2013/6/23 Albert ARIBAUD <albert.u.b...@aribaud.net>: > Hi Kuo-Jung, > > On Mon, 17 Jun 2013 20:06:52 +0800, Kuo-Jung Su <dant...@gmail.com> > wrote: > >> 3. No matter if MMU/D-cache is on or off, this patch >> always depends on previous patch: >> >> arm: dma_alloc_coherent: malloc() -> memalign(). >> >> Because the FTGMAC100 expects the tx/rx descriptors >> are always be aligned to 16-bytes boundary. > > Nitpick: as a general rule, the commit message does not need to mention > that a patch in a series depends on a previous one; this is implied by > the series itself. > > In fact, it should even /refrain/ from doing so because even though the > series is usually applied atomically, it is not impossible that a an > unrelated 'hot' bugfix patch slip in the middle of it, so "previous > patch", which makes sense at submission time, might not make sense any > more later. > > Personally I like the commit message to only provide a short explanation > of the whys and hows of the patch itself, and put any additional info > after the commit message separator "---" below or, if using patman, in > the cover letter. >
Got it, thanks -- Best wishes, Kuo-Jung Su _______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
