||
4004e90c NJ |
23 | discard_granularity (RO) |
24 | ----------------------- | |
25 | This shows the size of internal allocation of the device in bytes, if | |
26 | reported by the device. A value of '0' means device does not support | |
27 | the discard functionality. | |
28 | ||
0034af03 | 29 | discard_max_hw_bytes (RO) |
4004e90c NJ |
30 | ---------------------- |
31 | Devices that support discard functionality may have internal limits on | |
32 | the number of bytes that can be trimmed or unmapped in a single operation. | |
33 | The discard_max_bytes parameter is set by the device driver to the maximum | |
34 | number of bytes that can be discarded in a single operation. Discard | |
35 | requests issued to the device must not exceed this limit. A discard_max_bytes | |
36 | value of 0 means that the device does not support discard functionality. | |
37 | ||
0034af03 JA |
38 | discard_max_bytes (RW) |
39 | ---------------------- | |
40 | While discard_max_hw_bytes is the hardware limit for the device, this | |
41 | setting is the software limit. Some devices exhibit large latencies when | |
42 | large discards are issued, setting this value lower will make Linux issue | |
43 | smaller discards and potentially help reduce latencies induced by large | |
44 | discard operations. | |
45 | ||
4004e90c NJ |
46 | discard_zeroes_data (RO) |
47 | ------------------------ | |
48 | When read, this file will show if the discarded block are zeroed by the | |
49 | device or not. If its value is '1' the blocks are zeroed otherwise not. | |
50 | ||
cbb5901b JA |
51 | hw_sector_size (RO) |
52 | ------------------- | |
53 | This is the hardware sector size of the device, in bytes. | |
54 | ||
005411ea JL |
55 | io_poll (RW) |
56 | ------------ | |
57 | When read, this file shows the total number of block IO polls and how | |
58 | many returned success. Writing '0' to this file will disable polling | |
59 | for this device. Writing any non-zero value will enable this feature. | |
60 | ||
4004e90c NJ |
61 | iostats (RW) |
62 | ------------- | |
63 | This file is used to control (on/off) the iostats accounting of the | |
64 | disk. | |
65 | ||
66 | logical_block_size (RO) | |
67 | ----------------------- | |
141fd28c | 68 | This is the logical block size of the device, in bytes. |
4004e90c | 69 | |
cbb5901b JA |
70 | max_hw_sectors_kb (RO) |
71 | ---------------------- | |
72 | This is the maximum number of kilobytes supported in a single data transfer. | |
73 | ||
4004e90c NJ |
74 | max_integrity_segments (RO) |
75 | --------------------------- | |
76 | When read, this file shows the max limit of integrity segments as | |
77 | set by block layer which a hardware controller can handle. | |
78 | ||
cbb5901b JA |
79 | max_sectors_kb (RW) |
80 | ------------------- | |
81 | This is the maximum number of kilobytes that the block layer will allow | |
82 | for a filesystem request. Must be smaller than or equal to the maximum | |
83 | size allowed by the hardware. | |
84 | ||
4004e90c NJ |
85 | max_segments (RO) |
86 | ----------------- | |
87 | Maximum number of segments of the device. | |
88 | ||
89 | max_segment_size (RO) | |
90 | --------------------- | |
91 | Maximum segment size of the device. | |
92 | ||
93 | minimum_io_size (RO) | |
94 | -------------------- | |
db4ced14 | 95 | This is the smallest preferred IO size reported by the device. |
4004e90c | 96 | |
cbb5901b JA |
97 | nomerges (RW) |
98 | ------------- | |
488991e2 AB |
99 | This enables the user to disable the lookup logic involved with IO |
100 | merging requests in the block layer. By default (0) all merges are | |
101 | enabled. When set to 1 only simple one-hit merges will be tried. When | |
102 | set to 2 no merge algorithms will be tried (including one-hit or more | |
103 | complex tree/hash lookups). | |
cbb5901b JA |
104 | |
105 | nr_requests (RW) | |
106 | ---------------- | |
107 | This controls how many requests may be allocated in the block layer for | |
108 | read or write requests. Note that the total allocated number may be twice | |
109 | this amount, since it applies only to reads or writes (not the accumulated | |
110 | sum). | |
111 | ||
a051661c TH |
112 | To avoid priority inversion through request starvation, a request |
113 | queue maintains a separate request pool per each cgroup when | |
114 | CONFIG_BLK_CGROUP is enabled, and this parameter applies to each such | |
115 | per-block-cgroup request pool. IOW, if there are N block cgroups, | |
f884ab15 | 116 | each request queue may have up to N request pools, each independently |
a051661c TH |
117 | regulated by nr_requests. |
118 | ||
4004e90c NJ |
119 | optimal_io_size (RO) |
120 | -------------------- | |
db4ced14 | 121 | This is the optimal IO size reported by the device. |
4004e90c NJ |
122 | |
123 | physical_block_size (RO) | |
124 | ------------------------ | |
125 | This is the physical block size of device, in bytes. | |
126 | ||
cbb5901b JA |
127 | read_ahead_kb (RW) |
128 | ------------------ | |
129 | Maximum number of kilobytes to read-ahead for filesystems on this block | |
130 | device. | |
131 | ||
4004e90c NJ |
132 | rotational (RW) |
133 | --------------- | |
134 | This file is used to stat if the device is of rotational type or | |
135 | non-rotational type. | |
136 | ||
cbb5901b JA |
137 | rq_affinity (RW) |
138 | ---------------- | |
5757a6d7 DW |
139 | If this option is '1', the block layer will migrate request completions to the |
140 | cpu "group" that originally submitted the request. For some workloads this | |
141 | provides a significant reduction in CPU cycles due to caching effects. | |
142 | ||
143 | For storage configurations that need to maximize distribution of completion | |
144 | processing setting this option to '2' forces the completion to run on the | |
145 | requesting cpu (bypassing the "group" aggregation logic). | |
cbb5901b JA |
146 | |
147 | scheduler (RW) | |
148 | -------------- | |
149 | When read, this file will display the current and available IO schedulers | |
150 | for this block device. The currently active IO scheduler will be enclosed | |
151 | in [] brackets. Writing an IO scheduler name to this file will switch | |
152 | control of this block device to that new IO scheduler. Note that writing | |
153 | an IO scheduler name to this file will attempt to load that IO scheduler | |
154 | module, if it isn't already present in the system. | |
155 | ||
93e9d8e8 JA |
156 | write_cache (RW) |
157 | ---------------- | |
158 | When read, this file will display whether the device has write back | |
159 | caching enabled or not. It will return "write back" for the former | |
160 | case, and "write through" for the latter. Writing to this file can | |
161 | change the kernels view of the device, but it doesn't alter the | |
162 | device state. This means that it might not be safe to toggle the | |
163 | setting from "write back" to "write through", since that will also | |
164 | eliminate cache flushes issued by the kernel. | |
cbb5901b | 165 | |
005411ea JL |
166 | write_same_max_bytes (RO) |
167 | ------------------------- | |
168 | This is the number of bytes the device can write in a single write-same | |
169 | command. A value of '0' means write-same is not supported by this | |
170 | device. | |
171 | ||
cbb5901b JA |
172 | |
173 | Jens Axboe <jens.axboe@oracle.com>, February 2009 |