From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tom Rini Date: Tue, 25 Mar 2014 16:04:35 -0400 Subject: [U-Boot] [PATCH V3 09/13] ARM: tegra: use clrsetbits_le32 in pinmux driver In-Reply-To: <1395764855-23377-1-git-send-email-swarren@wwwdotorg.org> References: <1395764855-23377-1-git-send-email-swarren@wwwdotorg.org> Message-ID: <20140325200435.GV16360@bill-the-cat> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de On Tue, Mar 25, 2014 at 10:27:35AM -0600, Stephen Warren wrote: > From: Stephen Warren > > This removes a bunch of open-coded register IO, masking, and shifting. > I would have squashed this into "ARM: tegra: pinctrl: remove duplication" > except that keeping it a separate commit allows easier bisection of any > issues that are introduced by this patch. I also wrote this patch on top > of the series, and pushing it any lower in the series results in some > conflicts I didn't feel like fixing. Since things got a bit heated here while I was reading some other stuff... [snip] > +static inline void update_reg_mask_shift_val(u32 *reg, u32 mask, u32 shift, > + u32 val) > +{ > + clrsetbits_le32(reg, mask << shift, val << shift); > +} So, lack of comments bad. Intention, good. We have a bitfield of size M (that's all cleared in the mask) and value that may be less than M bits wide. The name is a mouthful (but I see where Simon was coming from, had I caught in time I might have suggested a comment instead. But as Wolfgang's v4 shows, it's also not hard to just call clrsetbits_le32 directly. Arguably the cases where mask==1 we should just call setbits_le32 but that's not a big deal. -- Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 836 bytes Desc: Digital signature URL: