sysfs-fs-ext4 3.9 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118
  1. What: /sys/fs/ext4/<disk>/mb_stats
  2. Date: March 2008
  3. Contact: "Theodore Ts'o" <tytso@mit.edu>
  4. Description:
  5. Controls whether the multiblock allocator should
  6. collect statistics, which are shown during the unmount.
  7. 1 means to collect statistics, 0 means not to collect
  8. statistics
  9. What: /sys/fs/ext4/<disk>/mb_group_prealloc
  10. Date: March 2008
  11. Contact: "Theodore Ts'o" <tytso@mit.edu>
  12. Description:
  13. The multiblock allocator will round up allocation
  14. requests to a multiple of this tuning parameter if the
  15. stripe size is not set in the ext4 superblock
  16. What: /sys/fs/ext4/<disk>/mb_max_to_scan
  17. Date: March 2008
  18. Contact: "Theodore Ts'o" <tytso@mit.edu>
  19. Description:
  20. The maximum number of extents the multiblock allocator
  21. will search to find the best extent
  22. What: /sys/fs/ext4/<disk>/mb_min_to_scan
  23. Date: March 2008
  24. Contact: "Theodore Ts'o" <tytso@mit.edu>
  25. Description:
  26. The minimum number of extents the multiblock allocator
  27. will search to find the best extent
  28. What: /sys/fs/ext4/<disk>/mb_order2_req
  29. Date: March 2008
  30. Contact: "Theodore Ts'o" <tytso@mit.edu>
  31. Description:
  32. Tuning parameter which controls the minimum size for
  33. requests (as a power of 2) where the buddy cache is
  34. used
  35. What: /sys/fs/ext4/<disk>/mb_stream_req
  36. Date: March 2008
  37. Contact: "Theodore Ts'o" <tytso@mit.edu>
  38. Description:
  39. Files which have fewer blocks than this tunable
  40. parameter will have their blocks allocated out of a
  41. block group specific preallocation pool, so that small
  42. files are packed closely together. Each large file
  43. will have its blocks allocated out of its own unique
  44. preallocation pool.
  45. What: /sys/fs/ext4/<disk>/inode_readahead_blks
  46. Date: March 2008
  47. Contact: "Theodore Ts'o" <tytso@mit.edu>
  48. Description:
  49. Tuning parameter which controls the maximum number of
  50. inode table blocks that ext4's inode table readahead
  51. algorithm will pre-read into the buffer cache
  52. What: /sys/fs/ext4/<disk>/delayed_allocation_blocks
  53. Date: March 2008
  54. Contact: "Theodore Ts'o" <tytso@mit.edu>
  55. Description:
  56. This file is read-only and shows the number of blocks
  57. that are dirty in the page cache, but which do not
  58. have their location in the filesystem allocated yet.
  59. What: /sys/fs/ext4/<disk>/lifetime_write_kbytes
  60. Date: March 2008
  61. Contact: "Theodore Ts'o" <tytso@mit.edu>
  62. Description:
  63. This file is read-only and shows the number of kilobytes
  64. of data that have been written to this filesystem since it was
  65. created.
  66. What: /sys/fs/ext4/<disk>/session_write_kbytes
  67. Date: March 2008
  68. Contact: "Theodore Ts'o" <tytso@mit.edu>
  69. Description:
  70. This file is read-only and shows the number of
  71. kilobytes of data that have been written to this
  72. filesystem since it was mounted.
  73. What: /sys/fs/ext4/<disk>/inode_goal
  74. Date: June 2008
  75. Contact: "Theodore Ts'o" <tytso@mit.edu>
  76. Description:
  77. Tuning parameter which (if non-zero) controls the goal
  78. inode used by the inode allocator in preference to
  79. all other allocation heuristics. This is intended for
  80. debugging use only, and should be 0 on production
  81. systems.
  82. What: /sys/fs/ext4/<disk>/max_writeback_mb_bump
  83. Date: September 2009
  84. Contact: "Theodore Ts'o" <tytso@mit.edu>
  85. Description:
  86. The maximum number of megabytes the writeback code will
  87. try to write out before move on to another inode.
  88. What: /sys/fs/ext4/<disk>/extent_max_zeroout_kb
  89. Date: August 2012
  90. Contact: "Theodore Ts'o" <tytso@mit.edu>
  91. Description:
  92. The maximum number of kilobytes which will be zeroed
  93. out in preference to creating a new uninitialized
  94. extent when manipulating an inode's extent tree. Note
  95. that using a larger value will increase the
  96. variability of time necessary to complete a random
  97. write operation (since a 4k random write might turn
  98. into a much larger write due to the zeroout
  99. operation).
  100. What: /sys/fs/ext4/<disk>/journal_task
  101. Date: February 2019
  102. Contact: "Theodore Ts'o" <tytso@mit.edu>
  103. Description:
  104. This file is read-only and shows the pid of journal thread in
  105. current pid-namespace or 0 if task is unreachable.