* [PATCH] ARM: Thumb-2: Enable ARM/Thumb interworking for v7 cache maintenance
@ 2011-02-15 17:56 Dave Martin
2011-02-15 18:00 ` Russell King - ARM Linux
2011-02-15 19:30 ` Nicolas Pitre
0 siblings, 2 replies; 4+ messages in thread
From: Dave Martin @ 2011-02-15 17:56 UTC (permalink / raw)
To: linux-arm-kernel
Cache maintenence functions may need to be called from low-level
platform code (such as for omap3).
Some low-level code may need to be built as ARM code even in a Thumb-2
kernel, if the code needs to talk to a non-Thumb-2-aware bootloader
or firmware (again, omap3 is an example).
In order to avoid the need for complex hacks to make calling
these cache maintenance routines work, simply use "bx lr" to
return from these functions.
There's no risk of build failures, since mm/*-v7.S are by definition
built with -march=armv7-a.
bx lr always does the right thing for a pure ARM kernel, so we
shouldn't need any conditional assembly here.
Signed-off-by: Dave Martin <dave.martin@linaro.org>
---
arch/arm/mm/cache-v7.S | 20 ++++++++++----------
1 files changed, 10 insertions(+), 10 deletions(-)
diff --git a/arch/arm/mm/cache-v7.S b/arch/arm/mm/cache-v7.S
index 6136e68..a5e5aa1 100644
--- a/arch/arm/mm/cache-v7.S
+++ b/arch/arm/mm/cache-v7.S
@@ -29,7 +29,7 @@ ENTRY(v7_flush_icache_all)
mov r0, #0
ALT_SMP(mcr p15, 0, r0, c7, c1, 0) @ invalidate I-cache inner shareable
ALT_UP(mcr p15, 0, r0, c7, c5, 0) @ I+BTB cache invalidate
- mov pc, lr
+ bx lr
ENDPROC(v7_flush_icache_all)
/*
@@ -87,7 +87,7 @@ finished:
mcr p15, 2, r10, c0, c0, 0 @ select current cache level in cssr
dsb
isb
- mov pc, lr
+ bx lr
ENDPROC(v7_flush_dcache_all)
/*
@@ -110,7 +110,7 @@ ENTRY(v7_flush_kern_cache_all)
ALT_UP(mcr p15, 0, r0, c7, c5, 0) @ I+BTB cache invalidate
ARM( ldmfd sp!, {r4-r5, r7, r9-r11, lr} )
THUMB( ldmfd sp!, {r4-r7, r9-r11, lr} )
- mov pc, lr
+ bx lr
ENDPROC(v7_flush_kern_cache_all)
/*
@@ -136,7 +136,7 @@ ENTRY(v7_flush_user_cache_all)
* - we have a VIPT cache.
*/
ENTRY(v7_flush_user_cache_range)
- mov pc, lr
+ bx lr
ENDPROC(v7_flush_user_cache_all)
ENDPROC(v7_flush_user_cache_range)
@@ -194,7 +194,7 @@ ENTRY(v7_coherent_user_range)
ALT_UP(mcr p15, 0, r0, c7, c5, 6) @ invalidate BTB
dsb
isb
- mov pc, lr
+ bx lr
/*
* Fault handling for the cache operation above. If the virtual address in r0
@@ -227,7 +227,7 @@ ENTRY(v7_flush_kern_dcache_area)
cmp r0, r1
blo 1b
dsb
- mov pc, lr
+ bx lr
ENDPROC(v7_flush_kern_dcache_area)
/*
@@ -256,7 +256,7 @@ v7_dma_inv_range:
cmp r0, r1
blo 1b
dsb
- mov pc, lr
+ bx lr
ENDPROC(v7_dma_inv_range)
/*
@@ -274,7 +274,7 @@ v7_dma_clean_range:
cmp r0, r1
blo 1b
dsb
- mov pc, lr
+ bx lr
ENDPROC(v7_dma_clean_range)
/*
@@ -292,7 +292,7 @@ ENTRY(v7_dma_flush_range)
cmp r0, r1
blo 1b
dsb
- mov pc, lr
+ bx lr
ENDPROC(v7_dma_flush_range)
/*
@@ -318,7 +318,7 @@ ENTRY(v7_dma_unmap_area)
add r1, r1, r0
teq r2, #DMA_TO_DEVICE
bne v7_dma_inv_range
- mov pc, lr
+ bx lr
ENDPROC(v7_dma_unmap_area)
__INITDATA
--
1.7.1
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH] ARM: Thumb-2: Enable ARM/Thumb interworking for v7 cache maintenance
2011-02-15 17:56 [PATCH] ARM: Thumb-2: Enable ARM/Thumb interworking for v7 cache maintenance Dave Martin
@ 2011-02-15 18:00 ` Russell King - ARM Linux
2011-02-16 9:26 ` Dave Martin
2011-02-15 19:30 ` Nicolas Pitre
1 sibling, 1 reply; 4+ messages in thread
From: Russell King - ARM Linux @ 2011-02-15 18:00 UTC (permalink / raw)
To: linux-arm-kernel
On Tue, Feb 15, 2011 at 05:56:05PM +0000, Dave Martin wrote:
> Cache maintenence functions may need to be called from low-level
> platform code (such as for omap3).
I'd rather wait until we have the suspend stuff sorted out. I'm
waiting for OMAP folk to sort out what's happening with the sleep*.S
code so that it can have the generic CPU suspend code integrated.
That could result in this change not being required.
^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH] ARM: Thumb-2: Enable ARM/Thumb interworking for v7 cache maintenance
2011-02-15 17:56 [PATCH] ARM: Thumb-2: Enable ARM/Thumb interworking for v7 cache maintenance Dave Martin
2011-02-15 18:00 ` Russell King - ARM Linux
@ 2011-02-15 19:30 ` Nicolas Pitre
1 sibling, 0 replies; 4+ messages in thread
From: Nicolas Pitre @ 2011-02-15 19:30 UTC (permalink / raw)
To: linux-arm-kernel
On Tue, 15 Feb 2011, Dave Martin wrote:
> Cache maintenence functions may need to be called from low-level
> platform code (such as for omap3).
>
> Some low-level code may need to be built as ARM code even in a Thumb-2
> kernel, if the code needs to talk to a non-Thumb-2-aware bootloader
> or firmware (again, omap3 is an example).
>
> In order to avoid the need for complex hacks to make calling
> these cache maintenance routines work, simply use "bx lr" to
> return from these functions.
>
> There's no risk of build failures, since mm/*-v7.S are by definition
> built with -march=armv7-a.
>
> bx lr always does the right thing for a pure ARM kernel, so we
> shouldn't need any conditional assembly here.
>
> Signed-off-by: Dave Martin <dave.martin@linaro.org>
Acked-by: Nicolas Pitre <nicolas.pitre@linaro.org>
> ---
> arch/arm/mm/cache-v7.S | 20 ++++++++++----------
> 1 files changed, 10 insertions(+), 10 deletions(-)
>
> diff --git a/arch/arm/mm/cache-v7.S b/arch/arm/mm/cache-v7.S
> index 6136e68..a5e5aa1 100644
> --- a/arch/arm/mm/cache-v7.S
> +++ b/arch/arm/mm/cache-v7.S
> @@ -29,7 +29,7 @@ ENTRY(v7_flush_icache_all)
> mov r0, #0
> ALT_SMP(mcr p15, 0, r0, c7, c1, 0) @ invalidate I-cache inner shareable
> ALT_UP(mcr p15, 0, r0, c7, c5, 0) @ I+BTB cache invalidate
> - mov pc, lr
> + bx lr
> ENDPROC(v7_flush_icache_all)
>
> /*
> @@ -87,7 +87,7 @@ finished:
> mcr p15, 2, r10, c0, c0, 0 @ select current cache level in cssr
> dsb
> isb
> - mov pc, lr
> + bx lr
> ENDPROC(v7_flush_dcache_all)
>
> /*
> @@ -110,7 +110,7 @@ ENTRY(v7_flush_kern_cache_all)
> ALT_UP(mcr p15, 0, r0, c7, c5, 0) @ I+BTB cache invalidate
> ARM( ldmfd sp!, {r4-r5, r7, r9-r11, lr} )
> THUMB( ldmfd sp!, {r4-r7, r9-r11, lr} )
> - mov pc, lr
> + bx lr
> ENDPROC(v7_flush_kern_cache_all)
>
> /*
> @@ -136,7 +136,7 @@ ENTRY(v7_flush_user_cache_all)
> * - we have a VIPT cache.
> */
> ENTRY(v7_flush_user_cache_range)
> - mov pc, lr
> + bx lr
> ENDPROC(v7_flush_user_cache_all)
> ENDPROC(v7_flush_user_cache_range)
>
> @@ -194,7 +194,7 @@ ENTRY(v7_coherent_user_range)
> ALT_UP(mcr p15, 0, r0, c7, c5, 6) @ invalidate BTB
> dsb
> isb
> - mov pc, lr
> + bx lr
>
> /*
> * Fault handling for the cache operation above. If the virtual address in r0
> @@ -227,7 +227,7 @@ ENTRY(v7_flush_kern_dcache_area)
> cmp r0, r1
> blo 1b
> dsb
> - mov pc, lr
> + bx lr
> ENDPROC(v7_flush_kern_dcache_area)
>
> /*
> @@ -256,7 +256,7 @@ v7_dma_inv_range:
> cmp r0, r1
> blo 1b
> dsb
> - mov pc, lr
> + bx lr
> ENDPROC(v7_dma_inv_range)
>
> /*
> @@ -274,7 +274,7 @@ v7_dma_clean_range:
> cmp r0, r1
> blo 1b
> dsb
> - mov pc, lr
> + bx lr
> ENDPROC(v7_dma_clean_range)
>
> /*
> @@ -292,7 +292,7 @@ ENTRY(v7_dma_flush_range)
> cmp r0, r1
> blo 1b
> dsb
> - mov pc, lr
> + bx lr
> ENDPROC(v7_dma_flush_range)
>
> /*
> @@ -318,7 +318,7 @@ ENTRY(v7_dma_unmap_area)
> add r1, r1, r0
> teq r2, #DMA_TO_DEVICE
> bne v7_dma_inv_range
> - mov pc, lr
> + bx lr
> ENDPROC(v7_dma_unmap_area)
>
> __INITDATA
> --
> 1.7.1
>
>
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH] ARM: Thumb-2: Enable ARM/Thumb interworking for v7 cache maintenance
2011-02-15 18:00 ` Russell King - ARM Linux
@ 2011-02-16 9:26 ` Dave Martin
0 siblings, 0 replies; 4+ messages in thread
From: Dave Martin @ 2011-02-16 9:26 UTC (permalink / raw)
To: linux-arm-kernel
On Tue, Feb 15, 2011 at 6:00 PM, Russell King - ARM Linux
<linux@arm.linux.org.uk> wrote:
> On Tue, Feb 15, 2011 at 05:56:05PM +0000, Dave Martin wrote:
>> Cache maintenence functions may need to be called from low-level
>> platform code (such as for omap3).
>
> I'd rather wait until we have the suspend stuff sorted out. ?I'm
> waiting for OMAP folk to sort out what's happening with the sleep*.S
> code so that it can have the generic CPU suspend code integrated.
> That could result in this change not being required.
>
OK; I guess we can work around this in the omap BSP for the time being.
---Dave
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2011-02-16 9:26 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-02-15 17:56 [PATCH] ARM: Thumb-2: Enable ARM/Thumb interworking for v7 cache maintenance Dave Martin
2011-02-15 18:00 ` Russell King - ARM Linux
2011-02-16 9:26 ` Dave Martin
2011-02-15 19:30 ` Nicolas Pitre
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).