• Home
  • History
  • Annotate
  • Line#
  • Navigate
  • Raw
  • Download
  • only in /netgear-R7000-V1.0.7.12_1.2.5/components/opensource/linux/linux-2.6.36/Documentation/DocBook/
1<?xml version="1.0" encoding="UTF-8"?>
2<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
3	"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
4
5<book id="MTD-NAND-Guide">
6 <bookinfo>
7  <title>MTD NAND Driver Programming Interface</title>
8  
9  <authorgroup>
10   <author>
11    <firstname>Thomas</firstname>
12    <surname>Gleixner</surname>
13    <affiliation>
14     <address>
15      <email>tglx@linutronix.de</email>
16     </address>
17    </affiliation>
18   </author>
19  </authorgroup>
20
21  <copyright>
22   <year>2004</year>
23   <holder>Thomas Gleixner</holder>
24  </copyright>
25
26  <legalnotice>
27   <para>
28     This documentation is free software; you can redistribute
29     it and/or modify it under the terms of the GNU General Public
30     License version 2 as published by the Free Software Foundation.
31   </para>
32      
33   <para>
34     This program is distributed in the hope that it will be
35     useful, but WITHOUT ANY WARRANTY; without even the implied
36     warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
37     See the GNU General Public License for more details.
38   </para>
39      
40   <para>
41     You should have received a copy of the GNU General Public
42     License along with this program; if not, write to the Free
43     Software Foundation, Inc., 59 Temple Place, Suite 330, Boston,
44     MA 02111-1307 USA
45   </para>
46      
47   <para>
48     For more details see the file COPYING in the source
49     distribution of Linux.
50   </para>
51  </legalnotice>
52 </bookinfo>
53
54<toc></toc>
55
56  <chapter id="intro">
57      <title>Introduction</title>
58  <para>
59  	The generic NAND driver supports almost all NAND and AG-AND based
60	chips and connects them to the Memory Technology Devices (MTD)
61	subsystem of the Linux Kernel.
62  </para>
63  <para>
64  	This documentation is provided for developers who want to implement
65	board drivers or filesystem drivers suitable for NAND devices.
66  </para>
67  </chapter>
68  
69  <chapter id="bugs">
70     <title>Known Bugs And Assumptions</title>
71  <para>
72	None.	
73  </para>
74  </chapter>
75
76  <chapter id="dochints">
77     <title>Documentation hints</title>
78     <para>
79     The function and structure docs are autogenerated. Each function and 
80     struct member has a short description which is marked with an [XXX] identifier.
81     The following chapters explain the meaning of those identifiers.
82     </para>
83     <sect1 id="Function_identifiers_XXX">
84	<title>Function identifiers [XXX]</title>
85     	<para>
86	The functions are marked with [XXX] identifiers in the short
87	comment. The identifiers explain the usage and scope of the
88	functions. Following identifiers are used:
89     	</para>
90	<itemizedlist>
91		<listitem><para>
92	  	[MTD Interface]</para><para>
93		These functions provide the interface to the MTD kernel API. 
94		They are not replacable and provide functionality
95		which is complete hardware independent.
96		</para></listitem>
97		<listitem><para>
98	  	[NAND Interface]</para><para>
99		These functions are exported and provide the interface to the NAND kernel API. 
100		</para></listitem>
101		<listitem><para>
102	  	[GENERIC]</para><para>
103		Generic functions are not replacable and provide functionality
104		which is complete hardware independent.
105		</para></listitem>
106		<listitem><para>
107	  	[DEFAULT]</para><para>
108		Default functions provide hardware related functionality which is suitable
109		for most of the implementations. These functions can be replaced by the
110		board driver if neccecary. Those functions are called via pointers in the
111		NAND chip description structure. The board driver can set the functions which
112		should be replaced by board dependent functions before calling nand_scan().
113		If the function pointer is NULL on entry to nand_scan() then the pointer
114		is set to the default function which is suitable for the detected chip type.
115		</para></listitem>
116	</itemizedlist>
117     </sect1>
118     <sect1 id="Struct_member_identifiers_XXX">
119	<title>Struct member identifiers [XXX]</title>
120     	<para>
121	The struct members are marked with [XXX] identifiers in the 
122	comment. The identifiers explain the usage and scope of the
123	members. Following identifiers are used:
124     	</para>
125	<itemizedlist>
126		<listitem><para>
127	  	[INTERN]</para><para>
128		These members are for NAND driver internal use only and must not be
129		modified. Most of these values are calculated from the chip geometry
130		information which is evaluated during nand_scan().
131		</para></listitem>
132		<listitem><para>
133	  	[REPLACEABLE]</para><para>
134		Replaceable members hold hardware related functions which can be 
135		provided by the board driver. The board driver can set the functions which
136		should be replaced by board dependent functions before calling nand_scan().
137		If the function pointer is NULL on entry to nand_scan() then the pointer
138		is set to the default function which is suitable for the detected chip type.
139		</para></listitem>
140		<listitem><para>
141	  	[BOARDSPECIFIC]</para><para>
142		Board specific members hold hardware related information which must
143		be provided by the board driver. The board driver must set the function
144		pointers and datafields before calling nand_scan().
145		</para></listitem>
146		<listitem><para>
147	  	[OPTIONAL]</para><para>
148		Optional members can hold information relevant for the board driver. The
149		generic NAND driver code does not use this information.
150		</para></listitem>
151	</itemizedlist>
152     </sect1>
153  </chapter>   
154
155  <chapter id="basicboarddriver">
156     	<title>Basic board driver</title>
157	<para>
158		For most boards it will be sufficient to provide just the
159		basic functions and fill out some really board dependent
160		members in the nand chip description structure.
161	</para>
162	<sect1 id="Basic_defines">
163		<title>Basic defines</title>
164		<para>
165			At least you have to provide a mtd structure and
166			a storage for the ioremap'ed chip address.
167			You can allocate the mtd structure using kmalloc
168			or you can allocate it statically.
169			In case of static allocation you have to allocate
170			a nand_chip structure too.
171		</para>
172		<para>
173			Kmalloc based example
174		</para>
175		<programlisting>
176static struct mtd_info *board_mtd;
177static void __iomem *baseaddr;
178		</programlisting>
179		<para>
180			Static example
181		</para>
182		<programlisting>
183static struct mtd_info board_mtd;
184static struct nand_chip board_chip;
185static void __iomem *baseaddr;
186		</programlisting>
187	</sect1>
188	<sect1 id="Partition_defines">
189		<title>Partition defines</title>
190		<para>
191			If you want to divide your device into partitions, then
192			enable the configuration switch CONFIG_MTD_PARTITIONS and define
193			a partitioning scheme suitable to your board.
194		</para>
195		<programlisting>
196#define NUM_PARTITIONS 2
197static struct mtd_partition partition_info[] = {
198	{ .name = "Flash partition 1",
199	  .offset =  0,
200	  .size =    8 * 1024 * 1024 },
201	{ .name = "Flash partition 2",
202	  .offset =  MTDPART_OFS_NEXT,
203	  .size =    MTDPART_SIZ_FULL },
204};
205		</programlisting>
206	</sect1>
207	<sect1 id="Hardware_control_functions">
208		<title>Hardware control function</title>
209		<para>
210			The hardware control function provides access to the 
211			control pins of the NAND chip(s). 
212			The access can be done by GPIO pins or by address lines.
213			If you use address lines, make sure that the timing
214			requirements are met.
215		</para>
216		<para>
217			<emphasis>GPIO based example</emphasis>
218		</para>
219		<programlisting>
220static void board_hwcontrol(struct mtd_info *mtd, int cmd)
221{
222	switch(cmd){
223		case NAND_CTL_SETCLE: /* Set CLE pin high */ break;
224		case NAND_CTL_CLRCLE: /* Set CLE pin low */ break;
225		case NAND_CTL_SETALE: /* Set ALE pin high */ break;
226		case NAND_CTL_CLRALE: /* Set ALE pin low */ break;
227		case NAND_CTL_SETNCE: /* Set nCE pin low */ break;
228		case NAND_CTL_CLRNCE: /* Set nCE pin high */ break;
229	}
230}
231		</programlisting>
232		<para>
233			<emphasis>Address lines based example.</emphasis> It's assumed that the
234			nCE pin is driven by a chip select decoder.
235		</para>
236		<programlisting>
237static void board_hwcontrol(struct mtd_info *mtd, int cmd)
238{
239	struct nand_chip *this = (struct nand_chip *) mtd->priv;
240	switch(cmd){
241		case NAND_CTL_SETCLE: this->IO_ADDR_W |= CLE_ADRR_BIT;  break;
242		case NAND_CTL_CLRCLE: this->IO_ADDR_W &amp;= ~CLE_ADRR_BIT; break;
243		case NAND_CTL_SETALE: this->IO_ADDR_W |= ALE_ADRR_BIT;  break;
244		case NAND_CTL_CLRALE: this->IO_ADDR_W &amp;= ~ALE_ADRR_BIT; break;
245	}
246}
247		</programlisting>
248	</sect1>
249	<sect1 id="Device_ready_function">
250		<title>Device ready function</title>
251		<para>
252			If the hardware interface has the ready busy pin of the NAND chip connected to a
253			GPIO or other accesible I/O pin, this function is used to read back the state of the
254			pin. The function has no arguments and should return 0, if the device is busy (R/B pin 
255			is low) and 1, if the device is ready (R/B pin is high).
256			If the hardware interface does not give access to the ready busy pin, then
257			the function must not be defined and the function pointer this->dev_ready is set to NULL.		
258		</para>
259	</sect1>
260	<sect1 id="Init_function">
261		<title>Init function</title>
262		<para>
263			The init function allocates memory and sets up all the board
264			specific parameters and function pointers. When everything
265			is set up nand_scan() is called. This function tries to
266			detect and identify then chip. If a chip is found all the
267			internal data fields are initialized accordingly.
268			The structure(s) have to be zeroed out first and then filled with the neccecary 
269			information about the device.
270		</para>
271		<programlisting>
272static int __init board_init (void)
273{
274	struct nand_chip *this;
275	int err = 0;
276
277	/* Allocate memory for MTD device structure and private data */
278	board_mtd = kzalloc(sizeof(struct mtd_info) + sizeof(struct nand_chip), GFP_KERNEL);
279	if (!board_mtd) {
280		printk ("Unable to allocate NAND MTD device structure.\n");
281		err = -ENOMEM;
282		goto out;
283	}
284
285	/* map physical address */
286	baseaddr = ioremap(CHIP_PHYSICAL_ADDRESS, 1024);
287	if (!baseaddr) {
288		printk("Ioremap to access NAND chip failed\n");
289		err = -EIO;
290		goto out_mtd;
291	}
292
293	/* Get pointer to private data */
294	this = (struct nand_chip *) ();
295	/* Link the private data with the MTD structure */
296	board_mtd->priv = this;
297
298	/* Set address of NAND IO lines */
299	this->IO_ADDR_R = baseaddr;
300	this->IO_ADDR_W = baseaddr;
301	/* Reference hardware control function */
302	this->hwcontrol = board_hwcontrol;
303	/* Set command delay time, see datasheet for correct value */
304	this->chip_delay = CHIP_DEPENDEND_COMMAND_DELAY;
305	/* Assign the device ready function, if available */
306	this->dev_ready = board_dev_ready;
307	this->eccmode = NAND_ECC_SOFT;
308
309	/* Scan to find existence of the device */
310	if (nand_scan (board_mtd, 1)) {
311		err = -ENXIO;
312		goto out_ior;
313	}
314	
315	add_mtd_partitions(board_mtd, partition_info, NUM_PARTITIONS);
316	goto out;
317
318out_ior:
319	iounmap(baseaddr);
320out_mtd:
321	kfree (board_mtd);
322out:
323	return err;
324}
325module_init(board_init);
326		</programlisting>
327	</sect1>
328	<sect1 id="Exit_function">
329		<title>Exit function</title>
330		<para>
331			The exit function is only neccecary if the driver is
332			compiled as a module. It releases all resources which
333			are held by the chip driver and unregisters the partitions
334			in the MTD layer.
335		</para>
336		<programlisting>
337#ifdef MODULE
338static void __exit board_cleanup (void)
339{
340	/* Release resources, unregister device */
341	nand_release (board_mtd);
342
343	/* unmap physical address */
344	iounmap(baseaddr);
345	
346	/* Free the MTD device structure */
347	kfree (board_mtd);
348}
349module_exit(board_cleanup);
350#endif
351		</programlisting>
352	</sect1>
353  </chapter>
354
355  <chapter id="boarddriversadvanced">
356     	<title>Advanced board driver functions</title>
357	<para>
358		This chapter describes the advanced functionality of the NAND
359		driver. For a list of functions which can be overridden by the board
360		driver see the documentation of the nand_chip structure.
361	</para>
362	<sect1 id="Multiple_chip_control">
363		<title>Multiple chip control</title>
364		<para>
365			The nand driver can control chip arrays. Therefore the
366			board driver must provide an own select_chip function. This
367			function must (de)select the requested chip.
368			The function pointer in the nand_chip structure must
369			be set before calling nand_scan(). The maxchip parameter
370			of nand_scan() defines the maximum number of chips to
371			scan for. Make sure that the select_chip function can
372			handle the requested number of chips.
373		</para>
374		<para>
375			The nand driver concatenates the chips to one virtual
376			chip and provides this virtual chip to the MTD layer.
377		</para>
378		<para>
379			<emphasis>Note: The driver can only handle linear chip arrays
380			of equally sized chips. There is no support for
381			parallel arrays which extend the buswidth.</emphasis>
382		</para>
383		<para>
384			<emphasis>GPIO based example</emphasis>
385		</para>
386		<programlisting>
387static void board_select_chip (struct mtd_info *mtd, int chip)
388{
389	/* Deselect all chips, set all nCE pins high */
390	GPIO(BOARD_NAND_NCE) |= 0xff;	
391	if (chip >= 0)
392		GPIO(BOARD_NAND_NCE) &amp;= ~ (1 &lt;&lt; chip);
393}
394		</programlisting>
395		<para>
396			<emphasis>Address lines based example.</emphasis>
397			Its assumed that the nCE pins are connected to an
398			address decoder.
399		</para>
400		<programlisting>
401static void board_select_chip (struct mtd_info *mtd, int chip)
402{
403	struct nand_chip *this = (struct nand_chip *) mtd->priv;
404	
405	/* Deselect all chips */
406	this->IO_ADDR_R &amp;= ~BOARD_NAND_ADDR_MASK;
407	this->IO_ADDR_W &amp;= ~BOARD_NAND_ADDR_MASK;
408	switch (chip) {
409	case 0:
410		this->IO_ADDR_R |= BOARD_NAND_ADDR_CHIP0;
411		this->IO_ADDR_W |= BOARD_NAND_ADDR_CHIP0;
412		break;
413	....	
414	case n:
415		this->IO_ADDR_R |= BOARD_NAND_ADDR_CHIPn;
416		this->IO_ADDR_W |= BOARD_NAND_ADDR_CHIPn;
417		break;
418	}	
419}
420		</programlisting>
421	</sect1>
422	<sect1 id="Hardware_ECC_support">
423		<title>Hardware ECC support</title>
424		<sect2 id="Functions_and_constants">
425			<title>Functions and constants</title>
426			<para>
427				The nand driver supports three different types of
428				hardware ECC.
429				<itemizedlist>
430				<listitem><para>NAND_ECC_HW3_256</para><para>
431				Hardware ECC generator providing 3 bytes ECC per
432				256 byte.
433				</para>	</listitem>
434				<listitem><para>NAND_ECC_HW3_512</para><para>
435				Hardware ECC generator providing 3 bytes ECC per
436				512 byte.
437				</para>	</listitem>
438				<listitem><para>NAND_ECC_HW6_512</para><para>
439				Hardware ECC generator providing 6 bytes ECC per
440				512 byte.
441				</para>	</listitem>
442				<listitem><para>NAND_ECC_HW8_512</para><para>
443				Hardware ECC generator providing 6 bytes ECC per
444				512 byte.
445				</para>	</listitem>
446				</itemizedlist>
447				If your hardware generator has a different functionality
448				add it at the appropriate place in nand_base.c
449			</para>
450			<para>
451				The board driver must provide following functions:
452				<itemizedlist>
453				<listitem><para>enable_hwecc</para><para>
454				This function is called before reading / writing to
455				the chip. Reset or initialize the hardware generator
456				in this function. The function is called with an
457				argument which let you distinguish between read 
458				and write operations.
459				</para>	</listitem>
460				<listitem><para>calculate_ecc</para><para>
461				This function is called after read / write from / to
462				the chip. Transfer the ECC from the hardware to
463				the buffer. If the option NAND_HWECC_SYNDROME is set
464				then the function is only called on write. See below.
465				</para>	</listitem>
466				<listitem><para>correct_data</para><para>
467				In case of an ECC error this function is called for
468				error detection and correction. Return 1 respectively 2
469				in case the error can be corrected. If the error is
470				not correctable return -1. If your hardware generator
471				matches the default algorithm of the nand_ecc software
472				generator then use the correction function provided
473				by nand_ecc instead of implementing duplicated code.
474				</para>	</listitem>
475				</itemizedlist>
476			</para>
477		</sect2>
478		<sect2 id="Hardware_ECC_with_syndrome_calculation">
479		<title>Hardware ECC with syndrome calculation</title>
480			<para>
481				Many hardware ECC implementations provide Reed-Solomon
482				codes and calculate an error syndrome on read. The syndrome
483				must be converted to a standard Reed-Solomon syndrome
484				before calling the error correction code in the generic
485				Reed-Solomon library.
486			</para>
487			<para>
488				The ECC bytes must be placed immidiately after the data
489				bytes in order to make the syndrome generator work. This
490				is contrary to the usual layout used by software ECC. The
491				separation of data and out of band area is not longer
492				possible. The nand driver code handles this layout and
493				the remaining free bytes in the oob area are managed by 
494				the autoplacement code. Provide a matching oob-layout
495				in this case. See rts_from4.c and diskonchip.c for 
496				implementation reference. In those cases we must also
497				use bad block tables on FLASH, because the ECC layout is
498				interferring with the bad block marker positions.
499				See bad block table support for details.
500			</para>
501		</sect2>
502	</sect1>
503	<sect1 id="Bad_Block_table_support">
504		<title>Bad block table support</title>
505		<para>
506			Most NAND chips mark the bad blocks at a defined
507			position in the spare area. Those blocks must 
508			not be erased under any circumstances as the bad 
509			block information would be lost.
510			It is possible to check the bad block mark each
511			time when the blocks are accessed by reading the
512			spare area of the first page in the block. This
513			is time consuming so a bad block table is used.
514		</para>
515		<para>
516			The nand driver supports various types of bad block
517			tables.
518			<itemizedlist>
519			<listitem><para>Per device</para><para>
520			The bad block table contains all bad block information
521			of the device which can consist of multiple chips.
522			</para>	</listitem>
523			<listitem><para>Per chip</para><para>
524			A bad block table is used per chip and contains the
525			bad block information for this particular chip.
526			</para>	</listitem>
527			<listitem><para>Fixed offset</para><para>
528			The bad block table is located at a fixed offset
529			in the chip (device). This applies to various
530			DiskOnChip devices.
531			</para>	</listitem>
532			<listitem><para>Automatic placed</para><para>
533			The bad block table is automatically placed and
534			detected either at the end or at the beginning
535			of a chip (device)
536			</para>	</listitem>
537			<listitem><para>Mirrored tables</para><para>
538			The bad block table is mirrored on the chip (device) to
539			allow updates of the bad block table without data loss.
540			</para>	</listitem>
541			</itemizedlist>
542		</para>
543		<para>	
544			nand_scan() calls the function nand_default_bbt(). 
545			nand_default_bbt() selects appropriate default
546			bad block table desriptors depending on the chip information
547			which was retrieved by nand_scan().
548		</para>
549		<para>
550			The standard policy is scanning the device for bad 
551			blocks and build a ram based bad block table which
552			allows faster access than always checking the
553			bad block information on the flash chip itself.
554		</para>
555		<sect2 id="Flash_based_tables">
556			<title>Flash based tables</title>
557			<para>
558				It may be desired or neccecary to keep a bad block table in FLASH. 
559				For AG-AND chips this is mandatory, as they have no factory marked
560				bad blocks. They have factory marked good blocks. The marker pattern
561				is erased when the block is erased to be reused. So in case of
562				powerloss before writing the pattern back to the chip this block 
563				would be lost and added to the bad blocks. Therefore we scan the 
564				chip(s) when we detect them the first time for good blocks and 
565				store this information in a bad block table before erasing any 
566				of the blocks.
567			</para>
568			<para>
569				The blocks in which the tables are stored are procteted against
570				accidental access by marking them bad in the memory bad block
571				table. The bad block table management functions are allowed
572				to circumvernt this protection.
573			</para>
574			<para>
575				The simplest way to activate the FLASH based bad block table support 
576				is to set the option NAND_USE_FLASH_BBT in the option field of
577				the nand chip structure before calling nand_scan(). For AG-AND
578				chips is this done by default.
579				This activates the default FLASH based bad block table functionality 
580				of the NAND driver. The default bad block table options are
581				<itemizedlist>
582				<listitem><para>Store bad block table per chip</para></listitem>
583				<listitem><para>Use 2 bits per block</para></listitem>
584				<listitem><para>Automatic placement at the end of the chip</para></listitem>
585				<listitem><para>Use mirrored tables with version numbers</para></listitem>
586				<listitem><para>Reserve 4 blocks at the end of the chip</para></listitem>
587				</itemizedlist>
588			</para>
589		</sect2>
590		<sect2 id="User_defined_tables">
591			<title>User defined tables</title>
592			<para>
593				User defined tables are created by filling out a 
594				nand_bbt_descr structure and storing the pointer in the
595				nand_chip structure member bbt_td before calling nand_scan(). 
596				If a mirror table is neccecary a second structure must be
597				created and a pointer to this structure must be stored
598				in bbt_md inside the nand_chip structure. If the bbt_md 
599				member is set to NULL then only the main table is used
600				and no scan for the mirrored table is performed.
601			</para>
602			<para>
603				The most important field in the nand_bbt_descr structure
604				is the options field. The options define most of the 
605				table properties. Use the predefined constants from
606				nand.h to define the options.
607				<itemizedlist>
608				<listitem><para>Number of bits per block</para>
609				<para>The supported number of bits is 1, 2, 4, 8.</para></listitem>
610				<listitem><para>Table per chip</para>
611				<para>Setting the constant NAND_BBT_PERCHIP selects that
612				a bad block table is managed for each chip in a chip array.
613				If this option is not set then a per device bad block table
614				is used.</para></listitem>
615				<listitem><para>Table location is absolute</para>
616				<para>Use the option constant NAND_BBT_ABSPAGE and
617				define the absolute page number where the bad block
618				table starts in the field pages. If you have selected bad block
619				tables per chip and you have a multi chip array then the start page
620				must be given for each chip in the chip array. Note: there is no scan
621				for a table ident pattern performed, so the fields 
622				pattern, veroffs, offs, len can be left uninitialized</para></listitem>
623				<listitem><para>Table location is automatically detected</para>
624				<para>The table can either be located in the first or the last good
625				blocks of the chip (device). Set NAND_BBT_LASTBLOCK to place
626				the bad block table at the end of the chip (device). The
627				bad block tables are marked and identified by a pattern which
628				is stored in the spare area of the first page in the block which
629				holds the bad block table. Store a pointer to the pattern  
630				in the pattern field. Further the length of the pattern has to be 
631				stored in len and the offset in the spare area must be given
632				in the offs member of the nand_bbt_descr stucture. For mirrored
633				bad block tables different patterns are mandatory.</para></listitem>
634				<listitem><para>Table creation</para>
635				<para>Set the option NAND_BBT_CREATE to enable the table creation
636				if no table can be found during the scan. Usually this is done only 
637				once if a new chip is found. </para></listitem>
638				<listitem><para>Table write support</para>
639				<para>Set the option NAND_BBT_WRITE to enable the table write support.
640				This allows the update of the bad block table(s) in case a block has
641				to be marked bad due to wear. The MTD interface function block_markbad
642				is calling the update function of the bad block table. If the write
643				support is enabled then the table is updated on FLASH.</para>
644				<para>
645				Note: Write support should only be enabled for mirrored tables with
646				version control.
647				</para></listitem>
648				<listitem><para>Table version control</para>
649				<para>Set the option NAND_BBT_VERSION to enable the table version control.
650				It's highly recommended to enable this for mirrored tables with write
651				support. It makes sure that the risk of loosing the bad block
652				table information is reduced to the loss of the information about the
653				one worn out block which should be marked bad. The version is stored in
654				4 consecutive bytes in the spare area of the device. The position of
655				the version number is defined by the member veroffs in the bad block table
656				descriptor.</para></listitem>
657				<listitem><para>Save block contents on write</para>
658				<para>
659				In case that the block which holds the bad block table does contain
660				other useful information, set the option NAND_BBT_SAVECONTENT. When
661				the bad block table is written then the whole block is read the bad
662				block table is updated and the block is erased and everything is 
663				written back. If this option is not set only the bad block table
664				is written and everything else in the block is ignored and erased.
665				</para></listitem>
666				<listitem><para>Number of reserved blocks</para>
667				<para>
668				For automatic placement some blocks must be reserved for
669				bad block table storage. The number of reserved blocks is defined 
670				in the maxblocks member of the babd block table description structure.
671				Reserving 4 blocks for mirrored tables should be a reasonable number. 
672				This also limits the number of blocks which are scanned for the bad
673				block table ident pattern.
674				</para></listitem>
675				</itemizedlist>
676			</para>
677		</sect2>
678	</sect1>
679	<sect1 id="Spare_area_placement">
680		<title>Spare area (auto)placement</title>
681		<para>
682			The nand driver implements different possibilities for
683			placement of filesystem data in the spare area, 
684			<itemizedlist>
685			<listitem><para>Placement defined by fs driver</para></listitem>
686			<listitem><para>Automatic placement</para></listitem>
687			</itemizedlist>
688			The default placement function is automatic placement. The
689			nand driver has built in default placement schemes for the
690			various chiptypes. If due to hardware ECC functionality the
691			default placement does not fit then the board driver can
692			provide a own placement scheme.
693		</para>
694		<para>
695			File system drivers can provide a own placement scheme which
696			is used instead of the default placement scheme.
697		</para>
698		<para>
699			Placement schemes are defined by a nand_oobinfo structure
700	     		<programlisting>
701struct nand_oobinfo {
702	int	useecc;
703	int	eccbytes;
704	int	eccpos[24];
705	int	oobfree[8][2];
706};
707	     		</programlisting>
708			<itemizedlist>
709			<listitem><para>useecc</para><para>
710				The useecc member controls the ecc and placement function. The header
711				file include/mtd/mtd-abi.h contains constants to select ecc and
712				placement. MTD_NANDECC_OFF switches off the ecc complete. This is
713				not recommended and available for testing and diagnosis only.
714				MTD_NANDECC_PLACE selects caller defined placement, MTD_NANDECC_AUTOPLACE
715				selects automatic placement.
716			</para></listitem>
717			<listitem><para>eccbytes</para><para>
718				The eccbytes member defines the number of ecc bytes per page.
719			</para></listitem>
720			<listitem><para>eccpos</para><para>
721				The eccpos array holds the byte offsets in the spare area where
722				the ecc codes are placed.
723			</para></listitem>
724			<listitem><para>oobfree</para><para>
725				The oobfree array defines the areas in the spare area which can be
726				used for automatic placement. The information is given in the format
727				{offset, size}. offset defines the start of the usable area, size the
728				length in bytes. More than one area can be defined. The list is terminated
729				by an {0, 0} entry.
730			</para></listitem>
731			</itemizedlist>
732		</para>
733		<sect2 id="Placement_defined_by_fs_driver">
734			<title>Placement defined by fs driver</title>
735			<para>
736				The calling function provides a pointer to a nand_oobinfo
737				structure which defines the ecc placement. For writes the
738				caller must provide a spare area buffer along with the
739				data buffer. The spare area buffer size is (number of pages) *
740				(size of spare area). For reads the buffer size is
741				(number of pages) * ((size of spare area) + (number of ecc
742				steps per page) * sizeof (int)). The driver stores the
743				result of the ecc check for each tuple in the spare buffer.
744				The storage sequence is 
745			</para>
746			<para>
747				&lt;spare data page 0&gt;&lt;ecc result 0&gt;...&lt;ecc result n&gt;
748			</para>
749			<para>
750				...
751			</para>
752			<para>
753				&lt;spare data page n&gt;&lt;ecc result 0&gt;...&lt;ecc result n&gt;
754			</para>
755			<para>
756				This is a legacy mode used by YAFFS1.
757			</para>
758			<para>
759				If the spare area buffer is NULL then only the ECC placement is
760				done according to the given scheme in the nand_oobinfo structure.
761			</para>
762		</sect2>
763		<sect2 id="Automatic_placement">
764			<title>Automatic placement</title>
765			<para>
766				Automatic placement uses the built in defaults to place the
767				ecc bytes in the spare area. If filesystem data have to be stored /
768				read into the spare area then the calling function must provide a
769				buffer. The buffer size per page is determined by the oobfree array in
770				the nand_oobinfo structure.
771			</para>
772			<para>
773				If the spare area buffer is NULL then only the ECC placement is
774				done according to the default builtin scheme.
775			</para>
776		</sect2>
777		<sect2 id="User_space_placement_selection">
778			<title>User space placement selection</title>
779		<para>
780			All non ecc functions like mtd->read and mtd->write use an internal 
781			structure, which can be set by an ioctl. This structure is preset 
782			to the autoplacement default.
783	     		<programlisting>
784	ioctl (fd, MEMSETOOBSEL, oobsel);
785	     		</programlisting>
786			oobsel is a pointer to a user supplied structure of type
787			nand_oobconfig. The contents of this structure must match the 
788			criteria of the filesystem, which will be used. See an example in utils/nandwrite.c.
789		</para>
790		</sect2>
791	</sect1>	
792	<sect1 id="Spare_area_autoplacement_default">
793		<title>Spare area autoplacement default schemes</title>
794		<sect2 id="pagesize_256">
795			<title>256 byte pagesize</title>
796<informaltable><tgroup cols="3"><tbody>
797<row>
798<entry>Offset</entry>
799<entry>Content</entry>
800<entry>Comment</entry>
801</row>
802<row>
803<entry>0x00</entry>
804<entry>ECC byte 0</entry>
805<entry>Error correction code byte 0</entry>
806</row>
807<row>
808<entry>0x01</entry>
809<entry>ECC byte 1</entry>
810<entry>Error correction code byte 1</entry>
811</row>
812<row>
813<entry>0x02</entry>
814<entry>ECC byte 2</entry>
815<entry>Error correction code byte 2</entry>
816</row>
817<row>
818<entry>0x03</entry>
819<entry>Autoplace 0</entry>
820<entry></entry>
821</row>
822<row>
823<entry>0x04</entry>
824<entry>Autoplace 1</entry>
825<entry></entry>
826</row>
827<row>
828<entry>0x05</entry>
829<entry>Bad block marker</entry>
830<entry>If any bit in this byte is zero, then this block is bad.
831This applies only to the first page in a block. In the remaining
832pages this byte is reserved</entry>
833</row>
834<row>
835<entry>0x06</entry>
836<entry>Autoplace 2</entry>
837<entry></entry>
838</row>
839<row>
840<entry>0x07</entry>
841<entry>Autoplace 3</entry>
842<entry></entry>
843</row>
844</tbody></tgroup></informaltable>
845		</sect2>
846		<sect2 id="pagesize_512">
847			<title>512 byte pagesize</title>
848<informaltable><tgroup cols="3"><tbody>
849<row>
850<entry>Offset</entry>
851<entry>Content</entry>
852<entry>Comment</entry>
853</row>
854<row>
855<entry>0x00</entry>
856<entry>ECC byte 0</entry>
857<entry>Error correction code byte 0 of the lower 256 Byte data in
858this page</entry>
859</row>
860<row>
861<entry>0x01</entry>
862<entry>ECC byte 1</entry>
863<entry>Error correction code byte 1 of the lower 256 Bytes of data
864in this page</entry>
865</row>
866<row>
867<entry>0x02</entry>
868<entry>ECC byte 2</entry>
869<entry>Error correction code byte 2 of the lower 256 Bytes of data
870in this page</entry>
871</row>
872<row>
873<entry>0x03</entry>
874<entry>ECC byte 3</entry>
875<entry>Error correction code byte 0 of the upper 256 Bytes of data
876in this page</entry>
877</row>
878<row>
879<entry>0x04</entry>
880<entry>reserved</entry>
881<entry>reserved</entry>
882</row>
883<row>
884<entry>0x05</entry>
885<entry>Bad block marker</entry>
886<entry>If any bit in this byte is zero, then this block is bad.
887This applies only to the first page in a block. In the remaining
888pages this byte is reserved</entry>
889</row>
890<row>
891<entry>0x06</entry>
892<entry>ECC byte 4</entry>
893<entry>Error correction code byte 1 of the upper 256 Bytes of data
894in this page</entry>
895</row>
896<row>
897<entry>0x07</entry>
898<entry>ECC byte 5</entry>
899<entry>Error correction code byte 2 of the upper 256 Bytes of data
900in this page</entry>
901</row>
902<row>
903<entry>0x08 - 0x0F</entry>
904<entry>Autoplace 0 - 7</entry>
905<entry></entry>
906</row>
907</tbody></tgroup></informaltable>
908		</sect2>
909		<sect2 id="pagesize_2048">
910			<title>2048 byte pagesize</title>
911<informaltable><tgroup cols="3"><tbody>
912<row>
913<entry>Offset</entry>
914<entry>Content</entry>
915<entry>Comment</entry>
916</row>
917<row>
918<entry>0x00</entry>
919<entry>Bad block marker</entry>
920<entry>If any bit in this byte is zero, then this block is bad.
921This applies only to the first page in a block. In the remaining
922pages this byte is reserved</entry>
923</row>
924<row>
925<entry>0x01</entry>
926<entry>Reserved</entry>
927<entry>Reserved</entry>
928</row>
929<row>
930<entry>0x02-0x27</entry>
931<entry>Autoplace 0 - 37</entry>
932<entry></entry>
933</row>
934<row>
935<entry>0x28</entry>
936<entry>ECC byte 0</entry>
937<entry>Error correction code byte 0 of the first 256 Byte data in
938this page</entry>
939</row>
940<row>
941<entry>0x29</entry>
942<entry>ECC byte 1</entry>
943<entry>Error correction code byte 1 of the first 256 Bytes of data
944in this page</entry>
945</row>
946<row>
947<entry>0x2A</entry>
948<entry>ECC byte 2</entry>
949<entry>Error correction code byte 2 of the first 256 Bytes data in
950this page</entry>
951</row>
952<row>
953<entry>0x2B</entry>
954<entry>ECC byte 3</entry>
955<entry>Error correction code byte 0 of the second 256 Bytes of data
956in this page</entry>
957</row>
958<row>
959<entry>0x2C</entry>
960<entry>ECC byte 4</entry>
961<entry>Error correction code byte 1 of the second 256 Bytes of data
962in this page</entry>
963</row>
964<row>
965<entry>0x2D</entry>
966<entry>ECC byte 5</entry>
967<entry>Error correction code byte 2 of the second 256 Bytes of data
968in this page</entry>
969</row>
970<row>
971<entry>0x2E</entry>
972<entry>ECC byte 6</entry>
973<entry>Error correction code byte 0 of the third 256 Bytes of data
974in this page</entry>
975</row>
976<row>
977<entry>0x2F</entry>
978<entry>ECC byte 7</entry>
979<entry>Error correction code byte 1 of the third 256 Bytes of data
980in this page</entry>
981</row>
982<row>
983<entry>0x30</entry>
984<entry>ECC byte 8</entry>
985<entry>Error correction code byte 2 of the third 256 Bytes of data
986in this page</entry>
987</row>
988<row>
989<entry>0x31</entry>
990<entry>ECC byte 9</entry>
991<entry>Error correction code byte 0 of the fourth 256 Bytes of data
992in this page</entry>
993</row>
994<row>
995<entry>0x32</entry>
996<entry>ECC byte 10</entry>
997<entry>Error correction code byte 1 of the fourth 256 Bytes of data
998in this page</entry>
999</row>
1000<row>
1001<entry>0x33</entry>
1002<entry>ECC byte 11</entry>
1003<entry>Error correction code byte 2 of the fourth 256 Bytes of data
1004in this page</entry>
1005</row>
1006<row>
1007<entry>0x34</entry>
1008<entry>ECC byte 12</entry>
1009<entry>Error correction code byte 0 of the fifth 256 Bytes of data
1010in this page</entry>
1011</row>
1012<row>
1013<entry>0x35</entry>
1014<entry>ECC byte 13</entry>
1015<entry>Error correction code byte 1 of the fifth 256 Bytes of data
1016in this page</entry>
1017</row>
1018<row>
1019<entry>0x36</entry>
1020<entry>ECC byte 14</entry>
1021<entry>Error correction code byte 2 of the fifth 256 Bytes of data
1022in this page</entry>
1023</row>
1024<row>
1025<entry>0x37</entry>
1026<entry>ECC byte 15</entry>
1027<entry>Error correction code byte 0 of the sixt 256 Bytes of data
1028in this page</entry>
1029</row>
1030<row>
1031<entry>0x38</entry>
1032<entry>ECC byte 16</entry>
1033<entry>Error correction code byte 1 of the sixt 256 Bytes of data
1034in this page</entry>
1035</row>
1036<row>
1037<entry>0x39</entry>
1038<entry>ECC byte 17</entry>
1039<entry>Error correction code byte 2 of the sixt 256 Bytes of data
1040in this page</entry>
1041</row>
1042<row>
1043<entry>0x3A</entry>
1044<entry>ECC byte 18</entry>
1045<entry>Error correction code byte 0 of the seventh 256 Bytes of
1046data in this page</entry>
1047</row>
1048<row>
1049<entry>0x3B</entry>
1050<entry>ECC byte 19</entry>
1051<entry>Error correction code byte 1 of the seventh 256 Bytes of
1052data in this page</entry>
1053</row>
1054<row>
1055<entry>0x3C</entry>
1056<entry>ECC byte 20</entry>
1057<entry>Error correction code byte 2 of the seventh 256 Bytes of
1058data in this page</entry>
1059</row>
1060<row>
1061<entry>0x3D</entry>
1062<entry>ECC byte 21</entry>
1063<entry>Error correction code byte 0 of the eigth 256 Bytes of data
1064in this page</entry>
1065</row>
1066<row>
1067<entry>0x3E</entry>
1068<entry>ECC byte 22</entry>
1069<entry>Error correction code byte 1 of the eigth 256 Bytes of data
1070in this page</entry>
1071</row>
1072<row>
1073<entry>0x3F</entry>
1074<entry>ECC byte 23</entry>
1075<entry>Error correction code byte 2 of the eigth 256 Bytes of data
1076in this page</entry>
1077</row>
1078</tbody></tgroup></informaltable>
1079		</sect2>
1080     	</sect1>
1081  </chapter>
1082
1083  <chapter id="filesystems">
1084     	<title>Filesystem support</title>
1085	<para>
1086		The NAND driver provides all neccecary functions for a
1087		filesystem via the MTD interface.
1088	</para>
1089	<para>
1090		Filesystems must be aware of the NAND pecularities and
1091		restrictions. One major restrictions of NAND Flash is, that you cannot 
1092		write as often as you want to a page. The consecutive writes to a page, 
1093		before erasing it again, are restricted to 1-3 writes, depending on the 
1094		manufacturers specifications. This applies similar to the spare area. 
1095	</para>
1096	<para>
1097		Therefore NAND aware filesystems must either write in page size chunks
1098		or hold a writebuffer to collect smaller writes until they sum up to 
1099		pagesize. Available NAND aware filesystems: JFFS2, YAFFS. 		
1100	</para>
1101	<para>
1102		The spare area usage to store filesystem data is controlled by
1103		the spare area placement functionality which is described in one
1104		of the earlier chapters.
1105	</para>
1106  </chapter>	
1107  <chapter id="tools">
1108     	<title>Tools</title>
1109	<para>
1110		The MTD project provides a couple of helpful tools to handle NAND Flash.
1111		<itemizedlist>
1112		<listitem><para>flasherase, flasheraseall: Erase and format FLASH partitions</para></listitem>
1113		<listitem><para>nandwrite: write filesystem images to NAND FLASH</para></listitem>
1114		<listitem><para>nanddump: dump the contents of a NAND FLASH partitions</para></listitem>
1115		</itemizedlist>
1116	</para>
1117	<para>
1118		These tools are aware of the NAND restrictions. Please use those tools
1119		instead of complaining about errors which are caused by non NAND aware
1120		access methods.
1121	</para>
1122  </chapter>	
1123
1124  <chapter id="defines">
1125     <title>Constants</title>
1126     <para>
1127     This chapter describes the constants which might be relevant for a driver developer.
1128     </para>
1129     <sect1 id="Chip_option_constants">
1130	<title>Chip option constants</title>
1131     	<sect2 id="Constants_for_chip_id_table">
1132		<title>Constants for chip id table</title>
1133     		<para>
1134		These constants are defined in nand.h. They are ored together to describe
1135		the chip functionality.
1136     		<programlisting>
1137/* Chip can not auto increment pages */
1138#define NAND_NO_AUTOINCR	0x00000001
1139/* Buswitdh is 16 bit */
1140#define NAND_BUSWIDTH_16	0x00000002
1141/* Device supports partial programming without padding */
1142#define NAND_NO_PADDING		0x00000004
1143/* Chip has cache program function */
1144#define NAND_CACHEPRG		0x00000008
1145/* Chip has copy back function */
1146#define NAND_COPYBACK		0x00000010
1147/* AND Chip which has 4 banks and a confusing page / block 
1148 * assignment. See Renesas datasheet for further information */
1149#define NAND_IS_AND		0x00000020
1150/* Chip has a array of 4 pages which can be read without
1151 * additional ready /busy waits */
1152#define NAND_4PAGE_ARRAY	0x00000040 
1153		</programlisting>
1154     		</para>
1155     	</sect2>
1156     	<sect2 id="Constants_for_runtime_options">
1157		<title>Constants for runtime options</title>
1158     		<para>
1159		These constants are defined in nand.h. They are ored together to describe
1160		the functionality.
1161     		<programlisting>
1162/* Use a flash based bad block table. This option is parsed by the
1163 * default bad block table function (nand_default_bbt). */
1164#define NAND_USE_FLASH_BBT	0x00010000
1165/* The hw ecc generator provides a syndrome instead a ecc value on read 
1166 * This can only work if we have the ecc bytes directly behind the 
1167 * data bytes. Applies for DOC and AG-AND Renesas HW Reed Solomon generators */
1168#define NAND_HWECC_SYNDROME	0x00020000
1169		</programlisting>
1170     		</para>
1171     	</sect2>
1172     </sect1>	
1173
1174     <sect1 id="EEC_selection_constants">
1175	<title>ECC selection constants</title>
1176	<para>
1177	Use these constants to select the ECC algorithm.
1178  	<programlisting>
1179/* No ECC. Usage is not recommended ! */
1180#define NAND_ECC_NONE		0
1181/* Software ECC 3 byte ECC per 256 Byte data */
1182#define NAND_ECC_SOFT		1
1183/* Hardware ECC 3 byte ECC per 256 Byte data */
1184#define NAND_ECC_HW3_256	2
1185/* Hardware ECC 3 byte ECC per 512 Byte data */
1186#define NAND_ECC_HW3_512	3
1187/* Hardware ECC 6 byte ECC per 512 Byte data */
1188#define NAND_ECC_HW6_512	4
1189/* Hardware ECC 6 byte ECC per 512 Byte data */
1190#define NAND_ECC_HW8_512	6
1191	</programlisting>
1192	</para>
1193     </sect1>	
1194
1195     <sect1 id="Hardware_control_related_constants">
1196	<title>Hardware control related constants</title>
1197	<para>
1198	These constants describe the requested hardware access function when
1199	the boardspecific hardware control function is called
1200  	<programlisting>
1201/* Select the chip by setting nCE to low */
1202#define NAND_CTL_SETNCE 	1
1203/* Deselect the chip by setting nCE to high */
1204#define NAND_CTL_CLRNCE		2
1205/* Select the command latch by setting CLE to high */
1206#define NAND_CTL_SETCLE		3
1207/* Deselect the command latch by setting CLE to low */
1208#define NAND_CTL_CLRCLE		4
1209/* Select the address latch by setting ALE to high */
1210#define NAND_CTL_SETALE		5
1211/* Deselect the address latch by setting ALE to low */
1212#define NAND_CTL_CLRALE		6
1213/* Set write protection by setting WP to high. Not used! */
1214#define NAND_CTL_SETWP		7
1215/* Clear write protection by setting WP to low. Not used! */
1216#define NAND_CTL_CLRWP		8
1217	</programlisting>
1218	</para>
1219     </sect1>	
1220
1221     <sect1 id="Bad_block_table_constants">
1222	<title>Bad block table related constants</title>
1223	<para>
1224	These constants describe the options used for bad block
1225	table descriptors.
1226  	<programlisting>
1227/* Options for the bad block table descriptors */
1228
1229/* The number of bits used per block in the bbt on the device */
1230#define NAND_BBT_NRBITS_MSK	0x0000000F
1231#define NAND_BBT_1BIT		0x00000001
1232#define NAND_BBT_2BIT		0x00000002
1233#define NAND_BBT_4BIT		0x00000004
1234#define NAND_BBT_8BIT		0x00000008
1235/* The bad block table is in the last good block of the device */
1236#define	NAND_BBT_LASTBLOCK	0x00000010
1237/* The bbt is at the given page, else we must scan for the bbt */
1238#define NAND_BBT_ABSPAGE	0x00000020
1239/* The bbt is at the given page, else we must scan for the bbt */
1240#define NAND_BBT_SEARCH		0x00000040
1241/* bbt is stored per chip on multichip devices */
1242#define NAND_BBT_PERCHIP	0x00000080
1243/* bbt has a version counter at offset veroffs */
1244#define NAND_BBT_VERSION	0x00000100
1245/* Create a bbt if none axists */
1246#define NAND_BBT_CREATE		0x00000200
1247/* Search good / bad pattern through all pages of a block */
1248#define NAND_BBT_SCANALLPAGES	0x00000400
1249/* Scan block empty during good / bad block scan */
1250#define NAND_BBT_SCANEMPTY	0x00000800
1251/* Write bbt if neccecary */
1252#define NAND_BBT_WRITE		0x00001000
1253/* Read and write back block contents when writing bbt */
1254#define NAND_BBT_SAVECONTENT	0x00002000
1255	</programlisting>
1256	</para>
1257     </sect1>	
1258
1259  </chapter>
1260  	
1261  <chapter id="structs">
1262     <title>Structures</title>
1263     <para>
1264     This chapter contains the autogenerated documentation of the structures which are
1265     used in the NAND driver and might be relevant for a driver developer. Each  
1266     struct member has a short description which is marked with an [XXX] identifier.
1267     See the chapter "Documentation hints" for an explanation.
1268     </para>
1269!Iinclude/linux/mtd/nand.h
1270  </chapter>
1271
1272  <chapter id="pubfunctions">
1273     <title>Public Functions Provided</title>
1274     <para>
1275     This chapter contains the autogenerated documentation of the NAND kernel API functions
1276      which are exported. Each function has a short description which is marked with an [XXX] identifier.
1277     See the chapter "Documentation hints" for an explanation.
1278     </para>
1279!Edrivers/mtd/nand/nand_base.c
1280!Edrivers/mtd/nand/nand_bbt.c
1281!Edrivers/mtd/nand/nand_ecc.c
1282  </chapter>
1283  
1284  <chapter id="intfunctions">
1285     <title>Internal Functions Provided</title>
1286     <para>
1287     This chapter contains the autogenerated documentation of the NAND driver internal functions.
1288     Each function has a short description which is marked with an [XXX] identifier.
1289     See the chapter "Documentation hints" for an explanation.
1290     The functions marked with [DEFAULT] might be relevant for a board driver developer.
1291     </para>
1292!Idrivers/mtd/nand/nand_base.c
1293!Idrivers/mtd/nand/nand_bbt.c
1294<!-- No internal functions for kernel-doc:
1295X!Idrivers/mtd/nand/nand_ecc.c
1296-->
1297  </chapter>
1298
1299  <chapter id="credits">
1300     <title>Credits</title>
1301	<para>
1302		The following people have contributed to the NAND driver:
1303		<orderedlist>
1304			<listitem><para>Steven J. Hill<email>sjhill@realitydiluted.com</email></para></listitem>
1305			<listitem><para>David Woodhouse<email>dwmw2@infradead.org</email></para></listitem>
1306			<listitem><para>Thomas Gleixner<email>tglx@linutronix.de</email></para></listitem>
1307		</orderedlist>
1308		A lot of users have provided bugfixes, improvements and helping hands for testing.
1309		Thanks a lot.
1310	</para>
1311	<para>
1312		The following people have contributed to this document:
1313		<orderedlist>
1314			<listitem><para>Thomas Gleixner<email>tglx@linutronix.de</email></para></listitem>
1315		</orderedlist>
1316	</para>
1317  </chapter>
1318</book>
1319