Re: linux-next: Tree for Aug 1

[Ming Lei]

On Thu, Aug 02, 2018 at 06:05:16AM -0700, Guenter Roeck wrote:
> On 08/02/2018 04:35 AM, Ming Lei wrote:
> > On Thu, Aug 2, 2018 at 12:58 PM, Guenter Roeck <linux@xxxxxxxxxxxx> wrote:
> > > On 08/01/2018 05:03 PM, James Bottomley wrote:
> > > > 
> > > > On Thu, 2018-08-02 at 07:57 +0800, Ming Lei wrote:
> > > > > 
> > > > > On Thu, Aug 2, 2018 at 7:47 AM, Guenter Roeck <linux@xxxxxxxxxxxx>
> > > > > wrote:
> > > > > > 
> > > > > > On Wed, Aug 01, 2018 at 03:52:45PM -0700, James Bottomley wrote:
> > > > > > > 
> > > > > > > On Wed, 2018-08-01 at 15:48 -0700, Guenter Roeck wrote:
> > > > > > > > 
> > > > > > > > On Wed, Aug 01, 2018 at 05:58:52PM +1000, Stephen Rothwell
> > > > > > > > wrote:
> > > > > > > > > 
> > > > > > > > > Hi all,
> > > > > > > > > 
> > > > > > > > > Changes since 20180731:
> > > > > > > > > 
> > > > > > > > > The pci tree gained a conflict against the pci-current tree.
> > > > > > > > > 
> > > > > > > > > The net-next tree gained a conflict against the bpf tree.
> > > > > > > > > 
> > > > > > > > > The block tree lost its build failure.
> > > > > > > > > 
> > > > > > > > > The staging tree still had its build failure due to an
> > > > > > > > > interaction
> > > > > > > > > with
> > > > > > > > > the vfs tree for which I disabled CONFIG_EROFS_FS.
> > > > > > > > > 
> > > > > > > > > The kspp tree lost its build failure.
> > > > > > > > > 
> > > > > > > > > Non-merge commits (relative to Linus' tree): 10070
> > > > > > > > >    9137 files changed, 417605 insertions(+), 179996 deletions(-
> > > > > > > > > )
> > > > > > > > > 
> > > > > > > > > -----------------------------------------------------------
> > > > > > > > > ------
> > > > > > > > > -----------
> > > > > > > > > 
> > > > > > > > 
> > > > > > > > The widespread kernel hang issues are still seen. I managed
> > > > > > > > to bisect it after working around the transient build failures.
> > > > > > > > Bisect log is attached below. Unfortunately, it doesn't help
> > > > > > > > much.
> > > > > > > > The culprit is reported as:
> > > > > > > > 
> > > > > > > > 2d542828c5e9 Merge remote-tracking branch 'scsi/for-next'
> > > > > > > > 
> > > > > > > > The preceding merge,
> > > > > > > > 
> > > > > > > > 453f1d821165 Merge remote-tracking branch 'cgroup/for-next'
> > > > > > > > 
> > > > > > > > checks out fine, as does the tip of scsi-next (commit
> > > > > > > > 103c7b7e0184,
> > > > > > > > "Merge branch 'misc' into for-next"). No idea how to proceed.
> > > > > > > 
> > > > > > > 
> > > > > > > This sounds like you may have a problem with this patch:
> > > > > > > 
> > > > > > >       commit d5038a13eca72fb216c07eb717169092e92284f1
> > > > > > >        Author: Johannes Thumshirn <jthumshirn@xxxxxxx>
> > > > > > >        Date:   Wed Jul 4 10:53:56 2018 +0200
> > > > > > > 
> > > > > > >            scsi: core: switch to scsi-mq by default
> > > > > > > 
> > > > > > > To verify, boot with the additional kernel parameter
> > > > > > > 
> > > > > > > scsi_mod.use_blk_mq=0
> > > > > > > 
> > > > > > > Which will reverse the effect of the above patch.
> > > > > > > 
> > > > > > 
> > > > > > Yes, that fixes the problem.
> > > > > 
> > > > > 
> > > > > That may not the root cause, given this issue is only started to
> > > > > see from next-20180731, but d5038a13eca7 (scsi: core: switch to
> > > > > scsi-mq by default)
> > > > > has been in -next for quite a while.
> > > > > 
> > > > > Seems something new causes this issue.
> > > > 
> > > > 
> > > > Read my other email about how to find this.
> > > > 
> > > > https://marc.info/?l=linux-scsi&m=153316446223676
> > > > 
> > > > Now that we've confirmed the issue, Gunter, could you attempt to bisect
> > > > it as that email describes?
> > > > 
> > > 
> > > So, I am more and more baffled.
> > > 
> > > I ran another round of bisect, this time each test executing twice,
> > > once with "scsi_mod.use_blk_mq=1" and once with "scsi_mod.use_blk_mq=0",
> > > requiring both to pass. Bisect still points to the merge as culprit.
> > > 
> > > Ok, one step further: Actually _revert_ commit d5038a13eca72 before running
> > > each test, meaning the default is use_blk_mq=0. Still run both tests.
> > > Bisect _still_ points to the merge of scsi-next as culprit.
> > > 
> > > So, to me it looks like the problem is triggered by _something_ in
> > > scsi-next, combined with _something_ in -next prior to the merge,
> > > not specifically associated with use_blk_mq=[0|1] or d5038a13eca72,
> > > but to a combination of some patch in scsi-next and some other patch.
> > 
> > Today I am a bit busy, and not trace it much.
> > 
> > So far, I found the code hangs in scsi_test_unit_ready()
> > <-get_capabilities()<-sr_probe(), and scsi_queue_rq()/ata_scsi_queuecmd()
> > has queued the command successfully, but never completed.
> > 
> > Also tried to revert commits merged to ata tree on 30th, 31th,
> > but no difference.
> > 
> 
> Looking at my commit logs, the problem started to happen after various DMA
> changes were introduced. The boot tests fail on ppc (few), mips (all 32 bit,
> most 64 bit), i386 (all), x86_64 (most). All other platform pass, even with
> the same type of boot tests. Here is an example from alpha:
> 
> Building alpha:defconfig:initrd ... running .... passed
> Building alpha:defconfig:sata:rootfs ... running ..... passed
> Building alpha:defconfig:usb:rootfs ... running ..... passed
> Building alpha:defconfig:usb-uas:rootfs ... running ...... passed
> Building alpha:defconfig:scsi[AM53C974]:rootfs ... running ....... passed
> Building alpha:defconfig:scsi[DC395]:rootfs ... running ....... passed
> Building alpha:defconfig:scsi[MEGASAS]:rootfs ... running ...... passed
> Building alpha:defconfig:scsi[MEGASAS2]:rootfs ... running ...... passed
> Building alpha:defconfig:scsi[FUSION]:rootfs ... running ...... passed
> Building alpha:defconfig:nvme:rootfs ... running ..... passed
> 
> arm64:
> 
> Building arm64:virt:defconfig:smp:initrd ... running ..... passed
> Building arm64:virt:defconfig:smp:usb:rootfs ... running ..... passed
> Building arm64:virt:defconfig:smp:usb-uas:rootfs ... running ..... passed
> Building arm64:virt:defconfig:smp:virtio:rootfs ... running ..... passed
> Building arm64:virt:defconfig:smp:nvme:rootfs ... running ..... passed
> Building arm64:virt:defconfig:smp:mmc:rootfs ... running ..... passed
> Building arm64:virt:defconfig:smp:scsi[DC395]:rootfs ... running ..... passed
> Building arm64:virt:defconfig:smp:scsi[AM53C974]:rootfs ... running ..... passed
> Building arm64:virt:defconfig:smp:scsi[MEGASAS]:rootfs ... running ..... passed
> Building arm64:virt:defconfig:smp:scsi[MEGASAS2]:rootfs ... running ..... passed
> Building arm64:virt:defconfig:smp:scsi[53C810]:rootfs ... running ...... passed
> Building arm64:virt:defconfig:smp:scsi[53C895A]:rootfs ... running ...... passed
> Building arm64:virt:defconfig:smp:scsi[FUSION]:rootfs ... running ...... passed
> Skipping arm64:xlnx-zcu102:defconfig:smp:initrd:xilinx/zynqmp-ep108 ...
> Skipping arm64:xlnx-zcu102:defconfig:smp:sd:rootfs:xilinx/zynqmp-ep108 ...
> Skipping arm64:xlnx-zcu102:defconfig:smp:sata:rootfs:xilinx/zynqmp-ep108 ...
> Building arm64:xlnx-zcu102:defconfig:smp:initrd:xilinx/zynqmp-zcu102-rev1.0 ... running ....... passed
> Building arm64:xlnx-zcu102:defconfig:smp:sd1:rootfs:xilinx/zynqmp-zcu102-rev1.0 ... running ......... passed
> Building arm64:xlnx-zcu102:defconfig:smp:sata:rootfs:xilinx/zynqmp-zcu102-rev1.0 ... running ...... passed
> Building arm64:raspi3:defconfig:smp:initrd:broadcom/bcm2837-rpi-3-b ... running ..... passed
> Building arm64:raspi3:defconfig:smp:sd:rootfs:broadcom/bcm2837-rpi-3-b ... running ........ passed
> Building arm64:virt:defconfig:nosmp:initrd ... running ..... passed
> Skipping arm64:xlnx-zcu102:defconfig:nosmp:initrd:xilinx/zynqmp-ep108 ...
> Skipping arm64:xlnx-zcu102:defconfig:nosmp:sd:rootfs:xilinx/zynqmp-ep108 ...
> Building arm64:xlnx-zcu102:defconfig:nosmp:initrd:xilinx/zynqmp-zcu102-rev1.0 ... running ......... passed
> Building arm64:xlnx-zcu102:defconfig:nosmp:sd1:rootfs:xilinx/zynqmp-zcu102-rev1.0 ... running ......... passed
> 
> ppc:
> 
> Building powerpc:mac99:qemu_ppc_book3s_defconfig:nosmp:rootfs ... running ....... passed
> Building powerpc:g3beige:qemu_ppc_book3s_defconfig:nosmp:rootfs ... running ...... passed
> Building powerpc:mac99:qemu_ppc_book3s_defconfig:smp:rootfs ... running ....... passed
> Building powerpc:virtex-ml507:44x/virtex5_defconfig:devtmpfs:initrd ... running .... passed
> Building powerpc:mpc8544ds:mpc85xx_defconfig:initrd ... running .... passed
> Building powerpc:mpc8544ds:mpc85xx_defconfig:scsi:rootfs ... running ..... passed
> Building powerpc:mpc8544ds:mpc85xx_defconfig:sata:rootfs ... running .... passed
> Building powerpc:mpc8544ds:mpc85xx_smp_defconfig:initrd ... running .... passed
> Building powerpc:mpc8544ds:mpc85xx_smp_defconfig:scsi:rootfs ... running ..... passed
> Building powerpc:mpc8544ds:mpc85xx_smp_defconfig:sata:rootfs ... running .... passed
> Building powerpc:bamboo:44x/bamboo_defconfig:devtmpfs:initrd ... running .... passed
> Building powerpc:bamboo:44x/bamboo_defconfig:devtmpfs:scsi[AM53C974]:rootfs ... running ..... passed
> Building powerpc:bamboo:44x/bamboo_defconfig:devtmpfs:smp:initrd ... running .... passed
> Building powerpc:bamboo:44x/bamboo_defconfig:devtmpfs:smp:scsi[AM53C974]:rootfs ... running ..... passed
> Building powerpc:sam460ex:44x/canyonlands_defconfig:devtmpfs:initrd ... running ..... passed
> Building powerpc:sam460ex:44x/canyonlands_defconfig:devtmpfs:usbdisk:rootfs ... running ...... passed
> Building powerpc:mac99:pmac32_defconfig:devtmpfs:zilog:initrd ... running .................................. failed (timeout)
> Building powerpc:mac99:pmac32_defconfig:devtmpfs:zilog:rootfs ... running .................................. failed (timeout)
> 
> Maybe that is a coincidence, but it is at least suspicious.

This issue can be fixed by reverting d250bf4e776ff09d5 ("blk-mq: only iterate over
inflight requests in blk_mq_tagset_busy_iter").

This patch looks wrong, because 'blk_mq_rq_state(rq) == MQ_RQ_IN_FLIGHT'
isn't completely same with 'blk_mq_request_started(req)'.

Thanks,
Ming