Commit b203886e authored by Dan Williams's avatar Dan Williams Committed by Neil Brown

md: kill STRIPE_OP_MOD_DMA in raid5 offload

From: Dan Williams <dan.j.williams@intel.com>

This micro-optimization allowed the raid code to skip a re-read of the
parity block after checking parity.  It took advantage of the fact that
xor-offload-engines have their own internal result buffer and can check
parity without writing to memory.  Remove it for the following reasons:

1/ It is a layering violation for MD to need to manage the DMA and
   non-DMA paths within async_xor_zero_sum
2/ Bad precedent to toggle the 'ops' flags outside the lock
3/ Hard to realize a performance gain as reads will not need an updated
   parity block and writes will dirty it anyways.
Signed-off-by: default avatarDan Williams <dan.j.williams@intel.com>
Signed-off-by: default avatarNeil Brown <neilb@suse.de>
parent 0cd17fec
...@@ -837,15 +837,10 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx, ...@@ -837,15 +837,10 @@ ops_run_postxor(struct stripe_head *sh, struct dma_async_tx_descriptor *tx,
static void ops_complete_check(void *stripe_head_ref) static void ops_complete_check(void *stripe_head_ref)
{ {
struct stripe_head *sh = stripe_head_ref; struct stripe_head *sh = stripe_head_ref;
int pd_idx = sh->pd_idx;
pr_debug("%s: stripe %llu\n", __func__, pr_debug("%s: stripe %llu\n", __func__,
(unsigned long long)sh->sector); (unsigned long long)sh->sector);
if (test_and_clear_bit(STRIPE_OP_MOD_DMA_CHECK, &sh->ops.pending) &&
sh->ops.zero_sum_result == 0)
set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags);
set_bit(STRIPE_OP_CHECK, &sh->ops.complete); set_bit(STRIPE_OP_CHECK, &sh->ops.complete);
set_bit(STRIPE_HANDLE, &sh->state); set_bit(STRIPE_HANDLE, &sh->state);
release_stripe(sh); release_stripe(sh);
...@@ -873,11 +868,6 @@ static void ops_run_check(struct stripe_head *sh) ...@@ -873,11 +868,6 @@ static void ops_run_check(struct stripe_head *sh)
tx = async_xor_zero_sum(xor_dest, xor_srcs, 0, count, STRIPE_SIZE, tx = async_xor_zero_sum(xor_dest, xor_srcs, 0, count, STRIPE_SIZE,
&sh->ops.zero_sum_result, 0, NULL, NULL, NULL); &sh->ops.zero_sum_result, 0, NULL, NULL, NULL);
if (tx)
set_bit(STRIPE_OP_MOD_DMA_CHECK, &sh->ops.pending);
else
clear_bit(STRIPE_OP_MOD_DMA_CHECK, &sh->ops.pending);
atomic_inc(&sh->count); atomic_inc(&sh->count);
tx = async_trigger_callback(ASYNC_TX_DEP_ACK | ASYNC_TX_ACK, tx, tx = async_trigger_callback(ASYNC_TX_DEP_ACK | ASYNC_TX_ACK, tx,
ops_complete_check, sh); ops_complete_check, sh);
......
...@@ -267,10 +267,8 @@ struct r6_state { ...@@ -267,10 +267,8 @@ struct r6_state {
/* modifiers to the base operations /* modifiers to the base operations
* STRIPE_OP_MOD_REPAIR_PD - compute the parity block and write it back * STRIPE_OP_MOD_REPAIR_PD - compute the parity block and write it back
* STRIPE_OP_MOD_DMA_CHECK - parity is not corrupted by the check
*/ */
#define STRIPE_OP_MOD_REPAIR_PD 7 #define STRIPE_OP_MOD_REPAIR_PD 7
#define STRIPE_OP_MOD_DMA_CHECK 8
/* /*
* Plugging: * Plugging:
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment