* [PATCH] raid5: fix possible oops in add_stripe_bio when enable pr_debug
@ 2012-09-20 6:34 Jianpeng Ma
2012-09-20 6:47 ` NeilBrown
0 siblings, 1 reply; 5+ messages in thread
From: Jianpeng Ma @ 2012-09-20 6:34 UTC (permalink / raw)
To: Neil Brown; +Cc: linux-raid
In func add_stripe_bio:
>> .....
>> bip = &sh->dev[dd_idx].toread;
>> ......
>>spin_unlock_irq(&sh->stripe_lock);
>> pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
>> (unsigned long long)(*bip)->bi_sector,
>> (unsigned long long)sh->sector, dd_idx);
After spin_unlock_irq, this thread scheded and toread may become null.
So it will be oops.
Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
---
drivers/md/raid5.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index adda94d..f172b1e 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -2356,6 +2356,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
struct bio **bip;
struct r5conf *conf = sh->raid_conf;
int firstwrite=0;
+ sector_t sector = bi->bi_sector;
pr_debug("adding bi b#%llu to stripe s#%llu\n",
(unsigned long long)bi->bi_sector,
@@ -2406,7 +2407,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
spin_unlock_irq(&sh->stripe_lock);
pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
- (unsigned long long)(*bip)->bi_sector,
+ (unsigned long long)sector,
(unsigned long long)sh->sector, dd_idx);
if (conf->mddev->bitmap && firstwrite) {
--
1.7.9.5
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] raid5: fix possible oops in add_stripe_bio when enable pr_debug
2012-09-20 6:34 [PATCH] raid5: fix possible oops in add_stripe_bio when enable pr_debug Jianpeng Ma
@ 2012-09-20 6:47 ` NeilBrown
2012-09-20 7:17 ` Jianpeng Ma
0 siblings, 1 reply; 5+ messages in thread
From: NeilBrown @ 2012-09-20 6:47 UTC (permalink / raw)
To: Jianpeng Ma; +Cc: linux-raid
[-- Attachment #1: Type: text/plain, Size: 1571 bytes --]
On Thu, 20 Sep 2012 14:34:00 +0800 "Jianpeng Ma" <majianpeng@gmail.com> wrote:
> In func add_stripe_bio:
> >> .....
> >> bip = &sh->dev[dd_idx].toread;
> >> ......
> >>spin_unlock_irq(&sh->stripe_lock);
>
> >> pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
> >> (unsigned long long)(*bip)->bi_sector,
> >> (unsigned long long)sh->sector, dd_idx);
> After spin_unlock_irq, this thread scheded and toread may become null.
> So it will be oops.
>
> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
> ---
> drivers/md/raid5.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
> index adda94d..f172b1e 100644
> --- a/drivers/md/raid5.c
> +++ b/drivers/md/raid5.c
> @@ -2356,6 +2356,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
> struct bio **bip;
> struct r5conf *conf = sh->raid_conf;
> int firstwrite=0;
> + sector_t sector = bi->bi_sector;
>
> pr_debug("adding bi b#%llu to stripe s#%llu\n",
> (unsigned long long)bi->bi_sector,
> @@ -2406,7 +2407,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
> spin_unlock_irq(&sh->stripe_lock);
>
> pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
> - (unsigned long long)(*bip)->bi_sector,
> + (unsigned long long)sector,
> (unsigned long long)sh->sector, dd_idx);
>
> if (conf->mddev->bitmap && firstwrite) {
how about we just move the spin_unlock_irq after the pr_debug??
NeilBrown
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Re: [PATCH] raid5: fix possible oops in add_stripe_bio when enable pr_debug
2012-09-20 6:47 ` NeilBrown
@ 2012-09-20 7:17 ` Jianpeng Ma
2012-09-25 6:45 ` NeilBrown
0 siblings, 1 reply; 5+ messages in thread
From: Jianpeng Ma @ 2012-09-20 7:17 UTC (permalink / raw)
To: Neil Brown; +Cc: linux-raid
On 2012-09-20 14:47 NeilBrown <neilb@suse.de> Wrote:
>On Thu, 20 Sep 2012 14:34:00 +0800 "Jianpeng Ma" <majianpeng@gmail.com> wrote:
>
>> In func add_stripe_bio:
>> >> .....
>> >> bip = &sh->dev[dd_idx].toread;
>> >> ......
>> >>spin_unlock_irq(&sh->stripe_lock);
>>
>> >> pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
>> >> (unsigned long long)(*bip)->bi_sector,
>> >> (unsigned long long)sh->sector, dd_idx);
>> After spin_unlock_irq, this thread scheded and toread may become null.
>> So it will be oops.
>>
>> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
>> ---
>> drivers/md/raid5.c | 3 ++-
>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
>> index adda94d..f172b1e 100644
>> --- a/drivers/md/raid5.c
>> +++ b/drivers/md/raid5.c
>> @@ -2356,6 +2356,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
>> struct bio **bip;
>> struct r5conf *conf = sh->raid_conf;
>> int firstwrite=0;
>> + sector_t sector = bi->bi_sector;
>>
>> pr_debug("adding bi b#%llu to stripe s#%llu\n",
>> (unsigned long long)bi->bi_sector,
>> @@ -2406,7 +2407,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
>> spin_unlock_irq(&sh->stripe_lock);
>>
>> pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
>> - (unsigned long long)(*bip)->bi_sector,
>> + (unsigned long long)sector,
>> (unsigned long long)sh->sector, dd_idx);
>>
>> if (conf->mddev->bitmap && firstwrite) {
>
>
>how about we just move the spin_unlock_irq after the pr_debug??
>
ah! Why are you think ? my method only add a parameter.
BTW, in func handle_failed_stripe:
>>if (!test_bit(R5_Wantfill, &sh->dev[i].flags) &&
>> (!test_bit(R5_Insync, &sh->dev[i].flags) ||
>> test_bit(R5_ReadError, &sh->dev[i].flags))) {
>> bi = sh->dev[i].toread;
>> sh->dev[i].toread = NULL;
>> if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
>> wake_up(&conf->wait_for_overlap);
Why use stripe_lock to protect toread?
Thanks!
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] raid5: fix possible oops in add_stripe_bio when enable pr_debug
2012-09-20 7:17 ` Jianpeng Ma
@ 2012-09-25 6:45 ` NeilBrown
2012-09-25 6:50 ` Jianpeng Ma
0 siblings, 1 reply; 5+ messages in thread
From: NeilBrown @ 2012-09-25 6:45 UTC (permalink / raw)
To: Jianpeng Ma; +Cc: linux-raid
[-- Attachment #1: Type: text/plain, Size: 2487 bytes --]
On Thu, 20 Sep 2012 15:17:54 +0800 "Jianpeng Ma" <majianpeng@gmail.com> wrote:
> On 2012-09-20 14:47 NeilBrown <neilb@suse.de> Wrote:
> >On Thu, 20 Sep 2012 14:34:00 +0800 "Jianpeng Ma" <majianpeng@gmail.com> wrote:
> >
> >> In func add_stripe_bio:
> >> >> .....
> >> >> bip = &sh->dev[dd_idx].toread;
> >> >> ......
> >> >>spin_unlock_irq(&sh->stripe_lock);
> >>
> >> >> pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
> >> >> (unsigned long long)(*bip)->bi_sector,
> >> >> (unsigned long long)sh->sector, dd_idx);
> >> After spin_unlock_irq, this thread scheded and toread may become null.
> >> So it will be oops.
> >>
> >> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
> >> ---
> >> drivers/md/raid5.c | 3 ++-
> >> 1 file changed, 2 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
> >> index adda94d..f172b1e 100644
> >> --- a/drivers/md/raid5.c
> >> +++ b/drivers/md/raid5.c
> >> @@ -2356,6 +2356,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
> >> struct bio **bip;
> >> struct r5conf *conf = sh->raid_conf;
> >> int firstwrite=0;
> >> + sector_t sector = bi->bi_sector;
> >>
> >> pr_debug("adding bi b#%llu to stripe s#%llu\n",
> >> (unsigned long long)bi->bi_sector,
> >> @@ -2406,7 +2407,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
> >> spin_unlock_irq(&sh->stripe_lock);
> >>
> >> pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
> >> - (unsigned long long)(*bip)->bi_sector,
> >> + (unsigned long long)sector,
> >> (unsigned long long)sh->sector, dd_idx);
> >>
> >> if (conf->mddev->bitmap && firstwrite) {
> >
> >
> >how about we just move the spin_unlock_irq after the pr_debug??
> >
> ah! Why are you think ? my method only add a parameter.
Yes.
> BTW, in func handle_failed_stripe:
> >>if (!test_bit(R5_Wantfill, &sh->dev[i].flags) &&
> >> (!test_bit(R5_Insync, &sh->dev[i].flags) ||
> >> test_bit(R5_ReadError, &sh->dev[i].flags))) {
> >> bi = sh->dev[i].toread;
> >> sh->dev[i].toread = NULL;
> >> if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
> >> wake_up(&conf->wait_for_overlap);
> Why use stripe_lock to protect toread?
I assume you mean that we should be holding the lock to protect toread, but
we aren't.
I've queued a patch to fix that.
Thanks.
NeilBrown
>
> Thanks!
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: Re: [PATCH] raid5: fix possible oops in add_stripe_bio when enable pr_debug
2012-09-25 6:45 ` NeilBrown
@ 2012-09-25 6:50 ` Jianpeng Ma
0 siblings, 0 replies; 5+ messages in thread
From: Jianpeng Ma @ 2012-09-25 6:50 UTC (permalink / raw)
To: Neil Brown; +Cc: linux-raid
On 2012-09-25 14:45 NeilBrown <neilb@suse.de> Wrote:
>On Thu, 20 Sep 2012 15:17:54 +0800 "Jianpeng Ma" <majianpeng@gmail.com> wrote:
>
>> On 2012-09-20 14:47 NeilBrown <neilb@suse.de> Wrote:
>> >On Thu, 20 Sep 2012 14:34:00 +0800 "Jianpeng Ma" <majianpeng@gmail.com> wrote:
>> >
>> >> In func add_stripe_bio:
>> >> >> .....
>> >> >> bip = &sh->dev[dd_idx].toread;
>> >> >> ......
>> >> >>spin_unlock_irq(&sh->stripe_lock);
>> >>
>> >> >> pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
>> >> >> (unsigned long long)(*bip)->bi_sector,
>> >> >> (unsigned long long)sh->sector, dd_idx);
>> >> After spin_unlock_irq, this thread scheded and toread may become null.
>> >> So it will be oops.
>> >>
>> >> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
>> >> ---
>> >> drivers/md/raid5.c | 3 ++-
>> >> 1 file changed, 2 insertions(+), 1 deletion(-)
>> >>
>> >> diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
>> >> index adda94d..f172b1e 100644
>> >> --- a/drivers/md/raid5.c
>> >> +++ b/drivers/md/raid5.c
>> >> @@ -2356,6 +2356,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
>> >> struct bio **bip;
>> >> struct r5conf *conf = sh->raid_conf;
>> >> int firstwrite=0;
>> >> + sector_t sector = bi->bi_sector;
>> >>
>> >> pr_debug("adding bi b#%llu to stripe s#%llu\n",
>> >> (unsigned long long)bi->bi_sector,
>> >> @@ -2406,7 +2407,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in
>> >> spin_unlock_irq(&sh->stripe_lock);
>> >>
>> >> pr_debug("added bi b#%llu to stripe s#%llu, disk %d.\n",
>> >> - (unsigned long long)(*bip)->bi_sector,
>> >> + (unsigned long long)sector,
>> >> (unsigned long long)sh->sector, dd_idx);
>> >>
>> >> if (conf->mddev->bitmap && firstwrite) {
>> >
>> >
>> >how about we just move the spin_unlock_irq after the pr_debug??
>> >
>> ah! Why are you think ? my method only add a parameter.
>
>Yes.
>
>> BTW, in func handle_failed_stripe:
>> >>if (!test_bit(R5_Wantfill, &sh->dev[i].flags) &&
>> >> (!test_bit(R5_Insync, &sh->dev[i].flags) ||
>> >> test_bit(R5_ReadError, &sh->dev[i].flags))) {
>> >> bi = sh->dev[i].toread;
>> >> sh->dev[i].toread = NULL;
>> >> if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags))
>> >> wake_up(&conf->wait_for_overlap);
>> Why use stripe_lock to protect toread?
>
>I assume you mean that we should be holding the lock to protect toread, but
>we aren't.
>I've queued a patch to fix that.
>
Hi,
Last Saturday, i sent a patch-set which contained a patch which fix this bug.
You can check your mail!
Thanks!
>Thanks.
>NeilBrown
>
>
>>
>> Thanks!
>
>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2012-09-25 6:50 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-09-20 6:34 [PATCH] raid5: fix possible oops in add_stripe_bio when enable pr_debug Jianpeng Ma
2012-09-20 6:47 ` NeilBrown
2012-09-20 7:17 ` Jianpeng Ma
2012-09-25 6:45 ` NeilBrown
2012-09-25 6:50 ` Jianpeng Ma
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).