提交 · c5a3100062cf277d3edd4e6f4a1f1e403524b464 · openeuler / Kernel

27 7月, 2011 3 次提交

md/raid5: move more code into common handle_stripe · c5a31000

由 NeilBrown 提交于 7月 27, 2011

The difference between the RAID5 and RAID6 code here is easily
resolved using conf->max_degraded.
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

c5a31000

md/raid5: Move code for finishing a reconstruction into handle_stripe. · 3687c061

由 NeilBrown 提交于 7月 27, 2011

Prior to commit ab69ae12 the code in handle_stripe5 and
handle_stripe6 to "Finish reconstruct operations initiated by the
expansion process" was identical.
That commit added an identical stanza of code to each function, but in
different places.  That was careless.

The raid5 code was correct, so move that out into handle_stripe and
remove raid6 version.
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

3687c061

md/raid5: Remove stripe_head_state arg from handle_stripe_expansion. · 86c374ba

由 NeilBrown 提交于 7月 27, 2011

This arg is only used to differentiate between RAID5 and RAID6 but
that is not needed.  For RAID5, raid5_compute_sector will set qd_idx
to "~0" so j with certainly not equals qd_idx, so there is no need
for a guard on that condition.

So remove the guard and remove the arg from the declaration and
callers of handle_stripe_expansion.
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

86c374ba

26 7月, 2011 7 次提交

md/raid5: move stripe_head_state and more code into handle_stripe. · cc94015a

由 NeilBrown 提交于 7月 26, 2011

By defining the 'stripe_head_state' in 'handle_stripe', we can move
some common code out of handle_stripe[56]() and into handle_stripe.

The means that all accesses for stripe_head_state in handle_stripe[56]
need to be 's->' instead of 's.', but the compiler should inline
those functions and just use a direct stack reference, and future
patches while hoist most of this code up into handle_stripe()
so we will revert to "s.".
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

cc94015a

md/raid5: add some more fields to stripe_head_state · c5709ef6

由 NeilBrown 提交于 7月 26, 2011

Adding these three fields will allow more common code to be moved
to handle_stripe()

struct field rearrangement by Namhyung Kim.
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

c5709ef6

md/raid5: unify stripe_head_state and r6_state · f2b3b44d

由 NeilBrown 提交于 7月 26, 2011

'struct stripe_head_state' stores state about the 'current' stripe
that is passed around while handling the stripe.
For RAID6 there is an extension structure: r6_state, which is also
passed around.
There is no value in keeping these separate, so move the fields from
the latter into the former.

This means that all code now needs to treat s->failed_num as an small
array, but this is a small cost.
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

f2b3b44d

md/raid5: move common code into handle_stripe · 82e5a171

由 NeilBrown 提交于 7月 26, 2011

There is common code at the start of handle_stripe5 and
handle_stripe6.  Move it into handle_stripe.
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

82e5a171

md/raid5: replace sh->lock with an 'active' flag. · c4c1663b

由 NeilBrown 提交于 7月 26, 2011

sh->lock is now mainly used to ensure that two threads aren't running
in the locked part of handle_stripe[56] at the same time.

That can more neatly be achieved with an 'active' flag which we set
while running handle_stripe.  If we find the flag is set, we simply
requeue the stripe for later by setting STRIPE_HANDLE.

For safety we take ->device_lock while examining the state of the
stripe and creating a summary in 'stripe_head_state / r6_state'.
This possibly isn't needed but as shared fields like ->toread,
->towrite are checked it is safer for now at least.

We leave the label after the old 'unlock' called "unlock" because it
will disappear in a few patches, so renaming seems pointless.

This leaves the stripe 'locked' for longer as we clear STRIPE_ACTIVE
later, but that is not a problem.
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

c4c1663b

md/raid5: Protect some more code with ->device_lock. · cbe47ec5

由 NeilBrown 提交于 7月 26, 2011

Other places that change or follow dev->towrite and dev->written take
the device_lock as well as the sh->lock.
So it should really be held in these places too.
Also, doing so will allow sh->lock to be discarded.

with merged fixes by: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

cbe47ec5

md/raid5: Remove use of sh->lock in sync_request · 83206d66

由 NeilBrown 提交于 7月 26, 2011

This is the start of a series of patches to remove sh->lock.

sync_request takes sh->lock before setting STRIPE_SYNCING to ensure
there is no race with testing it in handle_stripe[56].

Instead, use a new flag STRIPE_SYNC_REQUESTED and test it early
in handle_stripe[56] (after getting the same lock) and perform the
same set/clear operations if it was set.
Signed-off-by: NNeilBrown <neilb@suse.de>
Reviewed-by: NNamhyung Kim <namhyung@gmail.com>

83206d66

18 7月, 2011 5 次提交

md/raid5: get rid of duplicated call to bio_data_dir() · ffd96e35