The patch titled
Subject: md: convert to kvmalloc
has been added to the -mm tree. Its filename is
md-convert-to-kvmalloc.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/md-convert-to-kvmalloc.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/md-convert-to-kvmalloc.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Kent Overstreet <kent.overstreet@xxxxxxxxx>
Subject: md: convert to kvmalloc
The code really just wants a big flat buffer, so just do that.
Link: http://lkml.kernel.org/r/20181217131929.11727-3-kent.overstreet@xxxxxxxxx
Signed-off-by: Kent Overstreet <kent.overstreet@xxxxxxxxx>
Reviewed-by: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Shaohua Li <shli@xxxxxxxxxx>
Cc: Alexey Dobriyan <adobriyan@xxxxxxxxx>
Cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxx>
Cc: Eric Paris <eparis@xxxxxxxxxxxxxx>
Cc: Marcelo Ricardo Leitner <marcelo.leitner@xxxxxxxxx>
Cc: Neil Horman <nhorman@xxxxxxxxxxxxx>
Cc: Paul Moore <paul@xxxxxxxxxxxxxx>
Cc: Pravin B Shelar <pshelar@xxxxxxx>
Cc: Stephen Smalley <sds@xxxxxxxxxxxxx>
Cc: Vlad Yasevich <vyasevich@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
drivers/md/raid5-ppl.c | 6 --
drivers/md/raid5.c | 85 +++++++++++++++++----------------------
drivers/md/raid5.h | 9 ++--
3 files changed, 45 insertions(+), 55 deletions(-)
--- a/drivers/md/raid5.c~md-convert-to-kvmalloc
+++ a/drivers/md/raid5.c
@@ -54,7 +54,6 @@
#include <linux/slab.h>
#include <linux/ratelimit.h>
#include <linux/nodemask.h>
-#include <linux/flex_array.h>
#include <trace/events/block.h>
#include <linux/list_sort.h>
@@ -1394,22 +1393,16 @@ static void ops_complete_compute(void *s
}
/* return a pointer to the address conversion region of the scribble buffer */
-static addr_conv_t *to_addr_conv(struct stripe_head *sh,
- struct raid5_percpu *percpu, int i)
+static struct page **to_addr_page(struct raid5_percpu *percpu, int i)
{
- void *addr;
-
- addr = flex_array_get(percpu->scribble, i);
- return addr + sizeof(struct page *) * (sh->disks + 2);
+ return percpu->scribble + i * percpu->scribble_obj_size;
}
/* return a pointer to the address conversion region of the scribble buffer */
-static struct page **to_addr_page(struct raid5_percpu *percpu, int i)
+static addr_conv_t *to_addr_conv(struct stripe_head *sh,
+ struct raid5_percpu *percpu, int i)
{
- void *addr;
-
- addr = flex_array_get(percpu->scribble, i);
- return addr;
+ return (void *) (to_addr_page(percpu, i) + sh->disks + 2);
}
static struct dma_async_tx_descriptor *
@@ -2238,21 +2231,23 @@ static int grow_stripes(struct r5conf *c
* calculate over all devices (not just the data blocks), using zeros in place
* of the P and Q blocks.
*/
-static struct flex_array *scribble_alloc(int num, int cnt, gfp_t flags)
+static int scribble_alloc(struct raid5_percpu *percpu,
+ int num, int cnt, gfp_t flags)
{
- struct flex_array *ret;
- size_t len;
+ size_t obj_size =
+ sizeof(struct page *) * (num+2) +
+ sizeof(addr_conv_t) * (num+2);
+ void *scribble;
- len = sizeof(struct page *) * (num+2) + sizeof(addr_conv_t) * (num+2);
- ret = flex_array_alloc(len, cnt, flags);
- if (!ret)
- return NULL;
- /* always prealloc all elements, so no locking is required */
- if (flex_array_prealloc(ret, 0, cnt, flags)) {
- flex_array_free(ret);
- return NULL;
- }
- return ret;
+ scribble = kvmalloc_array(cnt, obj_size, flags);
+ if (!scribble)
+ return -ENOMEM;
+
+ kvfree(percpu->scribble);
+
+ percpu->scribble = scribble;
+ percpu->scribble_obj_size = obj_size;
+ return 0;
}
static int resize_chunks(struct r5conf *conf, int new_disks, int new_sectors)
@@ -2270,23 +2265,18 @@ static int resize_chunks(struct r5conf *
return 0;
mddev_suspend(conf->mddev);
get_online_cpus();
+
for_each_present_cpu(cpu) {
struct raid5_percpu *percpu;
- struct flex_array *scribble;
percpu = per_cpu_ptr(conf->percpu, cpu);
- scribble = scribble_alloc(new_disks,
- new_sectors / STRIPE_SECTORS,
- GFP_NOIO);
-
- if (scribble) {
- flex_array_free(percpu->scribble);
- percpu->scribble = scribble;
- } else {
- err = -ENOMEM;
+ err = scribble_alloc(percpu, new_disks,
+ new_sectors / STRIPE_SECTORS,
+ GFP_NOIO);
+ if (err)
break;
- }
}
+
put_online_cpus();
mddev_resume(conf->mddev);
if (!err) {
@@ -6738,25 +6728,26 @@ raid5_size(struct mddev *mddev, sector_t
static void free_scratch_buffer(struct r5conf *conf, struct raid5_percpu *percpu)
{
safe_put_page(percpu->spare_page);
- if (percpu->scribble)
- flex_array_free(percpu->scribble);
percpu->spare_page = NULL;
+ kvfree(percpu->scribble);
percpu->scribble = NULL;
}
static int alloc_scratch_buffer(struct r5conf *conf, struct raid5_percpu *percpu)
{
- if (conf->level == 6 && !percpu->spare_page)
+ if (conf->level == 6 && !percpu->spare_page) {
percpu->spare_page = alloc_page(GFP_KERNEL);
- if (!percpu->scribble)
- percpu->scribble = scribble_alloc(max(conf->raid_disks,
- conf->previous_raid_disks),
- max(conf->chunk_sectors,
- conf->prev_chunk_sectors)
- / STRIPE_SECTORS,
- GFP_KERNEL);
+ if (!percpu->spare_page)
+ return -ENOMEM;
+ }
- if (!percpu->scribble || (conf->level == 6 && !percpu->spare_page)) {
+ if (scribble_alloc(percpu,
+ max(conf->raid_disks,
+ conf->previous_raid_disks),
+ max(conf->chunk_sectors,
+ conf->prev_chunk_sectors)
+ / STRIPE_SECTORS,
+ GFP_KERNEL)) {
free_scratch_buffer(conf, percpu);
return -ENOMEM;
}
--- a/drivers/md/raid5.h~md-convert-to-kvmalloc
+++ a/drivers/md/raid5.h
@@ -638,10 +638,11 @@ struct r5conf {
/* per cpu variables */
struct raid5_percpu {
struct page *spare_page; /* Used when checking P/Q in raid6 */
- struct flex_array *scribble; /* space for constructing buffer
- * lists and performing address
- * conversions
- */
+ void *scribble; /* space for constructing buffer
+ * lists and performing address
+ * conversions
+ */
+ int scribble_obj_size;
} __percpu *percpu;
int scribble_disks;
int scribble_sectors;
--- a/drivers/md/raid5-ppl.c~md-convert-to-kvmalloc
+++ a/drivers/md/raid5-ppl.c
@@ -16,7 +16,6 @@
#include <linux/blkdev.h>
#include <linux/slab.h>
#include <linux/crc32c.h>
-#include <linux/flex_array.h>
#include <linux/async_tx.h>
#include <linux/raid/md_p.h>
#include "md.h"
@@ -165,7 +164,7 @@ ops_run_partial_parity(struct stripe_hea
struct dma_async_tx_descriptor *tx)
{
int disks = sh->disks;
- struct page **srcs = flex_array_get(percpu->scribble, 0);
+ struct page **srcs = percpu->scribble;
int count = 0, pd_idx = sh->pd_idx, i;
struct async_submit_ctl submit;
@@ -196,8 +195,7 @@ ops_run_partial_parity(struct stripe_hea
}
init_async_submit(&submit, ASYNC_TX_FENCE|ASYNC_TX_XOR_ZERO_DST, tx,
- NULL, sh, flex_array_get(percpu->scribble, 0)
- + sizeof(struct page *) * (sh->disks + 2));
+ NULL, sh, (void *) (srcs + sh->disks + 2));
if (count == 1)
tx = async_memcpy(sh->ppl_page, srcs[0], 0, 0, PAGE_SIZE,
_
Patches currently in -mm which might be from kent.overstreet@xxxxxxxxx are
openvswitch-convert-to-kvmalloc.patch
md-convert-to-kvmalloc.patch
selinux-convert-to-kvmalloc.patch
generic-radix-trees.patch
proc-commit-to-genradix.patch
sctp-convert-to-genradix.patch
drop-flex_arrays.patch
