From: Matthew Wilcox <willy@infradead.org>
To: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org,
netdev@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
linux-arm-kernel@lists.infradead.org, linux-mips@vger.kernel.org,
Ilias Apalodimas <ilias.apalodimas@linaro.org>,
Matteo Croce <mcroce@linux.microsoft.com>,
Grygorii Strashko <grygorii.strashko@ti.com>,
Arnd Bergmann <arnd@kernel.org>, Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH 1/1] mm: Fix struct page layout on 32-bit systems
Date: Wed, 14 Apr 2021 12:50:52 +0100 [thread overview]
Message-ID: <20210414115052.GS2531743@casper.infradead.org> (raw)
In-Reply-To: <20210414101044.19da09df@carbon>
On Wed, Apr 14, 2021 at 10:10:44AM +0200, Jesper Dangaard Brouer wrote:
> Yes, indeed! - And very frustrating. It's keeping me up at night.
> I'm dreaming about 32 vs 64 bit data structures. My fitbit stats tell
> me that I don't sleep well with these kind of dreams ;-)
Then you're going to love this ... even with the latest patch, there's
still a problem. Because dma_addr_t is still 64-bit aligned _as a type_,
that forces the union to be 64-bit aligned (as we already knew and worked
around), but what I'd forgotten is that forces the entirety of struct
page to be 64-bit aligned. Which means ...
/* size: 40, cachelines: 1, members: 4 */
/* padding: 4 */
/* forced alignments: 1 */
/* last cacheline: 40 bytes */
} __attribute__((__aligned__(8)));
.. that we still have a hole! It's just moved from being at offset 4
to being at offset 36.
> That said, I think we need to have a quicker fix for the immediate
> issue with 64-bit bit dma_addr on 32-bit arch and the misalignment hole
> it leaves[3] in struct page. In[3] you mention ppc32, does it only
> happens on certain 32-bit archs?
AFAICT it happens on mips32, ppc32, arm32 and arc. It doesn't happen
on x86-32 because dma_addr_t is 32-bit aligned.
Doing this fixes it:
+++ b/include/linux/types.h
@@ -140,7 +140,7 @@ typedef u64 blkcnt_t;
* so they don't care about the size of the actual bus addresses.
*/
#ifdef CONFIG_ARCH_DMA_ADDR_T_64BIT
-typedef u64 dma_addr_t;
+typedef u64 __attribute__((aligned(sizeof(void *)))) dma_addr_t;
#else
typedef u32 dma_addr_t;
#endif
> I'm seriously considering removing page_pool's support for doing/keeping
> DMA-mappings on 32-bit arch's. AFAIK only a single driver use this.
... if you're going to do that, then we don't need to do this.
next prev parent reply other threads:[~2021-04-14 11:51 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-10 20:52 [PATCH 0/1] Fix struct page layout on 32-bit systems Matthew Wilcox (Oracle)
2021-04-10 20:52 ` [PATCH 1/1] mm: " Matthew Wilcox (Oracle)
2021-04-11 9:43 ` Jesper Dangaard Brouer
2021-04-11 10:33 ` Matthew Wilcox
2021-04-12 1:15 ` Matthew Wilcox
2021-04-14 8:10 ` Jesper Dangaard Brouer
2021-04-14 11:50 ` Matthew Wilcox [this message]
2021-04-14 11:56 ` Ilias Apalodimas
2021-04-14 15:52 ` David Laight
2021-04-14 19:13 ` Jesper Dangaard Brouer
2021-04-14 21:35 ` Matthew Wilcox
2021-04-14 21:56 ` David Laight
2021-04-15 18:08 ` Jesper Dangaard Brouer
2021-04-15 18:21 ` Matthew Wilcox
2021-04-15 21:11 ` David Laight
2021-04-15 22:22 ` Matthew Wilcox
2021-04-16 7:32 ` David Laight
2021-04-16 11:05 ` Matthew Wilcox
2021-04-16 15:27 ` Matthew Wilcox
2021-04-16 17:08 ` Jesper Dangaard Brouer
2021-04-17 3:19 ` Matthew Wilcox
2021-04-17 10:31 ` Arnd Bergmann
2021-04-17 13:56 ` Matthew Wilcox
2021-04-17 17:30 ` Arnd Bergmann
2021-04-17 10:59 ` David Laight
2021-04-19 6:34 ` Christoph Hellwig
2021-04-19 7:15 ` Ilias Apalodimas
2021-04-12 18:23 ` Matthew Wilcox
2021-04-13 8:21 ` David Laight
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=20210414115052.GS2531743@casper.infradead.org \
--to=willy@infradead.org \
--cc=arnd@kernel.org \
--cc=brouer@redhat.com \
--cc=grygorii.strashko@ti.com \
--cc=hch@lst.de \
--cc=ilias.apalodimas@linaro.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mips@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mcroce@linux.microsoft.com \
--cc=netdev@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 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).