blob: 3adaace328a660fac9591e7a4b0726661dd3d832 [file] [log] [blame]
Linus Torvalds1da177e2005-04-16 15:20:36 -07001 Notes on the Generic Block Layer Rewrite in Linux 2.5
2 =====================================================
3
4Notes Written on Jan 15, 2002:
5 Jens Axboe <axboe@suse.de>
6 Suparna Bhattacharya <suparna@in.ibm.com>
7
8Last Updated May 2, 2002
9September 2003: Updated I/O Scheduler portions
10 Nick Piggin <piggin@cyberone.com.au>
11
12Introduction:
13
14These are some notes describing some aspects of the 2.5 block layer in the
15context of the bio rewrite. The idea is to bring out some of the key
16changes and a glimpse of the rationale behind those changes.
17
18Please mail corrections & suggestions to suparna@in.ibm.com.
19
20Credits:
21---------
22
232.5 bio rewrite:
24 Jens Axboe <axboe@suse.de>
25
26Many aspects of the generic block layer redesign were driven by and evolved
27over discussions, prior patches and the collective experience of several
28people. See sections 8 and 9 for a list of some related references.
29
30The following people helped with review comments and inputs for this
31document:
32 Christoph Hellwig <hch@infradead.org>
33 Arjan van de Ven <arjanv@redhat.com>
Adrian Bunkf4b09eb2006-01-03 13:37:51 +010034 Randy Dunlap <rdunlap@xenotime.net>
Linus Torvalds1da177e2005-04-16 15:20:36 -070035 Andre Hedrick <andre@linux-ide.org>
36
37The following people helped with fixes/contributions to the bio patches
38while it was still work-in-progress:
39 David S. Miller <davem@redhat.com>
40
41
42Description of Contents:
43------------------------
44
451. Scope for tuning of logic to various needs
46 1.1 Tuning based on device or low level driver capabilities
47 - Per-queue parameters
48 - Highmem I/O support
49 - I/O scheduler modularization
50 1.2 Tuning based on high level requirements/capabilities
51 1.2.1 I/O Barriers
52 1.2.2 Request Priority/Latency
53 1.3 Direct access/bypass to lower layers for diagnostics and special
54 device operations
55 1.3.1 Pre-built commands
562. New flexible and generic but minimalist i/o structure or descriptor
57 (instead of using buffer heads at the i/o layer)
58 2.1 Requirements/Goals addressed
59 2.2 The bio struct in detail (multi-page io unit)
60 2.3 Changes in the request structure
613. Using bios
62 3.1 Setup/teardown (allocation, splitting)
63 3.2 Generic bio helper routines
64 3.2.1 Traversing segments and completion units in a request
65 3.2.2 Setting up DMA scatterlists
66 3.2.3 I/O completion
67 3.2.4 Implications for drivers that do not interpret bios (don't handle
68 multiple segments)
69 3.2.5 Request command tagging
70 3.3 I/O submission
714. The I/O scheduler
725. Scalability related changes
73 5.1 Granular locking: Removal of io_request_lock
74 5.2 Prepare for transition to 64 bit sector_t
756. Other Changes/Implications
76 6.1 Partition re-mapping handled by the generic block layer
777. A few tips on migration of older drivers
788. A list of prior/related/impacted patches/ideas
799. Other References/Discussion Threads
80
81---------------------------------------------------------------------------
82
83Bio Notes
84--------
85
86Let us discuss the changes in the context of how some overall goals for the
87block layer are addressed.
88
891. Scope for tuning the generic logic to satisfy various requirements
90
91The block layer design supports adaptable abstractions to handle common
92processing with the ability to tune the logic to an appropriate extent
93depending on the nature of the device and the requirements of the caller.
94One of the objectives of the rewrite was to increase the degree of tunability
95and to enable higher level code to utilize underlying device/driver
96capabilities to the maximum extent for better i/o performance. This is
97important especially in the light of ever improving hardware capabilities
98and application/middleware software designed to take advantage of these
99capabilities.
100
1011.1 Tuning based on low level device / driver capabilities
102
103Sophisticated devices with large built-in caches, intelligent i/o scheduling
104optimizations, high memory DMA support, etc may find some of the
105generic processing an overhead, while for less capable devices the
106generic functionality is essential for performance or correctness reasons.
107Knowledge of some of the capabilities or parameters of the device should be
108used at the generic block layer to take the right decisions on
109behalf of the driver.
110
111How is this achieved ?
112
113Tuning at a per-queue level:
114
115i. Per-queue limits/values exported to the generic layer by the driver
116
117Various parameters that the generic i/o scheduler logic uses are set at
118a per-queue level (e.g maximum request size, maximum number of segments in
119a scatter-gather list, hardsect size)
120
121Some parameters that were earlier available as global arrays indexed by
122major/minor are now directly associated with the queue. Some of these may
123move into the block device structure in the future. Some characteristics
124have been incorporated into a queue flags field rather than separate fields
125in themselves. There are blk_queue_xxx functions to set the parameters,
126rather than update the fields directly
127
128Some new queue property settings:
129
130 blk_queue_bounce_limit(q, u64 dma_address)
131 Enable I/O to highmem pages, dma_address being the
132 limit. No highmem default.
133
134 blk_queue_max_sectors(q, max_sectors)
Mike Christie28832e82006-03-08 11:19:51 +0100135 Sets two variables that limit the size of the request.
136
137 - The request queue's max_sectors, which is a soft size in
Paolo Ornati670e9f32006-10-03 22:57:56 +0200138 units of 512 byte sectors, and could be dynamically varied
Mike Christie28832e82006-03-08 11:19:51 +0100139 by the core kernel.
140
141 - The request queue's max_hw_sectors, which is a hard limit
142 and reflects the maximum size request a driver can handle
143 in units of 512 byte sectors.
144
145 The default for both max_sectors and max_hw_sectors is
146 255. The upper limit of max_sectors is 1024.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700147
148 blk_queue_max_phys_segments(q, max_segments)
149 Maximum physical segments you can handle in a request. 128
150 default (driver limit). (See 3.2.2)
151
152 blk_queue_max_hw_segments(q, max_segments)
153 Maximum dma segments the hardware can handle in a request. 128
154 default (host adapter limit, after dma remapping).
155 (See 3.2.2)
156
157 blk_queue_max_segment_size(q, max_seg_size)
158 Maximum size of a clustered segment, 64kB default.
159
160 blk_queue_hardsect_size(q, hardsect_size)
161 Lowest possible sector size that the hardware can operate
162 on, 512 bytes default.
163
164New queue flags:
165
166 QUEUE_FLAG_CLUSTER (see 3.2.2)
167 QUEUE_FLAG_QUEUED (see 3.2.4)
168
169
170ii. High-mem i/o capabilities are now considered the default
171
172The generic bounce buffer logic, present in 2.4, where the block layer would
173by default copyin/out i/o requests on high-memory buffers to low-memory buffers
174assuming that the driver wouldn't be able to handle it directly, has been
175changed in 2.5. The bounce logic is now applied only for memory ranges
176for which the device cannot handle i/o. A driver can specify this by
177setting the queue bounce limit for the request queue for the device
178(blk_queue_bounce_limit()). This avoids the inefficiencies of the copyin/out
179where a device is capable of handling high memory i/o.
180
181In order to enable high-memory i/o where the device is capable of supporting
182it, the pci dma mapping routines and associated data structures have now been
183modified to accomplish a direct page -> bus translation, without requiring
184a virtual address mapping (unlike the earlier scheme of virtual address
185-> bus translation). So this works uniformly for high-memory pages (which
Matt LaPlante5d3f0832006-11-30 05:21:10 +0100186do not have a corresponding kernel virtual address space mapping) and
Linus Torvalds1da177e2005-04-16 15:20:36 -0700187low-memory pages.
188
189Note: Please refer to DMA-mapping.txt for a discussion on PCI high mem DMA
190aspects and mapping of scatter gather lists, and support for 64 bit PCI.
191
192Special handling is required only for cases where i/o needs to happen on
193pages at physical memory addresses beyond what the device can support. In these
194cases, a bounce bio representing a buffer from the supported memory range
195is used for performing the i/o with copyin/copyout as needed depending on
196the type of the operation. For example, in case of a read operation, the
197data read has to be copied to the original buffer on i/o completion, so a
198callback routine is set up to do this, while for write, the data is copied
199from the original buffer to the bounce buffer prior to issuing the
200operation. Since an original buffer may be in a high memory area that's not
201mapped in kernel virtual addr, a kmap operation may be required for
202performing the copy, and special care may be needed in the completion path
203as it may not be in irq context. Special care is also required (by way of
204GFP flags) when allocating bounce buffers, to avoid certain highmem
205deadlock possibilities.
206
207It is also possible that a bounce buffer may be allocated from high-memory
208area that's not mapped in kernel virtual addr, but within the range that the
209device can use directly; so the bounce page may need to be kmapped during
210copy operations. [Note: This does not hold in the current implementation,
211though]
212
213There are some situations when pages from high memory may need to
214be kmapped, even if bounce buffers are not necessary. For example a device
215may need to abort DMA operations and revert to PIO for the transfer, in
216which case a virtual mapping of the page is required. For SCSI it is also
217done in some scenarios where the low level driver cannot be trusted to
218handle a single sg entry correctly. The driver is expected to perform the
219kmaps as needed on such occasions using the __bio_kmap_atomic and bio_kmap_irq
220routines as appropriate. A driver could also use the blk_queue_bounce()
221routine on its own to bounce highmem i/o to low memory for specific requests
222if so desired.
223
224iii. The i/o scheduler algorithm itself can be replaced/set as appropriate
225
226As in 2.4, it is possible to plugin a brand new i/o scheduler for a particular
227queue or pick from (copy) existing generic schedulers and replace/override
228certain portions of it. The 2.5 rewrite provides improved modularization
229of the i/o scheduler. There are more pluggable callbacks, e.g for init,
230add request, extract request, which makes it possible to abstract specific
231i/o scheduling algorithm aspects and details outside of the generic loop.
232It also makes it possible to completely hide the implementation details of
233the i/o scheduler from block drivers.
234
235I/O scheduler wrappers are to be used instead of accessing the queue directly.
236See section 4. The I/O scheduler for details.
237
2381.2 Tuning Based on High level code capabilities
239
240i. Application capabilities for raw i/o
241
242This comes from some of the high-performance database/middleware
243requirements where an application prefers to make its own i/o scheduling
244decisions based on an understanding of the access patterns and i/o
245characteristics
246
247ii. High performance filesystems or other higher level kernel code's
248capabilities
249
250Kernel components like filesystems could also take their own i/o scheduling
251decisions for optimizing performance. Journalling filesystems may need
252some control over i/o ordering.
253
254What kind of support exists at the generic block layer for this ?
255
256The flags and rw fields in the bio structure can be used for some tuning
257from above e.g indicating that an i/o is just a readahead request, or for
258marking barrier requests (discussed next), or priority settings (currently
259unused). As far as user applications are concerned they would need an
260additional mechanism either via open flags or ioctls, or some other upper
261level mechanism to communicate such settings to block.
262
2631.2.1 I/O Barriers
264
265There is a way to enforce strict ordering for i/os through barriers.
266All requests before a barrier point must be serviced before the barrier
267request and any other requests arriving after the barrier will not be
268serviced until after the barrier has completed. This is useful for higher
269level control on write ordering, e.g flushing a log of committed updates
270to disk before the corresponding updates themselves.
271
272A flag in the bio structure, BIO_BARRIER is used to identify a barrier i/o.
273The generic i/o scheduler would make sure that it places the barrier request and
274all other requests coming after it after all the previous requests in the
275queue. Barriers may be implemented in different ways depending on the
Tejun Heoff5b8cf2006-01-06 09:58:37 +0100276driver. For more details regarding I/O barriers, please read barrier.txt
277in this directory.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700278
2791.2.2 Request Priority/Latency
280
281Todo/Under discussion:
282Arjan's proposed request priority scheme allows higher levels some broad
283 control (high/med/low) over the priority of an i/o request vs other pending
284 requests in the queue. For example it allows reads for bringing in an
285 executable page on demand to be given a higher priority over pending write
286 requests which haven't aged too much on the queue. Potentially this priority
287 could even be exposed to applications in some manner, providing higher level
288 tunability. Time based aging avoids starvation of lower priority
289 requests. Some bits in the bi_rw flags field in the bio structure are
290 intended to be used for this priority information.
291
292
2931.3 Direct Access to Low level Device/Driver Capabilities (Bypass mode)
294 (e.g Diagnostics, Systems Management)
295
296There are situations where high-level code needs to have direct access to
297the low level device capabilities or requires the ability to issue commands
298to the device bypassing some of the intermediate i/o layers.
299These could, for example, be special control commands issued through ioctl
300interfaces, or could be raw read/write commands that stress the drive's
301capabilities for certain kinds of fitness tests. Having direct interfaces at
302multiple levels without having to pass through upper layers makes
303it possible to perform bottom up validation of the i/o path, layer by
304layer, starting from the media.
305
306The normal i/o submission interfaces, e.g submit_bio, could be bypassed
307for specially crafted requests which such ioctl or diagnostics
308interfaces would typically use, and the elevator add_request routine
309can instead be used to directly insert such requests in the queue or preferably
310the blk_do_rq routine can be used to place the request on the queue and
311wait for completion. Alternatively, sometimes the caller might just
312invoke a lower level driver specific interface with the request as a
313parameter.
314
315If the request is a means for passing on special information associated with
316the command, then such information is associated with the request->special
317field (rather than misuse the request->buffer field which is meant for the
318request data buffer's virtual mapping).
319
320For passing request data, the caller must build up a bio descriptor
321representing the concerned memory buffer if the underlying driver interprets
322bio segments or uses the block layer end*request* functions for i/o
323completion. Alternatively one could directly use the request->buffer field to
324specify the virtual address of the buffer, if the driver expects buffer
325addresses passed in this way and ignores bio entries for the request type
326involved. In the latter case, the driver would modify and manage the
327request->buffer, request->sector and request->nr_sectors or
328request->current_nr_sectors fields itself rather than using the block layer
329end_request or end_that_request_first completion interfaces.
330(See 2.3 or Documentation/block/request.txt for a brief explanation of
331the request structure fields)
332
333[TBD: end_that_request_last should be usable even in this case;
334Perhaps an end_that_direct_request_first routine could be implemented to make
335handling direct requests easier for such drivers; Also for drivers that
336expect bios, a helper function could be provided for setting up a bio
337corresponding to a data buffer]
338
339<JENS: I dont understand the above, why is end_that_request_first() not
340usable? Or _last for that matter. I must be missing something>
341<SUP: What I meant here was that if the request doesn't have a bio, then
342 end_that_request_first doesn't modify nr_sectors or current_nr_sectors,
343 and hence can't be used for advancing request state settings on the
344 completion of partial transfers. The driver has to modify these fields
345 directly by hand.
346 This is because end_that_request_first only iterates over the bio list,
347 and always returns 0 if there are none associated with the request.
348 _last works OK in this case, and is not a problem, as I mentioned earlier
349>
350
3511.3.1 Pre-built Commands
352
353A request can be created with a pre-built custom command to be sent directly
354to the device. The cmd block in the request structure has room for filling
355in the command bytes. (i.e rq->cmd is now 16 bytes in size, and meant for
356command pre-building, and the type of the request is now indicated
357through rq->flags instead of via rq->cmd)
358
359The request structure flags can be set up to indicate the type of request
360in such cases (REQ_PC: direct packet command passed to driver, REQ_BLOCK_PC:
361packet command issued via blk_do_rq, REQ_SPECIAL: special request).
362
363It can help to pre-build device commands for requests in advance.
364Drivers can now specify a request prepare function (q->prep_rq_fn) that the
365block layer would invoke to pre-build device commands for a given request,
366or perform other preparatory processing for the request. This is routine is
367called by elv_next_request(), i.e. typically just before servicing a request.
368(The prepare function would not be called for requests that have REQ_DONTPREP
369enabled)
370
371Aside:
372 Pre-building could possibly even be done early, i.e before placing the
373 request on the queue, rather than construct the command on the fly in the
374 driver while servicing the request queue when it may affect latencies in
375 interrupt context or responsiveness in general. One way to add early
376 pre-building would be to do it whenever we fail to merge on a request.
377 Now REQ_NOMERGE is set in the request flags to skip this one in the future,
378 which means that it will not change before we feed it to the device. So
379 the pre-builder hook can be invoked there.
380
381
3822. Flexible and generic but minimalist i/o structure/descriptor.
383
3842.1 Reason for a new structure and requirements addressed
385
386Prior to 2.5, buffer heads were used as the unit of i/o at the generic block
387layer, and the low level request structure was associated with a chain of
388buffer heads for a contiguous i/o request. This led to certain inefficiencies
389when it came to large i/o requests and readv/writev style operations, as it
390forced such requests to be broken up into small chunks before being passed
391on to the generic block layer, only to be merged by the i/o scheduler
392when the underlying device was capable of handling the i/o in one shot.
393Also, using the buffer head as an i/o structure for i/os that didn't originate
Matt LaPlante4ae0edc2006-11-30 04:58:40 +0100394from the buffer cache unnecessarily added to the weight of the descriptors
Linus Torvalds1da177e2005-04-16 15:20:36 -0700395which were generated for each such chunk.
396
397The following were some of the goals and expectations considered in the
398redesign of the block i/o data structure in 2.5.
399
400i. Should be appropriate as a descriptor for both raw and buffered i/o -
401 avoid cache related fields which are irrelevant in the direct/page i/o path,
402 or filesystem block size alignment restrictions which may not be relevant
403 for raw i/o.
404ii. Ability to represent high-memory buffers (which do not have a virtual
405 address mapping in kernel address space).
Matt LaPlante4ae0edc2006-11-30 04:58:40 +0100406iii.Ability to represent large i/os w/o unnecessarily breaking them up (i.e
Linus Torvalds1da177e2005-04-16 15:20:36 -0700407 greater than PAGE_SIZE chunks in one shot)
408iv. At the same time, ability to retain independent identity of i/os from
409 different sources or i/o units requiring individual completion (e.g. for
410 latency reasons)
411v. Ability to represent an i/o involving multiple physical memory segments
412 (including non-page aligned page fragments, as specified via readv/writev)
Matt LaPlante4ae0edc2006-11-30 04:58:40 +0100413 without unnecessarily breaking it up, if the underlying device is capable of
Linus Torvalds1da177e2005-04-16 15:20:36 -0700414 handling it.
415vi. Preferably should be based on a memory descriptor structure that can be
416 passed around different types of subsystems or layers, maybe even
417 networking, without duplication or extra copies of data/descriptor fields
418 themselves in the process
419vii.Ability to handle the possibility of splits/merges as the structure passes
420 through layered drivers (lvm, md, evms), with minimal overhead.
421
422The solution was to define a new structure (bio) for the block layer,
423instead of using the buffer head structure (bh) directly, the idea being
424avoidance of some associated baggage and limitations. The bio structure
425is uniformly used for all i/o at the block layer ; it forms a part of the
426bh structure for buffered i/o, and in the case of raw/direct i/o kiobufs are
427mapped to bio structures.
428
4292.2 The bio struct
430
431The bio structure uses a vector representation pointing to an array of tuples
432of <page, offset, len> to describe the i/o buffer, and has various other
433fields describing i/o parameters and state that needs to be maintained for
434performing the i/o.
435
436Notice that this representation means that a bio has no virtual address
437mapping at all (unlike buffer heads).
438
439struct bio_vec {
440 struct page *bv_page;
441 unsigned short bv_len;
442 unsigned short bv_offset;
443};
444
445/*
446 * main unit of I/O for the block layer and lower layers (ie drivers)
447 */
448struct bio {
449 sector_t bi_sector;
450 struct bio *bi_next; /* request queue link */
451 struct block_device *bi_bdev; /* target device */
452 unsigned long bi_flags; /* status, command, etc */
453 unsigned long bi_rw; /* low bits: r/w, high: priority */
454
455 unsigned int bi_vcnt; /* how may bio_vec's */
456 unsigned int bi_idx; /* current index into bio_vec array */
457
458 unsigned int bi_size; /* total size in bytes */
459 unsigned short bi_phys_segments; /* segments after physaddr coalesce*/
460 unsigned short bi_hw_segments; /* segments after DMA remapping */
461 unsigned int bi_max; /* max bio_vecs we can hold
462 used as index into pool */
463 struct bio_vec *bi_io_vec; /* the actual vec list */
464 bio_end_io_t *bi_end_io; /* bi_end_io (bio) */
465 atomic_t bi_cnt; /* pin count: free when it hits zero */
466 void *bi_private;
467 bio_destructor_t *bi_destructor; /* bi_destructor (bio) */
468};
469
470With this multipage bio design:
471
472- Large i/os can be sent down in one go using a bio_vec list consisting
473 of an array of <page, offset, len> fragments (similar to the way fragments
474 are represented in the zero-copy network code)
475- Splitting of an i/o request across multiple devices (as in the case of
476 lvm or raid) is achieved by cloning the bio (where the clone points to
477 the same bi_io_vec array, but with the index and size accordingly modified)
478- A linked list of bios is used as before for unrelated merges (*) - this
479 avoids reallocs and makes independent completions easier to handle.
480- Code that traverses the req list needs to make a distinction between
481 segments of a request (bio_for_each_segment) and the distinct completion
482 units/bios (rq_for_each_bio).
483- Drivers which can't process a large bio in one shot can use the bi_idx
484 field to keep track of the next bio_vec entry to process.
485 (e.g a 1MB bio_vec needs to be handled in max 128kB chunks for IDE)
486 [TBD: Should preferably also have a bi_voffset and bi_vlen to avoid modifying
487 bi_offset an len fields]
488
489(*) unrelated merges -- a request ends up containing two or more bios that
490 didn't originate from the same place.
491
492bi_end_io() i/o callback gets called on i/o completion of the entire bio.
493
494At a lower level, drivers build a scatter gather list from the merged bios.
495The scatter gather list is in the form of an array of <page, offset, len>
496entries with their corresponding dma address mappings filled in at the
497appropriate time. As an optimization, contiguous physical pages can be
498covered by a single entry where <page> refers to the first page and <len>
499covers the range of pages (upto 16 contiguous pages could be covered this
500way). There is a helper routine (blk_rq_map_sg) which drivers can use to build
501the sg list.
502
503Note: Right now the only user of bios with more than one page is ll_rw_kio,
504which in turn means that only raw I/O uses it (direct i/o may not work
505right now). The intent however is to enable clustering of pages etc to
506become possible. The pagebuf abstraction layer from SGI also uses multi-page
507bios, but that is currently not included in the stock development kernels.
508The same is true of Andrew Morton's work-in-progress multipage bio writeout
509and readahead patches.
510
5112.3 Changes in the Request Structure
512
513The request structure is the structure that gets passed down to low level
514drivers. The block layer make_request function builds up a request structure,
515places it on the queue and invokes the drivers request_fn. The driver makes
516use of block layer helper routine elv_next_request to pull the next request
517off the queue. Control or diagnostic functions might bypass block and directly
518invoke underlying driver entry points passing in a specially constructed
519request structure.
520
521Only some relevant fields (mainly those which changed or may be referred
522to in some of the discussion here) are listed below, not necessarily in
523the order in which they occur in the structure (see include/linux/blkdev.h)
524Refer to Documentation/block/request.txt for details about all the request
525structure fields and a quick reference about the layers which are
526supposed to use or modify those fields.
527
528struct request {
529 struct list_head queuelist; /* Not meant to be directly accessed by
530 the driver.
531 Used by q->elv_next_request_fn
532 rq->queue is gone
533 */
534 .
535 .
536 unsigned char cmd[16]; /* prebuilt command data block */
537 unsigned long flags; /* also includes earlier rq->cmd settings */
538 .
539 .
540 sector_t sector; /* this field is now of type sector_t instead of int
541 preparation for 64 bit sectors */
542 .
543 .
544
545 /* Number of scatter-gather DMA addr+len pairs after
546 * physical address coalescing is performed.
547 */
548 unsigned short nr_phys_segments;
549
550 /* Number of scatter-gather addr+len pairs after
551 * physical and DMA remapping hardware coalescing is performed.
552 * This is the number of scatter-gather entries the driver
553 * will actually have to deal with after DMA mapping is done.
554 */
555 unsigned short nr_hw_segments;
556
557 /* Various sector counts */
558 unsigned long nr_sectors; /* no. of sectors left: driver modifiable */
559 unsigned long hard_nr_sectors; /* block internal copy of above */
560 unsigned int current_nr_sectors; /* no. of sectors left in the
561 current segment:driver modifiable */
562 unsigned long hard_cur_sectors; /* block internal copy of the above */
563 .
564 .
565 int tag; /* command tag associated with request */
566 void *special; /* same as before */
567 char *buffer; /* valid only for low memory buffers upto
568 current_nr_sectors */
569 .
570 .
571 struct bio *bio, *biotail; /* bio list instead of bh */
572 struct request_list *rl;
573}
574
575See the rq_flag_bits definitions for an explanation of the various flags
576available. Some bits are used by the block layer or i/o scheduler.
577
578The behaviour of the various sector counts are almost the same as before,
579except that since we have multi-segment bios, current_nr_sectors refers
580to the numbers of sectors in the current segment being processed which could
581be one of the many segments in the current bio (i.e i/o completion unit).
582The nr_sectors value refers to the total number of sectors in the whole
583request that remain to be transferred (no change). The purpose of the
584hard_xxx values is for block to remember these counts every time it hands
585over the request to the driver. These values are updated by block on
586end_that_request_first, i.e. every time the driver completes a part of the
587transfer and invokes block end*request helpers to mark this. The
588driver should not modify these values. The block layer sets up the
589nr_sectors and current_nr_sectors fields (based on the corresponding
590hard_xxx values and the number of bytes transferred) and updates it on
591every transfer that invokes end_that_request_first. It does the same for the
592buffer, bio, bio->bi_idx fields too.
593
594The buffer field is just a virtual address mapping of the current segment
595of the i/o buffer in cases where the buffer resides in low-memory. For high
596memory i/o, this field is not valid and must not be used by drivers.
597
598Code that sets up its own request structures and passes them down to
599a driver needs to be careful about interoperation with the block layer helper
600functions which the driver uses. (Section 1.3)
601
6023. Using bios
603
6043.1 Setup/Teardown
605
606There are routines for managing the allocation, and reference counting, and
607freeing of bios (bio_alloc, bio_get, bio_put).
608
609This makes use of Ingo Molnar's mempool implementation, which enables
610subsystems like bio to maintain their own reserve memory pools for guaranteed
611deadlock-free allocations during extreme VM load. For example, the VM
612subsystem makes use of the block layer to writeout dirty pages in order to be
613able to free up memory space, a case which needs careful handling. The
614allocation logic draws from the preallocated emergency reserve in situations
615where it cannot allocate through normal means. If the pool is empty and it
616can wait, then it would trigger action that would help free up memory or
617replenish the pool (without deadlocking) and wait for availability in the pool.
618If it is in IRQ context, and hence not in a position to do this, allocation
619could fail if the pool is empty. In general mempool always first tries to
620perform allocation without having to wait, even if it means digging into the
621pool as long it is not less that 50% full.
622
623On a free, memory is released to the pool or directly freed depending on
624the current availability in the pool. The mempool interface lets the
625subsystem specify the routines to be used for normal alloc and free. In the
626case of bio, these routines make use of the standard slab allocator.
627
628The caller of bio_alloc is expected to taken certain steps to avoid
629deadlocks, e.g. avoid trying to allocate more memory from the pool while
630already holding memory obtained from the pool.
631[TBD: This is a potential issue, though a rare possibility
632 in the bounce bio allocation that happens in the current code, since
633 it ends up allocating a second bio from the same pool while
634 holding the original bio ]
635
636Memory allocated from the pool should be released back within a limited
637amount of time (in the case of bio, that would be after the i/o is completed).
638This ensures that if part of the pool has been used up, some work (in this
639case i/o) must already be in progress and memory would be available when it
640is over. If allocating from multiple pools in the same code path, the order
641or hierarchy of allocation needs to be consistent, just the way one deals
642with multiple locks.
643
644The bio_alloc routine also needs to allocate the bio_vec_list (bvec_alloc())
645for a non-clone bio. There are the 6 pools setup for different size biovecs,
646so bio_alloc(gfp_mask, nr_iovecs) will allocate a vec_list of the
647given size from these slabs.
648
649The bi_destructor() routine takes into account the possibility of the bio
650having originated from a different source (see later discussions on
651n/w to block transfers and kvec_cb)
652
653The bio_get() routine may be used to hold an extra reference on a bio prior
654to i/o submission, if the bio fields are likely to be accessed after the
655i/o is issued (since the bio may otherwise get freed in case i/o completion
656happens in the meantime).
657
658The bio_clone() routine may be used to duplicate a bio, where the clone
659shares the bio_vec_list with the original bio (i.e. both point to the
660same bio_vec_list). This would typically be used for splitting i/o requests
661in lvm or md.
662
6633.2 Generic bio helper Routines
664
6653.2.1 Traversing segments and completion units in a request
666
667The macros bio_for_each_segment() and rq_for_each_bio() should be used for
668traversing the bios in the request list (drivers should avoid directly
669trying to do it themselves). Using these helpers should also make it easier
670to cope with block changes in the future.
671
672 rq_for_each_bio(bio, rq)
673 bio_for_each_segment(bio_vec, bio, i)
674 /* bio_vec is now current segment */
675
676I/O completion callbacks are per-bio rather than per-segment, so drivers
677that traverse bio chains on completion need to keep that in mind. Drivers
678which don't make a distinction between segments and completion units would
679need to be reorganized to support multi-segment bios.
680
6813.2.2 Setting up DMA scatterlists
682
683The blk_rq_map_sg() helper routine would be used for setting up scatter
684gather lists from a request, so a driver need not do it on its own.
685
686 nr_segments = blk_rq_map_sg(q, rq, scatterlist);
687
688The helper routine provides a level of abstraction which makes it easier
689to modify the internals of request to scatterlist conversion down the line
690without breaking drivers. The blk_rq_map_sg routine takes care of several
691things like collapsing physically contiguous segments (if QUEUE_FLAG_CLUSTER
692is set) and correct segment accounting to avoid exceeding the limits which
693the i/o hardware can handle, based on various queue properties.
694
695- Prevents a clustered segment from crossing a 4GB mem boundary
696- Avoids building segments that would exceed the number of physical
697 memory segments that the driver can handle (phys_segments) and the
698 number that the underlying hardware can handle at once, accounting for
699 DMA remapping (hw_segments) (i.e. IOMMU aware limits).
700
701Routines which the low level driver can use to set up the segment limits:
702
703blk_queue_max_hw_segments() : Sets an upper limit of the maximum number of
704hw data segments in a request (i.e. the maximum number of address/length
705pairs the host adapter can actually hand to the device at once)
706
707blk_queue_max_phys_segments() : Sets an upper limit on the maximum number
708of physical data segments in a request (i.e. the largest sized scatter list
709a driver could handle)
710
7113.2.3 I/O completion
712
713The existing generic block layer helper routines end_request,
714end_that_request_first and end_that_request_last can be used for i/o
715completion (and setting things up so the rest of the i/o or the next
716request can be kicked of) as before. With the introduction of multi-page
717bio support, end_that_request_first requires an additional argument indicating
718the number of sectors completed.
719
7203.2.4 Implications for drivers that do not interpret bios (don't handle
721 multiple segments)
722
723Drivers that do not interpret bios e.g those which do not handle multiple
724segments and do not support i/o into high memory addresses (require bounce
725buffers) and expect only virtually mapped buffers, can access the rq->buffer
726field. As before the driver should use current_nr_sectors to determine the
727size of remaining data in the current segment (that is the maximum it can
728transfer in one go unless it interprets segments), and rely on the block layer
729end_request, or end_that_request_first/last to take care of all accounting
730and transparent mapping of the next bio segment when a segment boundary
731is crossed on completion of a transfer. (The end*request* functions should
732be used if only if the request has come down from block/bio path, not for
733direct access requests which only specify rq->buffer without a valid rq->bio)
734
7353.2.5 Generic request command tagging
736
7373.2.5.1 Tag helpers
738
739Block now offers some simple generic functionality to help support command
740queueing (typically known as tagged command queueing), ie manage more than
741one outstanding command on a queue at any given time.
742
743 blk_queue_init_tags(request_queue_t *q, int depth)
744
745 Initialize internal command tagging structures for a maximum
746 depth of 'depth'.
747
748 blk_queue_free_tags((request_queue_t *q)
749
750 Teardown tag info associated with the queue. This will be done
751 automatically by block if blk_queue_cleanup() is called on a queue
752 that is using tagging.
753
754The above are initialization and exit management, the main helpers during
755normal operations are:
756
757 blk_queue_start_tag(request_queue_t *q, struct request *rq)
758
759 Start tagged operation for this request. A free tag number between
760 0 and 'depth' is assigned to the request (rq->tag holds this number),
761 and 'rq' is added to the internal tag management. If the maximum depth
762 for this queue is already achieved (or if the tag wasn't started for
763 some other reason), 1 is returned. Otherwise 0 is returned.
764
765 blk_queue_end_tag(request_queue_t *q, struct request *rq)
766
767 End tagged operation on this request. 'rq' is removed from the internal
768 book keeping structures.
769
770To minimize struct request and queue overhead, the tag helpers utilize some
771of the same request members that are used for normal request queue management.
772This means that a request cannot both be an active tag and be on the queue
773list at the same time. blk_queue_start_tag() will remove the request, but
774the driver must remember to call blk_queue_end_tag() before signalling
775completion of the request to the block layer. This means ending tag
776operations before calling end_that_request_last()! For an example of a user
777of these helpers, see the IDE tagged command queueing support.
778
779Certain hardware conditions may dictate a need to invalidate the block tag
780queue. For instance, on IDE any tagged request error needs to clear both
781the hardware and software block queue and enable the driver to sanely restart
782all the outstanding requests. There's a third helper to do that:
783
784 blk_queue_invalidate_tags(request_queue_t *q)
785
Matt LaPlanted6bc8ac2006-10-03 22:54:15 +0200786 Clear the internal block tag queue and re-add all the pending requests
Linus Torvalds1da177e2005-04-16 15:20:36 -0700787 to the request queue. The driver will receive them again on the
788 next request_fn run, just like it did the first time it encountered
789 them.
790
7913.2.5.2 Tag info
792
793Some block functions exist to query current tag status or to go from a
794tag number to the associated request. These are, in no particular order:
795
796 blk_queue_tagged(q)
797
798 Returns 1 if the queue 'q' is using tagging, 0 if not.
799
800 blk_queue_tag_request(q, tag)
801
802 Returns a pointer to the request associated with tag 'tag'.
803
804 blk_queue_tag_depth(q)
805
806 Return current queue depth.
807
808 blk_queue_tag_queue(q)
809
810 Returns 1 if the queue can accept a new queued command, 0 if we are
811 at the maximum depth already.
812
813 blk_queue_rq_tagged(rq)
814
815 Returns 1 if the request 'rq' is tagged.
816
8173.2.5.2 Internal structure
818
819Internally, block manages tags in the blk_queue_tag structure:
820
821 struct blk_queue_tag {
822 struct request **tag_index; /* array or pointers to rq */
823 unsigned long *tag_map; /* bitmap of free tags */
824 struct list_head busy_list; /* fifo list of busy tags */
825 int busy; /* queue depth */
826 int max_depth; /* max queue depth */
827 };
828
829Most of the above is simple and straight forward, however busy_list may need
830a bit of explaining. Normally we don't care too much about request ordering,
831but in the event of any barrier requests in the tag queue we need to ensure
832that requests are restarted in the order they were queue. This may happen
833if the driver needs to use blk_queue_invalidate_tags().
834
835Tagging also defines a new request flag, REQ_QUEUED. This is set whenever
836a request is currently tagged. You should not use this flag directly,
837blk_rq_tagged(rq) is the portable way to do so.
838
8393.3 I/O Submission
840
841The routine submit_bio() is used to submit a single io. Higher level i/o
842routines make use of this:
843
844(a) Buffered i/o:
845The routine submit_bh() invokes submit_bio() on a bio corresponding to the
846bh, allocating the bio if required. ll_rw_block() uses submit_bh() as before.
847
848(b) Kiobuf i/o (for raw/direct i/o):
849The ll_rw_kio() routine breaks up the kiobuf into page sized chunks and
850maps the array to one or more multi-page bios, issuing submit_bio() to
851perform the i/o on each of these.
852
853The embedded bh array in the kiobuf structure has been removed and no
854preallocation of bios is done for kiobufs. [The intent is to remove the
855blocks array as well, but it's currently in there to kludge around direct i/o.]
856Thus kiobuf allocation has switched back to using kmalloc rather than vmalloc.
857
858Todo/Observation:
859
860 A single kiobuf structure is assumed to correspond to a contiguous range
861 of data, so brw_kiovec() invokes ll_rw_kio for each kiobuf in a kiovec.
862 So right now it wouldn't work for direct i/o on non-contiguous blocks.
863 This is to be resolved. The eventual direction is to replace kiobuf
864 by kvec's.
865
866 Badari Pulavarty has a patch to implement direct i/o correctly using
867 bio and kvec.
868
869
870(c) Page i/o:
871Todo/Under discussion:
872
873 Andrew Morton's multi-page bio patches attempt to issue multi-page
874 writeouts (and reads) from the page cache, by directly building up
875 large bios for submission completely bypassing the usage of buffer
876 heads. This work is still in progress.
877
878 Christoph Hellwig had some code that uses bios for page-io (rather than
879 bh). This isn't included in bio as yet. Christoph was also working on a
880 design for representing virtual/real extents as an entity and modifying
881 some of the address space ops interfaces to utilize this abstraction rather
882 than buffer_heads. (This is somewhat along the lines of the SGI XFS pagebuf
883 abstraction, but intended to be as lightweight as possible).
884
885(d) Direct access i/o:
886Direct access requests that do not contain bios would be submitted differently
887as discussed earlier in section 1.3.
888
889Aside:
890
891 Kvec i/o:
892
Matt LaPlante53cb4722006-10-03 22:55:17 +0200893 Ben LaHaise's aio code uses a slightly different structure instead
Linus Torvalds1da177e2005-04-16 15:20:36 -0700894 of kiobufs, called a kvec_cb. This contains an array of <page, offset, len>
895 tuples (very much like the networking code), together with a callback function
896 and data pointer. This is embedded into a brw_cb structure when passed
897 to brw_kvec_async().
898
899 Now it should be possible to directly map these kvecs to a bio. Just as while
900 cloning, in this case rather than PRE_BUILT bio_vecs, we set the bi_io_vec
901 array pointer to point to the veclet array in kvecs.
902
903 TBD: In order for this to work, some changes are needed in the way multi-page
904 bios are handled today. The values of the tuples in such a vector passed in
905 from higher level code should not be modified by the block layer in the course
906 of its request processing, since that would make it hard for the higher layer
907 to continue to use the vector descriptor (kvec) after i/o completes. Instead,
908 all such transient state should either be maintained in the request structure,
909 and passed on in some way to the endio completion routine.
910
911
9124. The I/O scheduler
Tejun Heo4c9f7832005-10-20 16:47:40 +0200913I/O scheduler, a.k.a. elevator, is implemented in two layers. Generic dispatch
914queue and specific I/O schedulers. Unless stated otherwise, elevator is used
915to refer to both parts and I/O scheduler to specific I/O schedulers.
916
917Block layer implements generic dispatch queue in ll_rw_blk.c and elevator.c.
918The generic dispatch queue is responsible for properly ordering barrier
919requests, requeueing, handling non-fs requests and all other subtleties.
920
921Specific I/O schedulers are responsible for ordering normal filesystem
922requests. They can also choose to delay certain requests to improve
923throughput or whatever purpose. As the plural form indicates, there are
924multiple I/O schedulers. They can be built as modules but at least one should
925be built inside the kernel. Each queue can choose different one and can also
926change to another one dynamically.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700927
928A block layer call to the i/o scheduler follows the convention elv_xxx(). This
929calls elevator_xxx_fn in the elevator switch (drivers/block/elevator.c). Oh,
930xxx and xxx might not match exactly, but use your imagination. If an elevator
931doesn't implement a function, the switch does nothing or some minimal house
932keeping work.
933
9344.1. I/O scheduler API
935
936The functions an elevator may implement are: (* are mandatory)
937elevator_merge_fn called to query requests for merge with a bio
938
Tejun Heo4c9f7832005-10-20 16:47:40 +0200939elevator_merge_req_fn called when two requests get merged. the one
940 which gets merged into the other one will be
941 never seen by I/O scheduler again. IOW, after
942 being merged, the request is gone.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700943
944elevator_merged_fn called when a request in the scheduler has been
945 involved in a merge. It is used in the deadline
946 scheduler for example, to reposition the request
947 if its sorting order has changed.
948
Jens Axboe126ec9a62006-12-20 11:06:15 +0100949elevator_allow_merge_fn called whenever the block layer determines
950 that a bio can be merged into an existing
951 request safely. The io scheduler may still
952 want to stop a merge at this point if it
953 results in some sort of conflict internally,
954 this hook allows it to do that.
955
Tejun Heo4c9f7832005-10-20 16:47:40 +0200956elevator_dispatch_fn fills the dispatch queue with ready requests.
957 I/O schedulers are free to postpone requests by
958 not filling the dispatch queue unless @force
959 is non-zero. Once dispatched, I/O schedulers
960 are not allowed to manipulate the requests -
961 they belong to generic dispatch queue.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700962
Tejun Heo4c9f7832005-10-20 16:47:40 +0200963elevator_add_req_fn called to add a new request into the scheduler
Linus Torvalds1da177e2005-04-16 15:20:36 -0700964
965elevator_queue_empty_fn returns true if the merge queue is empty.
966 Drivers shouldn't use this, but rather check
967 if elv_next_request is NULL (without losing the
968 request if one exists!)
969
Linus Torvalds1da177e2005-04-16 15:20:36 -0700970elevator_former_req_fn
971elevator_latter_req_fn These return the request before or after the
972 one specified in disk sort order. Used by the
973 block layer to find merge possibilities.
974
Tejun Heo4c9f7832005-10-20 16:47:40 +0200975elevator_completed_req_fn called when a request is completed.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700976
977elevator_may_queue_fn returns true if the scheduler wants to allow the
978 current context to queue a new request even if
979 it is over the queue limit. This must be used
980 very carefully!!
981
982elevator_set_req_fn
983elevator_put_req_fn Must be used to allocate and free any elevator
Tejun Heo4c9f7832005-10-20 16:47:40 +0200984 specific storage for a request.
985
986elevator_activate_req_fn Called when device driver first sees a request.
987 I/O schedulers can use this callback to
988 determine when actual execution of a request
989 starts.
990elevator_deactivate_req_fn Called when device driver decides to delay
991 a request by requeueing it.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700992
993elevator_init_fn
994elevator_exit_fn Allocate and free any elevator specific storage
995 for a queue.
996
Tejun Heo4c9f7832005-10-20 16:47:40 +02009974.2 Request flows seen by I/O schedulers
Matt LaPlante53cb4722006-10-03 22:55:17 +0200998All requests seen by I/O schedulers strictly follow one of the following three
Tejun Heo4c9f7832005-10-20 16:47:40 +0200999flows.
1000
1001 set_req_fn ->
1002
1003 i. add_req_fn -> (merged_fn ->)* -> dispatch_fn -> activate_req_fn ->
1004 (deactivate_req_fn -> activate_req_fn ->)* -> completed_req_fn
1005 ii. add_req_fn -> (merged_fn ->)* -> merge_req_fn
1006 iii. [none]
1007
1008 -> put_req_fn
1009
10104.3 I/O scheduler implementation
Linus Torvalds1da177e2005-04-16 15:20:36 -07001011The generic i/o scheduler algorithm attempts to sort/merge/batch requests for
1012optimal disk scan and request servicing performance (based on generic
1013principles and device capabilities), optimized for:
1014i. improved throughput
1015ii. improved latency
1016iii. better utilization of h/w & CPU time
1017
1018Characteristics:
1019
1020i. Binary tree
1021AS and deadline i/o schedulers use red black binary trees for disk position
1022sorting and searching, and a fifo linked list for time-based searching. This
Matt LaPlante5d3f0832006-11-30 05:21:10 +01001023gives good scalability and good availability of information. Requests are
Linus Torvalds1da177e2005-04-16 15:20:36 -07001024almost always dispatched in disk sort order, so a cache is kept of the next
1025request in sort order to prevent binary tree lookups.
1026
1027This arrangement is not a generic block layer characteristic however, so
1028elevators may implement queues as they please.
1029
Tejun Heo4c9f7832005-10-20 16:47:40 +02001030ii. Merge hash
Linus Torvalds1da177e2005-04-16 15:20:36 -07001031AS and deadline use a hash table indexed by the last sector of a request. This
1032enables merging code to quickly look up "back merge" candidates, even when
1033multiple I/O streams are being performed at once on one disk.
1034
1035"Front merges", a new request being merged at the front of an existing request,
1036are far less common than "back merges" due to the nature of most I/O patterns.
1037Front merges are handled by the binary trees in AS and deadline schedulers.
1038
Tejun Heo4c9f7832005-10-20 16:47:40 +02001039iii. Plugging the queue to batch requests in anticipation of opportunities for
1040 merge/sort optimizations
Linus Torvalds1da177e2005-04-16 15:20:36 -07001041
1042This is just the same as in 2.4 so far, though per-device unplugging
1043support is anticipated for 2.5. Also with a priority-based i/o scheduler,
1044such decisions could be based on request priorities.
1045
1046Plugging is an approach that the current i/o scheduling algorithm resorts to so
1047that it collects up enough requests in the queue to be able to take
1048advantage of the sorting/merging logic in the elevator. If the
1049queue is empty when a request comes in, then it plugs the request queue
1050(sort of like plugging the bottom of a vessel to get fluid to build up)
1051till it fills up with a few more requests, before starting to service
1052the requests. This provides an opportunity to merge/sort the requests before
1053passing them down to the device. There are various conditions when the queue is
1054unplugged (to open up the flow again), either through a scheduled task or
1055could be on demand. For example wait_on_buffer sets the unplugging going
1056(by running tq_disk) so the read gets satisfied soon. So in the read case,
1057the queue gets explicitly unplugged as part of waiting for completion,
1058in fact all queues get unplugged as a side-effect.
1059
1060Aside:
1061 This is kind of controversial territory, as it's not clear if plugging is
1062 always the right thing to do. Devices typically have their own queues,
1063 and allowing a big queue to build up in software, while letting the device be
1064 idle for a while may not always make sense. The trick is to handle the fine
1065 balance between when to plug and when to open up. Also now that we have
1066 multi-page bios being queued in one shot, we may not need to wait to merge
1067 a big request from the broken up pieces coming by.
1068
1069 Per-queue granularity unplugging (still a Todo) may help reduce some of the
1070 concerns with just a single tq_disk flush approach. Something like
1071 blk_kick_queue() to unplug a specific queue (right away ?)
1072 or optionally, all queues, is in the plan.
1073
Tejun Heo4c9f7832005-10-20 16:47:40 +020010744.4 I/O contexts
Linus Torvalds1da177e2005-04-16 15:20:36 -07001075I/O contexts provide a dynamically allocated per process data area. They may
1076be used in I/O schedulers, and in the block layer (could be used for IO statis,
Ben Collins1d193f42005-11-15 00:09:21 -08001077priorities for example). See *io_context in block/ll_rw_blk.c, and as-iosched.c
1078for an example of usage in an i/o scheduler.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001079
1080
10815. Scalability related changes
1082
10835.1 Granular Locking: io_request_lock replaced by a per-queue lock
1084
1085The global io_request_lock has been removed as of 2.5, to avoid
1086the scalability bottleneck it was causing, and has been replaced by more
1087granular locking. The request queue structure has a pointer to the
1088lock to be used for that queue. As a result, locking can now be
1089per-queue, with a provision for sharing a lock across queues if
1090necessary (e.g the scsi layer sets the queue lock pointers to the
1091corresponding adapter lock, which results in a per host locking
1092granularity). The locking semantics are the same, i.e. locking is
1093still imposed by the block layer, grabbing the lock before
1094request_fn execution which it means that lots of older drivers
1095should still be SMP safe. Drivers are free to drop the queue
1096lock themselves, if required. Drivers that explicitly used the
1097io_request_lock for serialization need to be modified accordingly.
1098Usually it's as easy as adding a global lock:
1099
1100 static spinlock_t my_driver_lock = SPIN_LOCK_UNLOCKED;
1101
1102and passing the address to that lock to blk_init_queue().
1103
11045.2 64 bit sector numbers (sector_t prepares for 64 bit support)
1105
1106The sector number used in the bio structure has been changed to sector_t,
1107which could be defined as 64 bit in preparation for 64 bit sector support.
1108
11096. Other Changes/Implications
1110
11116.1 Partition re-mapping handled by the generic block layer
1112
1113In 2.5 some of the gendisk/partition related code has been reorganized.
1114Now the generic block layer performs partition-remapping early and thus
1115provides drivers with a sector number relative to whole device, rather than
1116having to take partition number into account in order to arrive at the true
1117sector number. The routine blk_partition_remap() is invoked by
1118generic_make_request even before invoking the queue specific make_request_fn,
1119so the i/o scheduler also gets to operate on whole disk sector numbers. This
1120should typically not require changes to block drivers, it just never gets
1121to invoke its own partition sector offset calculations since all bios
1122sent are offset from the beginning of the device.
1123
1124
11257. A Few Tips on Migration of older drivers
1126
1127Old-style drivers that just use CURRENT and ignores clustered requests,
1128may not need much change. The generic layer will automatically handle
1129clustered requests, multi-page bios, etc for the driver.
1130
1131For a low performance driver or hardware that is PIO driven or just doesn't
1132support scatter-gather changes should be minimal too.
1133
1134The following are some points to keep in mind when converting old drivers
1135to bio.
1136
1137Drivers should use elv_next_request to pick up requests and are no longer
1138supposed to handle looping directly over the request list.
1139(struct request->queue has been removed)
1140
1141Now end_that_request_first takes an additional number_of_sectors argument.
1142It used to handle always just the first buffer_head in a request, now
1143it will loop and handle as many sectors (on a bio-segment granularity)
1144as specified.
1145
1146Now bh->b_end_io is replaced by bio->bi_end_io, but most of the time the
1147right thing to use is bio_endio(bio, uptodate) instead.
1148
1149If the driver is dropping the io_request_lock from its request_fn strategy,
1150then it just needs to replace that with q->queue_lock instead.
1151
1152As described in Sec 1.1, drivers can set max sector size, max segment size
1153etc per queue now. Drivers that used to define their own merge functions i
1154to handle things like this can now just use the blk_queue_* functions at
1155blk_init_queue time.
1156
1157Drivers no longer have to map a {partition, sector offset} into the
1158correct absolute location anymore, this is done by the block layer, so
1159where a driver received a request ala this before:
1160
1161 rq->rq_dev = mk_kdev(3, 5); /* /dev/hda5 */
1162 rq->sector = 0; /* first sector on hda5 */
1163
1164 it will now see
1165
1166 rq->rq_dev = mk_kdev(3, 0); /* /dev/hda */
1167 rq->sector = 123128; /* offset from start of disk */
1168
1169As mentioned, there is no virtual mapping of a bio. For DMA, this is
1170not a problem as the driver probably never will need a virtual mapping.
1171Instead it needs a bus mapping (pci_map_page for a single segment or
1172use blk_rq_map_sg for scatter gather) to be able to ship it to the driver. For
1173PIO drivers (or drivers that need to revert to PIO transfer once in a
1174while (IDE for example)), where the CPU is doing the actual data
1175transfer a virtual mapping is needed. If the driver supports highmem I/O,
1176(Sec 1.1, (ii) ) it needs to use __bio_kmap_atomic and bio_kmap_irq to
1177temporarily map a bio into the virtual address space.
1178
1179
11808. Prior/Related/Impacted patches
1181
11828.1. Earlier kiobuf patches (sct/axboe/chait/hch/mkp)
1183- orig kiobuf & raw i/o patches (now in 2.4 tree)
1184- direct kiobuf based i/o to devices (no intermediate bh's)
1185- page i/o using kiobuf
1186- kiobuf splitting for lvm (mkp)
1187- elevator support for kiobuf request merging (axboe)
11888.2. Zero-copy networking (Dave Miller)
11898.3. SGI XFS - pagebuf patches - use of kiobufs
11908.4. Multi-page pioent patch for bio (Christoph Hellwig)
11918.5. Direct i/o implementation (Andrea Arcangeli) since 2.4.10-pre11
11928.6. Async i/o implementation patch (Ben LaHaise)
11938.7. EVMS layering design (IBM EVMS team)
11948.8. Larger page cache size patch (Ben LaHaise) and
1195 Large page size (Daniel Phillips)
1196 => larger contiguous physical memory buffers
11978.9. VM reservations patch (Ben LaHaise)
11988.10. Write clustering patches ? (Marcelo/Quintela/Riel ?)
11998.11. Block device in page cache patch (Andrea Archangeli) - now in 2.4.10+
12008.12. Multiple block-size transfers for faster raw i/o (Shailabh Nagar,
1201 Badari)
12028.13 Priority based i/o scheduler - prepatches (Arjan van de Ven)
12038.14 IDE Taskfile i/o patch (Andre Hedrick)
12048.15 Multi-page writeout and readahead patches (Andrew Morton)
12058.16 Direct i/o patches for 2.5 using kvec and bio (Badari Pulavarthy)
1206
12079. Other References:
1208
12099.1 The Splice I/O Model - Larry McVoy (and subsequent discussions on lkml,
1210and Linus' comments - Jan 2001)
12119.2 Discussions about kiobuf and bh design on lkml between sct, linus, alan
1212et al - Feb-March 2001 (many of the initial thoughts that led to bio were
Matt LaPlantefff92892006-10-03 22:47:42 +02001213brought up in this discussion thread)
Linus Torvalds1da177e2005-04-16 15:20:36 -070012149.3 Discussions on mempool on lkml - Dec 2001.
1215