Hi Han,

Thanks for your suggestions about this patch. I will update it in driver next version patch set. 

Do you have some other comments for the patch set except for removing  "+ ret = of_property_read_bool(np, "m25p,fast-read");" in driver ?


Thanks
Yunhui
-----Original Message-----
From: Han Xu [mailto:xhnjupt at gmail.com] 
Sent: Tuesday, March 01, 2016 4:17 AM
To: Yunhui Cui
Cc: Yunhui Cui; dwmw2 at infradead.org; computersforpeace at gmail.com; han.xu at freescale.com; linux-mtd at lists.infradead.org; linux-kernel at vger.kernel.org; linux-arm-kernel at lists.infradead.org; Yao Yuan
Subject: Re: [PATCH v2 3/4] mtd:spi-nor:fsl-quadspi:Add fast-read mode support

On Mon, Feb 29, 2016 at 03:43:36AM +0000, Yunhui Cui wrote:

Hi Han,

But I don't think QuadSPI driver need to check the m25p,fast-read property again since spi-nor layer has already done that. Adding the property in flash node should work in the same way.

[Yunhui]: There are three modes in fsl-quadspi driver , fast mode, quad mode, ddr quad read. The last parameter mode of spi_nor_scan() I have to specify . Otherwise, flash is still set to quad mode.

If the Nor chip behaves as you said only support fasst-read, the (info->flags &
SPI_NOR_QUAD_READ) can only take false path.

spi-nor.c: 

1419         if (mode == SPI_NOR_QUAD && info->flags & SPI_NOR_QUAD_READ) {
1420                 ret = set_quad_mode(nor, info);
1421                 if (ret) {
1422                         dev_err(dev, "quad mode not supported\n");
1423                         return ret;
1424                 }
1425                 nor->flash_read = SPI_NOR_QUAD;
1426         } else if (mode == SPI_NOR_DUAL && info->flags & SPI_NOR_DUAL_READ) {
1427                 nor->flash_read = SPI_NOR_DUAL;
1428         }


Thanks
Yunhui
-----Original Message-----
From: Han Xu [mailto:xhnjupt at gmail.com]
Sent: Saturday, February 27, 2016 12:32 AM
To: Yunhui Cui
Cc: Yunhui Cui; dwmw2 at infradead.org; computersforpeace at gmail.com; 
han.xu at freescale.com; linux-mtd at lists.infradead.org; 
linux-kernel at vger.kernel.org; linux-arm-kernel at lists.infradead.org; 
Yao Yuan
Subject: Re: [PATCH v2 3/4] mtd:spi-nor:fsl-quadspi:Add fast-read mode 
support

On Thu, Feb 25, 2016 at 08:07:22AM +0000, Yunhui Cui wrote:

quoted

Hi Han,

I have provided the options " m25p,fast-read ", because there are probable some flashes can't support quad mode.
So we should support fast-read mode in our driver. Moreover,  There is a option to select fast-read mode in spi_nor.c :
               /* If we were instantiated by DT, use it */
                 if (of_property_read_bool(np, "m25p,fast-read"))
                         nor->flash_read = SPI_NOR_FAST;

Did you have some REAL cases using SPI NOR that only supports upto fast-read with Quad SPI driver? Neither fast-read or normal-read, which is actually more general, supported in the driver, just because I didn't see any REAL cases till now.

I didn't run against the patch, although IMO it's not that necessary. But I don't think QuadSPI driver need to check the m25p,fast-read property again since spi-nor layer has already done that. Adding the property in flash node should work in the same way.
 

quoted

Thanks
Yunhui

-----Original Message-----
From: Han Xu [mailto:xhnjupt at gmail.com]
Sent: Thursday, February 18, 2016 2:08 AM
To: Yunhui Cui
Cc: dwmw2 at infradead.org; computersforpeace at gmail.com; 
han.xu at freescale.com; linux-mtd at lists.infradead.org; 
linux-kernel at vger.kernel.org; linux-arm-kernel at lists.infradead.org;
Yao Yuan
Subject: Re: [PATCH v2 3/4] mtd:spi-nor:fsl-quadspi:Add fast-read 
mode support

On Mon, Feb 01, 2016 at 07:30:07PM +0800, Yunhui Cui wrote:

quoted

The qspi driver add generic fast-read mode for different flash 
venders. There are some different board flash work on different 
mode, such fast-read, quad-mode.
So we have to modify the third entrace parameter of spi_nor_scan().

Signed-off-by: Yunhui Cui <redacted>
---
 drivers/mtd/spi-nor/fsl-quadspi.c | 27 
+++++++++++++++++++++------
 1 file changed, 21 insertions(+), 6 deletions(-)

diff --git a/drivers/mtd/spi-nor/fsl-quadspi.c

b/drivers/mtd/spi-nor/fsl-quadspi.c
index 9861290..0a31cb1 100644

--- a/drivers/mtd/spi-nor/fsl-quadspi.c
+++ b/drivers/mtd/spi-nor/fsl-quadspi.c

@@ -389,11 +389,21 @@ static void fsl_qspi_init_lut(struct fsl_qspi *q)
 	/* Read */
 	lut_base = SEQID_READ * 4;
 
-	qspi_writel(q, LUT0(CMD, PAD1, read_op) | LUT1(ADDR, PAD1, addrlen),
-			base + QUADSPI_LUT(lut_base));
-	qspi_writel(q, LUT0(DUMMY, PAD1, read_dm) |
-		    LUT1(FSL_READ, PAD4, rxfifo),
-			base + QUADSPI_LUT(lut_base + 1));
+	if (nor->flash_read == SPI_NOR_FAST) {
+		qspi_writel(q, LUT0(CMD, PAD1, read_op) |
+			    LUT1(ADDR, PAD1, addrlen),
+				base + QUADSPI_LUT(lut_base));
+		qspi_writel(q,  LUT0(DUMMY, PAD1, read_dm) |
+			    LUT1(FSL_READ, PAD1, rxfifo),
+				base + QUADSPI_LUT(lut_base + 1));
+	} else if (nor->flash_read == SPI_NOR_QUAD) {
+		qspi_writel(q, LUT0(CMD, PAD1, read_op) |
+			    LUT1(ADDR, PAD1, addrlen),
+				base + QUADSPI_LUT(lut_base));
+		qspi_writel(q, LUT0(DUMMY, PAD1, read_dm) |
+			    LUT1(FSL_READ, PAD4, rxfifo),
+				base + QUADSPI_LUT(lut_base + 1));
+	}
 
 	/* Write enable */
 	lut_base = SEQID_WREN * 4;

@@ -468,6 +478,7 @@ static int fsl_qspi_get_seqid(struct fsl_qspi 

*q,
u8 cmd)  {
 	switch (cmd) {
 	case SPINOR_OP_READ_1_1_4:
+	case SPINOR_OP_READ_FAST:
 		return SEQID_READ;
 	case SPINOR_OP_WREN:
 		return SEQID_WREN;

@@ -963,6 +974,7 @@ static int fsl_qspi_probe(struct platform_device *pdev)
 	struct spi_nor *nor;
 	struct mtd_info *mtd;
 	int ret, i = 0;
+	enum read_mode mode = SPI_NOR_QUAD;
 
 	q = devm_kzalloc(dev, sizeof(*q), GFP_KERNEL);
 	if (!q)

@@ -1064,7 +1076,10 @@ static int fsl_qspi_probe(struct platform_device *pdev)
 		/* set the chip address for READID */
 		fsl_qspi_set_base_addr(q, nor);
 
-		ret = spi_nor_scan(nor, NULL, SPI_NOR_QUAD);
+		ret = of_property_read_bool(np, "m25p,fast-read");
+		mode = (ret) ? SPI_NOR_FAST : SPI_NOR_QUAD;
+
+		ret = spi_nor_scan(nor, NULL, mode);
 		if (ret)
 			goto mutex_failed;
 

I understand you want to provide a more generic read mode, but this is a quad spi driver, I prefer to provide the options, either quad read/quad IO read, or just normal read.

quoted

--
2.1.0.27.g96db324


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/