linux_dsm_epyc7002/arch/arm/lib/findbit.S

197 lines
4.8 KiB
ArmAsm
Raw Normal View History

/*
* linux/arch/arm/lib/findbit.S
*
* Copyright (C) 1995-2000 Russell King
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*
* 16th March 2001 - John Ripley <jripley@sonicblue.com>
* Fixed so that "size" is an exclusive not an inclusive quantity.
* All users of these functions expect exclusive sizes, and may
* also call with zero size.
* Reworked by rmk.
*/
#include <linux/linkage.h>
#include <asm/assembler.h>
.text
/*
* Purpose : Find a 'zero' bit
* Prototype: int find_first_zero_bit(void *addr, unsigned int maxbit);
*/
ENTRY(_find_first_zero_bit_le)
teq r1, #0
beq 3f
mov r2, #0
1:
ARM( ldrb r3, [r0, r2, lsr #3] )
THUMB( lsr r3, r2, #3 )
THUMB( ldrb r3, [r0, r3] )
eors r3, r3, #0xff @ invert bits
bne .L_found @ any now set - found zero bit
add r2, r2, #8 @ next bit pointer
2: cmp r2, r1 @ any more?
blo 1b
3: mov r0, r1 @ no free bits
ARM: convert all "mov.* pc, reg" to "bx reg" for ARMv6+ ARMv6 and greater introduced a new instruction ("bx") which can be used to return from function calls. Recent CPUs perform better when the "bx lr" instruction is used rather than the "mov pc, lr" instruction, and this sequence is strongly recommended to be used by the ARM architecture manual (section A.4.1.1). We provide a new macro "ret" with all its variants for the condition code which will resolve to the appropriate instruction. Rather than doing this piecemeal, and miss some instances, change all the "mov pc" instances to use the new macro, with the exception of the "movs" instruction and the kprobes code. This allows us to detect the "mov pc, lr" case and fix it up - and also gives us the possibility of deploying this for other registers depending on the CPU selection. Reported-by: Will Deacon <will.deacon@arm.com> Tested-by: Stephen Warren <swarren@nvidia.com> # Tegra Jetson TK1 Tested-by: Robert Jarzmik <robert.jarzmik@free.fr> # mioa701_bootresume.S Tested-by: Andrew Lunn <andrew@lunn.ch> # Kirkwood Tested-by: Shawn Guo <shawn.guo@freescale.com> Tested-by: Tony Lindgren <tony@atomide.com> # OMAPs Tested-by: Gregory CLEMENT <gregory.clement@free-electrons.com> # Armada XP, 375, 385 Acked-by: Sekhar Nori <nsekhar@ti.com> # DaVinci Acked-by: Christoffer Dall <christoffer.dall@linaro.org> # kvm/hyp Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com> # PXA3xx Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> # Xen Tested-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> # ARMv7M Tested-by: Simon Horman <horms+renesas@verge.net.au> # Shmobile Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-06-30 22:29:12 +07:00
ret lr
ENDPROC(_find_first_zero_bit_le)
/*
* Purpose : Find next 'zero' bit
* Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset)
*/
ENTRY(_find_next_zero_bit_le)
teq r1, #0
beq 3b
ands ip, r2, #7
beq 1b @ If new byte, goto old routine
ARM( ldrb r3, [r0, r2, lsr #3] )
THUMB( lsr r3, r2, #3 )
THUMB( ldrb r3, [r0, r3] )
eor r3, r3, #0xff @ now looking for a 1 bit
movs r3, r3, lsr ip @ shift off unused bits
bne .L_found
orr r2, r2, #7 @ if zero, then no bits here
add r2, r2, #1 @ align bit pointer
b 2b @ loop for next bit
ENDPROC(_find_next_zero_bit_le)
/*
* Purpose : Find a 'one' bit
* Prototype: int find_first_bit(const unsigned long *addr, unsigned int maxbit);
*/
ENTRY(_find_first_bit_le)
teq r1, #0
beq 3f
mov r2, #0
1:
ARM( ldrb r3, [r0, r2, lsr #3] )
THUMB( lsr r3, r2, #3 )
THUMB( ldrb r3, [r0, r3] )
movs r3, r3
bne .L_found @ any now set - found zero bit
add r2, r2, #8 @ next bit pointer
2: cmp r2, r1 @ any more?
blo 1b
3: mov r0, r1 @ no free bits
ARM: convert all "mov.* pc, reg" to "bx reg" for ARMv6+ ARMv6 and greater introduced a new instruction ("bx") which can be used to return from function calls. Recent CPUs perform better when the "bx lr" instruction is used rather than the "mov pc, lr" instruction, and this sequence is strongly recommended to be used by the ARM architecture manual (section A.4.1.1). We provide a new macro "ret" with all its variants for the condition code which will resolve to the appropriate instruction. Rather than doing this piecemeal, and miss some instances, change all the "mov pc" instances to use the new macro, with the exception of the "movs" instruction and the kprobes code. This allows us to detect the "mov pc, lr" case and fix it up - and also gives us the possibility of deploying this for other registers depending on the CPU selection. Reported-by: Will Deacon <will.deacon@arm.com> Tested-by: Stephen Warren <swarren@nvidia.com> # Tegra Jetson TK1 Tested-by: Robert Jarzmik <robert.jarzmik@free.fr> # mioa701_bootresume.S Tested-by: Andrew Lunn <andrew@lunn.ch> # Kirkwood Tested-by: Shawn Guo <shawn.guo@freescale.com> Tested-by: Tony Lindgren <tony@atomide.com> # OMAPs Tested-by: Gregory CLEMENT <gregory.clement@free-electrons.com> # Armada XP, 375, 385 Acked-by: Sekhar Nori <nsekhar@ti.com> # DaVinci Acked-by: Christoffer Dall <christoffer.dall@linaro.org> # kvm/hyp Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com> # PXA3xx Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> # Xen Tested-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> # ARMv7M Tested-by: Simon Horman <horms+renesas@verge.net.au> # Shmobile Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-06-30 22:29:12 +07:00
ret lr
ENDPROC(_find_first_bit_le)
/*
* Purpose : Find next 'one' bit
* Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset)
*/
ENTRY(_find_next_bit_le)
teq r1, #0
beq 3b
ands ip, r2, #7
beq 1b @ If new byte, goto old routine
ARM( ldrb r3, [r0, r2, lsr #3] )
THUMB( lsr r3, r2, #3 )
THUMB( ldrb r3, [r0, r3] )
movs r3, r3, lsr ip @ shift off unused bits
bne .L_found
orr r2, r2, #7 @ if zero, then no bits here
add r2, r2, #1 @ align bit pointer
b 2b @ loop for next bit
ENDPROC(_find_next_bit_le)
#ifdef __ARMEB__
ENTRY(_find_first_zero_bit_be)
teq r1, #0
beq 3f
mov r2, #0
1: eor r3, r2, #0x18 @ big endian byte ordering
ARM( ldrb r3, [r0, r3, lsr #3] )
THUMB( lsr r3, #3 )
THUMB( ldrb r3, [r0, r3] )
eors r3, r3, #0xff @ invert bits
bne .L_found @ any now set - found zero bit
add r2, r2, #8 @ next bit pointer
2: cmp r2, r1 @ any more?
blo 1b
3: mov r0, r1 @ no free bits
ARM: convert all "mov.* pc, reg" to "bx reg" for ARMv6+ ARMv6 and greater introduced a new instruction ("bx") which can be used to return from function calls. Recent CPUs perform better when the "bx lr" instruction is used rather than the "mov pc, lr" instruction, and this sequence is strongly recommended to be used by the ARM architecture manual (section A.4.1.1). We provide a new macro "ret" with all its variants for the condition code which will resolve to the appropriate instruction. Rather than doing this piecemeal, and miss some instances, change all the "mov pc" instances to use the new macro, with the exception of the "movs" instruction and the kprobes code. This allows us to detect the "mov pc, lr" case and fix it up - and also gives us the possibility of deploying this for other registers depending on the CPU selection. Reported-by: Will Deacon <will.deacon@arm.com> Tested-by: Stephen Warren <swarren@nvidia.com> # Tegra Jetson TK1 Tested-by: Robert Jarzmik <robert.jarzmik@free.fr> # mioa701_bootresume.S Tested-by: Andrew Lunn <andrew@lunn.ch> # Kirkwood Tested-by: Shawn Guo <shawn.guo@freescale.com> Tested-by: Tony Lindgren <tony@atomide.com> # OMAPs Tested-by: Gregory CLEMENT <gregory.clement@free-electrons.com> # Armada XP, 375, 385 Acked-by: Sekhar Nori <nsekhar@ti.com> # DaVinci Acked-by: Christoffer Dall <christoffer.dall@linaro.org> # kvm/hyp Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com> # PXA3xx Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> # Xen Tested-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> # ARMv7M Tested-by: Simon Horman <horms+renesas@verge.net.au> # Shmobile Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-06-30 22:29:12 +07:00
ret lr
ENDPROC(_find_first_zero_bit_be)
ENTRY(_find_next_zero_bit_be)
teq r1, #0
beq 3b
ands ip, r2, #7
beq 1b @ If new byte, goto old routine
eor r3, r2, #0x18 @ big endian byte ordering
ARM( ldrb r3, [r0, r3, lsr #3] )
THUMB( lsr r3, #3 )
THUMB( ldrb r3, [r0, r3] )
eor r3, r3, #0xff @ now looking for a 1 bit
movs r3, r3, lsr ip @ shift off unused bits
bne .L_found
orr r2, r2, #7 @ if zero, then no bits here
add r2, r2, #1 @ align bit pointer
b 2b @ loop for next bit
ENDPROC(_find_next_zero_bit_be)
ENTRY(_find_first_bit_be)
teq r1, #0
beq 3f
mov r2, #0
1: eor r3, r2, #0x18 @ big endian byte ordering
ARM( ldrb r3, [r0, r3, lsr #3] )
THUMB( lsr r3, #3 )
THUMB( ldrb r3, [r0, r3] )
movs r3, r3
bne .L_found @ any now set - found zero bit
add r2, r2, #8 @ next bit pointer
2: cmp r2, r1 @ any more?
blo 1b
3: mov r0, r1 @ no free bits
ARM: convert all "mov.* pc, reg" to "bx reg" for ARMv6+ ARMv6 and greater introduced a new instruction ("bx") which can be used to return from function calls. Recent CPUs perform better when the "bx lr" instruction is used rather than the "mov pc, lr" instruction, and this sequence is strongly recommended to be used by the ARM architecture manual (section A.4.1.1). We provide a new macro "ret" with all its variants for the condition code which will resolve to the appropriate instruction. Rather than doing this piecemeal, and miss some instances, change all the "mov pc" instances to use the new macro, with the exception of the "movs" instruction and the kprobes code. This allows us to detect the "mov pc, lr" case and fix it up - and also gives us the possibility of deploying this for other registers depending on the CPU selection. Reported-by: Will Deacon <will.deacon@arm.com> Tested-by: Stephen Warren <swarren@nvidia.com> # Tegra Jetson TK1 Tested-by: Robert Jarzmik <robert.jarzmik@free.fr> # mioa701_bootresume.S Tested-by: Andrew Lunn <andrew@lunn.ch> # Kirkwood Tested-by: Shawn Guo <shawn.guo@freescale.com> Tested-by: Tony Lindgren <tony@atomide.com> # OMAPs Tested-by: Gregory CLEMENT <gregory.clement@free-electrons.com> # Armada XP, 375, 385 Acked-by: Sekhar Nori <nsekhar@ti.com> # DaVinci Acked-by: Christoffer Dall <christoffer.dall@linaro.org> # kvm/hyp Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com> # PXA3xx Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> # Xen Tested-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> # ARMv7M Tested-by: Simon Horman <horms+renesas@verge.net.au> # Shmobile Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-06-30 22:29:12 +07:00
ret lr
ENDPROC(_find_first_bit_be)
ENTRY(_find_next_bit_be)
teq r1, #0
beq 3b
ands ip, r2, #7
beq 1b @ If new byte, goto old routine
eor r3, r2, #0x18 @ big endian byte ordering
ARM( ldrb r3, [r0, r3, lsr #3] )
THUMB( lsr r3, #3 )
THUMB( ldrb r3, [r0, r3] )
movs r3, r3, lsr ip @ shift off unused bits
bne .L_found
orr r2, r2, #7 @ if zero, then no bits here
add r2, r2, #1 @ align bit pointer
b 2b @ loop for next bit
ENDPROC(_find_next_bit_be)
#endif
/*
* One or more bits in the LSB of r3 are assumed to be set.
*/
.L_found:
#if __LINUX_ARM_ARCH__ >= 5
rsb r0, r3, #0
and r3, r3, r0
clz r3, r3
rsb r3, r3, #31
add r0, r2, r3
#else
tst r3, #0x0f
addeq r2, r2, #4
movne r3, r3, lsl #4
tst r3, #0x30
addeq r2, r2, #2
movne r3, r3, lsl #2
tst r3, #0x40
addeq r2, r2, #1
mov r0, r2
#endif
cmp r1, r0 @ Clamp to maxbit
movlo r0, r1
ARM: convert all "mov.* pc, reg" to "bx reg" for ARMv6+ ARMv6 and greater introduced a new instruction ("bx") which can be used to return from function calls. Recent CPUs perform better when the "bx lr" instruction is used rather than the "mov pc, lr" instruction, and this sequence is strongly recommended to be used by the ARM architecture manual (section A.4.1.1). We provide a new macro "ret" with all its variants for the condition code which will resolve to the appropriate instruction. Rather than doing this piecemeal, and miss some instances, change all the "mov pc" instances to use the new macro, with the exception of the "movs" instruction and the kprobes code. This allows us to detect the "mov pc, lr" case and fix it up - and also gives us the possibility of deploying this for other registers depending on the CPU selection. Reported-by: Will Deacon <will.deacon@arm.com> Tested-by: Stephen Warren <swarren@nvidia.com> # Tegra Jetson TK1 Tested-by: Robert Jarzmik <robert.jarzmik@free.fr> # mioa701_bootresume.S Tested-by: Andrew Lunn <andrew@lunn.ch> # Kirkwood Tested-by: Shawn Guo <shawn.guo@freescale.com> Tested-by: Tony Lindgren <tony@atomide.com> # OMAPs Tested-by: Gregory CLEMENT <gregory.clement@free-electrons.com> # Armada XP, 375, 385 Acked-by: Sekhar Nori <nsekhar@ti.com> # DaVinci Acked-by: Christoffer Dall <christoffer.dall@linaro.org> # kvm/hyp Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com> # PXA3xx Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> # Xen Tested-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> # ARMv7M Tested-by: Simon Horman <horms+renesas@verge.net.au> # Shmobile Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2014-06-30 22:29:12 +07:00
ret lr