> -----Original Message----- > From: Wood Scott-B07421 > Sent: Friday, October 15, 2010 0:01 AM > To: Zang Roy-R61911 > Cc: Wood Scott-B07421; Anton Vorontsov; linux-...@lists.infradead.org; > dw...@infradead.org; dedeki...@gmail.com; a...@linux-foundation.org; Lan > Chunhe-B25806; Gala Kumar-B11780; linuxppc-...@ozlabs.org; Hu Mingkai-B21284 > Subject: Re: [PATCH 2/3 v4] P4080/mtd: Only make elbc nand driver detect nand > flash partitions > > On Wed, 13 Oct 2010 20:09:02 -0700 > "Zang Roy-R61911" <r61...@freescale.com> wrote: > > > > > > > + struct fsl_elbc_fcm_ctrl *elbc_fcm_ctrl = NULL; > > > > > > > > > > No need for = NULL. > > > > Any harm? Or just personal habit or style? Can you explain about > > why? > > > > > > Besides not wanting superfluous code on general principle, it could > > > hide a bug if in the future the real initialization is missing on some > > > code path. It would become a runtime NULL dereference rather than a > > > compiler warning. > > > > Not exactly. > > Per my understand, if the pointer will definitely be assigned in code > > path, > > it is not necessary to init it when define. for example, > > > > char c; > > char b; > > char *a; > > if (condition) > > a = &c; > > else > > a = &b; > > ... > > > > for other case, if the path will not ensure the pointer assignment, it > > will be inited > > when define to avoid warning. for example, > > > > char c; > > char *a = NULL; > > if (condition) > > a = &c; > > ... > > Yes, but this patch looks like the former case, not the > latter. That is right. > Is GCC giving a warning without the initializer? no. we are on the same point. Thanks. Roy
_______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev