All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg KH <greg@kroah.com>
To: Jianxiong Gao <jxgao@google.com>
Cc: stable@vger.kernel.org
Subject: Re: [PATCH v5.10 0/8] preserve DMA offsets when using swiotlb
Date: Sat, 24 Apr 2021 16:43:48 +0200	[thread overview]
Message-ID: <YIQupJuzbdgif6WA@kroah.com> (raw)
In-Reply-To: <CAMGD6P2gUpUuX5cdPi1Q0nqRFmsBPctUR+hBt+DnPK+H4jHiiQ@mail.gmail.com>

On Fri, Apr 23, 2021 at 10:28:32AM -0700, Jianxiong Gao wrote:
> > How?  Anything that installed 5.10 when it was released never had this
> > working, they had to move to 5.12 to get that to work.
> 
> I wasn't clear. The bug is not specific to SEV virtualization. We
> simply encountered it while working on SEV virtualization. This is a
> pre-existing bug.
> 
> Briefly, the NVMe spec expects a page offset to be retained from the
> memory address space to the IO address space.
> 
> Before these patches, the SWIOTLB truncates any page offset.
> 
> Thus, all NVMe + SWIOTLB systems are broken due to this bug without
> these patches.

Ok, and what prevents you from adding this new feature do your "custom"
kernel, or to use 5.12 instead?

This is a new feature that Linux has never supported, and these patches
are not "trivial" at all.  I also do not see the maintainer of the
subsystem agreeing that these are needed to be backported, which is not
a good sign.

So I recommend just using a newer kernel version, that way all will be
good and no need to backport anything.  What is preventing you from
doing that today?

thanks,

greg k-h

  parent reply	other threads:[~2021-04-24 14:43 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-05 21:02 [PATCH v5.10 0/8] preserve DMA offsets when using swiotlb Jianxiong Gao
2021-04-05 21:02 ` [PATCH v5.10 1/8] driver core: add a min_align_mask field to struct device_dma_parameters Jianxiong Gao
2021-04-07 12:50   ` Greg KH
2021-04-05 21:02 ` [PATCH v5.10 2/8] swiotlb: factor out an io_tlb_offset helper Jianxiong Gao
2021-04-05 21:02 ` [PATCH v5.10 3/8] swiotlb: factor out a nr_slots helper Jianxiong Gao
2021-04-05 21:02 ` [PATCH v5.10 4/8] swiotlb: clean up swiotlb_tbl_unmap_single Jianxiong Gao
2021-04-05 21:02 ` [PATCH v5.10 5/8] swiotlb: refactor swiotlb_tbl_map_single Jianxiong Gao
2021-04-05 21:02 ` [PATCH v5.10 6/8] swiotlb: don't modify orig_addr in swiotlb_tbl_sync_single Jianxiong Gao
2021-04-05 21:02 ` [PATCH v5.10 7/8] swiotlb: respect min_align_mask Jianxiong Gao
2021-04-05 21:02 ` [PATCH v5.10 8/8] nvme-pci: set min_align_mask Jianxiong Gao
2021-04-07 12:51 ` [PATCH v5.10 0/8] preserve DMA offsets when using swiotlb Greg KH
2021-04-20 23:38   ` Jianxiong Gao
2021-04-23 15:14     ` Greg KH
2021-04-23 17:28       ` Jianxiong Gao
2021-04-23 22:10         ` Jianxiong Gao
2021-04-23 22:33           ` Marc Orr
2021-04-24 14:43         ` Greg KH [this message]
2021-04-26 18:00           ` Jianxiong Gao
2021-04-28 16:18             ` Sasha Levin

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=YIQupJuzbdgif6WA@kroah.com \
    --to=greg@kroah.com \
    --cc=jxgao@google.com \
    --cc=stable@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.