Re: [PULL] mtd: nand: material for 4.9

From: Brian Norris
Date: Mon Oct 03 2016 - 12:55:52 EST

Hi Boris,

On Sun, Sep 25, 2016 at 08:53:26PM +0200, Boris Brezillon wrote:
> Hi Brian,
> Here is my PR for 4.9. Let me know if you see any problem.
> Also, please let us know ahead of time if you can't take this PR (I know you
> are busy lately, and I guess I could send this PR directly to Linus or ask
> Richard to take it in its UBIFS/UBI tree).

I made some time for this one. We'll see about the pairing one
(hopefully trivial? but I'm not sure if there's value in merging without

> Thanks,
> Boris
> The following changes since commit 29b4817d4018df78086157ea3a55c1d9424a7cfc:
> Linux 4.8-rc1 (2016-08-07 18:18:00 -0700)
> are available in the git repository at:
> git@xxxxxxxxxx:linux-nand/linux.git tags/for-4.9
> for you to fetch changes up to d44154f969a44269a9288c274c1c2fd9e85df8a5:
> mtd: nand: Provide nand_cleanup() function to free NAND related resources (2016-09-23 09:35:16 +0200)

Thanks. I only added one trivial fixup that's spelling.txt
(what??) caught:

diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
index 7a00245ebada..e5718e5ecf92 100644
--- a/drivers/mtd/nand/nand_base.c
+++ b/drivers/mtd/nand/nand_base.c
@@ -746,7 +746,7 @@ static void nand_command_lp(struct mtd_info *mtd, unsigned int command,
chip->cmd_ctrl(mtd, column, ctrl);

- /* Only ouput a single addr cycle for 8bits opcodes. */
+ /* Only output a single addr cycle for 8bits opcodes. */
if (!nand_opcode_8bits(command))
chip->cmd_ctrl(mtd, column >> 8, ctrl);

> ----------------------------------------------------------------
> Notable core changes:
> - add the infrastructure to automate NAND timings configuration
> - provide a generic DT property to maximize ECC strength
> The rest is just a bunch of minor drivers and core fixes/cleanup
> patches.
> ----------------------------------------------------------------

I added a small note in the merge commit about the BBT changes you made.
That's an especially fragile piece of code I think (and used by every
NAND driver), but it looked OK to me.