From: Andre Noll <maan@systemlinux.org>
To: linux-raid@vger.kernel.org
Cc: Andre Noll <maan@systemlinux.org>
Subject: [PATCH 6/6] md: Convert remaining 1k representations in linear.c to sectors.
Date: Fri, 25 Jul 2008 17:27:52 +0200 [thread overview]
Message-ID: <1216999672-4943-7-git-send-email-maan@systemlinux.org> (raw)
In-Reply-To: <1216999672-4943-1-git-send-email-maan@systemlinux.org>
This patch renames hash_spacing and preshift to spacing and
sector_shift respectively with the following change of semantics:
Case 1: (sizeof(sector_t) <= sizeof(u32)).
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In this case, we have sector_shift = preshift = 0 and spacing =
2 * hash_spacing.
Hence, the index for the hash table which is computed by the new code
in which_dev() as sector / spacing equals the old value which was
(sector/2) / hash_spacing.
Note also that the value of nb_zone stays the same because both sz
and base double.
Case 2: (sizeof(sector_t) > sizeof(u32)).
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
(aka the shifting dance case). Here we have sector_shift = preshift +
1 and
spacing = 2 * hash_spacing
during the computation of nb_zone and curr_sector, but
spacing = hash_spacing
in which_dev() because in the last hunk of the patch for linear.c we
shift down conf->spacing (= 2 * hash_spacing) by one more bit than
in the old code.
Hence in the computation of nb_zone, sz and base have the same value
as before, so nb_zone is not affected. Also curr_sector in the next
hunk stays the same.
In which_dev() the hash table index is computed as
(sector >> sector_shift) / spacing
In view of sector_shift = preshift + 1 and spacing = hash_spacing,
this equals
((sector/2) >> preshift) / hash_spacing
which is the value computed by the old code.
Signed-off-by: Andre Noll <maan@systemlinux.org>
---
drivers/md/linear.c | 35 +++++++++++++++++------------------
include/linux/raid/linear.h | 6 ++++--
2 files changed, 21 insertions(+), 20 deletions(-)
diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index 441b7fa..71ffebd 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -33,14 +33,13 @@ static inline dev_info_t *which_dev(mddev_t *mddev, sector_t sector)
{
dev_info_t *hash;
linear_conf_t *conf = mddev_to_conf(mddev);
- sector_t block = sector >> 1;
/*
* sector_div(a,b) returns the remainer and sets a to a/b
*/
- block >>= conf->preshift;
- (void)sector_div(block, conf->hash_spacing);
- hash = conf->hash_table[block];
+ sector >>= conf->sector_shift;
+ (void)sector_div(sector, conf->spacing);
+ hash = conf->hash_table[sector];
while (sector >= hash->num_sectors + hash->start_sector)
hash++;
@@ -162,25 +161,25 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
* that is larger than min_sectors and use the size of that as
* the actual spacing
*/
- conf->hash_spacing = conf->array_sectors / 2;
+ conf->spacing = conf->array_sectors;
for (i=0; i < cnt-1 ; i++) {
sector_t tmp = 0;
int j;
for (j = i; j < cnt - 1 && tmp < min_sectors; j++)
tmp += conf->disks[j].num_sectors;
- if (tmp >= min_sectors && tmp < conf->hash_spacing * 2)
- conf->hash_spacing = tmp / 2;
+ if (tmp >= min_sectors && tmp < conf->spacing)
+ conf->spacing = tmp;
}
- /* hash_spacing may be too large for sector_div to work with,
+ /* spacing may be too large for sector_div to work with,
* so we might need to pre-shift
*/
- conf->preshift = 0;
+ conf->sector_shift = 0;
if (sizeof(sector_t) > sizeof(u32)) {
- sector_t space = conf->hash_spacing;
+ sector_t space = conf->spacing;
while (space > (sector_t)(~(u32)0)) {
space >>= 1;
- conf->preshift++;
+ conf->sector_shift++;
}
}
/*
@@ -192,9 +191,9 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
unsigned round;
unsigned long base;
- sz = conf->array_sectors >> (conf->preshift + 1);
+ sz = conf->array_sectors >> conf->sector_shift;
sz += 1; /* force round-up */
- base = conf->hash_spacing >> conf->preshift;
+ base = conf->spacing >> conf->sector_shift;
round = sector_div(sz, base);
nb_zone = sz + (round ? 1 : 0);
}
@@ -219,7 +218,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
i = 0;
for (curr_sector = 0;
curr_sector < conf->array_sectors;
- curr_sector += conf->hash_spacing * 2) {
+ curr_sector += conf->spacing) {
while (i < raid_disks-1 &&
curr_sector >= conf->disks[i+1].start_sector)
@@ -228,12 +227,12 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
*table ++ = conf->disks + i;
}
- if (conf->preshift) {
- conf->hash_spacing >>= conf->preshift;
- /* round hash_spacing up so that when we divide by it,
+ if (conf->sector_shift) {
+ conf->spacing >>= conf->sector_shift;
+ /* round spacing up so that when we divide by it,
* we err on the side of "too-low", which is safest.
*/
- conf->hash_spacing++;
+ conf->spacing++;
}
BUG_ON(table - conf->hash_table > nb_zone);
diff --git a/include/linux/raid/linear.h b/include/linux/raid/linear.h
index 87090e9..f38b9c5 100644
--- a/include/linux/raid/linear.h
+++ b/include/linux/raid/linear.h
@@ -15,9 +15,11 @@ struct linear_private_data
{
struct linear_private_data *prev; /* earlier version */
dev_info_t **hash_table;
- sector_t hash_spacing;
+ sector_t spacing;
sector_t array_sectors;
- int preshift; /* shift before dividing by hash_spacing */
+ int sector_shift; /* shift before dividing
+ * by spacing
+ */
dev_info_t disks[0];
};
--
1.5.4.3
next prev parent reply other threads:[~2008-07-25 15:27 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-07-25 15:27 [PATCH 0/6] md: Make linear raid use sector_t for internal representations Andre Noll
2008-07-25 15:27 ` [PATCH 1/6] md: linear.c: Fix typo in comment Andre Noll
2008-07-25 15:27 ` [PATCH 2/6] md: linear.c: Remove pointless initialization of curr_offset Andre Noll
2008-07-25 15:27 ` [PATCH 3/6] md: linear.c: Remove broken debug code Andre Noll
2008-07-25 15:27 ` [PATCH 4/6] md: linear: Represent dev_info->size and dev_info->offset in sectors Andre Noll
2008-07-25 15:27 ` [PATCH 5/6] md: linear.c: Make two local variables sector-based Andre Noll
2008-07-25 15:27 ` Andre Noll [this message]
-- strict thread matches above, loose matches on Subject: below --
2008-08-21 13:35 [PATCH 0/6] md: Make linear raid use sector_t for internal representations Andre Noll
2008-08-21 13:35 ` [PATCH 6/6] md: Convert remaining 1k representations in linear.c to sectors Andre Noll
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1216999672-4943-7-git-send-email-maan@systemlinux.org \
--to=maan@systemlinux.org \
--cc=linux-raid@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.