On Mon, August 26, 2013, Ulf Hansson wrote:
> On 26 August 2013 08:47, Seungwon Jeon <tgih.jun@xxxxxxxxxxx> wrote:
> > On Friday, August 23 2013, Jaehoon Chung wrote:
> >> On 08/23/2013 06:23 PM, Ulf Hansson wrote:
> >> > On 21 August 2013 14:42, Seungwon Jeon <tgih.jun@xxxxxxxxxxx> wrote:
> >> >> From 029a839ddf6f13a1e1a8bf4d4bc32b67712593ec Mon Sep 17 00:00:00 2001
> >> >> From: Seungwon Jeon <tgih.jun@xxxxxxxxxxx>
> >> >> Date: Wed, 21 Aug 2013 17:30:02 +0900
> >> >> Subject: [PATCH 3/3] mmc: fix the remove of blk on suspend
> >> >>
> >> >> As mmc_cleanup_queue() is moved, NULL pointer access to card of
> >> >> mmc_queue is happened since commit fdfa20c1(mmc: reordered shutdown
> >> >> sequence mmc_bld_remove_req).
> >> >>
> >> >> Here, mmc_cleanup_queue is split into two parts.
> >> >> One is to quit the mmc_queue and two is to clean up the resource
> >> >> of mmc_queue.
> >> >>
> >> >> The following is log message related to the problem.
> >> >>
> >> >> Unable to handle kernel NULL pointer dereference at virtual address 000002a8
> >> >> pgd = ecd9c000
> >> >> [000002a8] *pgd=6d082831, *pte=00000000, *ppte=00000000
> >> >> Internal error: Oops: 17 [#1] SMP ARM
> >> >> Modules linked in: bnep rfcomm smsc95xx usbnet mii bluetooth nfsd lockd nfs_acl exportfs
> >> auth_rpcgss sunrpc oid_registry vfat fat btrfs raid6_pq xor zlib_deflate
> >> >> CPU: 3 PID: 2384 Comm: bash Not tainted 3.11.0-rc4-00869-ga7143f1-dirty #60
> >> >> task: c46d9b00 ti: ecefc000 task.ti: ecefc000
> >> >> PC is at mmc_blk_remove_req+0x58/0x88
> >> >> LR is at _raw_spin_unlock_irqrestore+0xc/0x14
> >> >> pc : [<c034e7d8>] lr : [<c0494ac8>] psr: 200f0053
> >> >> sp : ecefddf8 ip : 00000000 fp : 000dc1e8
> >> >> r10: c058ead8 r9 : ecce3f18 r8 : 00100100
> >> >> r7 : 00200200 r6 : c26b7118 r5 : 00000000 r4 : c26b1dc0
> >> >> r3 : 00000002 r2 : 00000000 r1 : 200f0053 r0 : 00000000
> >> >> Flags: nzCv IRQs on FIQs off Mode SVC_32 ISA ARM Segment user
> >> >> Control: 10c5387d Table: 6cd9c04a DAC: 00000015
> >> >> Process bash (pid: 2384, stack limit = 0xecefc240)
> >> >> Stack: (0xecefddf8 to 0xecefe000)
> >> >> <...>
> >> >> [<c034e7d8>] (mmc_blk_remove_req+0x58/0x88) from [<c03512d0>]
> >> >> (mmc_blk_remove_parts.isra.5+0x90/0xa8)
> >> >> [<c03512d0>] (mmc_blk_remove_parts.isra.5+0x90/0xa8) from
> >> >> [<c0351308>] (mmc_blk_remove+0x20/0x128)
> >> >> [<c0351308>] (mmc_blk_remove+0x20/0x128) from [<c034409c>] (mmc_bus_remove+0x18/0x20)
> >> >> [<c034409c>] (mmc_bus_remove+0x18/0x20) from [<c0265a20>] (__device_release_driver+0x7c/0xc8)
> >> >> [<c0265a20>] (__device_release_driver+0x7c/0xc8) from [<c0265a88>]
> (device_release_driver+0x1c/0x28)
> >> >> [<c0265a88>] (device_release_driver+0x1c/0x28) from [<c0265410>] (bus_remove_device+0x100/0x11c)
> >> >> [<c0265410>] (bus_remove_device+0x100/0x11c) from [<c0262c04>] (device_del+0x110/0x174)
> >> >> [<c0262c04>] (device_del+0x110/0x174) from [<c034463c>] (mmc_remove_card+0x64/0x78)
> >> >> [<c034463c>] (mmc_remove_card+0x64/0x78) from [<c0345124>] (mmc_remove+0x24/0x30)
> >> >> [<c0345124>] (mmc_remove+0x24/0x30) from [<c0343fb0>] (mmc_pm_notify+0x94/0xf8)
> >> >> [<c0343fb0>] (mmc_pm_notify+0x94/0xf8) from [<c00413b4>] (notifier_call_chain+0x44/0x84)
> >> >> [<c00413b4>] (notifier_call_chain+0x44/0x84) from [<c00417b4>]
> >> (__blocking_notifier_call_chain+0x48/0x60)
> >> >> [<c00417b4>] (__blocking_notifier_call_chain+0x48/0x60) from [<c00417e4>]
> >> (blocking_notifier_call_chain+0x18/0x20)
> >> >> [<c00417e4>] (blocking_notifier_call_chain+0x18/0x20) from [<c0059d48>]
> >> (pm_notifier_call_chain+0x14/0x2c)
> >> >> [<c0059d48>] (pm_notifier_call_chain+0x14/0x2c) from [<c005aa9c>] (pm_suspend+0xac/0x24c)
> >> >> [<c005aa9c>] (pm_suspend+0xac/0x24c) from [<c0059a68>] (state_store+0xb0/0xc4)
> >> >> [<c0059a68>] (state_store+0xb0/0xc4) from [<c01d610c>] (kobj_attr_store+0x14/0x20)
> >> >> [<c01d610c>] (kobj_attr_store+0x14/0x20) from [<c012b224>] (sysfs_write_file+0x118/0x164)
> >> >> [<c012b224>] (sysfs_write_file+0x118/0x164) from [<c00d59d4>] (vfs_write+0xd8/0x178)
> >> >> [<c00d59d4>] (vfs_write+0xd8/0x178) from [<c00d5d3c>] (SyS_write+0x40/0x68)
> >> >> [<c00d5d3c>] (SyS_write+0x40/0x68) from [<c000ea20>] (ret_fast_syscall+0x0/0x30)
> >> >> Code: ebfc509b e59432dc e3130002 0a000006 (e5d532a8)
> >> >>
> >> >> Reported-by: Alban Browaeys <prahal@xxxxxxxxx>
> >> >> Signed-off-by: Seungwon Jeon <tgih.jun@xxxxxxxxxxx>
> >> >
> >> > Acked-by: Ulf Hansson <ulf.hansson@xxxxxxxxxx>
> >>
> >> Acked-by: Jaehoon Chung <jh80.chung@xxxxxxxxxxx>
> >>
> >> Best Regards,
> >> Jaehoon Chung
> >
> > Ulf and Jaehoon,
> > Thank you for acks.
> >
> > But it could be replaced with [PATCH] mmc: fix null pointer use in mmc_blk_remove_req.
> > We would have found that. It has been merged.
>
> I noticed that now as well.
>
> So this patch can be completely dropped then?
Yes, I guess it's enough.
But if separating mmc_cleanup_queue() is meaningful, it can be another patch.
Do you have any idea?
Thanks,
Seungwon Jeon
>
> Kind regards
> Ulf Hansson
>
> >
> > Thanks,
> > Seungwon Jeon
> >
> >> >
> >> >> ---
> >> >> drivers/mmc/card/block.c | 9 ++++++---
> >> >> drivers/mmc/card/queue.c | 11 ++++++++---
> >> >> drivers/mmc/card/queue.h | 1 +
> >> >> 3 files changed, 15 insertions(+), 6 deletions(-)
> >> >>
> >> >> diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c
> >> >> index cd0b7f4..1e6726d 100644
> >> >> --- a/drivers/mmc/card/block.c
> >> >> +++ b/drivers/mmc/card/block.c
> >> >> @@ -2191,9 +2191,7 @@ static void mmc_blk_remove_req(struct mmc_blk_data *md)
> >> >> * is freeing the queue that stops new requests
> >> >> * from being accepted.
> >> >> */
> >> >> - mmc_cleanup_queue(&md->queue);
> >> >> - if (md->flags & MMC_BLK_PACKED_CMD)
> >> >> - mmc_packed_clean(&md->queue);
> >> >> + mmc_quit_queue(&md->queue);
> >> >> card = md->queue.card;
> >> >> if (md->disk->flags & GENHD_FL_UP) {
> >> >> device_remove_file(disk_to_dev(md->disk), &md->force_ro);
> >> >> @@ -2204,6 +2202,11 @@ static void mmc_blk_remove_req(struct mmc_blk_data *md)
> >> >>
> >> >> del_gendisk(md->disk);
> >> >> }
> >> >> +
> >> >> + mmc_cleanup_queue(&md->queue);
> >> >> + if (md->flags & MMC_BLK_PACKED_CMD)
> >> >> + mmc_packed_clean(&md->queue);
> >> >> +
> >> >> mmc_blk_put(md);
> >> >> }
> >> >> }
> >> >> diff --git a/drivers/mmc/card/queue.c b/drivers/mmc/card/queue.c
> >> >> index fa9632e..82e5550 100644
> >> >> --- a/drivers/mmc/card/queue.c
> >> >> +++ b/drivers/mmc/card/queue.c
> >> >> @@ -318,12 +318,10 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card,
> >> >> return ret;
> >> >> }
> >> >>
> >> >> -void mmc_cleanup_queue(struct mmc_queue *mq)
> >> >> +void mmc_quit_queue(struct mmc_queue *mq)
> >> >> {
> >> >> struct request_queue *q = mq->queue;
> >> >> unsigned long flags;
> >> >> - struct mmc_queue_req *mqrq_cur = mq->mqrq_cur;
> >> >> - struct mmc_queue_req *mqrq_prev = mq->mqrq_prev;
> >> >>
> >> >> /* Make sure the queue isn't suspended, as that will deadlock */
> >> >> mmc_queue_resume(mq);
> >> >> @@ -336,6 +334,13 @@ void mmc_cleanup_queue(struct mmc_queue *mq)
> >> >> q->queuedata = NULL;
> >> >> blk_start_queue(q);
> >> >> spin_unlock_irqrestore(q->queue_lock, flags);
> >> >> +}
> >> >> +EXPORT_SYMBOL(mmc_quit_queue);
> >> >> +
> >> >> +void mmc_cleanup_queue(struct mmc_queue *mq)
> >> >> +{
> >> >> + struct mmc_queue_req *mqrq_cur = mq->mqrq_cur;
> >> >> + struct mmc_queue_req *mqrq_prev = mq->mqrq_prev;
> >> >>
> >> >> kfree(mqrq_cur->bounce_sg);
> >> >> mqrq_cur->bounce_sg = NULL;
> >> >> diff --git a/drivers/mmc/card/queue.h b/drivers/mmc/card/queue.h
> >> >> index 5752d50..a3d89a2 100644
> >> >> --- a/drivers/mmc/card/queue.h
> >> >> +++ b/drivers/mmc/card/queue.h
> >> >> @@ -61,6 +61,7 @@ struct mmc_queue {
> >> >>
> >> >> extern int mmc_init_queue(struct mmc_queue *, struct mmc_card *, spinlock_t *,
> >> >> const char *);
> >> >> +extern void mmc_quit_queue(struct mmc_queue *);
> >> >> extern void mmc_cleanup_queue(struct mmc_queue *);
> >> >> extern void mmc_queue_suspend(struct mmc_queue *);
> >> >> extern void mmc_queue_resume(struct mmc_queue *);
> >> >> --
> >> >> 1.7.0.4
> >> >>
> >> >>
> >> >> --
> >> >> To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
> >> >> the body of a message to majordomo@xxxxxxxxxxxxxxx
> >> >> More majordomo info at http://vger.kernel.org/majordomo-info.html
> >> > --
> >> > To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
> >> > the body of a message to majordomo@xxxxxxxxxxxxxxx
> >> > More majordomo info at http://vger.kernel.org/majordomo-info.html
> >> >
> >>
> >> --
> >> To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
> >> the body of a message to majordomo@xxxxxxxxxxxxxxx
> >> More majordomo info at http://vger.kernel.org/majordomo-info.html
> >
> --
> To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
