libata: Cavium OCTEON SOC Compact Flash driver (v5)

As part of our efforts to get the Cavium OCTEON processor support merged (see: http://marc.info/?l=linux-mips&m=122834773330212), we have this CF driver for your consideration.

This fifth version is split into three separate patches. The first adds some entries to the ata_timing table in libata-core.c. The Second adds some special handling to ata_pio_need_iordy() for Compact Flash devices, The final patch is is the driver proper.

I will reply with the three patches.

David Daney (3):
libata: Add another column to the ata_timing table.
libata: Add special ata_pio_need_iordy() handling for Compact Flash.
libata: New driver for OCTEON SOC Compact Flash interface (v5).

 drivers/ata/Kconfig          |    9 +
 drivers/ata/Makefile         |    1 +
 drivers/ata/libata-core.c    |   76 ++--
 drivers/ata/pata_octeon_cf.c |  956 
++++++++++++++++++++++++++++++++++++++++++
 include/linux/libata.h       |   10 +-
 5 files changed, 1014 insertions(+), 38 deletions(-)
 create mode 100644 drivers/ata/pata_octeon_cf.c

The forthcoming OCTEON SOC Compact Flash driver needs an additional timing value that was not available in the ata_timing table. I add a new column for dmack_hold time. The values were obtained from the Compact Flash specification Rev 4.1.

---
 drivers/ata/libata-core.c |   72 +++++++++++++++++++++++----------------------
 include/linux/libata.h    |   10 ++++--
 2 files changed, 44 insertions(+), 38 deletions(-)

index 5e2eb74..1be2dde 100644
@@ -2953,33 +2953,33 @@ int sata_set_spd(struct ata_link *link)
  */
 
 static const struct ata_timing ata_timing[] = {
-/*	{ XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 960,   0 }, */
-	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 600,   0 },
-	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 383,   0 },
-	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 240,   0 },
-	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 180,   0 },
-	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 120,   0 },
-	{ XFER_PIO_5,     15,  65,  25, 100,  65,  25, 100,   0 },
-	{ XFER_PIO_6,     10,  55,  20,  80,  55,  20,  80,   0 },
-
-	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 960,   0 },
-	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 480,   0 },
-	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 240,   0 },
-
-	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 480,   0 },
-	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 150,   0 },
-	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 120,   0 },
-	{ XFER_MW_DMA_3,  25,   0,   0,   0,  65,  25, 100,   0 },
-	{ XFER_MW_DMA_4,  25,   0,   0,   0,  55,  20,  80,   0 },
-
-/*	{ XFER_UDMA_SLOW,  0,   0,   0,   0,   0,   0,   0, 150 }, */
-	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0,   0, 120 },
-	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0,   0,  80 },
-	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0,   0,  60 },
-	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0,   0,  45 },
-	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0,   0,  30 },
-	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0,   0,  20 },
-	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0,   0,  15 },
+/*	{ XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 0,  960,   0 }, */
+	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 0,  600,   0 },
+	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 0,  383,   0 },
+	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 0,  240,   0 },
+	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 0,  180,   0 },
+	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 0,  120,   0 },
+	{ XFER_PIO_5,     15,  65,  25, 100,  65,  25, 0,  100,   0 },
+	{ XFER_PIO_6,     10,  55,  20,  80,  55,  20, 0,   80,   0 },
+
+	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 50, 960,   0 },
+	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 30, 480,   0 },
+	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 20, 240,   0 },
+
+	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 20, 480,   0 },
+	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 5,  150,   0 },
+	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 5,  120,   0 },
+	{ XFER_MW_DMA_3,  25,   0,   0,   0,  65,  25, 5,  100,   0 },
+	{ XFER_MW_DMA_4,  25,   0,   0,   0,  55,  20, 5,   80,   0 },
+
+/*	{ XFER_UDMA_SLOW,  0,   0,   0,   0,   0,   0, 0,    0, 150 }, */
+	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0, 0,    0, 120 },
+	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0, 0,    0,  80 },
+	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0, 0,    0,  60 },
+	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0, 0,    0,  45 },
+	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0, 0,    0,  30 },
+	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0, 0,    0,  20 },
+	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0, 0,    0,  15 },
 
 	{ 0xFF }
 };

@@ -2989,14 +2989,15 @@ static const struct ata_timing ata_timing[] = {
 
 static void ata_timing_quantize(const struct ata_timing *t, struct ata_timing *q, int T, int UT)
 {
-	q->setup   = EZ(t->setup   * 1000,  T);
-	q->act8b   = EZ(t->act8b   * 1000,  T);
-	q->rec8b   = EZ(t->rec8b   * 1000,  T);
-	q->cyc8b   = EZ(t->cyc8b   * 1000,  T);
-	q->active  = EZ(t->active  * 1000,  T);
-	q->recover = EZ(t->recover * 1000,  T);
-	q->cycle   = EZ(t->cycle   * 1000,  T);
-	q->udma    = EZ(t->udma    * 1000, UT);
+	q->setup	= EZ(t->setup      * 1000,  T);
+	q->act8b	= EZ(t->act8b      * 1000,  T);
+	q->rec8b	= EZ(t->rec8b      * 1000,  T);
+	q->cyc8b	= EZ(t->cyc8b      * 1000,  T);
+	q->active	= EZ(t->active     * 1000,  T);
+	q->recover	= EZ(t->recover    * 1000,  T);
+	q->dmack_hold	= EZ(t->dmack_hold * 1000,  T);
+	q->cycle	= EZ(t->cycle      * 1000,  T);
+	q->udma		= EZ(t->udma       * 1000, UT);
 }
 
 void ata_timing_merge(const struct ata_timing *a, const struct ata_timing *b,

@@ -3008,6 +3009,7 @@ void ata_timing_merge(const struct ata_timing *a, const struct ata_timing *b,
 	if (what & ATA_TIMING_CYC8B  ) m->cyc8b   = max(a->cyc8b,   b->cyc8b);
 	if (what & ATA_TIMING_ACTIVE ) m->active  = max(a->active,  b->active);
 	if (what & ATA_TIMING_RECOVER) m->recover = max(a->recover, b->recover);
+	if (what & ATA_TIMING_DMACK_HOLD) m->dmack_hold = max(a->dmack_hold, b->dmack_hold);
 	if (what & ATA_TIMING_CYCLE  ) m->cycle   = max(a->cycle,   b->cycle);
 	if (what & ATA_TIMING_UDMA   ) m->udma    = max(a->udma,    b->udma);
 }

index ed3f26e..3c872ae 100644
@@ -399,12 +399,14 @@ enum {
 				  ATA_TIMING_CYC8B,
 	ATA_TIMING_ACTIVE	= (1 << 4),
 	ATA_TIMING_RECOVER	= (1 << 5),
-	ATA_TIMING_CYCLE	= (1 << 6),
-	ATA_TIMING_UDMA		= (1 << 7),
+	ATA_TIMING_DMACK_HOLD	= (1 << 6),
+	ATA_TIMING_CYCLE	= (1 << 7),
+	ATA_TIMING_UDMA		= (1 << 8),
 	ATA_TIMING_ALL		= ATA_TIMING_SETUP | ATA_TIMING_ACT8B |
 				  ATA_TIMING_REC8B | ATA_TIMING_CYC8B |
 				  ATA_TIMING_ACTIVE | ATA_TIMING_RECOVER |
-				  ATA_TIMING_CYCLE | ATA_TIMING_UDMA,
+				  ATA_TIMING_DMACK_HOLD | ATA_TIMING_CYCLE |
+				  ATA_TIMING_UDMA,
 };
 
 enum ata_xfer_mask {

@@ -864,6 +866,8 @@ struct ata_timing {
 	unsigned short cyc8b;		/* t0 for 8-bit I/O */
 	unsigned short active;		/* t2 or tD */
 	unsigned short recover;		/* t2i or tK */
+	unsigned short write_hold;	/* t4 */
+	unsigned short dmack_hold;	/* tj */
 	unsigned short cycle;		/* t0 */
 	unsigned short udma;		/* t2CYCTYP/2 */
 };

1.5.6.5

Hello.

Note that I wasn't telling you to drop t4, just to correct its value
for SWDMA modes.

NAK, patch broken.

[...]

I decided I don't need it, so I tried to drop it...

[...]

Sorry about that. I am testing a fixed version.

David Daney

The forthcoming OCTEON SOC Compact Flash driver needs an additional timing value that was not available in the ata_timing table. I add a new column for dmack_hold time. The values were obtained from the Compact Flash specification Rev 4.1.

---

This is a small correction. In the first version I forgot to completely remove a field from the ata_timing struct that was present in an earlier version of this patch set. This one should be good.

drivers/ata/libata-core.c | 72 +++++++++++++++++++++++----------------------
include/linux/libata.h | 9 ++++--

2 files changed, 43 insertions(+), 38 deletions(-)

index 5e2eb74..1be2dde 100644
@@ -2953,33 +2953,33 @@ int sata_set_spd(struct ata_link *link)
 */
 

static const struct ata_timing ata_timing[] = {
-/* { XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 960, 0 }, */

-	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 600,   0 },
-	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 383,   0 },
-	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 240,   0 },
-	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 180,   0 },
-	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 120,   0 },
-	{ XFER_PIO_5,     15,  65,  25, 100,  65,  25, 100,   0 },
-	{ XFER_PIO_6,     10,  55,  20,  80,  55,  20,  80,   0 },
-
-	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 960,   0 },
-	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 480,   0 },
-	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 240,   0 },
-
-	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 480,   0 },
-	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 150,   0 },
-	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 120,   0 },
-	{ XFER_MW_DMA_3,  25,   0,   0,   0,  65,  25, 100,   0 },
-	{ XFER_MW_DMA_4,  25,   0,   0,   0,  55,  20,  80,   0 },
-
-/*	{ XFER_UDMA_SLOW,  0,   0,   0,   0,   0,   0,   0, 150 }, */
-	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0,   0, 120 },
-	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0,   0,  80 },
-	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0,   0,  60 },
-	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0,   0,  45 },
-	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0,   0,  30 },
-	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0,   0,  20 },
-	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0,   0,  15 },
+/*	{ XFER_PIO_SLOW, 120, 290, 240, 960, 290, 240, 0,  960,   0 }, */
+	{ XFER_PIO_0,     70, 290, 240, 600, 165, 150, 0,  600,   0 },
+	{ XFER_PIO_1,     50, 290,  93, 383, 125, 100, 0,  383,   0 },
+	{ XFER_PIO_2,     30, 290,  40, 330, 100,  90, 0,  240,   0 },
+	{ XFER_PIO_3,     30,  80,  70, 180,  80,  70, 0,  180,   0 },
+	{ XFER_PIO_4,     25,  70,  25, 120,  70,  25, 0,  120,   0 },
+	{ XFER_PIO_5,     15,  65,  25, 100,  65,  25, 0,  100,   0 },
+	{ XFER_PIO_6,     10,  55,  20,  80,  55,  20, 0,   80,   0 },
+
+	{ XFER_SW_DMA_0, 120,   0,   0,   0, 480, 480, 50, 960,   0 },
+	{ XFER_SW_DMA_1,  90,   0,   0,   0, 240, 240, 30, 480,   0 },
+	{ XFER_SW_DMA_2,  60,   0,   0,   0, 120, 120, 20, 240,   0 },
+
+	{ XFER_MW_DMA_0,  60,   0,   0,   0, 215, 215, 20, 480,   0 },
+	{ XFER_MW_DMA_1,  45,   0,   0,   0,  80,  50, 5,  150,   0 },
+	{ XFER_MW_DMA_2,  25,   0,   0,   0,  70,  25, 5,  120,   0 },
+	{ XFER_MW_DMA_3,  25,   0,   0,   0,  65,  25, 5,  100,   0 },
+	{ XFER_MW_DMA_4,  25,   0,   0,   0,  55,  20, 5,   80,   0 },
+
+/*	{ XFER_UDMA_SLOW,  0,   0,   0,   0,   0,   0, 0,    0, 150 }, */
+	{ XFER_UDMA_0,     0,   0,   0,   0,   0,   0, 0,    0, 120 },
+	{ XFER_UDMA_1,     0,   0,   0,   0,   0,   0, 0,    0,  80 },
+	{ XFER_UDMA_2,     0,   0,   0,   0,   0,   0, 0,    0,  60 },
+	{ XFER_UDMA_3,     0,   0,   0,   0,   0,   0, 0,    0,  45 },
+	{ XFER_UDMA_4,     0,   0,   0,   0,   0,   0, 0,    0,  30 },
+	{ XFER_UDMA_5,     0,   0,   0,   0,   0,   0, 0,    0,  20 },
+	{ XFER_UDMA_6,     0,   0,   0,   0,   0,   0, 0,    0,  15 },

	{ 0xFF }
};

@@ -2989,14 +2989,15 @@ static const struct ata_timing ata_timing[] = {

static void ata_timing_quantize(const struct ata_timing *t, struct ata_timing *q, int T, int UT) {

-	q->setup   = EZ(t->setup   * 1000,  T);
-	q->act8b   = EZ(t->act8b   * 1000,  T);
-	q->rec8b   = EZ(t->rec8b   * 1000,  T);
-	q->cyc8b   = EZ(t->cyc8b   * 1000,  T);
-	q->active  = EZ(t->active  * 1000,  T);
-	q->recover = EZ(t->recover * 1000,  T);
-	q->cycle   = EZ(t->cycle   * 1000,  T);
-	q->udma    = EZ(t->udma    * 1000, UT);
+	q->setup	= EZ(t->setup      * 1000,  T);
+	q->act8b	= EZ(t->act8b      * 1000,  T);
+	q->rec8b	= EZ(t->rec8b      * 1000,  T);
+	q->cyc8b	= EZ(t->cyc8b      * 1000,  T);

+ q->active = EZ(t->active * 1000, T);
+ q->recover = EZ(t->recover * 1000, T);
+ q->dmack_hold = EZ(t->dmack_hold * 1000, T);
+ q->cycle = EZ(t->cycle * 1000, T);
+ q->udma = EZ(t->udma * 1000, UT);
}

void ata_timing_merge(const struct ata_timing *a, const struct ata_timing *b,
@@ -3008,6 +3009,7 @@ void ata_timing_merge(const struct ata_timing *a, const struct ata_timing *b,

	if (what & ATA_TIMING_CYC8B  ) m->cyc8b   = max(a->cyc8b,   b->cyc8b);
	if (what & ATA_TIMING_ACTIVE ) m->active  = max(a->active,  b->active);
	if (what & ATA_TIMING_RECOVER) m->recover = max(a->recover, b->recover);
+	if (what & ATA_TIMING_DMACK_HOLD) m->dmack_hold = max(a->dmack_hold, b->dmack_hold);
	if (what & ATA_TIMING_CYCLE  ) m->cycle   = max(a->cycle,   b->cycle);
	if (what & ATA_TIMING_UDMA   ) m->udma    = max(a->udma,    b->udma);
}

index ed3f26e..cf24a2a 100644
@@ -399,12 +399,14 @@ enum {

				  ATA_TIMING_CYC8B,
	ATA_TIMING_ACTIVE	= (1 << 4),
	ATA_TIMING_RECOVER	= (1 << 5),

- ATA_TIMING_CYCLE = (1 << 6),
- ATA_TIMING_UDMA = (1 << 7),
+ ATA_TIMING_DMACK_HOLD = (1 << 6),
+ ATA_TIMING_CYCLE = (1 << 7),
+ ATA_TIMING_UDMA = (1 << 8),
ATA_TIMING_ALL = ATA_TIMING_SETUP | ATA_TIMING_ACT8B |
ATA_TIMING_REC8B | ATA_TIMING_CYC8B | ATA_TIMING_ACTIVE | ATA_TIMING_RECOVER |

-				  ATA_TIMING_CYCLE | ATA_TIMING_UDMA,
+				  ATA_TIMING_DMACK_HOLD | ATA_TIMING_CYCLE |
+				  ATA_TIMING_UDMA,

};

enum ata_xfer_mask {
@@ -864,6 +866,7 @@ struct ata_timing {
unsigned short cyc8b; /* t0 for 8-bit I/O */
unsigned short active; /* t2 or tD */
unsigned short recover; /* t2i or tK */
+ unsigned short dmack_hold; /* tj */

	unsigned short cycle;		/* t0 */
	unsigned short udma;		/* t2CYCTYP/2 */
};

According to the Compact Flash specification r4.1, PIO modes 5 and 6 do not use iordy.

---
 drivers/ata/libata-core.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

index 1be2dde..b4a35d8 100644
@@ -1923,6 +1923,10 @@ unsigned int ata_pio_need_iordy(const struct ata_device *adev)
 	   as the caller should know this */
 	if (adev->link->ap->flags & ATA_FLAG_NO_IORDY)
 		return 0;
+	/* CF spec. r4.1 Table 22 says no iordy on PIO5 and PIO6.  */
+	if (ata_id_is_cfa(adev->id)
+	    && (adev->pio_mode == XFER_PIO_5 || adev->pio_mode == XFER_PIO_6))
+		return 0;
 	/* PIO3 and higher it is mandatory */
 	if (adev->pio_mode > XFER_PIO_2)
 		return 1;

1.5.6.5

As part of our efforts to get the Cavium OCTEON processor support merged (see: http://marc.info/?l=linux-mips&m=122834773330212), we have this CF driver for your consideration.

Most OCTEON variants have *no* DMA or interrupt support on the CF interface so for these, only PIO is supported. Although if DMA is available, we do take advantage of it.

The register definitions are part of the chip support patch set mentioned above, and are not included here.

In this fifth version, I have tried to address Sergei Shtylyov's comments regarding the various timing parameters. Several timing parameters were adjusted based on testing with a logic analyzer.

The status polling tasklet was changed to a workqueue as the polling with a tasklet was using 100% of one CPU for extended periods.

Thanks,

---
 drivers/ata/Kconfig          |    9 +
 drivers/ata/Makefile         |    1 +
 drivers/ata/pata_octeon_cf.c |  956 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 966 insertions(+), 0 deletions(-)
 create mode 100644 drivers/ata/pata_octeon_cf.c

index 78fbec8..b59904b 100644
@@ -697,6 +697,15 @@ config PATA_IXP4XX_CF
 
 	  If unsure, say N.
 
+config PATA_OCTEON_CF
+	tristate "OCTEON Boot Bus Compact Flash support"
+	depends on CPU_CAVIUM_OCTEON
+	help
+	  This option enables a polled compact flash driver for use with
+	  compact flash cards attached to the OCTEON boot bus.
+
+	  If unsure, say N.
+
 config PATA_SCC
 	tristate "Toshiba's Cell Reference Set IDE support"
 	depends on PCI && PPC_CELLEB

index 674965f..7f1ecf9 100644
@@ -69,6 +69,7 @@ obj-$(CONFIG_PATA_IXP4XX_CF)	+= pata_ixp4xx_cf.o
 obj-$(CONFIG_PATA_SCC)		+= pata_scc.o
 obj-$(CONFIG_PATA_SCH)		+= pata_sch.o
 obj-$(CONFIG_PATA_BF54X)	+= pata_bf54x.o
+obj-$(CONFIG_PATA_OCTEON_CF)	+= pata_octeon_cf.o
 obj-$(CONFIG_PATA_PLATFORM)	+= pata_platform.o
 obj-$(CONFIG_PATA_OF_PLATFORM)	+= pata_of_platform.o
 obj-$(CONFIG_PATA_ICSIDE)	+= pata_icside.o

1.5.6.5