linux_dsm_epyc7002/arch/blackfin/lib/memset.S
Robin Getz 648eee52cc Blackfin: optimize strncpy a bit
Add a little strncpy optimization which can easily cut boot time by 20%.

When the kernel is booting with initramfs, it builds up the filesystem
from a cpio archive by calling strncpy_from_user() via fs/namei.c's
do_getname() on every file in the archive (which can be lots) with a
length of PATH_MAX (1024).  This causes the dest of the strncpy to be
padded with many NUL bytes.

This optimization mostly causes these NUL bytes to be padded with a call
to memset() which is already optimized for filling memory quickly, but
the hardware loop helps a little bit as well.

Boot time measured with 'loglevel=0' so UART speed doesn't get in the way.

Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2010-05-22 14:19:11 -04:00

88 lines
1.8 KiB
ArmAsm

/*
* Copyright 2004-2009 Analog Devices Inc.
*
* Licensed under the ADI BSD license or the GPL-2 (or later)
*/
#include <linux/linkage.h>
.align 2
#ifdef CONFIG_MEMSET_L1
.section .l1.text
#else
.text
#endif
/*
* C Library function MEMSET
* R0 = address (leave unchanged to form result)
* R1 = filler byte
* R2 = count
* Favours word aligned data.
* The strncpy assumes that I0 and I1 are not used in this function
*/
ENTRY(_memset)
P0 = R0 ; /* P0 = address */
P2 = R2 ; /* P2 = count */
R3 = R0 + R2; /* end */
CC = R2 <= 7(IU);
IF CC JUMP .Ltoo_small;
R1 = R1.B (Z); /* R1 = fill char */
R2 = 3;
R2 = R0 & R2; /* addr bottom two bits */
CC = R2 == 0; /* AZ set if zero. */
IF !CC JUMP .Lforce_align ; /* Jump if addr not aligned. */
.Laligned:
P1 = P2 >> 2; /* count = n/4 */
R2 = R1 << 8; /* create quad filler */
R2.L = R2.L + R1.L(NS);
R2.H = R2.L + R1.H(NS);
P2 = R3;
LSETUP (.Lquad_loop , .Lquad_loop) LC0=P1;
.Lquad_loop:
[P0++] = R2;
CC = P0 == P2;
IF !CC JUMP .Lbytes_left;
RTS;
.Lbytes_left:
R2 = R3; /* end point */
R3 = P0; /* current position */
R2 = R2 - R3; /* bytes left */
P2 = R2;
.Ltoo_small:
CC = P2 == 0; /* Check zero count */
IF CC JUMP .Lfinished; /* Unusual */
.Lbytes:
LSETUP (.Lbyte_loop , .Lbyte_loop) LC0=P2;
.Lbyte_loop:
B[P0++] = R1;
.Lfinished:
RTS;
.Lforce_align:
CC = BITTST (R0, 0); /* odd byte */
R0 = 4;
R0 = R0 - R2;
P1 = R0;
R0 = P0; /* Recover return address */
IF !CC JUMP .Lskip1;
B[P0++] = R1;
.Lskip1:
CC = R2 <= 2; /* 2 bytes */
P2 -= P1; /* reduce count */
IF !CC JUMP .Laligned;
B[P0++] = R1;
B[P0++] = R1;
JUMP .Laligned;
ENDPROC(_memset)