1What: /sys/block/<disk>/stat 2Date: February 2008 3Contact: Jerome Marchand <jmarchan@redhat.com> 4Description: 5 The /sys/block/<disk>/stat files displays the I/O 6 statistics of disk <disk>. They contain 11 fields: 7 1 - reads completed successfully 8 2 - reads merged 9 3 - sectors read 10 4 - time spent reading (ms) 11 5 - writes completed 12 6 - writes merged 13 7 - sectors written 14 8 - time spent writing (ms) 15 9 - I/Os currently in progress 16 10 - time spent doing I/Os (ms) 17 11 - weighted time spent doing I/Os (ms) 18 For more details refer Documentation/iostats.txt 19 20 21What: /sys/block/<disk>/<part>/stat 22Date: February 2008 23Contact: Jerome Marchand <jmarchan@redhat.com> 24Description: 25 The /sys/block/<disk>/<part>/stat files display the 26 I/O statistics of partition <part>. The format is the 27 same as the above-written /sys/block/<disk>/stat 28 format. 29 30 31What: /sys/block/<disk>/integrity/format 32Date: June 2008 33Contact: Martin K. Petersen <martin.petersen@oracle.com> 34Description: 35 Metadata format for integrity capable block device. 36 E.g. T10-DIF-TYPE1-CRC. 37 38 39What: /sys/block/<disk>/integrity/read_verify 40Date: June 2008 41Contact: Martin K. Petersen <martin.petersen@oracle.com> 42Description: 43 Indicates whether the block layer should verify the 44 integrity of read requests serviced by devices that 45 support sending integrity metadata. 46 47 48What: /sys/block/<disk>/integrity/tag_size 49Date: June 2008 50Contact: Martin K. Petersen <martin.petersen@oracle.com> 51Description: 52 Number of bytes of integrity tag space available per 53 512 bytes of data. 54 55 56What: /sys/block/<disk>/integrity/device_is_integrity_capable 57Date: July 2014 58Contact: Martin K. Petersen <martin.petersen@oracle.com> 59Description: 60 Indicates whether a storage device is capable of storing 61 integrity metadata. Set if the device is T10 PI-capable. 62 63 64What: /sys/block/<disk>/integrity/write_generate 65Date: June 2008 66Contact: Martin K. Petersen <martin.petersen@oracle.com> 67Description: 68 Indicates whether the block layer should automatically 69 generate checksums for write requests bound for 70 devices that support receiving integrity metadata. 71 72What: /sys/block/<disk>/alignment_offset 73Date: April 2009 74Contact: Martin K. Petersen <martin.petersen@oracle.com> 75Description: 76 Storage devices may report a physical block size that is 77 bigger than the logical block size (for instance a drive 78 with 4KB physical sectors exposing 512-byte logical 79 blocks to the operating system). This parameter 80 indicates how many bytes the beginning of the device is 81 offset from the disk's natural alignment. 82 83What: /sys/block/<disk>/<partition>/alignment_offset 84Date: April 2009 85Contact: Martin K. Petersen <martin.petersen@oracle.com> 86Description: 87 Storage devices may report a physical block size that is 88 bigger than the logical block size (for instance a drive 89 with 4KB physical sectors exposing 512-byte logical 90 blocks to the operating system). This parameter 91 indicates how many bytes the beginning of the partition 92 is offset from the disk's natural alignment. 93 94What: /sys/block/<disk>/queue/logical_block_size 95Date: May 2009 96Contact: Martin K. Petersen <martin.petersen@oracle.com> 97Description: 98 This is the smallest unit the storage device can 99 address. It is typically 512 bytes. 100 101What: /sys/block/<disk>/queue/physical_block_size 102Date: May 2009 103Contact: Martin K. Petersen <martin.petersen@oracle.com> 104Description: 105 This is the smallest unit a physical storage device can 106 write atomically. It is usually the same as the logical 107 block size but may be bigger. One example is SATA 108 drives with 4KB sectors that expose a 512-byte logical 109 block size to the operating system. For stacked block 110 devices the physical_block_size variable contains the 111 maximum physical_block_size of the component devices. 112 113What: /sys/block/<disk>/queue/minimum_io_size 114Date: April 2009 115Contact: Martin K. Petersen <martin.petersen@oracle.com> 116Description: 117 Storage devices may report a granularity or preferred 118 minimum I/O size which is the smallest request the 119 device can perform without incurring a performance 120 penalty. For disk drives this is often the physical 121 block size. For RAID arrays it is often the stripe 122 chunk size. A properly aligned multiple of 123 minimum_io_size is the preferred request size for 124 workloads where a high number of I/O operations is 125 desired. 126 127What: /sys/block/<disk>/queue/optimal_io_size 128Date: April 2009 129Contact: Martin K. Petersen <martin.petersen@oracle.com> 130Description: 131 Storage devices may report an optimal I/O size, which is 132 the device's preferred unit for sustained I/O. This is 133 rarely reported for disk drives. For RAID arrays it is 134 usually the stripe width or the internal track size. A 135 properly aligned multiple of optimal_io_size is the 136 preferred request size for workloads where sustained 137 throughput is desired. If no optimal I/O size is 138 reported this file contains 0. 139 140What: /sys/block/<disk>/queue/nomerges 141Date: January 2010 142Contact: 143Description: 144 Standard I/O elevator operations include attempts to 145 merge contiguous I/Os. For known random I/O loads these 146 attempts will always fail and result in extra cycles 147 being spent in the kernel. This allows one to turn off 148 this behavior on one of two ways: When set to 1, complex 149 merge checks are disabled, but the simple one-shot merges 150 with the previous I/O request are enabled. When set to 2, 151 all merge tries are disabled. The default value is 0 - 152 which enables all types of merge tries. 153 154What: /sys/block/<disk>/discard_alignment 155Date: May 2011 156Contact: Martin K. Petersen <martin.petersen@oracle.com> 157Description: 158 Devices that support discard functionality may 159 internally allocate space in units that are bigger than 160 the exported logical block size. The discard_alignment 161 parameter indicates how many bytes the beginning of the 162 device is offset from the internal allocation unit's 163 natural alignment. 164 165What: /sys/block/<disk>/<partition>/discard_alignment 166Date: May 2011 167Contact: Martin K. Petersen <martin.petersen@oracle.com> 168Description: 169 Devices that support discard functionality may 170 internally allocate space in units that are bigger than 171 the exported logical block size. The discard_alignment 172 parameter indicates how many bytes the beginning of the 173 partition is offset from the internal allocation unit's 174 natural alignment. 175 176What: /sys/block/<disk>/queue/discard_granularity 177Date: May 2011 178Contact: Martin K. Petersen <martin.petersen@oracle.com> 179Description: 180 Devices that support discard functionality may 181 internally allocate space using units that are bigger 182 than the logical block size. The discard_granularity 183 parameter indicates the size of the internal allocation 184 unit in bytes if reported by the device. Otherwise the 185 discard_granularity will be set to match the device's 186 physical block size. A discard_granularity of 0 means 187 that the device does not support discard functionality. 188 189What: /sys/block/<disk>/queue/discard_max_bytes 190Date: May 2011 191Contact: Martin K. Petersen <martin.petersen@oracle.com> 192Description: 193 Devices that support discard functionality may have 194 internal limits on the number of bytes that can be 195 trimmed or unmapped in a single operation. Some storage 196 protocols also have inherent limits on the number of 197 blocks that can be described in a single command. The 198 discard_max_bytes parameter is set by the device driver 199 to the maximum number of bytes that can be discarded in 200 a single operation. Discard requests issued to the 201 device must not exceed this limit. A discard_max_bytes 202 value of 0 means that the device does not support 203 discard functionality. 204 205What: /sys/block/<disk>/queue/discard_zeroes_data 206Date: May 2011 207Contact: Martin K. Petersen <martin.petersen@oracle.com> 208Description: 209 Devices that support discard functionality may return 210 stale or random data when a previously discarded block 211 is read back. This can cause problems if the filesystem 212 expects discarded blocks to be explicitly cleared. If a 213 device reports that it deterministically returns zeroes 214 when a discarded area is read the discard_zeroes_data 215 parameter will be set to one. Otherwise it will be 0 and 216 the result of reading a discarded area is undefined. 217 218What: /sys/block/<disk>/queue/write_same_max_bytes 219Date: January 2012 220Contact: Martin K. Petersen <martin.petersen@oracle.com> 221Description: 222 Some devices support a write same operation in which a 223 single data block can be written to a range of several 224 contiguous blocks on storage. This can be used to wipe 225 areas on disk or to initialize drives in a RAID 226 configuration. write_same_max_bytes indicates how many 227 bytes can be written in a single write same command. If 228 write_same_max_bytes is 0, write same is not supported 229 by the device. 230 231