On Thu, Oct 22, 2015 at 04:40:47PM +0300, Nikita Kiryanov wrote: > On Thu, Oct 22, 2015 at 02:47:29PM +0200, Hans de Goede wrote: > Hi Hans, > > Hi, > > > > On 22-10-15 14:01, Nikita Kiryanov wrote: > > >Implement defaults for the raw partition image loading so that the #ifdef > > >CONFIG_SYS_MMCSD_RAW_MODE_U_BOOT_PARTITION in spl_mmc_load_image() will no > > >longer be necessary. > > > > > >This change makes it possible for mmc_load_image_raw_partition() and > > >mmc_load_image_raw_sector() to coexist. > > > > > >Signed-off-by: Nikita Kiryanov <nik...@compulab.co.il> > > >Cc: Igor Grinberg <grinb...@compulab.co.il> > > >Cc: Paul Kocialkowski <cont...@paulk.fr> > > >Cc: Pantelis Antoniou <pa...@antoniou-consulting.com> > > >Cc: Tom Rini <tr...@konsulko.com> > > >Cc: Simon Glass <s...@chromium.org> > > > > Same remark as with the previous patch, I'm not happy to see all these > > patches removing #ifdef-s given that spl is severely size constrained > > on some devices (e.g. recently we had a patchset for the rockchip 3036, > > which has only 8k space for the SPL). > > > > And I really do not see a need for this, boards which want to use > > multiple methods can simple define the CONFIG_SPL_FOO for all of them. > > The goal of the first part of the series is to make the spl_mmc file > easier to maintain. I do not believe that the function stubs would add > much to the binary size, but if you're concerned about bloat, I can do > a buildman comparison.
Please do. We want this to be as small a change in size as possible (and may be why we need to look harder at LTO). -- Tom
signature.asc
Description: Digital signature
_______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
