On 11/26/2012 04:33:08 AM, Phil Sutter wrote:
The basic idea is taken from the linux-kernel, but further optimized.
First align the buffer to 8 bytes, then use ldrd/strd to read and store in 8 byte quantities, then do the final bytes.
Tested using: 'date ; nand read.raw 0xE00000 0x0 0x10000 ; date'. Without this patch, NAND read of 132MB took 49s (~2.69MB/s). With this patch in place, reading the same amount of data was done in 27s (~4.89MB/s). So read performance is increased by ~80%!
Signed-off-by: Nico Erfurth ne@erfurth.eu Tested-by: Phil Sutter phil.sutter@viprinet.com Cc: Prafulla Wadaskar prafulla@marvell.com
drivers/mtd/nand/kirkwood_nand.c | 29 +++++++++++++++++++++++++++++ 1 files changed, 29 insertions(+), 0 deletions(-)
diff --git a/drivers/mtd/nand/kirkwood_nand.c b/drivers/mtd/nand/kirkwood_nand.c index bdab5aa..e04a59f 100644 --- a/drivers/mtd/nand/kirkwood_nand.c +++ b/drivers/mtd/nand/kirkwood_nand.c @@ -38,6 +38,34 @@ struct kwnandf_registers { static struct kwnandf_registers *nf_reg = (struct kwnandf_registers *)KW_NANDF_BASE;
+/* The basic idea is stolen from the linux kernel, but the inner loop is optimized a bit more */ +static void kw_nand_read_buf(struct mtd_info *mtd, uint8_t *buf, int len) +{
- struct nand_chip *chip = mtd->priv;
- while (len && (unsigned long)buf & 7)
- {
Brace goes on the previous line.
*buf++ = readb(chip->IO_ADDR_R);len--;- };
- asm volatile (
".LFlashLoop:\n"" subs\t%0, #8\n"" ldrpld\tr2, [%2]\n" // Read 2 words" strpld\tr2, [%1], #8\n" // Read 2 words" bpl\t.LFlashLoop\n" // This results in oneadditional loop if len%8 <> 0
" addne\t%0, #8\n": "+&r" (len), "+&r" (buf): "r" (chip->IO_ADDR_R): "r2", "r3", "memory", "cc"- );
Use a real tab (or a space) rather than \t (which only helps readability in the asm output, rather than the C source that people actually look at).
Should probably use a numeric label to avoid any possibility of conflict.
Would this make more sense as a more generic optimized memcpy_fromio() or similar?
-Scott