[PATCHv1 8/8] unicore32 additional architecture files: low-level lib: misc
From: Guan Xuetao
Date: Mon Jan 03 2011 - 06:54:46 EST
From: Guan Xuetao <guanxuetao@xxxxxxxxxxxxxxx>
Patch 8 implements the rest low-level libraries.
Signed-off-by: Guan Xuetao <guanxuetao@xxxxxxxxxxxxxxx>
---
arch/unicore32/include/asm/assembler.h | 131 +++++++++++++++++++++
arch/unicore32/include/asm/bitops.h | 47 ++++++++
arch/unicore32/include/asm/delay.h | 52 ++++++++
arch/unicore32/include/asm/futex.h | 143 +++++++++++++++++++++++
arch/unicore32/include/asm/mutex.h | 20 +++
arch/unicore32/include/asm/swab.h | 28 +++++
arch/unicore32/lib/Makefile | 16 +++
arch/unicore32/lib/delay.S | 51 ++++++++
arch/unicore32/lib/findbit.S | 98 ++++++++++++++++
arch/unicore32/lib/sha1.S | 200 ++++++++++++++++++++++++++++++++
10 files changed, 786 insertions(+), 0 deletions(-)
diff --git a/arch/unicore32/include/asm/assembler.h b/arch/unicore32/include/asm/assembler.h
new file mode 100644
index 0000000..8e87ed7
--- /dev/null
+++ b/arch/unicore32/include/asm/assembler.h
@@ -0,0 +1,131 @@
+/*
+ * linux/arch/unicore32/include/asm/assembler.h
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * 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.
+ *
+ * Do not include any C declarations in this file - it is included by
+ * assembler source.
+ */
+#ifndef __ASSEMBLY__
+#error "Only include this from assembly code"
+#endif
+
+#include <asm/ptrace.h>
+
+/*
+ * Little Endian independent macros for shifting bytes within registers.
+ */
+#define pull >>
+#define push <<
+#define get_byte_0 << #0
+#define get_byte_1 >> #8
+#define get_byte_2 >> #16
+#define get_byte_3 >> #24
+#define put_byte_0 << #0
+#define put_byte_1 << #8
+#define put_byte_2 << #16
+#define put_byte_3 << #24
+
+#define cadd cmpadd
+#define cand cmpand
+#define csub cmpsub
+#define cxor cmpxor
+
+/*
+ * Enable and disable interrupts
+ */
+ .macro disable_irq, temp
+ mov \temp, asr
+ andn \temp, \temp, #0xFF
+ or \temp, \temp, #PSR_I_BIT | PRIV_MODE
+ mov.a asr, \temp
+ .endm
+
+ .macro enable_irq, temp
+ mov \temp, asr
+ andn \temp, \temp, #0xFF
+ or \temp, \temp, #PRIV_MODE
+ mov.a asr, \temp
+ .endm
+
+#define USER(x...) \
+9999: x; \
+ .pushsection __ex_table, "a"; \
+ .align 3; \
+ .long 9999b, 9001f; \
+ .popsection
+
+ .macro notcond, cond, nexti = .+8
+ .ifc \cond, eq
+ bne \nexti
+ .else; .ifc \cond, ne
+ beq \nexti
+ .else; .ifc \cond, ea
+ bub \nexti
+ .else; .ifc \cond, ub
+ bea \nexti
+ .else; .ifc \cond, fs
+ bns \nexti
+ .else; .ifc \cond, ns
+ bfs \nexti
+ .else; .ifc \cond, fv
+ bnv \nexti
+ .else; .ifc \cond, nv
+ bfv \nexti
+ .else; .ifc \cond, ua
+ beb \nexti
+ .else; .ifc \cond, eb
+ bua \nexti
+ .else; .ifc \cond, eg
+ bsl \nexti
+ .else; .ifc \cond, sl
+ beg \nexti
+ .else; .ifc \cond, sg
+ bel \nexti
+ .else; .ifc \cond, el
+ bsg \nexti
+ .else; .ifnc \cond, al
+ .error "Unknown cond in notcond macro argument"
+ .endif; .endif; .endif; .endif; .endif; .endif; .endif
+ .endif; .endif; .endif; .endif; .endif; .endif; .endif
+ .endif
+ .endm
+
+ .macro usracc, instr, reg, ptr, inc, cond, rept, abort
+ .rept \rept
+ notcond \cond, .+8
+9999 :
+ .if \inc == 1
+ \instr\()b.u \reg, [\ptr], #\inc
+ .elseif \inc == 4
+ \instr\()w.u \reg, [\ptr], #\inc
+ .else
+ .error "Unsupported inc macro argument"
+ .endif
+
+ .pushsection __ex_table, "a"
+ .align 3
+ .long 9999b, \abort
+ .popsection
+ .endr
+ .endm
+
+ .macro strusr, reg, ptr, inc, cond = al, rept = 1, abort = 9001f
+ usracc st, \reg, \ptr, \inc, \cond, \rept, \abort
+ .endm
+
+ .macro ldrusr, reg, ptr, inc, cond = al, rept = 1, abort = 9001f
+ usracc ld, \reg, \ptr, \inc, \cond, \rept, \abort
+ .endm
+
+ .macro nop8
+ .rept 8
+ nop
+ .endr
+ .endm
diff --git a/arch/unicore32/include/asm/bitops.h b/arch/unicore32/include/asm/bitops.h
new file mode 100644
index 0000000..1628a63
--- /dev/null
+++ b/arch/unicore32/include/asm/bitops.h
@@ -0,0 +1,47 @@
+/*
+ * linux/arch/unicore32/include/asm/bitops.h
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * 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.
+ */
+
+#ifndef __UNICORE_BITOPS_H__
+#define __UNICORE_BITOPS_H__
+
+#define find_next_bit __uc32_find_next_bit
+#define find_next_zero_bit __uc32_find_next_zero_bit
+
+#define find_first_bit __uc32_find_first_bit
+#define find_first_zero_bit __uc32_find_first_zero_bit
+
+#define _ASM_GENERIC_BITOPS_FLS_H_
+#define _ASM_GENERIC_BITOPS___FLS_H_
+#define _ASM_GENERIC_BITOPS_FFS_H_
+#define _ASM_GENERIC_BITOPS___FFS_H_
+/*
+ * On UNICORE, those functions can be implemented around
+ * the cntlz instruction for much better code efficiency.
+ */
+
+static inline int fls(int x)
+{
+ int ret;
+
+ asm("cntlz\t%0, %1" : "=r" (ret) : "r" (x) : "cc");
+ ret = 32 - ret;
+
+ return ret;
+}
+
+#define __fls(x) (fls(x) - 1)
+#define ffs(x) ({ unsigned long __t = (x); fls(__t & -__t); })
+#define __ffs(x) (ffs(x) - 1)
+
+#include <asm-generic/bitops.h>
+
+#endif /* __UNICORE_BITOPS_H__ */
diff --git a/arch/unicore32/include/asm/delay.h b/arch/unicore32/include/asm/delay.h
new file mode 100644
index 0000000..164ae61
--- /dev/null
+++ b/arch/unicore32/include/asm/delay.h
@@ -0,0 +1,52 @@
+/*
+ * linux/arch/unicore32/include/asm/delay.h
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * 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.
+ *
+ * Delay routines, using a pre-computed "loops_per_second" value.
+ */
+#ifndef __UNICORE_DELAY_H__
+#define __UNICORE_DELAY_H__
+
+#include <asm/param.h> /* HZ */
+
+extern void __delay(int loops);
+
+/*
+ * This function intentionally does not exist; if you see references to
+ * it, it means that you're calling udelay() with an out of range value.
+ *
+ * With currently imposed limits, this means that we support a max delay
+ * of 2000us. Further limits: HZ<=1000 and bogomips<=3355
+ */
+extern void __bad_udelay(void);
+
+/*
+ * division by multiplication: you don't have to worry about
+ * loss of precision.
+ *
+ * Use only for very small delays ( < 1 msec). Should probably use a
+ * lookup table, really, as the multiplications take much too long with
+ * short delays. This is a "reasonable" implementation, though (and the
+ * first constant multiplications gets optimized away if the delay is
+ * a constant)
+ */
+extern void __udelay(unsigned long usecs);
+extern void __const_udelay(unsigned long);
+
+#define MAX_UDELAY_MS 2
+
+#define udelay(n) \
+ (__builtin_constant_p(n) ? \
+ ((n) > (MAX_UDELAY_MS * 1000) ? __bad_udelay() : \
+ __const_udelay((n) * ((2199023U*HZ)>>11))) : \
+ __udelay(n))
+
+#endif /* __UNICORE_DELAY_H__ */
+
diff --git a/arch/unicore32/include/asm/futex.h b/arch/unicore32/include/asm/futex.h
new file mode 100644
index 0000000..07dea61
--- /dev/null
+++ b/arch/unicore32/include/asm/futex.h
@@ -0,0 +1,143 @@
+/*
+ * linux/arch/unicore32/include/asm/futex.h
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * 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.
+ */
+
+#ifndef __UNICORE_FUTEX_H__
+#define __UNICORE_FUTEX_H__
+
+#ifdef __KERNEL__
+
+#include <linux/futex.h>
+#include <linux/preempt.h>
+#include <linux/uaccess.h>
+#include <linux/errno.h>
+
+#define __futex_atomic_op(insn, ret, oldval, uaddr, oparg) \
+ __asm__ __volatile__( \
+ "1: ldw.u %1, [%2]\n" \
+ " " insn "\n" \
+ "2: stw.u %0, [%2]\n" \
+ " mov %0, #0\n" \
+ "3:\n" \
+ " .pushsection __ex_table,\"a\"\n" \
+ " .align 3\n" \
+ " .long 1b, 4f, 2b, 4f\n" \
+ " .popsection\n" \
+ " .pushsection .fixup,\"ax\"\n" \
+ "4: mov %0, %4\n" \
+ " b 3b\n" \
+ " .popsection" \
+ : "=&r" (ret), "=&r" (oldval) \
+ : "r" (uaddr), "r" (oparg), "Ir" (-EFAULT) \
+ : "cc", "memory")
+
+static inline int
+futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
+{
+ int op = (encoded_op >> 28) & 7;
+ int cmp = (encoded_op >> 24) & 15;
+ int oparg = (encoded_op << 8) >> 20;
+ int cmparg = (encoded_op << 20) >> 20;
+ int oldval = 0, ret;
+
+ if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
+ oparg = 1 << oparg;
+
+ if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
+ return -EFAULT;
+
+ pagefault_disable(); /* implies preempt_disable() */
+
+ switch (op) {
+ case FUTEX_OP_SET:
+ __futex_atomic_op("mov %0, %3", ret, oldval, uaddr, oparg);
+ break;
+ case FUTEX_OP_ADD:
+ __futex_atomic_op("add %0, %1, %3", ret, oldval, uaddr, oparg);
+ break;
+ case FUTEX_OP_OR:
+ __futex_atomic_op("or %0, %1, %3", ret, oldval, uaddr, oparg);
+ break;
+ case FUTEX_OP_ANDN:
+ __futex_atomic_op("and %0, %1, %3",
+ ret, oldval, uaddr, ~oparg);
+ break;
+ case FUTEX_OP_XOR:
+ __futex_atomic_op("xor %0, %1, %3", ret, oldval, uaddr, oparg);
+ break;
+ default:
+ ret = -ENOSYS;
+ }
+
+ pagefault_enable(); /* subsumes preempt_enable() */
+
+ if (!ret) {
+ switch (cmp) {
+ case FUTEX_OP_CMP_EQ:
+ ret = (oldval == cmparg);
+ break;
+ case FUTEX_OP_CMP_NE:
+ ret = (oldval != cmparg);
+ break;
+ case FUTEX_OP_CMP_LT:
+ ret = (oldval < cmparg);
+ break;
+ case FUTEX_OP_CMP_GE:
+ ret = (oldval >= cmparg);
+ break;
+ case FUTEX_OP_CMP_LE:
+ ret = (oldval <= cmparg);
+ break;
+ case FUTEX_OP_CMP_GT:
+ ret = (oldval > cmparg);
+ break;
+ default:
+ ret = -ENOSYS;
+ }
+ }
+ return ret;
+}
+
+static inline int
+futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
+{
+ int val;
+
+ if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
+ return -EFAULT;
+
+ pagefault_disable(); /* implies preempt_disable() */
+
+ __asm__ __volatile__("@futex_atomic_cmpxchg_inatomic\n"
+ "1: ldw.u %0, [%3]\n"
+ " cmpxor.a %0, %1\n"
+ " bne 3f\n"
+ "2: stw.u %2, [%3]\n"
+ "3:\n"
+ " .pushsection __ex_table,\"a\"\n"
+ " .align 3\n"
+ " .long 1b, 4f, 2b, 4f\n"
+ " .popsection\n"
+ " .pushsection .fixup,\"ax\"\n"
+ "4: mov %0, %4\n"
+ " b 3b\n"
+ " .popsection"
+ : "=&r" (val)
+ : "r" (oldval), "r" (newval), "r" (uaddr), "Ir" (-EFAULT)
+ : "cc", "memory");
+
+ pagefault_enable(); /* subsumes preempt_enable() */
+
+ return val;
+}
+
+#endif /* __KERNEL__ */
+#endif /* __UNICORE_FUTEX_H__ */
diff --git a/arch/unicore32/include/asm/mutex.h b/arch/unicore32/include/asm/mutex.h
new file mode 100644
index 0000000..fab7d0e
--- /dev/null
+++ b/arch/unicore32/include/asm/mutex.h
@@ -0,0 +1,20 @@
+/*
+ * linux/arch/unicore32/include/asm/mutex.h
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * 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.
+ *
+ * UniCore optimized mutex locking primitives
+ *
+ * Please look into asm-generic/mutex-xchg.h for a formal definition.
+ */
+#ifndef __UNICORE_MUTEX_H__
+#define __UNICORE_MUTEX_H__
+
+# include <asm-generic/mutex-xchg.h>
+#endif
diff --git a/arch/unicore32/include/asm/swab.h b/arch/unicore32/include/asm/swab.h
new file mode 100644
index 0000000..8f0521f
--- /dev/null
+++ b/arch/unicore32/include/asm/swab.h
@@ -0,0 +1,28 @@
+/*
+ * linux/arch/unicore32/include/asm/swab.h
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * 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.
+ *
+ * In little endian mode, the data bus is connected such
+ * that byte accesses appear as:
+ * 0 = d0...d7, 1 = d8...d15, 2 = d16...d23, 3 = d24...d31
+ * and word accesses (data or instruction) appear as:
+ * d0...d31
+ */
+#ifndef __UNICORE_SWAB_H__
+#define __UNICORE_SWAB_H__
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+#include <asm-generic/swab.h>
+
+#define __arch_swab32(x) __builtin_bswap32(x)
+
+#endif
+
diff --git a/arch/unicore32/lib/Makefile b/arch/unicore32/lib/Makefile
new file mode 100644
index 0000000..f96b55c
--- /dev/null
+++ b/arch/unicore32/lib/Makefile
@@ -0,0 +1,16 @@
+#
+# linux/arch/unicore32/lib/Makefile
+#
+# Copyright (C) 2001-2010 GUAN Xue-tao
+#
+
+lib-y := backtrace.o delay.o findbit.o sha1.o \
+ csumipv6.o csumpartial.o csumpartialcopy.o csumpartialcopyuser.o \
+ strncpy_from_user.o strnlen_user.o \
+ io-readsb.o io-writesb.o io-readsl.o \
+ io-writesl.o io-readsw.o io-writesw.o \
+ clear_user.o copy_page.o getuser.o putuser.o \
+ copy_from_user.o copy_to_user.o
+
+$(obj)/csumpartialcopy.o: $(obj)/csumpartialcopygeneric.S
+$(obj)/csumpartialcopyuser.o: $(obj)/csumpartialcopygeneric.S
diff --git a/arch/unicore32/lib/delay.S b/arch/unicore32/lib/delay.S
new file mode 100644
index 0000000..24664c0
--- /dev/null
+++ b/arch/unicore32/lib/delay.S
@@ -0,0 +1,51 @@
+/*
+ * linux/arch/unicore32/lib/delay.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * 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.
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/param.h>
+ .text
+
+.LC0: .word loops_per_jiffy
+.LC1: .word (2199023*HZ)>>11
+
+/*
+ * r0 <= 2000
+ * lpj <= 0x01ffffff (max. 3355 bogomips)
+ * HZ <= 1000
+ */
+
+ENTRY(__udelay)
+ ldw r2, .LC1
+ mul r0, r2, r0
+ENTRY(__const_udelay) @ 0 <= r0 <= 0x7fffff06
+ ldw r2, .LC0
+ ldw r2, [r2] @ max = 0x01ffffff
+ mov r0, r0 >> #14 @ max = 0x0001ffff
+ mov r2, r2 >> #10 @ max = 0x00007fff
+ mul r0, r2, r0 @ max = 2^32-1
+ mov.a r0, r0 >> #6
+ cmoveq pc, lr
+
+/*
+ * loops = r0 * HZ * loops_per_jiffy / 1000000
+ *
+ * Oh, if only we had a cycle counter...
+ */
+
+@ Delay routine
+ENTRY(__delay)
+ sub.a r0, r0, #2
+ bua __delay
+ mov pc, lr
+ENDPROC(__udelay)
+ENDPROC(__const_udelay)
+ENDPROC(__delay)
diff --git a/arch/unicore32/lib/findbit.S b/arch/unicore32/lib/findbit.S
new file mode 100644
index 0000000..c360ce9
--- /dev/null
+++ b/arch/unicore32/lib/findbit.S
@@ -0,0 +1,98 @@
+/*
+ * linux/arch/unicore32/lib/findbit.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * 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.
+ */
+#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);
+ */
+__uc32_find_first_zero_bit:
+ cxor.a r1, #0
+ beq 3f
+ mov r2, #0
+1: ldb r3, [r0+], r2 >> #3
+ xor.a r3, r3, #0xff @ invert bits
+ bne .L_found @ any now set - found zero bit
+ add r2, r2, #8 @ next bit pointer
+2: csub.a r2, r1 @ any more?
+ bub 1b
+3: mov r0, r1 @ no free bits
+ mov pc, lr
+
+/*
+ * Purpose : Find next 'zero' bit
+ * Prototype: int find_next_zero_bit
+ * (void *addr, unsigned int maxbit, int offset)
+ */
+ENTRY(__uc32_find_next_zero_bit)
+ cxor.a r1, #0
+ beq 3b
+ and.a ip, r2, #7
+ beq 1b @ If new byte, goto old routine
+ ldb r3, [r0+], r2 >> #3
+ xor r3, r3, #0xff @ now looking for a 1 bit
+ mov.a r3, r3 >> ip @ shift off unused bits
+ bne .L_found
+ or r2, r2, #7 @ if zero, then no bits here
+ add r2, r2, #1 @ align bit pointer
+ b 2b @ loop for next bit
+ENDPROC(__uc32_find_next_zero_bit)
+
+/*
+ * Purpose : Find a 'one' bit
+ * Prototype: int find_first_bit
+ * (const unsigned long *addr, unsigned int maxbit);
+ */
+__uc32_find_first_bit:
+ cxor.a r1, #0
+ beq 3f
+ mov r2, #0
+1: ldb r3, [r0+], r2 >> #3
+ mov.a r3, r3
+ bne .L_found @ any now set - found zero bit
+ add r2, r2, #8 @ next bit pointer
+2: csub.a r2, r1 @ any more?
+ bub 1b
+3: mov r0, r1 @ no free bits
+ mov pc, lr
+
+/*
+ * Purpose : Find next 'one' bit
+ * Prototype: int find_next_zero_bit
+ * (void *addr, unsigned int maxbit, int offset)
+ */
+ENTRY(__uc32_find_next_bit)
+ cxor.a r1, #0
+ beq 3b
+ and.a ip, r2, #7
+ beq 1b @ If new byte, goto old routine
+ ldb r3, [r0+], r2 >> #3
+ mov.a r3, r3 >> ip @ shift off unused bits
+ bne .L_found
+ or r2, r2, #7 @ if zero, then no bits here
+ add r2, r2, #1 @ align bit pointer
+ b 2b @ loop for next bit
+ENDPROC(__uc32_find_next_bit)
+
+/*
+ * One or more bits in the LSB of r3 are assumed to be set.
+ */
+.L_found:
+ rsub r1, r3, #0
+ and r3, r3, r1
+ cntlz r3, r3
+ rsub r3, r3, #31
+ add r0, r2, r3
+ mov pc, lr
+
diff --git a/arch/unicore32/lib/sha1.S b/arch/unicore32/lib/sha1.S
new file mode 100644
index 0000000..0c50895
--- /dev/null
+++ b/arch/unicore32/lib/sha1.S
@@ -0,0 +1,200 @@
+/*
+ * linux/arch/unicore32/lib/sha1.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * 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.
+ *
+ * The reference implementation for this code is linux/lib/sha1.c
+ */
+
+#include <linux/linkage.h>
+
+ .text
+
+
+/*
+ * void sha_transform(__u32 *digest, const char *in, __u32 *W)
+ *
+ * Note: the "in" ptr may be unaligned.
+ */
+
+ENTRY(sha_transform)
+
+ stm.w (lr), [sp-]
+
+ @ for (i = 0; i < 16; i++)
+ @ W[i] = be32_to_cpu(in[i]);
+
+ mov r3, r2
+ mov lr, #16
+1: ldb.w r4, [r1]+, #1
+ ldb.w r5, [r1]+, #1
+ ldb.w r6, [r1]+, #1
+ ldb.w r7, [r1]+, #1
+ sub.a lr, lr, #1
+ or r5, r5, r4 << #8
+ or r6, r6, r5 << #8
+ or r7, r7, r6 << #8
+ stw.w r7, [r3]+, #4
+ bne 1b
+
+ @ for (i = 0; i < 64; i++)
+ @ W[i+16] = ror(W[i+13] ^ W[i+8] ^ W[i+2] ^ W[i], 31);
+
+ sub r3, r2, #4
+ mov lr, #64
+2: ldw.w r4, [r3+], #4
+ sub.a lr, lr, #1
+ ldw r5, [r3+], #8
+ ldw r6, [r3+], #32
+ ldw r7, [r3+], #52
+ xor r4, r4, r5
+ xor r4, r4, r6
+ xor r4, r4, r7
+ mov r4, r4 <> #31
+ stw r4, [r3+], #64
+ bne 2b
+
+ /*
+ * The SHA functions are:
+ *
+ * f1(B,C,D) = (D ^ (B & (C ^ D)))
+ * f2(B,C,D) = (B ^ C ^ D)
+ * f3(B,C,D) = ((B & C) | (D & (B | C)))
+ *
+ * Then the sub-blocks are processed as follows:
+ *
+ * A' = ror(A, 27) + f(B,C,D) + E + K + *W++
+ * B' = A
+ * C' = ror(B, 2)
+ * D' = C
+ * E' = D
+ *
+ * We therefore unroll each loop 5 times to avoid register shuffling.
+ * Also the ror for C (and also D and E which are successivelyderived
+ * from it) is applied in place to cut on an additional mov insn for
+ * each round.
+ */
+
+ .macro sha_f1, A, B, C, D, E
+ ldw.w r3, [r2]+, #4
+ xor ip, \C, \D
+ add \E, r1, \E <> #2
+ and ip, \B, ip <> #2
+ add \E, \E, \A <> #27
+ xor ip, ip, \D <> #2
+ add \E, \E, r3
+ add \E, \E, ip
+ .endm
+
+ .macro sha_f2, A, B, C, D, E
+ ldw.w r3, [r2]+, #4
+ add \E, r1, \E <> #2
+ xor ip, \B, \C <> #2
+ add \E, \E, \A <> #27
+ xor ip, ip, \D <> #2
+ add \E, \E, r3
+ add \E, \E, ip
+ .endm
+
+ .macro sha_f3, A, B, C, D, E
+ ldw.w r3, [r2]+, #4
+ add \E, r1, \E <> #2
+ or ip, \B, \C <> #2
+ add \E, \E, \A <> #27
+ and ip, ip, \D <> #2
+ add \E, \E, r3
+ and r3, \B, \C <> #2
+ or ip, ip, r3
+ add \E, \E, ip
+ .endm
+
+ ldm (r4 - r8), [r0]+
+
+ mov lr, #4
+ ldw r1, .L_sha_K + 0
+
+ /* adjust initial values */
+ mov r6, r6 <> #30
+ mov r7, r7 <> #30
+ mov r8, r8 <> #30
+
+3: sub.a lr, lr, #1
+ sha_f1 r4, r5, r6, r7, r8
+ sha_f1 r8, r4, r5, r6, r7
+ sha_f1 r7, r8, r4, r5, r6
+ sha_f1 r6, r7, r8, r4, r5
+ sha_f1 r5, r6, r7, r8, r4
+ bne 3b
+
+ ldw r1, .L_sha_K + 4
+ mov lr, #4
+
+4: sub.a lr, lr, #1
+ sha_f2 r4, r5, r6, r7, r8
+ sha_f2 r8, r4, r5, r6, r7
+ sha_f2 r7, r8, r4, r5, r6
+ sha_f2 r6, r7, r8, r4, r5
+ sha_f2 r5, r6, r7, r8, r4
+ bne 4b
+
+ ldw r1, .L_sha_K + 8
+ mov lr, #4
+
+5: sub.a lr, lr, #1
+ sha_f3 r4, r5, r6, r7, r8
+ sha_f3 r8, r4, r5, r6, r7
+ sha_f3 r7, r8, r4, r5, r6
+ sha_f3 r6, r7, r8, r4, r5
+ sha_f3 r5, r6, r7, r8, r4
+ bne 5b
+
+ ldw r1, .L_sha_K + 12
+ mov lr, #4
+
+6: sub.a lr, lr, #1
+ sha_f2 r4, r5, r6, r7, r8
+ sha_f2 r8, r4, r5, r6, r7
+ sha_f2 r7, r8, r4, r5, r6
+ sha_f2 r6, r7, r8, r4, r5
+ sha_f2 r5, r6, r7, r8, r4
+ bne 6b
+
+ ldm (r1, r2, r3, r9, r10), [r0]+
+ add r4, r1, r4
+ add r5, r2, r5
+ add r6, r3, r6 <> #2
+ add r7, r9, r7 <> #2
+ add r8, r10, r8 <> #2
+ stm (r4 - r8), [r0]+
+
+ ldm.w (pc), [sp]+
+
+ENDPROC(sha_transform)
+
+ .align 2
+.L_sha_K:
+ .word 0x5a827999, 0x6ed9eba1, 0x8f1bbcdc, 0xca62c1d6
+
+
+/*
+ * void sha_init(__u32 *buf)
+ */
+
+ .align 2
+.L_sha_initial_digest:
+ .word 0x67452301, 0xefcdab89, 0x98badcfe, 0x10325476, 0xc3d2e1f0
+
+ENTRY(sha_init)
+
+ adr r1, .L_sha_initial_digest
+ ldm (r1 - r5), [r1]+
+ stm (r1 - r5), [r0]+
+ mov pc, lr
+
+ENDPROC(sha_init)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/