From: Thorsten Leemhuis <regressions@leemhuis.info>
To: Mykola Marzhan <mykola@meshstor.io>,
Kiran Kumar Modukuri <kmodukuri@nvidia.com>,
Chaitanya Kulkarni <kch@nvidia.com>, Jens Axboe <axboe@kernel.dk>
Cc: Li Nan <magiclinan@didiglobal.com>, Xiao Ni <xiao@kernel.org>,
Song Liu <song@kernel.org>,
linux-block@vger.kernel.org, Christoph Hellwig <hch@lst.de>,
linux-nvme@lists.infradead.org,
Guoqing Jiang <guoqing.jiang@linux.dev>,
linux-raid@vger.kernel.org, Leon Romanovsky <leon@kernel.org>,
Jason Gunthorpe <jgg@ziepe.ca>,
Logan Gunthorpe <logang@deltatee.com>,
Bjorn Helgaas <bhelgaas@google.com>,
Shivaji Kant <shivajikant@google.com>,
Pranjal Shrivastava <praan@google.com>,
Henrique Carvalho <henrique.carvalho@suse.com>,
stable@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-rdma@vger.kernel.org, linux-pci@vger.kernel.org,
Sagi Grimberg <sagi@grimberg.me>, Yu Kuai <yukuai@fygo.io>,
Keith Busch <kbusch@kernel.org>,
Linux kernel regressions list <regressions@lists.linux.dev>
Subject: Re: [PATCH v5 00/10] block,md,nvme: correct handling of unsupported P2PDMA transfers
Date: Fri, 24 Jul 2026 12:21:29 +0200 [thread overview]
Message-ID: <c969f48f-833c-4a0d-8e3c-c7e88f9a052d@leemhuis.info> (raw)
In-Reply-To: <20260723204206.76930-1-mykola@meshstor.io>
On 7/23/26 22:41, Mykola Marzhan wrote:
> md treats an unsupported P2PDMA transfer to a member as success, and
> nvme-rdma retries one forever. Whether peer memory can be DMA-mapped
> depends on the PCIe topology between the two devices, and since v6.17
> the failure completes as BLK_STS_INVAL: retryable for multipath and
> deliberately ignored by md for member failures, so mirrors silently
> diverge.
> [...]
> Routing: patch 1 block, 2-8 md, 9-10 nvme; 7, 8 and 10 use the new
> status. Patch 1 alone stops the divergence and the dm-multipath
> requeue loop; the nvme-rdma loop also needs 9-10; everything else
> refines the fallout, so the series bisects safely. The md patches
> fix a v7.2-rc1 regression (02666132403a), except 3 (older bug,
> stable) and 5 (refactor).
This landed on my radar due to the word "regression" here. Given the
complexity of this patch-set and that we are at v5 please allow me to ask:
Should 02666132403aec ("md: propagate BLK_FEAT_PCI_P2PDMA from member
devices to RAID device") [v7.2-rc1] maybe be removed for now and
reapplied later once this was sorted out? Or is that for some reason
unneeded? Note, no need for a long explanation needed in the latter
case, a brief statement from long-term developer will suffice.
Ciao, Thorsten
prev parent reply other threads:[~2026-07-24 10:21 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-23 20:41 Mykola Marzhan
2026-07-23 20:41 ` [PATCH v5 01/10] block: add BLK_STS_P2PDMA for unsupported peer-to-peer transfers Mykola Marzhan
2026-07-23 20:41 ` [PATCH v5 02/10] md: ensure REQ_NOMERGE is set on P2PDMA bios Mykola Marzhan
2026-07-23 20:41 ` [PATCH v5 03/10] md/raid1: serialize non-write-behind writes on CollisionCheck rdevs Mykola Marzhan
2026-07-23 20:42 ` [PATCH v5 04/10] md/raid1: don't use write-behind for P2PDMA bios Mykola Marzhan
2026-07-23 20:42 ` [PATCH v5 05/10] md/raid1,raid10: factor out raid1_write_error() helper Mykola Marzhan
2026-07-23 20:42 ` [PATCH v5 06/10] md/raid1,raid10: keep REQ_NOMERGE on narrow_write_error() retry clones Mykola Marzhan
2026-07-23 20:42 ` [PATCH v5 07/10] md/raid1,raid10: skip futile retries on P2PDMA mapping failures Mykola Marzhan
2026-07-24 15:11 ` Logan Gunthorpe
2026-07-23 20:42 ` [PATCH v5 08/10] md/raid1,raid10: set IO_BLOCKED in case of BLK_STS_P2PDMA Mykola Marzhan
2026-07-23 20:42 ` [PATCH v5 09/10] nvme-rdma: use ib_dma_map_sgtable_attrs() Mykola Marzhan
2026-07-24 18:47 ` Logan Gunthorpe
2026-07-23 20:42 ` [PATCH v5 10/10] nvme-rdma: return BLK_STS_P2PDMA for unsupported P2P transfers Mykola Marzhan
2026-07-24 18:50 ` Logan Gunthorpe
2026-07-24 10:21 ` Thorsten Leemhuis [this message]
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=c969f48f-833c-4a0d-8e3c-c7e88f9a052d@leemhuis.info \
--to=regressions@leemhuis.info \
--cc=axboe@kernel.dk \
--cc=bhelgaas@google.com \
--cc=guoqing.jiang@linux.dev \
--cc=hch@lst.de \
--cc=henrique.carvalho@suse.com \
--cc=jgg@ziepe.ca \
--cc=kbusch@kernel.org \
--cc=kch@nvidia.com \
--cc=kmodukuri@nvidia.com \
--cc=leon@kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=linux-pci@vger.kernel.org \
--cc=linux-raid@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=logang@deltatee.com \
--cc=magiclinan@didiglobal.com \
--cc=mykola@meshstor.io \
--cc=praan@google.com \
--cc=regressions@lists.linux.dev \
--cc=sagi@grimberg.me \
--cc=shivajikant@google.com \
--cc=song@kernel.org \
--cc=stable@vger.kernel.org \
--cc=xiao@kernel.org \
--cc=yukuai@fygo.io \
/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
all inboxes | Powered by JetHome®