From mboxrd@z Thu Jan  1 00:00:00 1970
From: Scott Wood <scottwood@freescale.com>
Date: Tue, 15 May 2012 15:32:19 -0500
Subject: [U-Boot] nand_get_device() and nand_release_device()
In-Reply-To: <4FA7F4FA.1080900@freescale.com>
References: <4FA7F4FA.1080900@freescale.com>
Message-ID: <4FB2BD53.90301@freescale.com>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

On 05/07/2012 11:14 AM, Timur Tabi wrote:
> Scott,
> 
> As you know, the P1022 has this weird LBC/DIU muxing problem.  I'd like to
> add support for the 'nand' command while the DIU is active.  Looking at
> the NAND code, I see functions nand_get_device() and
> nand_release_device().  What is the idea behind these functions? 

In U-Boot they don't do much.  In Linux they handle locking

> Would it
> make sense to do something like this:
> 
> 
> static int __board_start_nand(void)
> {
> 	return 0;
> }
> int board_start_nand(void) __attribute__((weak, alias("__board_start_nand")));
> 
> static void __board_finish_nand(void)
> {
> }
> void board_finish_nand(void) __attribute__((weak,
> alias("__board_finish_nand")));
> 
> 
> static int nand_get_device(struct nand_chip *chip, struct mtd_info *mtd,
> int new_state)
> {
> 	chip->state = new_state;
> 
> 	return board_start_nand();
> }
> 
> static void nand_release_device(struct mtd_info *mtd)
> {
> 	struct nand_chip *chip = mtd->priv;
> 
> 	board_finish_nand();
> 
> 	/* De-select the NAND device */
> 	chip->select_chip(mtd, -1);
> }
> 

What's wrong with the existing select_chip interface?

And if something new is needed, please use the function pointer
interface rather than weak symbols, and do it first in Linux.

-Scott