Message ID | 50867DA1.5070801@gmail.com |
---|---|
State | Rejected |
Delegated to: | Tom Rini |
Headers | show |
On Tue, Oct 23, 2012 at 04:51:05PM +0530, Vikram Narayanan wrote: > Split up the selection of nand_chip from nand_init to a separate > function. This is just to make it a little consistent as we've > nand_deselect function. > > Also use it in spl_nand_load_image function. > > Signed-off-by: Vikram Narayanan <vikram186@gmail.com> > Cc: Stefan Roese <sr@denx.de> What does this allow for us to do next? We've added a few bytes to SPL and we're already close to being too large sometimes.
On 10/23/2012 9:17 PM, Tom Rini wrote: > On Tue, Oct 23, 2012 at 04:51:05PM +0530, Vikram Narayanan wrote: > >> Split up the selection of nand_chip from nand_init to a separate >> function. This is just to make it a little consistent as we've >> nand_deselect function. >> >> Also use it in spl_nand_load_image function. >> >> Signed-off-by: Vikram Narayanan<vikram186@gmail.com> >> Cc: Stefan Roese<sr@denx.de> > > What does this allow for us to do next? We've added a few bytes to SPL > and we're already close to being too large sometimes. > Yep. 12 bytes added. If SPL is too much loaded to include this, the patch shall be dropped. ~Vikram
On Tue, Oct 23, 2012 at 11:01:31PM +0530, Vikram Narayanan wrote: > On 10/23/2012 9:17 PM, Tom Rini wrote: > >On Tue, Oct 23, 2012 at 04:51:05PM +0530, Vikram Narayanan wrote: > > > >>Split up the selection of nand_chip from nand_init to a separate > >>function. This is just to make it a little consistent as we've > >>nand_deselect function. > >> > >>Also use it in spl_nand_load_image function. > >> > >>Signed-off-by: Vikram Narayanan<vikram186@gmail.com> > >>Cc: Stefan Roese<sr@denx.de> > > > >What does this allow for us to do next? We've added a few bytes to SPL > >and we're already close to being too large sometimes. > > Yep. 12 bytes added. If SPL is too much loaded to include this, the > patch shall be dropped. Since we're trying to squish down to 4kb and every byte counts, yes, this might well be. Thanks.
diff --git a/common/spl/spl_nand.c b/common/spl/spl_nand.c index 61de5a4..5ebeb38 100644 --- a/common/spl/spl_nand.c +++ b/common/spl/spl_nand.c @@ -34,6 +34,7 @@ void spl_nand_load_image(void) debug("spl: nand - using hw ecc\n"); nand_init(); + nand_select(); /*use CONFIG_SYS_TEXT_BASE as temporary storage area */ header = (struct image_header *)(CONFIG_SYS_TEXT_BASE); diff --git a/drivers/mtd/nand/nand_spl_simple.c b/drivers/mtd/nand/nand_spl_simple.c index 4a4d02f..77c56fc 100644 --- a/drivers/mtd/nand/nand_spl_simple.c +++ b/drivers/mtd/nand/nand_spl_simple.c @@ -270,12 +270,16 @@ void nand_init(void) nand_chip.ecc.correct = nand_correct_data; } #endif +} +/* Select before operation */ +void nand_select(void) +{ if (nand_chip.select_chip) nand_chip.select_chip(&mtd, 0); } -/* Unselect after operation */ +/* Deselect after operation */ void nand_deselect(void) { if (nand_chip.select_chip)
Split up the selection of nand_chip from nand_init to a separate function. This is just to make it a little consistent as we've nand_deselect function. Also use it in spl_nand_load_image function. Signed-off-by: Vikram Narayanan <vikram186@gmail.com> Cc: Stefan Roese <sr@denx.de> --- common/spl/spl_nand.c | 1 + drivers/mtd/nand/nand_spl_simple.c | 6 +++++- 2 files changed, 6 insertions(+), 1 deletions(-)