> -----Original Message----- > From: Russell King - ARM Linux [mailto:li...@arm.linux.org.uk] > Sent: 2013年5月22日 8:15 > To: Yang, Wenyou > Cc: linux-arm-ker...@lists.infradead.org; linux-kernel@vger.kernel.org; > plagn...@jcrosoft.com; Ferre, Nicolas; li...@maxim.org.za > Subject: Re: [PATCH] ARM: at91: Fix: Change internal SRAM memory type to > "MT_MEMORY_SO" > > On Mon, May 20, 2013 at 09:06:19AM +0800, Wenyou Yang wrote: > > Signed-off-by: Wenyou Yang <wenyou.y...@atmel.com> > > This needs more of a description. Also, for a single patch, it's silly > to send two mails, the first being a cover which has a little more > information in it about the patch than the patch itself. > > You need to explain _why_ you're making this change. What I want to see > is that you've thought about the implications of this - particularly that > you know that strongly ordered memory does *not* imply any ordering with > any other memory types. > > In other words, I want to know that this change is not a bodge but there's > a real reason behind it. The story is: for sama5d3x with Cortex-A5 core, if not so, when copying code snippet to the internal SRAM, then jump to run this code, but fail to run. So, refer to other code(such as omap4), do such change. I also test it on at91sam9 with 926ej-s core. As what you said, I am digging it. Thank you very much.
Best Regards, Wenyou Yang