From: Konrad Rzeszutek Wilk <konrad.wilk-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>
To: Akinobu Mita <akinobu.mita-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: Andi Kleen <andi-Vw/NltI1exuRpAAqCnN02g@public.gmane.org>,
x86-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
LKML <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
Ingo Molnar <mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
"H. Peter Anvin" <hpa-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org>,
Andrew Morton
<akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>,
Thomas Gleixner <tglx-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org>
Subject: Re: [PATCH v2 1/5] x86: make dma_alloc_coherent() return zeroed memory if CMA is enabled
Date: Fri, 24 Jan 2014 13:07:41 -0500 [thread overview]
Message-ID: <20140124180741.GC15785@phenom.dumpdata.com> (raw)
In-Reply-To: <CAC5umygcr0a_BfDJ9tB1EDry4UEwThG7WvkA8Dn27zO9RQab8Q-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
On Fri, Jan 17, 2014 at 08:34:38AM +0900, Akinobu Mita wrote:
> 2014/1/16 Konrad Rzeszutek Wilk <konrad.wilk-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>:
> > On Tue, Jan 14, 2014 at 11:13:46PM +0900, Akinobu Mita wrote:
> >> Calling dma_alloc_coherent() with __GFP_ZERO must return zeroed memory.
> >>
> >> But when the contiguous memory allocator (CMA) is enabled on x86 and
> >> the memory region is allocated by dma_alloc_from_contiguous(), it
> >> doesn't return zeroed memory. Because dma_generic_alloc_coherent()
> >
> > So why not fix it there to return zeroed out memory?
>
> I thought it looked nicer than this patch as we can remove memset
> from all caller of dma_alloc_from_contiguous(). But if I look at
> the caller on arm, we can't simply remove the memset because
> __dma_clear_buffer() is used there for ensuring cache flushing and
> it is used in many places.
OK, that should be part of the commit description.
>
> Of course we can do redundant memset in dma_alloc_from_contiguous(),
> but now I think this patch is less impact for fixing this problem.
OK.
WARNING: multiple messages have this Message-ID (diff)
From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: Akinobu Mita <akinobu.mita@gmail.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Marek Szyprowski <m.szyprowski@samsung.com>,
David Woodhouse <dwmw2@infradead.org>,
Don Dutile <ddutile@redhat.com>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
Andi Kleen <andi@firstfloor.org>,
x86@kernel.org, iommu@lists.linux-foundation.org
Subject: Re: [PATCH v2 1/5] x86: make dma_alloc_coherent() return zeroed memory if CMA is enabled
Date: Fri, 24 Jan 2014 13:07:41 -0500 [thread overview]
Message-ID: <20140124180741.GC15785@phenom.dumpdata.com> (raw)
In-Reply-To: <CAC5umygcr0a_BfDJ9tB1EDry4UEwThG7WvkA8Dn27zO9RQab8Q@mail.gmail.com>
On Fri, Jan 17, 2014 at 08:34:38AM +0900, Akinobu Mita wrote:
> 2014/1/16 Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>:
> > On Tue, Jan 14, 2014 at 11:13:46PM +0900, Akinobu Mita wrote:
> >> Calling dma_alloc_coherent() with __GFP_ZERO must return zeroed memory.
> >>
> >> But when the contiguous memory allocator (CMA) is enabled on x86 and
> >> the memory region is allocated by dma_alloc_from_contiguous(), it
> >> doesn't return zeroed memory. Because dma_generic_alloc_coherent()
> >
> > So why not fix it there to return zeroed out memory?
>
> I thought it looked nicer than this patch as we can remove memset
> from all caller of dma_alloc_from_contiguous(). But if I look at
> the caller on arm, we can't simply remove the memset because
> __dma_clear_buffer() is used there for ensuring cache flushing and
> it is used in many places.
OK, that should be part of the commit description.
>
> Of course we can do redundant memset in dma_alloc_from_contiguous(),
> but now I think this patch is less impact for fixing this problem.
OK.
next prev parent reply other threads:[~2014-01-24 18:07 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-01-14 14:13 [PATCH v2 0/5] enhance DMA CMA on x86 Akinobu Mita
2014-01-14 14:13 ` Akinobu Mita
[not found] ` <1389708830-5841-1-git-send-email-akinobu.mita-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-01-14 14:13 ` [PATCH v2 1/5] x86: make dma_alloc_coherent() return zeroed memory if CMA is enabled Akinobu Mita
2014-01-14 14:13 ` Akinobu Mita
[not found] ` <1389708830-5841-2-git-send-email-akinobu.mita-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-01-15 20:09 ` Konrad Rzeszutek Wilk
2014-01-15 20:09 ` Konrad Rzeszutek Wilk
[not found] ` <20140115200934.GC5201-6K5HmflnPlqSPmnEAIUT9EEOCMrvLtNR@public.gmane.org>
2014-01-16 23:34 ` Akinobu Mita
2014-01-16 23:34 ` Akinobu Mita
[not found] ` <CAC5umygcr0a_BfDJ9tB1EDry4UEwThG7WvkA8Dn27zO9RQab8Q-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-01-24 18:07 ` Konrad Rzeszutek Wilk [this message]
2014-01-24 18:07 ` Konrad Rzeszutek Wilk
2014-01-27 13:54 ` Marek Szyprowski
2014-01-27 13:54 ` Marek Szyprowski
[not found] ` <52E66527.8010000-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org>
2014-01-27 15:23 ` Konrad Rzeszutek Wilk
2014-01-27 15:23 ` Konrad Rzeszutek Wilk
[not found] ` <20140127152342.GC20968-6K5HmflnPlqSPmnEAIUT9EEOCMrvLtNR@public.gmane.org>
2014-01-28 23:20 ` Akinobu Mita
2014-01-28 23:20 ` Akinobu Mita
[not found] ` <CAC5umyjJfO9UyAkoG-bEX-vdO+X0fM_YD2dwj2AHT6T7hKgv0A-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-02-03 12:14 ` Akinobu Mita
2014-02-03 12:14 ` Akinobu Mita
2014-01-14 14:13 ` [PATCH v2 2/5] x86: enable DMA CMA with swiotlb Akinobu Mita
2014-01-14 14:13 ` Akinobu Mita
[not found] ` <1389708830-5841-3-git-send-email-akinobu.mita-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-01-15 20:12 ` Konrad Rzeszutek Wilk
2014-01-15 20:12 ` Konrad Rzeszutek Wilk
[not found] ` <20140115201224.GD5201-6K5HmflnPlqSPmnEAIUT9EEOCMrvLtNR@public.gmane.org>
2014-01-16 23:32 ` Akinobu Mita
2014-01-16 23:32 ` Akinobu Mita
[not found] ` <CAC5umyitWHvWiWn9B3wtTNBhbO7Rz_ghqmxhwu9caq40msnfAA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-01-24 18:06 ` Konrad Rzeszutek Wilk
2014-01-24 18:06 ` Konrad Rzeszutek Wilk
2014-01-14 14:13 ` [PATCH v2 3/5] intel-iommu: integrate DMA CMA Akinobu Mita
2014-01-14 14:13 ` Akinobu Mita
[not found] ` <1389708830-5841-4-git-send-email-akinobu.mita-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2014-01-16 14:28 ` Marek Szyprowski
2014-01-16 14:28 ` Marek Szyprowski
[not found] ` <52D7EC74.7030401-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org>
2014-01-16 23:28 ` Akinobu Mita
2014-01-16 23:28 ` Akinobu Mita
2014-01-14 14:13 ` [PATCH v2 4/5] memblock: introduce memblock_alloc_range() Akinobu Mita
2014-01-14 14:13 ` Akinobu Mita
2014-01-14 14:13 ` [PATCH v2 5/5] cma: add placement specifier for "cma=" kernel parameter Akinobu Mita
2014-01-14 14:13 ` Akinobu Mita
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=20140124180741.GC15785@phenom.dumpdata.com \
--to=konrad.wilk-qhclzuegtsvqt0dzr+alfa@public.gmane.org \
--cc=akinobu.mita-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org \
--cc=andi-Vw/NltI1exuRpAAqCnN02g@public.gmane.org \
--cc=dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
--cc=hpa-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org \
--cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=tglx-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org \
--cc=x86-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.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.