On Fri, Mar 22, 2019 at 9:36 AM Ming Lei <tom.leiming@xxxxxxxxx> wrote:
>
> On Fri, Mar 22, 2019 at 2:39 AM Bart Van Assche <bvanassche@xxxxxxx> wrote:
> >
> > On Sat, 2019-03-16 at 10:09 +0800, Jason Yan wrote:
> > > If we remove the scsi disk when running io with fio, oops occured with
> > > the following condition.
> > >
> > > [scsi_eh_0] [fio]
> > > scsi_end_request
> > > ->blk_update_request
> > > ->end_bio(io returned to userspace)
> > > close
> > > ->sd_release
> > > ->scsi_disk_put
> > > ->scsi_disk_release
> > > ->disk->private_data = NULL;
> > >
> > > ->scsi_mq_uninit_cmd
> > > ->scsi_uninit_cmd
> > > ->scsi_cmd_to_driver
> > > ->drv is NULL, Oops
> > >
> > > There is a small window between blk_update_request() and
> > > scsi_mq_uninit_cmd() that scsi disk may have been released. This will
> > > cause a oops like below:
> > >
> > > Unable to handle kernel NULL pointer dereference at virtual address
> > > 0000000000000000
> > > s/sync.c:67, func=xfer, error=In[11347.116050] Mem abort info:
> > > put/output error
> > > [11347.121598] ESR = 0x96000006
> > > [11347.126200] Exception class = DABT (current EL), IL = 32 bits
> > > [11347.132117] SET = 0, FnV = 0
> > > [11347.135170] EA = 0, S1PTW = 0
> > > [11347.138308] Data abort info:
> > > [11347.141186] ISV = 0, ISS = 0x00000006
> > > [11347.145019] CM = 0, WnR = 0
> > > [11347.147977] user pgtable: 4k pages, 48-bit VAs, pgdp =
> > > 00000000a67aece2
> > > [11347.154591] [0000000000000000] pgd=0000002f90774003,
> > > pud=0000002fab098003, pmd=0000000000000000
> > > [11347.163304] Internal error: Oops: 96000006 [#1] PREEMPT SMP
> > > [11347.168870] Modules linked in: hisi_sas_v3_hw hisi_sas_main libsas
> > > [11347.175044] CPU: 56 PID: 4294 Comm: scsi_eh_2 Not tainted
> > > 4.19.0-g8052059-dirty #2
> > > [11347.182600] Hardware name: Huawei D06/D06, BIOS Hisilicon D06 UEFI
> > > RC0 - B601 (V6.01) 11/08/2018
> > > [11347.191370] pstate: a0c00009 (NzCv daif 㰃繐ε흾㯗
> >
> > Please verify whether the following patch is a valid alternative for your patch:
> >
> > diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
> > index ed34bfbc3844..745ffdda1bc1 100644
> > --- a/drivers/scsi/sd.c
> > +++ b/drivers/scsi/sd.c
> > @@ -1408,6 +1408,7 @@ static void sd_release(struct gendisk *disk, fmode_t mode)
> > {
> > struct scsi_disk *sdkp = scsi_disk(disk);
> > struct scsi_device *sdev = sdkp->device;
> > + struct request_queue *q = sdkp->disk->queue;
> >
> > SCSI_LOG_HLQUEUE(3, sd_printk(KERN_INFO, sdkp, "sd_release\n"));
> >
> > @@ -1417,9 +1418,12 @@ static void sd_release(struct gendisk *disk, fmode_t mode)
> > }
> >
> > /*
> > - * XXX and what if there are packets in flight and this close()
> > - * XXX is followed by a "rmmod sd_mod"?
> > + * Wait until any requests that are in progress have completed.
> > + * This is necessary to avoid that e.g. scsi_end_request() crashes
> > + * due to scsi_disk_relase() clearing the disk->private_data pointer.
> > */
> > + blk_mq_freeze_queue(q);
> > + blk_mq_unfreeze_queue(q);
>
> It is over-kill to drain any requests here, what we want is to just
> drain any in-flight
> IO requests.
Not only over-kill, actually it can cause big performance issue, since any
block/scsi utility may trigger the freeze/unfreeze.
Thanks,
Ming Lei
