From: Christoph Hellwig <hch@lst.de>
To: Halil Pasic <pasic@linux.ibm.com>
Cc: "Christoph Hellwig" <hch@lst.de>,
"Petr Tesařík" <petr@tesarici.cz>,
"Niklas Schnelle" <schnelle@linux.ibm.com>,
"Bjorn Helgaas" <bhelgaas@google.com>,
"Marek Szyprowski" <m.szyprowski@samsung.com>,
"Robin Murphy" <robin.murphy@arm.com>,
"Petr Tesarik" <petr.tesarik1@huawei-partners.com>,
"Ross Lagerwall" <ross.lagerwall@citrix.com>,
linux-pci <linux-pci@vger.kernel.org>,
linux-kernel@vger.kernel.org, iommu@lists.linux.dev,
"Matthew Rosato" <mjrosato@linux.ibm.com>,
"Jianxiong Gao" <jxgao@google.com>
Subject: Re: Memory corruption with CONFIG_SWIOTLB_DYNAMIC=y
Date: Wed, 8 Nov 2023 08:30:48 +0100 [thread overview]
Message-ID: <20231108073048.GA5277@lst.de> (raw)
In-Reply-To: <20231107182437.06632f6e.pasic@linux.ibm.com>
On Tue, Nov 07, 2023 at 06:24:37PM +0100, Halil Pasic wrote:
> Thanks Christoph! So for NVMe in certain contexts the low
> bits of addresses get discarded, but in other contexts the high bits
> of addresses get discarded and the low bits need to remain the same
> after the bounce (and that's why we need commits 36950f2da1ea ("driver
> core: add a min_align_mask) and 1f221a0d0dbf ("swiotlb: respect
> min_align_mask").
Nothing really gets discarded. NVMe basicaly requires all pages
(where the page is a device page, currently hard coded to 4k in
Linux) except for the first to not have an offset. This is what
the driver sets the virt boundary to in Linux. When bounce buffering
thus all segments (except possibly the first) need to keep their
alignment.
next prev parent reply other threads:[~2023-11-08 7:30 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-03 15:13 Memory corruption with CONFIG_SWIOTLB_DYNAMIC=y Niklas Schnelle
2023-11-03 16:14 ` Halil Pasic
2023-11-03 20:50 ` Petr Tesařík
2023-11-06 7:42 ` Christoph Hellwig
2023-11-07 17:24 ` Halil Pasic
2023-11-08 7:30 ` Christoph Hellwig [this message]
2023-11-06 10:08 ` Halil Pasic
2023-11-07 17:24 ` Halil Pasic
2023-11-08 9:13 ` Petr Tesařík
2023-11-23 10:16 ` Petr Tesařík
2023-11-27 15:59 ` Christoph Hellwig
2023-11-28 7:16 ` Petr Tesařík
2023-11-03 18:59 ` Petr Tesařík
2023-11-06 7:44 ` Christoph Hellwig
2023-11-06 12:46 ` Petr Tesarik
2023-11-08 10:52 ` Halil Pasic
2023-11-08 11:04 ` Petr Tesarik
2023-11-08 14:32 ` Halil Pasic
2023-11-08 14:45 ` Petr Tesarik
2023-11-10 9:22 ` Halil Pasic
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=20231108073048.GA5277@lst.de \
--to=hch@lst.de \
--cc=bhelgaas@google.com \
--cc=iommu@lists.linux.dev \
--cc=jxgao@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=m.szyprowski@samsung.com \
--cc=mjrosato@linux.ibm.com \
--cc=pasic@linux.ibm.com \
--cc=petr.tesarik1@huawei-partners.com \
--cc=petr@tesarici.cz \
--cc=robin.murphy@arm.com \
--cc=ross.lagerwall@citrix.com \
--cc=schnelle@linux.ibm.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;
as well as URLs for NNTP newsgroup(s).