diff options
| author | Russell King <[email protected]> | 2014-06-30 16:29:12 +0100 |
|---|---|---|
| committer | Russell King <[email protected]> | 2014-07-18 12:29:04 +0100 |
| commit | 6ebbf2ce437b33022d30badd49dc94d33ecfa498 (patch) | |
| tree | bc015e35b456a28bb0e501803a454dc0c0d3291a /arch/arm/mm/cache-v4wb.S | |
| parent | af040ffc9ba1e079ee4c0748aff64fa3d4716fa5 (diff) | |
ARM: convert all "mov.* pc, reg" to "bx reg" for ARMv6+
ARMv6 and greater introduced a new instruction ("bx") which can be used
to return from function calls. Recent CPUs perform better when the
"bx lr" instruction is used rather than the "mov pc, lr" instruction,
and this sequence is strongly recommended to be used by the ARM
architecture manual (section A.4.1.1).
We provide a new macro "ret" with all its variants for the condition
code which will resolve to the appropriate instruction.
Rather than doing this piecemeal, and miss some instances, change all
the "mov pc" instances to use the new macro, with the exception of
the "movs" instruction and the kprobes code. This allows us to detect
the "mov pc, lr" case and fix it up - and also gives us the possibility
of deploying this for other registers depending on the CPU selection.
Reported-by: Will Deacon <[email protected]>
Tested-by: Stephen Warren <[email protected]> # Tegra Jetson TK1
Tested-by: Robert Jarzmik <[email protected]> # mioa701_bootresume.S
Tested-by: Andrew Lunn <[email protected]> # Kirkwood
Tested-by: Shawn Guo <[email protected]>
Tested-by: Tony Lindgren <[email protected]> # OMAPs
Tested-by: Gregory CLEMENT <[email protected]> # Armada XP, 375, 385
Acked-by: Sekhar Nori <[email protected]> # DaVinci
Acked-by: Christoffer Dall <[email protected]> # kvm/hyp
Acked-by: Haojian Zhuang <[email protected]> # PXA3xx
Acked-by: Stefano Stabellini <[email protected]> # Xen
Tested-by: Uwe Kleine-König <[email protected]> # ARMv7M
Tested-by: Simon Horman <[email protected]> # Shmobile
Signed-off-by: Russell King <[email protected]>
Diffstat (limited to 'arch/arm/mm/cache-v4wb.S')
| -rw-r--r-- | arch/arm/mm/cache-v4wb.S | 15 |
1 files changed, 8 insertions, 7 deletions
diff --git a/arch/arm/mm/cache-v4wb.S b/arch/arm/mm/cache-v4wb.S index cd4945321407..2522f8c8fbb1 100644 --- a/arch/arm/mm/cache-v4wb.S +++ b/arch/arm/mm/cache-v4wb.S @@ -9,6 +9,7 @@ */ #include <linux/linkage.h> #include <linux/init.h> +#include <asm/assembler.h> #include <asm/memory.h> #include <asm/page.h> #include "proc-macros.S" @@ -58,7 +59,7 @@ flush_base: ENTRY(v4wb_flush_icache_all) mov r0, #0 mcr p15, 0, r0, c7, c5, 0 @ invalidate I cache - mov pc, lr + ret lr ENDPROC(v4wb_flush_icache_all) /* @@ -94,7 +95,7 @@ __flush_whole_cache: blo 1b #endif mcr p15, 0, ip, c7, c10, 4 @ drain write buffer - mov pc, lr + ret lr /* * flush_user_cache_range(start, end, flags) @@ -122,7 +123,7 @@ ENTRY(v4wb_flush_user_cache_range) blo 1b tst r2, #VM_EXEC mcrne p15, 0, ip, c7, c10, 4 @ drain write buffer - mov pc, lr + ret lr /* * flush_kern_dcache_area(void *addr, size_t size) @@ -170,7 +171,7 @@ ENTRY(v4wb_coherent_user_range) mov r0, #0 mcr p15, 0, r0, c7, c5, 0 @ invalidate I cache mcr p15, 0, r0, c7, c10, 4 @ drain WB - mov pc, lr + ret lr /* @@ -195,7 +196,7 @@ v4wb_dma_inv_range: cmp r0, r1 blo 1b mcr p15, 0, r0, c7, c10, 4 @ drain write buffer - mov pc, lr + ret lr /* * dma_clean_range(start, end) @@ -212,7 +213,7 @@ v4wb_dma_clean_range: cmp r0, r1 blo 1b mcr p15, 0, r0, c7, c10, 4 @ drain write buffer - mov pc, lr + ret lr /* * dma_flush_range(start, end) @@ -248,7 +249,7 @@ ENDPROC(v4wb_dma_map_area) * - dir - DMA direction */ ENTRY(v4wb_dma_unmap_area) - mov pc, lr + ret lr ENDPROC(v4wb_dma_unmap_area) .globl v4wb_flush_kern_cache_louis |