diff options
author | Michael Hennerich <michael.hennerich@analog.com> | 2007-11-17 23:46:58 +0800 |
---|---|---|
committer | Bryan Wu <bryan.wu@analog.com> | 2007-11-17 23:46:58 +0800 |
commit | 5c91fb902d4e6f6006faf45edd3f25932cb7d58c (patch) | |
tree | 0bdf4457456a6bc8f794078e22a885dc44b2d159 | |
parent | 1754a5d9f97f16f729066b8f125351af4951d6fe (diff) | |
download | kernel_goldelico_gta04-5c91fb902d4e6f6006faf45edd3f25932cb7d58c.zip kernel_goldelico_gta04-5c91fb902d4e6f6006faf45edd3f25932cb7d58c.tar.gz kernel_goldelico_gta04-5c91fb902d4e6f6006faf45edd3f25932cb7d58c.tar.bz2 |
Blackfin arch: Add assembly function insl_16
/*
* CPUs often take a performance hit when accessing unaligned memory
* locations. The actual performance hit varies, it can be small if the
* hardware handles it or large if we have to take an exception and fix
* it
* in software.
*
* Since an ethernet header is 14 bytes network drivers often end up
* with
* the IP header at an unaligned offset. The IP header can be aligned by
* shifting the start of the packet by 2 bytes. Drivers should do this
* with:
*
* skb_reserve(NET_IP_ALIGN);
*
* The downside to this alignment of the IP header is that the DMA is
* now
* unaligned. On some architectures the cost of an unaligned DMA is high
* and this cost outweighs the gains made by aligning the IP header.
*
* Since this trade off varies between architectures, we allow
* NET_IP_ALIGN
* to be overridden.
*/
This new function insl_16 allows to read form 32-bit IO and writes to
16-bit aligned memory. This is useful in above described scenario -
In particular with the AXIS AX88180 Gigabit Ethernet MAC.
Once the device is in 32-bit mode, reads from the RX FIFO always
decrements 4bytes.
While on the other side the destination address in SDRAM is always
16-bit aligned.
If we use skb_reserve(0) the receive buffer is 32-bit aligned but later
we hit a unaligned exception in the IP code.
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <bryan.wu@analog.com>
-rw-r--r-- | arch/blackfin/kernel/bfin_ksyms.c | 1 | ||||
-rw-r--r-- | arch/blackfin/lib/ins.S | 19 | ||||
-rw-r--r-- | include/asm-blackfin/io.h | 1 |
3 files changed, 21 insertions, 0 deletions
diff --git a/arch/blackfin/kernel/bfin_ksyms.c b/arch/blackfin/kernel/bfin_ksyms.c index 5dad9d3..b3fa7d8 100644 --- a/arch/blackfin/kernel/bfin_ksyms.c +++ b/arch/blackfin/kernel/bfin_ksyms.c @@ -100,6 +100,7 @@ EXPORT_SYMBOL(outsw); EXPORT_SYMBOL(insw); EXPORT_SYMBOL(outsl); EXPORT_SYMBOL(insl); +EXPORT_SYMBOL(insl_16); EXPORT_SYMBOL(irq_flags); EXPORT_SYMBOL(iounmap); EXPORT_SYMBOL(blackfin_dcache_invalidate_range); diff --git a/arch/blackfin/lib/ins.S b/arch/blackfin/lib/ins.S index a17cc77..df7b883 100644 --- a/arch/blackfin/lib/ins.S +++ b/arch/blackfin/lib/ins.S @@ -77,3 +77,22 @@ ENTRY(_insb) sti R3; RTS; ENDPROC(_insb) + + + +ENTRY(_insl_16) + P0 = R0; /* P0 = port */ + cli R3; + P1 = R1; /* P1 = address */ + P2 = R2; /* P2 = count */ + SSYNC; + LSETUP( .Llong16_loop_s, .Llong16_loop_e) LC0 = P2; +.Llong16_loop_s: R0 = [P0]; + W[P1++] = R0; + R0 = R0 >> 16; + W[P1++] = R0; + NOP; +.Llong16_loop_e: NOP; + sti R3; + RTS; +ENDPROC(_insl_16) diff --git a/include/asm-blackfin/io.h b/include/asm-blackfin/io.h index d1d2e6b..1601d62 100644 --- a/include/asm-blackfin/io.h +++ b/include/asm-blackfin/io.h @@ -122,6 +122,7 @@ extern void outsl(unsigned long port, const void *addr, unsigned long count); extern void insb(unsigned long port, void *addr, unsigned long count); extern void insw(unsigned long port, void *addr, unsigned long count); extern void insl(unsigned long port, void *addr, unsigned long count); +extern void insl_16(unsigned long port, void *addr, unsigned long count); extern void dma_outsb(unsigned long port, const void *addr, unsigned short count); extern void dma_outsw(unsigned long port, const void *addr, unsigned short count); |