1HDPARM(8) System Manager's Manual HDPARM(8)
2
6 hdparm - get/set SATA/IDE device parameters
7
9 hdparm [ flags ] [device] ..
10
12 hdparm provides a command line interface to various kernel interfaces
13 supported by the Linux SATA/PATA/SAS "libata" subsystem and the older
14 IDE driver subsystem. Many newer (2008 and later) USB drive enclosures
15 now also support "SAT" (SCSI-ATA Command Translation) and therefore may
16 also work with hdparm. Eg. recent WD "Passport" models and recent
17 NexStar-3 enclosures. Some options may work correctly only with the
18 latest kernels.
19
21 When no flags are given, -acdgkmur is assumed. For Get/Set options, a
22 query without an optional parameter (e.g. -d) will query (get) the
23 device state, and with a parameter (e.g., -d0) will set the device
24 state.
25 -a Get/set sector count for filesystem (software) read-ahead. This
27 is used to improve performance in sequential reads of large
28 files, by prefetching additional blocks in anticipation of them
29 being needed by the running task. Many IDE drives also have a
30 separate built-in read-ahead function, which augments this
31 filesystem (software) read-ahead function.
32 -A Get/set the IDE drive´s read-lookahead feature (usually ON by
34 default). Usage: -A0 (disable) or -A1 (enable).
35 -b Get/set bus state.
37 -B Query/set Advanced Power Management feature, if the drive sup‐
39 ports it. A low value means aggressive power management and a
40 high value means better performance. Possible settings range
41 from values 1 through 127 (which permit spin-down), and values
42 128 through 254 (which do not permit spin-down). The highest
43 degree of power management is attained with a setting of 1, and
44 the highest I/O performance with a setting of 254. A value of
45 255 tells hdparm to disable Advanced Power Management altogether
46 on the drive (not all drives support disabling it, but most do).
47 -c Query/enable (E)IDE 32-bit I/O support. A numeric parameter can
49 be used to enable/disable 32-bit I/O support: Currently sup‐
50 ported values include 0 to disable 32-bit I/O support, 1 to
51 enable 32-bit data transfers, and 3 to enable 32-bit data trans‐
52 fers with a special sync sequence required by many chipsets.
53 The value 3 works with nearly all 32-bit IDE chipsets, but
54 incurs slightly more overhead. Note that "32-bit" refers to
55 data transfers across a PCI or VLB bus to the interface card
56 only; all (E)IDE drives still have only a 16-bit connection over
57 the ribbon cable from the interface card.
58 -C Check the current IDE power mode status, which will always be
60 one of unknown (drive does not support this command),
61 active/idle (normal operation), standby (low power mode, drive
62 has spun down), or sleeping (lowest power mode, drive is com‐
63 pletely shut down). The -S, -y, -Y, and -Z flags can be used to
64 manipulate the IDE power modes.
65 -d Disable/enable the "using_dma" flag for this drive. This option
67 now works with most combinations of drives and PCI interfaces
68 which support DMA and which are known to the kernel IDE driver.
69 It is also a good idea to use the appropriate -X option in com‐
70 bination with -d1 to ensure that the drive itself is programmed
71 for the correct DMA mode, although most BIOSs should do this for
72 you at boot time. Using DMA nearly always gives the best per‐
73 formance, with fast I/O throughput and low CPU usage. But there
74 are at least a few configurations of chipsets and drives for
75 which DMA does not make much of a difference, or may even slow
76 things down (on really messed up hardware!). Your mileage may
77 vary.
78 --dco-freeze
80 DCO stands for Device Configuration Overlay, a way for vendors
81 to selectively disable certain features of a drive. The --dco-
82 freeze flag will freeze/lock the current drive configuration,
83 thereby preventing software (or malware) from changing any DCO
84 settings until after the next power-on reset.
85 --dco-identify
87 Query and dump information regarding drive configuration set‐
88 tings which can be disabled by the vendor or OEM installer.
89 These settings show capabilities of the drive which might be
90 disabled by the vendor for "enhanced compatibility". When dis‐
91 abled, they are otherwise hidden and will not show in the -I
92 identify output. For example, system vendors sometimes disable
93 48_bit addressing on large drives, for compatibility (and loss
94 of capacity) with a specific BIOS. In such cases, --dco-iden‐
95 tify will show that the drive is 48_bit capable, but -I will not
96 show it, and nor will the drive accept 48_bit commands.
97 --dco-restore
99 Reset all drive settings, features, and accessible capacities
100 back to factory defaults and full capabilities. This command
101 will fail if DCO is frozen/locked, or if a -Np maximum size
102 restriction has also been set. This is EXTREMELY DANGEROUS and
103 will very likely cause massive loss of data. DO NOT USE THIS
104 COMMAND.
105 --direct
107 Use the kernel O_DIRECT flag when performing a -t timing test.
108 This bypasses the page cache, causing the reads to go directly
109 from the drive into hdparm's buffers, using so-called "raw" I/O.
110 In many cases, this can produce results that appear much faster
111 than the usual page cache method, giving a better indication of
112 raw device and driver performance.
113 --drq-hsm-error
115 VERY DANGEROUS, DON'T EVEN THINK ABOUT USING IT. This flag
116 causes hdparm to issue an IDENTIFY command to the kernel, but
117 incorrectly marked as a "non-data" command. This results in the
118 drive being left with its DataReQust(DRQ) line "stuck" high.
119 This confuses the kernel drivers, and may crash the system imme‐
120 diately with massive data loss. The option exists to help in
121 testing and fortifying the kernel against similar real-world
122 drive malfunctions. VERY DANGEROUS, DO NOT USE!!
123 -D Enable/disable the on-drive defect management feature, whereby
125 the drive firmware tries to automatically manage defective sec‐
126 tors by relocating them to "spare" sectors reserved by the fac‐
127 tory for such. Control of this feature via the -D flag is not
128 supported for most modern drives since ATA-4; thus this command
129 may fail.
130 -E Set cd/dvd drive speed. This is NOT necessary for regular oper‐
132 ation, as the drive will automatically switch speeds on its own.
133 But if you want to play with it, just supply a speed number
134 after the option, usually a number like 2 or 4. This can be
135 useful in some cases, though, to smooth out DVD video playback.
136 -f Sync and flush the buffer cache for the device on exit. This
138 operation is also performed internally as part of the -t and -T
139 timings and other flags.
140 --fallocate
142 This flag currently works only on ext4 and xfs filesystem types.
143 When used, this must be the only flag given. It requires two
144 parameters: the desired file size in kilo-bytes (byte count
145 divided by 1024), followed by the pathname for the new file. It
146 will create a new file of the specified size, but without actu‐
147 ally having to write any data to the file. This will normally
148 complete very quickly, and without thrashing the storage device.
149 Eg. Create a 10KByte file: hdparm --fallocate 10 temp_file
151 --fibmap
153 When used, this must be the only flag given. It requires a file
154 path as a parameter, and will print out a list of the block
155 extents (sector ranges) occupied by that file on disk. Sector
156 numbers are given as absolute LBA numbers, referenced from sec‐
157 tor 0 of the physical device rather than from the partition or
158 filesystem. This information can then be used for a variety of
159 purposes, such as examining the degree of fragmenation of larger
160 files, or determining appropriate sectors to deliberately cor‐
161 rupt during fault-injection testing procedures.
162 This flag uses the new FIEMAP (file extent map) ioctl() when
164 available, and falls back to the older FIBMAP (file block map)
165 ioctl() otherwise. Note that FIBMAP suffers from a 32-bit
166 block-number interface, and thus not work beyond 8TB or 16TB.
167 FIBMAP is also very slow, and does not deal well with preallo‐
168 cated uncommitted extents in ext4/xfs file systems, unless a
169 sync() is done before using this flag.
170 --fwdownload
172 When used, this should be the only flag given. It requires a
173 file path immediately after the flag, indicating where the new
174 drive firmware should be read from. The contents of this file
175 will be sent to the drive using the (S)ATA DOWNLOAD MICROCODE
176 command, using either transfer protocol 7 (entire file at once),
177 or, if the drive supports it, transfer protocol 3 (segmented
178 download). This command is EXTREMELY DANGEROUS and could
179 destroy both the drive and all data on it. DO NOT USE THIS COM‐
180 MAND. The --fwdownload-mode3 , --fwdownload-mode3-max , and
181 --fwdownload-mode7 variations on basic --fwdownload allow over‐
182 riding automatic protocol detection in favour of forcing hdparm
183 to use a specific transfer protocol, for testing purposes only.
184 -F Flush the on-drive write cache buffer (older drives may not
186 implement this).
187 -g Display the drive geometry (cylinders, heads, sectors), the size
189 (in sectors) of the device, and the starting offset (in sectors)
190 of the device from the beginning of the drive.
191 -h Display terse usage information (help).
193 -i Display the identification info which the kernel drivers (IDE,
195 libata) have stored from boot/configuration time. This may dif‐
196 fer from the current information obtainable directly from the
197 drive itself with the -I flag. The data returned may or may not
198 be current, depending on activity since booting the system. For
199 a more detailed interpretation of the identification info, refer
200 to AT Attachment Interface for Disk Drives, ANSI ASC X3T9.2
201 working draft, revision 4a, April 19/93, and later editions.
202 -I Request identification info directly from the drive, which is
204 displayed in a new expanded format with considerably more detail
205 than with the older -i flag.
206 --idle-immediate
208 Issue an ATA IDLE_IMMEDIATE command, to put the drive into a
209 lower power state. Usually the device remains spun-up.
210 --idle-unload
212 Issue an ATA IDLE_IMMEDIATE_WITH_UNLOAD command, to unload or
213 park the heads and put the drive into a lower power state. Usu‐
214 ally the device remains spun-up.
215 --Istdin
217 This is a special variation on the -I option, which accepts a
218 drive identification block as standard input instead of using a
219 /dev/hd* parameter. The format of this block must be exactly
220 the same as that found in the /proc/ide/*/hd*/identify "files",
221 or that produced by the --Istdout option described below. This
222 variation is designed for use with collected "libraries" of
223 drive identification information, and can also be used on ATAPI
224 drives which may give media errors with the standard mechanism.
225 When --Istdin is used, it must be the *only* parameter given.
226 --Istdout
228 This option dumps the drive's identify data in hex to stdout, in
229 a format similar to that from /proc/ide/*/identify, and suitable
230 for later use with the --Istdin option.
231 -k Get/set the keep_settings_over_reset flag for the drive. When
233 this flag is set, the driver will preserve the -dmu options over
234 a soft reset, (as done during the error recovery sequence).
235 This flag defaults to off, to prevent drive reset loops which
236 could be caused by combinations of -dmu settings. The -k flag
237 should therefore only be set after one has achieved confidence
238 in correct system operation with a chosen set of configuration
239 settings. In practice, all that is typically necessary to test
240 a configuration (prior to using -k) is to verify that the drive
241 can be read/written, and that no error logs (kernel messages)
242 are generated in the process (look in /var/adm/messages on most
243 systems).
244 -K Set the drive´s keep_features_over_reset flag. Setting this
246 enables the drive to retain the settings for -APSWXZ over a soft
247 reset (as done during the error recovery sequence). Not all
248 drives support this feature.
249 -L Set the drive´s doorlock flag. Setting this to 1 will lock the
251 door mechanism of some removable hard drives (eg. Syquest, ZIP,
252 Jazz..), and setting it to 0 will unlock the door mechanism.
253 Normally, Linux maintains the door locking mechanism automati‐
254 cally, depending on drive usage (locked whenever a filesystem is
255 mounted). But on system shutdown, this can be a nuisance if the
256 root partition is on a removable disk, since the root partition
257 is left mounted (read-only) after shutdown. So, by using this
258 command to unlock the door after the root filesystem is
259 remounted read-only, one can then remove the cartridge from the
260 drive after shutdown.
261 -m Get/set sector count for multiple sector I/O on the drive. A
263 setting of 0 disables this feature. Multiple sector mode (aka
264 IDE Block Mode), is a feature of most modern IDE hard drives,
265 permitting the transfer of multiple sectors per I/O interrupt,
266 rather than the usual one sector per interrupt. When this fea‐
267 ture is enabled, it typically reduces operating system overhead
268 for disk I/O by 30-50%. On many systems, it also provides
269 increased data throughput of anywhere from 5% to 50%. Some
270 drives, however (most notably the WD Caviar series), seem to run
271 slower with multiple mode enabled. Your mileage may vary. Most
272 drives support the minimum settings of 2, 4, 8, or 16 (sectors).
273 Larger settings may also be possible, depending on the drive. A
274 setting of 16 or 32 seems optimal on many systems. Western Dig‐
275 ital recommends lower settings of 4 to 8 on many of their
276 drives, due tiny (32kB) drive buffers and non-optimized buffer‐
277 ing algorithms. The -i flag can be used to find the maximum
278 setting supported by an installed drive (look for MaxMultSect in
279 the output). Some drives claim to support multiple mode, but
280 lose data at some settings. Under rare circumstances, such
281 failures can result in massive filesystem corruption.
282 --make-bad-sector
284 Deliberately create a bad sector (aka. "media error") on the
285 disk. EXCEPTIONALLY DANGEROUS. DO NOT USE THIS FLAG!! This can
286 be useful for testing of device/RAID error recovery mechanisms.
287 The sector number is given as a (base10) parameter after the
288 flag. Depending on the device, hdparm will choose one of two
289 possible ATA commands for corrupting the sector. The WRITE_LONG
290 works on most drives, but only up to the 28-bit sector boundary.
291 Some very recent drives (2008) may support the new WRITE_UNCOR‐
292 RECTABLE_EXT command, which works for any LBA48 sector. If
293 available, hdparm will use that in preference to WRITE_LONG.
294 The WRITE_UNCORRECTABLE_EXT command itself presents a choice of
295 how the new bad sector should behave. By default, it will look
296 like any other bad sector, and the drive may take some time to
297 retry and fail on subsequent READs of the sector. However, if a
298 single letter f is prepended immediately in front of the first
299 digit of the sector number parameter, then hdparm will issue a
300 "flagged" WRITE_UNCORRECTABLE_EXT, which causes the drive to
301 merely flag the sector as bad (rather than genuinely corrupt
302 it), and subsequent READs of the sector will fail immediately
303 (rather than after several retries). Note also that the
304 --repair-sector flag can be used to restore (any) bad sectors
305 when they are no longer needed, including sectors that were gen‐
306 uinely bad (the drive will likely remap those to a fresh area on
307 the media).
308 -M Get/set Automatic Acoustic Management (AAM) setting. Most modern
310 harddisk drives have the ability to speed down the head move‐
311 ments to reduce their noise output. The possible values are
312 between 0 and 254. 128 is the most quiet (and therefore slowest)
313 setting and 254 the fastest (and loudest). Some drives have only
314 two levels (quiet / fast), while others may have different lev‐
315 els between 128 and 254. At the moment, most drives only sup‐
316 port 3 options, off, quiet, and fast. These have been assigned
317 the values 0, 128, and 254 at present, respectively, but integer
318 space has been incorporated for future expansion, should this
319 change.
320 -N Get/set max visible number of sectors, also known as the Host
322 Protected Area setting. Without a parameter, -N displays the
323 current setting, which is reported as two values: the first
324 gives the current max sectors setting, and the second shows the
325 native (real) hardware limit for the disk. The difference
326 between these two values indicates how many sectors of the disk
327 are currently hidden from the operating system, in the form of a
328 Host Protected Area (HPA). This area is often used by computer
329 makers to hold diagnostic software, and/or a copy of the origi‐
330 nally provided operating system for recovery purposes. To
331 change the current max (VERY DANGEROUS, DATA LOSS IS EXTREMELY
332 LIKELY), a new value should be provided (in base10) immediately
333 following the -N flag. This value is specified as a count of
334 sectors, rather than the "max sector address" of the drive.
335 Drives have the concept of a temporary (volatile) setting which
336 is lost on the next hardware reset, as well as a more permanent
337 (non-volatile) value which survives resets and power cycles. By
338 default, -N affects only the temporary (volatile) setting. To
339 change the permanent (non-volatile) value, prepend a leading p
340 character immediately before the first digit of the value.
341 Drives are supposed to allow only a single permanent change per
342 session. A hardware reset (or power cycle) is required before
343 another permanent -N operation can succeed. Note that any
344 attempt to set this value may fail if the disk is being accessed
345 by other software at the same time. This is because setting the
346 value requires a pair of back-to-back drive commands, but there
347 is no way to prevent some other command from being inserted
348 between them by the kernel. So if it fails initially, just try
349 again. Kernel support for -N is buggy for many adapter types
350 across many kernel versions, in that an incorrect (too small)
351 max size value is sometimes reported. As of the 2.6.27 kernel,
352 this does finally seem to be working on most hardware.
353 -n Get or set the "ignore write errors" flag in the driver. Do NOT
355 play with this without grokking the driver source code first.
356 -p Attempt to reprogram the IDE interface chipset for the specified
358 PIO mode, or attempt to auto-tune for the "best" PIO mode sup‐
359 ported by the drive. This feature is supported in the kernel
360 for only a few "known" chipsets, and even then the support is
361 iffy at best. Some IDE chipsets are unable to alter the PIO
362 mode for a single drive, in which case this flag may cause the
363 PIO mode for both drives to be set. Many IDE chipsets support
364 either fewer or more than the standard six (0 to 5) PIO modes,
365 so the exact speed setting that is actually implemented will
366 vary by chipset/driver sophistication. Use with extreme cau‐
367 tion! This feature includes zero protection for the unwary, and
368 an unsuccessful outcome may result in severe filesystem corrup‐
369 tion!
370 -P Set the maximum sector count for the drive´s internal prefetch
372 mechanism. Not all drives support this feature, and it was
373 dropped from the offical spec as of ATA-4.
374 --prefer-ata12
376 When using the SAT (SCSI ATA Translation) protocol, hdparm nor‐
377 mally prefers to use the 16-byte command format whenever possi‐
378 ble. But some USB drive enclosures don't work correctly with
379 16-byte commands. This flag can be used to force use of the
380 smaller 12-byte command format with such drives. hdparm will
381 still revert to 16-byte commands for things that cannot be done
382 with the 12-byte format (eg. sector accesses beyond 28-bits).
383 -q Handle the next flag quietly, suppressing normal output (but not
385 error messages). This is useful for reducing screen clutter
386 when running from system startup scripts. Not applicable to the
387 -i or -v or -t or -T flags.
388 -Q Get or set the device's command queue_depth, if supported by the
390 hardware. This only works with 2.6.xx (or later) kernels, and
391 only with device and driver combinations which support changing
392 the queue_depth. For SATA disks, this is the Native Command
393 Queuing (NCQ) queue depth.
394 -r Get/set read-only flag for the device. When set, Linux disal‐
396 lows write operations on the device.
397 --read-sector
399 Reads from the specified sector number, and dumps the contents
400 in hex to standard output. The sector number must be given
401 (base10) after this flag. hdparm will issue a low-level read
402 (completely bypassing the usual block layer read/write mecha‐
403 nisms) for the specified sector. This can be used to defini‐
404 tively check whether a given sector is bad (media error) or not
405 (doing so through the usual mechanisms can sometimes give false
406 positives).
407 --repair-sector
409 This is an alias for the --write-sector flag. VERY DANGEROUS.
410 -R Register an IDE interface (DANGEROUS). See the -U option for
412 more information.
413 -s Enable/disable the power-on in standby feature, if supported by
415 the drive. VERY DANGEROUS. Do not use unless you are abso‐
416 lutely certain that both the system BIOS (or firmware) and the
417 operating system kernel (Linux >= 2.6.22) support probing for
418 drives that use this feature. When enabled, the drive is pow‐
419 ered-up in the standby mode to allow the controller to sequence
420 the spin-up of devices, reducing the instantaneous current draw
421 burden when many drives share a power supply. Primarily for use
422 in large RAID setups. This feature is usually disabled and the
423 drive is powered-up in the active mode (see -C above). Note
424 that a drive may also allow enabling this feature by a jumper.
425 Some SATA drives support the control of this feature by pin 11
426 of the SATA power connector. In these cases, this command may be
427 unsupported or may have no effect.
428 -S Put the drive into idle (low-power) mode, and also set the
430 standby (spindown) timeout for the drive. This timeout value is
431 used by the drive to determine how long to wait (with no disk
432 activity) before turning off the spindle motor to save power.
433 Under such circumstances, the drive may take as long as 30 sec‐
434 onds to respond to a subsequent disk access, though most drives
435 are much quicker. The encoding of the timeout value is somewhat
436 peculiar. A value of zero means "timeouts are disabled": the
437 device will not automatically enter standby mode. Values from 1
438 to 240 specify multiples of 5 seconds, yielding timeouts from 5
439 seconds to 20 minutes. Values from 241 to 251 specify from 1 to
440 11 units of 30 minutes, yielding timeouts from 30 minutes to 5.5
441 hours. A value of 252 signifies a timeout of 21 minutes. A
442 value of 253 sets a vendor-defined timeout period between 8 and
443 12 hours, and the value 254 is reserved. 255 is interpreted as
444 21 minutes plus 15 seconds. Note that some older drives may
445 have very different interpretations of these values.
446 -T Perform timings of cache reads for benchmark and comparison pur‐
448 poses. For meaningful results, this operation should be
449 repeated 2-3 times on an otherwise inactive system (no other
450 active processes) with at least a couple of megabytes of free
451 memory. This displays the speed of reading directly from the
452 Linux buffer cache without disk access. This measurement is
453 essentially an indication of the throughput of the processor,
454 cache, and memory of the system under test.
455 -t Perform timings of device reads for benchmark and comparison
457 purposes. For meaningful results, this operation should be
458 repeated 2-3 times on an otherwise inactive system (no other
459 active processes) with at least a couple of megabytes of free
460 memory. This displays the speed of reading through the buffer
461 cache to the disk without any prior caching of data. This mea‐
462 surement is an indication of how fast the drive can sustain
463 sequential data reads under Linux, without any filesystem over‐
464 head. To ensure accurate measurements, the buffer cache is
465 flushed during the processing of -t using the BLKFLSBUF ioctl.
466 --trim-sector-ranges
468 For Solid State Drives (SSDs). EXCEPTIONALLY DANGEROUS. DO NOT
469 USE THIS FLAG!! Tells the drive firmware to discard unneeded
470 data sectors, destroying any data that may have been present
471 within them. This makes those sectors available for immediate
472 use by the firmware's garbage collection mechanism, to improve
473 scheduling for wear-leveling of the flash media. This option
474 expects one or more sector range pairs immediately after the
475 flag: an LBA starting address, a colon, and a sector count, with
476 no intervening spaces. EXCEPTIONALLY DANGEROUS. DO NOT USE THIS
477 FLAG!!
478 Eg. hdparm --trim-sector-ranges 1000:4 7894:16 /dev/sdz
480 --trim-sector-ranges-stdin
482 Identical to --trim-sector-ranges above, except the list of
483 lba:count pairs is read from stdin rather than being specified
484 on the command line. This can be used to avoid problems with
485 excessively long command lines. It also permits batching of
486 many more sector ranges into single commands to the drive, up to
487 the currently configured transfer limit (max_sectors_kb).
488 -u Get/set interrupt-unmask flag for the drive. A setting of 1
490 permits the driver to unmask other interrupts during processing
491 of a disk interrupt, which greatly improves Linux´s responsive‐
492 ness and eliminates "serial port overrun" errors. Use this fea‐
493 ture with caution: some drive/controller combinations do not
494 tolerate the increased I/O latencies possible when this feature
495 is enabled, resulting in massive filesystem corruption. In par‐
496 ticular, CMD-640B and RZ1000 (E)IDE interfaces can be unreliable
497 (due to a hardware flaw) when this option is used with kernel
498 versions earlier than 2.0.13. Disabling the IDE prefetch fea‐
499 ture of these interfaces (usually a BIOS/CMOS setting) provides
500 a safe fix for the problem for use with earlier kernels.
501 -U Un-register an IDE interface (DANGEROUS). The companion for the
503 -R option. Intended for use with hardware made specifically for
504 hot-swapping (very rare!). Use with knowledge and extreme cau‐
505 tion as this can easily hang or damage your system. The hdparm
506 source distribution includes a ´contrib´ directory with some
507 user-donated scripts for hot-swapping on the UltraBay of a
508 ThinkPad 600E. Use at your own risk.
509 -v Display some basic settings, similar to -acdgkmur for IDE. This
511 is also the default behaviour when no flags are specified.
512 --verbose
514 Display extra diagnostics from some commands.
515 -w Perform a device reset (DANGEROUS). Do NOT use this option. It
517 exists for unlikely situations where a reboot might otherwise be
518 required to get a confused drive back into a useable state.
519 --write-sector
521 Writes zeros to the specified sector number. VERY DANGEROUS.
522 The sector number must be given (base10) after this flag.
523 hdparm will issue a low-level write (completely bypassing the
524 usual block layer read/write mechanisms) to the specified sec‐
525 tor. This can be used to force a drive to repair a bad sector
526 (media error).
527 -W Get/set the IDE/SATA drive´s write-caching feature.
529 -x Tristate device for hotswap (DANGEROUS).
531 -X Set the IDE transfer mode for (E)IDE/ATA drives. This is typi‐
533 cally used in combination with -d1 when enabling DMA to/from a
534 drive on a supported interface chipset, where -X mdma2 is used
535 to select multiword DMA mode2 transfers and -X sdma1 is used to
536 select simple mode 1 DMA transfers. With systems which support
537 UltraDMA burst timings, -X udma2 is used to select UltraDMA
538 mode2 transfers (you´ll need to prepare the chipset for UltraDMA
539 beforehand). Apart from that, use of this flag is seldom neces‐
540 sary since most/all modern IDE drives default to their fastest
541 PIO transfer mode at power-on. Fiddling with this can be both
542 needless and risky. On drives which support alternate transfer
543 modes, -X can be used to switch the mode of the drive only.
544 Prior to changing the transfer mode, the IDE interface should be
545 jumpered or programmed (see -p flag) for the new mode setting to
546 prevent loss and/or corruption of data. Use this with extreme
547 caution! For the PIO (Programmed Input/Output) transfer modes
548 used by Linux, this value is simply the desired PIO mode number
549 plus 8. Thus, a value of 09 sets PIO mode1, 10 enables PIO
550 mode2, and 11 selects PIO mode3. Setting 00 restores the
551 drive´s "default" PIO mode, and 01 disables IORDY. For multi‐
552 word DMA, the value used is the desired DMA mode number plus 32.
553 for UltraDMA, the value is the desired UltraDMA mode number plus
554 64.
555 -y Force an IDE drive to immediately enter the low power consump‐
557 tion standby mode, usually causing it to spin down. The current
558 power mode status can be checked using the -C flag.
559 -Y Force an IDE drive to immediately enter the lowest power con‐
561 sumption sleep mode, causing it to shut down completely. A hard
562 or soft reset is required before the drive can be accessed again
563 (the Linux IDE driver will automatically handle issuing a reset
564 if/when needed). The current power mode status can be checked
565 using the -C flag.
566 -z Force a kernel re-read of the partition table of the specified
568 device(s).
569 -Z Disable the automatic power-saving function of certain Seagate
571 drives (ST3xxx models?), to prevent them from idling/spinning-
572 down at inconvenient times.
573 -H Read the temperature from some (mostly Hitachi) drives. Also
575 reports if the temperature is within operating condition range
576 (this may not be reliable). Does not cause the drive to spin up
577 if idle.
578 ATA Security Feature Set
581 These switches are DANGEROUS to experiment with, and might not work
583 with every kernel. USE AT YOUR OWN RISK.
584 --security-help
586 Display terse usage info for all of the --security-* flags.
587 --security-freeze
589 Freeze the drive´s security settings. The drive does not accept
590 any security commands until next power-on reset. Use this func‐
591 tion in combination with --security-unlock to protect drive from
592 any attempt to set a new password. Can be used standalone, too.
593 No other flags are permitted on the command line with this one.
594 --security-unlock PWD
596 Unlock the drive, using password PWD. Password is given as an
597 ASCII string and is padded with NULs to reach 32 bytes. The
598 applicable drive password is selected with the --user-master
599 switch. No other flags are permitted on the command line with
600 this one. THIS FEATURE IS EXPERIMENTAL AND NOT WELL TESTED. USE
601 AT YOUR OWN RISK.
602 --security-set-pass PWD
604 Lock the drive, using password PWD (Set Password) (DANGEROUS).
605 Password is given as an ASCII string and is padded with NULs to
606 reach 32 bytes. Use the special password NULL to set an empty
607 password. The applicable drive password is selected with the
608 --user-master switch and the applicable security mode with the
609 --security-mode switch. No other flags are permitted on the
610 command line with this one. THIS FEATURE IS EXPERIMENTAL AND
611 NOT WELL TESTED. USE AT YOUR OWN RISK.
612 --security-disable PWD
614 Disable drive locking, using password PWD. Password is given as
615 an ASCII string and is padded with NULs to reach 32 bytes. The
616 applicable drive password is selected with the --user-master
617 switch. No other flags are permitted on the command line with
618 this one. THIS FEATURE IS EXPERIMENTAL AND NOT WELL TESTED. USE
619 AT YOUR OWN RISK.
620 --security-erase PWD
622 Erase (locked) drive, using password PWD (DANGEROUS). Password
623 is given as an ASCII string and is padded with NULs to reach 32
624 bytes. Use the special password NULL to represent an empty
625 password. The applicable drive password is selected with the
626 --user-master switch. No other flags are permitted on the com‐
627 mand line with this one. THIS FEATURE IS EXPERIMENTAL AND NOT
628 WELL TESTED. USE AT YOUR OWN RISK.
629 --security-erase-enhanced PWD
631 Enhanced erase (locked) drive, using password PWD (DANGEROUS).
632 Password is given as an ASCII string and is padded with NULs to
633 reach 32 bytes. The applicable drive password is selected with
634 the --user-master switch. No other flags are permitted on the
635 command line with this one. THIS FEATURE IS EXPERIMENTAL AND
636 NOT WELL TESTED. USE AT YOUR OWN RISK.
637 --user-master USER
639 Specifies which password (user/master) to select. Defaults to
640 master. Only useful in combination with --security-unlock,
641 --security-set-pass, --security-disable, --security-erase or
642 --security-erase-enhanced.
643 u user password
644 m master password
645 THIS FEATURE IS EXPERIMENTAL AND NOT WELL TESTED. USE AT YOUR
647 OWN RISK.
648 --security-mode MODE
650 Specifies which security mode (high/maximum) to set. Defaults
651 to high. Only useful in combination with --security-set-pass.
652 h high security
653 m maximum security
654 THIS FEATURE IS EXPERIMENTAL AND NOT WELL TESTED. USE AT YOUR
656 OWN RISK.
657
659 /etc/hdparm.conf
660
662 As noted above, the -m sectcount and -u 1 options should be used with
663 caution at first, preferably on a read-only filesystem. Most drives
664 work well with these features, but a few drive/controller combinations
665 are not 100% compatible. Filesystem corruption may result. Backup
666 everything before experimenting!
667 Some options (eg. -r for SCSI) may not work with old kernels as neces‐
669 sary ioctl()´s were not supported.
670 Although this utility is intended primarily for use with SATA/IDE hard
672 disk devices, several of the options are also valid (and permitted) for
673 use with SCSI hard disk devices and MFM/RLL hard disks with XT inter‐
674 faces.
675 The Linux kernel up until 2.6.12 (and probably later) doesn´t handle
677 the security unlock and disable commands gracefully and will segfault
678 and in some cases even panic. The security commands however might
679 indeed have been executed by the drive. This poor kernel behaviour
680 makes the PIO data security commands rather useless at the moment.
681 Note that the "security erase" and "security disable" commands have
683 been implemented as two consecutive PIO data commands and will not suc‐
684 ceed on a locked drive because the second command will not be issued
685 after the segfault. See the code for hints how patch it to work around
686 this problem. Despite the segfault it is often still possible to run
687 two instances of hdparm consecutively and issue the two necessary com‐
688 mands that way.
689
691 hdparm has been written by Mark Lord <mlord@pobox.com>, the original
692 primary developer and maintainer of the (E)IDE driver for Linux, and
693 current contributer to the libata subsystem, along with suggestions and
694 patches from many netfolk.
695 The disable Seagate auto-powersaving code is courtesy of Tomi Leppikan‐
697 gas(tomilepp@paju.oulu.fi).
698 Security freeze command by Benjamin Benz, 2005.
700 PIO data out security commands by Leonard den Ottolander , 2005. Some
702 other parts by Benjamin Benz and others.
703
705 http://www.t13.org/ Technical Committee T13 AT Attachment (ATA/ATAPI)
706 Interface.
707 http://www.serialata.org/ Serial ATA International Organization.
709 http://www.compactflash.org/ CompactFlash Association
711Version 9.27 September 2009 HDPARM(8)