raid6main.c

来自「Linux Kernel 2.6.9 for OMAP1710」· C语言 代码 · 共 2,095 行 · 第 1/4 页

			unsigned int * pd_idx, raid6_conf_t *conf)
{
	long stripe;
	unsigned long chunk_number;
	unsigned int chunk_offset;
	sector_t new_sector;
	int sectors_per_chunk = conf->chunk_size >> 9;

	/* First compute the information on this sector */

	/*
	 * Compute the chunk number and the sector offset inside the chunk
	 */
	chunk_offset = sector_div(r_sector, sectors_per_chunk);
	chunk_number = r_sector;
	if ( r_sector != chunk_number ) {
		printk(KERN_CRIT "raid6: ERROR: r_sector = %llu, chunk_number = %lu\n",
		       (unsigned long long)r_sector, (unsigned long)chunk_number);
		BUG();
	}

	/*
	 * Compute the stripe number
	 */
	stripe = chunk_number / data_disks;

	/*
	 * Compute the data disk and parity disk indexes inside the stripe
	 */
	*dd_idx = chunk_number % data_disks;

	/*
	 * Select the parity disk based on the user selected algorithm.
	 */

	/**** FIX THIS ****/
	switch (conf->algorithm) {
	case ALGORITHM_LEFT_ASYMMETRIC:
		*pd_idx = raid_disks - 1 - (stripe % raid_disks);
		if (*pd_idx == raid_disks-1)
		  	(*dd_idx)++; 	/* Q D D D P */
		else if (*dd_idx >= *pd_idx)
		  	(*dd_idx) += 2; /* D D P Q D */
		break;
	case ALGORITHM_RIGHT_ASYMMETRIC:
		*pd_idx = stripe % raid_disks;
		if (*pd_idx == raid_disks-1)
		  	(*dd_idx)++; 	/* Q D D D P */
		else if (*dd_idx >= *pd_idx)
		  	(*dd_idx) += 2; /* D D P Q D */
		break;
	case ALGORITHM_LEFT_SYMMETRIC:
		*pd_idx = raid_disks - 1 - (stripe % raid_disks);
		*dd_idx = (*pd_idx + 2 + *dd_idx) % raid_disks;
		break;
	case ALGORITHM_RIGHT_SYMMETRIC:
		*pd_idx = stripe % raid_disks;
		*dd_idx = (*pd_idx + 2 + *dd_idx) % raid_disks;
		break;
	default:
		printk (KERN_CRIT "raid6: unsupported algorithm %d\n",
			conf->algorithm);
	}

	PRINTK("raid6: chunk_number = %lu, pd_idx = %u, dd_idx = %u\n",
	       chunk_number, *pd_idx, *dd_idx);

	/*
	 * Finally, compute the new sector number
	 */
	new_sector = (sector_t) stripe * sectors_per_chunk + chunk_offset;
	return new_sector;
}


static sector_t compute_blocknr(struct stripe_head *sh, int i)
{
	raid6_conf_t *conf = sh->raid_conf;
	int raid_disks = conf->raid_disks, data_disks = raid_disks - 2;
	sector_t new_sector = sh->sector, check;
	int sectors_per_chunk = conf->chunk_size >> 9;
	sector_t stripe;
	int chunk_offset;
	int chunk_number, dummy1, dummy2, dd_idx = i;
	sector_t r_sector;
	int i0 = i;

	chunk_offset = sector_div(new_sector, sectors_per_chunk);
	stripe = new_sector;
	if ( new_sector != stripe ) {
		printk(KERN_CRIT "raid6: ERROR: new_sector = %llu, stripe = %lu\n",
		       (unsigned long long)new_sector, (unsigned long)stripe);
		BUG();
	}

	switch (conf->algorithm) {
		case ALGORITHM_LEFT_ASYMMETRIC:
		case ALGORITHM_RIGHT_ASYMMETRIC:
		  	if (sh->pd_idx == raid_disks-1)
				i--; 	/* Q D D D P */
			else if (i > sh->pd_idx)
				i -= 2; /* D D P Q D */
			break;
		case ALGORITHM_LEFT_SYMMETRIC:
		case ALGORITHM_RIGHT_SYMMETRIC:
			if (sh->pd_idx == raid_disks-1)
				i--; /* Q D D D P */
			else {
				/* D D P Q D */
				if (i < sh->pd_idx)
					i += raid_disks;
				i -= (sh->pd_idx + 2);
			}
			break;
		default:
			printk (KERN_CRIT "raid6: unsupported algorithm %d\n",
				conf->algorithm);
	}

	PRINTK("raid6: compute_blocknr: pd_idx = %u, i0 = %u, i = %u\n", sh->pd_idx, i0, i);

	chunk_number = stripe * data_disks + i;
	r_sector = (sector_t)chunk_number * sectors_per_chunk + chunk_offset;

	check = raid6_compute_sector (r_sector, raid_disks, data_disks, &dummy1, &dummy2, conf);
	if (check != sh->sector || dummy1 != dd_idx || dummy2 != sh->pd_idx) {
		printk(KERN_CRIT "raid6: compute_blocknr: map not correct\n");
		return 0;
	}
	return r_sector;
}



/*
 * Copy data between a page in the stripe cache, and one or more bion
 * The page could align with the middle of the bio, or there could be
 * several bion, each with several bio_vecs, which cover part of the page
 * Multiple bion are linked together on bi_next.  There may be extras
 * at the end of this list.  We ignore them.
 */
static void copy_data(int frombio, struct bio *bio,
		     struct page *page,
		     sector_t sector)
{
	char *pa = page_address(page);
	struct bio_vec *bvl;
	int i;

	for (;bio && bio->bi_sector < sector+STRIPE_SECTORS;
	      bio = r5_next_bio(bio, sector) ) {
		int page_offset;
		if (bio->bi_sector >= sector)
			page_offset = (signed)(bio->bi_sector - sector) * 512;
		else
			page_offset = (signed)(sector - bio->bi_sector) * -512;
		bio_for_each_segment(bvl, bio, i) {
			int len = bio_iovec_idx(bio,i)->bv_len;
			int clen;
			int b_offset = 0;

			if (page_offset < 0) {
				b_offset = -page_offset;
				page_offset += b_offset;
				len -= b_offset;
			}

			if (len > 0 && page_offset + len > STRIPE_SIZE)
				clen = STRIPE_SIZE - page_offset;
			else clen = len;

			if (clen > 0) {
				char *ba = __bio_kmap_atomic(bio, i, KM_USER0);
				if (frombio)
					memcpy(pa+page_offset, ba+b_offset, clen);
				else
					memcpy(ba+b_offset, pa+page_offset, clen);
				__bio_kunmap_atomic(ba, KM_USER0);
			}
			if (clen < len) /* hit end of page */
				break;
			page_offset +=  len;
		}
	}
}

#define check_xor() 	do { 						\
			   if (count == MAX_XOR_BLOCKS) {		\
				xor_block(count, STRIPE_SIZE, ptr);	\
				count = 1;				\
			   }						\
			} while(0)

/* Compute P and Q syndromes */
static void compute_parity(struct stripe_head *sh, int method)
{
	raid6_conf_t *conf = sh->raid_conf;
	int i, pd_idx = sh->pd_idx, qd_idx, d0_idx, disks = conf->raid_disks, count;
	struct bio *chosen;
	/**** FIX THIS: This could be very bad if disks is close to 256 ****/
	void *ptrs[disks];

	qd_idx = raid6_next_disk(pd_idx, disks);
	d0_idx = raid6_next_disk(qd_idx, disks);

	PRINTK("compute_parity, stripe %llu, method %d\n",
		(unsigned long long)sh->sector, method);

	switch(method) {
	case READ_MODIFY_WRITE:
		BUG();		/* READ_MODIFY_WRITE N/A for RAID-6 */
	case RECONSTRUCT_WRITE:
	case UPDATE_PARITY:	/* Is this right? */
		for (i= disks; i-- ;)
			if ( i != pd_idx && i != qd_idx && sh->dev[i].towrite ) {
				chosen = sh->dev[i].towrite;
				sh->dev[i].towrite = NULL;
				if (sh->dev[i].written) BUG();
				sh->dev[i].written = chosen;
			}
		break;
	case CHECK_PARITY:
		BUG();		/* Not implemented yet */
	}

	for (i = disks; i--;)
		if (sh->dev[i].written) {
			sector_t sector = sh->dev[i].sector;
			struct bio *wbi = sh->dev[i].written;
			while (wbi && wbi->bi_sector < sector + STRIPE_SECTORS) {
				copy_data(1, wbi, sh->dev[i].page, sector);
				wbi = r5_next_bio(wbi, sector);
			}

			set_bit(R5_LOCKED, &sh->dev[i].flags);
			set_bit(R5_UPTODATE, &sh->dev[i].flags);
		}

//	switch(method) {
//	case RECONSTRUCT_WRITE:
//	case CHECK_PARITY:
//	case UPDATE_PARITY:
		/* Note that unlike RAID-5, the ordering of the disks matters greatly. */
		/* FIX: Is this ordering of drives even remotely optimal? */
		count = 0;
		i = d0_idx;
		do {
			ptrs[count++] = page_address(sh->dev[i].page);

			i = raid6_next_disk(i, disks);
		} while ( i != d0_idx );
//		break;
//	}

	raid6_call.gen_syndrome(disks, STRIPE_SIZE, ptrs);

	switch(method) {
	case RECONSTRUCT_WRITE:
		set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
		set_bit(R5_UPTODATE, &sh->dev[qd_idx].flags);
		set_bit(R5_LOCKED,   &sh->dev[pd_idx].flags);
		set_bit(R5_LOCKED,   &sh->dev[qd_idx].flags);
		break;
	case UPDATE_PARITY:
		set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
		set_bit(R5_UPTODATE, &sh->dev[qd_idx].flags);
		break;
	}
}

/* Compute one missing block */
static void compute_block_1(struct stripe_head *sh, int dd_idx)
{
	raid6_conf_t *conf = sh->raid_conf;
	int i, count, disks = conf->raid_disks;
	void *ptr[MAX_XOR_BLOCKS], *p;
	int pd_idx = sh->pd_idx;
	int qd_idx = raid6_next_disk(pd_idx, disks);

	PRINTK("compute_block_1, stripe %llu, idx %d\n",
		(unsigned long long)sh->sector, dd_idx);

	if ( dd_idx == qd_idx ) {
		/* We're actually computing the Q drive */
		compute_parity(sh, UPDATE_PARITY);
	} else {
		ptr[0] = page_address(sh->dev[dd_idx].page);
		memset(ptr[0], 0, STRIPE_SIZE);
		count = 1;
		for (i = disks ; i--; ) {
			if (i == dd_idx || i == qd_idx)
				continue;
			p = page_address(sh->dev[i].page);
			if (test_bit(R5_UPTODATE, &sh->dev[i].flags))
				ptr[count++] = p;
			else
				PRINTK("compute_block() %d, stripe %llu, %d"
				       " not present\n", dd_idx,
				       (unsigned long long)sh->sector, i);

			check_xor();
		}
		if (count != 1)
			xor_block(count, STRIPE_SIZE, ptr);
		set_bit(R5_UPTODATE, &sh->dev[dd_idx].flags);
	}
}

/* Compute two missing blocks */
static void compute_block_2(struct stripe_head *sh, int dd_idx1, int dd_idx2)
{
	raid6_conf_t *conf = sh->raid_conf;
	int i, count, disks = conf->raid_disks;
	int pd_idx = sh->pd_idx;
	int qd_idx = raid6_next_disk(pd_idx, disks);
	int d0_idx = raid6_next_disk(qd_idx, disks);
	int faila, failb;

	/* faila and failb are disk numbers relative to d0_idx */
	/* pd_idx become disks-2 and qd_idx become disks-1 */
	faila = (dd_idx1 < d0_idx) ? dd_idx1+(disks-d0_idx) : dd_idx1-d0_idx;
	failb = (dd_idx2 < d0_idx) ? dd_idx2+(disks-d0_idx) : dd_idx2-d0_idx;

	BUG_ON(faila == failb);
	if ( failb < faila ) { int tmp = faila; faila = failb; failb = tmp; }

	PRINTK("compute_block_2, stripe %llu, idx %d,%d (%d,%d)\n",
	       (unsigned long long)sh->sector, dd_idx1, dd_idx2, faila, failb);

	if ( failb == disks-1 ) {
		/* Q disk is one of the missing disks */
		if ( faila == disks-2 ) {
			/* Missing P+Q, just recompute */
			compute_parity(sh, UPDATE_PARITY);
			return;
		} else {
			/* We're missing D+Q; recompute D from P */
			compute_block_1(sh, (dd_idx1 == qd_idx) ? dd_idx2 : dd_idx1);
			compute_parity(sh, UPDATE_PARITY); /* Is this necessary? */
			return;
		}
	}

	/* We're missing D+P or D+D; build pointer table */
	{
		/**** FIX THIS: This could be very bad if disks is close to 256 ****/
		void *ptrs[disks];

		count = 0;
		i = d0_idx;
		do {
			ptrs[count++] = page_address(sh->dev[i].page);
			i = raid6_next_disk(i, disks);
		} while ( i != d0_idx );

		if ( failb == disks-2 ) {
			/* We're missing D+P. */
			raid6_datap_recov(disks, STRIPE_SIZE, faila, ptrs);
		} else {
			/* We're missing D+D. */
			raid6_2data_recov(disks, STRIPE_SIZE, faila, failb, ptrs);
		}

		/* Both the above update both missing blocks */
		set_bit(R5_UPTODATE, &sh->dev[dd_idx1].flags);
		set_bit(R5_UPTODATE, &sh->dev[dd_idx2].flags);
	}
}


/*
 * Each stripe/dev can have one or more bion attached.
 * toread/towrite point to the first in a chain.
 * The bi_next chain must be in order.
 */
static void add_stripe_bio (struct stripe_head *sh, struct bio *bi, int dd_idx, int forwrite)
{
	struct bio **bip;
	raid6_conf_t *conf = sh->raid_conf;

	PRINTK("adding bh b#%llu to stripe s#%llu\n",
		(unsigned long long)bi->bi_sector,
		(unsigned long long)sh->sector);


	spin_lock(&sh->lock);
	spin_lock_irq(&conf->device_lock);
	if (forwrite)
		bip = &sh->dev[dd_idx].towrite;
	else
		bip = &sh->dev[dd_idx].toread;
	while (*bip && (*bip)->bi_sector < bi->bi_sector) {
		BUG_ON((*bip)->bi_sector + ((*bip)->bi_size >> 9) > bi->bi_sector);
		bip = & (*bip)->bi_next;
	}
/* FIXME do I need to worry about overlapping bion */
	if (*bip && bi->bi_next && (*bip) != bi->bi_next)
		BUG();
	if (*bip)
		bi->bi_next = *bip;
	*bip = bi;
	bi->bi_phys_segments ++;
	spin_unlock_irq(&conf->device_lock);
	spin_unlock(&sh->lock);

	PRINTK("added bi b#%llu to stripe s#%llu, disk %d.\n",
		(unsigned long long)bi->bi_sector,
		(unsigned long long)sh->sector, dd_idx);

	if (forwrite) {
		/* check if page is coverred */
		sector_t sector = sh->dev[dd_idx].sector;
		for (bi=sh->dev[dd_idx].towrite;
		     sector < sh->dev[dd_idx].sector + STRIPE_SECTORS &&
			     bi && bi->bi_sector <= sector;
		     bi = r5_next_bio(bi, sh->dev[dd_idx].sector)) {
			if (bi->bi_sector + (bi->bi_size>>9) >= sector)
				sector = bi->bi_sector + (bi->bi_size>>9);
		}
		if (sector >= sh->dev[dd_idx].sector + STRIPE_SECTORS)
			set_bit(R5_OVERWRITE, &sh->dev[dd_idx].flags);
	}
}


/*
 * handle_stripe - do things to a stripe.
 *
 * We lock the stripe and then examine the state of various bits
 * to see what needs to be done.
 * Possible results:
 *    return some read request which now have data
 *    return some write requests which are safely on disc
 *    schedule a read on some buffers
 *    schedule a write of some buffers
 *    return confirmation of parity correctness
 *
 * Parity calculations are done inside the stripe lock
 * buffers are taken off read_list or write_list, and bh_cache buffers
 * get BH_Lock set before the stripe lock is released.
 *
 */

static void handle_stripe(struct stripe_head *sh)
{
	raid6_conf_t *conf = sh->raid_conf;
	int disks = conf->raid_disks;
	struct bio *return_bi= NULL;
	struct bio *bi;
	int i;
	int syncing;
	int locked=0, uptodate=0, to_read=0, to_write=0, failed=0, written=0;
	int non_overwrite = 0;
	int failed_num[2] = {0, 0};
	struct r5dev *dev, *pdev, *qdev;
	int pd_idx = sh->pd_idx;
	int qd_idx = raid6_next_disk(pd_idx, disks);
	int p_failed, q_failed;

	PRINTK("handling stripe %llu, state=%#lx cnt=%d, pd_idx=%d, qd_idx=%d\n",
	       (unsigned long long)sh->sector, sh->state, atomic_read(&sh->count),
	       pd_idx, qd_idx);

	spin_lock(&sh->lock);
	clear_bit(STRIPE_HANDLE, &sh->state);
	clear_bit(STRIPE_DELAYED, &sh->state);

	syncing = test_bit(STRIPE_SYNCING, &sh->state);
	/* Now to look around and see what can be done */

	for (i=disks; i--; ) {
		mdk_rdev_t *rdev;
		dev = &sh->dev[i];
		clear_bit(R5_Insync, &dev->flags);
		clear_bit(R5_Syncio, &dev->flags);

		PRINTK("check %d: state 0x%lx read %p write %p written %p\n",
			i, dev->flags, dev->toread, dev->towrite, dev->written);
		/* maybe we can reply to a read */
		if (test_bit(R5_UPTODATE, &dev->flags) && dev->toread) {
			struct bio *rbi, *rbi2;
			PRINTK("Return read for disc %d\n", i);
			spin_lock_irq(&conf->device_lock);
			rbi = dev->toread;
			dev->toread = NULL;
			spin_unlock_irq(&conf->device_lock);
			while (rbi && rbi->bi_sector < dev->sector + STRIPE_SECTORS) {
				copy_data(0, rbi, dev->page, dev->sector);
				rbi2 = r5_next_bio(rbi, dev->sector);
				spin_lock_irq(&conf->device_lock);
				if (--rbi->bi_phys_segments == 0) {
					rbi->bi_next = return_bi;
					return_bi = rbi;
				}
				spin_unlock_irq(&conf->device_lock);
				rbi = rbi2;
			}
		}

		/* now count some things */
		if (test_bit(R5_LOCKED, &dev->flags)) locked++;
		if (test_bit(R5_UPTODATE, &dev->flags)) uptodate++;


		if (dev->toread) to_read++;
		if (dev->towrite) {
			to_write++;
			if (!test_bit(R5_OVERWRITE, &dev->flags))
				non_overwrite++;
		}
		if (dev->written) written++;
		rdev = conf->disks[i].rdev; /* FIXME, should I be looking rdev */
		if (!rdev || !rdev->in_sync) {
			if ( failed < 2 )
				failed_num[failed] = i;
			failed++;
		} else
			set_bit(R5_Insync, &dev->flags);
	}
	PRINTK("locked=%d uptodate=%d to_read=%d"
	       " to_write=%d failed=%d failed_num=%d,%d\n",
	       locked, uptodate, to_read, to_write, failed,
	       failed_num[0], failed_num[1]);
	/* check if the array has lost >2 devices and, if so, some requests might