[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAGsJ_4z-KRNTJsce_6pCOjEDtsiHHOh4C9K9iY1o_3--RMM-=w@mail.gmail.com>
Date: Tue, 23 Dec 2025 15:58:28 +1300
From: Barry Song <21cnbao@...il.com>
To: Qinxin Xia <xiaqinxin@...wei.com>
Cc: jonathan.cameron@...wei.com, wangzhou1@...ilicon.com,
iommu@...ts.linux.dev, prime.zeng@...wei.com, fanghao11@...wei.com,
linux-kernel@...r.kernel.org, linuxarm@...wei.com
Subject: Re: [PATCH v5 1/3] dma-mapping: benchmark: modify the framework to
adapt to more map modes
> >> + if (!mb_ops->prepare_data)
> >> + mb_ops->prepare_data(mparam);
> >
> > Did you actually test it? and why don't you need prepare_data() sometimes?
> >
> > Thanks
> > Barry
> 'if (mb_ops->prepare_data)' is correct.
> If you want to do some tests before I release the next version, you can
> modify it in this way. I have tested it and it is work.
I’ve reviewed it many times. You are:
+ if (!mb_ops->prepare_data)
+ mb_ops->prepare_data(mparam);
not
if (mb_ops->prepare_data)
mb_ops->prepare_data(mparam);
Am I crazy?
Thanks
Barry
Powered by blists - more mailing lists