Hi Boris,

Boris Brezillon boris.brezillon@collabora.com wrote on Mon, 8 Apr 2019 10:07:47 +0200:

On Sun, 7 Apr 2019 21:34:01 +0200 Miquel Raynal miquel.raynal@bootlin.com wrote:

...

Since the migration of the driver to stop using the legacy ->select_chip() hook, there is nothing deselecting the target anymore, thus the selection is not forced at the next access. Ensure the ND_RUN bit is always in a clean state.

Cc: Daniel Mack daniel@zonque.org Cc: stable@vger.kernel.org Fixes: b25251414f6e00 ("mtd: rawnand: marvell: Stop implementing ->select_chip()") Suggested-by: Boris Brezillon boris.brezillon@collabora.com Signed-off-by: Miquel Raynal miquel.raynal@bootlin.com

---

Hi Daniel, could you please test this version of the fix?

Thanks, Miquèl

drivers/mtd/nand/raw/marvell_nand.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/mtd/nand/raw/marvell_nand.c b/drivers/mtd/nand/raw/marvell_nand.c index f38e5c1b87e4..a7d7f17d8682 100644 --- a/drivers/mtd/nand/raw/marvell_nand.c +++ b/drivers/mtd/nand/raw/marvell_nand.c @@ -722,12 +722,6 @@ static void marvell_nfc_select_target(struct nand_chip *chip, struct marvell_nfc *nfc = to_marvell_nfc(chip->controller); u32 ndcr_generic;

• if (chip == nfc->selected_chip && die_nr == marvell_nand->selected_die)

• return;
  

• writel_relaxed(marvell_nand->ndtr0, nfc->regs + NDTR0);
• writel_relaxed(marvell_nand->ndtr1, nfc->regs + NDTR1);
• /*
  • Reset the NDCR register to a clean state for this particular chip,
  • also clear ND_RUN bit.

@@ -736,6 +730,12 @@ static void marvell_nfc_select_target(struct nand_chip *chip, NDCR_GENERIC_FIELDS_MASK & ~NDCR_ND_RUN; writel_relaxed(ndcr_generic | marvell_nand->ndcr, nfc->regs + NDCR);

• if (chip == nfc->selected_chip && die_nr == marvell_nand->selected_die)

• return;
  

• writel_relaxed(marvell_nand->ndtr0, nfc->regs + NDTR0);
• writel_relaxed(marvell_nand->ndtr1, nfc->regs + NDTR1);
• /* Also reset the interrupt status register */ marvell_nfc_clear_int(nfc, NDCR_ALL_INT);

I think this should also be moved before the

if (chip == nfc->selected_chip && die_nr == marvell_nand->selected_die)

check.

Yes, I think it is not needed, I wanted Daniel to test. But maybe it is better to ensure interrupts are always cleared anyway. I'll send a v2 then.

Thanks, Miquèl