Linux kernel mirror (for testing) git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel os linux

libata-sff: avoid byte swapping in ata_sff_data_xfer()

Handling of the trailing byte in ata_sff_data_xfer() is suboptimal bacause:

- it always initializes the padding buffer to 0 which is not really needed in
both the read and write cases;

- it has to use memcpy() to transfer a single byte from/to the padding buffer;

- it uses io{read|write}16() accessors which swap bytes on the big endian CPUs
and so have to additionally convert the data from/to the little endian format
instead of using io{read|write}16_rep() accessors which are not supposed to
change the byte ordering.

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>

authored by

Sergei Shtylyov and committed by
Jeff Garzik
2102d749 4da646b7

+13 -7
+13 -7
drivers/ata/libata-sff.c
··· 727 727 else 728 728 iowrite16_rep(data_addr, buf, words); 729 729 730 - /* Transfer trailing 1 byte, if any. */ 730 + /* Transfer trailing byte, if any. */ 731 731 if (unlikely(buflen & 0x01)) { 732 - __le16 align_buf[1] = { 0 }; 733 - unsigned char *trailing_buf = buf + buflen - 1; 732 + unsigned char pad[2]; 734 733 734 + /* Point buf to the tail of buffer */ 735 + buf += buflen - 1; 736 + 737 + /* 738 + * Use io*16_rep() accessors here as well to avoid pointlessly 739 + * swapping bytes to and fro on the big endian machines... 740 + */ 735 741 if (rw == READ) { 736 - align_buf[0] = cpu_to_le16(ioread16(data_addr)); 737 - memcpy(trailing_buf, align_buf, 1); 742 + ioread16_rep(data_addr, pad, 1); 743 + *buf = pad[0]; 738 744 } else { 739 - memcpy(align_buf, trailing_buf, 1); 740 - iowrite16(le16_to_cpu(align_buf[0]), data_addr); 745 + pad[0] = *buf; 746 + iowrite16_rep(data_addr, pad, 1); 741 747 } 742 748 words++; 743 749 }