Re: [PATCH 20/32] Blackfin arch: dma_memcpy borken for > 64K
On 5/21/07, Pekka Enberg <[EMAIL PROTECTED]> wrote: On 5/21/07, Bryan Wu <[EMAIL PROTECTED]> wrote: > -void *dma_memcpy(void *dest, const void *src, size_t size) > +void *_dma_memcpy(void *dest, const void *src, size_t size) Minor nitpick: the established naming convention is two underscores, not one. that should be marked as static now actually since the function isnt for use outside of this file > +void *dma_memcpy(void *dest, const void *src, size_t size) > +{ > + size_t bulk; > + size_t rest; > + void * addr; > + > + bulk = (size >> 16) << 16; I assume this is significantly faster on your architecture than: bulk = size & ~0xUL; which is more readable? actually they should get optimized to the same thing -mike - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 20/32] Blackfin arch: dma_memcpy borken for > 64K
Hi Bryan, On 5/21/07, Bryan Wu <[EMAIL PROTECTED]> wrote: -void *dma_memcpy(void *dest, const void *src, size_t size) +void *_dma_memcpy(void *dest, const void *src, size_t size) Minor nitpick: the established naming convention is two underscores, not one. +void *dma_memcpy(void *dest, const void *src, size_t size) +{ + size_t bulk; + size_t rest; + void * addr; + + bulk = (size >> 16) << 16; I assume this is significantly faster on your architecture than: bulk = size & ~0xUL; which is more readable? + rest = size - bulk; + if (bulk) + _dma_memcpy(dest, src, bulk); + addr = _dma_memcpy(dest+bulk, src+bulk, rest); + return addr; Does this work for 128K and up? Why? - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
[PATCH 20/32] Blackfin arch: dma_memcpy borken for > 64K
From: Aubrey Li <[EMAIL PROTECTED]> Signed-off-by: Aubrey Li <[EMAIL PROTECTED]> Signed-off-by: Bryan Wu <[EMAIL PROTECTED]> --- arch/blackfin/kernel/bfin_dma_5xx.c | 17 - 1 files changed, 16 insertions(+), 1 deletions(-) diff --git a/arch/blackfin/kernel/bfin_dma_5xx.c b/arch/blackfin/kernel/bfin_dma_5xx.c index 1c344ac..0ccb0dc 100644 --- a/arch/blackfin/kernel/bfin_dma_5xx.c +++ b/arch/blackfin/kernel/bfin_dma_5xx.c @@ -595,7 +595,7 @@ unsigned short get_dma_curr_ycount(unsigned int channel) } EXPORT_SYMBOL(get_dma_curr_ycount); -void *dma_memcpy(void *dest, const void *src, size_t size) +void *_dma_memcpy(void *dest, const void *src, size_t size) { int direction; /* 1 - address decrease, 0 - address increase */ int flag_align; /* 1 - address aligned, 0 - address unaligned */ @@ -734,6 +734,21 @@ void *dma_memcpy(void *dest, const void *src, size_t size) return dest; } + +void *dma_memcpy(void *dest, const void *src, size_t size) +{ + size_t bulk; + size_t rest; + void * addr; + + bulk = (size >> 16) << 16; + rest = size - bulk; + if (bulk) + _dma_memcpy(dest, src, bulk); + addr = _dma_memcpy(dest+bulk, src+bulk, rest); + return addr; +} + EXPORT_SYMBOL(dma_memcpy); void *safe_dma_memcpy(void *dest, const void *src, size_t size) -- 1.5.1.2 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
[PATCH 20/32] Blackfin arch: dma_memcpy borken for 64K
From: Aubrey Li [EMAIL PROTECTED] Signed-off-by: Aubrey Li [EMAIL PROTECTED] Signed-off-by: Bryan Wu [EMAIL PROTECTED] --- arch/blackfin/kernel/bfin_dma_5xx.c | 17 - 1 files changed, 16 insertions(+), 1 deletions(-) diff --git a/arch/blackfin/kernel/bfin_dma_5xx.c b/arch/blackfin/kernel/bfin_dma_5xx.c index 1c344ac..0ccb0dc 100644 --- a/arch/blackfin/kernel/bfin_dma_5xx.c +++ b/arch/blackfin/kernel/bfin_dma_5xx.c @@ -595,7 +595,7 @@ unsigned short get_dma_curr_ycount(unsigned int channel) } EXPORT_SYMBOL(get_dma_curr_ycount); -void *dma_memcpy(void *dest, const void *src, size_t size) +void *_dma_memcpy(void *dest, const void *src, size_t size) { int direction; /* 1 - address decrease, 0 - address increase */ int flag_align; /* 1 - address aligned, 0 - address unaligned */ @@ -734,6 +734,21 @@ void *dma_memcpy(void *dest, const void *src, size_t size) return dest; } + +void *dma_memcpy(void *dest, const void *src, size_t size) +{ + size_t bulk; + size_t rest; + void * addr; + + bulk = (size 16) 16; + rest = size - bulk; + if (bulk) + _dma_memcpy(dest, src, bulk); + addr = _dma_memcpy(dest+bulk, src+bulk, rest); + return addr; +} + EXPORT_SYMBOL(dma_memcpy); void *safe_dma_memcpy(void *dest, const void *src, size_t size) -- 1.5.1.2 - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 20/32] Blackfin arch: dma_memcpy borken for 64K
Hi Bryan, On 5/21/07, Bryan Wu [EMAIL PROTECTED] wrote: -void *dma_memcpy(void *dest, const void *src, size_t size) +void *_dma_memcpy(void *dest, const void *src, size_t size) Minor nitpick: the established naming convention is two underscores, not one. +void *dma_memcpy(void *dest, const void *src, size_t size) +{ + size_t bulk; + size_t rest; + void * addr; + + bulk = (size 16) 16; I assume this is significantly faster on your architecture than: bulk = size ~0xUL; which is more readable? + rest = size - bulk; + if (bulk) + _dma_memcpy(dest, src, bulk); + addr = _dma_memcpy(dest+bulk, src+bulk, rest); + return addr; Does this work for 128K and up? Why? - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH 20/32] Blackfin arch: dma_memcpy borken for 64K
On 5/21/07, Pekka Enberg [EMAIL PROTECTED] wrote: On 5/21/07, Bryan Wu [EMAIL PROTECTED] wrote: -void *dma_memcpy(void *dest, const void *src, size_t size) +void *_dma_memcpy(void *dest, const void *src, size_t size) Minor nitpick: the established naming convention is two underscores, not one. that should be marked as static now actually since the function isnt for use outside of this file +void *dma_memcpy(void *dest, const void *src, size_t size) +{ + size_t bulk; + size_t rest; + void * addr; + + bulk = (size 16) 16; I assume this is significantly faster on your architecture than: bulk = size ~0xUL; which is more readable? actually they should get optimized to the same thing -mike - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/