From: Ming Lei <ming.lei@redhat.com>
To: Johannes Thumshirn <jthumshirn@suse.de>
Cc: Jens Axboe <axboe@fb.com>, Omar Sandoval <osandov@osandov.com>,
Bart Van Assche <Bart.VanAssche@sandisk.com>,
Hannes Reinecke <hare@suse.de>, Christoph Hellwig <hch@lst.de>,
Linux Block Layer Mailinglist <linux-block@vger.kernel.org>,
Linux Kernel Mailinglist <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] block: bios with an offset are always gappy
Date: Thu, 13 Apr 2017 20:11:40 +0800 [thread overview]
Message-ID: <20170413121134.GA32362@ming.t460p> (raw)
In-Reply-To: <20170413115328.GH6734@linux-x5ow.site>
[-- Attachment #1: Type: text/plain, Size: 1032 bytes --]
On Thu, Apr 13, 2017 at 01:53:28PM +0200, Johannes Thumshirn wrote:
> On Thu, Apr 13, 2017 at 06:02:21PM +0800, Ming Lei wrote:
> > On Thu, Apr 13, 2017 at 10:06:29AM +0200, Johannes Thumshirn wrote:
> > > Doing a mkfs.btrfs on a (qemu emulated) PCIe NVMe causes a kernel panic
> > > in nvme_setup_prps() because the dma_len will drop below zero but the
> > > length not.
> >
> > Looks I can't reproduce the issue in QEMU(32G nvme, either partitioned
> > or not, just use 'mkfs.btrfs /dev/nvme0n1p1'), could you share the exact
> > mkfs command line and size of your emulated NVMe?
>
> the exact cmdline is mkfs.btrfs -f /dev/nvme0n1p1 (-f because there was a
> existing btrfs on the image). The image is 17179869184 (a.k.a 16G) bytes.
>
> [...]
>
> > Could you try the following patch to see if it fixes your issue?
>
> It's back to the old, erratic behaviour, see log below.
Ok, could you apply the attached debug patch and collect the
ftrace log? (ftrace_dump_on_oops need to be passed to kernel cmd line).
Thanks,
Ming
[-- Attachment #2: dump-rq.patch --]
[-- Type: text/plain, Size: 2383 bytes --]
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 26a5fd05fe88..a813a36d48d9 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -491,6 +491,8 @@ static bool nvme_setup_prps(struct nvme_dev *dev, struct request *req)
break;
if (dma_len > 0)
continue;
+ if (dma_len < 0)
+ blk_dump_rq(req, "nvme dma sg gap");
BUG_ON(dma_len < 0);
sg = sg_next(sg);
dma_addr = sg_dma_address(sg);
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 8e521194f6fc..f3b001e401d2 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -811,5 +811,29 @@ static inline int bio_integrity_add_page(struct bio *bio, struct page *page,
#endif /* CONFIG_BLK_DEV_INTEGRITY */
+static inline void blk_dump_bio(struct bio *bio, const char *msg)
+{
+ struct bvec_iter iter;
+ struct bio_vec bvec;
+ int i = 0;
+ unsigned sectors = 0;
+
+ trace_printk("%s-%p: %hx/%hx %u %llu %u\n",
+ msg, bio,
+ bio->bi_flags, bio->bi_opf,
+ bio->bi_phys_segments,
+ (unsigned long long)bio->bi_iter.bi_sector,
+ bio->bi_iter.bi_size);
+ bio_for_each_segment(bvec, bio, iter) {
+ sectors += bvec.bv_len >> 9;
+ trace_printk("\t %d: %lu %u %u(%u)\n", i++,
+ (unsigned long)page_to_pfn(bvec.bv_page),
+ bvec.bv_offset,
+ bvec.bv_len, bvec.bv_len >> 12);
+ }
+ trace_printk("\t total sectors %u\n", sectors);
+}
+
+
#endif /* CONFIG_BLOCK */
#endif /* __LINUX_BIO_H */
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 7548f332121a..b75d6fe5a1b9 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1698,6 +1698,22 @@ static inline bool req_gap_front_merge(struct request *req, struct bio *bio)
return bio_will_gap(req->q, bio, req->bio);
}
+static inline void blk_dump_rq(const struct request *req, const char *msg)
+{
+ struct bio *bio;
+ int i = 0;
+
+ trace_printk("%s: dump bvec for %p(f:%x, seg: %d)\n",
+ msg, req, req->cmd_flags,
+ req->nr_phys_segments);
+
+ __rq_for_each_bio(bio, req) {
+ char num[16];
+ snprintf(num, 16, "%d", i++);
+ blk_dump_bio(bio, num);
+ }
+}
+
int kblockd_schedule_work(struct work_struct *work);
int kblockd_schedule_work_on(int cpu, struct work_struct *work);
int kblockd_schedule_delayed_work(struct delayed_work *dwork, unsigned long delay);
next prev parent reply other threads:[~2017-04-13 12:11 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-13 8:06 [PATCH] block: bios with an offset are always gappy Johannes Thumshirn
2017-04-13 9:48 ` Christoph Hellwig
2017-04-13 9:56 ` Johannes Thumshirn
2017-04-13 10:01 ` Johannes Thumshirn
2017-04-13 10:02 ` Ming Lei
2017-04-13 10:10 ` Johannes Thumshirn
2017-04-13 11:53 ` Johannes Thumshirn
2017-04-13 12:11 ` Ming Lei [this message]
[not found] ` <20170413122010.GJ6734@linux-x5ow.site>
2017-04-13 13:44 ` Ming Lei
2017-04-13 14:45 ` Ming Lei
2017-04-13 14:50 ` Johannes Thumshirn
2017-04-13 20:35 ` Andreas Mohr
2017-04-14 1:15 ` Ming Lei
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170413121134.GA32362@ming.t460p \
--to=ming.lei@redhat.com \
--cc=Bart.VanAssche@sandisk.com \
--cc=axboe@fb.com \
--cc=hare@suse.de \
--cc=hch@lst.de \
--cc=jthumshirn@suse.de \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=osandov@osandov.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox