Skip to contentSkip to conversation

ide: add at91_ide driver

PATCH In review Stanislaw Gruszka 3 participants 10 messages Feb 18, 2009 – Mar 13, 2009

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

PATCH v3 2/3 ide: add at91_ide driver

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data Register. Moreover some extra things need to be done when setting PIO mode. Only PIO mode is used, hardware have no DMA support. If interrupt line is connected through GPIO extra quirk is needed to cope with fake interrupts.

---
 arch/arm/mach-at91/include/mach/board.h |    3 +
 drivers/ide/Kconfig                     |    5 +
 drivers/ide/Makefile                    |    1 +
 drivers/ide/at91_ide.c                  |  467 +++++++++++++++++++++++++++++++
 4 files changed, 476 insertions(+), 0 deletions(-)
 create mode 100644 drivers/ide/at91_ide.c
arch/arm/mach-at91/include/mach/board.h +3 -0
index fb51f0e..2bc35a8 100644
@@ -56,6 +56,9 @@ struct at91_cf_data {
 	u8	vcc_pin;		/* power switching */
 	u8	rst_pin;		/* card reset */
 	u8	chipselect;		/* EBI Chip Select number */
+	u8	flags;
+#define AT91_CF_TRUE_IDE	0x01
+#define AT91_IDE_SWAP_A0_A2	0x02
 };
 extern void __init at91_add_device_cf(struct at91_cf_data *data);
 
drivers/ide/Kconfig +5 -0
index 3dad229..df2817b 100644
@@ -721,6 +721,11 @@ config BLK_DEV_IDE_TX4939
 	depends on SOC_TX4939
 	select BLK_DEV_IDEDMA_SFF
 
+config BLK_DEV_IDE_AT91
+	tristate "Atmel AT91 (SAM9, CAP9, AT572D940HF) IDE support"
+	depends on ARM && ARCH_AT91 && !ARCH_AT91RM9200 && !ARCH_AT91X40
+	select IDE_TIMINGS
+
 config IDE_ARM
 	tristate "ARM IDE support"
 	depends on ARM && (ARCH_RPC || ARCH_SHARK)
drivers/ide/Makefile +1 -0
index d0e3d7d..1c326d9 100644
@@ -116,3 +116,4 @@ obj-$(CONFIG_BLK_DEV_IDE_AU1XXX)	+= au1xxx-ide.o
 
 obj-$(CONFIG_BLK_DEV_IDE_TX4938)	+= tx4938ide.o
 obj-$(CONFIG_BLK_DEV_IDE_TX4939)	+= tx4939ide.o
+obj-$(CONFIG_BLK_DEV_IDE_AT91)		+= at91_ide.o
drivers/ide/at91_ide.c new file +467 -0

467 lines not on this page · raw message

Signature · 1 line
1.5.2.5
Acked-by on v3 2/3
#1 lore raw

Hello.

Quoted 10 lines · Stanislaw Gruszka

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data Register. Moreover some extra things need to be done when setting PIO mode. Only PIO mode is used, hardware have no DMA support. If interrupt line is connected through GPIO extra quirk is needed to cope with fake interrupts.

Quoted 6 lines
drivers/ide/at91_ide.c new file +0 -0
new file mode 100644
index 0000000..1bb50f4

[...]

drivers/ide/at91_ide.c line 74

Quoted 4 lines
drivers/ide/at91_ide.c new file +3 -0
+static void set_smc_timings(const u8 chipselect, const u16 cycle,
+			    const u16 setup, const u16 pulse,
+			    const u16 data_float, int use_iordy)

This bunch of *const* specifiers is pretty pointless...

drivers/ide/at91_ide.c line 199

Quoted 13 lines
drivers/ide/at91_ide.c new file +12 -0
+static void at91_ide_tf_load(ide_drive_t *drive, ide_task_t *task)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	struct ide_io_ports *io_ports = &hwif->io_ports;
+	struct ide_taskfile *tf = &task->tf;
+	u8 HIHI = (task->tf_flags & IDE_TFLAG_LBA48) ? 0xE0 : 0xEF;
+
+	if (task->tf_flags & IDE_TFLAG_FLAGGED)
+		HIHI = 0xFF;
+
+	if (task->tf_flags & IDE_TFLAG_OUT_DATA) {
+		u16 data = (tf->hob_data << 8) | tf->data;

It's not quite correct -- should be:

		u8 data[2] = { tf->data, tf->hob_data };

Although, if this is a little endian only code, it should be
equivivalent. I shouldn't probably have asked to change this part at all... :-/

drivers/ide/at91_ide.c line 237

Quoted 9 lines
drivers/ide/at91_ide.c new file +8 -0
+static void at91_ide_tf_read(ide_drive_t *drive, ide_task_t *task)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	struct ide_io_ports *io_ports = &hwif->io_ports;
+	struct ide_taskfile *tf = &task->tf;
+
+	if (task->tf_flags & IDE_TFLAG_IN_DATA) {
+		u16 data;

It would have been easier to just declare it as:

		u8 data[2];

drivers/ide/at91_ide.c line 241

Quoted 5 lines
drivers/ide/at91_ide.c new file +4 -0
+
+		at91_ide_input_data(drive, NULL, &data, 2);
+		tf->data = data & 0xff;
+		tf->hob_data = (data >> 8) & 0xff;
		tf->data = data[0];
		tf->hob_data = data[1];

MBR, Sergei

#2 lore raw
In reply to Stanislaw Gruszka
Quoted 9 lines · Stanislaw Gruszka

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data Register. Moreover some extra things need to be done when setting PIO mode. Only PIO mode is used, hardware have no DMA support. If interrupt line is connected through GPIO extra quirk is needed to cope with fake interrupts.

This is prolly only me, but I afil to compile this driver with the current
arch/arm/configs/at91sam9263ek_defconfig:

CC drivers/ide/at91_ide.o
drivers/ide/at91_ide.c: In function ‘at91_ide_input_data’:
drivers/ide/at91_ide.c:159: error: implicit declaration of function ‘__ide_mm_insw’ drivers/ide/at91_ide.c: In function ‘at91_ide_output_data’:
drivers/ide/at91_ide.c:174: error: implicit declaration of function ‘__ide_mm_outsw’ drivers/ide/at91_ide.c: In function ‘at91_ide_tf_load’:
drivers/ide/at91_ide.c:192: error: ‘task’ undeclared (first use in this function) drivers/ide/at91_ide.c:192: error: (Each undeclared identifier is reported only once
drivers/ide/at91_ide.c:192: error: for each function it appears in.) drivers/ide/at91_ide.c: At top level:
drivers/ide/at91_ide.c:303: warning: initialization from incompatible pointer type drivers/ide/at91_ide.c:304: warning: initialization from incompatible pointer type make[1]: *** [drivers/ide/at91_ide.o] Error 1
make: *** [drivers/ide/] Error 2

MBR, Sergei

#3 lore raw
Quoted 13 lines · Sergei Shtylyov
Quoted 9 lines

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data Register. Moreover some extra things need to be done when setting PIO mode. Only PIO mode is used, hardware have no DMA support. If interrupt line is connected through GPIO extra quirk is needed to cope with fake interrupts.

This is prolly only me, but I afil to compile this driver with the
current arch/arm/configs/at91sam9263ek_defconfig:

Quoted 15 lines

CC drivers/ide/at91_ide.o
drivers/ide/at91_ide.c: In function ‘at91_ide_input_data’:
drivers/ide/at91_ide.c:159: error: implicit declaration of function ‘__ide_mm_insw’ drivers/ide/at91_ide.c: In function ‘at91_ide_output_data’:
drivers/ide/at91_ide.c:174: error: implicit declaration of function ‘__ide_mm_outsw’ drivers/ide/at91_ide.c: In function ‘at91_ide_tf_load’:
drivers/ide/at91_ide.c:192: error: ‘task’ undeclared (first use in this function) drivers/ide/at91_ide.c:192: error: (Each undeclared identifier is reported only once
drivers/ide/at91_ide.c:192: error: for each function it appears in.) drivers/ide/at91_ide.c: At top level:
drivers/ide/at91_ide.c:303: warning: initialization from incompatible pointer type drivers/ide/at91_ide.c:304: warning: initialization from incompatible pointer type make[1]: *** [drivers/ide/at91_ide.o] Error 1
make: *** [drivers/ide/] Error 2

Ah, it's only with pata-2.6 series applied. Bart, you clearly need to
include this driver to some of your patches...

MBR, Sergei

#4 lore raw
Quoted 29 lines · I wrote:
Quoted 13 lines
Quoted 9 lines

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data Register. Moreover some extra things need to be done when setting PIO mode. Only PIO mode is used, hardware have no DMA support. If interrupt line is connected through GPIO extra quirk is needed to cope with fake interrupts.

This is prolly only me, but I afil to compile this driver with the
current arch/arm/configs/at91sam9263ek_defconfig:

Quoted 15 lines

CC drivers/ide/at91_ide.o
drivers/ide/at91_ide.c: In function ‘at91_ide_input_data’:
drivers/ide/at91_ide.c:159: error: implicit declaration of function ‘__ide_mm_insw’ drivers/ide/at91_ide.c: In function ‘at91_ide_output_data’:
drivers/ide/at91_ide.c:174: error: implicit declaration of function ‘__ide_mm_outsw’ drivers/ide/at91_ide.c: In function ‘at91_ide_tf_load’:
drivers/ide/at91_ide.c:192: error: ‘task’ undeclared (first use in this function) drivers/ide/at91_ide.c:192: error: (Each undeclared identifier is reported only once
drivers/ide/at91_ide.c:192: error: for each function it appears in.) drivers/ide/at91_ide.c: At top level:
drivers/ide/at91_ide.c:303: warning: initialization from incompatible pointer type drivers/ide/at91_ide.c:304: warning: initialization from incompatible pointer type make[1]: *** [drivers/ide/at91_ide.o] Error 1
make: *** [drivers/ide/] Error 2

Quoted 2 lines

Ah, it's only with pata-2.6 series applied. Bart, you clearly need to
include this driver to some of your patches...

... and ide-add-flagged_tf_flags-to-struct-ide_taskfile-v2.patch too.

MBR, Sergei

#5 lore raw
Quoted 29 lines · I wrote:
Quoted 28 lines
Quoted 28 lines
Quoted 9 lines

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data Register. Moreover some extra things need to be done when setting PIO mode. Only PIO mode is used, hardware have no DMA support. If interrupt line is connected through GPIO extra quirk is needed to cope with fake interrupts.

This is prolly only me, but I afil to compile this driver with the
current arch/arm/configs/at91sam9263ek_defconfig:

CC drivers/ide/at91_ide.o
drivers/ide/at91_ide.c: In function ‘at91_ide_input_data’:
drivers/ide/at91_ide.c:159: error: implicit declaration of function ‘__ide_mm_insw’ drivers/ide/at91_ide.c: In function ‘at91_ide_output_data’:
drivers/ide/at91_ide.c:174: error: implicit declaration of function ‘__ide_mm_outsw’ drivers/ide/at91_ide.c: In function ‘at91_ide_tf_load’:
drivers/ide/at91_ide.c:192: error: ‘task’ undeclared (first use in this function) drivers/ide/at91_ide.c:192: error: (Each undeclared identifier is reported only once
drivers/ide/at91_ide.c:192: error: for each function it appears in.) drivers/ide/at91_ide.c: At top level:
drivers/ide/at91_ide.c:303: warning: initialization from incompatible pointer type drivers/ide/at91_ide.c:304: warning: initialization from incompatible pointer type make[1]: *** [drivers/ide/at91_ide.o] Error 1
make: *** [drivers/ide/] Error 2

Quoted 2 lines
Quoted 2 lines

Ah, it's only with pata-2.6 series applied. Bart, you clearly need
to include this driver to some of your patches...

Oops, got muddled in too many follow-ups. :-) Namely, ide-remove-ide_task_t-typedef.patch is incomplete...

Quoted 1 line

... and ide-add-flagged_tf_flags-to-struct-ide_taskfile-v2.patch too.

MBR, Sergei

#6 lore raw
Quoted 42 lines · Sergei Shtylyov
Quoted 39 lines · I wrote:
Quoted 39 lines
Quoted 39 lines
Quoted 13 lines

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory
Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data
Register. Moreover some extra things need to be done when setting PIO mode.
Only PIO mode is used, hardware have no DMA support. If interrupt line is
connected through GPIO extra quirk is needed to cope with fake interrupts.

Quoted 1 line

This is prolly only me, but I afil to compile this driver with the
current arch/arm/configs/at91sam9263ek_defconfig:

CC drivers/ide/at91_ide.o
drivers/ide/at91_ide.c: In function ‘at91_ide_input_data’:
drivers/ide/at91_ide.c:159: error: implicit declaration of function ‘__ide_mm_insw’
drivers/ide/at91_ide.c: In function ‘at91_ide_output_data’:
drivers/ide/at91_ide.c:174: error: implicit declaration of function ‘__ide_mm_outsw’
drivers/ide/at91_ide.c: In function ‘at91_ide_tf_load’:
drivers/ide/at91_ide.c:192: error: ‘task’ undeclared (first use in this function)
drivers/ide/at91_ide.c:192: error: (Each undeclared identifier is reported only
once
drivers/ide/at91_ide.c:192: error: for each function it appears in.) drivers/ide/at91_ide.c: At top level:
drivers/ide/at91_ide.c:303: warning: initialization from incompatible pointer type
drivers/ide/at91_ide.c:304: warning: initialization from incompatible pointer type
make[1]: *** [drivers/ide/at91_ide.o] Error 1
make: *** [drivers/ide/] Error 2

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Ah, it's only with pata-2.6 series applied. Bart, you clearly need
to include this driver to some of your patches...

I did and asked people to test the final result (see my mail on merging at91_ide)... unfortunately it seems like I missed few spots... :)

Quoted 2 lines

Oops, got muddled in too many follow-ups. :-) Namely, ide-remove-ide_task_t-typedef.patch is incomplete...

Was a bit more than that, I also needed to merge __ide_mm_* fixup below...

[ Anyway please verify that I got it right this time, thanks! ]

From: Bartlomiej Zolnierkiewicz <[email protected]>
Subject: [PATCH] at91_ide: use readsw()/writesw() directly

Use readsw()/writesw() directly intead of __ide_mm_{in,out}sw() macros.

Cc Stanislaw Gruszka <[email protected]>
Cc Sergei Shtylyov <[email protected]>

---
needed for "ide: include <asm/ide.h> only when needed" patch in pata tree

 drivers/ide/at91_ide.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/ide/at91_ide.c +2 -2
Index: b/drivers/ide/at91_ide.c
===================================================================
@@ -156,7 +156,7 @@ static void at91_ide_input_data(ide_driv
 	len++;
 
 	enter_16bit(chipselect, mode);
-	__ide_mm_insw((void __iomem *) io_ports->data_addr, buf, len / 2);
+	readsw((void __iomem *)io_ports->data_addr, buf, len / 2);
 	leave_16bit(chipselect, mode);
 }
 
@@ -171,7 +171,7 @@ static void at91_ide_output_data(ide_dri
 	pdbg("cs %u buf %p len %d\n", chipselect,  buf, len);
 
 	enter_16bit(chipselect, mode);
-	__ide_mm_outsw((void __iomem *) io_ports->data_addr, buf, len / 2);
+	writesw((void __iomem *)io_ports->data_addr, buf, len / 2);
 	leave_16bit(chipselect, mode);
 }
#7 lore raw

Hello.

Quoted 34 lines · Bartlomiej Zolnierkiewicz
Quoted 28 lines
Quoted 27 lines
Quoted 27 lines
Quoted 27 lines
Quoted 8 lines

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data Register. Moreover some extra things need to be done when setting PIO mode. Only PIO mode is used, hardware have no DMA support. If interrupt line is connected through GPIO extra quirk is needed to cope with fake interrupts.

This is prolly only me, but I afil to compile this driver with the
current arch/arm/configs/at91sam9263ek_defconfig:

CC drivers/ide/at91_ide.o
drivers/ide/at91_ide.c: In function ‘at91_ide_input_data’:
drivers/ide/at91_ide.c:159: error: implicit declaration of function ‘__ide_mm_insw’ drivers/ide/at91_ide.c: In function ‘at91_ide_output_data’:
drivers/ide/at91_ide.c:174: error: implicit declaration of function ‘__ide_mm_outsw’ drivers/ide/at91_ide.c: In function ‘at91_ide_tf_load’:
drivers/ide/at91_ide.c:192: error: ‘task’ undeclared (first use in this function) drivers/ide/at91_ide.c:192: error: (Each undeclared identifier is reported only once
drivers/ide/at91_ide.c:192: error: for each function it appears in.) drivers/ide/at91_ide.c: At top level:
drivers/ide/at91_ide.c:303: warning: initialization from incompatible pointer type drivers/ide/at91_ide.c:304: warning: initialization from incompatible pointer type make[1]: *** [drivers/ide/at91_ide.o] Error 1
make: *** [drivers/ide/] Error 2

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Ah, it's only with pata-2.6 series applied. Bart, you clearly need
to include this driver to some of your patches...

I did and asked people to test the final result (see my mail on merging at91_ide)... unfortunately it seems like I missed few spots... :)

Quoted 6 lines
Quoted 2 lines

Oops, got muddled in too many follow-ups. :-)

Quoted 1 line

Namely, ide-remove-ide_task_t-typedef.patch is incomplete...

Was a bit more than that, I also needed to merge __ide_mm_* fixup below...

[ Anyway please verify that I got it right this time, thanks! ]

No, looks like you haven't again:

Applying patch ide-pass-command-instead-of-request-to-ide_pio_datablock.patch
1 out of 3 hunks FAILED
missing header for unified diff at line 208 of patch
The text leading up to this was:
--------------------------
| printk(KERN_ERR "%s: multimode not set!\n",
| drive->name);
--------------------------
No file to patch. Skipping patch.
11 out of 11 hunks ignored
Patch ide-pass-command-instead-of-request-to-ide_pio_datablock.patch does not apply (enforce with -f)

It appears that this patch has 2 hunks repeated twice now...

MBR, Sergei

#8 lore raw
Quoted 71 lines · Sergei Shtylyov

Hello.

Quoted 45 lines · Bartlomiej Zolnierkiewicz
Quoted 39 lines
Quoted 38 lines
Quoted 38 lines
Quoted 38 lines
Quoted 12 lines

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data
Register. Moreover some extra things need to be done when setting PIO mode.
Only PIO mode is used, hardware have no DMA support. If interrupt line is
connected through GPIO extra quirk is needed to cope with fake interrupts.

Quoted 1 line

This is prolly only me, but I afil to compile this driver with the
current arch/arm/configs/at91sam9263ek_defconfig:

CC drivers/ide/at91_ide.o
drivers/ide/at91_ide.c: In function ‘at91_ide_input_data’:
drivers/ide/at91_ide.c:159: error: implicit declaration of function ‘__ide_mm_insw’
drivers/ide/at91_ide.c: In function ‘at91_ide_output_data’:
drivers/ide/at91_ide.c:174: error: implicit declaration of function ‘__ide_mm_outsw’
drivers/ide/at91_ide.c: In function ‘at91_ide_tf_load’:
drivers/ide/at91_ide.c:192: error: ‘task’ undeclared (first use in this function)
drivers/ide/at91_ide.c:192: error: (Each undeclared identifier is reported only
once
drivers/ide/at91_ide.c:192: error: for each function it appears in.) drivers/ide/at91_ide.c: At top level:
drivers/ide/at91_ide.c:303: warning: initialization from incompatible pointer type
drivers/ide/at91_ide.c:304: warning: initialization from incompatible pointer type
make[1]: *** [drivers/ide/at91_ide.o] Error 1
make: *** [drivers/ide/] Error 2

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Ah, it's only with pata-2.6 series applied. Bart, you clearly need
to include this driver to some of your patches...

I did and asked people to test the final result (see my mail on merging at91_ide)... unfortunately it seems like I missed few spots... :)

Quoted 6 lines
Quoted 2 lines

Oops, got muddled in too many follow-ups. :-)

Quoted 1 line

Namely, ide-remove-ide_task_t-typedef.patch is incomplete...

Was a bit more than that, I also needed to merge __ide_mm_* fixup below...

[ Anyway please verify that I got it right this time, thanks! ]

No, looks like you haven't again:

Applying patch ide-pass-command-instead-of-request-to-ide_pio_datablock.patch 1 out of 3 hunks FAILED
missing header for unified diff at line 208 of patch
The text leading up to this was:
--------------------------
| printk(KERN_ERR "%s: multimode not set!\n",
| drive->name);
--------------------------
No file to patch. Skipping patch.
11 out of 11 hunks ignored
Patch ide-pass-command-instead-of-request-to-ide_pio_datablock.patch does not apply (enforce with -f)

Quoted 1 line

It appears that this patch has 2 hunks repeated twice now...

Strangely, the problem seems to be on your side this time as the patch applies just fine here and AFAIK Stephen has successfully applied it to linux-next (also md5sum of patch on kernel.org matches the local copy).

Did you forgot to re-sync the whole pata-2.6 tree by any chance?

Thanks,
Bart

#9 lore raw
Quoted 56 lines · Bartlomiej Zolnierkiewicz
Quoted 55 lines
Quoted 34 lines
Quoted 28 lines
Quoted 27 lines
Quoted 27 lines
Quoted 27 lines
Quoted 8 lines

This is IDE host driver for AT91 (SAM9, CAP9, AT572D940HF) Static Memory Controller with Compact Flash True IDE Mode logic.

Driver have to switch 8/16 bit bus width when accessing Task Tile or Data Register. Moreover some extra things need to be done when setting PIO mode. Only PIO mode is used, hardware have no DMA support. If interrupt line is connected through GPIO extra quirk is needed to cope with fake interrupts.

This is prolly only me, but I afil to compile this driver with the
current arch/arm/configs/at91sam9263ek_defconfig:

CC drivers/ide/at91_ide.o
drivers/ide/at91_ide.c: In function ‘at91_ide_input_data’:
drivers/ide/at91_ide.c:159: error: implicit declaration of function ‘__ide_mm_insw’ drivers/ide/at91_ide.c: In function ‘at91_ide_output_data’:
drivers/ide/at91_ide.c:174: error: implicit declaration of function ‘__ide_mm_outsw’ drivers/ide/at91_ide.c: In function ‘at91_ide_tf_load’:
drivers/ide/at91_ide.c:192: error: ‘task’ undeclared (first use in this function) drivers/ide/at91_ide.c:192: error: (Each undeclared identifier is reported only once
drivers/ide/at91_ide.c:192: error: for each function it appears in.) drivers/ide/at91_ide.c: At top level:
drivers/ide/at91_ide.c:303: warning: initialization from incompatible pointer type drivers/ide/at91_ide.c:304: warning: initialization from incompatible pointer type make[1]: *** [drivers/ide/at91_ide.o] Error 1
make: *** [drivers/ide/] Error 2

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Ah, it's only with pata-2.6 series applied. Bart, you clearly need
to include this driver to some of your patches...

I did and asked people to test the final result (see my mail on merging at91_ide)... unfortunately it seems like I missed few spots... :)

Quoted 7 lines
Quoted 3 lines

Oops, got muddled in too many follow-ups. :-)

Quoted 1 line

Namely, ide-remove-ide_task_t-typedef.patch is incomplete...

Was a bit more than that, I also needed to merge __ide_mm_* fixup below...

[ Anyway please verify that I got it right this time, thanks! ]

No, looks like you haven't again:

Applying patch ide-pass-command-instead-of-request-to-ide_pio_datablock.patch
1 out of 3 hunks FAILED
missing header for unified diff at line 208 of patch
The text leading up to this was:
--------------------------
| printk(KERN_ERR "%s: multimode not set!\n",
| drive->name);
--------------------------
No file to patch. Skipping patch.
11 out of 11 hunks ignored
Patch ide-pass-command-instead-of-request-to-ide_pio_datablock.patch does not apply (enforce with -f)

Quoted 7 lines
Quoted 1 line

It appears that this patch has 2 hunks repeated twice now...

Strangely, the problem seems to be on your side this time as the patch applies just fine here and AFAIK Stephen has successfully applied it to linux-next (also md5sum of patch on kernel.org matches the local copy).

Did you forgot to re-sync the whole pata-2.6 tree by any chance?

This was a result of re-sync... something went awry when downloading the
patch indeed -- perhaps I didn't really overwrite it and appended instead. :-<

Quoted 2 lines

Thanks,
Bart

WBR, Sergei