Skip to contentSkip to conversation

arm: omap: complete removal of machine_desc.io_pg_offst and .phys_io

PATCH New Anand Gadiyar 3 participants 4 messages Oct 18, 2010

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 arm: omap: complete removal of machine_desc.io_pg_offst and .phys_io

Commit 861bd81ee6 in linux-next as of 20101018 (arm: remove machine_desc.io_pg_offst and .phys_io) missed out on updating a few boards. This causes the build of omap2plus_defconfig to fail. Fix this.

Cc Nicolas Pitre <[email protected]>
Cc Eric Miao <[email protected]>
Cc Tony Lindgren <[email protected]>

---
I've not boot tested on these three boards, and am simply following what the original patch does.

 arch/arm/mach-omap2/board-cm-t3517.c   |    2 --
 arch/arm/mach-omap2/board-igep0030.c   |    2 --
 arch/arm/mach-omap2/board-omap3logic.c |    4 ----
 3 files changed, 8 deletions(-)
arch/arm/mach-omap2/board-cm-t3517.c +0 -2
Index: mainline/arch/arm/mach-omap2/board-cm-t3517.c
===================================================================
@@ -283,8 +283,6 @@ static void __init cm_t3517_init(void)
 }
 
 MACHINE_START(CM_T3517, "Compulab CM-T3517")
-	.phys_io	= 0x48000000,
-	.io_pg_offst	= ((0xd8000000) >> 18) & 0xfffc,
 	.boot_params	= 0x80000100,
 	.map_io		= omap3_map_io,
 	.reserve        = omap_reserve,
arch/arm/mach-omap2/board-igep0030.c +0 -2
Index: mainline/arch/arm/mach-omap2/board-igep0030.c
===================================================================
@@ -392,8 +392,6 @@ static void __init igep3_init(void)
 }
 
 MACHINE_START(IGEP0030, "IGEP OMAP3 module")
-	.phys_io	= 0x48000000,
-	.io_pg_offst	= ((0xfa000000) >> 18) & 0xfffc,
 	.boot_params	= 0x80000100,
 	.map_io		= omap3_map_io,
 	.init_irq	= igep3_init_irq,
arch/arm/mach-omap2/board-omap3logic.c +0 -4
Index: mainline/arch/arm/mach-omap2/board-omap3logic.c
===================================================================
@@ -225,8 +225,6 @@ static void __init omap3logic_init(void)
 }
 
 MACHINE_START(OMAP3_TORPEDO, "Logic OMAP3 Torpedo board")
-	.phys_io	= 0x48000000,
-	.io_pg_offst	= ((0xd8000000) >> 18) & 0xfffc,
 	.boot_params	= 0x80000100,
 	.map_io		= omap3_map_io,
 	.init_irq	= omap3logic_init_irq,
@@ -235,8 +233,6 @@ MACHINE_START(OMAP3_TORPEDO, "Logic OMAP
 MACHINE_END
 
 MACHINE_START(OMAP3530_LV_SOM, "OMAP Logic 3530 LV SOM board")
-	.phys_io	= 0x48000000,
-	.io_pg_offst	= ((0xd8000000) >> 18) & 0xfffc,
 	.boot_params	= 0x80000100,
 	.map_io		= omap3_map_io,
 	.init_irq	= omap3logic_init_irq,
#1 lore raw

PATCH 1/1 arm: omap: complete removal of machine_desc.io_pg_offst and .phys_io

Quoted 4 lines · Anand Gadiyar

Commit 861bd81ee6 in linux-next as of 20101018 (arm: remove machine_desc.io_pg_offst and .phys_io) missed out on updating a few boards. This causes the build of omap2plus_defconfig to fail. Fix this.

Not quite true. The patch merged into my tree eliminates io_pg_offst and phys_io for everything that is in my tree.

For new board support code which isn't in my tree, there is a problem - patches to it can't be applied to my tree.

However, applying these patches to the trees which have the new code in won't have the other changes related to the removal of io_pg_offst, and so the support code will subtly break until it's all merged together.

There's two solutions to this. One is to wait merging new board code until after this merge window. The other is to ensure that new board code is in _my_ tree before the io_pg_offst changes were submitted...

#2 lore raw

PATCH 1/1 arm: omap: complete removal of machine_desc.io_pg_offst and .phys_io

Quoted 12 lines · Russell King - ARM Linux
Quoted 4 lines · Anand Gadiyar

Commit 861bd81ee6 in linux-next as of 20101018 (arm: remove machine_desc.io_pg_offst and .phys_io) missed out on updating a few boards. This causes the build of omap2plus_defconfig to fail. Fix this.

Not quite true. The patch merged into my tree eliminates io_pg_offst and phys_io for everything that is in my tree.

For new board support code which isn't in my tree, there is a problem - patches to it can't be applied to my tree.

Quoted 7 lines

However, applying these patches to the trees which have the new code in won't have the other changes related to the removal of io_pg_offst, and so the support code will subtly break until it's all merged together.

There's two solutions to this. One is to wait merging new board code until after this merge window. The other is to ensure that new board code is in _my_ tree before the io_pg_offst changes were submitted...

I'm going to drop the io_pg_offst patches for a couple of days. If this new board support can be sorted out to come into my tree in the next 24 hours, then I'm sure we can get this fixed properly.

I don't think I can afford to hold the io_pg_offst patches for more than 48 hours though, so Wednesday evening they'll be merged whatever.

maintainer
#3 lore raw

PATCH 1/1 arm: omap: complete removal of machine_desc.io_pg_offst and .phys_io

Quoted 21 lines · Russell King - ARM Linux
Quoted 12 lines · Russell King - ARM Linux
Quoted 4 lines · Anand Gadiyar

Commit 861bd81ee6 in linux-next as of 20101018 (arm: remove machine_desc.io_pg_offst and .phys_io) missed out on updating a few boards. This causes the build of omap2plus_defconfig to fail. Fix this.

Not quite true. The patch merged into my tree eliminates io_pg_offst and phys_io for everything that is in my tree.

For new board support code which isn't in my tree, there is a problem - patches to it can't be applied to my tree.

Quoted 7 lines

However, applying these patches to the trees which have the new code in won't have the other changes related to the removal of io_pg_offst, and so the support code will subtly break until it's all merged together.

There's two solutions to this. One is to wait merging new board code until after this merge window. The other is to ensure that new board code is in _my_ tree before the io_pg_offst changes were submitted...

Quoted 6 lines

I'm going to drop the io_pg_offst patches for a couple of days. If this new board support can be sorted out to come into my tree in the next 24 hours, then I'm sure we can get this fixed properly.

I don't think I can afford to hold the io_pg_offst patches for more than 48 hours though, so Wednesday evening they'll be merged whatever.

Go ahead, I'll patch up accordingly in the omap tree so it will all merge nicely in for-next and for the merge window.

Regards,

Tony