x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

When using eBPF to probe CS CALL instructions within a function,
a crash can be triggered.

The eBPF tool probes offset 257 of the __hrtimer_run_queues()
function:

<__hrtimer_run_queues+249>:  nopl   0x0(%rax,%rax,1)
<__hrtimer_run_queues+254>:  mov    %r14,%rdi
<__hrtimer_run_queues+257>:  cs call <__x86_indirect_thunk_r12>
<__hrtimer_run_queues+263>:  mov    %eax,%r12d
<__hrtimer_run_queues+266>:  xchg   %ax,%ax
<__hrtimer_run_queues+268>:  mov    %r13,%rdi

Which triggers this crash:

  BUG: unable to handle page fault for address: 00000000000f41c9
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 0 P4D 0
  Oops: 0002 [#1] SMP NOPTI
  CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P
  RIP: 0010:__hrtimer_run_queues+0x106/0x230

Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is
at the 6th byte of the above CS CALL instruction. Since the CS CALL
instruction occupies 6 bytes, the exception occurred in the middle of that
call instruction.

The root cause is that when using eBPF tools to probe in the middle of a
function, a kprobe with INT3 is used as the underlying implementation.

During single-step emulation of the original CALL instruction,
int3_emulate_call() assumes that the probed CALL instruction is 5 bytes
long. However, the actual CS-prefixed CALL instruction occupies 6 bytes,
so it constructs an incorrect exception return address. When the CPU
returns from the kprobe handler, the next instruction to be executed is at
the address of the last byte of that CS CALL instruction. Coincidentally,
starting from that address, the CPU fetches and decodes a completely
different instruction, which ultimately triggers a kernel crash.

Fix the issue by using the actual instruction length obtained from
the instruction decoder when constructing the exception return
address, rather than relying on the hardcoded CALL_INSN_SIZE macro.

[ mingo: Refined the changelog ]

Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step")
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

21 files changed, 9945 insertions, 0 deletions

diff --git a/lib/crypto/powerpc/.gitignore b/lib/crypto/powerpc/.gitignore
new file mode 100644
index 000000000..7aa71d83f
--- /dev/null
+++ b/lib/crypto/powerpc/.gitignore
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0-only
+aesp8-ppc.S
+ghashp8-ppc.S
diff --git a/lib/crypto/powerpc/aes-spe-core.S b/lib/crypto/powerpc/aes-spe-core.S
new file mode 100644
index 000000000..8e00eccc3
--- /dev/null
+++ b/lib/crypto/powerpc/aes-spe-core.S
@@ -0,0 +1,346 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Fast AES implementation for SPE instruction set (PPC)
+ *
+ * This code makes use of the SPE SIMD instruction set as defined in
+ * http://cache.freescale.com/files/32bit/doc/ref_manual/SPEPIM.pdf
+ * Implementation is based on optimization guide notes from
+ * http://cache.freescale.com/files/32bit/doc/app_note/AN2665.pdf
+ *
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ */
+
+#include <asm/ppc_asm.h>
+#include "aes-spe-regs.h"
+
+#define	EAD(in, bpos) \
+	rlwimi		rT0,in,28-((bpos+3)%4)*8,20,27;
+
+#define DAD(in, bpos) \
+	rlwimi		rT1,in,24-((bpos+3)%4)*8,24,31;
+
+#define LWH(out, off) \
+	evlwwsplat	out,off(rT0);	/* load word high		*/
+
+#define LWL(out, off) \
+	lwz		out,off(rT0);	/* load word low		*/
+
+#define LBZ(out, tab, off) \
+	lbz		out,off(tab);	/* load byte			*/
+
+#define LAH(out, in, bpos, off) \
+	EAD(in, bpos)			/* calc addr + load word high	*/ \
+	LWH(out, off)
+
+#define LAL(out, in, bpos, off) \
+	EAD(in, bpos)			/* calc addr + load word low	*/ \
+	LWL(out, off)
+
+#define LAE(out, in, bpos) \
+	EAD(in, bpos)			/* calc addr + load enc byte	*/ \
+	LBZ(out, rT0, 8)
+
+#define LBE(out) \
+	LBZ(out, rT0, 8)		/* load enc byte		*/
+
+#define LAD(out, in, bpos) \
+	DAD(in, bpos)			/* calc addr + load dec byte	*/ \
+	LBZ(out, rT1, 0)
+
+#define LBD(out) \
+	LBZ(out, rT1, 0)
+
+/*
+ * ppc_encrypt_block: The central encryption function for a single 16 bytes
+ * block. It does no stack handling or register saving to support fast calls
+ * via bl/blr. It expects that caller has pre-xored input data with first
+ * 4 words of encryption key into rD0-rD3. Pointer/counter registers must
+ * have also been set up before (rT0, rKP, CTR). Output is stored in rD0-rD3
+ * and rW0-rW3 and caller must execute a final xor on the output registers.
+ * All working registers rD0-rD3 & rW0-rW7 are overwritten during processing.
+ *
+ */
+_GLOBAL(ppc_encrypt_block)
+	LAH(rW4, rD1, 2, 4)
+	LAH(rW6, rD0, 3, 0)
+	LAH(rW3, rD0, 1, 8)
+ppc_encrypt_block_loop:
+	LAH(rW0, rD3, 0, 12)
+	LAL(rW0, rD0, 0, 12)
+	LAH(rW1, rD1, 0, 12)
+	LAH(rW2, rD2, 1, 8)
+	LAL(rW2, rD3, 1, 8)
+	LAL(rW3, rD1, 1, 8)
+	LAL(rW4, rD2, 2, 4)
+	LAL(rW6, rD1, 3, 0)
+	LAH(rW5, rD3, 2, 4)
+	LAL(rW5, rD0, 2, 4)
+	LAH(rW7, rD2, 3, 0)
+	evldw		rD1,16(rKP)
+	EAD(rD3, 3)
+	evxor		rW2,rW2,rW4
+	LWL(rW7, 0)
+	evxor		rW2,rW2,rW6
+	EAD(rD2, 0)
+	evxor		rD1,rD1,rW2
+	LWL(rW1, 12)
+	evxor		rD1,rD1,rW0
+	evldw		rD3,24(rKP)
+	evmergehi	rD0,rD0,rD1
+	EAD(rD1, 2)
+	evxor		rW3,rW3,rW5
+	LWH(rW4, 4)
+	evxor		rW3,rW3,rW7
+	EAD(rD0, 3)
+	evxor		rD3,rD3,rW3
+	LWH(rW6, 0)
+	evxor		rD3,rD3,rW1
+	EAD(rD0, 1)
+	evmergehi	rD2,rD2,rD3
+	LWH(rW3, 8)
+	LAH(rW0, rD3, 0, 12)
+	LAL(rW0, rD0, 0, 12)
+	LAH(rW1, rD1, 0, 12)
+	LAH(rW2, rD2, 1, 8)
+	LAL(rW2, rD3, 1, 8)
+	LAL(rW3, rD1, 1, 8)
+	LAL(rW4, rD2, 2, 4)
+	LAL(rW6, rD1, 3, 0)
+	LAH(rW5, rD3, 2, 4)
+	LAL(rW5, rD0, 2, 4)
+	LAH(rW7, rD2, 3, 0)
+	evldw		rD1,32(rKP)
+	EAD(rD3, 3)
+	evxor		rW2,rW2,rW4
+	LWL(rW7, 0)
+	evxor		rW2,rW2,rW6
+	EAD(rD2, 0)
+	evxor		rD1,rD1,rW2
+	LWL(rW1, 12)
+	evxor		rD1,rD1,rW0
+	evldw		rD3,40(rKP)
+	evmergehi	rD0,rD0,rD1
+	EAD(rD1, 2)
+	evxor		rW3,rW3,rW5
+	LWH(rW4, 4)
+	evxor		rW3,rW3,rW7
+	EAD(rD0, 3)
+	evxor		rD3,rD3,rW3
+	LWH(rW6, 0)
+	evxor		rD3,rD3,rW1
+	EAD(rD0, 1)
+	evmergehi	rD2,rD2,rD3
+	LWH(rW3, 8)
+	addi		rKP,rKP,32
+	bdnz		ppc_encrypt_block_loop
+	LAH(rW0, rD3, 0, 12)
+	LAL(rW0, rD0, 0, 12)
+	LAH(rW1, rD1, 0, 12)
+	LAH(rW2, rD2, 1, 8)
+	LAL(rW2, rD3, 1, 8)
+	LAL(rW3, rD1, 1, 8)
+	LAL(rW4, rD2, 2, 4)
+	LAH(rW5, rD3, 2, 4)
+	LAL(rW6, rD1, 3, 0)
+	LAL(rW5, rD0, 2, 4)
+	LAH(rW7, rD2, 3, 0)
+	evldw		rD1,16(rKP)
+	EAD(rD3, 3)
+	evxor		rW2,rW2,rW4
+	LWL(rW7, 0)
+	evxor		rW2,rW2,rW6
+	EAD(rD2, 0)
+	evxor		rD1,rD1,rW2
+	LWL(rW1, 12)
+	evxor		rD1,rD1,rW0
+	evldw		rD3,24(rKP)
+	evmergehi	rD0,rD0,rD1
+	EAD(rD1, 0)
+	evxor		rW3,rW3,rW5
+	LBE(rW2)
+	evxor		rW3,rW3,rW7
+	EAD(rD0, 1)
+	evxor		rD3,rD3,rW3
+	LBE(rW6)
+	evxor		rD3,rD3,rW1
+	EAD(rD0, 0)
+	evmergehi	rD2,rD2,rD3
+	LBE(rW1)
+	LAE(rW0, rD3, 0)
+	LAE(rW1, rD0, 0)
+	LAE(rW4, rD2, 1)
+	LAE(rW5, rD3, 1)
+	LAE(rW3, rD2, 0)
+	LAE(rW7, rD1, 1)
+	rlwimi		rW0,rW4,8,16,23
+	rlwimi		rW1,rW5,8,16,23
+	LAE(rW4, rD1, 2)
+	LAE(rW5, rD2, 2)
+	rlwimi		rW2,rW6,8,16,23
+	rlwimi		rW3,rW7,8,16,23
+	LAE(rW6, rD3, 2)
+	LAE(rW7, rD0, 2)
+	rlwimi		rW0,rW4,16,8,15
+	rlwimi		rW1,rW5,16,8,15
+	LAE(rW4, rD0, 3)
+	LAE(rW5, rD1, 3)
+	rlwimi		rW2,rW6,16,8,15
+	lwz		rD0,32(rKP)
+	rlwimi		rW3,rW7,16,8,15
+	lwz		rD1,36(rKP)
+	LAE(rW6, rD2, 3)
+	LAE(rW7, rD3, 3)
+	rlwimi		rW0,rW4,24,0,7
+	lwz		rD2,40(rKP)
+	rlwimi		rW1,rW5,24,0,7
+	lwz		rD3,44(rKP)
+	rlwimi		rW2,rW6,24,0,7
+	rlwimi		rW3,rW7,24,0,7
+	blr
+
+/*
+ * ppc_decrypt_block: The central decryption function for a single 16 bytes
+ * block. It does no stack handling or register saving to support fast calls
+ * via bl/blr. It expects that caller has pre-xored input data with first
+ * 4 words of encryption key into rD0-rD3. Pointer/counter registers must
+ * have also been set up before (rT0, rKP, CTR). Output is stored in rD0-rD3
+ * and rW0-rW3 and caller must execute a final xor on the output registers.
+ * All working registers rD0-rD3 & rW0-rW7 are overwritten during processing.
+ *
+ */
+_GLOBAL(ppc_decrypt_block)
+	LAH(rW0, rD1, 0, 12)
+	LAH(rW6, rD0, 3, 0)
+	LAH(rW3, rD0, 1, 8)
+ppc_decrypt_block_loop:
+	LAH(rW1, rD3, 0, 12)
+	LAL(rW0, rD2, 0, 12)
+	LAH(rW2, rD2, 1, 8)
+	LAL(rW2, rD3, 1, 8)
+	LAH(rW4, rD3, 2, 4)
+	LAL(rW4, rD0, 2, 4)
+	LAL(rW6, rD1, 3, 0)
+	LAH(rW5, rD1, 2, 4)
+	LAH(rW7, rD2, 3, 0)
+	LAL(rW7, rD3, 3, 0)
+	LAL(rW3, rD1, 1, 8)
+	evldw		rD1,16(rKP)
+	EAD(rD0, 0)
+	evxor		rW4,rW4,rW6
+	LWL(rW1, 12)
+	evxor		rW0,rW0,rW4
+	EAD(rD2, 2)
+	evxor		rW0,rW0,rW2
+	LWL(rW5, 4)
+	evxor		rD1,rD1,rW0
+	evldw		rD3,24(rKP)
+	evmergehi	rD0,rD0,rD1
+	EAD(rD1, 0)
+	evxor		rW3,rW3,rW7
+	LWH(rW0, 12)
+	evxor		rW3,rW3,rW1
+	EAD(rD0, 3)
+	evxor		rD3,rD3,rW3
+	LWH(rW6, 0)
+	evxor		rD3,rD3,rW5
+	EAD(rD0, 1)
+	evmergehi	rD2,rD2,rD3
+	LWH(rW3, 8)
+	LAH(rW1, rD3, 0, 12)
+	LAL(rW0, rD2, 0, 12)
+	LAH(rW2, rD2, 1, 8)
+	LAL(rW2, rD3, 1, 8)
+	LAH(rW4, rD3, 2, 4)
+	LAL(rW4, rD0, 2, 4)
+	LAL(rW6, rD1, 3, 0)
+	LAH(rW5, rD1, 2, 4)
+	LAH(rW7, rD2, 3, 0)
+	LAL(rW7, rD3, 3, 0)
+	LAL(rW3, rD1, 1, 8)
+	evldw		 rD1,32(rKP)
+	EAD(rD0, 0)
+	evxor		rW4,rW4,rW6
+	LWL(rW1, 12)
+	evxor		rW0,rW0,rW4
+	EAD(rD2, 2)
+	evxor		rW0,rW0,rW2
+	LWL(rW5, 4)
+	evxor		rD1,rD1,rW0
+	evldw		rD3,40(rKP)
+	evmergehi	rD0,rD0,rD1
+	EAD(rD1, 0)
+	evxor		rW3,rW3,rW7
+	LWH(rW0, 12)
+	evxor		rW3,rW3,rW1
+	EAD(rD0, 3)
+	evxor		rD3,rD3,rW3
+	LWH(rW6, 0)
+	evxor		rD3,rD3,rW5
+	EAD(rD0, 1)
+	evmergehi	rD2,rD2,rD3
+	LWH(rW3, 8)
+	addi		rKP,rKP,32
+	bdnz		ppc_decrypt_block_loop
+	LAH(rW1, rD3, 0, 12)
+	LAL(rW0, rD2, 0, 12)
+	LAH(rW2, rD2, 1, 8)
+	LAL(rW2, rD3, 1, 8)
+	LAH(rW4, rD3, 2, 4)
+	LAL(rW4, rD0, 2, 4)
+	LAL(rW6, rD1, 3, 0)
+	LAH(rW5, rD1, 2, 4)
+	LAH(rW7, rD2, 3, 0)
+	LAL(rW7, rD3, 3, 0)
+	LAL(rW3, rD1, 1, 8)
+	evldw		 rD1,16(rKP)
+	EAD(rD0, 0)
+	evxor		rW4,rW4,rW6
+	LWL(rW1, 12)
+	evxor		rW0,rW0,rW4
+	EAD(rD2, 2)
+	evxor		rW0,rW0,rW2
+	LWL(rW5, 4)
+	evxor		rD1,rD1,rW0
+	evldw		rD3,24(rKP)
+	evmergehi	rD0,rD0,rD1
+	DAD(rD1, 0)
+	evxor		rW3,rW3,rW7
+	LBD(rW0)
+	evxor		rW3,rW3,rW1
+	DAD(rD0, 1)
+	evxor		rD3,rD3,rW3
+	LBD(rW6)
+	evxor		rD3,rD3,rW5
+	DAD(rD0, 0)
+	evmergehi	rD2,rD2,rD3
+	LBD(rW3)
+	LAD(rW2, rD3, 0)
+	LAD(rW1, rD2, 0)
+	LAD(rW4, rD2, 1)
+	LAD(rW5, rD3, 1)
+	LAD(rW7, rD1, 1)
+	rlwimi		rW0,rW4,8,16,23
+	rlwimi		rW1,rW5,8,16,23
+	LAD(rW4, rD3, 2)
+	LAD(rW5, rD0, 2)
+	rlwimi		rW2,rW6,8,16,23
+	rlwimi		rW3,rW7,8,16,23
+	LAD(rW6, rD1, 2)
+	LAD(rW7, rD2, 2)
+	rlwimi		rW0,rW4,16,8,15
+	rlwimi		rW1,rW5,16,8,15
+	LAD(rW4, rD0, 3)
+	LAD(rW5, rD1, 3)
+	rlwimi		rW2,rW6,16,8,15
+	lwz		rD0,32(rKP)
+	rlwimi		rW3,rW7,16,8,15
+	lwz		rD1,36(rKP)
+	LAD(rW6, rD2, 3)
+	LAD(rW7, rD3, 3)
+	rlwimi		rW0,rW4,24,0,7
+	lwz		rD2,40(rKP)
+	rlwimi		rW1,rW5,24,0,7
+	lwz		rD3,44(rKP)
+	rlwimi		rW2,rW6,24,0,7
+	rlwimi		rW3,rW7,24,0,7
+	blr
diff --git a/lib/crypto/powerpc/aes-spe-keys.S b/lib/crypto/powerpc/aes-spe-keys.S
new file mode 100644
index 000000000..2e1bc0d09
--- /dev/null
+++ b/lib/crypto/powerpc/aes-spe-keys.S
@@ -0,0 +1,278 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Key handling functions for PPC AES implementation
+ *
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ */
+
+#include <asm/ppc_asm.h>
+
+#ifdef __BIG_ENDIAN__
+#define LOAD_KEY(d, s, off) \
+	lwz		d,off(s);
+#else
+#define LOAD_KEY(d, s, off) \
+	li		r0,off; \
+	lwbrx		d,s,r0;
+#endif
+
+#define INITIALIZE_KEY \
+	stwu		r1,-32(r1);	/* create stack frame		*/ \
+	stw		r14,8(r1);	/* save registers		*/ \
+	stw		r15,12(r1);					   \
+	stw		r16,16(r1);
+
+#define FINALIZE_KEY \
+	lwz		r14,8(r1);	/* restore registers		*/ \
+	lwz		r15,12(r1);					   \
+	lwz		r16,16(r1);					   \
+	xor		r5,r5,r5;	/* clear sensitive data		*/ \
+	xor		r6,r6,r6;					   \
+	xor		r7,r7,r7;					   \
+	xor		r8,r8,r8;					   \
+	xor		r9,r9,r9;					   \
+	xor		r10,r10,r10;					   \
+	xor		r11,r11,r11;					   \
+	xor		r12,r12,r12;					   \
+	addi		r1,r1,32;	/* cleanup stack		*/
+
+#define LS_BOX(r, t1, t2) \
+	lis		t2,PPC_AES_4K_ENCTAB@h;				   \
+	ori		t2,t2,PPC_AES_4K_ENCTAB@l;			   \
+	rlwimi		t2,r,4,20,27;					   \
+	lbz		t1,8(t2);					   \
+	rlwimi		r,t1,0,24,31;					   \
+	rlwimi		t2,r,28,20,27;					   \
+	lbz		t1,8(t2);					   \
+	rlwimi		r,t1,8,16,23;					   \
+	rlwimi		t2,r,20,20,27;					   \
+	lbz		t1,8(t2);					   \
+	rlwimi		r,t1,16,8,15;					   \
+	rlwimi		t2,r,12,20,27;					   \
+	lbz		t1,8(t2);					   \
+	rlwimi		r,t1,24,0,7;
+
+#define GF8_MUL(out, in, t1, t2) \
+	lis t1,0x8080;			/* multiplication in GF8	*/ \
+	ori t1,t1,0x8080; 						   \
+	and t1,t1,in; 							   \
+	srwi t1,t1,7; 							   \
+	mulli t1,t1,0x1b; 						   \
+	lis t2,0x7f7f; 							   \
+	ori t2,t2,0x7f7f; 						   \
+	and t2,t2,in; 							   \
+	slwi t2,t2,1; 							   \
+	xor out,t1,t2;
+
+/*
+ * ppc_expand_key_128(u32 *key_enc, const u8 *key)
+ *
+ * Expand 128 bit key into 176 bytes encryption key. It consists of
+ * key itself plus 10 rounds with 16 bytes each
+ *
+ */
+_GLOBAL(ppc_expand_key_128)
+	INITIALIZE_KEY
+	LOAD_KEY(r5,r4,0)
+	LOAD_KEY(r6,r4,4)
+	LOAD_KEY(r7,r4,8)
+	LOAD_KEY(r8,r4,12)
+	stw		r5,0(r3)	/* key[0..3] = input data	*/
+	stw		r6,4(r3)
+	stw		r7,8(r3)
+	stw		r8,12(r3)
+	li		r16,10		/* 10 expansion rounds		*/
+	lis		r0,0x0100	/* RCO(1)			*/
+ppc_expand_128_loop:
+	addi		r3,r3,16
+	mr		r14,r8		/* apply LS_BOX to 4th temp	*/
+	rotlwi		r14,r14,8
+	LS_BOX(r14, r15, r4)
+	xor		r14,r14,r0
+	xor		r5,r5,r14	/* xor next 4 keys		*/
+	xor		r6,r6,r5
+	xor		r7,r7,r6
+	xor		r8,r8,r7
+	stw		r5,0(r3)	/* store next 4 keys		*/
+	stw		r6,4(r3)
+	stw		r7,8(r3)
+	stw		r8,12(r3)
+	GF8_MUL(r0, r0, r4, r14)	/* multiply RCO by 2 in GF	*/
+	subi		r16,r16,1
+	cmpwi		r16,0
+	bt		eq,ppc_expand_128_end
+	b		ppc_expand_128_loop
+ppc_expand_128_end:
+	FINALIZE_KEY
+	blr
+
+/*
+ * ppc_expand_key_192(u32 *key_enc, const u8 *key)
+ *
+ * Expand 192 bit key into 208 bytes encryption key. It consists of key
+ * itself plus 12 rounds with 16 bytes each
+ *
+ */
+_GLOBAL(ppc_expand_key_192)
+	INITIALIZE_KEY
+	LOAD_KEY(r5,r4,0)
+	LOAD_KEY(r6,r4,4)
+	LOAD_KEY(r7,r4,8)
+	LOAD_KEY(r8,r4,12)
+	LOAD_KEY(r9,r4,16)
+	LOAD_KEY(r10,r4,20)
+	stw		r5,0(r3)
+	stw		r6,4(r3)
+	stw		r7,8(r3)
+	stw		r8,12(r3)
+	stw		r9,16(r3)
+	stw		r10,20(r3)
+	li		r16,8		/* 8 expansion rounds		*/
+	lis		r0,0x0100	/* RCO(1)			*/
+ppc_expand_192_loop:
+	addi		r3,r3,24
+	mr		r14,r10		/* apply LS_BOX to 6th temp	*/
+	rotlwi		r14,r14,8
+	LS_BOX(r14, r15, r4)
+	xor		r14,r14,r0
+	xor		r5,r5,r14	/* xor next 6 keys		*/
+	xor		r6,r6,r5
+	xor		r7,r7,r6
+	xor		r8,r8,r7
+	xor		r9,r9,r8
+	xor		r10,r10,r9
+	stw		r5,0(r3)
+	stw		r6,4(r3)
+	stw		r7,8(r3)
+	stw		r8,12(r3)
+	subi		r16,r16,1
+	cmpwi		r16,0		/* last round early kick out	*/
+	bt		eq,ppc_expand_192_end
+	stw		r9,16(r3)
+	stw		r10,20(r3)
+	GF8_MUL(r0, r0, r4, r14)	/* multiply RCO GF8		*/
+	b		ppc_expand_192_loop
+ppc_expand_192_end:
+	FINALIZE_KEY
+	blr
+
+/*
+ * ppc_expand_key_256(u32 *key_enc, const u8 *key)
+ *
+ * Expand 256 bit key into 240 bytes encryption key. It consists of key
+ * itself plus 14 rounds with 16 bytes each
+ *
+ */
+_GLOBAL(ppc_expand_key_256)
+	INITIALIZE_KEY
+	LOAD_KEY(r5,r4,0)
+	LOAD_KEY(r6,r4,4)
+	LOAD_KEY(r7,r4,8)
+	LOAD_KEY(r8,r4,12)
+	LOAD_KEY(r9,r4,16)
+	LOAD_KEY(r10,r4,20)
+	LOAD_KEY(r11,r4,24)
+	LOAD_KEY(r12,r4,28)
+	stw		r5,0(r3)
+	stw		r6,4(r3)
+	stw		r7,8(r3)
+	stw		r8,12(r3)
+	stw		r9,16(r3)
+	stw		r10,20(r3)
+	stw		r11,24(r3)
+	stw		r12,28(r3)
+	li		r16,7		/* 7 expansion rounds		*/
+	lis		r0,0x0100	/* RCO(1)			*/
+ppc_expand_256_loop:
+	addi		r3,r3,32
+	mr		r14,r12		/* apply LS_BOX to 8th temp	*/
+	rotlwi		r14,r14,8
+	LS_BOX(r14, r15, r4)
+	xor		r14,r14,r0
+	xor		r5,r5,r14	/* xor 4 keys			*/
+	xor		r6,r6,r5
+	xor		r7,r7,r6
+	xor		r8,r8,r7
+	mr		r14,r8
+	LS_BOX(r14, r15, r4)		/* apply LS_BOX to 4th temp	*/
+	xor		r9,r9,r14	/* xor 4 keys			*/
+	xor		r10,r10,r9
+	xor		r11,r11,r10
+	xor		r12,r12,r11
+	stw		r5,0(r3)
+	stw		r6,4(r3)
+	stw		r7,8(r3)
+	stw		r8,12(r3)
+	subi		r16,r16,1
+	cmpwi		r16,0		/* last round early kick out	*/
+	bt		eq,ppc_expand_256_end
+	stw		r9,16(r3)
+	stw		r10,20(r3)
+	stw		r11,24(r3)
+	stw		r12,28(r3)
+	GF8_MUL(r0, r0, r4, r14)
+	b		ppc_expand_256_loop
+ppc_expand_256_end:
+	FINALIZE_KEY
+	blr
+
+/*
+ * ppc_generate_decrypt_key: derive decryption key from encryption key
+ * number of bytes to handle are calculated from length of key (16/24/32)
+ *
+ */
+_GLOBAL(ppc_generate_decrypt_key)
+	addi		r6,r5,24
+	slwi		r6,r6,2
+	lwzx		r7,r4,r6	/* first/last 4 words are same	*/
+	stw		r7,0(r3)
+	lwz		r7,0(r4)
+	stwx		r7,r3,r6
+	addi		r6,r6,4
+	lwzx		r7,r4,r6
+	stw		r7,4(r3)
+	lwz		r7,4(r4)
+	stwx		r7,r3,r6
+	addi		r6,r6,4
+	lwzx		r7,r4,r6
+	stw		r7,8(r3)
+	lwz		r7,8(r4)
+	stwx		r7,r3,r6
+	addi		r6,r6,4
+	lwzx		r7,r4,r6
+	stw		r7,12(r3)
+	lwz		r7,12(r4)
+	stwx		r7,r3,r6
+	addi		r3,r3,16
+	add		r4,r4,r6
+	subi		r4,r4,28
+	addi		r5,r5,20
+	srwi		r5,r5,2
+ppc_generate_decrypt_block:
+	li	r6,4
+	mtctr	r6
+ppc_generate_decrypt_word:
+	lwz		r6,0(r4)
+	GF8_MUL(r7, r6, r0, r7)
+	GF8_MUL(r8, r7, r0, r8)
+	GF8_MUL(r9, r8, r0, r9)
+	xor		r10,r9,r6
+	xor		r11,r7,r8
+	xor		r11,r11,r9
+	xor		r12,r7,r10
+	rotrwi		r12,r12,24
+	xor		r11,r11,r12
+	xor		r12,r8,r10
+	rotrwi		r12,r12,16
+	xor		r11,r11,r12
+	rotrwi		r12,r10,8
+	xor		r11,r11,r12
+	stw		r11,0(r3)
+	addi		r3,r3,4
+	addi		r4,r4,4
+	bdnz		ppc_generate_decrypt_word
+	subi		r4,r4,32
+	subi		r5,r5,1
+	cmpwi		r5,0
+	bt		gt,ppc_generate_decrypt_block
+	blr
diff --git a/lib/crypto/powerpc/aes-spe-modes.S b/lib/crypto/powerpc/aes-spe-modes.S
new file mode 100644
index 000000000..3f92a6a85
--- /dev/null
+++ b/lib/crypto/powerpc/aes-spe-modes.S
@@ -0,0 +1,625 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * AES modes (ECB/CBC/CTR/XTS) for PPC AES implementation
+ *
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ */
+
+#include <asm/ppc_asm.h>
+#include "aes-spe-regs.h"
+
+#ifdef __BIG_ENDIAN__			/* Macros for big endian builds	*/
+
+#define LOAD_DATA(reg, off) \
+	lwz		reg,off(rSP);	/* load with offset		*/
+#define SAVE_DATA(reg, off) \
+	stw		reg,off(rDP);	/* save with offset		*/
+#define NEXT_BLOCK \
+	addi		rSP,rSP,16;	/* increment pointers per bloc	*/ \
+	addi		rDP,rDP,16;
+#define LOAD_IV(reg, off) \
+	lwz		reg,off(rIP);	/* IV loading with offset	*/
+#define SAVE_IV(reg, off) \
+	stw		reg,off(rIP);	/* IV saving with offset	*/
+#define START_IV			/* nothing to reset		*/
+#define CBC_DEC 16			/* CBC decrement per block	*/
+#define CTR_DEC 1			/* CTR decrement one byte	*/
+
+#else					/* Macros for little endian	*/
+
+#define LOAD_DATA(reg, off) \
+	lwbrx		reg,0,rSP;	/* load reversed		*/ \
+	addi		rSP,rSP,4;	/* and increment pointer	*/
+#define SAVE_DATA(reg, off) \
+	stwbrx		reg,0,rDP;	/* save reversed		*/ \
+	addi		rDP,rDP,4;	/* and increment pointer	*/
+#define NEXT_BLOCK			/* nothing todo			*/
+#define LOAD_IV(reg, off) \
+	lwbrx		reg,0,rIP;	/* load reversed		*/ \
+	addi		rIP,rIP,4;	/* and increment pointer	*/
+#define SAVE_IV(reg, off) \
+	stwbrx		reg,0,rIP;	/* load reversed		*/ \
+	addi		rIP,rIP,4;	/* and increment pointer	*/
+#define START_IV \
+	subi		rIP,rIP,16;	/* must reset pointer		*/
+#define CBC_DEC 32			/* 2 blocks because of incs	*/
+#define CTR_DEC 17			/* 1 block because of incs	*/
+
+#endif
+
+#define SAVE_0_REGS
+#define LOAD_0_REGS
+
+#define SAVE_4_REGS \
+	stw		rI0,96(r1);	/* save 32 bit registers	*/ \
+	stw		rI1,100(r1);					   \
+	stw		rI2,104(r1);					   \
+	stw		rI3,108(r1);
+
+#define LOAD_4_REGS \
+	lwz		rI0,96(r1);	/* restore 32 bit registers	*/ \
+	lwz		rI1,100(r1);					   \
+	lwz		rI2,104(r1);					   \
+	lwz		rI3,108(r1);
+
+#define SAVE_8_REGS \
+	SAVE_4_REGS							   \
+	stw		rG0,112(r1);	/* save 32 bit registers	*/ \
+	stw		rG1,116(r1);					   \
+	stw		rG2,120(r1);					   \
+	stw		rG3,124(r1);
+
+#define LOAD_8_REGS \
+	LOAD_4_REGS							   \
+	lwz		rG0,112(r1);	/* restore 32 bit registers	*/ \
+	lwz		rG1,116(r1);					   \
+	lwz		rG2,120(r1);					   \
+	lwz		rG3,124(r1);
+
+#define INITIALIZE_CRYPT(tab,nr32bitregs) \
+	mflr		r0;						   \
+	stwu		r1,-160(r1);	/* create stack frame		*/ \
+	lis		rT0,tab@h;	/* en-/decryption table pointer	*/ \
+	stw		r0,8(r1);	/* save link register		*/ \
+	ori		rT0,rT0,tab@l;					   \
+	evstdw		r14,16(r1);					   \
+	mr		rKS,rKP;					   \
+	evstdw		r15,24(r1);	/* We must save non volatile	*/ \
+	evstdw		r16,32(r1);	/* registers. Take the chance	*/ \
+	evstdw		r17,40(r1);	/* and save the SPE part too	*/ \
+	evstdw		r18,48(r1);					   \
+	evstdw		r19,56(r1);					   \
+	evstdw		r20,64(r1);					   \
+	evstdw		r21,72(r1);					   \
+	evstdw		r22,80(r1);					   \
+	evstdw		r23,88(r1);					   \
+	SAVE_##nr32bitregs##_REGS
+
+#define FINALIZE_CRYPT(nr32bitregs) \
+	lwz		r0,8(r1);					   \
+	evldw		r14,16(r1);	/* restore SPE registers	*/ \
+	evldw		r15,24(r1);					   \
+	evldw		r16,32(r1);					   \
+	evldw		r17,40(r1);					   \
+	evldw		r18,48(r1);					   \
+	evldw		r19,56(r1);					   \
+	evldw		r20,64(r1);					   \
+	evldw		r21,72(r1);					   \
+	evldw		r22,80(r1);					   \
+	evldw		r23,88(r1);					   \
+	LOAD_##nr32bitregs##_REGS					   \
+	mtlr		r0;		/* restore link register	*/ \
+	xor		r0,r0,r0;					   \
+	stw		r0,16(r1);	/* delete sensitive data	*/ \
+	stw		r0,24(r1);	/* that we might have pushed	*/ \
+	stw		r0,32(r1);	/* from other context that runs	*/ \
+	stw		r0,40(r1);	/* the same code		*/ \
+	stw		r0,48(r1);					   \
+	stw		r0,56(r1);					   \
+	stw		r0,64(r1);					   \
+	stw		r0,72(r1);					   \
+	stw		r0,80(r1);					   \
+	stw		r0,88(r1);					   \
+	addi		r1,r1,160;	/* cleanup stack frame		*/
+
+#define ENDIAN_SWAP(t0, t1, s0, s1) \
+	rotrwi		t0,s0,8;	/* swap endianness for 2 GPRs	*/ \
+	rotrwi		t1,s1,8;					   \
+	rlwimi		t0,s0,8,8,15;					   \
+	rlwimi		t1,s1,8,8,15;					   \
+	rlwimi		t0,s0,8,24,31;					   \
+	rlwimi		t1,s1,8,24,31;
+
+#define GF128_MUL(d0, d1, d2, d3, t0) \
+	li		t0,0x87;	/* multiplication in GF128	*/ \
+	cmpwi		d3,-1;						   \
+	iselgt		t0,0,t0;					   \
+	rlwimi		d3,d2,0,0,0;	/* propagate "carry" bits	*/ \
+	rotlwi		d3,d3,1;					   \
+	rlwimi		d2,d1,0,0,0;					   \
+	rotlwi		d2,d2,1;					   \
+	rlwimi		d1,d0,0,0,0;					   \
+	slwi		d0,d0,1;	/* shift left 128 bit		*/ \
+	rotlwi		d1,d1,1;					   \
+	xor		d0,d0,t0;
+
+#define START_KEY(d0, d1, d2, d3) \
+	lwz		rW0,0(rKP);					   \
+	mtctr		rRR;						   \
+	lwz		rW1,4(rKP);					   \
+	lwz		rW2,8(rKP);					   \
+	lwz		rW3,12(rKP);					   \
+	xor		rD0,d0,rW0;					   \
+	xor		rD1,d1,rW1;					   \
+	xor		rD2,d2,rW2;					   \
+	xor		rD3,d3,rW3;
+
+/*
+ * ppc_encrypt_aes(u8 *out, const u8 *in, u32 *key_enc,
+ *		   u32 rounds)
+ *
+ * called from glue layer to encrypt a single 16 byte block
+ * round values are AES128 = 4, AES192 = 5, AES256 = 6
+ *
+ */
+_GLOBAL(ppc_encrypt_aes)
+	INITIALIZE_CRYPT(PPC_AES_4K_ENCTAB, 0)
+	LOAD_DATA(rD0, 0)
+	LOAD_DATA(rD1, 4)
+	LOAD_DATA(rD2, 8)
+	LOAD_DATA(rD3, 12)
+	START_KEY(rD0, rD1, rD2, rD3)
+	bl		ppc_encrypt_block
+	xor		rD0,rD0,rW0
+	SAVE_DATA(rD0, 0)
+	xor		rD1,rD1,rW1
+	SAVE_DATA(rD1, 4)
+	xor		rD2,rD2,rW2
+	SAVE_DATA(rD2, 8)
+	xor		rD3,rD3,rW3
+	SAVE_DATA(rD3, 12)
+	FINALIZE_CRYPT(0)
+	blr
+
+/*
+ * ppc_decrypt_aes(u8 *out, const u8 *in, u32 *key_dec,
+ *		   u32 rounds)
+ *
+ * called from glue layer to decrypt a single 16 byte block
+ * round values are AES128 = 4, AES192 = 5, AES256 = 6
+ *
+ */
+_GLOBAL(ppc_decrypt_aes)
+	INITIALIZE_CRYPT(PPC_AES_4K_DECTAB,0)
+	LOAD_DATA(rD0, 0)
+	addi		rT1,rT0,4096
+	LOAD_DATA(rD1, 4)
+	LOAD_DATA(rD2, 8)
+	LOAD_DATA(rD3, 12)
+	START_KEY(rD0, rD1, rD2, rD3)
+	bl		ppc_decrypt_block
+	xor		rD0,rD0,rW0
+	SAVE_DATA(rD0, 0)
+	xor		rD1,rD1,rW1
+	SAVE_DATA(rD1, 4)
+	xor		rD2,rD2,rW2
+	SAVE_DATA(rD2, 8)
+	xor		rD3,rD3,rW3
+	SAVE_DATA(rD3, 12)
+	FINALIZE_CRYPT(0)
+	blr
+
+/*
+ * ppc_encrypt_ecb(u8 *out, const u8 *in, u32 *key_enc,
+ *		   u32 rounds, u32 bytes);
+ *
+ * called from glue layer to encrypt multiple blocks via ECB
+ * Bytes must be larger or equal 16 and only whole blocks are
+ * processed. round values are AES128 = 4, AES192 = 5 and
+ * AES256 = 6
+ *
+ */
+_GLOBAL(ppc_encrypt_ecb)
+	INITIALIZE_CRYPT(PPC_AES_4K_ENCTAB, 0)
+ppc_encrypt_ecb_loop:
+	LOAD_DATA(rD0, 0)
+	mr		rKP,rKS
+	LOAD_DATA(rD1, 4)
+	subi		rLN,rLN,16
+	LOAD_DATA(rD2, 8)
+	cmpwi		rLN,15
+	LOAD_DATA(rD3, 12)
+	START_KEY(rD0, rD1, rD2, rD3)
+	bl		ppc_encrypt_block
+	xor		rD0,rD0,rW0
+	SAVE_DATA(rD0, 0)
+	xor		rD1,rD1,rW1
+	SAVE_DATA(rD1, 4)
+	xor		rD2,rD2,rW2
+	SAVE_DATA(rD2, 8)
+	xor		rD3,rD3,rW3
+	SAVE_DATA(rD3, 12)
+	NEXT_BLOCK
+	bt		gt,ppc_encrypt_ecb_loop
+	FINALIZE_CRYPT(0)
+	blr
+
+/*
+ * ppc_decrypt_ecb(u8 *out, const u8 *in, u32 *key_dec,
+ *		   u32 rounds, u32 bytes);
+ *
+ * called from glue layer to decrypt multiple blocks via ECB
+ * Bytes must be larger or equal 16 and only whole blocks are
+ * processed. round values are AES128 = 4, AES192 = 5 and
+ * AES256 = 6
+ *
+ */
+_GLOBAL(ppc_decrypt_ecb)
+	INITIALIZE_CRYPT(PPC_AES_4K_DECTAB, 0)
+	addi		rT1,rT0,4096
+ppc_decrypt_ecb_loop:
+	LOAD_DATA(rD0, 0)
+	mr		rKP,rKS
+	LOAD_DATA(rD1, 4)
+	subi		rLN,rLN,16
+	LOAD_DATA(rD2, 8)
+	cmpwi		rLN,15
+	LOAD_DATA(rD3, 12)
+	START_KEY(rD0, rD1, rD2, rD3)
+	bl		ppc_decrypt_block
+	xor		rD0,rD0,rW0
+	SAVE_DATA(rD0, 0)
+	xor		rD1,rD1,rW1
+	SAVE_DATA(rD1, 4)
+	xor		rD2,rD2,rW2
+	SAVE_DATA(rD2, 8)
+	xor		rD3,rD3,rW3
+	SAVE_DATA(rD3, 12)
+	NEXT_BLOCK
+	bt		gt,ppc_decrypt_ecb_loop
+	FINALIZE_CRYPT(0)
+	blr
+
+/*
+ * ppc_encrypt_cbc(u8 *out, const u8 *in, u32 *key_enc,
+ *		   32 rounds, u32 bytes, u8 *iv);
+ *
+ * called from glue layer to encrypt multiple blocks via CBC
+ * Bytes must be larger or equal 16 and only whole blocks are
+ * processed. round values are AES128 = 4, AES192 = 5 and
+ * AES256 = 6
+ *
+ */
+_GLOBAL(ppc_encrypt_cbc)
+	INITIALIZE_CRYPT(PPC_AES_4K_ENCTAB, 4)
+	LOAD_IV(rI0, 0)
+	LOAD_IV(rI1, 4)
+	LOAD_IV(rI2, 8)
+	LOAD_IV(rI3, 12)
+ppc_encrypt_cbc_loop:
+	LOAD_DATA(rD0, 0)
+	mr		rKP,rKS
+	LOAD_DATA(rD1, 4)
+	subi		rLN,rLN,16
+	LOAD_DATA(rD2, 8)
+	cmpwi		rLN,15
+	LOAD_DATA(rD3, 12)
+	xor		rD0,rD0,rI0
+	xor		rD1,rD1,rI1
+	xor		rD2,rD2,rI2
+	xor		rD3,rD3,rI3
+	START_KEY(rD0, rD1, rD2, rD3)
+	bl		ppc_encrypt_block
+	xor		rI0,rD0,rW0
+	SAVE_DATA(rI0, 0)
+	xor		rI1,rD1,rW1
+	SAVE_DATA(rI1, 4)
+	xor		rI2,rD2,rW2
+	SAVE_DATA(rI2, 8)
+	xor		rI3,rD3,rW3
+	SAVE_DATA(rI3, 12)
+	NEXT_BLOCK
+	bt		gt,ppc_encrypt_cbc_loop
+	START_IV
+	SAVE_IV(rI0, 0)
+	SAVE_IV(rI1, 4)
+	SAVE_IV(rI2, 8)
+	SAVE_IV(rI3, 12)
+	FINALIZE_CRYPT(4)
+	blr
+
+/*
+ * ppc_decrypt_cbc(u8 *out, const u8 *in, u32 *key_dec,
+ *		   u32 rounds, u32 bytes, u8 *iv);
+ *
+ * called from glue layer to decrypt multiple blocks via CBC
+ * round values are AES128 = 4, AES192 = 5, AES256 = 6
+ *
+ */
+_GLOBAL(ppc_decrypt_cbc)
+	INITIALIZE_CRYPT(PPC_AES_4K_DECTAB, 4)
+	li		rT1,15
+	LOAD_IV(rI0, 0)
+	andc		rLN,rLN,rT1
+	LOAD_IV(rI1, 4)
+	subi		rLN,rLN,16
+	LOAD_IV(rI2, 8)
+	add		rSP,rSP,rLN	/* reverse processing		*/
+	LOAD_IV(rI3, 12)
+	add		rDP,rDP,rLN
+	LOAD_DATA(rD0, 0)
+	addi		rT1,rT0,4096
+	LOAD_DATA(rD1, 4)
+	LOAD_DATA(rD2, 8)
+	LOAD_DATA(rD3, 12)
+	START_IV
+	SAVE_IV(rD0, 0)
+	SAVE_IV(rD1, 4)
+	SAVE_IV(rD2, 8)
+	cmpwi		rLN,16
+	SAVE_IV(rD3, 12)
+	bt		lt,ppc_decrypt_cbc_end
+ppc_decrypt_cbc_loop:
+	mr		rKP,rKS
+	START_KEY(rD0, rD1, rD2, rD3)
+	bl		ppc_decrypt_block
+	subi		rLN,rLN,16
+	subi		rSP,rSP,CBC_DEC
+	xor		rW0,rD0,rW0
+	LOAD_DATA(rD0, 0)
+	xor		rW1,rD1,rW1
+	LOAD_DATA(rD1, 4)
+	xor		rW2,rD2,rW2
+	LOAD_DATA(rD2, 8)
+	xor		rW3,rD3,rW3
+	LOAD_DATA(rD3, 12)
+	xor		rW0,rW0,rD0
+	SAVE_DATA(rW0, 0)
+	xor		rW1,rW1,rD1
+	SAVE_DATA(rW1, 4)
+	xor		rW2,rW2,rD2
+	SAVE_DATA(rW2, 8)
+	xor		rW3,rW3,rD3
+	SAVE_DATA(rW3, 12)
+	cmpwi		rLN,15
+	subi		rDP,rDP,CBC_DEC
+	bt		gt,ppc_decrypt_cbc_loop
+ppc_decrypt_cbc_end:
+	mr		rKP,rKS
+	START_KEY(rD0, rD1, rD2, rD3)
+	bl		ppc_decrypt_block
+	xor		rW0,rW0,rD0
+	xor		rW1,rW1,rD1
+	xor		rW2,rW2,rD2
+	xor		rW3,rW3,rD3
+	xor		rW0,rW0,rI0	/* decrypt with initial IV	*/
+	SAVE_DATA(rW0, 0)
+	xor		rW1,rW1,rI1
+	SAVE_DATA(rW1, 4)
+	xor		rW2,rW2,rI2
+	SAVE_DATA(rW2, 8)
+	xor		rW3,rW3,rI3
+	SAVE_DATA(rW3, 12)
+	FINALIZE_CRYPT(4)
+	blr
+
+/*
+ * ppc_crypt_ctr(u8 *out, const u8 *in, u32 *key_enc,
+ *		 u32 rounds, u32 bytes, u8 *iv);
+ *
+ * called from glue layer to encrypt/decrypt multiple blocks
+ * via CTR. Number of bytes does not need to be a multiple of
+ * 16. Round values are AES128 = 4, AES192 = 5, AES256 = 6
+ *
+ */
+_GLOBAL(ppc_crypt_ctr)
+	INITIALIZE_CRYPT(PPC_AES_4K_ENCTAB, 4)
+	LOAD_IV(rI0, 0)
+	LOAD_IV(rI1, 4)
+	LOAD_IV(rI2, 8)
+	cmpwi		rLN,16
+	LOAD_IV(rI3, 12)
+	START_IV
+	bt		lt,ppc_crypt_ctr_partial
+ppc_crypt_ctr_loop:
+	mr		rKP,rKS
+	START_KEY(rI0, rI1, rI2, rI3)
+	bl		ppc_encrypt_block
+	xor		rW0,rD0,rW0
+	xor		rW1,rD1,rW1
+	xor		rW2,rD2,rW2
+	xor		rW3,rD3,rW3
+	LOAD_DATA(rD0, 0)
+	subi		rLN,rLN,16
+	LOAD_DATA(rD1, 4)
+	LOAD_DATA(rD2, 8)
+	LOAD_DATA(rD3, 12)
+	xor		rD0,rD0,rW0
+	SAVE_DATA(rD0, 0)
+	xor		rD1,rD1,rW1
+	SAVE_DATA(rD1, 4)
+	xor		rD2,rD2,rW2
+	SAVE_DATA(rD2, 8)
+	xor		rD3,rD3,rW3
+	SAVE_DATA(rD3, 12)
+	addic		rI3,rI3,1	/* increase counter			*/
+	addze		rI2,rI2
+	addze		rI1,rI1
+	addze		rI0,rI0
+	NEXT_BLOCK
+	cmpwi		rLN,15
+	bt		gt,ppc_crypt_ctr_loop
+ppc_crypt_ctr_partial:
+	cmpwi		rLN,0
+	bt		eq,ppc_crypt_ctr_end
+	mr		rKP,rKS
+	START_KEY(rI0, rI1, rI2, rI3)
+	bl		ppc_encrypt_block
+	xor		rW0,rD0,rW0
+	SAVE_IV(rW0, 0)
+	xor		rW1,rD1,rW1
+	SAVE_IV(rW1, 4)
+	xor		rW2,rD2,rW2
+	SAVE_IV(rW2, 8)
+	xor		rW3,rD3,rW3
+	SAVE_IV(rW3, 12)
+	mtctr		rLN
+	subi		rIP,rIP,CTR_DEC
+	subi		rSP,rSP,1
+	subi		rDP,rDP,1
+ppc_crypt_ctr_xorbyte:
+	lbzu		rW4,1(rIP)	/* bytewise xor for partial block	*/
+	lbzu		rW5,1(rSP)
+	xor		rW4,rW4,rW5
+	stbu		rW4,1(rDP)
+	bdnz		ppc_crypt_ctr_xorbyte
+	subf		rIP,rLN,rIP
+	addi		rIP,rIP,1
+	addic		rI3,rI3,1
+	addze		rI2,rI2
+	addze		rI1,rI1
+	addze		rI0,rI0
+ppc_crypt_ctr_end:
+	SAVE_IV(rI0, 0)
+	SAVE_IV(rI1, 4)
+	SAVE_IV(rI2, 8)
+	SAVE_IV(rI3, 12)
+	FINALIZE_CRYPT(4)
+	blr
+
+/*
+ * ppc_encrypt_xts(u8 *out, const u8 *in, u32 *key_enc,
+ *		   u32 rounds, u32 bytes, u8 *iv, u32 *key_twk);
+ *
+ * called from glue layer to encrypt multiple blocks via XTS
+ * If key_twk is given, the initial IV encryption will be
+ * processed too. Round values are AES128 = 4, AES192 = 5,
+ * AES256 = 6
+ *
+ */
+_GLOBAL(ppc_encrypt_xts)
+	INITIALIZE_CRYPT(PPC_AES_4K_ENCTAB, 8)
+	LOAD_IV(rI0, 0)
+	LOAD_IV(rI1, 4)
+	LOAD_IV(rI2, 8)
+	cmpwi		rKT,0
+	LOAD_IV(rI3, 12)
+	bt		eq,ppc_encrypt_xts_notweak
+	mr		rKP,rKT
+	START_KEY(rI0, rI1, rI2, rI3)
+	bl		ppc_encrypt_block
+	xor		rI0,rD0,rW0
+	xor		rI1,rD1,rW1
+	xor		rI2,rD2,rW2
+	xor		rI3,rD3,rW3
+ppc_encrypt_xts_notweak:
+	ENDIAN_SWAP(rG0, rG1, rI0, rI1)
+	ENDIAN_SWAP(rG2, rG3, rI2, rI3)
+ppc_encrypt_xts_loop:
+	LOAD_DATA(rD0, 0)
+	mr		rKP,rKS
+	LOAD_DATA(rD1, 4)
+	subi		rLN,rLN,16
+	LOAD_DATA(rD2, 8)
+	LOAD_DATA(rD3, 12)
+	xor		rD0,rD0,rI0
+	xor		rD1,rD1,rI1
+	xor		rD2,rD2,rI2
+	xor		rD3,rD3,rI3
+	START_KEY(rD0, rD1, rD2, rD3)
+	bl		ppc_encrypt_block
+	xor		rD0,rD0,rW0
+	xor		rD1,rD1,rW1
+	xor		rD2,rD2,rW2
+	xor		rD3,rD3,rW3
+	xor		rD0,rD0,rI0
+	SAVE_DATA(rD0, 0)
+	xor		rD1,rD1,rI1
+	SAVE_DATA(rD1, 4)
+	xor		rD2,rD2,rI2
+	SAVE_DATA(rD2, 8)
+	xor		rD3,rD3,rI3
+	SAVE_DATA(rD3, 12)
+	GF128_MUL(rG0, rG1, rG2, rG3, rW0)
+	ENDIAN_SWAP(rI0, rI1, rG0, rG1)
+	ENDIAN_SWAP(rI2, rI3, rG2, rG3)
+	cmpwi		rLN,0
+	NEXT_BLOCK
+	bt		gt,ppc_encrypt_xts_loop
+	START_IV
+	SAVE_IV(rI0, 0)
+	SAVE_IV(rI1, 4)
+	SAVE_IV(rI2, 8)
+	SAVE_IV(rI3, 12)
+	FINALIZE_CRYPT(8)
+	blr
+
+/*
+ * ppc_decrypt_xts(u8 *out, const u8 *in, u32 *key_dec,
+ *		   u32 rounds, u32 blocks, u8 *iv, u32 *key_twk);
+ *
+ * called from glue layer to decrypt multiple blocks via XTS
+ * If key_twk is given, the initial IV encryption will be
+ * processed too. Round values are AES128 = 4, AES192 = 5,
+ * AES256 = 6
+ *
+ */
+_GLOBAL(ppc_decrypt_xts)
+	INITIALIZE_CRYPT(PPC_AES_4K_DECTAB, 8)
+	LOAD_IV(rI0, 0)
+	addi		rT1,rT0,4096
+	LOAD_IV(rI1, 4)
+	LOAD_IV(rI2, 8)
+	cmpwi		rKT,0
+	LOAD_IV(rI3, 12)
+	bt		eq,ppc_decrypt_xts_notweak
+	subi		rT0,rT0,4096
+	mr		rKP,rKT
+	START_KEY(rI0, rI1, rI2, rI3)
+	bl		ppc_encrypt_block
+	xor		rI0,rD0,rW0
+	xor		rI1,rD1,rW1
+	xor		rI2,rD2,rW2
+	xor		rI3,rD3,rW3
+	addi		rT0,rT0,4096
+ppc_decrypt_xts_notweak:
+	ENDIAN_SWAP(rG0, rG1, rI0, rI1)
+	ENDIAN_SWAP(rG2, rG3, rI2, rI3)
+ppc_decrypt_xts_loop:
+	LOAD_DATA(rD0, 0)
+	mr		rKP,rKS
+	LOAD_DATA(rD1, 4)
+	subi		rLN,rLN,16
+	LOAD_DATA(rD2, 8)
+	LOAD_DATA(rD3, 12)
+	xor		rD0,rD0,rI0
+	xor		rD1,rD1,rI1
+	xor		rD2,rD2,rI2
+	xor		rD3,rD3,rI3
+	START_KEY(rD0, rD1, rD2, rD3)
+	bl		ppc_decrypt_block
+	xor		rD0,rD0,rW0
+	xor		rD1,rD1,rW1
+	xor		rD2,rD2,rW2
+	xor		rD3,rD3,rW3
+	xor		rD0,rD0,rI0
+	SAVE_DATA(rD0, 0)
+	xor		rD1,rD1,rI1
+	SAVE_DATA(rD1, 4)
+	xor		rD2,rD2,rI2
+	SAVE_DATA(rD2, 8)
+	xor		rD3,rD3,rI3
+	SAVE_DATA(rD3, 12)
+	GF128_MUL(rG0, rG1, rG2, rG3, rW0)
+	ENDIAN_SWAP(rI0, rI1, rG0, rG1)
+	ENDIAN_SWAP(rI2, rI3, rG2, rG3)
+	cmpwi		rLN,0
+	NEXT_BLOCK
+	bt		gt,ppc_decrypt_xts_loop
+	START_IV
+	SAVE_IV(rI0, 0)
+	SAVE_IV(rI1, 4)
+	SAVE_IV(rI2, 8)
+	SAVE_IV(rI3, 12)
+	FINALIZE_CRYPT(8)
+	blr
diff --git a/lib/crypto/powerpc/aes-spe-regs.h b/lib/crypto/powerpc/aes-spe-regs.h
new file mode 100644
index 000000000..2eb4c9b94
--- /dev/null
+++ b/lib/crypto/powerpc/aes-spe-regs.h
@@ -0,0 +1,37 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Common registers for PPC AES implementation
+ *
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ */
+
+#define rKS r0	/* copy of en-/decryption key pointer			*/
+#define rDP r3	/* destination pointer					*/
+#define rSP r4	/* source pointer					*/
+#define rKP r5	/* pointer to en-/decryption key pointer		*/
+#define rRR r6	/* en-/decryption rounds				*/
+#define rLN r7	/* length of data to be processed			*/
+#define rIP r8	/* potiner to IV (CBC/CTR/XTS modes)			*/
+#define rKT r9	/* pointer to tweak key (XTS mode)			*/
+#define rT0 r11	/* pointers to en-/decryption tables			*/
+#define rT1 r10
+#define rD0 r9	/* data 						*/
+#define rD1 r14
+#define rD2 r12
+#define rD3 r15
+#define rW0 r16	/* working registers					*/
+#define rW1 r17
+#define rW2 r18
+#define rW3 r19
+#define rW4 r20
+#define rW5 r21
+#define rW6 r22
+#define rW7 r23
+#define rI0 r24	/* IV							*/
+#define rI1 r25
+#define rI2 r26
+#define rI3 r27
+#define rG0 r28	/* endian reversed tweak (XTS mode)			*/
+#define rG1 r29
+#define rG2 r30
+#define rG3 r31
diff --git a/lib/crypto/powerpc/aes-tab-4k.S b/lib/crypto/powerpc/aes-tab-4k.S
new file mode 100644
index 000000000..ceb604bc6
--- /dev/null
+++ b/lib/crypto/powerpc/aes-tab-4k.S
@@ -0,0 +1,326 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * 4K AES tables for PPC AES implementation
+ *
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ */
+
+/*
+ * These big endian AES encryption/decryption tables have been taken from
+ * crypto/aes_generic.c and are designed to be simply accessed by a combination
+ * of rlwimi/lwz instructions with a minimum of table registers (usually only
+ * one required). Thus they are aligned to 4K. The locality of rotated values
+ * is derived from the reduced offsets that are available in the SPE load
+ * instructions. E.g. evldw, evlwwsplat, ...
+ *
+ * For the safety-conscious it has to be noted that they might be vulnerable
+ * to cache timing attacks because of their size. Nevertheless in contrast to
+ * the generic tables they have been reduced from 16KB to 8KB + 256 bytes.
+ * This is a quite good tradeoff for low power devices (e.g. routers) without
+ * dedicated encryption hardware where we usually have no multiuser
+ * environment.
+ *
+ */
+
+#define R(a, b, c, d) \
+	0x##a##b##c##d, 0x##d##a##b##c, 0x##c##d##a##b, 0x##b##c##d##a
+
+.data
+.align 12
+.globl PPC_AES_4K_ENCTAB
+PPC_AES_4K_ENCTAB:
+/* encryption table, same as crypto_ft_tab in crypto/aes-generic.c */
+	.long R(c6, 63, 63, a5), R(f8, 7c, 7c, 84)
+	.long R(ee, 77, 77, 99), R(f6, 7b, 7b, 8d)
+	.long R(ff, f2, f2, 0d), R(d6, 6b, 6b, bd)
+	.long R(de, 6f, 6f, b1), R(91, c5, c5, 54)
+	.long R(60, 30, 30, 50), R(02, 01, 01, 03)
+	.long R(ce, 67, 67, a9), R(56, 2b, 2b, 7d)
+	.long R(e7, fe, fe, 19), R(b5, d7, d7, 62)
+	.long R(4d, ab, ab, e6), R(ec, 76, 76, 9a)
+	.long R(8f, ca, ca, 45), R(1f, 82, 82, 9d)
+	.long R(89, c9, c9, 40), R(fa, 7d, 7d, 87)
+	.long R(ef, fa, fa, 15), R(b2, 59, 59, eb)
+	.long R(8e, 47, 47, c9), R(fb, f0, f0, 0b)
+	.long R(41, ad, ad, ec), R(b3, d4, d4, 67)
+	.long R(5f, a2, a2, fd), R(45, af, af, ea)
+	.long R(23, 9c, 9c, bf), R(53, a4, a4, f7)
+	.long R(e4, 72, 72, 96), R(9b, c0, c0, 5b)
+	.long R(75, b7, b7, c2), R(e1, fd, fd, 1c)
+	.long R(3d, 93, 93, ae), R(4c, 26, 26, 6a)
+	.long R(6c, 36, 36, 5a), R(7e, 3f, 3f, 41)
+	.long R(f5, f7, f7, 02), R(83, cc, cc, 4f)
+	.long R(68, 34, 34, 5c), R(51, a5, a5, f4)
+	.long R(d1, e5, e5, 34), R(f9, f1, f1, 08)
+	.long R(e2, 71, 71, 93), R(ab, d8, d8, 73)
+	.long R(62, 31, 31, 53), R(2a, 15, 15, 3f)
+	.long R(08, 04, 04, 0c), R(95, c7, c7, 52)
+	.long R(46, 23, 23, 65), R(9d, c3, c3, 5e)
+	.long R(30, 18, 18, 28), R(37, 96, 96, a1)
+	.long R(0a, 05, 05, 0f), R(2f, 9a, 9a, b5)
+	.long R(0e, 07, 07, 09), R(24, 12, 12, 36)
+	.long R(1b, 80, 80, 9b), R(df, e2, e2, 3d)
+	.long R(cd, eb, eb, 26), R(4e, 27, 27, 69)
+	.long R(7f, b2, b2, cd), R(ea, 75, 75, 9f)
+	.long R(12, 09, 09, 1b), R(1d, 83, 83, 9e)
+	.long R(58, 2c, 2c, 74), R(34, 1a, 1a, 2e)
+	.long R(36, 1b, 1b, 2d), R(dc, 6e, 6e, b2)
+	.long R(b4, 5a, 5a, ee), R(5b, a0, a0, fb)
+	.long R(a4, 52, 52, f6), R(76, 3b, 3b, 4d)
+	.long R(b7, d6, d6, 61), R(7d, b3, b3, ce)
+	.long R(52, 29, 29, 7b), R(dd, e3, e3, 3e)
+	.long R(5e, 2f, 2f, 71), R(13, 84, 84, 97)
+	.long R(a6, 53, 53, f5), R(b9, d1, d1, 68)
+	.long R(00, 00, 00, 00), R(c1, ed, ed, 2c)
+	.long R(40, 20, 20, 60), R(e3, fc, fc, 1f)
+	.long R(79, b1, b1, c8), R(b6, 5b, 5b, ed)
+	.long R(d4, 6a, 6a, be), R(8d, cb, cb, 46)
+	.long R(67, be, be, d9), R(72, 39, 39, 4b)
+	.long R(94, 4a, 4a, de), R(98, 4c, 4c, d4)
+	.long R(b0, 58, 58, e8), R(85, cf, cf, 4a)
+	.long R(bb, d0, d0, 6b), R(c5, ef, ef, 2a)
+	.long R(4f, aa, aa, e5), R(ed, fb, fb, 16)
+	.long R(86, 43, 43, c5), R(9a, 4d, 4d, d7)
+	.long R(66, 33, 33, 55), R(11, 85, 85, 94)
+	.long R(8a, 45, 45, cf), R(e9, f9, f9, 10)
+	.long R(04, 02, 02, 06), R(fe, 7f, 7f, 81)
+	.long R(a0, 50, 50, f0), R(78, 3c, 3c, 44)
+	.long R(25, 9f, 9f, ba), R(4b, a8, a8, e3)
+	.long R(a2, 51, 51, f3), R(5d, a3, a3, fe)
+	.long R(80, 40, 40, c0), R(05, 8f, 8f, 8a)
+	.long R(3f, 92, 92, ad), R(21, 9d, 9d, bc)
+	.long R(70, 38, 38, 48), R(f1, f5, f5, 04)
+	.long R(63, bc, bc, df), R(77, b6, b6, c1)
+	.long R(af, da, da, 75), R(42, 21, 21, 63)
+	.long R(20, 10, 10, 30), R(e5, ff, ff, 1a)
+	.long R(fd, f3, f3, 0e), R(bf, d2, d2, 6d)
+	.long R(81, cd, cd, 4c), R(18, 0c, 0c, 14)
+	.long R(26, 13, 13, 35), R(c3, ec, ec, 2f)
+	.long R(be, 5f, 5f, e1), R(35, 97, 97, a2)
+	.long R(88, 44, 44, cc), R(2e, 17, 17, 39)
+	.long R(93, c4, c4, 57), R(55, a7, a7, f2)
+	.long R(fc, 7e, 7e, 82), R(7a, 3d, 3d, 47)
+	.long R(c8, 64, 64, ac), R(ba, 5d, 5d, e7)
+	.long R(32, 19, 19, 2b), R(e6, 73, 73, 95)
+	.long R(c0, 60, 60, a0), R(19, 81, 81, 98)
+	.long R(9e, 4f, 4f, d1), R(a3, dc, dc, 7f)
+	.long R(44, 22, 22, 66), R(54, 2a, 2a, 7e)
+	.long R(3b, 90, 90, ab), R(0b, 88, 88, 83)
+	.long R(8c, 46, 46, ca), R(c7, ee, ee, 29)
+	.long R(6b, b8, b8, d3), R(28, 14, 14, 3c)
+	.long R(a7, de, de, 79), R(bc, 5e, 5e, e2)
+	.long R(16, 0b, 0b, 1d), R(ad, db, db, 76)
+	.long R(db, e0, e0, 3b), R(64, 32, 32, 56)
+	.long R(74, 3a, 3a, 4e), R(14, 0a, 0a, 1e)
+	.long R(92, 49, 49, db), R(0c, 06, 06, 0a)
+	.long R(48, 24, 24, 6c), R(b8, 5c, 5c, e4)
+	.long R(9f, c2, c2, 5d), R(bd, d3, d3, 6e)
+	.long R(43, ac, ac, ef), R(c4, 62, 62, a6)
+	.long R(39, 91, 91, a8), R(31, 95, 95, a4)
+	.long R(d3, e4, e4, 37), R(f2, 79, 79, 8b)
+	.long R(d5, e7, e7, 32), R(8b, c8, c8, 43)
+	.long R(6e, 37, 37, 59), R(da, 6d, 6d, b7)
+	.long R(01, 8d, 8d, 8c), R(b1, d5, d5, 64)
+	.long R(9c, 4e, 4e, d2), R(49, a9, a9, e0)
+	.long R(d8, 6c, 6c, b4), R(ac, 56, 56, fa)
+	.long R(f3, f4, f4, 07), R(cf, ea, ea, 25)
+	.long R(ca, 65, 65, af), R(f4, 7a, 7a, 8e)
+	.long R(47, ae, ae, e9), R(10, 08, 08, 18)
+	.long R(6f, ba, ba, d5), R(f0, 78, 78, 88)
+	.long R(4a, 25, 25, 6f), R(5c, 2e, 2e, 72)
+	.long R(38, 1c, 1c, 24), R(57, a6, a6, f1)
+	.long R(73, b4, b4, c7), R(97, c6, c6, 51)
+	.long R(cb, e8, e8, 23), R(a1, dd, dd, 7c)
+	.long R(e8, 74, 74, 9c), R(3e, 1f, 1f, 21)
+	.long R(96, 4b, 4b, dd), R(61, bd, bd, dc)
+	.long R(0d, 8b, 8b, 86), R(0f, 8a, 8a, 85)
+	.long R(e0, 70, 70, 90), R(7c, 3e, 3e, 42)
+	.long R(71, b5, b5, c4), R(cc, 66, 66, aa)
+	.long R(90, 48, 48, d8), R(06, 03, 03, 05)
+	.long R(f7, f6, f6, 01), R(1c, 0e, 0e, 12)
+	.long R(c2, 61, 61, a3), R(6a, 35, 35, 5f)
+	.long R(ae, 57, 57, f9), R(69, b9, b9, d0)
+	.long R(17, 86, 86, 91), R(99, c1, c1, 58)
+	.long R(3a, 1d, 1d, 27), R(27, 9e, 9e, b9)
+	.long R(d9, e1, e1, 38), R(eb, f8, f8, 13)
+	.long R(2b, 98, 98, b3), R(22, 11, 11, 33)
+	.long R(d2, 69, 69, bb), R(a9, d9, d9, 70)
+	.long R(07, 8e, 8e, 89), R(33, 94, 94, a7)
+	.long R(2d, 9b, 9b, b6), R(3c, 1e, 1e, 22)
+	.long R(15, 87, 87, 92), R(c9, e9, e9, 20)
+	.long R(87, ce, ce, 49), R(aa, 55, 55, ff)
+	.long R(50, 28, 28, 78), R(a5, df, df, 7a)
+	.long R(03, 8c, 8c, 8f), R(59, a1, a1, f8)
+	.long R(09, 89, 89, 80), R(1a, 0d, 0d, 17)
+	.long R(65, bf, bf, da), R(d7, e6, e6, 31)
+	.long R(84, 42, 42, c6), R(d0, 68, 68, b8)
+	.long R(82, 41, 41, c3), R(29, 99, 99, b0)
+	.long R(5a, 2d, 2d, 77), R(1e, 0f, 0f, 11)
+	.long R(7b, b0, b0, cb), R(a8, 54, 54, fc)
+	.long R(6d, bb, bb, d6), R(2c, 16, 16, 3a)
+.globl PPC_AES_4K_DECTAB
+PPC_AES_4K_DECTAB:
+/* decryption table, same as crypto_it_tab in crypto/aes-generic.c */
+	.long R(51, f4, a7, 50), R(7e, 41, 65, 53)
+	.long R(1a, 17, a4, c3), R(3a, 27, 5e, 96)
+	.long R(3b, ab, 6b, cb), R(1f, 9d, 45, f1)
+	.long R(ac, fa, 58, ab), R(4b, e3, 03, 93)
+	.long R(20, 30, fa, 55), R(ad, 76, 6d, f6)
+	.long R(88, cc, 76, 91), R(f5, 02, 4c, 25)
+	.long R(4f, e5, d7, fc), R(c5, 2a, cb, d7)
+	.long R(26, 35, 44, 80), R(b5, 62, a3, 8f)
+	.long R(de, b1, 5a, 49), R(25, ba, 1b, 67)
+	.long R(45, ea, 0e, 98), R(5d, fe, c0, e1)
+	.long R(c3, 2f, 75, 02), R(81, 4c, f0, 12)
+	.long R(8d, 46, 97, a3), R(6b, d3, f9, c6)
+	.long R(03, 8f, 5f, e7), R(15, 92, 9c, 95)
+	.long R(bf, 6d, 7a, eb), R(95, 52, 59, da)
+	.long R(d4, be, 83, 2d), R(58, 74, 21, d3)
+	.long R(49, e0, 69, 29), R(8e, c9, c8, 44)
+	.long R(75, c2, 89, 6a), R(f4, 8e, 79, 78)
+	.long R(99, 58, 3e, 6b), R(27, b9, 71, dd)
+	.long R(be, e1, 4f, b6), R(f0, 88, ad, 17)
+	.long R(c9, 20, ac, 66), R(7d, ce, 3a, b4)
+	.long R(63, df, 4a, 18), R(e5, 1a, 31, 82)
+	.long R(97, 51, 33, 60), R(62, 53, 7f, 45)
+	.long R(b1, 64, 77, e0), R(bb, 6b, ae, 84)
+	.long R(fe, 81, a0, 1c), R(f9, 08, 2b, 94)
+	.long R(70, 48, 68, 58), R(8f, 45, fd, 19)
+	.long R(94, de, 6c, 87), R(52, 7b, f8, b7)
+	.long R(ab, 73, d3, 23), R(72, 4b, 02, e2)
+	.long R(e3, 1f, 8f, 57), R(66, 55, ab, 2a)
+	.long R(b2, eb, 28, 07), R(2f, b5, c2, 03)
+	.long R(86, c5, 7b, 9a), R(d3, 37, 08, a5)
+	.long R(30, 28, 87, f2), R(23, bf, a5, b2)
+	.long R(02, 03, 6a, ba), R(ed, 16, 82, 5c)
+	.long R(8a, cf, 1c, 2b), R(a7, 79, b4, 92)
+	.long R(f3, 07, f2, f0), R(4e, 69, e2, a1)
+	.long R(65, da, f4, cd), R(06, 05, be, d5)
+	.long R(d1, 34, 62, 1f), R(c4, a6, fe, 8a)
+	.long R(34, 2e, 53, 9d), R(a2, f3, 55, a0)
+	.long R(05, 8a, e1, 32), R(a4, f6, eb, 75)
+	.long R(0b, 83, ec, 39), R(40, 60, ef, aa)
+	.long R(5e, 71, 9f, 06), R(bd, 6e, 10, 51)
+	.long R(3e, 21, 8a, f9), R(96, dd, 06, 3d)
+	.long R(dd, 3e, 05, ae), R(4d, e6, bd, 46)
+	.long R(91, 54, 8d, b5), R(71, c4, 5d, 05)
+	.long R(04, 06, d4, 6f), R(60, 50, 15, ff)
+	.long R(19, 98, fb, 24), R(d6, bd, e9, 97)
+	.long R(89, 40, 43, cc), R(67, d9, 9e, 77)
+	.long R(b0, e8, 42, bd), R(07, 89, 8b, 88)
+	.long R(e7, 19, 5b, 38), R(79, c8, ee, db)
+	.long R(a1, 7c, 0a, 47), R(7c, 42, 0f, e9)
+	.long R(f8, 84, 1e, c9), R(00, 00, 00, 00)
+	.long R(09, 80, 86, 83), R(32, 2b, ed, 48)
+	.long R(1e, 11, 70, ac), R(6c, 5a, 72, 4e)
+	.long R(fd, 0e, ff, fb), R(0f, 85, 38, 56)
+	.long R(3d, ae, d5, 1e), R(36, 2d, 39, 27)
+	.long R(0a, 0f, d9, 64), R(68, 5c, a6, 21)
+	.long R(9b, 5b, 54, d1), R(24, 36, 2e, 3a)
+	.long R(0c, 0a, 67, b1), R(93, 57, e7, 0f)
+	.long R(b4, ee, 96, d2), R(1b, 9b, 91, 9e)
+	.long R(80, c0, c5, 4f), R(61, dc, 20, a2)
+	.long R(5a, 77, 4b, 69), R(1c, 12, 1a, 16)
+	.long R(e2, 93, ba, 0a), R(c0, a0, 2a, e5)
+	.long R(3c, 22, e0, 43), R(12, 1b, 17, 1d)
+	.long R(0e, 09, 0d, 0b), R(f2, 8b, c7, ad)
+	.long R(2d, b6, a8, b9), R(14, 1e, a9, c8)
+	.long R(57, f1, 19, 85), R(af, 75, 07, 4c)
+	.long R(ee, 99, dd, bb), R(a3, 7f, 60, fd)
+	.long R(f7, 01, 26, 9f), R(5c, 72, f5, bc)
+	.long R(44, 66, 3b, c5), R(5b, fb, 7e, 34)
+	.long R(8b, 43, 29, 76), R(cb, 23, c6, dc)
+	.long R(b6, ed, fc, 68), R(b8, e4, f1, 63)
+	.long R(d7, 31, dc, ca), R(42, 63, 85, 10)
+	.long R(13, 97, 22, 40), R(84, c6, 11, 20)
+	.long R(85, 4a, 24, 7d), R(d2, bb, 3d, f8)
+	.long R(ae, f9, 32, 11), R(c7, 29, a1, 6d)
+	.long R(1d, 9e, 2f, 4b), R(dc, b2, 30, f3)
+	.long R(0d, 86, 52, ec), R(77, c1, e3, d0)
+	.long R(2b, b3, 16, 6c), R(a9, 70, b9, 99)
+	.long R(11, 94, 48, fa), R(47, e9, 64, 22)
+	.long R(a8, fc, 8c, c4), R(a0, f0, 3f, 1a)
+	.long R(56, 7d, 2c, d8), R(22, 33, 90, ef)
+	.long R(87, 49, 4e, c7), R(d9, 38, d1, c1)
+	.long R(8c, ca, a2, fe), R(98, d4, 0b, 36)
+	.long R(a6, f5, 81, cf), R(a5, 7a, de, 28)
+	.long R(da, b7, 8e, 26), R(3f, ad, bf, a4)
+	.long R(2c, 3a, 9d, e4), R(50, 78, 92, 0d)
+	.long R(6a, 5f, cc, 9b), R(54, 7e, 46, 62)
+	.long R(f6, 8d, 13, c2), R(90, d8, b8, e8)
+	.long R(2e, 39, f7, 5e), R(82, c3, af, f5)
+	.long R(9f, 5d, 80, be), R(69, d0, 93, 7c)
+	.long R(6f, d5, 2d, a9), R(cf, 25, 12, b3)
+	.long R(c8, ac, 99, 3b), R(10, 18, 7d, a7)
+	.long R(e8, 9c, 63, 6e), R(db, 3b, bb, 7b)
+	.long R(cd, 26, 78, 09), R(6e, 59, 18, f4)
+	.long R(ec, 9a, b7, 01), R(83, 4f, 9a, a8)
+	.long R(e6, 95, 6e, 65), R(aa, ff, e6, 7e)
+	.long R(21, bc, cf, 08), R(ef, 15, e8, e6)
+	.long R(ba, e7, 9b, d9), R(4a, 6f, 36, ce)
+	.long R(ea, 9f, 09, d4), R(29, b0, 7c, d6)
+	.long R(31, a4, b2, af), R(2a, 3f, 23, 31)
+	.long R(c6, a5, 94, 30), R(35, a2, 66, c0)
+	.long R(74, 4e, bc, 37), R(fc, 82, ca, a6)
+	.long R(e0, 90, d0, b0), R(33, a7, d8, 15)
+	.long R(f1, 04, 98, 4a), R(41, ec, da, f7)
+	.long R(7f, cd, 50, 0e), R(17, 91, f6, 2f)
+	.long R(76, 4d, d6, 8d), R(43, ef, b0, 4d)
+	.long R(cc, aa, 4d, 54), R(e4, 96, 04, df)
+	.long R(9e, d1, b5, e3), R(4c, 6a, 88, 1b)
+	.long R(c1, 2c, 1f, b8), R(46, 65, 51, 7f)
+	.long R(9d, 5e, ea, 04), R(01, 8c, 35, 5d)
+	.long R(fa, 87, 74, 73), R(fb, 0b, 41, 2e)
+	.long R(b3, 67, 1d, 5a), R(92, db, d2, 52)
+	.long R(e9, 10, 56, 33), R(6d, d6, 47, 13)
+	.long R(9a, d7, 61, 8c), R(37, a1, 0c, 7a)
+	.long R(59, f8, 14, 8e), R(eb, 13, 3c, 89)
+	.long R(ce, a9, 27, ee), R(b7, 61, c9, 35)
+	.long R(e1, 1c, e5, ed), R(7a, 47, b1, 3c)
+	.long R(9c, d2, df, 59), R(55, f2, 73, 3f)
+	.long R(18, 14, ce, 79), R(73, c7, 37, bf)
+	.long R(53, f7, cd, ea), R(5f, fd, aa, 5b)
+	.long R(df, 3d, 6f, 14), R(78, 44, db, 86)
+	.long R(ca, af, f3, 81), R(b9, 68, c4, 3e)
+	.long R(38, 24, 34, 2c), R(c2, a3, 40, 5f)
+	.long R(16, 1d, c3, 72), R(bc, e2, 25, 0c)
+	.long R(28, 3c, 49, 8b), R(ff, 0d, 95, 41)
+	.long R(39, a8, 01, 71), R(08, 0c, b3, de)
+	.long R(d8, b4, e4, 9c), R(64, 56, c1, 90)
+	.long R(7b, cb, 84, 61), R(d5, 32, b6, 70)
+	.long R(48, 6c, 5c, 74), R(d0, b8, 57, 42)
+.globl PPC_AES_4K_DECTAB2
+PPC_AES_4K_DECTAB2:
+/* decryption table, same as crypto_il_tab in crypto/aes-generic.c */
+	.byte 0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38
+	.byte 0xbf, 0x40, 0xa3, 0x9e, 0x81, 0xf3, 0xd7, 0xfb
+	.byte 0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87
+	.byte 0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb
+	.byte 0x54, 0x7b, 0x94, 0x32, 0xa6, 0xc2, 0x23, 0x3d
+	.byte 0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e
+	.byte 0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2
+	.byte 0x76, 0x5b, 0xa2, 0x49, 0x6d, 0x8b, 0xd1, 0x25
+	.byte 0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16
+	.byte 0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92
+	.byte 0x6c, 0x70, 0x48, 0x50, 0xfd, 0xed, 0xb9, 0xda
+	.byte 0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84
+	.byte 0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a
+	.byte 0xf7, 0xe4, 0x58, 0x05, 0xb8, 0xb3, 0x45, 0x06
+	.byte 0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02
+	.byte 0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b
+	.byte 0x3a, 0x91, 0x11, 0x41, 0x4f, 0x67, 0xdc, 0xea
+	.byte 0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73
+	.byte 0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85
+	.byte 0xe2, 0xf9, 0x37, 0xe8, 0x1c, 0x75, 0xdf, 0x6e
+	.byte 0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89
+	.byte 0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b
+	.byte 0xfc, 0x56, 0x3e, 0x4b, 0xc6, 0xd2, 0x79, 0x20
+	.byte 0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4
+	.byte 0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31
+	.byte 0xb1, 0x12, 0x10, 0x59, 0x27, 0x80, 0xec, 0x5f
+	.byte 0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d
+	.byte 0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef
+	.byte 0xa0, 0xe0, 0x3b, 0x4d, 0xae, 0x2a, 0xf5, 0xb0
+	.byte 0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61
+	.byte 0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26
+	.byte 0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d
diff --git a/lib/crypto/powerpc/aes.h b/lib/crypto/powerpc/aes.h
new file mode 100644
index 000000000..5a36b637e
--- /dev/null
+++ b/lib/crypto/powerpc/aes.h
@@ -0,0 +1,240 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ * Copyright (C) 2015 International Business Machines Inc.
+ * Copyright 2026 Google LLC
+ */
+#include <asm/simd.h>
+#include <asm/switch_to.h>
+#include <linux/cpufeature.h>
+#include <linux/jump_label.h>
+#include <linux/preempt.h>
+#include <linux/uaccess.h>
+
+#ifdef CONFIG_SPE
+
+EXPORT_SYMBOL_GPL(ppc_expand_key_128);
+EXPORT_SYMBOL_GPL(ppc_expand_key_192);
+EXPORT_SYMBOL_GPL(ppc_expand_key_256);
+EXPORT_SYMBOL_GPL(ppc_generate_decrypt_key);
+EXPORT_SYMBOL_GPL(ppc_encrypt_ecb);
+EXPORT_SYMBOL_GPL(ppc_decrypt_ecb);
+EXPORT_SYMBOL_GPL(ppc_encrypt_cbc);
+EXPORT_SYMBOL_GPL(ppc_decrypt_cbc);
+EXPORT_SYMBOL_GPL(ppc_crypt_ctr);
+EXPORT_SYMBOL_GPL(ppc_encrypt_xts);
+EXPORT_SYMBOL_GPL(ppc_decrypt_xts);
+
+void ppc_encrypt_aes(u8 *out, const u8 *in, const u32 *key_enc, u32 rounds);
+void ppc_decrypt_aes(u8 *out, const u8 *in, const u32 *key_dec, u32 rounds);
+
+static void spe_begin(void)
+{
+	/* disable preemption and save users SPE registers if required */
+	preempt_disable();
+	enable_kernel_spe();
+}
+
+static void spe_end(void)
+{
+	disable_kernel_spe();
+	/* reenable preemption */
+	preempt_enable();
+}
+
+static void aes_preparekey_arch(union aes_enckey_arch *k,
+				union aes_invkey_arch *inv_k,
+				const u8 *in_key, int key_len, int nrounds)
+{
+	if (key_len == AES_KEYSIZE_128)
+		ppc_expand_key_128(k->spe_enc_key, in_key);
+	else if (key_len == AES_KEYSIZE_192)
+		ppc_expand_key_192(k->spe_enc_key, in_key);
+	else
+		ppc_expand_key_256(k->spe_enc_key, in_key);
+
+	if (inv_k)
+		ppc_generate_decrypt_key(inv_k->spe_dec_key, k->spe_enc_key,
+					 key_len);
+}
+
+static void aes_encrypt_arch(const struct aes_enckey *key,
+			     u8 out[AES_BLOCK_SIZE],
+			     const u8 in[AES_BLOCK_SIZE])
+{
+	spe_begin();
+	ppc_encrypt_aes(out, in, key->k.spe_enc_key, key->nrounds / 2 - 1);
+	spe_end();
+}
+
+static void aes_decrypt_arch(const struct aes_key *key,
+			     u8 out[AES_BLOCK_SIZE],
+			     const u8 in[AES_BLOCK_SIZE])
+{
+	spe_begin();
+	ppc_decrypt_aes(out, in, key->inv_k.spe_dec_key, key->nrounds / 2 - 1);
+	spe_end();
+}
+
+#else /* CONFIG_SPE */
+
+static __ro_after_init DEFINE_STATIC_KEY_FALSE(have_vec_crypto);
+
+EXPORT_SYMBOL_GPL(aes_p8_set_encrypt_key);
+EXPORT_SYMBOL_GPL(aes_p8_set_decrypt_key);
+EXPORT_SYMBOL_GPL(aes_p8_encrypt);
+EXPORT_SYMBOL_GPL(aes_p8_decrypt);
+EXPORT_SYMBOL_GPL(aes_p8_cbc_encrypt);
+EXPORT_SYMBOL_GPL(aes_p8_ctr32_encrypt_blocks);
+EXPORT_SYMBOL_GPL(aes_p8_xts_encrypt);
+EXPORT_SYMBOL_GPL(aes_p8_xts_decrypt);
+
+static inline bool is_vsx_format(const struct p8_aes_key *key)
+{
+	return key->nrounds != 0;
+}
+
+/*
+ * Convert a round key from VSX to generic format by reflecting all 16 bytes (if
+ * little endian) or reflecting the bytes in each 4-byte word (if big endian),
+ * and (if apply_inv_mix=true) applying InvMixColumn to each column.
+ *
+ * It would be nice if the VSX and generic key formats would be compatible.  But
+ * that's very difficult to do, with the assembly code having been borrowed from
+ * OpenSSL and also targeted to POWER8 rather than POWER9.
+ *
+ * Fortunately, this conversion should only be needed in extremely rare cases,
+ * possibly not at all in practice.  It's just included for full correctness.
+ */
+static void rndkey_from_vsx(u32 out[4], const u32 in[4], bool apply_inv_mix)
+{
+	const bool be = IS_ENABLED(CONFIG_CPU_BIG_ENDIAN);
+	u32 k0 = swab32(in[0]);
+	u32 k1 = swab32(in[1]);
+	u32 k2 = swab32(in[2]);
+	u32 k3 = swab32(in[3]);
+
+	if (apply_inv_mix) {
+		k0 = inv_mix_columns(k0);
+		k1 = inv_mix_columns(k1);
+		k2 = inv_mix_columns(k2);
+		k3 = inv_mix_columns(k3);
+	}
+	out[0] = be ? k0 : k3;
+	out[1] = be ? k1 : k2;
+	out[2] = be ? k2 : k1;
+	out[3] = be ? k3 : k0;
+}
+
+static void aes_preparekey_arch(union aes_enckey_arch *k,
+				union aes_invkey_arch *inv_k,
+				const u8 *in_key, int key_len, int nrounds)
+{
+	const int keybits = 8 * key_len;
+	int ret;
+
+	if (static_branch_likely(&have_vec_crypto) && likely(may_use_simd())) {
+		preempt_disable();
+		pagefault_disable();
+		enable_kernel_vsx();
+		ret = aes_p8_set_encrypt_key(in_key, keybits, &k->p8);
+		/*
+		 * aes_p8_set_encrypt_key() should never fail here, since the
+		 * key length was already validated.
+		 */
+		WARN_ON_ONCE(ret);
+		if (inv_k) {
+			ret = aes_p8_set_decrypt_key(in_key, keybits,
+						     &inv_k->p8);
+			/* ... and likewise for aes_p8_set_decrypt_key(). */
+			WARN_ON_ONCE(ret);
+		}
+		disable_kernel_vsx();
+		pagefault_enable();
+		preempt_enable();
+	} else {
+		aes_expandkey_generic(k->rndkeys,
+				      inv_k ? inv_k->inv_rndkeys : NULL,
+				      in_key, key_len);
+		/* Mark the key as using the generic format. */
+		k->p8.nrounds = 0;
+		if (inv_k)
+			inv_k->p8.nrounds = 0;
+	}
+}
+
+static void aes_encrypt_arch(const struct aes_enckey *key,
+			     u8 out[AES_BLOCK_SIZE],
+			     const u8 in[AES_BLOCK_SIZE])
+{
+	if (static_branch_likely(&have_vec_crypto) &&
+	    likely(is_vsx_format(&key->k.p8) && may_use_simd())) {
+		preempt_disable();
+		pagefault_disable();
+		enable_kernel_vsx();
+		aes_p8_encrypt(in, out, &key->k.p8);
+		disable_kernel_vsx();
+		pagefault_enable();
+		preempt_enable();
+	} else if (unlikely(is_vsx_format(&key->k.p8))) {
+		/*
+		 * This handles (the hopefully extremely rare) case where a key
+		 * was prepared using the VSX optimized format, then encryption
+		 * is done in a context that cannot use VSX instructions.
+		 */
+		u32 rndkeys[AES_MAX_KEYLENGTH_U32];
+
+		for (int i = 0; i < 4 * (key->nrounds + 1); i += 4)
+			rndkey_from_vsx(&rndkeys[i],
+					&key->k.p8.rndkeys[i], false);
+		aes_encrypt_generic(rndkeys, key->nrounds, out, in);
+	} else {
+		aes_encrypt_generic(key->k.rndkeys, key->nrounds, out, in);
+	}
+}
+
+static void aes_decrypt_arch(const struct aes_key *key, u8 out[AES_BLOCK_SIZE],
+			     const u8 in[AES_BLOCK_SIZE])
+{
+	if (static_branch_likely(&have_vec_crypto) &&
+	    likely(is_vsx_format(&key->inv_k.p8) && may_use_simd())) {
+		preempt_disable();
+		pagefault_disable();
+		enable_kernel_vsx();
+		aes_p8_decrypt(in, out, &key->inv_k.p8);
+		disable_kernel_vsx();
+		pagefault_enable();
+		preempt_enable();
+	} else if (unlikely(is_vsx_format(&key->inv_k.p8))) {
+		/*
+		 * This handles (the hopefully extremely rare) case where a key
+		 * was prepared using the VSX optimized format, then decryption
+		 * is done in a context that cannot use VSX instructions.
+		 */
+		u32 inv_rndkeys[AES_MAX_KEYLENGTH_U32];
+		int i;
+
+		rndkey_from_vsx(&inv_rndkeys[0],
+				&key->inv_k.p8.rndkeys[0], false);
+		for (i = 4; i < 4 * key->nrounds; i += 4) {
+			rndkey_from_vsx(&inv_rndkeys[i],
+					&key->inv_k.p8.rndkeys[i], true);
+		}
+		rndkey_from_vsx(&inv_rndkeys[i],
+				&key->inv_k.p8.rndkeys[i], false);
+		aes_decrypt_generic(inv_rndkeys, key->nrounds, out, in);
+	} else {
+		aes_decrypt_generic(key->inv_k.inv_rndkeys, key->nrounds,
+				    out, in);
+	}
+}
+
+#define aes_mod_init_arch aes_mod_init_arch
+static void aes_mod_init_arch(void)
+{
+	if (cpu_has_feature(CPU_FTR_ARCH_207S) &&
+	    (cur_cpu_spec->cpu_user_features2 & PPC_FEATURE2_VEC_CRYPTO))
+		static_branch_enable(&have_vec_crypto);
+}
+
+#endif /* !CONFIG_SPE */
diff --git a/lib/crypto/powerpc/aesp8-ppc.pl b/lib/crypto/powerpc/aesp8-ppc.pl
new file mode 100644
index 000000000..253a06758
--- /dev/null
+++ b/lib/crypto/powerpc/aesp8-ppc.pl
@@ -0,0 +1,3890 @@
+#! /usr/bin/env perl
+# SPDX-License-Identifier: GPL-2.0
+
+# This code is taken from CRYPTOGAMs[1] and is included here using the option
+# in the license to distribute the code under the GPL. Therefore 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.
+#
+# [1] https://www.openssl.org/~appro/cryptogams/
+
+# Copyright (c) 2006-2017, CRYPTOGAMS by <appro@openssl.org>
+# All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+#
+#       * Redistributions of source code must retain copyright notices,
+#         this list of conditions and the following disclaimer.
+#
+#       * Redistributions in binary form must reproduce the above
+#         copyright notice, this list of conditions and the following
+#         disclaimer in the documentation and/or other materials
+#         provided with the distribution.
+#
+#       * Neither the name of the CRYPTOGAMS nor the names of its
+#         copyright holder and contributors may be used to endorse or
+#         promote products derived from this software without specific
+#         prior written permission.
+#
+# ALTERNATIVELY, provided that this notice is retained in full, this
+# product may be distributed under the terms of the GNU General Public
+# License (GPL), in which case the provisions of the GPL apply INSTEAD OF
+# those given above.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER AND CONTRIBUTORS
+# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+# ====================================================================
+# Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
+# project. The module is, however, dual licensed under OpenSSL and
+# CRYPTOGAMS licenses depending on where you obtain it. For further
+# details see https://www.openssl.org/~appro/cryptogams/.
+# ====================================================================
+#
+# This module implements support for AES instructions as per PowerISA
+# specification version 2.07, first implemented by POWER8 processor.
+# The module is endian-agnostic in sense that it supports both big-
+# and little-endian cases. Data alignment in parallelizable modes is
+# handled with VSX loads and stores, which implies MSR.VSX flag being
+# set. It should also be noted that ISA specification doesn't prohibit
+# alignment exceptions for these instructions on page boundaries.
+# Initially alignment was handled in pure AltiVec/VMX way [when data
+# is aligned programmatically, which in turn guarantees exception-
+# free execution], but it turned to hamper performance when vcipher
+# instructions are interleaved. It's reckoned that eventual
+# misalignment penalties at page boundaries are in average lower
+# than additional overhead in pure AltiVec approach.
+#
+# May 2016
+#
+# Add XTS subroutine, 9x on little- and 12x improvement on big-endian
+# systems were measured.
+#
+######################################################################
+# Current large-block performance in cycles per byte processed with
+# 128-bit key (less is better).
+#
+#		CBC en-/decrypt	CTR	XTS
+# POWER8[le]	3.96/0.72	0.74	1.1
+# POWER8[be]	3.75/0.65	0.66	1.0
+
+$flavour = shift;
+
+if ($flavour =~ /64/) {
+	$SIZE_T	=8;
+	$LRSAVE	=2*$SIZE_T;
+	$STU	="stdu";
+	$POP	="ld";
+	$PUSH	="std";
+	$UCMP	="cmpld";
+	$SHL	="sldi";
+} elsif ($flavour =~ /32/) {
+	$SIZE_T	=4;
+	$LRSAVE	=$SIZE_T;
+	$STU	="stwu";
+	$POP	="lwz";
+	$PUSH	="stw";
+	$UCMP	="cmplw";
+	$SHL	="slwi";
+} else { die "nonsense $flavour"; }
+
+$LITTLE_ENDIAN = ($flavour=~/le$/) ? $SIZE_T : 0;
+
+$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
+( $xlate="${dir}ppc-xlate.pl" and -f $xlate ) or
+( $xlate="${dir}../../perlasm/ppc-xlate.pl" and -f $xlate) or
+( $xlate="${dir}../../../arch/powerpc/crypto/ppc-xlate.pl" and -f $xlate) or
+die "can't locate ppc-xlate.pl";
+
+open STDOUT,"| $^X $xlate $flavour ".shift || die "can't call $xlate: $!";
+
+$FRAME=8*$SIZE_T;
+$prefix="aes_p8";
+
+$sp="r1";
+$vrsave="r12";
+
+#########################################################################
+{{{	# Key setup procedures						#
+my ($inp,$bits,$out,$ptr,$cnt,$rounds)=map("r$_",(3..8));
+my ($zero,$in0,$in1,$key,$rcon,$mask,$tmp)=map("v$_",(0..6));
+my ($stage,$outperm,$outmask,$outhead,$outtail)=map("v$_",(7..11));
+
+$code.=<<___;
+.machine	"any"
+
+.text
+
+.align	7
+rcon:
+.long	0x01000000, 0x01000000, 0x01000000, 0x01000000	?rev
+.long	0x1b000000, 0x1b000000, 0x1b000000, 0x1b000000	?rev
+.long	0x0d0e0f0c, 0x0d0e0f0c, 0x0d0e0f0c, 0x0d0e0f0c	?rev
+.long	0,0,0,0						?asis
+.long	0x0f102132, 0x43546576, 0x8798a9ba, 0xcbdcedfe
+Lconsts:
+	mflr	r0
+	bcl	20,31,\$+4
+	mflr	$ptr	 #vvvvv "distance between . and rcon
+	addi	$ptr,$ptr,-0x58
+	mtlr	r0
+	blr
+	.long	0
+	.byte	0,12,0x14,0,0,0,0,0
+.asciz	"AES for PowerISA 2.07, CRYPTOGAMS by <appro\@openssl.org>"
+
+.globl	.${prefix}_set_encrypt_key
+Lset_encrypt_key:
+	mflr		r11
+	$PUSH		r11,$LRSAVE($sp)
+
+	li		$ptr,-1
+	${UCMP}i	$inp,0
+	beq-		Lenc_key_abort		# if ($inp==0) return -1;
+	${UCMP}i	$out,0
+	beq-		Lenc_key_abort		# if ($out==0) return -1;
+	li		$ptr,-2
+	cmpwi		$bits,128
+	blt-		Lenc_key_abort
+	cmpwi		$bits,256
+	bgt-		Lenc_key_abort
+	andi.		r0,$bits,0x3f
+	bne-		Lenc_key_abort
+
+	lis		r0,0xfff0
+	mfspr		$vrsave,256
+	mtspr		256,r0
+
+	bl		Lconsts
+	mtlr		r11
+
+	neg		r9,$inp
+	lvx		$in0,0,$inp
+	addi		$inp,$inp,15		# 15 is not typo
+	lvsr		$key,0,r9		# borrow $key
+	li		r8,0x20
+	cmpwi		$bits,192
+	lvx		$in1,0,$inp
+	le?vspltisb	$mask,0x0f		# borrow $mask
+	lvx		$rcon,0,$ptr
+	le?vxor		$key,$key,$mask		# adjust for byte swap
+	lvx		$mask,r8,$ptr
+	addi		$ptr,$ptr,0x10
+	vperm		$in0,$in0,$in1,$key	# align [and byte swap in LE]
+	li		$cnt,8
+	vxor		$zero,$zero,$zero
+	mtctr		$cnt
+
+	?lvsr		$outperm,0,$out
+	vspltisb	$outmask,-1
+	lvx		$outhead,0,$out
+	?vperm		$outmask,$zero,$outmask,$outperm
+
+	blt		Loop128
+	addi		$inp,$inp,8
+	beq		L192
+	addi		$inp,$inp,8
+	b		L256
+
+.align	4
+Loop128:
+	vperm		$key,$in0,$in0,$mask	# rotate-n-splat
+	vsldoi		$tmp,$zero,$in0,12	# >>32
+	 vperm		$outtail,$in0,$in0,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	vcipherlast	$key,$key,$rcon
+	 stvx		$stage,0,$out
+	 addi		$out,$out,16
+
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	 vadduwm	$rcon,$rcon,$rcon
+	vxor		$in0,$in0,$key
+	bdnz		Loop128
+
+	lvx		$rcon,0,$ptr		# last two round keys
+
+	vperm		$key,$in0,$in0,$mask	# rotate-n-splat
+	vsldoi		$tmp,$zero,$in0,12	# >>32
+	 vperm		$outtail,$in0,$in0,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	vcipherlast	$key,$key,$rcon
+	 stvx		$stage,0,$out
+	 addi		$out,$out,16
+
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	 vadduwm	$rcon,$rcon,$rcon
+	vxor		$in0,$in0,$key
+
+	vperm		$key,$in0,$in0,$mask	# rotate-n-splat
+	vsldoi		$tmp,$zero,$in0,12	# >>32
+	 vperm		$outtail,$in0,$in0,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	vcipherlast	$key,$key,$rcon
+	 stvx		$stage,0,$out
+	 addi		$out,$out,16
+
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	vxor		$in0,$in0,$key
+	 vperm		$outtail,$in0,$in0,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	 stvx		$stage,0,$out
+
+	addi		$inp,$out,15		# 15 is not typo
+	addi		$out,$out,0x50
+
+	li		$rounds,10
+	b		Ldone
+
+.align	4
+L192:
+	lvx		$tmp,0,$inp
+	li		$cnt,4
+	 vperm		$outtail,$in0,$in0,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	 stvx		$stage,0,$out
+	 addi		$out,$out,16
+	vperm		$in1,$in1,$tmp,$key	# align [and byte swap in LE]
+	vspltisb	$key,8			# borrow $key
+	mtctr		$cnt
+	vsububm		$mask,$mask,$key	# adjust the mask
+
+Loop192:
+	vperm		$key,$in1,$in1,$mask	# roate-n-splat
+	vsldoi		$tmp,$zero,$in0,12	# >>32
+	vcipherlast	$key,$key,$rcon
+
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+
+	 vsldoi		$stage,$zero,$in1,8
+	vspltw		$tmp,$in0,3
+	vxor		$tmp,$tmp,$in1
+	vsldoi		$in1,$zero,$in1,12	# >>32
+	 vadduwm	$rcon,$rcon,$rcon
+	vxor		$in1,$in1,$tmp
+	vxor		$in0,$in0,$key
+	vxor		$in1,$in1,$key
+	 vsldoi		$stage,$stage,$in0,8
+
+	vperm		$key,$in1,$in1,$mask	# rotate-n-splat
+	vsldoi		$tmp,$zero,$in0,12	# >>32
+	 vperm		$outtail,$stage,$stage,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	vcipherlast	$key,$key,$rcon
+	 stvx		$stage,0,$out
+	 addi		$out,$out,16
+
+	 vsldoi		$stage,$in0,$in1,8
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	 vperm		$outtail,$stage,$stage,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	 stvx		$stage,0,$out
+	 addi		$out,$out,16
+
+	vspltw		$tmp,$in0,3
+	vxor		$tmp,$tmp,$in1
+	vsldoi		$in1,$zero,$in1,12	# >>32
+	 vadduwm	$rcon,$rcon,$rcon
+	vxor		$in1,$in1,$tmp
+	vxor		$in0,$in0,$key
+	vxor		$in1,$in1,$key
+	 vperm		$outtail,$in0,$in0,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	 stvx		$stage,0,$out
+	 addi		$inp,$out,15		# 15 is not typo
+	 addi		$out,$out,16
+	bdnz		Loop192
+
+	li		$rounds,12
+	addi		$out,$out,0x20
+	b		Ldone
+
+.align	4
+L256:
+	lvx		$tmp,0,$inp
+	li		$cnt,7
+	li		$rounds,14
+	 vperm		$outtail,$in0,$in0,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	 stvx		$stage,0,$out
+	 addi		$out,$out,16
+	vperm		$in1,$in1,$tmp,$key	# align [and byte swap in LE]
+	mtctr		$cnt
+
+Loop256:
+	vperm		$key,$in1,$in1,$mask	# rotate-n-splat
+	vsldoi		$tmp,$zero,$in0,12	# >>32
+	 vperm		$outtail,$in1,$in1,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	vcipherlast	$key,$key,$rcon
+	 stvx		$stage,0,$out
+	 addi		$out,$out,16
+
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in0,$in0,$tmp
+	 vadduwm	$rcon,$rcon,$rcon
+	vxor		$in0,$in0,$key
+	 vperm		$outtail,$in0,$in0,$outperm	# rotate
+	 vsel		$stage,$outhead,$outtail,$outmask
+	 vmr		$outhead,$outtail
+	 stvx		$stage,0,$out
+	 addi		$inp,$out,15		# 15 is not typo
+	 addi		$out,$out,16
+	bdz		Ldone
+
+	vspltw		$key,$in0,3		# just splat
+	vsldoi		$tmp,$zero,$in1,12	# >>32
+	vsbox		$key,$key
+
+	vxor		$in1,$in1,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in1,$in1,$tmp
+	vsldoi		$tmp,$zero,$tmp,12	# >>32
+	vxor		$in1,$in1,$tmp
+
+	vxor		$in1,$in1,$key
+	b		Loop256
+
+.align	4
+Ldone:
+	lvx		$in1,0,$inp		# redundant in aligned case
+	vsel		$in1,$outhead,$in1,$outmask
+	stvx		$in1,0,$inp
+	li		$ptr,0
+	mtspr		256,$vrsave
+	stw		$rounds,0($out)
+
+Lenc_key_abort:
+	mr		r3,$ptr
+	blr
+	.long		0
+	.byte		0,12,0x14,1,0,0,3,0
+	.long		0
+.size	.${prefix}_set_encrypt_key,.-.${prefix}_set_encrypt_key
+
+.globl	.${prefix}_set_decrypt_key
+	$STU		$sp,-$FRAME($sp)
+	mflr		r10
+	$PUSH		r10,$FRAME+$LRSAVE($sp)
+	bl		Lset_encrypt_key
+	mtlr		r10
+
+	cmpwi		r3,0
+	bne-		Ldec_key_abort
+
+	slwi		$cnt,$rounds,4
+	subi		$inp,$out,240		# first round key
+	srwi		$rounds,$rounds,1
+	add		$out,$inp,$cnt		# last round key
+	mtctr		$rounds
+
+Ldeckey:
+	lwz		r0, 0($inp)
+	lwz		r6, 4($inp)
+	lwz		r7, 8($inp)
+	lwz		r8, 12($inp)
+	addi		$inp,$inp,16
+	lwz		r9, 0($out)
+	lwz		r10,4($out)
+	lwz		r11,8($out)
+	lwz		r12,12($out)
+	stw		r0, 0($out)
+	stw		r6, 4($out)
+	stw		r7, 8($out)
+	stw		r8, 12($out)
+	subi		$out,$out,16
+	stw		r9, -16($inp)
+	stw		r10,-12($inp)
+	stw		r11,-8($inp)
+	stw		r12,-4($inp)
+	bdnz		Ldeckey
+
+	xor		r3,r3,r3		# return value
+Ldec_key_abort:
+	addi		$sp,$sp,$FRAME
+	blr
+	.long		0
+	.byte		0,12,4,1,0x80,0,3,0
+	.long		0
+.size	.${prefix}_set_decrypt_key,.-.${prefix}_set_decrypt_key
+___
+}}}
+#########################################################################
+{{{	# Single block en- and decrypt procedures			#
+sub gen_block () {
+my $dir = shift;
+my $n   = $dir eq "de" ? "n" : "";
+my ($inp,$out,$key,$rounds,$idx)=map("r$_",(3..7));
+
+$code.=<<___;
+.globl	.${prefix}_${dir}crypt
+	lwz		$rounds,240($key)
+	lis		r0,0xfc00
+	mfspr		$vrsave,256
+	li		$idx,15			# 15 is not typo
+	mtspr		256,r0
+
+	lvx		v0,0,$inp
+	neg		r11,$out
+	lvx		v1,$idx,$inp
+	lvsl		v2,0,$inp		# inpperm
+	le?vspltisb	v4,0x0f
+	?lvsl		v3,0,r11		# outperm
+	le?vxor		v2,v2,v4
+	li		$idx,16
+	vperm		v0,v0,v1,v2		# align [and byte swap in LE]
+	lvx		v1,0,$key
+	?lvsl		v5,0,$key		# keyperm
+	srwi		$rounds,$rounds,1
+	lvx		v2,$idx,$key
+	addi		$idx,$idx,16
+	subi		$rounds,$rounds,1
+	?vperm		v1,v1,v2,v5		# align round key
+
+	vxor		v0,v0,v1
+	lvx		v1,$idx,$key
+	addi		$idx,$idx,16
+	mtctr		$rounds
+
+Loop_${dir}c:
+	?vperm		v2,v2,v1,v5
+	v${n}cipher	v0,v0,v2
+	lvx		v2,$idx,$key
+	addi		$idx,$idx,16
+	?vperm		v1,v1,v2,v5
+	v${n}cipher	v0,v0,v1
+	lvx		v1,$idx,$key
+	addi		$idx,$idx,16
+	bdnz		Loop_${dir}c
+
+	?vperm		v2,v2,v1,v5
+	v${n}cipher	v0,v0,v2
+	lvx		v2,$idx,$key
+	?vperm		v1,v1,v2,v5
+	v${n}cipherlast	v0,v0,v1
+
+	vspltisb	v2,-1
+	vxor		v1,v1,v1
+	li		$idx,15			# 15 is not typo
+	?vperm		v2,v1,v2,v3		# outmask
+	le?vxor		v3,v3,v4
+	lvx		v1,0,$out		# outhead
+	vperm		v0,v0,v0,v3		# rotate [and byte swap in LE]
+	vsel		v1,v1,v0,v2
+	lvx		v4,$idx,$out
+	stvx		v1,0,$out
+	vsel		v0,v0,v4,v2
+	stvx		v0,$idx,$out
+
+	mtspr		256,$vrsave
+	blr
+	.long		0
+	.byte		0,12,0x14,0,0,0,3,0
+	.long		0
+.size	.${prefix}_${dir}crypt,.-.${prefix}_${dir}crypt
+___
+}
+&gen_block("en");
+&gen_block("de");
+}}}
+#########################################################################
+{{{	# CBC en- and decrypt procedures				#
+my ($inp,$out,$len,$key,$ivp,$enc,$rounds,$idx)=map("r$_",(3..10));
+my ($rndkey0,$rndkey1,$inout,$tmp)=		map("v$_",(0..3));
+my ($ivec,$inptail,$inpperm,$outhead,$outperm,$outmask,$keyperm)=
+						map("v$_",(4..10));
+$code.=<<___;
+.globl	.${prefix}_cbc_encrypt
+	${UCMP}i	$len,16
+	bltlr-
+
+	cmpwi		$enc,0			# test direction
+	lis		r0,0xffe0
+	mfspr		$vrsave,256
+	mtspr		256,r0
+
+	li		$idx,15
+	vxor		$rndkey0,$rndkey0,$rndkey0
+	le?vspltisb	$tmp,0x0f
+
+	lvx		$ivec,0,$ivp		# load [unaligned] iv
+	lvsl		$inpperm,0,$ivp
+	lvx		$inptail,$idx,$ivp
+	le?vxor		$inpperm,$inpperm,$tmp
+	vperm		$ivec,$ivec,$inptail,$inpperm
+
+	neg		r11,$inp
+	?lvsl		$keyperm,0,$key		# prepare for unaligned key
+	lwz		$rounds,240($key)
+
+	lvsr		$inpperm,0,r11		# prepare for unaligned load
+	lvx		$inptail,0,$inp
+	addi		$inp,$inp,15		# 15 is not typo
+	le?vxor		$inpperm,$inpperm,$tmp
+
+	?lvsr		$outperm,0,$out		# prepare for unaligned store
+	vspltisb	$outmask,-1
+	lvx		$outhead,0,$out
+	?vperm		$outmask,$rndkey0,$outmask,$outperm
+	le?vxor		$outperm,$outperm,$tmp
+
+	srwi		$rounds,$rounds,1
+	li		$idx,16
+	subi		$rounds,$rounds,1
+	beq		Lcbc_dec
+
+Lcbc_enc:
+	vmr		$inout,$inptail
+	lvx		$inptail,0,$inp
+	addi		$inp,$inp,16
+	mtctr		$rounds
+	subi		$len,$len,16		# len-=16
+
+	lvx		$rndkey0,0,$key
+	 vperm		$inout,$inout,$inptail,$inpperm
+	lvx		$rndkey1,$idx,$key
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key
+	addi		$idx,$idx,16
+	vxor		$inout,$inout,$ivec
+
+Loop_cbc_enc:
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vcipher		$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key
+	addi		$idx,$idx,16
+	bdnz		Loop_cbc_enc
+
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key
+	li		$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vcipherlast	$ivec,$inout,$rndkey0
+	${UCMP}i	$len,16
+
+	vperm		$tmp,$ivec,$ivec,$outperm
+	vsel		$inout,$outhead,$tmp,$outmask
+	vmr		$outhead,$tmp
+	stvx		$inout,0,$out
+	addi		$out,$out,16
+	bge		Lcbc_enc
+
+	b		Lcbc_done
+
+.align	4
+Lcbc_dec:
+	${UCMP}i	$len,128
+	bge		_aesp8_cbc_decrypt8x
+	vmr		$tmp,$inptail
+	lvx		$inptail,0,$inp
+	addi		$inp,$inp,16
+	mtctr		$rounds
+	subi		$len,$len,16		# len-=16
+
+	lvx		$rndkey0,0,$key
+	 vperm		$tmp,$tmp,$inptail,$inpperm
+	lvx		$rndkey1,$idx,$key
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$inout,$tmp,$rndkey0
+	lvx		$rndkey0,$idx,$key
+	addi		$idx,$idx,16
+
+Loop_cbc_dec:
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vncipher	$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vncipher	$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key
+	addi		$idx,$idx,16
+	bdnz		Loop_cbc_dec
+
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vncipher	$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key
+	li		$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vncipherlast	$inout,$inout,$rndkey0
+	${UCMP}i	$len,16
+
+	vxor		$inout,$inout,$ivec
+	vmr		$ivec,$tmp
+	vperm		$tmp,$inout,$inout,$outperm
+	vsel		$inout,$outhead,$tmp,$outmask
+	vmr		$outhead,$tmp
+	stvx		$inout,0,$out
+	addi		$out,$out,16
+	bge		Lcbc_dec
+
+Lcbc_done:
+	addi		$out,$out,-1
+	lvx		$inout,0,$out		# redundant in aligned case
+	vsel		$inout,$outhead,$inout,$outmask
+	stvx		$inout,0,$out
+
+	neg		$enc,$ivp		# write [unaligned] iv
+	li		$idx,15			# 15 is not typo
+	vxor		$rndkey0,$rndkey0,$rndkey0
+	vspltisb	$outmask,-1
+	le?vspltisb	$tmp,0x0f
+	?lvsl		$outperm,0,$enc
+	?vperm		$outmask,$rndkey0,$outmask,$outperm
+	le?vxor		$outperm,$outperm,$tmp
+	lvx		$outhead,0,$ivp
+	vperm		$ivec,$ivec,$ivec,$outperm
+	vsel		$inout,$outhead,$ivec,$outmask
+	lvx		$inptail,$idx,$ivp
+	stvx		$inout,0,$ivp
+	vsel		$inout,$ivec,$inptail,$outmask
+	stvx		$inout,$idx,$ivp
+
+	mtspr		256,$vrsave
+	blr
+	.long		0
+	.byte		0,12,0x14,0,0,0,6,0
+	.long		0
+___
+#########################################################################
+{{	# Optimized CBC decrypt procedure				#
+my $key_="r11";
+my ($x00,$x10,$x20,$x30,$x40,$x50,$x60,$x70)=map("r$_",(0,8,26..31));
+my ($in0, $in1, $in2, $in3, $in4, $in5, $in6, $in7 )=map("v$_",(0..3,10..13));
+my ($out0,$out1,$out2,$out3,$out4,$out5,$out6,$out7)=map("v$_",(14..21));
+my $rndkey0="v23";	# v24-v25 rotating buffer for first found keys
+			# v26-v31 last 6 round keys
+my ($tmp,$keyperm)=($in3,$in4);	# aliases with "caller", redundant assignment
+
+$code.=<<___;
+.align	5
+_aesp8_cbc_decrypt8x:
+	$STU		$sp,-`($FRAME+21*16+6*$SIZE_T)`($sp)
+	li		r10,`$FRAME+8*16+15`
+	li		r11,`$FRAME+8*16+31`
+	stvx		v20,r10,$sp		# ABI says so
+	addi		r10,r10,32
+	stvx		v21,r11,$sp
+	addi		r11,r11,32
+	stvx		v22,r10,$sp
+	addi		r10,r10,32
+	stvx		v23,r11,$sp
+	addi		r11,r11,32
+	stvx		v24,r10,$sp
+	addi		r10,r10,32
+	stvx		v25,r11,$sp
+	addi		r11,r11,32
+	stvx		v26,r10,$sp
+	addi		r10,r10,32
+	stvx		v27,r11,$sp
+	addi		r11,r11,32
+	stvx		v28,r10,$sp
+	addi		r10,r10,32
+	stvx		v29,r11,$sp
+	addi		r11,r11,32
+	stvx		v30,r10,$sp
+	stvx		v31,r11,$sp
+	li		r0,-1
+	stw		$vrsave,`$FRAME+21*16-4`($sp)	# save vrsave
+	li		$x10,0x10
+	$PUSH		r26,`$FRAME+21*16+0*$SIZE_T`($sp)
+	li		$x20,0x20
+	$PUSH		r27,`$FRAME+21*16+1*$SIZE_T`($sp)
+	li		$x30,0x30
+	$PUSH		r28,`$FRAME+21*16+2*$SIZE_T`($sp)
+	li		$x40,0x40
+	$PUSH		r29,`$FRAME+21*16+3*$SIZE_T`($sp)
+	li		$x50,0x50
+	$PUSH		r30,`$FRAME+21*16+4*$SIZE_T`($sp)
+	li		$x60,0x60
+	$PUSH		r31,`$FRAME+21*16+5*$SIZE_T`($sp)
+	li		$x70,0x70
+	mtspr		256,r0
+
+	subi		$rounds,$rounds,3	# -4 in total
+	subi		$len,$len,128		# bias
+
+	lvx		$rndkey0,$x00,$key	# load key schedule
+	lvx		v30,$x10,$key
+	addi		$key,$key,0x20
+	lvx		v31,$x00,$key
+	?vperm		$rndkey0,$rndkey0,v30,$keyperm
+	addi		$key_,$sp,$FRAME+15
+	mtctr		$rounds
+
+Load_cbc_dec_key:
+	?vperm		v24,v30,v31,$keyperm
+	lvx		v30,$x10,$key
+	addi		$key,$key,0x20
+	stvx		v24,$x00,$key_		# off-load round[1]
+	?vperm		v25,v31,v30,$keyperm
+	lvx		v31,$x00,$key
+	stvx		v25,$x10,$key_		# off-load round[2]
+	addi		$key_,$key_,0x20
+	bdnz		Load_cbc_dec_key
+
+	lvx		v26,$x10,$key
+	?vperm		v24,v30,v31,$keyperm
+	lvx		v27,$x20,$key
+	stvx		v24,$x00,$key_		# off-load round[3]
+	?vperm		v25,v31,v26,$keyperm
+	lvx		v28,$x30,$key
+	stvx		v25,$x10,$key_		# off-load round[4]
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	?vperm		v26,v26,v27,$keyperm
+	lvx		v29,$x40,$key
+	?vperm		v27,v27,v28,$keyperm
+	lvx		v30,$x50,$key
+	?vperm		v28,v28,v29,$keyperm
+	lvx		v31,$x60,$key
+	?vperm		v29,v29,v30,$keyperm
+	lvx		$out0,$x70,$key		# borrow $out0
+	?vperm		v30,v30,v31,$keyperm
+	lvx		v24,$x00,$key_		# pre-load round[1]
+	?vperm		v31,v31,$out0,$keyperm
+	lvx		v25,$x10,$key_		# pre-load round[2]
+
+	#lvx		$inptail,0,$inp		# "caller" already did this
+	#addi		$inp,$inp,15		# 15 is not typo
+	subi		$inp,$inp,15		# undo "caller"
+
+	 le?li		$idx,8
+	lvx_u		$in0,$x00,$inp		# load first 8 "words"
+	 le?lvsl	$inpperm,0,$idx
+	 le?vspltisb	$tmp,0x0f
+	lvx_u		$in1,$x10,$inp
+	 le?vxor	$inpperm,$inpperm,$tmp	# transform for lvx_u/stvx_u
+	lvx_u		$in2,$x20,$inp
+	 le?vperm	$in0,$in0,$in0,$inpperm
+	lvx_u		$in3,$x30,$inp
+	 le?vperm	$in1,$in1,$in1,$inpperm
+	lvx_u		$in4,$x40,$inp
+	 le?vperm	$in2,$in2,$in2,$inpperm
+	vxor		$out0,$in0,$rndkey0
+	lvx_u		$in5,$x50,$inp
+	 le?vperm	$in3,$in3,$in3,$inpperm
+	vxor		$out1,$in1,$rndkey0
+	lvx_u		$in6,$x60,$inp
+	 le?vperm	$in4,$in4,$in4,$inpperm
+	vxor		$out2,$in2,$rndkey0
+	lvx_u		$in7,$x70,$inp
+	addi		$inp,$inp,0x80
+	 le?vperm	$in5,$in5,$in5,$inpperm
+	vxor		$out3,$in3,$rndkey0
+	 le?vperm	$in6,$in6,$in6,$inpperm
+	vxor		$out4,$in4,$rndkey0
+	 le?vperm	$in7,$in7,$in7,$inpperm
+	vxor		$out5,$in5,$rndkey0
+	vxor		$out6,$in6,$rndkey0
+	vxor		$out7,$in7,$rndkey0
+
+	mtctr		$rounds
+	b		Loop_cbc_dec8x
+.align	5
+Loop_cbc_dec8x:
+	vncipher	$out0,$out0,v24
+	vncipher	$out1,$out1,v24
+	vncipher	$out2,$out2,v24
+	vncipher	$out3,$out3,v24
+	vncipher	$out4,$out4,v24
+	vncipher	$out5,$out5,v24
+	vncipher	$out6,$out6,v24
+	vncipher	$out7,$out7,v24
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vncipher	$out0,$out0,v25
+	vncipher	$out1,$out1,v25
+	vncipher	$out2,$out2,v25
+	vncipher	$out3,$out3,v25
+	vncipher	$out4,$out4,v25
+	vncipher	$out5,$out5,v25
+	vncipher	$out6,$out6,v25
+	vncipher	$out7,$out7,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		Loop_cbc_dec8x
+
+	subic		$len,$len,128		# $len-=128
+	vncipher	$out0,$out0,v24
+	vncipher	$out1,$out1,v24
+	vncipher	$out2,$out2,v24
+	vncipher	$out3,$out3,v24
+	vncipher	$out4,$out4,v24
+	vncipher	$out5,$out5,v24
+	vncipher	$out6,$out6,v24
+	vncipher	$out7,$out7,v24
+
+	subfe.		r0,r0,r0		# borrow?-1:0
+	vncipher	$out0,$out0,v25
+	vncipher	$out1,$out1,v25
+	vncipher	$out2,$out2,v25
+	vncipher	$out3,$out3,v25
+	vncipher	$out4,$out4,v25
+	vncipher	$out5,$out5,v25
+	vncipher	$out6,$out6,v25
+	vncipher	$out7,$out7,v25
+
+	and		r0,r0,$len
+	vncipher	$out0,$out0,v26
+	vncipher	$out1,$out1,v26
+	vncipher	$out2,$out2,v26
+	vncipher	$out3,$out3,v26
+	vncipher	$out4,$out4,v26
+	vncipher	$out5,$out5,v26
+	vncipher	$out6,$out6,v26
+	vncipher	$out7,$out7,v26
+
+	add		$inp,$inp,r0		# $inp is adjusted in such
+						# way that at exit from the
+						# loop inX-in7 are loaded
+						# with last "words"
+	vncipher	$out0,$out0,v27
+	vncipher	$out1,$out1,v27
+	vncipher	$out2,$out2,v27
+	vncipher	$out3,$out3,v27
+	vncipher	$out4,$out4,v27
+	vncipher	$out5,$out5,v27
+	vncipher	$out6,$out6,v27
+	vncipher	$out7,$out7,v27
+
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	vncipher	$out0,$out0,v28
+	vncipher	$out1,$out1,v28
+	vncipher	$out2,$out2,v28
+	vncipher	$out3,$out3,v28
+	vncipher	$out4,$out4,v28
+	vncipher	$out5,$out5,v28
+	vncipher	$out6,$out6,v28
+	vncipher	$out7,$out7,v28
+	lvx		v24,$x00,$key_		# re-pre-load round[1]
+
+	vncipher	$out0,$out0,v29
+	vncipher	$out1,$out1,v29
+	vncipher	$out2,$out2,v29
+	vncipher	$out3,$out3,v29
+	vncipher	$out4,$out4,v29
+	vncipher	$out5,$out5,v29
+	vncipher	$out6,$out6,v29
+	vncipher	$out7,$out7,v29
+	lvx		v25,$x10,$key_		# re-pre-load round[2]
+
+	vncipher	$out0,$out0,v30
+	 vxor		$ivec,$ivec,v31		# xor with last round key
+	vncipher	$out1,$out1,v30
+	 vxor		$in0,$in0,v31
+	vncipher	$out2,$out2,v30
+	 vxor		$in1,$in1,v31
+	vncipher	$out3,$out3,v30
+	 vxor		$in2,$in2,v31
+	vncipher	$out4,$out4,v30
+	 vxor		$in3,$in3,v31
+	vncipher	$out5,$out5,v30
+	 vxor		$in4,$in4,v31
+	vncipher	$out6,$out6,v30
+	 vxor		$in5,$in5,v31
+	vncipher	$out7,$out7,v30
+	 vxor		$in6,$in6,v31
+
+	vncipherlast	$out0,$out0,$ivec
+	vncipherlast	$out1,$out1,$in0
+	 lvx_u		$in0,$x00,$inp		# load next input block
+	vncipherlast	$out2,$out2,$in1
+	 lvx_u		$in1,$x10,$inp
+	vncipherlast	$out3,$out3,$in2
+	 le?vperm	$in0,$in0,$in0,$inpperm
+	 lvx_u		$in2,$x20,$inp
+	vncipherlast	$out4,$out4,$in3
+	 le?vperm	$in1,$in1,$in1,$inpperm
+	 lvx_u		$in3,$x30,$inp
+	vncipherlast	$out5,$out5,$in4
+	 le?vperm	$in2,$in2,$in2,$inpperm
+	 lvx_u		$in4,$x40,$inp
+	vncipherlast	$out6,$out6,$in5
+	 le?vperm	$in3,$in3,$in3,$inpperm
+	 lvx_u		$in5,$x50,$inp
+	vncipherlast	$out7,$out7,$in6
+	 le?vperm	$in4,$in4,$in4,$inpperm
+	 lvx_u		$in6,$x60,$inp
+	vmr		$ivec,$in7
+	 le?vperm	$in5,$in5,$in5,$inpperm
+	 lvx_u		$in7,$x70,$inp
+	 addi		$inp,$inp,0x80
+
+	le?vperm	$out0,$out0,$out0,$inpperm
+	le?vperm	$out1,$out1,$out1,$inpperm
+	stvx_u		$out0,$x00,$out
+	 le?vperm	$in6,$in6,$in6,$inpperm
+	 vxor		$out0,$in0,$rndkey0
+	le?vperm	$out2,$out2,$out2,$inpperm
+	stvx_u		$out1,$x10,$out
+	 le?vperm	$in7,$in7,$in7,$inpperm
+	 vxor		$out1,$in1,$rndkey0
+	le?vperm	$out3,$out3,$out3,$inpperm
+	stvx_u		$out2,$x20,$out
+	 vxor		$out2,$in2,$rndkey0
+	le?vperm	$out4,$out4,$out4,$inpperm
+	stvx_u		$out3,$x30,$out
+	 vxor		$out3,$in3,$rndkey0
+	le?vperm	$out5,$out5,$out5,$inpperm
+	stvx_u		$out4,$x40,$out
+	 vxor		$out4,$in4,$rndkey0
+	le?vperm	$out6,$out6,$out6,$inpperm
+	stvx_u		$out5,$x50,$out
+	 vxor		$out5,$in5,$rndkey0
+	le?vperm	$out7,$out7,$out7,$inpperm
+	stvx_u		$out6,$x60,$out
+	 vxor		$out6,$in6,$rndkey0
+	stvx_u		$out7,$x70,$out
+	addi		$out,$out,0x80
+	 vxor		$out7,$in7,$rndkey0
+
+	mtctr		$rounds
+	beq		Loop_cbc_dec8x		# did $len-=128 borrow?
+
+	addic.		$len,$len,128
+	beq		Lcbc_dec8x_done
+	nop
+	nop
+
+Loop_cbc_dec8x_tail:				# up to 7 "words" tail...
+	vncipher	$out1,$out1,v24
+	vncipher	$out2,$out2,v24
+	vncipher	$out3,$out3,v24
+	vncipher	$out4,$out4,v24
+	vncipher	$out5,$out5,v24
+	vncipher	$out6,$out6,v24
+	vncipher	$out7,$out7,v24
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vncipher	$out1,$out1,v25
+	vncipher	$out2,$out2,v25
+	vncipher	$out3,$out3,v25
+	vncipher	$out4,$out4,v25
+	vncipher	$out5,$out5,v25
+	vncipher	$out6,$out6,v25
+	vncipher	$out7,$out7,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		Loop_cbc_dec8x_tail
+
+	vncipher	$out1,$out1,v24
+	vncipher	$out2,$out2,v24
+	vncipher	$out3,$out3,v24
+	vncipher	$out4,$out4,v24
+	vncipher	$out5,$out5,v24
+	vncipher	$out6,$out6,v24
+	vncipher	$out7,$out7,v24
+
+	vncipher	$out1,$out1,v25
+	vncipher	$out2,$out2,v25
+	vncipher	$out3,$out3,v25
+	vncipher	$out4,$out4,v25
+	vncipher	$out5,$out5,v25
+	vncipher	$out6,$out6,v25
+	vncipher	$out7,$out7,v25
+
+	vncipher	$out1,$out1,v26
+	vncipher	$out2,$out2,v26
+	vncipher	$out3,$out3,v26
+	vncipher	$out4,$out4,v26
+	vncipher	$out5,$out5,v26
+	vncipher	$out6,$out6,v26
+	vncipher	$out7,$out7,v26
+
+	vncipher	$out1,$out1,v27
+	vncipher	$out2,$out2,v27
+	vncipher	$out3,$out3,v27
+	vncipher	$out4,$out4,v27
+	vncipher	$out5,$out5,v27
+	vncipher	$out6,$out6,v27
+	vncipher	$out7,$out7,v27
+
+	vncipher	$out1,$out1,v28
+	vncipher	$out2,$out2,v28
+	vncipher	$out3,$out3,v28
+	vncipher	$out4,$out4,v28
+	vncipher	$out5,$out5,v28
+	vncipher	$out6,$out6,v28
+	vncipher	$out7,$out7,v28
+
+	vncipher	$out1,$out1,v29
+	vncipher	$out2,$out2,v29
+	vncipher	$out3,$out3,v29
+	vncipher	$out4,$out4,v29
+	vncipher	$out5,$out5,v29
+	vncipher	$out6,$out6,v29
+	vncipher	$out7,$out7,v29
+
+	vncipher	$out1,$out1,v30
+	 vxor		$ivec,$ivec,v31		# last round key
+	vncipher	$out2,$out2,v30
+	 vxor		$in1,$in1,v31
+	vncipher	$out3,$out3,v30
+	 vxor		$in2,$in2,v31
+	vncipher	$out4,$out4,v30
+	 vxor		$in3,$in3,v31
+	vncipher	$out5,$out5,v30
+	 vxor		$in4,$in4,v31
+	vncipher	$out6,$out6,v30
+	 vxor		$in5,$in5,v31
+	vncipher	$out7,$out7,v30
+	 vxor		$in6,$in6,v31
+
+	cmplwi		$len,32			# switch($len)
+	blt		Lcbc_dec8x_one
+	nop
+	beq		Lcbc_dec8x_two
+	cmplwi		$len,64
+	blt		Lcbc_dec8x_three
+	nop
+	beq		Lcbc_dec8x_four
+	cmplwi		$len,96
+	blt		Lcbc_dec8x_five
+	nop
+	beq		Lcbc_dec8x_six
+
+Lcbc_dec8x_seven:
+	vncipherlast	$out1,$out1,$ivec
+	vncipherlast	$out2,$out2,$in1
+	vncipherlast	$out3,$out3,$in2
+	vncipherlast	$out4,$out4,$in3
+	vncipherlast	$out5,$out5,$in4
+	vncipherlast	$out6,$out6,$in5
+	vncipherlast	$out7,$out7,$in6
+	vmr		$ivec,$in7
+
+	le?vperm	$out1,$out1,$out1,$inpperm
+	le?vperm	$out2,$out2,$out2,$inpperm
+	stvx_u		$out1,$x00,$out
+	le?vperm	$out3,$out3,$out3,$inpperm
+	stvx_u		$out2,$x10,$out
+	le?vperm	$out4,$out4,$out4,$inpperm
+	stvx_u		$out3,$x20,$out
+	le?vperm	$out5,$out5,$out5,$inpperm
+	stvx_u		$out4,$x30,$out
+	le?vperm	$out6,$out6,$out6,$inpperm
+	stvx_u		$out5,$x40,$out
+	le?vperm	$out7,$out7,$out7,$inpperm
+	stvx_u		$out6,$x50,$out
+	stvx_u		$out7,$x60,$out
+	addi		$out,$out,0x70
+	b		Lcbc_dec8x_done
+
+.align	5
+Lcbc_dec8x_six:
+	vncipherlast	$out2,$out2,$ivec
+	vncipherlast	$out3,$out3,$in2
+	vncipherlast	$out4,$out4,$in3
+	vncipherlast	$out5,$out5,$in4
+	vncipherlast	$out6,$out6,$in5
+	vncipherlast	$out7,$out7,$in6
+	vmr		$ivec,$in7
+
+	le?vperm	$out2,$out2,$out2,$inpperm
+	le?vperm	$out3,$out3,$out3,$inpperm
+	stvx_u		$out2,$x00,$out
+	le?vperm	$out4,$out4,$out4,$inpperm
+	stvx_u		$out3,$x10,$out
+	le?vperm	$out5,$out5,$out5,$inpperm
+	stvx_u		$out4,$x20,$out
+	le?vperm	$out6,$out6,$out6,$inpperm
+	stvx_u		$out5,$x30,$out
+	le?vperm	$out7,$out7,$out7,$inpperm
+	stvx_u		$out6,$x40,$out
+	stvx_u		$out7,$x50,$out
+	addi		$out,$out,0x60
+	b		Lcbc_dec8x_done
+
+.align	5
+Lcbc_dec8x_five:
+	vncipherlast	$out3,$out3,$ivec
+	vncipherlast	$out4,$out4,$in3
+	vncipherlast	$out5,$out5,$in4
+	vncipherlast	$out6,$out6,$in5
+	vncipherlast	$out7,$out7,$in6
+	vmr		$ivec,$in7
+
+	le?vperm	$out3,$out3,$out3,$inpperm
+	le?vperm	$out4,$out4,$out4,$inpperm
+	stvx_u		$out3,$x00,$out
+	le?vperm	$out5,$out5,$out5,$inpperm
+	stvx_u		$out4,$x10,$out
+	le?vperm	$out6,$out6,$out6,$inpperm
+	stvx_u		$out5,$x20,$out
+	le?vperm	$out7,$out7,$out7,$inpperm
+	stvx_u		$out6,$x30,$out
+	stvx_u		$out7,$x40,$out
+	addi		$out,$out,0x50
+	b		Lcbc_dec8x_done
+
+.align	5
+Lcbc_dec8x_four:
+	vncipherlast	$out4,$out4,$ivec
+	vncipherlast	$out5,$out5,$in4
+	vncipherlast	$out6,$out6,$in5
+	vncipherlast	$out7,$out7,$in6
+	vmr		$ivec,$in7
+
+	le?vperm	$out4,$out4,$out4,$inpperm
+	le?vperm	$out5,$out5,$out5,$inpperm
+	stvx_u		$out4,$x00,$out
+	le?vperm	$out6,$out6,$out6,$inpperm
+	stvx_u		$out5,$x10,$out
+	le?vperm	$out7,$out7,$out7,$inpperm
+	stvx_u		$out6,$x20,$out
+	stvx_u		$out7,$x30,$out
+	addi		$out,$out,0x40
+	b		Lcbc_dec8x_done
+
+.align	5
+Lcbc_dec8x_three:
+	vncipherlast	$out5,$out5,$ivec
+	vncipherlast	$out6,$out6,$in5
+	vncipherlast	$out7,$out7,$in6
+	vmr		$ivec,$in7
+
+	le?vperm	$out5,$out5,$out5,$inpperm
+	le?vperm	$out6,$out6,$out6,$inpperm
+	stvx_u		$out5,$x00,$out
+	le?vperm	$out7,$out7,$out7,$inpperm
+	stvx_u		$out6,$x10,$out
+	stvx_u		$out7,$x20,$out
+	addi		$out,$out,0x30
+	b		Lcbc_dec8x_done
+
+.align	5
+Lcbc_dec8x_two:
+	vncipherlast	$out6,$out6,$ivec
+	vncipherlast	$out7,$out7,$in6
+	vmr		$ivec,$in7
+
+	le?vperm	$out6,$out6,$out6,$inpperm
+	le?vperm	$out7,$out7,$out7,$inpperm
+	stvx_u		$out6,$x00,$out
+	stvx_u		$out7,$x10,$out
+	addi		$out,$out,0x20
+	b		Lcbc_dec8x_done
+
+.align	5
+Lcbc_dec8x_one:
+	vncipherlast	$out7,$out7,$ivec
+	vmr		$ivec,$in7
+
+	le?vperm	$out7,$out7,$out7,$inpperm
+	stvx_u		$out7,0,$out
+	addi		$out,$out,0x10
+
+Lcbc_dec8x_done:
+	le?vperm	$ivec,$ivec,$ivec,$inpperm
+	stvx_u		$ivec,0,$ivp		# write [unaligned] iv
+
+	li		r10,`$FRAME+15`
+	li		r11,`$FRAME+31`
+	stvx		$inpperm,r10,$sp	# wipe copies of round keys
+	addi		r10,r10,32
+	stvx		$inpperm,r11,$sp
+	addi		r11,r11,32
+	stvx		$inpperm,r10,$sp
+	addi		r10,r10,32
+	stvx		$inpperm,r11,$sp
+	addi		r11,r11,32
+	stvx		$inpperm,r10,$sp
+	addi		r10,r10,32
+	stvx		$inpperm,r11,$sp
+	addi		r11,r11,32
+	stvx		$inpperm,r10,$sp
+	addi		r10,r10,32
+	stvx		$inpperm,r11,$sp
+	addi		r11,r11,32
+
+	mtspr		256,$vrsave
+	lvx		v20,r10,$sp		# ABI says so
+	addi		r10,r10,32
+	lvx		v21,r11,$sp
+	addi		r11,r11,32
+	lvx		v22,r10,$sp
+	addi		r10,r10,32
+	lvx		v23,r11,$sp
+	addi		r11,r11,32
+	lvx		v24,r10,$sp
+	addi		r10,r10,32
+	lvx		v25,r11,$sp
+	addi		r11,r11,32
+	lvx		v26,r10,$sp
+	addi		r10,r10,32
+	lvx		v27,r11,$sp
+	addi		r11,r11,32
+	lvx		v28,r10,$sp
+	addi		r10,r10,32
+	lvx		v29,r11,$sp
+	addi		r11,r11,32
+	lvx		v30,r10,$sp
+	lvx		v31,r11,$sp
+	$POP		r26,`$FRAME+21*16+0*$SIZE_T`($sp)
+	$POP		r27,`$FRAME+21*16+1*$SIZE_T`($sp)
+	$POP		r28,`$FRAME+21*16+2*$SIZE_T`($sp)
+	$POP		r29,`$FRAME+21*16+3*$SIZE_T`($sp)
+	$POP		r30,`$FRAME+21*16+4*$SIZE_T`($sp)
+	$POP		r31,`$FRAME+21*16+5*$SIZE_T`($sp)
+	addi		$sp,$sp,`$FRAME+21*16+6*$SIZE_T`
+	blr
+	.long		0
+	.byte		0,12,0x14,0,0x80,6,6,0
+	.long		0
+.size	.${prefix}_cbc_encrypt,.-.${prefix}_cbc_encrypt
+___
+}}	}}}
+
+#########################################################################
+{{{	# CTR procedure[s]						#
+
+####################### WARNING: Here be dragons! #######################
+#
+# This code is written as 'ctr32', based on a 32-bit counter used
+# upstream. The kernel does *not* use a 32-bit counter. The kernel uses
+# a 128-bit counter.
+#
+# This leads to subtle changes from the upstream code: the counter
+# is incremented with vaddu_q_m rather than vaddu_w_m. This occurs in
+# both the bulk (8 blocks at a time) path, and in the individual block
+# path. Be aware of this when doing updates.
+#
+# See:
+# 1d4aa0b4c181 ("crypto: vmx - Fixing AES-CTR counter bug")
+# 009b30ac7444 ("crypto: vmx - CTR: always increment IV as quadword")
+# https://github.com/openssl/openssl/pull/8942
+#
+#########################################################################
+my ($inp,$out,$len,$key,$ivp,$x10,$rounds,$idx)=map("r$_",(3..10));
+my ($rndkey0,$rndkey1,$inout,$tmp)=		map("v$_",(0..3));
+my ($ivec,$inptail,$inpperm,$outhead,$outperm,$outmask,$keyperm,$one)=
+						map("v$_",(4..11));
+my $dat=$tmp;
+
+$code.=<<___;
+.globl	.${prefix}_ctr32_encrypt_blocks
+	${UCMP}i	$len,1
+	bltlr-
+
+	lis		r0,0xfff0
+	mfspr		$vrsave,256
+	mtspr		256,r0
+
+	li		$idx,15
+	vxor		$rndkey0,$rndkey0,$rndkey0
+	le?vspltisb	$tmp,0x0f
+
+	lvx		$ivec,0,$ivp		# load [unaligned] iv
+	lvsl		$inpperm,0,$ivp
+	lvx		$inptail,$idx,$ivp
+	 vspltisb	$one,1
+	le?vxor		$inpperm,$inpperm,$tmp
+	vperm		$ivec,$ivec,$inptail,$inpperm
+	 vsldoi		$one,$rndkey0,$one,1
+
+	neg		r11,$inp
+	?lvsl		$keyperm,0,$key		# prepare for unaligned key
+	lwz		$rounds,240($key)
+
+	lvsr		$inpperm,0,r11		# prepare for unaligned load
+	lvx		$inptail,0,$inp
+	addi		$inp,$inp,15		# 15 is not typo
+	le?vxor		$inpperm,$inpperm,$tmp
+
+	srwi		$rounds,$rounds,1
+	li		$idx,16
+	subi		$rounds,$rounds,1
+
+	${UCMP}i	$len,8
+	bge		_aesp8_ctr32_encrypt8x
+
+	?lvsr		$outperm,0,$out		# prepare for unaligned store
+	vspltisb	$outmask,-1
+	lvx		$outhead,0,$out
+	?vperm		$outmask,$rndkey0,$outmask,$outperm
+	le?vxor		$outperm,$outperm,$tmp
+
+	lvx		$rndkey0,0,$key
+	mtctr		$rounds
+	lvx		$rndkey1,$idx,$key
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$inout,$ivec,$rndkey0
+	lvx		$rndkey0,$idx,$key
+	addi		$idx,$idx,16
+	b		Loop_ctr32_enc
+
+.align	5
+Loop_ctr32_enc:
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vcipher		$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key
+	addi		$idx,$idx,16
+	bdnz		Loop_ctr32_enc
+
+	vadduqm		$ivec,$ivec,$one	# Kernel change for 128-bit
+	 vmr		$dat,$inptail
+	 lvx		$inptail,0,$inp
+	 addi		$inp,$inp,16
+	 subic.		$len,$len,1		# blocks--
+
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key
+	 vperm		$dat,$dat,$inptail,$inpperm
+	 li		$idx,16
+	?vperm		$rndkey1,$rndkey0,$rndkey1,$keyperm
+	 lvx		$rndkey0,0,$key
+	vxor		$dat,$dat,$rndkey1	# last round key
+	vcipherlast	$inout,$inout,$dat
+
+	 lvx		$rndkey1,$idx,$key
+	 addi		$idx,$idx,16
+	vperm		$inout,$inout,$inout,$outperm
+	vsel		$dat,$outhead,$inout,$outmask
+	 mtctr		$rounds
+	 ?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vmr		$outhead,$inout
+	 vxor		$inout,$ivec,$rndkey0
+	 lvx		$rndkey0,$idx,$key
+	 addi		$idx,$idx,16
+	stvx		$dat,0,$out
+	addi		$out,$out,16
+	bne		Loop_ctr32_enc
+
+	addi		$out,$out,-1
+	lvx		$inout,0,$out		# redundant in aligned case
+	vsel		$inout,$outhead,$inout,$outmask
+	stvx		$inout,0,$out
+
+	mtspr		256,$vrsave
+	blr
+	.long		0
+	.byte		0,12,0x14,0,0,0,6,0
+	.long		0
+___
+#########################################################################
+{{	# Optimized CTR procedure					#
+my $key_="r11";
+my ($x00,$x10,$x20,$x30,$x40,$x50,$x60,$x70)=map("r$_",(0,8,26..31));
+my ($in0, $in1, $in2, $in3, $in4, $in5, $in6, $in7 )=map("v$_",(0..3,10,12..14));
+my ($out0,$out1,$out2,$out3,$out4,$out5,$out6,$out7)=map("v$_",(15..22));
+my $rndkey0="v23";	# v24-v25 rotating buffer for first found keys
+			# v26-v31 last 6 round keys
+my ($tmp,$keyperm)=($in3,$in4);	# aliases with "caller", redundant assignment
+my ($two,$three,$four)=($outhead,$outperm,$outmask);
+
+$code.=<<___;
+.align	5
+_aesp8_ctr32_encrypt8x:
+	$STU		$sp,-`($FRAME+21*16+6*$SIZE_T)`($sp)
+	li		r10,`$FRAME+8*16+15`
+	li		r11,`$FRAME+8*16+31`
+	stvx		v20,r10,$sp		# ABI says so
+	addi		r10,r10,32
+	stvx		v21,r11,$sp
+	addi		r11,r11,32
+	stvx		v22,r10,$sp
+	addi		r10,r10,32
+	stvx		v23,r11,$sp
+	addi		r11,r11,32
+	stvx		v24,r10,$sp
+	addi		r10,r10,32
+	stvx		v25,r11,$sp
+	addi		r11,r11,32
+	stvx		v26,r10,$sp
+	addi		r10,r10,32
+	stvx		v27,r11,$sp
+	addi		r11,r11,32
+	stvx		v28,r10,$sp
+	addi		r10,r10,32
+	stvx		v29,r11,$sp
+	addi		r11,r11,32
+	stvx		v30,r10,$sp
+	stvx		v31,r11,$sp
+	li		r0,-1
+	stw		$vrsave,`$FRAME+21*16-4`($sp)	# save vrsave
+	li		$x10,0x10
+	$PUSH		r26,`$FRAME+21*16+0*$SIZE_T`($sp)
+	li		$x20,0x20
+	$PUSH		r27,`$FRAME+21*16+1*$SIZE_T`($sp)
+	li		$x30,0x30
+	$PUSH		r28,`$FRAME+21*16+2*$SIZE_T`($sp)
+	li		$x40,0x40
+	$PUSH		r29,`$FRAME+21*16+3*$SIZE_T`($sp)
+	li		$x50,0x50
+	$PUSH		r30,`$FRAME+21*16+4*$SIZE_T`($sp)
+	li		$x60,0x60
+	$PUSH		r31,`$FRAME+21*16+5*$SIZE_T`($sp)
+	li		$x70,0x70
+	mtspr		256,r0
+
+	subi		$rounds,$rounds,3	# -4 in total
+
+	lvx		$rndkey0,$x00,$key	# load key schedule
+	lvx		v30,$x10,$key
+	addi		$key,$key,0x20
+	lvx		v31,$x00,$key
+	?vperm		$rndkey0,$rndkey0,v30,$keyperm
+	addi		$key_,$sp,$FRAME+15
+	mtctr		$rounds
+
+Load_ctr32_enc_key:
+	?vperm		v24,v30,v31,$keyperm
+	lvx		v30,$x10,$key
+	addi		$key,$key,0x20
+	stvx		v24,$x00,$key_		# off-load round[1]
+	?vperm		v25,v31,v30,$keyperm
+	lvx		v31,$x00,$key
+	stvx		v25,$x10,$key_		# off-load round[2]
+	addi		$key_,$key_,0x20
+	bdnz		Load_ctr32_enc_key
+
+	lvx		v26,$x10,$key
+	?vperm		v24,v30,v31,$keyperm
+	lvx		v27,$x20,$key
+	stvx		v24,$x00,$key_		# off-load round[3]
+	?vperm		v25,v31,v26,$keyperm
+	lvx		v28,$x30,$key
+	stvx		v25,$x10,$key_		# off-load round[4]
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	?vperm		v26,v26,v27,$keyperm
+	lvx		v29,$x40,$key
+	?vperm		v27,v27,v28,$keyperm
+	lvx		v30,$x50,$key
+	?vperm		v28,v28,v29,$keyperm
+	lvx		v31,$x60,$key
+	?vperm		v29,v29,v30,$keyperm
+	lvx		$out0,$x70,$key		# borrow $out0
+	?vperm		v30,v30,v31,$keyperm
+	lvx		v24,$x00,$key_		# pre-load round[1]
+	?vperm		v31,v31,$out0,$keyperm
+	lvx		v25,$x10,$key_		# pre-load round[2]
+
+	vadduqm		$two,$one,$one
+	subi		$inp,$inp,15		# undo "caller"
+	$SHL		$len,$len,4
+
+	vadduqm		$out1,$ivec,$one	# counter values ...
+	vadduqm		$out2,$ivec,$two	# (do all ctr adds as 128-bit)
+	vxor		$out0,$ivec,$rndkey0	# ... xored with rndkey[0]
+	 le?li		$idx,8
+	vadduqm		$out3,$out1,$two
+	vxor		$out1,$out1,$rndkey0
+	 le?lvsl	$inpperm,0,$idx
+	vadduqm		$out4,$out2,$two
+	vxor		$out2,$out2,$rndkey0
+	 le?vspltisb	$tmp,0x0f
+	vadduqm		$out5,$out3,$two
+	vxor		$out3,$out3,$rndkey0
+	 le?vxor	$inpperm,$inpperm,$tmp	# transform for lvx_u/stvx_u
+	vadduqm		$out6,$out4,$two
+	vxor		$out4,$out4,$rndkey0
+	vadduqm		$out7,$out5,$two
+	vxor		$out5,$out5,$rndkey0
+	vadduqm		$ivec,$out6,$two	# next counter value
+	vxor		$out6,$out6,$rndkey0
+	vxor		$out7,$out7,$rndkey0
+
+	mtctr		$rounds
+	b		Loop_ctr32_enc8x
+.align	5
+Loop_ctr32_enc8x:
+	vcipher 	$out0,$out0,v24
+	vcipher 	$out1,$out1,v24
+	vcipher 	$out2,$out2,v24
+	vcipher 	$out3,$out3,v24
+	vcipher 	$out4,$out4,v24
+	vcipher 	$out5,$out5,v24
+	vcipher 	$out6,$out6,v24
+	vcipher 	$out7,$out7,v24
+Loop_ctr32_enc8x_middle:
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vcipher 	$out0,$out0,v25
+	vcipher 	$out1,$out1,v25
+	vcipher 	$out2,$out2,v25
+	vcipher 	$out3,$out3,v25
+	vcipher 	$out4,$out4,v25
+	vcipher 	$out5,$out5,v25
+	vcipher 	$out6,$out6,v25
+	vcipher 	$out7,$out7,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		Loop_ctr32_enc8x
+
+	subic		r11,$len,256		# $len-256, borrow $key_
+	vcipher 	$out0,$out0,v24
+	vcipher 	$out1,$out1,v24
+	vcipher 	$out2,$out2,v24
+	vcipher 	$out3,$out3,v24
+	vcipher 	$out4,$out4,v24
+	vcipher 	$out5,$out5,v24
+	vcipher 	$out6,$out6,v24
+	vcipher 	$out7,$out7,v24
+
+	subfe		r0,r0,r0		# borrow?-1:0
+	vcipher 	$out0,$out0,v25
+	vcipher 	$out1,$out1,v25
+	vcipher 	$out2,$out2,v25
+	vcipher 	$out3,$out3,v25
+	vcipher 	$out4,$out4,v25
+	vcipher		$out5,$out5,v25
+	vcipher		$out6,$out6,v25
+	vcipher		$out7,$out7,v25
+
+	and		r0,r0,r11
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	vcipher		$out0,$out0,v26
+	vcipher		$out1,$out1,v26
+	vcipher		$out2,$out2,v26
+	vcipher		$out3,$out3,v26
+	vcipher		$out4,$out4,v26
+	vcipher		$out5,$out5,v26
+	vcipher		$out6,$out6,v26
+	vcipher		$out7,$out7,v26
+	lvx		v24,$x00,$key_		# re-pre-load round[1]
+
+	subic		$len,$len,129		# $len-=129
+	vcipher		$out0,$out0,v27
+	addi		$len,$len,1		# $len-=128 really
+	vcipher		$out1,$out1,v27
+	vcipher		$out2,$out2,v27
+	vcipher		$out3,$out3,v27
+	vcipher		$out4,$out4,v27
+	vcipher		$out5,$out5,v27
+	vcipher		$out6,$out6,v27
+	vcipher		$out7,$out7,v27
+	lvx		v25,$x10,$key_		# re-pre-load round[2]
+
+	vcipher		$out0,$out0,v28
+	 lvx_u		$in0,$x00,$inp		# load input
+	vcipher		$out1,$out1,v28
+	 lvx_u		$in1,$x10,$inp
+	vcipher		$out2,$out2,v28
+	 lvx_u		$in2,$x20,$inp
+	vcipher		$out3,$out3,v28
+	 lvx_u		$in3,$x30,$inp
+	vcipher		$out4,$out4,v28
+	 lvx_u		$in4,$x40,$inp
+	vcipher		$out5,$out5,v28
+	 lvx_u		$in5,$x50,$inp
+	vcipher		$out6,$out6,v28
+	 lvx_u		$in6,$x60,$inp
+	vcipher		$out7,$out7,v28
+	 lvx_u		$in7,$x70,$inp
+	 addi		$inp,$inp,0x80
+
+	vcipher		$out0,$out0,v29
+	 le?vperm	$in0,$in0,$in0,$inpperm
+	vcipher		$out1,$out1,v29
+	 le?vperm	$in1,$in1,$in1,$inpperm
+	vcipher		$out2,$out2,v29
+	 le?vperm	$in2,$in2,$in2,$inpperm
+	vcipher		$out3,$out3,v29
+	 le?vperm	$in3,$in3,$in3,$inpperm
+	vcipher		$out4,$out4,v29
+	 le?vperm	$in4,$in4,$in4,$inpperm
+	vcipher		$out5,$out5,v29
+	 le?vperm	$in5,$in5,$in5,$inpperm
+	vcipher		$out6,$out6,v29
+	 le?vperm	$in6,$in6,$in6,$inpperm
+	vcipher		$out7,$out7,v29
+	 le?vperm	$in7,$in7,$in7,$inpperm
+
+	add		$inp,$inp,r0		# $inp is adjusted in such
+						# way that at exit from the
+						# loop inX-in7 are loaded
+						# with last "words"
+	subfe.		r0,r0,r0		# borrow?-1:0
+	vcipher		$out0,$out0,v30
+	 vxor		$in0,$in0,v31		# xor with last round key
+	vcipher		$out1,$out1,v30
+	 vxor		$in1,$in1,v31
+	vcipher		$out2,$out2,v30
+	 vxor		$in2,$in2,v31
+	vcipher		$out3,$out3,v30
+	 vxor		$in3,$in3,v31
+	vcipher		$out4,$out4,v30
+	 vxor		$in4,$in4,v31
+	vcipher		$out5,$out5,v30
+	 vxor		$in5,$in5,v31
+	vcipher		$out6,$out6,v30
+	 vxor		$in6,$in6,v31
+	vcipher		$out7,$out7,v30
+	 vxor		$in7,$in7,v31
+
+	bne		Lctr32_enc8x_break	# did $len-129 borrow?
+
+	vcipherlast	$in0,$out0,$in0
+	vcipherlast	$in1,$out1,$in1
+	 vadduqm	$out1,$ivec,$one	# counter values ...
+	vcipherlast	$in2,$out2,$in2
+	 vadduqm	$out2,$ivec,$two
+	 vxor		$out0,$ivec,$rndkey0	# ... xored with rndkey[0]
+	vcipherlast	$in3,$out3,$in3
+	 vadduqm	$out3,$out1,$two
+	 vxor		$out1,$out1,$rndkey0
+	vcipherlast	$in4,$out4,$in4
+	 vadduqm	$out4,$out2,$two
+	 vxor		$out2,$out2,$rndkey0
+	vcipherlast	$in5,$out5,$in5
+	 vadduqm	$out5,$out3,$two
+	 vxor		$out3,$out3,$rndkey0
+	vcipherlast	$in6,$out6,$in6
+	 vadduqm	$out6,$out4,$two
+	 vxor		$out4,$out4,$rndkey0
+	vcipherlast	$in7,$out7,$in7
+	 vadduqm	$out7,$out5,$two
+	 vxor		$out5,$out5,$rndkey0
+	le?vperm	$in0,$in0,$in0,$inpperm
+	 vadduqm	$ivec,$out6,$two	# next counter value
+	 vxor		$out6,$out6,$rndkey0
+	le?vperm	$in1,$in1,$in1,$inpperm
+	 vxor		$out7,$out7,$rndkey0
+	mtctr		$rounds
+
+	 vcipher	$out0,$out0,v24
+	stvx_u		$in0,$x00,$out
+	le?vperm	$in2,$in2,$in2,$inpperm
+	 vcipher	$out1,$out1,v24
+	stvx_u		$in1,$x10,$out
+	le?vperm	$in3,$in3,$in3,$inpperm
+	 vcipher	$out2,$out2,v24
+	stvx_u		$in2,$x20,$out
+	le?vperm	$in4,$in4,$in4,$inpperm
+	 vcipher	$out3,$out3,v24
+	stvx_u		$in3,$x30,$out
+	le?vperm	$in5,$in5,$in5,$inpperm
+	 vcipher	$out4,$out4,v24
+	stvx_u		$in4,$x40,$out
+	le?vperm	$in6,$in6,$in6,$inpperm
+	 vcipher	$out5,$out5,v24
+	stvx_u		$in5,$x50,$out
+	le?vperm	$in7,$in7,$in7,$inpperm
+	 vcipher	$out6,$out6,v24
+	stvx_u		$in6,$x60,$out
+	 vcipher	$out7,$out7,v24
+	stvx_u		$in7,$x70,$out
+	addi		$out,$out,0x80
+
+	b		Loop_ctr32_enc8x_middle
+
+.align	5
+Lctr32_enc8x_break:
+	cmpwi		$len,-0x60
+	blt		Lctr32_enc8x_one
+	nop
+	beq		Lctr32_enc8x_two
+	cmpwi		$len,-0x40
+	blt		Lctr32_enc8x_three
+	nop
+	beq		Lctr32_enc8x_four
+	cmpwi		$len,-0x20
+	blt		Lctr32_enc8x_five
+	nop
+	beq		Lctr32_enc8x_six
+	cmpwi		$len,0x00
+	blt		Lctr32_enc8x_seven
+
+Lctr32_enc8x_eight:
+	vcipherlast	$out0,$out0,$in0
+	vcipherlast	$out1,$out1,$in1
+	vcipherlast	$out2,$out2,$in2
+	vcipherlast	$out3,$out3,$in3
+	vcipherlast	$out4,$out4,$in4
+	vcipherlast	$out5,$out5,$in5
+	vcipherlast	$out6,$out6,$in6
+	vcipherlast	$out7,$out7,$in7
+
+	le?vperm	$out0,$out0,$out0,$inpperm
+	le?vperm	$out1,$out1,$out1,$inpperm
+	stvx_u		$out0,$x00,$out
+	le?vperm	$out2,$out2,$out2,$inpperm
+	stvx_u		$out1,$x10,$out
+	le?vperm	$out3,$out3,$out3,$inpperm
+	stvx_u		$out2,$x20,$out
+	le?vperm	$out4,$out4,$out4,$inpperm
+	stvx_u		$out3,$x30,$out
+	le?vperm	$out5,$out5,$out5,$inpperm
+	stvx_u		$out4,$x40,$out
+	le?vperm	$out6,$out6,$out6,$inpperm
+	stvx_u		$out5,$x50,$out
+	le?vperm	$out7,$out7,$out7,$inpperm
+	stvx_u		$out6,$x60,$out
+	stvx_u		$out7,$x70,$out
+	addi		$out,$out,0x80
+	b		Lctr32_enc8x_done
+
+.align	5
+Lctr32_enc8x_seven:
+	vcipherlast	$out0,$out0,$in1
+	vcipherlast	$out1,$out1,$in2
+	vcipherlast	$out2,$out2,$in3
+	vcipherlast	$out3,$out3,$in4
+	vcipherlast	$out4,$out4,$in5
+	vcipherlast	$out5,$out5,$in6
+	vcipherlast	$out6,$out6,$in7
+
+	le?vperm	$out0,$out0,$out0,$inpperm
+	le?vperm	$out1,$out1,$out1,$inpperm
+	stvx_u		$out0,$x00,$out
+	le?vperm	$out2,$out2,$out2,$inpperm
+	stvx_u		$out1,$x10,$out
+	le?vperm	$out3,$out3,$out3,$inpperm
+	stvx_u		$out2,$x20,$out
+	le?vperm	$out4,$out4,$out4,$inpperm
+	stvx_u		$out3,$x30,$out
+	le?vperm	$out5,$out5,$out5,$inpperm
+	stvx_u		$out4,$x40,$out
+	le?vperm	$out6,$out6,$out6,$inpperm
+	stvx_u		$out5,$x50,$out
+	stvx_u		$out6,$x60,$out
+	addi		$out,$out,0x70
+	b		Lctr32_enc8x_done
+
+.align	5
+Lctr32_enc8x_six:
+	vcipherlast	$out0,$out0,$in2
+	vcipherlast	$out1,$out1,$in3
+	vcipherlast	$out2,$out2,$in4
+	vcipherlast	$out3,$out3,$in5
+	vcipherlast	$out4,$out4,$in6
+	vcipherlast	$out5,$out5,$in7
+
+	le?vperm	$out0,$out0,$out0,$inpperm
+	le?vperm	$out1,$out1,$out1,$inpperm
+	stvx_u		$out0,$x00,$out
+	le?vperm	$out2,$out2,$out2,$inpperm
+	stvx_u		$out1,$x10,$out
+	le?vperm	$out3,$out3,$out3,$inpperm
+	stvx_u		$out2,$x20,$out
+	le?vperm	$out4,$out4,$out4,$inpperm
+	stvx_u		$out3,$x30,$out
+	le?vperm	$out5,$out5,$out5,$inpperm
+	stvx_u		$out4,$x40,$out
+	stvx_u		$out5,$x50,$out
+	addi		$out,$out,0x60
+	b		Lctr32_enc8x_done
+
+.align	5
+Lctr32_enc8x_five:
+	vcipherlast	$out0,$out0,$in3
+	vcipherlast	$out1,$out1,$in4
+	vcipherlast	$out2,$out2,$in5
+	vcipherlast	$out3,$out3,$in6
+	vcipherlast	$out4,$out4,$in7
+
+	le?vperm	$out0,$out0,$out0,$inpperm
+	le?vperm	$out1,$out1,$out1,$inpperm
+	stvx_u		$out0,$x00,$out
+	le?vperm	$out2,$out2,$out2,$inpperm
+	stvx_u		$out1,$x10,$out
+	le?vperm	$out3,$out3,$out3,$inpperm
+	stvx_u		$out2,$x20,$out
+	le?vperm	$out4,$out4,$out4,$inpperm
+	stvx_u		$out3,$x30,$out
+	stvx_u		$out4,$x40,$out
+	addi		$out,$out,0x50
+	b		Lctr32_enc8x_done
+
+.align	5
+Lctr32_enc8x_four:
+	vcipherlast	$out0,$out0,$in4
+	vcipherlast	$out1,$out1,$in5
+	vcipherlast	$out2,$out2,$in6
+	vcipherlast	$out3,$out3,$in7
+
+	le?vperm	$out0,$out0,$out0,$inpperm
+	le?vperm	$out1,$out1,$out1,$inpperm
+	stvx_u		$out0,$x00,$out
+	le?vperm	$out2,$out2,$out2,$inpperm
+	stvx_u		$out1,$x10,$out
+	le?vperm	$out3,$out3,$out3,$inpperm
+	stvx_u		$out2,$x20,$out
+	stvx_u		$out3,$x30,$out
+	addi		$out,$out,0x40
+	b		Lctr32_enc8x_done
+
+.align	5
+Lctr32_enc8x_three:
+	vcipherlast	$out0,$out0,$in5
+	vcipherlast	$out1,$out1,$in6
+	vcipherlast	$out2,$out2,$in7
+
+	le?vperm	$out0,$out0,$out0,$inpperm
+	le?vperm	$out1,$out1,$out1,$inpperm
+	stvx_u		$out0,$x00,$out
+	le?vperm	$out2,$out2,$out2,$inpperm
+	stvx_u		$out1,$x10,$out
+	stvx_u		$out2,$x20,$out
+	addi		$out,$out,0x30
+	b		Lctr32_enc8x_done
+
+.align	5
+Lctr32_enc8x_two:
+	vcipherlast	$out0,$out0,$in6
+	vcipherlast	$out1,$out1,$in7
+
+	le?vperm	$out0,$out0,$out0,$inpperm
+	le?vperm	$out1,$out1,$out1,$inpperm
+	stvx_u		$out0,$x00,$out
+	stvx_u		$out1,$x10,$out
+	addi		$out,$out,0x20
+	b		Lctr32_enc8x_done
+
+.align	5
+Lctr32_enc8x_one:
+	vcipherlast	$out0,$out0,$in7
+
+	le?vperm	$out0,$out0,$out0,$inpperm
+	stvx_u		$out0,0,$out
+	addi		$out,$out,0x10
+
+Lctr32_enc8x_done:
+	li		r10,`$FRAME+15`
+	li		r11,`$FRAME+31`
+	stvx		$inpperm,r10,$sp	# wipe copies of round keys
+	addi		r10,r10,32
+	stvx		$inpperm,r11,$sp
+	addi		r11,r11,32
+	stvx		$inpperm,r10,$sp
+	addi		r10,r10,32
+	stvx		$inpperm,r11,$sp
+	addi		r11,r11,32
+	stvx		$inpperm,r10,$sp
+	addi		r10,r10,32
+	stvx		$inpperm,r11,$sp
+	addi		r11,r11,32
+	stvx		$inpperm,r10,$sp
+	addi		r10,r10,32
+	stvx		$inpperm,r11,$sp
+	addi		r11,r11,32
+
+	mtspr		256,$vrsave
+	lvx		v20,r10,$sp		# ABI says so
+	addi		r10,r10,32
+	lvx		v21,r11,$sp
+	addi		r11,r11,32
+	lvx		v22,r10,$sp
+	addi		r10,r10,32
+	lvx		v23,r11,$sp
+	addi		r11,r11,32
+	lvx		v24,r10,$sp
+	addi		r10,r10,32
+	lvx		v25,r11,$sp
+	addi		r11,r11,32
+	lvx		v26,r10,$sp
+	addi		r10,r10,32
+	lvx		v27,r11,$sp
+	addi		r11,r11,32
+	lvx		v28,r10,$sp
+	addi		r10,r10,32
+	lvx		v29,r11,$sp
+	addi		r11,r11,32
+	lvx		v30,r10,$sp
+	lvx		v31,r11,$sp
+	$POP		r26,`$FRAME+21*16+0*$SIZE_T`($sp)
+	$POP		r27,`$FRAME+21*16+1*$SIZE_T`($sp)
+	$POP		r28,`$FRAME+21*16+2*$SIZE_T`($sp)
+	$POP		r29,`$FRAME+21*16+3*$SIZE_T`($sp)
+	$POP		r30,`$FRAME+21*16+4*$SIZE_T`($sp)
+	$POP		r31,`$FRAME+21*16+5*$SIZE_T`($sp)
+	addi		$sp,$sp,`$FRAME+21*16+6*$SIZE_T`
+	blr
+	.long		0
+	.byte		0,12,0x14,0,0x80,6,6,0
+	.long		0
+.size	.${prefix}_ctr32_encrypt_blocks,.-.${prefix}_ctr32_encrypt_blocks
+___
+}}	}}}
+
+#########################################################################
+{{{	# XTS procedures						#
+# int aes_p8_xts_[en|de]crypt(const char *inp, char *out, size_t len,	#
+#                             const AES_KEY *key1, const AES_KEY *key2,	#
+#                             [const] unsigned char iv[16]);		#
+# If $key2 is NULL, then a "tweak chaining" mode is engaged, in which	#
+# input tweak value is assumed to be encrypted already, and last tweak	#
+# value, one suitable for consecutive call on same chunk of data, is	#
+# written back to original buffer. In addition, in "tweak chaining"	#
+# mode only complete input blocks are processed.			#
+
+my ($inp,$out,$len,$key1,$key2,$ivp,$rounds,$idx) =	map("r$_",(3..10));
+my ($rndkey0,$rndkey1,$inout) =				map("v$_",(0..2));
+my ($output,$inptail,$inpperm,$leperm,$keyperm) =	map("v$_",(3..7));
+my ($tweak,$seven,$eighty7,$tmp,$tweak1) =		map("v$_",(8..12));
+my $taillen = $key2;
+
+   ($inp,$idx) = ($idx,$inp);				# reassign
+
+$code.=<<___;
+.globl	.${prefix}_xts_encrypt
+	mr		$inp,r3				# reassign
+	li		r3,-1
+	${UCMP}i	$len,16
+	bltlr-
+
+	lis		r0,0xfff0
+	mfspr		r12,256				# save vrsave
+	li		r11,0
+	mtspr		256,r0
+
+	vspltisb	$seven,0x07			# 0x070707..07
+	le?lvsl		$leperm,r11,r11
+	le?vspltisb	$tmp,0x0f
+	le?vxor		$leperm,$leperm,$seven
+
+	li		$idx,15
+	lvx		$tweak,0,$ivp			# load [unaligned] iv
+	lvsl		$inpperm,0,$ivp
+	lvx		$inptail,$idx,$ivp
+	le?vxor		$inpperm,$inpperm,$tmp
+	vperm		$tweak,$tweak,$inptail,$inpperm
+
+	neg		r11,$inp
+	lvsr		$inpperm,0,r11			# prepare for unaligned load
+	lvx		$inout,0,$inp
+	addi		$inp,$inp,15			# 15 is not typo
+	le?vxor		$inpperm,$inpperm,$tmp
+
+	${UCMP}i	$key2,0				# key2==NULL?
+	beq		Lxts_enc_no_key2
+
+	?lvsl		$keyperm,0,$key2		# prepare for unaligned key
+	lwz		$rounds,240($key2)
+	srwi		$rounds,$rounds,1
+	subi		$rounds,$rounds,1
+	li		$idx,16
+
+	lvx		$rndkey0,0,$key2
+	lvx		$rndkey1,$idx,$key2
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$tweak,$tweak,$rndkey0
+	lvx		$rndkey0,$idx,$key2
+	addi		$idx,$idx,16
+	mtctr		$rounds
+
+Ltweak_xts_enc:
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$tweak,$tweak,$rndkey1
+	lvx		$rndkey1,$idx,$key2
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vcipher		$tweak,$tweak,$rndkey0
+	lvx		$rndkey0,$idx,$key2
+	addi		$idx,$idx,16
+	bdnz		Ltweak_xts_enc
+
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$tweak,$tweak,$rndkey1
+	lvx		$rndkey1,$idx,$key2
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vcipherlast	$tweak,$tweak,$rndkey0
+
+	li		$ivp,0				# don't chain the tweak
+	b		Lxts_enc
+
+Lxts_enc_no_key2:
+	li		$idx,-16
+	and		$len,$len,$idx			# in "tweak chaining"
+							# mode only complete
+							# blocks are processed
+Lxts_enc:
+	lvx		$inptail,0,$inp
+	addi		$inp,$inp,16
+
+	?lvsl		$keyperm,0,$key1		# prepare for unaligned key
+	lwz		$rounds,240($key1)
+	srwi		$rounds,$rounds,1
+	subi		$rounds,$rounds,1
+	li		$idx,16
+
+	vslb		$eighty7,$seven,$seven		# 0x808080..80
+	vor		$eighty7,$eighty7,$seven	# 0x878787..87
+	vspltisb	$tmp,1				# 0x010101..01
+	vsldoi		$eighty7,$eighty7,$tmp,15	# 0x870101..01
+
+	${UCMP}i	$len,96
+	bge		_aesp8_xts_encrypt6x
+
+	andi.		$taillen,$len,15
+	subic		r0,$len,32
+	subi		$taillen,$taillen,16
+	subfe		r0,r0,r0
+	and		r0,r0,$taillen
+	add		$inp,$inp,r0
+
+	lvx		$rndkey0,0,$key1
+	lvx		$rndkey1,$idx,$key1
+	addi		$idx,$idx,16
+	vperm		$inout,$inout,$inptail,$inpperm
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$inout,$inout,$tweak
+	vxor		$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key1
+	addi		$idx,$idx,16
+	mtctr		$rounds
+	b		Loop_xts_enc
+
+.align	5
+Loop_xts_enc:
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key1
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vcipher		$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key1
+	addi		$idx,$idx,16
+	bdnz		Loop_xts_enc
+
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key1
+	li		$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$rndkey0,$rndkey0,$tweak
+	vcipherlast	$output,$inout,$rndkey0
+
+	le?vperm	$tmp,$output,$output,$leperm
+	be?nop
+	le?stvx_u	$tmp,0,$out
+	be?stvx_u	$output,0,$out
+	addi		$out,$out,16
+
+	subic.		$len,$len,16
+	beq		Lxts_enc_done
+
+	vmr		$inout,$inptail
+	lvx		$inptail,0,$inp
+	addi		$inp,$inp,16
+	lvx		$rndkey0,0,$key1
+	lvx		$rndkey1,$idx,$key1
+	addi		$idx,$idx,16
+
+	subic		r0,$len,32
+	subfe		r0,r0,r0
+	and		r0,r0,$taillen
+	add		$inp,$inp,r0
+
+	vsrab		$tmp,$tweak,$seven		# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	vsldoi		$tmp,$tmp,$tmp,15
+	vand		$tmp,$tmp,$eighty7
+	vxor		$tweak,$tweak,$tmp
+
+	vperm		$inout,$inout,$inptail,$inpperm
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$inout,$inout,$tweak
+	vxor		$output,$output,$rndkey0	# just in case $len<16
+	vxor		$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key1
+	addi		$idx,$idx,16
+
+	mtctr		$rounds
+	${UCMP}i	$len,16
+	bge		Loop_xts_enc
+
+	vxor		$output,$output,$tweak
+	lvsr		$inpperm,0,$len			# $inpperm is no longer needed
+	vxor		$inptail,$inptail,$inptail	# $inptail is no longer needed
+	vspltisb	$tmp,-1
+	vperm		$inptail,$inptail,$tmp,$inpperm
+	vsel		$inout,$inout,$output,$inptail
+
+	subi		r11,$out,17
+	subi		$out,$out,16
+	mtctr		$len
+	li		$len,16
+Loop_xts_enc_steal:
+	lbzu		r0,1(r11)
+	stb		r0,16(r11)
+	bdnz		Loop_xts_enc_steal
+
+	mtctr		$rounds
+	b		Loop_xts_enc			# one more time...
+
+Lxts_enc_done:
+	${UCMP}i	$ivp,0
+	beq		Lxts_enc_ret
+
+	vsrab		$tmp,$tweak,$seven		# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	vsldoi		$tmp,$tmp,$tmp,15
+	vand		$tmp,$tmp,$eighty7
+	vxor		$tweak,$tweak,$tmp
+
+	le?vperm	$tweak,$tweak,$tweak,$leperm
+	stvx_u		$tweak,0,$ivp
+
+Lxts_enc_ret:
+	mtspr		256,r12				# restore vrsave
+	li		r3,0
+	blr
+	.long		0
+	.byte		0,12,0x04,0,0x80,6,6,0
+	.long		0
+.size	.${prefix}_xts_encrypt,.-.${prefix}_xts_encrypt
+
+.globl	.${prefix}_xts_decrypt
+	mr		$inp,r3				# reassign
+	li		r3,-1
+	${UCMP}i	$len,16
+	bltlr-
+
+	lis		r0,0xfff8
+	mfspr		r12,256				# save vrsave
+	li		r11,0
+	mtspr		256,r0
+
+	andi.		r0,$len,15
+	neg		r0,r0
+	andi.		r0,r0,16
+	sub		$len,$len,r0
+
+	vspltisb	$seven,0x07			# 0x070707..07
+	le?lvsl		$leperm,r11,r11
+	le?vspltisb	$tmp,0x0f
+	le?vxor		$leperm,$leperm,$seven
+
+	li		$idx,15
+	lvx		$tweak,0,$ivp			# load [unaligned] iv
+	lvsl		$inpperm,0,$ivp
+	lvx		$inptail,$idx,$ivp
+	le?vxor		$inpperm,$inpperm,$tmp
+	vperm		$tweak,$tweak,$inptail,$inpperm
+
+	neg		r11,$inp
+	lvsr		$inpperm,0,r11			# prepare for unaligned load
+	lvx		$inout,0,$inp
+	addi		$inp,$inp,15			# 15 is not typo
+	le?vxor		$inpperm,$inpperm,$tmp
+
+	${UCMP}i	$key2,0				# key2==NULL?
+	beq		Lxts_dec_no_key2
+
+	?lvsl		$keyperm,0,$key2		# prepare for unaligned key
+	lwz		$rounds,240($key2)
+	srwi		$rounds,$rounds,1
+	subi		$rounds,$rounds,1
+	li		$idx,16
+
+	lvx		$rndkey0,0,$key2
+	lvx		$rndkey1,$idx,$key2
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$tweak,$tweak,$rndkey0
+	lvx		$rndkey0,$idx,$key2
+	addi		$idx,$idx,16
+	mtctr		$rounds
+
+Ltweak_xts_dec:
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$tweak,$tweak,$rndkey1
+	lvx		$rndkey1,$idx,$key2
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vcipher		$tweak,$tweak,$rndkey0
+	lvx		$rndkey0,$idx,$key2
+	addi		$idx,$idx,16
+	bdnz		Ltweak_xts_dec
+
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vcipher		$tweak,$tweak,$rndkey1
+	lvx		$rndkey1,$idx,$key2
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vcipherlast	$tweak,$tweak,$rndkey0
+
+	li		$ivp,0				# don't chain the tweak
+	b		Lxts_dec
+
+Lxts_dec_no_key2:
+	neg		$idx,$len
+	andi.		$idx,$idx,15
+	add		$len,$len,$idx			# in "tweak chaining"
+							# mode only complete
+							# blocks are processed
+Lxts_dec:
+	lvx		$inptail,0,$inp
+	addi		$inp,$inp,16
+
+	?lvsl		$keyperm,0,$key1		# prepare for unaligned key
+	lwz		$rounds,240($key1)
+	srwi		$rounds,$rounds,1
+	subi		$rounds,$rounds,1
+	li		$idx,16
+
+	vslb		$eighty7,$seven,$seven		# 0x808080..80
+	vor		$eighty7,$eighty7,$seven	# 0x878787..87
+	vspltisb	$tmp,1				# 0x010101..01
+	vsldoi		$eighty7,$eighty7,$tmp,15	# 0x870101..01
+
+	${UCMP}i	$len,96
+	bge		_aesp8_xts_decrypt6x
+
+	lvx		$rndkey0,0,$key1
+	lvx		$rndkey1,$idx,$key1
+	addi		$idx,$idx,16
+	vperm		$inout,$inout,$inptail,$inpperm
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$inout,$inout,$tweak
+	vxor		$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key1
+	addi		$idx,$idx,16
+	mtctr		$rounds
+
+	${UCMP}i	$len,16
+	blt		Ltail_xts_dec
+	be?b		Loop_xts_dec
+
+.align	5
+Loop_xts_dec:
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vncipher	$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key1
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vncipher	$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key1
+	addi		$idx,$idx,16
+	bdnz		Loop_xts_dec
+
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vncipher	$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key1
+	li		$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$rndkey0,$rndkey0,$tweak
+	vncipherlast	$output,$inout,$rndkey0
+
+	le?vperm	$tmp,$output,$output,$leperm
+	be?nop
+	le?stvx_u	$tmp,0,$out
+	be?stvx_u	$output,0,$out
+	addi		$out,$out,16
+
+	subic.		$len,$len,16
+	beq		Lxts_dec_done
+
+	vmr		$inout,$inptail
+	lvx		$inptail,0,$inp
+	addi		$inp,$inp,16
+	lvx		$rndkey0,0,$key1
+	lvx		$rndkey1,$idx,$key1
+	addi		$idx,$idx,16
+
+	vsrab		$tmp,$tweak,$seven		# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	vsldoi		$tmp,$tmp,$tmp,15
+	vand		$tmp,$tmp,$eighty7
+	vxor		$tweak,$tweak,$tmp
+
+	vperm		$inout,$inout,$inptail,$inpperm
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$inout,$inout,$tweak
+	vxor		$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key1
+	addi		$idx,$idx,16
+
+	mtctr		$rounds
+	${UCMP}i	$len,16
+	bge		Loop_xts_dec
+
+Ltail_xts_dec:
+	vsrab		$tmp,$tweak,$seven		# next tweak value
+	vaddubm		$tweak1,$tweak,$tweak
+	vsldoi		$tmp,$tmp,$tmp,15
+	vand		$tmp,$tmp,$eighty7
+	vxor		$tweak1,$tweak1,$tmp
+
+	subi		$inp,$inp,16
+	add		$inp,$inp,$len
+
+	vxor		$inout,$inout,$tweak		# :-(
+	vxor		$inout,$inout,$tweak1		# :-)
+
+Loop_xts_dec_short:
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vncipher	$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key1
+	addi		$idx,$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vncipher	$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key1
+	addi		$idx,$idx,16
+	bdnz		Loop_xts_dec_short
+
+	?vperm		$rndkey1,$rndkey1,$rndkey0,$keyperm
+	vncipher	$inout,$inout,$rndkey1
+	lvx		$rndkey1,$idx,$key1
+	li		$idx,16
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+	vxor		$rndkey0,$rndkey0,$tweak1
+	vncipherlast	$output,$inout,$rndkey0
+
+	le?vperm	$tmp,$output,$output,$leperm
+	be?nop
+	le?stvx_u	$tmp,0,$out
+	be?stvx_u	$output,0,$out
+
+	vmr		$inout,$inptail
+	lvx		$inptail,0,$inp
+	#addi		$inp,$inp,16
+	lvx		$rndkey0,0,$key1
+	lvx		$rndkey1,$idx,$key1
+	addi		$idx,$idx,16
+	vperm		$inout,$inout,$inptail,$inpperm
+	?vperm		$rndkey0,$rndkey0,$rndkey1,$keyperm
+
+	lvsr		$inpperm,0,$len			# $inpperm is no longer needed
+	vxor		$inptail,$inptail,$inptail	# $inptail is no longer needed
+	vspltisb	$tmp,-1
+	vperm		$inptail,$inptail,$tmp,$inpperm
+	vsel		$inout,$inout,$output,$inptail
+
+	vxor		$rndkey0,$rndkey0,$tweak
+	vxor		$inout,$inout,$rndkey0
+	lvx		$rndkey0,$idx,$key1
+	addi		$idx,$idx,16
+
+	subi		r11,$out,1
+	mtctr		$len
+	li		$len,16
+Loop_xts_dec_steal:
+	lbzu		r0,1(r11)
+	stb		r0,16(r11)
+	bdnz		Loop_xts_dec_steal
+
+	mtctr		$rounds
+	b		Loop_xts_dec			# one more time...
+
+Lxts_dec_done:
+	${UCMP}i	$ivp,0
+	beq		Lxts_dec_ret
+
+	vsrab		$tmp,$tweak,$seven		# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	vsldoi		$tmp,$tmp,$tmp,15
+	vand		$tmp,$tmp,$eighty7
+	vxor		$tweak,$tweak,$tmp
+
+	le?vperm	$tweak,$tweak,$tweak,$leperm
+	stvx_u		$tweak,0,$ivp
+
+Lxts_dec_ret:
+	mtspr		256,r12				# restore vrsave
+	li		r3,0
+	blr
+	.long		0
+	.byte		0,12,0x04,0,0x80,6,6,0
+	.long		0
+.size	.${prefix}_xts_decrypt,.-.${prefix}_xts_decrypt
+___
+#########################################################################
+{{	# Optimized XTS procedures					#
+my $key_=$key2;
+my ($x00,$x10,$x20,$x30,$x40,$x50,$x60,$x70)=map("r$_",(0,3,26..31));
+    $x00=0 if ($flavour =~ /osx/);
+my ($in0,  $in1,  $in2,  $in3,  $in4,  $in5 )=map("v$_",(0..5));
+my ($out0, $out1, $out2, $out3, $out4, $out5)=map("v$_",(7,12..16));
+my ($twk0, $twk1, $twk2, $twk3, $twk4, $twk5)=map("v$_",(17..22));
+my $rndkey0="v23";	# v24-v25 rotating buffer for first found keys
+			# v26-v31 last 6 round keys
+my ($keyperm)=($out0);	# aliases with "caller", redundant assignment
+my $taillen=$x70;
+
+$code.=<<___;
+.align	5
+_aesp8_xts_encrypt6x:
+	$STU		$sp,-`($FRAME+21*16+6*$SIZE_T)`($sp)
+	mflr		r11
+	li		r7,`$FRAME+8*16+15`
+	li		r3,`$FRAME+8*16+31`
+	$PUSH		r11,`$FRAME+21*16+6*$SIZE_T+$LRSAVE`($sp)
+	stvx		v20,r7,$sp		# ABI says so
+	addi		r7,r7,32
+	stvx		v21,r3,$sp
+	addi		r3,r3,32
+	stvx		v22,r7,$sp
+	addi		r7,r7,32
+	stvx		v23,r3,$sp
+	addi		r3,r3,32
+	stvx		v24,r7,$sp
+	addi		r7,r7,32
+	stvx		v25,r3,$sp
+	addi		r3,r3,32
+	stvx		v26,r7,$sp
+	addi		r7,r7,32
+	stvx		v27,r3,$sp
+	addi		r3,r3,32
+	stvx		v28,r7,$sp
+	addi		r7,r7,32
+	stvx		v29,r3,$sp
+	addi		r3,r3,32
+	stvx		v30,r7,$sp
+	stvx		v31,r3,$sp
+	li		r0,-1
+	stw		$vrsave,`$FRAME+21*16-4`($sp)	# save vrsave
+	li		$x10,0x10
+	$PUSH		r26,`$FRAME+21*16+0*$SIZE_T`($sp)
+	li		$x20,0x20
+	$PUSH		r27,`$FRAME+21*16+1*$SIZE_T`($sp)
+	li		$x30,0x30
+	$PUSH		r28,`$FRAME+21*16+2*$SIZE_T`($sp)
+	li		$x40,0x40
+	$PUSH		r29,`$FRAME+21*16+3*$SIZE_T`($sp)
+	li		$x50,0x50
+	$PUSH		r30,`$FRAME+21*16+4*$SIZE_T`($sp)
+	li		$x60,0x60
+	$PUSH		r31,`$FRAME+21*16+5*$SIZE_T`($sp)
+	li		$x70,0x70
+	mtspr		256,r0
+
+	xxlor		2, 32+$eighty7, 32+$eighty7
+	vsldoi		$eighty7,$tmp,$eighty7,1        # 0x010101..87
+	xxlor		1, 32+$eighty7, 32+$eighty7
+
+	# Load XOR Lconsts.
+	mr		$x70, r6
+	bl		Lconsts
+	lxvw4x		0, $x40, r6		# load XOR contents
+	mr		r6, $x70
+	li		$x70,0x70
+
+	subi		$rounds,$rounds,3	# -4 in total
+
+	lvx		$rndkey0,$x00,$key1	# load key schedule
+	lvx		v30,$x10,$key1
+	addi		$key1,$key1,0x20
+	lvx		v31,$x00,$key1
+	?vperm		$rndkey0,$rndkey0,v30,$keyperm
+	addi		$key_,$sp,$FRAME+15
+	mtctr		$rounds
+
+Load_xts_enc_key:
+	?vperm		v24,v30,v31,$keyperm
+	lvx		v30,$x10,$key1
+	addi		$key1,$key1,0x20
+	stvx		v24,$x00,$key_		# off-load round[1]
+	?vperm		v25,v31,v30,$keyperm
+	lvx		v31,$x00,$key1
+	stvx		v25,$x10,$key_		# off-load round[2]
+	addi		$key_,$key_,0x20
+	bdnz		Load_xts_enc_key
+
+	lvx		v26,$x10,$key1
+	?vperm		v24,v30,v31,$keyperm
+	lvx		v27,$x20,$key1
+	stvx		v24,$x00,$key_		# off-load round[3]
+	?vperm		v25,v31,v26,$keyperm
+	lvx		v28,$x30,$key1
+	stvx		v25,$x10,$key_		# off-load round[4]
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	?vperm		v26,v26,v27,$keyperm
+	lvx		v29,$x40,$key1
+	?vperm		v27,v27,v28,$keyperm
+	lvx		v30,$x50,$key1
+	?vperm		v28,v28,v29,$keyperm
+	lvx		v31,$x60,$key1
+	?vperm		v29,v29,v30,$keyperm
+	lvx		$twk5,$x70,$key1	# borrow $twk5
+	?vperm		v30,v30,v31,$keyperm
+	lvx		v24,$x00,$key_		# pre-load round[1]
+	?vperm		v31,v31,$twk5,$keyperm
+	lvx		v25,$x10,$key_		# pre-load round[2]
+
+	# Switch to use the following codes with 0x010101..87 to generate tweak.
+	#     eighty7 = 0x010101..87
+	# vsrab         tmp, tweak, seven       # next tweak value, right shift 7 bits
+	# vand          tmp, tmp, eighty7       # last byte with carry
+	# vaddubm       tweak, tweak, tweak     # left shift 1 bit (x2)
+	# xxlor         vsx, 0, 0
+	# vpermxor      tweak, tweak, tmp, vsx
+
+	 vperm		$in0,$inout,$inptail,$inpperm
+	 subi		$inp,$inp,31		# undo "caller"
+	vxor		$twk0,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out0,$in0,$twk0
+	xxlor		32+$in1, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in1
+
+	 lvx_u		$in1,$x10,$inp
+	vxor		$twk1,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in1,$in1,$in1,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out1,$in1,$twk1
+	xxlor		32+$in2, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in2
+
+	 lvx_u		$in2,$x20,$inp
+	 andi.		$taillen,$len,15
+	vxor		$twk2,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in2,$in2,$in2,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out2,$in2,$twk2
+	xxlor		32+$in3, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in3
+
+	 lvx_u		$in3,$x30,$inp
+	 sub		$len,$len,$taillen
+	vxor		$twk3,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in3,$in3,$in3,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out3,$in3,$twk3
+	xxlor		32+$in4, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in4
+
+	 lvx_u		$in4,$x40,$inp
+	 subi		$len,$len,0x60
+	vxor		$twk4,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in4,$in4,$in4,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out4,$in4,$twk4
+	xxlor		32+$in5, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in5
+
+	 lvx_u		$in5,$x50,$inp
+	 addi		$inp,$inp,0x60
+	vxor		$twk5,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in5,$in5,$in5,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out5,$in5,$twk5
+	xxlor		32+$in0, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in0
+
+	vxor		v31,v31,$rndkey0
+	mtctr		$rounds
+	b		Loop_xts_enc6x
+
+.align	5
+Loop_xts_enc6x:
+	vcipher		$out0,$out0,v24
+	vcipher		$out1,$out1,v24
+	vcipher		$out2,$out2,v24
+	vcipher		$out3,$out3,v24
+	vcipher		$out4,$out4,v24
+	vcipher		$out5,$out5,v24
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vcipher		$out0,$out0,v25
+	vcipher		$out1,$out1,v25
+	vcipher		$out2,$out2,v25
+	vcipher		$out3,$out3,v25
+	vcipher		$out4,$out4,v25
+	vcipher		$out5,$out5,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		Loop_xts_enc6x
+
+	xxlor		32+$eighty7, 1, 1	# 0x010101..87
+
+	subic		$len,$len,96		# $len-=96
+	 vxor		$in0,$twk0,v31		# xor with last round key
+	vcipher		$out0,$out0,v24
+	vcipher		$out1,$out1,v24
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk0,$tweak,$rndkey0
+	 vaddubm	$tweak,$tweak,$tweak
+	vcipher		$out2,$out2,v24
+	vcipher		$out3,$out3,v24
+	vcipher		$out4,$out4,v24
+	vcipher		$out5,$out5,v24
+
+	subfe.		r0,r0,r0		# borrow?-1:0
+	 vand		$tmp,$tmp,$eighty7
+	vcipher		$out0,$out0,v25
+	vcipher		$out1,$out1,v25
+	 xxlor		32+$in1, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in1
+	vcipher		$out2,$out2,v25
+	vcipher		$out3,$out3,v25
+	 vxor		$in1,$twk1,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk1,$tweak,$rndkey0
+	vcipher		$out4,$out4,v25
+	vcipher		$out5,$out5,v25
+
+	and		r0,r0,$len
+	 vaddubm	$tweak,$tweak,$tweak
+	vcipher		$out0,$out0,v26
+	vcipher		$out1,$out1,v26
+	 vand		$tmp,$tmp,$eighty7
+	vcipher		$out2,$out2,v26
+	vcipher		$out3,$out3,v26
+	 xxlor		32+$in2, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in2
+	vcipher		$out4,$out4,v26
+	vcipher		$out5,$out5,v26
+
+	add		$inp,$inp,r0		# $inp is adjusted in such
+						# way that at exit from the
+						# loop inX-in5 are loaded
+						# with last "words"
+	 vxor		$in2,$twk2,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk2,$tweak,$rndkey0
+	 vaddubm	$tweak,$tweak,$tweak
+	vcipher		$out0,$out0,v27
+	vcipher		$out1,$out1,v27
+	vcipher		$out2,$out2,v27
+	vcipher		$out3,$out3,v27
+	 vand		$tmp,$tmp,$eighty7
+	vcipher		$out4,$out4,v27
+	vcipher		$out5,$out5,v27
+
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	 xxlor		32+$in3, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in3
+	vcipher		$out0,$out0,v28
+	vcipher		$out1,$out1,v28
+	 vxor		$in3,$twk3,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk3,$tweak,$rndkey0
+	vcipher		$out2,$out2,v28
+	vcipher		$out3,$out3,v28
+	 vaddubm	$tweak,$tweak,$tweak
+	vcipher		$out4,$out4,v28
+	vcipher		$out5,$out5,v28
+	lvx		v24,$x00,$key_		# re-pre-load round[1]
+	 vand		$tmp,$tmp,$eighty7
+
+	vcipher		$out0,$out0,v29
+	vcipher		$out1,$out1,v29
+	 xxlor		32+$in4, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in4
+	vcipher		$out2,$out2,v29
+	vcipher		$out3,$out3,v29
+	 vxor		$in4,$twk4,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk4,$tweak,$rndkey0
+	vcipher		$out4,$out4,v29
+	vcipher		$out5,$out5,v29
+	lvx		v25,$x10,$key_		# re-pre-load round[2]
+	 vaddubm	$tweak,$tweak,$tweak
+
+	vcipher		$out0,$out0,v30
+	vcipher		$out1,$out1,v30
+	 vand		$tmp,$tmp,$eighty7
+	vcipher		$out2,$out2,v30
+	vcipher		$out3,$out3,v30
+	 xxlor		32+$in5, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in5
+	vcipher		$out4,$out4,v30
+	vcipher		$out5,$out5,v30
+	 vxor		$in5,$twk5,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk5,$tweak,$rndkey0
+
+	vcipherlast	$out0,$out0,$in0
+	 lvx_u		$in0,$x00,$inp		# load next input block
+	 vaddubm	$tweak,$tweak,$tweak
+	vcipherlast	$out1,$out1,$in1
+	 lvx_u		$in1,$x10,$inp
+	vcipherlast	$out2,$out2,$in2
+	 le?vperm	$in0,$in0,$in0,$leperm
+	 lvx_u		$in2,$x20,$inp
+	 vand		$tmp,$tmp,$eighty7
+	vcipherlast	$out3,$out3,$in3
+	 le?vperm	$in1,$in1,$in1,$leperm
+	 lvx_u		$in3,$x30,$inp
+	vcipherlast	$out4,$out4,$in4
+	 le?vperm	$in2,$in2,$in2,$leperm
+	 lvx_u		$in4,$x40,$inp
+	 xxlor		10, 32+$in0, 32+$in0
+	 xxlor		32+$in0, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in0
+	 xxlor		32+$in0, 10, 10
+	vcipherlast	$tmp,$out5,$in5		# last block might be needed
+						# in stealing mode
+	 le?vperm	$in3,$in3,$in3,$leperm
+	 lvx_u		$in5,$x50,$inp
+	 addi		$inp,$inp,0x60
+	 le?vperm	$in4,$in4,$in4,$leperm
+	 le?vperm	$in5,$in5,$in5,$leperm
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	 vxor		$out0,$in0,$twk0
+	le?vperm	$out2,$out2,$out2,$leperm
+	stvx_u		$out1,$x10,$out
+	 vxor		$out1,$in1,$twk1
+	le?vperm	$out3,$out3,$out3,$leperm
+	stvx_u		$out2,$x20,$out
+	 vxor		$out2,$in2,$twk2
+	le?vperm	$out4,$out4,$out4,$leperm
+	stvx_u		$out3,$x30,$out
+	 vxor		$out3,$in3,$twk3
+	le?vperm	$out5,$tmp,$tmp,$leperm
+	stvx_u		$out4,$x40,$out
+	 vxor		$out4,$in4,$twk4
+	le?stvx_u	$out5,$x50,$out
+	be?stvx_u	$tmp, $x50,$out
+	 vxor		$out5,$in5,$twk5
+	addi		$out,$out,0x60
+
+	mtctr		$rounds
+	beq		Loop_xts_enc6x		# did $len-=96 borrow?
+
+	xxlor		32+$eighty7, 2, 2	# 0x010101..87
+
+	addic.		$len,$len,0x60
+	beq		Lxts_enc6x_zero
+	cmpwi		$len,0x20
+	blt		Lxts_enc6x_one
+	nop
+	beq		Lxts_enc6x_two
+	cmpwi		$len,0x40
+	blt		Lxts_enc6x_three
+	nop
+	beq		Lxts_enc6x_four
+
+Lxts_enc6x_five:
+	vxor		$out0,$in1,$twk0
+	vxor		$out1,$in2,$twk1
+	vxor		$out2,$in3,$twk2
+	vxor		$out3,$in4,$twk3
+	vxor		$out4,$in5,$twk4
+
+	bl		_aesp8_xts_enc5x
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	vmr		$twk0,$twk5		# unused tweak
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	le?vperm	$out2,$out2,$out2,$leperm
+	stvx_u		$out1,$x10,$out
+	le?vperm	$out3,$out3,$out3,$leperm
+	stvx_u		$out2,$x20,$out
+	vxor		$tmp,$out4,$twk5	# last block prep for stealing
+	le?vperm	$out4,$out4,$out4,$leperm
+	stvx_u		$out3,$x30,$out
+	stvx_u		$out4,$x40,$out
+	addi		$out,$out,0x50
+	bne		Lxts_enc6x_steal
+	b		Lxts_enc6x_done
+
+.align	4
+Lxts_enc6x_four:
+	vxor		$out0,$in2,$twk0
+	vxor		$out1,$in3,$twk1
+	vxor		$out2,$in4,$twk2
+	vxor		$out3,$in5,$twk3
+	vxor		$out4,$out4,$out4
+
+	bl		_aesp8_xts_enc5x
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	vmr		$twk0,$twk4		# unused tweak
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	le?vperm	$out2,$out2,$out2,$leperm
+	stvx_u		$out1,$x10,$out
+	vxor		$tmp,$out3,$twk4	# last block prep for stealing
+	le?vperm	$out3,$out3,$out3,$leperm
+	stvx_u		$out2,$x20,$out
+	stvx_u		$out3,$x30,$out
+	addi		$out,$out,0x40
+	bne		Lxts_enc6x_steal
+	b		Lxts_enc6x_done
+
+.align	4
+Lxts_enc6x_three:
+	vxor		$out0,$in3,$twk0
+	vxor		$out1,$in4,$twk1
+	vxor		$out2,$in5,$twk2
+	vxor		$out3,$out3,$out3
+	vxor		$out4,$out4,$out4
+
+	bl		_aesp8_xts_enc5x
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	vmr		$twk0,$twk3		# unused tweak
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	vxor		$tmp,$out2,$twk3	# last block prep for stealing
+	le?vperm	$out2,$out2,$out2,$leperm
+	stvx_u		$out1,$x10,$out
+	stvx_u		$out2,$x20,$out
+	addi		$out,$out,0x30
+	bne		Lxts_enc6x_steal
+	b		Lxts_enc6x_done
+
+.align	4
+Lxts_enc6x_two:
+	vxor		$out0,$in4,$twk0
+	vxor		$out1,$in5,$twk1
+	vxor		$out2,$out2,$out2
+	vxor		$out3,$out3,$out3
+	vxor		$out4,$out4,$out4
+
+	bl		_aesp8_xts_enc5x
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	vmr		$twk0,$twk2		# unused tweak
+	vxor		$tmp,$out1,$twk2	# last block prep for stealing
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	stvx_u		$out1,$x10,$out
+	addi		$out,$out,0x20
+	bne		Lxts_enc6x_steal
+	b		Lxts_enc6x_done
+
+.align	4
+Lxts_enc6x_one:
+	vxor		$out0,$in5,$twk0
+	nop
+Loop_xts_enc1x:
+	vcipher		$out0,$out0,v24
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vcipher		$out0,$out0,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		Loop_xts_enc1x
+
+	add		$inp,$inp,$taillen
+	cmpwi		$taillen,0
+	vcipher		$out0,$out0,v24
+
+	subi		$inp,$inp,16
+	vcipher		$out0,$out0,v25
+
+	lvsr		$inpperm,0,$taillen
+	vcipher		$out0,$out0,v26
+
+	lvx_u		$in0,0,$inp
+	vcipher		$out0,$out0,v27
+
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	vcipher		$out0,$out0,v28
+	lvx		v24,$x00,$key_		# re-pre-load round[1]
+
+	vcipher		$out0,$out0,v29
+	lvx		v25,$x10,$key_		# re-pre-load round[2]
+	 vxor		$twk0,$twk0,v31
+
+	le?vperm	$in0,$in0,$in0,$leperm
+	vcipher		$out0,$out0,v30
+
+	vperm		$in0,$in0,$in0,$inpperm
+	vcipherlast	$out0,$out0,$twk0
+
+	vmr		$twk0,$twk1		# unused tweak
+	vxor		$tmp,$out0,$twk1	# last block prep for stealing
+	le?vperm	$out0,$out0,$out0,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	addi		$out,$out,0x10
+	bne		Lxts_enc6x_steal
+	b		Lxts_enc6x_done
+
+.align	4
+Lxts_enc6x_zero:
+	cmpwi		$taillen,0
+	beq		Lxts_enc6x_done
+
+	add		$inp,$inp,$taillen
+	subi		$inp,$inp,16
+	lvx_u		$in0,0,$inp
+	lvsr		$inpperm,0,$taillen	# $in5 is no more
+	le?vperm	$in0,$in0,$in0,$leperm
+	vperm		$in0,$in0,$in0,$inpperm
+	vxor		$tmp,$tmp,$twk0
+Lxts_enc6x_steal:
+	vxor		$in0,$in0,$twk0
+	vxor		$out0,$out0,$out0
+	vspltisb	$out1,-1
+	vperm		$out0,$out0,$out1,$inpperm
+	vsel		$out0,$in0,$tmp,$out0	# $tmp is last block, remember?
+
+	subi		r30,$out,17
+	subi		$out,$out,16
+	mtctr		$taillen
+Loop_xts_enc6x_steal:
+	lbzu		r0,1(r30)
+	stb		r0,16(r30)
+	bdnz		Loop_xts_enc6x_steal
+
+	li		$taillen,0
+	mtctr		$rounds
+	b		Loop_xts_enc1x		# one more time...
+
+.align	4
+Lxts_enc6x_done:
+	${UCMP}i	$ivp,0
+	beq		Lxts_enc6x_ret
+
+	vxor		$tweak,$twk0,$rndkey0
+	le?vperm	$tweak,$tweak,$tweak,$leperm
+	stvx_u		$tweak,0,$ivp
+
+Lxts_enc6x_ret:
+	mtlr		r11
+	li		r10,`$FRAME+15`
+	li		r11,`$FRAME+31`
+	stvx		$seven,r10,$sp		# wipe copies of round keys
+	addi		r10,r10,32
+	stvx		$seven,r11,$sp
+	addi		r11,r11,32
+	stvx		$seven,r10,$sp
+	addi		r10,r10,32
+	stvx		$seven,r11,$sp
+	addi		r11,r11,32
+	stvx		$seven,r10,$sp
+	addi		r10,r10,32
+	stvx		$seven,r11,$sp
+	addi		r11,r11,32
+	stvx		$seven,r10,$sp
+	addi		r10,r10,32
+	stvx		$seven,r11,$sp
+	addi		r11,r11,32
+
+	mtspr		256,$vrsave
+	lvx		v20,r10,$sp		# ABI says so
+	addi		r10,r10,32
+	lvx		v21,r11,$sp
+	addi		r11,r11,32
+	lvx		v22,r10,$sp
+	addi		r10,r10,32
+	lvx		v23,r11,$sp
+	addi		r11,r11,32
+	lvx		v24,r10,$sp
+	addi		r10,r10,32
+	lvx		v25,r11,$sp
+	addi		r11,r11,32
+	lvx		v26,r10,$sp
+	addi		r10,r10,32
+	lvx		v27,r11,$sp
+	addi		r11,r11,32
+	lvx		v28,r10,$sp
+	addi		r10,r10,32
+	lvx		v29,r11,$sp
+	addi		r11,r11,32
+	lvx		v30,r10,$sp
+	lvx		v31,r11,$sp
+	$POP		r26,`$FRAME+21*16+0*$SIZE_T`($sp)
+	$POP		r27,`$FRAME+21*16+1*$SIZE_T`($sp)
+	$POP		r28,`$FRAME+21*16+2*$SIZE_T`($sp)
+	$POP		r29,`$FRAME+21*16+3*$SIZE_T`($sp)
+	$POP		r30,`$FRAME+21*16+4*$SIZE_T`($sp)
+	$POP		r31,`$FRAME+21*16+5*$SIZE_T`($sp)
+	addi		$sp,$sp,`$FRAME+21*16+6*$SIZE_T`
+	blr
+	.long		0
+	.byte		0,12,0x04,1,0x80,6,6,0
+	.long		0
+
+.align	5
+_aesp8_xts_enc5x:
+	vcipher		$out0,$out0,v24
+	vcipher		$out1,$out1,v24
+	vcipher		$out2,$out2,v24
+	vcipher		$out3,$out3,v24
+	vcipher		$out4,$out4,v24
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vcipher		$out0,$out0,v25
+	vcipher		$out1,$out1,v25
+	vcipher		$out2,$out2,v25
+	vcipher		$out3,$out3,v25
+	vcipher		$out4,$out4,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		_aesp8_xts_enc5x
+
+	add		$inp,$inp,$taillen
+	cmpwi		$taillen,0
+	vcipher		$out0,$out0,v24
+	vcipher		$out1,$out1,v24
+	vcipher		$out2,$out2,v24
+	vcipher		$out3,$out3,v24
+	vcipher		$out4,$out4,v24
+
+	subi		$inp,$inp,16
+	vcipher		$out0,$out0,v25
+	vcipher		$out1,$out1,v25
+	vcipher		$out2,$out2,v25
+	vcipher		$out3,$out3,v25
+	vcipher		$out4,$out4,v25
+	 vxor		$twk0,$twk0,v31
+
+	vcipher		$out0,$out0,v26
+	lvsr		$inpperm,r0,$taillen	# $in5 is no more
+	vcipher		$out1,$out1,v26
+	vcipher		$out2,$out2,v26
+	vcipher		$out3,$out3,v26
+	vcipher		$out4,$out4,v26
+	 vxor		$in1,$twk1,v31
+
+	vcipher		$out0,$out0,v27
+	lvx_u		$in0,0,$inp
+	vcipher		$out1,$out1,v27
+	vcipher		$out2,$out2,v27
+	vcipher		$out3,$out3,v27
+	vcipher		$out4,$out4,v27
+	 vxor		$in2,$twk2,v31
+
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	vcipher		$out0,$out0,v28
+	vcipher		$out1,$out1,v28
+	vcipher		$out2,$out2,v28
+	vcipher		$out3,$out3,v28
+	vcipher		$out4,$out4,v28
+	lvx		v24,$x00,$key_		# re-pre-load round[1]
+	 vxor		$in3,$twk3,v31
+
+	vcipher		$out0,$out0,v29
+	le?vperm	$in0,$in0,$in0,$leperm
+	vcipher		$out1,$out1,v29
+	vcipher		$out2,$out2,v29
+	vcipher		$out3,$out3,v29
+	vcipher		$out4,$out4,v29
+	lvx		v25,$x10,$key_		# re-pre-load round[2]
+	 vxor		$in4,$twk4,v31
+
+	vcipher		$out0,$out0,v30
+	vperm		$in0,$in0,$in0,$inpperm
+	vcipher		$out1,$out1,v30
+	vcipher		$out2,$out2,v30
+	vcipher		$out3,$out3,v30
+	vcipher		$out4,$out4,v30
+
+	vcipherlast	$out0,$out0,$twk0
+	vcipherlast	$out1,$out1,$in1
+	vcipherlast	$out2,$out2,$in2
+	vcipherlast	$out3,$out3,$in3
+	vcipherlast	$out4,$out4,$in4
+	blr
+        .long   	0
+        .byte   	0,12,0x14,0,0,0,0,0
+
+.align	5
+_aesp8_xts_decrypt6x:
+	$STU		$sp,-`($FRAME+21*16+6*$SIZE_T)`($sp)
+	mflr		r11
+	li		r7,`$FRAME+8*16+15`
+	li		r3,`$FRAME+8*16+31`
+	$PUSH		r11,`$FRAME+21*16+6*$SIZE_T+$LRSAVE`($sp)
+	stvx		v20,r7,$sp		# ABI says so
+	addi		r7,r7,32
+	stvx		v21,r3,$sp
+	addi		r3,r3,32
+	stvx		v22,r7,$sp
+	addi		r7,r7,32
+	stvx		v23,r3,$sp
+	addi		r3,r3,32
+	stvx		v24,r7,$sp
+	addi		r7,r7,32
+	stvx		v25,r3,$sp
+	addi		r3,r3,32
+	stvx		v26,r7,$sp
+	addi		r7,r7,32
+	stvx		v27,r3,$sp
+	addi		r3,r3,32
+	stvx		v28,r7,$sp
+	addi		r7,r7,32
+	stvx		v29,r3,$sp
+	addi		r3,r3,32
+	stvx		v30,r7,$sp
+	stvx		v31,r3,$sp
+	li		r0,-1
+	stw		$vrsave,`$FRAME+21*16-4`($sp)	# save vrsave
+	li		$x10,0x10
+	$PUSH		r26,`$FRAME+21*16+0*$SIZE_T`($sp)
+	li		$x20,0x20
+	$PUSH		r27,`$FRAME+21*16+1*$SIZE_T`($sp)
+	li		$x30,0x30
+	$PUSH		r28,`$FRAME+21*16+2*$SIZE_T`($sp)
+	li		$x40,0x40
+	$PUSH		r29,`$FRAME+21*16+3*$SIZE_T`($sp)
+	li		$x50,0x50
+	$PUSH		r30,`$FRAME+21*16+4*$SIZE_T`($sp)
+	li		$x60,0x60
+	$PUSH		r31,`$FRAME+21*16+5*$SIZE_T`($sp)
+	li		$x70,0x70
+	mtspr		256,r0
+
+	xxlor		2, 32+$eighty7, 32+$eighty7
+	vsldoi		$eighty7,$tmp,$eighty7,1        # 0x010101..87
+	xxlor		1, 32+$eighty7, 32+$eighty7
+
+	# Load XOR Lconsts.
+	mr		$x70, r6
+	bl		Lconsts
+	lxvw4x		0, $x40, r6		# load XOR contents
+	mr		r6, $x70
+	li		$x70,0x70
+
+	subi		$rounds,$rounds,3	# -4 in total
+
+	lvx		$rndkey0,$x00,$key1	# load key schedule
+	lvx		v30,$x10,$key1
+	addi		$key1,$key1,0x20
+	lvx		v31,$x00,$key1
+	?vperm		$rndkey0,$rndkey0,v30,$keyperm
+	addi		$key_,$sp,$FRAME+15
+	mtctr		$rounds
+
+Load_xts_dec_key:
+	?vperm		v24,v30,v31,$keyperm
+	lvx		v30,$x10,$key1
+	addi		$key1,$key1,0x20
+	stvx		v24,$x00,$key_		# off-load round[1]
+	?vperm		v25,v31,v30,$keyperm
+	lvx		v31,$x00,$key1
+	stvx		v25,$x10,$key_		# off-load round[2]
+	addi		$key_,$key_,0x20
+	bdnz		Load_xts_dec_key
+
+	lvx		v26,$x10,$key1
+	?vperm		v24,v30,v31,$keyperm
+	lvx		v27,$x20,$key1
+	stvx		v24,$x00,$key_		# off-load round[3]
+	?vperm		v25,v31,v26,$keyperm
+	lvx		v28,$x30,$key1
+	stvx		v25,$x10,$key_		# off-load round[4]
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	?vperm		v26,v26,v27,$keyperm
+	lvx		v29,$x40,$key1
+	?vperm		v27,v27,v28,$keyperm
+	lvx		v30,$x50,$key1
+	?vperm		v28,v28,v29,$keyperm
+	lvx		v31,$x60,$key1
+	?vperm		v29,v29,v30,$keyperm
+	lvx		$twk5,$x70,$key1	# borrow $twk5
+	?vperm		v30,v30,v31,$keyperm
+	lvx		v24,$x00,$key_		# pre-load round[1]
+	?vperm		v31,v31,$twk5,$keyperm
+	lvx		v25,$x10,$key_		# pre-load round[2]
+
+	 vperm		$in0,$inout,$inptail,$inpperm
+	 subi		$inp,$inp,31		# undo "caller"
+	vxor		$twk0,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out0,$in0,$twk0
+	xxlor		32+$in1, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in1
+
+	 lvx_u		$in1,$x10,$inp
+	vxor		$twk1,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in1,$in1,$in1,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out1,$in1,$twk1
+	xxlor		32+$in2, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in2
+
+	 lvx_u		$in2,$x20,$inp
+	 andi.		$taillen,$len,15
+	vxor		$twk2,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in2,$in2,$in2,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out2,$in2,$twk2
+	xxlor		32+$in3, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in3
+
+	 lvx_u		$in3,$x30,$inp
+	 sub		$len,$len,$taillen
+	vxor		$twk3,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in3,$in3,$in3,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out3,$in3,$twk3
+	xxlor		32+$in4, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in4
+
+	 lvx_u		$in4,$x40,$inp
+	 subi		$len,$len,0x60
+	vxor		$twk4,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in4,$in4,$in4,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out4,$in4,$twk4
+	xxlor		32+$in5, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in5
+
+	 lvx_u		$in5,$x50,$inp
+	 addi		$inp,$inp,0x60
+	vxor		$twk5,$tweak,$rndkey0
+	vsrab		$tmp,$tweak,$seven	# next tweak value
+	vaddubm		$tweak,$tweak,$tweak
+	 le?vperm	$in5,$in5,$in5,$leperm
+	vand		$tmp,$tmp,$eighty7
+	 vxor		$out5,$in5,$twk5
+	xxlor		32+$in0, 0, 0
+	vpermxor	$tweak, $tweak, $tmp, $in0
+
+	vxor		v31,v31,$rndkey0
+	mtctr		$rounds
+	b		Loop_xts_dec6x
+
+.align	5
+Loop_xts_dec6x:
+	vncipher	$out0,$out0,v24
+	vncipher	$out1,$out1,v24
+	vncipher	$out2,$out2,v24
+	vncipher	$out3,$out3,v24
+	vncipher	$out4,$out4,v24
+	vncipher	$out5,$out5,v24
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vncipher	$out0,$out0,v25
+	vncipher	$out1,$out1,v25
+	vncipher	$out2,$out2,v25
+	vncipher	$out3,$out3,v25
+	vncipher	$out4,$out4,v25
+	vncipher	$out5,$out5,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		Loop_xts_dec6x
+
+	xxlor		32+$eighty7, 1, 1	# 0x010101..87
+
+	subic		$len,$len,96		# $len-=96
+	 vxor		$in0,$twk0,v31		# xor with last round key
+	vncipher	$out0,$out0,v24
+	vncipher	$out1,$out1,v24
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk0,$tweak,$rndkey0
+	 vaddubm	$tweak,$tweak,$tweak
+	vncipher	$out2,$out2,v24
+	vncipher	$out3,$out3,v24
+	vncipher	$out4,$out4,v24
+	vncipher	$out5,$out5,v24
+
+	subfe.		r0,r0,r0		# borrow?-1:0
+	 vand		$tmp,$tmp,$eighty7
+	vncipher	$out0,$out0,v25
+	vncipher	$out1,$out1,v25
+	 xxlor		32+$in1, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in1
+	vncipher	$out2,$out2,v25
+	vncipher	$out3,$out3,v25
+	 vxor		$in1,$twk1,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk1,$tweak,$rndkey0
+	vncipher	$out4,$out4,v25
+	vncipher	$out5,$out5,v25
+
+	and		r0,r0,$len
+	 vaddubm	$tweak,$tweak,$tweak
+	vncipher	$out0,$out0,v26
+	vncipher	$out1,$out1,v26
+	 vand		$tmp,$tmp,$eighty7
+	vncipher	$out2,$out2,v26
+	vncipher	$out3,$out3,v26
+	 xxlor		32+$in2, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in2
+	vncipher	$out4,$out4,v26
+	vncipher	$out5,$out5,v26
+
+	add		$inp,$inp,r0		# $inp is adjusted in such
+						# way that at exit from the
+						# loop inX-in5 are loaded
+						# with last "words"
+	 vxor		$in2,$twk2,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk2,$tweak,$rndkey0
+	 vaddubm	$tweak,$tweak,$tweak
+	vncipher	$out0,$out0,v27
+	vncipher	$out1,$out1,v27
+	vncipher	$out2,$out2,v27
+	vncipher	$out3,$out3,v27
+	 vand		$tmp,$tmp,$eighty7
+	vncipher	$out4,$out4,v27
+	vncipher	$out5,$out5,v27
+
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	 xxlor		32+$in3, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in3
+	vncipher	$out0,$out0,v28
+	vncipher	$out1,$out1,v28
+	 vxor		$in3,$twk3,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk3,$tweak,$rndkey0
+	vncipher	$out2,$out2,v28
+	vncipher	$out3,$out3,v28
+	 vaddubm	$tweak,$tweak,$tweak
+	vncipher	$out4,$out4,v28
+	vncipher	$out5,$out5,v28
+	lvx		v24,$x00,$key_		# re-pre-load round[1]
+	 vand		$tmp,$tmp,$eighty7
+
+	vncipher	$out0,$out0,v29
+	vncipher	$out1,$out1,v29
+	 xxlor		32+$in4, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in4
+	vncipher	$out2,$out2,v29
+	vncipher	$out3,$out3,v29
+	 vxor		$in4,$twk4,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk4,$tweak,$rndkey0
+	vncipher	$out4,$out4,v29
+	vncipher	$out5,$out5,v29
+	lvx		v25,$x10,$key_		# re-pre-load round[2]
+	 vaddubm	$tweak,$tweak,$tweak
+
+	vncipher	$out0,$out0,v30
+	vncipher	$out1,$out1,v30
+	 vand		$tmp,$tmp,$eighty7
+	vncipher	$out2,$out2,v30
+	vncipher	$out3,$out3,v30
+	 xxlor		32+$in5, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in5
+	vncipher	$out4,$out4,v30
+	vncipher	$out5,$out5,v30
+	 vxor		$in5,$twk5,v31
+	 vsrab		$tmp,$tweak,$seven	# next tweak value
+	 vxor		$twk5,$tweak,$rndkey0
+
+	vncipherlast	$out0,$out0,$in0
+	 lvx_u		$in0,$x00,$inp		# load next input block
+	 vaddubm	$tweak,$tweak,$tweak
+	vncipherlast	$out1,$out1,$in1
+	 lvx_u		$in1,$x10,$inp
+	vncipherlast	$out2,$out2,$in2
+	 le?vperm	$in0,$in0,$in0,$leperm
+	 lvx_u		$in2,$x20,$inp
+	 vand		$tmp,$tmp,$eighty7
+	vncipherlast	$out3,$out3,$in3
+	 le?vperm	$in1,$in1,$in1,$leperm
+	 lvx_u		$in3,$x30,$inp
+	vncipherlast	$out4,$out4,$in4
+	 le?vperm	$in2,$in2,$in2,$leperm
+	 lvx_u		$in4,$x40,$inp
+	 xxlor		10, 32+$in0, 32+$in0
+	 xxlor		32+$in0, 0, 0
+	 vpermxor	$tweak, $tweak, $tmp, $in0
+	 xxlor		32+$in0, 10, 10
+	vncipherlast	$out5,$out5,$in5
+	 le?vperm	$in3,$in3,$in3,$leperm
+	 lvx_u		$in5,$x50,$inp
+	 addi		$inp,$inp,0x60
+	 le?vperm	$in4,$in4,$in4,$leperm
+	 le?vperm	$in5,$in5,$in5,$leperm
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	 vxor		$out0,$in0,$twk0
+	le?vperm	$out2,$out2,$out2,$leperm
+	stvx_u		$out1,$x10,$out
+	 vxor		$out1,$in1,$twk1
+	le?vperm	$out3,$out3,$out3,$leperm
+	stvx_u		$out2,$x20,$out
+	 vxor		$out2,$in2,$twk2
+	le?vperm	$out4,$out4,$out4,$leperm
+	stvx_u		$out3,$x30,$out
+	 vxor		$out3,$in3,$twk3
+	le?vperm	$out5,$out5,$out5,$leperm
+	stvx_u		$out4,$x40,$out
+	 vxor		$out4,$in4,$twk4
+	stvx_u		$out5,$x50,$out
+	 vxor		$out5,$in5,$twk5
+	addi		$out,$out,0x60
+
+	mtctr		$rounds
+	beq		Loop_xts_dec6x		# did $len-=96 borrow?
+
+	xxlor		32+$eighty7, 2, 2	# 0x010101..87
+
+	addic.		$len,$len,0x60
+	beq		Lxts_dec6x_zero
+	cmpwi		$len,0x20
+	blt		Lxts_dec6x_one
+	nop
+	beq		Lxts_dec6x_two
+	cmpwi		$len,0x40
+	blt		Lxts_dec6x_three
+	nop
+	beq		Lxts_dec6x_four
+
+Lxts_dec6x_five:
+	vxor		$out0,$in1,$twk0
+	vxor		$out1,$in2,$twk1
+	vxor		$out2,$in3,$twk2
+	vxor		$out3,$in4,$twk3
+	vxor		$out4,$in5,$twk4
+
+	bl		_aesp8_xts_dec5x
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	vmr		$twk0,$twk5		# unused tweak
+	vxor		$twk1,$tweak,$rndkey0
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	vxor		$out0,$in0,$twk1
+	le?vperm	$out2,$out2,$out2,$leperm
+	stvx_u		$out1,$x10,$out
+	le?vperm	$out3,$out3,$out3,$leperm
+	stvx_u		$out2,$x20,$out
+	le?vperm	$out4,$out4,$out4,$leperm
+	stvx_u		$out3,$x30,$out
+	stvx_u		$out4,$x40,$out
+	addi		$out,$out,0x50
+	bne		Lxts_dec6x_steal
+	b		Lxts_dec6x_done
+
+.align	4
+Lxts_dec6x_four:
+	vxor		$out0,$in2,$twk0
+	vxor		$out1,$in3,$twk1
+	vxor		$out2,$in4,$twk2
+	vxor		$out3,$in5,$twk3
+	vxor		$out4,$out4,$out4
+
+	bl		_aesp8_xts_dec5x
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	vmr		$twk0,$twk4		# unused tweak
+	vmr		$twk1,$twk5
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	vxor		$out0,$in0,$twk5
+	le?vperm	$out2,$out2,$out2,$leperm
+	stvx_u		$out1,$x10,$out
+	le?vperm	$out3,$out3,$out3,$leperm
+	stvx_u		$out2,$x20,$out
+	stvx_u		$out3,$x30,$out
+	addi		$out,$out,0x40
+	bne		Lxts_dec6x_steal
+	b		Lxts_dec6x_done
+
+.align	4
+Lxts_dec6x_three:
+	vxor		$out0,$in3,$twk0
+	vxor		$out1,$in4,$twk1
+	vxor		$out2,$in5,$twk2
+	vxor		$out3,$out3,$out3
+	vxor		$out4,$out4,$out4
+
+	bl		_aesp8_xts_dec5x
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	vmr		$twk0,$twk3		# unused tweak
+	vmr		$twk1,$twk4
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	vxor		$out0,$in0,$twk4
+	le?vperm	$out2,$out2,$out2,$leperm
+	stvx_u		$out1,$x10,$out
+	stvx_u		$out2,$x20,$out
+	addi		$out,$out,0x30
+	bne		Lxts_dec6x_steal
+	b		Lxts_dec6x_done
+
+.align	4
+Lxts_dec6x_two:
+	vxor		$out0,$in4,$twk0
+	vxor		$out1,$in5,$twk1
+	vxor		$out2,$out2,$out2
+	vxor		$out3,$out3,$out3
+	vxor		$out4,$out4,$out4
+
+	bl		_aesp8_xts_dec5x
+
+	le?vperm	$out0,$out0,$out0,$leperm
+	vmr		$twk0,$twk2		# unused tweak
+	vmr		$twk1,$twk3
+	le?vperm	$out1,$out1,$out1,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	vxor		$out0,$in0,$twk3
+	stvx_u		$out1,$x10,$out
+	addi		$out,$out,0x20
+	bne		Lxts_dec6x_steal
+	b		Lxts_dec6x_done
+
+.align	4
+Lxts_dec6x_one:
+	vxor		$out0,$in5,$twk0
+	nop
+Loop_xts_dec1x:
+	vncipher	$out0,$out0,v24
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vncipher	$out0,$out0,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		Loop_xts_dec1x
+
+	subi		r0,$taillen,1
+	vncipher	$out0,$out0,v24
+
+	andi.		r0,r0,16
+	cmpwi		$taillen,0
+	vncipher	$out0,$out0,v25
+
+	sub		$inp,$inp,r0
+	vncipher	$out0,$out0,v26
+
+	lvx_u		$in0,0,$inp
+	vncipher	$out0,$out0,v27
+
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	vncipher	$out0,$out0,v28
+	lvx		v24,$x00,$key_		# re-pre-load round[1]
+
+	vncipher	$out0,$out0,v29
+	lvx		v25,$x10,$key_		# re-pre-load round[2]
+	 vxor		$twk0,$twk0,v31
+
+	le?vperm	$in0,$in0,$in0,$leperm
+	vncipher	$out0,$out0,v30
+
+	mtctr		$rounds
+	vncipherlast	$out0,$out0,$twk0
+
+	vmr		$twk0,$twk1		# unused tweak
+	vmr		$twk1,$twk2
+	le?vperm	$out0,$out0,$out0,$leperm
+	stvx_u		$out0,$x00,$out		# store output
+	addi		$out,$out,0x10
+	vxor		$out0,$in0,$twk2
+	bne		Lxts_dec6x_steal
+	b		Lxts_dec6x_done
+
+.align	4
+Lxts_dec6x_zero:
+	cmpwi		$taillen,0
+	beq		Lxts_dec6x_done
+
+	lvx_u		$in0,0,$inp
+	le?vperm	$in0,$in0,$in0,$leperm
+	vxor		$out0,$in0,$twk1
+Lxts_dec6x_steal:
+	vncipher	$out0,$out0,v24
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vncipher	$out0,$out0,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		Lxts_dec6x_steal
+
+	add		$inp,$inp,$taillen
+	vncipher	$out0,$out0,v24
+
+	cmpwi		$taillen,0
+	vncipher	$out0,$out0,v25
+
+	lvx_u		$in0,0,$inp
+	vncipher	$out0,$out0,v26
+
+	lvsr		$inpperm,0,$taillen	# $in5 is no more
+	vncipher	$out0,$out0,v27
+
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	vncipher	$out0,$out0,v28
+	lvx		v24,$x00,$key_		# re-pre-load round[1]
+
+	vncipher	$out0,$out0,v29
+	lvx		v25,$x10,$key_		# re-pre-load round[2]
+	 vxor		$twk1,$twk1,v31
+
+	le?vperm	$in0,$in0,$in0,$leperm
+	vncipher	$out0,$out0,v30
+
+	vperm		$in0,$in0,$in0,$inpperm
+	vncipherlast	$tmp,$out0,$twk1
+
+	le?vperm	$out0,$tmp,$tmp,$leperm
+	le?stvx_u	$out0,0,$out
+	be?stvx_u	$tmp,0,$out
+
+	vxor		$out0,$out0,$out0
+	vspltisb	$out1,-1
+	vperm		$out0,$out0,$out1,$inpperm
+	vsel		$out0,$in0,$tmp,$out0
+	vxor		$out0,$out0,$twk0
+
+	subi		r30,$out,1
+	mtctr		$taillen
+Loop_xts_dec6x_steal:
+	lbzu		r0,1(r30)
+	stb		r0,16(r30)
+	bdnz		Loop_xts_dec6x_steal
+
+	li		$taillen,0
+	mtctr		$rounds
+	b		Loop_xts_dec1x		# one more time...
+
+.align	4
+Lxts_dec6x_done:
+	${UCMP}i	$ivp,0
+	beq		Lxts_dec6x_ret
+
+	vxor		$tweak,$twk0,$rndkey0
+	le?vperm	$tweak,$tweak,$tweak,$leperm
+	stvx_u		$tweak,0,$ivp
+
+Lxts_dec6x_ret:
+	mtlr		r11
+	li		r10,`$FRAME+15`
+	li		r11,`$FRAME+31`
+	stvx		$seven,r10,$sp		# wipe copies of round keys
+	addi		r10,r10,32
+	stvx		$seven,r11,$sp
+	addi		r11,r11,32
+	stvx		$seven,r10,$sp
+	addi		r10,r10,32
+	stvx		$seven,r11,$sp
+	addi		r11,r11,32
+	stvx		$seven,r10,$sp
+	addi		r10,r10,32
+	stvx		$seven,r11,$sp
+	addi		r11,r11,32
+	stvx		$seven,r10,$sp
+	addi		r10,r10,32
+	stvx		$seven,r11,$sp
+	addi		r11,r11,32
+
+	mtspr		256,$vrsave
+	lvx		v20,r10,$sp		# ABI says so
+	addi		r10,r10,32
+	lvx		v21,r11,$sp
+	addi		r11,r11,32
+	lvx		v22,r10,$sp
+	addi		r10,r10,32
+	lvx		v23,r11,$sp
+	addi		r11,r11,32
+	lvx		v24,r10,$sp
+	addi		r10,r10,32
+	lvx		v25,r11,$sp
+	addi		r11,r11,32
+	lvx		v26,r10,$sp
+	addi		r10,r10,32
+	lvx		v27,r11,$sp
+	addi		r11,r11,32
+	lvx		v28,r10,$sp
+	addi		r10,r10,32
+	lvx		v29,r11,$sp
+	addi		r11,r11,32
+	lvx		v30,r10,$sp
+	lvx		v31,r11,$sp
+	$POP		r26,`$FRAME+21*16+0*$SIZE_T`($sp)
+	$POP		r27,`$FRAME+21*16+1*$SIZE_T`($sp)
+	$POP		r28,`$FRAME+21*16+2*$SIZE_T`($sp)
+	$POP		r29,`$FRAME+21*16+3*$SIZE_T`($sp)
+	$POP		r30,`$FRAME+21*16+4*$SIZE_T`($sp)
+	$POP		r31,`$FRAME+21*16+5*$SIZE_T`($sp)
+	addi		$sp,$sp,`$FRAME+21*16+6*$SIZE_T`
+	blr
+	.long		0
+	.byte		0,12,0x04,1,0x80,6,6,0
+	.long		0
+
+.align	5
+_aesp8_xts_dec5x:
+	vncipher	$out0,$out0,v24
+	vncipher	$out1,$out1,v24
+	vncipher	$out2,$out2,v24
+	vncipher	$out3,$out3,v24
+	vncipher	$out4,$out4,v24
+	lvx		v24,$x20,$key_		# round[3]
+	addi		$key_,$key_,0x20
+
+	vncipher	$out0,$out0,v25
+	vncipher	$out1,$out1,v25
+	vncipher	$out2,$out2,v25
+	vncipher	$out3,$out3,v25
+	vncipher	$out4,$out4,v25
+	lvx		v25,$x10,$key_		# round[4]
+	bdnz		_aesp8_xts_dec5x
+
+	subi		r0,$taillen,1
+	vncipher	$out0,$out0,v24
+	vncipher	$out1,$out1,v24
+	vncipher	$out2,$out2,v24
+	vncipher	$out3,$out3,v24
+	vncipher	$out4,$out4,v24
+
+	andi.		r0,r0,16
+	cmpwi		$taillen,0
+	vncipher	$out0,$out0,v25
+	vncipher	$out1,$out1,v25
+	vncipher	$out2,$out2,v25
+	vncipher	$out3,$out3,v25
+	vncipher	$out4,$out4,v25
+	 vxor		$twk0,$twk0,v31
+
+	sub		$inp,$inp,r0
+	vncipher	$out0,$out0,v26
+	vncipher	$out1,$out1,v26
+	vncipher	$out2,$out2,v26
+	vncipher	$out3,$out3,v26
+	vncipher	$out4,$out4,v26
+	 vxor		$in1,$twk1,v31
+
+	vncipher	$out0,$out0,v27
+	lvx_u		$in0,0,$inp
+	vncipher	$out1,$out1,v27
+	vncipher	$out2,$out2,v27
+	vncipher	$out3,$out3,v27
+	vncipher	$out4,$out4,v27
+	 vxor		$in2,$twk2,v31
+
+	addi		$key_,$sp,$FRAME+15	# rewind $key_
+	vncipher	$out0,$out0,v28
+	vncipher	$out1,$out1,v28
+	vncipher	$out2,$out2,v28
+	vncipher	$out3,$out3,v28
+	vncipher	$out4,$out4,v28
+	lvx		v24,$x00,$key_		# re-pre-load round[1]
+	 vxor		$in3,$twk3,v31
+
+	vncipher	$out0,$out0,v29
+	le?vperm	$in0,$in0,$in0,$leperm
+	vncipher	$out1,$out1,v29
+	vncipher	$out2,$out2,v29
+	vncipher	$out3,$out3,v29
+	vncipher	$out4,$out4,v29
+	lvx		v25,$x10,$key_		# re-pre-load round[2]
+	 vxor		$in4,$twk4,v31
+
+	vncipher	$out0,$out0,v30
+	vncipher	$out1,$out1,v30
+	vncipher	$out2,$out2,v30
+	vncipher	$out3,$out3,v30
+	vncipher	$out4,$out4,v30
+
+	vncipherlast	$out0,$out0,$twk0
+	vncipherlast	$out1,$out1,$in1
+	vncipherlast	$out2,$out2,$in2
+	vncipherlast	$out3,$out3,$in3
+	vncipherlast	$out4,$out4,$in4
+	mtctr		$rounds
+	blr
+        .long   	0
+        .byte   	0,12,0x14,0,0,0,0,0
+___
+}}	}}}
+
+my $consts=1;
+foreach(split("\n",$code)) {
+        s/\`([^\`]*)\`/eval($1)/geo;
+
+	# constants table endian-specific conversion
+	if ($consts && m/\.(long|byte)\s+(.+)\s+(\?[a-z]*)$/o) {
+	    my $conv=$3;
+	    my @bytes=();
+
+	    # convert to endian-agnostic format
+	    if ($1 eq "long") {
+	      foreach (split(/,\s*/,$2)) {
+		my $l = /^0/?oct:int;
+		push @bytes,($l>>24)&0xff,($l>>16)&0xff,($l>>8)&0xff,$l&0xff;
+	      }
+	    } else {
+		@bytes = map(/^0/?oct:int,split(/,\s*/,$2));
+	    }
+
+	    # little-endian conversion
+	    if ($flavour =~ /le$/o) {
+		SWITCH: for($conv)  {
+		    /\?inv/ && do   { @bytes=map($_^0xf,@bytes); last; };
+		    /\?rev/ && do   { @bytes=reverse(@bytes);    last; };
+		}
+	    }
+
+	    #emit
+	    print ".byte\t",join(',',map (sprintf("0x%02x",$_),@bytes)),"\n";
+	    next;
+	}
+	$consts=0 if (m/Lconsts:/o);	# end of table
+
+	# instructions prefixed with '?' are endian-specific and need
+	# to be adjusted accordingly...
+	if ($flavour =~ /le$/o) {	# little-endian
+	    s/le\?//o		or
+	    s/be\?/#be#/o	or
+	    s/\?lvsr/lvsl/o	or
+	    s/\?lvsl/lvsr/o	or
+	    s/\?(vperm\s+v[0-9]+,\s*)(v[0-9]+,\s*)(v[0-9]+,\s*)(v[0-9]+)/$1$3$2$4/o or
+	    s/\?(vsldoi\s+v[0-9]+,\s*)(v[0-9]+,)\s*(v[0-9]+,\s*)([0-9]+)/$1$3$2 16-$4/o or
+	    s/\?(vspltw\s+v[0-9]+,\s*)(v[0-9]+,)\s*([0-9])/$1$2 3-$3/o;
+	} else {			# big-endian
+	    s/le\?/#le#/o	or
+	    s/be\?//o		or
+	    s/\?([a-z]+)/$1/o;
+	}
+
+        print $_,"\n";
+}
+
+close STDOUT;
diff --git a/lib/crypto/powerpc/chacha-p10le-8x.S b/lib/crypto/powerpc/chacha-p10le-8x.S
new file mode 100644
index 000000000..b29562bd5
--- /dev/null
+++ b/lib/crypto/powerpc/chacha-p10le-8x.S
@@ -0,0 +1,840 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+#
+# Accelerated chacha20 implementation for ppc64le.
+#
+# Copyright 2023- IBM Corp. All rights reserved
+#
+#===================================================================================
+# Written by Danny Tsen <dtsen@us.ibm.com>
+#
+# do rounds,  8 quarter rounds
+# 1.  a += b; d ^= a; d <<<= 16;
+# 2.  c += d; b ^= c; b <<<= 12;
+# 3.  a += b; d ^= a; d <<<= 8;
+# 4.  c += d; b ^= c; b <<<= 7
+#
+# row1 = (row1 + row2),  row4 = row1 xor row4,  row4 rotate each word by 16
+# row3 = (row3 + row4),  row2 = row3 xor row2,  row2 rotate each word by 12
+# row1 = (row1 + row2), row4 = row1 xor row4,  row4 rotate each word by 8
+# row3 = (row3 + row4), row2 = row3 xor row2,  row2 rotate each word by 7
+#
+# 4 blocks (a b c d)
+#
+# a0 b0 c0 d0
+# a1 b1 c1 d1
+# ...
+# a4 b4 c4 d4
+# ...
+# a8 b8 c8 d8
+# ...
+# a12 b12 c12 d12
+# a13 ...
+# a14 ...
+# a15 b15 c15 d15
+#
+# Column round (v0, v4,  v8, v12, v1, v5,  v9, v13, v2, v6, v10, v14, v3, v7, v11, v15)
+# Diagnal round (v0, v5, v10, v15, v1, v6, v11, v12, v2, v7,  v8, v13, v3, v4,  v9, v14)
+#
+
+#include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/asm-compat.h>
+#include <linux/linkage.h>
+
+.machine	"any"
+.text
+
+.macro	SAVE_GPR GPR OFFSET FRAME
+	std	\GPR,\OFFSET(\FRAME)
+.endm
+
+.macro	SAVE_VRS VRS OFFSET FRAME
+	li	16, \OFFSET
+	stvx	\VRS, 16, \FRAME
+.endm
+
+.macro	SAVE_VSX VSX OFFSET FRAME
+	li	16, \OFFSET
+	stxvx	\VSX, 16, \FRAME
+.endm
+
+.macro	RESTORE_GPR GPR OFFSET FRAME
+	ld	\GPR,\OFFSET(\FRAME)
+.endm
+
+.macro	RESTORE_VRS VRS OFFSET FRAME
+	li	16, \OFFSET
+	lvx	\VRS, 16, \FRAME
+.endm
+
+.macro	RESTORE_VSX VSX OFFSET FRAME
+	li	16, \OFFSET
+	lxvx	\VSX, 16, \FRAME
+.endm
+
+.macro SAVE_REGS
+	mflr 0
+	std 0, 16(1)
+	stdu 1,-752(1)
+
+	SAVE_GPR 14, 112, 1
+	SAVE_GPR 15, 120, 1
+	SAVE_GPR 16, 128, 1
+	SAVE_GPR 17, 136, 1
+	SAVE_GPR 18, 144, 1
+	SAVE_GPR 19, 152, 1
+	SAVE_GPR 20, 160, 1
+	SAVE_GPR 21, 168, 1
+	SAVE_GPR 22, 176, 1
+	SAVE_GPR 23, 184, 1
+	SAVE_GPR 24, 192, 1
+	SAVE_GPR 25, 200, 1
+	SAVE_GPR 26, 208, 1
+	SAVE_GPR 27, 216, 1
+	SAVE_GPR 28, 224, 1
+	SAVE_GPR 29, 232, 1
+	SAVE_GPR 30, 240, 1
+	SAVE_GPR 31, 248, 1
+
+	addi	9, 1, 256
+	SAVE_VRS 20, 0, 9
+	SAVE_VRS 21, 16, 9
+	SAVE_VRS 22, 32, 9
+	SAVE_VRS 23, 48, 9
+	SAVE_VRS 24, 64, 9
+	SAVE_VRS 25, 80, 9
+	SAVE_VRS 26, 96, 9
+	SAVE_VRS 27, 112, 9
+	SAVE_VRS 28, 128, 9
+	SAVE_VRS 29, 144, 9
+	SAVE_VRS 30, 160, 9
+	SAVE_VRS 31, 176, 9
+
+	SAVE_VSX 14, 192, 9
+	SAVE_VSX 15, 208, 9
+	SAVE_VSX 16, 224, 9
+	SAVE_VSX 17, 240, 9
+	SAVE_VSX 18, 256, 9
+	SAVE_VSX 19, 272, 9
+	SAVE_VSX 20, 288, 9
+	SAVE_VSX 21, 304, 9
+	SAVE_VSX 22, 320, 9
+	SAVE_VSX 23, 336, 9
+	SAVE_VSX 24, 352, 9
+	SAVE_VSX 25, 368, 9
+	SAVE_VSX 26, 384, 9
+	SAVE_VSX 27, 400, 9
+	SAVE_VSX 28, 416, 9
+	SAVE_VSX 29, 432, 9
+	SAVE_VSX 30, 448, 9
+	SAVE_VSX 31, 464, 9
+.endm # SAVE_REGS
+
+.macro RESTORE_REGS
+	addi	9, 1, 256
+	RESTORE_VRS 20, 0, 9
+	RESTORE_VRS 21, 16, 9
+	RESTORE_VRS 22, 32, 9
+	RESTORE_VRS 23, 48, 9
+	RESTORE_VRS 24, 64, 9
+	RESTORE_VRS 25, 80, 9
+	RESTORE_VRS 26, 96, 9
+	RESTORE_VRS 27, 112, 9
+	RESTORE_VRS 28, 128, 9
+	RESTORE_VRS 29, 144, 9
+	RESTORE_VRS 30, 160, 9
+	RESTORE_VRS 31, 176, 9
+
+	RESTORE_VSX 14, 192, 9
+	RESTORE_VSX 15, 208, 9
+	RESTORE_VSX 16, 224, 9
+	RESTORE_VSX 17, 240, 9
+	RESTORE_VSX 18, 256, 9
+	RESTORE_VSX 19, 272, 9
+	RESTORE_VSX 20, 288, 9
+	RESTORE_VSX 21, 304, 9
+	RESTORE_VSX 22, 320, 9
+	RESTORE_VSX 23, 336, 9
+	RESTORE_VSX 24, 352, 9
+	RESTORE_VSX 25, 368, 9
+	RESTORE_VSX 26, 384, 9
+	RESTORE_VSX 27, 400, 9
+	RESTORE_VSX 28, 416, 9
+	RESTORE_VSX 29, 432, 9
+	RESTORE_VSX 30, 448, 9
+	RESTORE_VSX 31, 464, 9
+
+	RESTORE_GPR 14, 112, 1
+	RESTORE_GPR 15, 120, 1
+	RESTORE_GPR 16, 128, 1
+	RESTORE_GPR 17, 136, 1
+	RESTORE_GPR 18, 144, 1
+	RESTORE_GPR 19, 152, 1
+	RESTORE_GPR 20, 160, 1
+	RESTORE_GPR 21, 168, 1
+	RESTORE_GPR 22, 176, 1
+	RESTORE_GPR 23, 184, 1
+	RESTORE_GPR 24, 192, 1
+	RESTORE_GPR 25, 200, 1
+	RESTORE_GPR 26, 208, 1
+	RESTORE_GPR 27, 216, 1
+	RESTORE_GPR 28, 224, 1
+	RESTORE_GPR 29, 232, 1
+	RESTORE_GPR 30, 240, 1
+	RESTORE_GPR 31, 248, 1
+
+	addi    1, 1, 752
+	ld 0, 16(1)
+	mtlr 0
+.endm # RESTORE_REGS
+
+.macro QT_loop_8x
+	# QR(v0, v4,  v8, v12, v1, v5,  v9, v13, v2, v6, v10, v14, v3, v7, v11, v15)
+	xxlor	0, 32+25, 32+25
+	xxlor	32+25, 20, 20
+	vadduwm 0, 0, 4
+	vadduwm 1, 1, 5
+	vadduwm 2, 2, 6
+	vadduwm 3, 3, 7
+	  vadduwm 16, 16, 20
+	  vadduwm 17, 17, 21
+	  vadduwm 18, 18, 22
+	  vadduwm 19, 19, 23
+
+	  vpermxor 12, 12, 0, 25
+	  vpermxor 13, 13, 1, 25
+	  vpermxor 14, 14, 2, 25
+	  vpermxor 15, 15, 3, 25
+	  vpermxor 28, 28, 16, 25
+	  vpermxor 29, 29, 17, 25
+	  vpermxor 30, 30, 18, 25
+	  vpermxor 31, 31, 19, 25
+	xxlor	32+25, 0, 0
+	vadduwm 8, 8, 12
+	vadduwm 9, 9, 13
+	vadduwm 10, 10, 14
+	vadduwm 11, 11, 15
+	  vadduwm 24, 24, 28
+	  vadduwm 25, 25, 29
+	  vadduwm 26, 26, 30
+	  vadduwm 27, 27, 31
+	vxor 4, 4, 8
+	vxor 5, 5, 9
+	vxor 6, 6, 10
+	vxor 7, 7, 11
+	  vxor 20, 20, 24
+	  vxor 21, 21, 25
+	  vxor 22, 22, 26
+	  vxor 23, 23, 27
+
+	xxlor	0, 32+25, 32+25
+	xxlor	32+25, 21, 21
+	vrlw 4, 4, 25  #
+	vrlw 5, 5, 25
+	vrlw 6, 6, 25
+	vrlw 7, 7, 25
+	  vrlw 20, 20, 25  #
+	  vrlw 21, 21, 25
+	  vrlw 22, 22, 25
+	  vrlw 23, 23, 25
+	xxlor	32+25, 0, 0
+	vadduwm 0, 0, 4
+	vadduwm 1, 1, 5
+	vadduwm 2, 2, 6
+	vadduwm 3, 3, 7
+	  vadduwm 16, 16, 20
+	  vadduwm 17, 17, 21
+	  vadduwm 18, 18, 22
+	  vadduwm 19, 19, 23
+
+	xxlor	0, 32+25, 32+25
+	xxlor	32+25, 22, 22
+	  vpermxor 12, 12, 0, 25
+	  vpermxor 13, 13, 1, 25
+	  vpermxor 14, 14, 2, 25
+	  vpermxor 15, 15, 3, 25
+	  vpermxor 28, 28, 16, 25
+	  vpermxor 29, 29, 17, 25
+	  vpermxor 30, 30, 18, 25
+	  vpermxor 31, 31, 19, 25
+	xxlor	32+25, 0, 0
+	vadduwm 8, 8, 12
+	vadduwm 9, 9, 13
+	vadduwm 10, 10, 14
+	vadduwm 11, 11, 15
+	  vadduwm 24, 24, 28
+	  vadduwm 25, 25, 29
+	  vadduwm 26, 26, 30
+	  vadduwm 27, 27, 31
+	xxlor	0, 32+28, 32+28
+	xxlor	32+28, 23, 23
+	vxor 4, 4, 8
+	vxor 5, 5, 9
+	vxor 6, 6, 10
+	vxor 7, 7, 11
+	  vxor 20, 20, 24
+	  vxor 21, 21, 25
+	  vxor 22, 22, 26
+	  vxor 23, 23, 27
+	vrlw 4, 4, 28  #
+	vrlw 5, 5, 28
+	vrlw 6, 6, 28
+	vrlw 7, 7, 28
+	  vrlw 20, 20, 28  #
+	  vrlw 21, 21, 28
+	  vrlw 22, 22, 28
+	  vrlw 23, 23, 28
+	xxlor	32+28, 0, 0
+
+	# QR(v0, v5, v10, v15, v1, v6, v11, v12, v2, v7,  v8, v13, v3, v4,  v9, v14)
+	xxlor	0, 32+25, 32+25
+	xxlor	32+25, 20, 20
+	vadduwm 0, 0, 5
+	vadduwm 1, 1, 6
+	vadduwm 2, 2, 7
+	vadduwm 3, 3, 4
+	  vadduwm 16, 16, 21
+	  vadduwm 17, 17, 22
+	  vadduwm 18, 18, 23
+	  vadduwm 19, 19, 20
+
+	  vpermxor 15, 15, 0, 25
+	  vpermxor 12, 12, 1, 25
+	  vpermxor 13, 13, 2, 25
+	  vpermxor 14, 14, 3, 25
+	  vpermxor 31, 31, 16, 25
+	  vpermxor 28, 28, 17, 25
+	  vpermxor 29, 29, 18, 25
+	  vpermxor 30, 30, 19, 25
+
+	xxlor	32+25, 0, 0
+	vadduwm 10, 10, 15
+	vadduwm 11, 11, 12
+	vadduwm 8, 8, 13
+	vadduwm 9, 9, 14
+	  vadduwm 26, 26, 31
+	  vadduwm 27, 27, 28
+	  vadduwm 24, 24, 29
+	  vadduwm 25, 25, 30
+	vxor 5, 5, 10
+	vxor 6, 6, 11
+	vxor 7, 7, 8
+	vxor 4, 4, 9
+	  vxor 21, 21, 26
+	  vxor 22, 22, 27
+	  vxor 23, 23, 24
+	  vxor 20, 20, 25
+
+	xxlor	0, 32+25, 32+25
+	xxlor	32+25, 21, 21
+	vrlw 5, 5, 25
+	vrlw 6, 6, 25
+	vrlw 7, 7, 25
+	vrlw 4, 4, 25
+	  vrlw 21, 21, 25
+	  vrlw 22, 22, 25
+	  vrlw 23, 23, 25
+	  vrlw 20, 20, 25
+	xxlor	32+25, 0, 0
+
+	vadduwm 0, 0, 5
+	vadduwm 1, 1, 6
+	vadduwm 2, 2, 7
+	vadduwm 3, 3, 4
+	  vadduwm 16, 16, 21
+	  vadduwm 17, 17, 22
+	  vadduwm 18, 18, 23
+	  vadduwm 19, 19, 20
+
+	xxlor	0, 32+25, 32+25
+	xxlor	32+25, 22, 22
+	  vpermxor 15, 15, 0, 25
+	  vpermxor 12, 12, 1, 25
+	  vpermxor 13, 13, 2, 25
+	  vpermxor 14, 14, 3, 25
+	  vpermxor 31, 31, 16, 25
+	  vpermxor 28, 28, 17, 25
+	  vpermxor 29, 29, 18, 25
+	  vpermxor 30, 30, 19, 25
+	xxlor	32+25, 0, 0
+
+	vadduwm 10, 10, 15
+	vadduwm 11, 11, 12
+	vadduwm 8, 8, 13
+	vadduwm 9, 9, 14
+	  vadduwm 26, 26, 31
+	  vadduwm 27, 27, 28
+	  vadduwm 24, 24, 29
+	  vadduwm 25, 25, 30
+
+	xxlor	0, 32+28, 32+28
+	xxlor	32+28, 23, 23
+	vxor 5, 5, 10
+	vxor 6, 6, 11
+	vxor 7, 7, 8
+	vxor 4, 4, 9
+	  vxor 21, 21, 26
+	  vxor 22, 22, 27
+	  vxor 23, 23, 24
+	  vxor 20, 20, 25
+	vrlw 5, 5, 28
+	vrlw 6, 6, 28
+	vrlw 7, 7, 28
+	vrlw 4, 4, 28
+	  vrlw 21, 21, 28
+	  vrlw 22, 22, 28
+	  vrlw 23, 23, 28
+	  vrlw 20, 20, 28
+	xxlor	32+28, 0, 0
+.endm
+
+.macro QT_loop_4x
+	# QR(v0, v4,  v8, v12, v1, v5,  v9, v13, v2, v6, v10, v14, v3, v7, v11, v15)
+	vadduwm 0, 0, 4
+	vadduwm 1, 1, 5
+	vadduwm 2, 2, 6
+	vadduwm 3, 3, 7
+	  vpermxor 12, 12, 0, 20
+	  vpermxor 13, 13, 1, 20
+	  vpermxor 14, 14, 2, 20
+	  vpermxor 15, 15, 3, 20
+	vadduwm 8, 8, 12
+	vadduwm 9, 9, 13
+	vadduwm 10, 10, 14
+	vadduwm 11, 11, 15
+	vxor 4, 4, 8
+	vxor 5, 5, 9
+	vxor 6, 6, 10
+	vxor 7, 7, 11
+	vrlw 4, 4, 21
+	vrlw 5, 5, 21
+	vrlw 6, 6, 21
+	vrlw 7, 7, 21
+	vadduwm 0, 0, 4
+	vadduwm 1, 1, 5
+	vadduwm 2, 2, 6
+	vadduwm 3, 3, 7
+	  vpermxor 12, 12, 0, 22
+	  vpermxor 13, 13, 1, 22
+	  vpermxor 14, 14, 2, 22
+	  vpermxor 15, 15, 3, 22
+	vadduwm 8, 8, 12
+	vadduwm 9, 9, 13
+	vadduwm 10, 10, 14
+	vadduwm 11, 11, 15
+	vxor 4, 4, 8
+	vxor 5, 5, 9
+	vxor 6, 6, 10
+	vxor 7, 7, 11
+	vrlw 4, 4, 23
+	vrlw 5, 5, 23
+	vrlw 6, 6, 23
+	vrlw 7, 7, 23
+
+	# QR(v0, v5, v10, v15, v1, v6, v11, v12, v2, v7,  v8, v13, v3, v4,  v9, v14)
+	vadduwm 0, 0, 5
+	vadduwm 1, 1, 6
+	vadduwm 2, 2, 7
+	vadduwm 3, 3, 4
+	  vpermxor 15, 15, 0, 20
+	  vpermxor 12, 12, 1, 20
+	  vpermxor 13, 13, 2, 20
+	  vpermxor 14, 14, 3, 20
+	vadduwm 10, 10, 15
+	vadduwm 11, 11, 12
+	vadduwm 8, 8, 13
+	vadduwm 9, 9, 14
+	vxor 5, 5, 10
+	vxor 6, 6, 11
+	vxor 7, 7, 8
+	vxor 4, 4, 9
+	vrlw 5, 5, 21
+	vrlw 6, 6, 21
+	vrlw 7, 7, 21
+	vrlw 4, 4, 21
+	vadduwm 0, 0, 5
+	vadduwm 1, 1, 6
+	vadduwm 2, 2, 7
+	vadduwm 3, 3, 4
+	  vpermxor 15, 15, 0, 22
+	  vpermxor 12, 12, 1, 22
+	  vpermxor 13, 13, 2, 22
+	  vpermxor 14, 14, 3, 22
+	vadduwm 10, 10, 15
+	vadduwm 11, 11, 12
+	vadduwm 8, 8, 13
+	vadduwm 9, 9, 14
+	vxor 5, 5, 10
+	vxor 6, 6, 11
+	vxor 7, 7, 8
+	vxor 4, 4, 9
+	vrlw 5, 5, 23
+	vrlw 6, 6, 23
+	vrlw 7, 7, 23
+	vrlw 4, 4, 23
+.endm
+
+# Transpose
+.macro TP_4x a0 a1 a2 a3
+	xxmrghw  10, 32+\a0, 32+\a1	# a0, a1, b0, b1
+	xxmrghw  11, 32+\a2, 32+\a3	# a2, a3, b2, b3
+	xxmrglw  12, 32+\a0, 32+\a1	# c0, c1, d0, d1
+	xxmrglw  13, 32+\a2, 32+\a3	# c2, c3, d2, d3
+	xxpermdi	32+\a0, 10, 11, 0	# a0, a1, a2, a3
+	xxpermdi	32+\a1, 10, 11, 3	# b0, b1, b2, b3
+	xxpermdi	32+\a2, 12, 13, 0	# c0, c1, c2, c3
+	xxpermdi	32+\a3, 12, 13, 3	# d0, d1, d2, d3
+.endm
+
+# key stream = working state + state
+.macro Add_state S
+	vadduwm \S+0, \S+0, 16-\S
+	vadduwm \S+4, \S+4, 17-\S
+	vadduwm \S+8, \S+8, 18-\S
+	vadduwm \S+12, \S+12, 19-\S
+
+	vadduwm \S+1, \S+1, 16-\S
+	vadduwm \S+5, \S+5, 17-\S
+	vadduwm \S+9, \S+9, 18-\S
+	vadduwm \S+13, \S+13, 19-\S
+
+	vadduwm \S+2, \S+2, 16-\S
+	vadduwm \S+6, \S+6, 17-\S
+	vadduwm \S+10, \S+10, 18-\S
+	vadduwm \S+14, \S+14, 19-\S
+
+	vadduwm	\S+3, \S+3, 16-\S
+	vadduwm	\S+7, \S+7, 17-\S
+	vadduwm	\S+11, \S+11, 18-\S
+	vadduwm	\S+15, \S+15, 19-\S
+.endm
+
+#
+# write 256 bytes
+#
+.macro Write_256 S
+	add 9, 14, 5
+	add 16, 14, 4
+	lxvw4x 0, 0, 9
+	lxvw4x 1, 17, 9
+	lxvw4x 2, 18, 9
+	lxvw4x 3, 19, 9
+	lxvw4x 4, 20, 9
+	lxvw4x 5, 21, 9
+	lxvw4x 6, 22, 9
+	lxvw4x 7, 23, 9
+	lxvw4x 8, 24, 9
+	lxvw4x 9, 25, 9
+	lxvw4x 10, 26, 9
+	lxvw4x 11, 27, 9
+	lxvw4x 12, 28, 9
+	lxvw4x 13, 29, 9
+	lxvw4x 14, 30, 9
+	lxvw4x 15, 31, 9
+
+	xxlxor \S+32, \S+32, 0
+	xxlxor \S+36, \S+36, 1
+	xxlxor \S+40, \S+40, 2
+	xxlxor \S+44, \S+44, 3
+	xxlxor \S+33, \S+33, 4
+	xxlxor \S+37, \S+37, 5
+	xxlxor \S+41, \S+41, 6
+	xxlxor \S+45, \S+45, 7
+	xxlxor \S+34, \S+34, 8
+	xxlxor \S+38, \S+38, 9
+	xxlxor \S+42, \S+42, 10
+	xxlxor \S+46, \S+46, 11
+	xxlxor \S+35, \S+35, 12
+	xxlxor \S+39, \S+39, 13
+	xxlxor \S+43, \S+43, 14
+	xxlxor \S+47, \S+47, 15
+
+	stxvw4x \S+32, 0, 16
+	stxvw4x \S+36, 17, 16
+	stxvw4x \S+40, 18, 16
+	stxvw4x \S+44, 19, 16
+
+	stxvw4x \S+33, 20, 16
+	stxvw4x \S+37, 21, 16
+	stxvw4x \S+41, 22, 16
+	stxvw4x \S+45, 23, 16
+
+	stxvw4x \S+34, 24, 16
+	stxvw4x \S+38, 25, 16
+	stxvw4x \S+42, 26, 16
+	stxvw4x \S+46, 27, 16
+
+	stxvw4x \S+35, 28, 16
+	stxvw4x \S+39, 29, 16
+	stxvw4x \S+43, 30, 16
+	stxvw4x \S+47, 31, 16
+
+.endm
+
+#
+# void chacha_p10le_8x(const struct chacha_state *state, u8 *dst, const u8 *src,
+#		       unsigned int len, int nrounds);
+#
+SYM_FUNC_START(chacha_p10le_8x)
+.align 5
+	cmpdi	6, 0
+	ble	Out_no_chacha
+
+	SAVE_REGS
+
+	# r17 - r31 mainly for Write_256 macro.
+	li	17, 16
+	li	18, 32
+	li	19, 48
+	li	20, 64
+	li	21, 80
+	li	22, 96
+	li	23, 112
+	li	24, 128
+	li	25, 144
+	li	26, 160
+	li	27, 176
+	li	28, 192
+	li	29, 208
+	li	30, 224
+	li	31, 240
+
+	mr 15, 6			# len
+	li 14, 0			# offset to inp and outp
+
+        lxvw4x	48, 0, 3		#  vr16, constants
+	lxvw4x	49, 17, 3		#  vr17, key 1
+	lxvw4x	50, 18, 3		#  vr18, key 2
+	lxvw4x	51, 19, 3		#  vr19, counter, nonce
+
+	# create (0, 1, 2, 3) counters
+	vspltisw 0, 0
+	vspltisw 1, 1
+	vspltisw 2, 2
+	vspltisw 3, 3
+	vmrghw	4, 0, 1
+	vmrglw	5, 2, 3
+	vsldoi	30, 4, 5, 8		# vr30 counter, 4 (0, 1, 2, 3)
+
+	vspltisw 21, 12
+	vspltisw 23, 7
+
+	addis	11, 2, permx@toc@ha
+	addi	11, 11, permx@toc@l
+	lxvw4x	32+20, 0, 11
+	lxvw4x	32+22, 17, 11
+
+	sradi	8, 7, 1
+
+	mtctr 8
+
+	# save constants to vsx
+	xxlor	16, 48, 48
+	xxlor	17, 49, 49
+	xxlor	18, 50, 50
+	xxlor	19, 51, 51
+
+	vspltisw 25, 4
+	vspltisw 26, 8
+
+	xxlor	25, 32+26, 32+26
+	xxlor	24, 32+25, 32+25
+
+	vadduwm	31, 30, 25		# counter = (0, 1, 2, 3) + (4, 4, 4, 4)
+	xxlor	30, 32+30, 32+30
+	xxlor	31, 32+31, 32+31
+
+	xxlor	20, 32+20, 32+20
+	xxlor	21, 32+21, 32+21
+	xxlor	22, 32+22, 32+22
+	xxlor	23, 32+23, 32+23
+
+	cmpdi	6, 512
+	blt	Loop_last
+
+Loop_8x:
+	xxspltw  32+0, 16, 0
+	xxspltw  32+1, 16, 1
+	xxspltw  32+2, 16, 2
+	xxspltw  32+3, 16, 3
+
+	xxspltw  32+4, 17, 0
+	xxspltw  32+5, 17, 1
+	xxspltw  32+6, 17, 2
+	xxspltw  32+7, 17, 3
+	xxspltw  32+8, 18, 0
+	xxspltw  32+9, 18, 1
+	xxspltw  32+10, 18, 2
+	xxspltw  32+11, 18, 3
+	xxspltw  32+12, 19, 0
+	xxspltw  32+13, 19, 1
+	xxspltw  32+14, 19, 2
+	xxspltw  32+15, 19, 3
+	vadduwm	12, 12, 30	# increase counter
+
+	xxspltw  32+16, 16, 0
+	xxspltw  32+17, 16, 1
+	xxspltw  32+18, 16, 2
+	xxspltw  32+19, 16, 3
+
+	xxspltw  32+20, 17, 0
+	xxspltw  32+21, 17, 1
+	xxspltw  32+22, 17, 2
+	xxspltw  32+23, 17, 3
+	xxspltw  32+24, 18, 0
+	xxspltw  32+25, 18, 1
+	xxspltw  32+26, 18, 2
+	xxspltw  32+27, 18, 3
+	xxspltw  32+28, 19, 0
+	xxspltw  32+29, 19, 1
+	vadduwm	28, 28, 31	# increase counter
+	xxspltw  32+30, 19, 2
+	xxspltw  32+31, 19, 3
+
+.align 5
+quarter_loop_8x:
+	QT_loop_8x
+
+	bdnz	quarter_loop_8x
+
+	xxlor	0, 32+30, 32+30
+	xxlor	32+30, 30, 30
+	vadduwm	12, 12, 30
+	xxlor	32+30, 0, 0
+	TP_4x 0, 1, 2, 3
+	TP_4x 4, 5, 6, 7
+	TP_4x 8, 9, 10, 11
+	TP_4x 12, 13, 14, 15
+
+	xxlor	0, 48, 48
+	xxlor	1, 49, 49
+	xxlor	2, 50, 50
+	xxlor	3, 51, 51
+	xxlor	48, 16, 16
+	xxlor	49, 17, 17
+	xxlor	50, 18, 18
+	xxlor	51, 19, 19
+	Add_state 0
+	xxlor	48, 0, 0
+	xxlor	49, 1, 1
+	xxlor	50, 2, 2
+	xxlor	51, 3, 3
+	Write_256 0
+	addi	14, 14, 256	# offset +=256
+	addi	15, 15, -256	# len -=256
+
+	xxlor	5, 32+31, 32+31
+	xxlor	32+31, 31, 31
+	vadduwm	28, 28, 31
+	xxlor	32+31, 5, 5
+	TP_4x 16+0, 16+1, 16+2, 16+3
+	TP_4x 16+4, 16+5, 16+6, 16+7
+	TP_4x 16+8, 16+9, 16+10, 16+11
+	TP_4x 16+12, 16+13, 16+14, 16+15
+
+	xxlor	32, 16, 16
+	xxlor	33, 17, 17
+	xxlor	34, 18, 18
+	xxlor	35, 19, 19
+	Add_state 16
+	Write_256 16
+	addi	14, 14, 256	# offset +=256
+	addi	15, 15, -256	# len +=256
+
+	xxlor	32+24, 24, 24
+	xxlor	32+25, 25, 25
+	xxlor	32+30, 30, 30
+	vadduwm	30, 30, 25
+	vadduwm	31, 30, 24
+	xxlor	30, 32+30, 32+30
+	xxlor	31, 32+31, 32+31
+
+	cmpdi	15, 0
+	beq	Out_loop
+
+	cmpdi	15, 512
+	blt	Loop_last
+
+	mtctr 8
+	b Loop_8x
+
+Loop_last:
+        lxvw4x	48, 0, 3		#  vr16, constants
+	lxvw4x	49, 17, 3		#  vr17, key 1
+	lxvw4x	50, 18, 3		#  vr18, key 2
+	lxvw4x	51, 19, 3		#  vr19, counter, nonce
+
+	vspltisw 21, 12
+	vspltisw 23, 7
+	addis	11, 2, permx@toc@ha
+	addi	11, 11, permx@toc@l
+	lxvw4x	32+20, 0, 11
+	lxvw4x	32+22, 17, 11
+
+	sradi	8, 7, 1
+	mtctr 8
+
+Loop_4x:
+	vspltw  0, 16, 0
+	vspltw  1, 16, 1
+	vspltw  2, 16, 2
+	vspltw  3, 16, 3
+
+	vspltw  4, 17, 0
+	vspltw  5, 17, 1
+	vspltw  6, 17, 2
+	vspltw  7, 17, 3
+	vspltw  8, 18, 0
+	vspltw  9, 18, 1
+	vspltw  10, 18, 2
+	vspltw  11, 18, 3
+	vspltw  12, 19, 0
+	vadduwm	12, 12, 30	# increase counter
+	vspltw  13, 19, 1
+	vspltw  14, 19, 2
+	vspltw  15, 19, 3
+
+.align 5
+quarter_loop:
+	QT_loop_4x
+
+	bdnz	quarter_loop
+
+	vadduwm	12, 12, 30
+	TP_4x 0, 1, 2, 3
+	TP_4x 4, 5, 6, 7
+	TP_4x 8, 9, 10, 11
+	TP_4x 12, 13, 14, 15
+
+	Add_state 0
+	Write_256 0
+	addi	14, 14, 256	# offset += 256
+	addi	15, 15, -256	# len += 256
+
+	# Update state counter
+	vspltisw 25, 4
+	vadduwm	30, 30, 25
+
+	cmpdi	15, 0
+	beq	Out_loop
+	cmpdi	15, 256
+	blt	Out_loop
+
+	mtctr 8
+	b Loop_4x
+
+Out_loop:
+	RESTORE_REGS
+	blr
+
+Out_no_chacha:
+	li	3, 0
+	blr
+SYM_FUNC_END(chacha_p10le_8x)
+
+SYM_DATA_START_LOCAL(PERMX)
+.align 5
+permx:
+.long 0x22330011, 0x66774455, 0xaabb8899, 0xeeffccdd
+.long 0x11223300, 0x55667744, 0x99aabb88, 0xddeeffcc
+SYM_DATA_END(PERMX)
diff --git a/lib/crypto/powerpc/chacha.h b/lib/crypto/powerpc/chacha.h
new file mode 100644
index 000000000..1df6e1ce3
--- /dev/null
+++ b/lib/crypto/powerpc/chacha.h
@@ -0,0 +1,76 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * ChaCha stream cipher (P10 accelerated)
+ *
+ * Copyright 2023- IBM Corp. All rights reserved.
+ */
+
+#include <crypto/internal/simd.h>
+#include <linux/kernel.h>
+#include <linux/cpufeature.h>
+#include <linux/sizes.h>
+#include <asm/simd.h>
+#include <asm/switch_to.h>
+
+asmlinkage void chacha_p10le_8x(const struct chacha_state *state, u8 *dst,
+				const u8 *src, unsigned int len, int nrounds);
+
+static __ro_after_init DEFINE_STATIC_KEY_FALSE(have_p10);
+
+static void vsx_begin(void)
+{
+	preempt_disable();
+	enable_kernel_vsx();
+}
+
+static void vsx_end(void)
+{
+	disable_kernel_vsx();
+	preempt_enable();
+}
+
+static void chacha_p10_do_8x(struct chacha_state *state, u8 *dst, const u8 *src,
+			     unsigned int bytes, int nrounds)
+{
+	unsigned int l = bytes & ~0x0FF;
+
+	if (l > 0) {
+		chacha_p10le_8x(state, dst, src, l, nrounds);
+		bytes -= l;
+		src += l;
+		dst += l;
+		state->x[12] += l / CHACHA_BLOCK_SIZE;
+	}
+
+	if (bytes > 0)
+		chacha_crypt_generic(state, dst, src, bytes, nrounds);
+}
+
+#define hchacha_block_arch hchacha_block_generic /* not implemented yet */
+
+static void chacha_crypt_arch(struct chacha_state *state, u8 *dst,
+			      const u8 *src, unsigned int bytes, int nrounds)
+{
+	if (!static_branch_likely(&have_p10) || bytes <= CHACHA_BLOCK_SIZE ||
+	    !crypto_simd_usable())
+		return chacha_crypt_generic(state, dst, src, bytes, nrounds);
+
+	do {
+		unsigned int todo = min_t(unsigned int, bytes, SZ_4K);
+
+		vsx_begin();
+		chacha_p10_do_8x(state, dst, src, todo, nrounds);
+		vsx_end();
+
+		bytes -= todo;
+		src += todo;
+		dst += todo;
+	} while (bytes);
+}
+
+#define chacha_mod_init_arch chacha_mod_init_arch
+static void chacha_mod_init_arch(void)
+{
+	if (cpu_has_feature(CPU_FTR_ARCH_31))
+		static_branch_enable(&have_p10);
+}
diff --git a/lib/crypto/powerpc/curve25519-ppc64le_asm.S b/lib/crypto/powerpc/curve25519-ppc64le_asm.S
new file mode 100644
index 000000000..06c1febe2
--- /dev/null
+++ b/lib/crypto/powerpc/curve25519-ppc64le_asm.S
@@ -0,0 +1,671 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+#
+# This code is taken from CRYPTOGAMs[1] and is included here using the option
+# in the license to distribute the code under the GPL. Therefore 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.
+#
+# [1] https://github.com/dot-asm/cryptogams/
+
+# Copyright (c) 2006-2017, CRYPTOGAMS by <appro@openssl.org>
+# All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+#
+#       * Redistributions of source code must retain copyright notices,
+#         this list of conditions and the following disclaimer.
+#
+#       * Redistributions in binary form must reproduce the above
+#         copyright notice, this list of conditions and the following
+#         disclaimer in the documentation and/or other materials
+#         provided with the distribution.
+#
+#       * Neither the name of the CRYPTOGAMS nor the names of its
+#         copyright holder and contributors may be used to endorse or
+#         promote products derived from this software without specific
+#         prior written permission.
+#
+# ALTERNATIVELY, provided that this notice is retained in full, this
+# product may be distributed under the terms of the GNU General Public
+# License (GPL), in which case the provisions of the GPL apply INSTEAD OF
+# those given above.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER AND CONTRIBUTORS
+# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+# ====================================================================
+# Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
+# project. The module is, however, dual licensed under OpenSSL and
+# CRYPTOGAMS licenses depending on where you obtain it. For further
+# details see https://www.openssl.org/~appro/cryptogams/.
+# ====================================================================
+
+#
+# ====================================================================
+# Written and Modified by Danny Tsen <dtsen@us.ibm.com>
+# - Added x25519_fe51_sqr_times, x25519_fe51_frombytes, x25519_fe51_tobytes
+#   and x25519_cswap
+#
+# Copyright 2024- IBM Corp.
+#
+# X25519 lower-level primitives for PPC64.
+#
+
+#include <linux/linkage.h>
+
+.text
+
+.align	5
+SYM_FUNC_START(x25519_fe51_mul)
+
+	stdu	1,-144(1)
+	std	21,56(1)
+	std	22,64(1)
+	std	23,72(1)
+	std	24,80(1)
+	std	25,88(1)
+	std	26,96(1)
+	std	27,104(1)
+	std	28,112(1)
+	std	29,120(1)
+	std	30,128(1)
+	std	31,136(1)
+
+	ld	6,0(5)
+	ld	7,0(4)
+	ld	8,8(4)
+	ld	9,16(4)
+	ld	10,24(4)
+	ld	11,32(4)
+
+	mulld	22,7,6
+	mulhdu	23,7,6
+
+	mulld	24,8,6
+	mulhdu	25,8,6
+
+	mulld	30,11,6
+	mulhdu	31,11,6
+	ld	4,8(5)
+	mulli	11,11,19
+
+	mulld	26,9,6
+	mulhdu	27,9,6
+
+	mulld	28,10,6
+	mulhdu	29,10,6
+	mulld	12,11,4
+	mulhdu	21,11,4
+	addc	22,22,12
+	adde	23,23,21
+
+	mulld	12,7,4
+	mulhdu	21,7,4
+	addc	24,24,12
+	adde	25,25,21
+
+	mulld	12,10,4
+	mulhdu	21,10,4
+	ld	6,16(5)
+	mulli	10,10,19
+	addc	30,30,12
+	adde	31,31,21
+
+	mulld	12,8,4
+	mulhdu	21,8,4
+	addc	26,26,12
+	adde	27,27,21
+
+	mulld	12,9,4
+	mulhdu	21,9,4
+	addc	28,28,12
+	adde	29,29,21
+	mulld	12,10,6
+	mulhdu	21,10,6
+	addc	22,22,12
+	adde	23,23,21
+
+	mulld	12,11,6
+	mulhdu	21,11,6
+	addc	24,24,12
+	adde	25,25,21
+
+	mulld	12,9,6
+	mulhdu	21,9,6
+	ld	4,24(5)
+	mulli	9,9,19
+	addc	30,30,12
+	adde	31,31,21
+
+	mulld	12,7,6
+	mulhdu	21,7,6
+	addc	26,26,12
+	adde	27,27,21
+
+	mulld	12,8,6
+	mulhdu	21,8,6
+	addc	28,28,12
+	adde	29,29,21
+	mulld	12,9,4
+	mulhdu	21,9,4
+	addc	22,22,12
+	adde	23,23,21
+
+	mulld	12,10,4
+	mulhdu	21,10,4
+	addc	24,24,12
+	adde	25,25,21
+
+	mulld	12,8,4
+	mulhdu	21,8,4
+	ld	6,32(5)
+	mulli	8,8,19
+	addc	30,30,12
+	adde	31,31,21
+
+	mulld	12,11,4
+	mulhdu	21,11,4
+	addc	26,26,12
+	adde	27,27,21
+
+	mulld	12,7,4
+	mulhdu	21,7,4
+	addc	28,28,12
+	adde	29,29,21
+	mulld	12,8,6
+	mulhdu	21,8,6
+	addc	22,22,12
+	adde	23,23,21
+
+	mulld	12,9,6
+	mulhdu	21,9,6
+	addc	24,24,12
+	adde	25,25,21
+
+	mulld	12,10,6
+	mulhdu	21,10,6
+	addc	26,26,12
+	adde	27,27,21
+
+	mulld	12,11,6
+	mulhdu	21,11,6
+	addc	28,28,12
+	adde	29,29,21
+
+	mulld	12,7,6
+	mulhdu	21,7,6
+	addc	30,30,12
+	adde	31,31,21
+
+.Lfe51_reduce:
+	li	0,-1
+	srdi	0,0,13
+
+	srdi	12,26,51
+	and	9,26,0
+	insrdi	12,27,51,0
+	srdi	21,22,51
+	and	7,22,0
+	insrdi	21,23,51,0
+	addc	28,28,12
+	addze	29,29
+	addc	24,24,21
+	addze	25,25
+
+	srdi	12,28,51
+	and	10,28,0
+	insrdi	12,29,51,0
+	srdi	21,24,51
+	and	8,24,0
+	insrdi	21,25,51,0
+	addc	30,30,12
+	addze	31,31
+	add	9,9,21
+
+	srdi	12,30,51
+	and	11,30,0
+	insrdi	12,31,51,0
+	mulli	12,12,19
+
+	add	7,7,12
+
+	srdi	21,9,51
+	and	9,9,0
+	add	10,10,21
+
+	srdi	12,7,51
+	and	7,7,0
+	add	8,8,12
+
+	std	9,16(3)
+	std	10,24(3)
+	std	11,32(3)
+	std	7,0(3)
+	std	8,8(3)
+
+	ld	21,56(1)
+	ld	22,64(1)
+	ld	23,72(1)
+	ld	24,80(1)
+	ld	25,88(1)
+	ld	26,96(1)
+	ld	27,104(1)
+	ld	28,112(1)
+	ld	29,120(1)
+	ld	30,128(1)
+	ld	31,136(1)
+	addi	1,1,144
+	blr
+SYM_FUNC_END(x25519_fe51_mul)
+
+.align	5
+SYM_FUNC_START(x25519_fe51_sqr)
+
+	stdu	1,-144(1)
+	std	21,56(1)
+	std	22,64(1)
+	std	23,72(1)
+	std	24,80(1)
+	std	25,88(1)
+	std	26,96(1)
+	std	27,104(1)
+	std	28,112(1)
+	std	29,120(1)
+	std	30,128(1)
+	std	31,136(1)
+
+	ld	7,0(4)
+	ld	8,8(4)
+	ld	9,16(4)
+	ld	10,24(4)
+	ld	11,32(4)
+
+	add	6,7,7
+	mulli	21,11,19
+
+	mulld	22,7,7
+	mulhdu	23,7,7
+	mulld	24,8,6
+	mulhdu	25,8,6
+	mulld	26,9,6
+	mulhdu	27,9,6
+	mulld	28,10,6
+	mulhdu	29,10,6
+	mulld	30,11,6
+	mulhdu	31,11,6
+	add	6,8,8
+	mulld	12,11,21
+	mulhdu	11,11,21
+	addc	28,28,12
+	adde	29,29,11
+
+	mulli	5,10,19
+
+	mulld	12,8,8
+	mulhdu	11,8,8
+	addc	26,26,12
+	adde	27,27,11
+	mulld	12,9,6
+	mulhdu	11,9,6
+	addc	28,28,12
+	adde	29,29,11
+	mulld	12,10,6
+	mulhdu	11,10,6
+	addc	30,30,12
+	adde	31,31,11
+	mulld	12,21,6
+	mulhdu	11,21,6
+	add	6,10,10
+	addc	22,22,12
+	adde	23,23,11
+	mulld	12,10,5
+	mulhdu	10,10,5
+	addc	24,24,12
+	adde	25,25,10
+	mulld	12,6,21
+	mulhdu	10,6,21
+	add	6,9,9
+	addc	26,26,12
+	adde	27,27,10
+
+	mulld	12,9,9
+	mulhdu	10,9,9
+	addc	30,30,12
+	adde	31,31,10
+	mulld	12,5,6
+	mulhdu	10,5,6
+	addc	22,22,12
+	adde	23,23,10
+	mulld	12,21,6
+	mulhdu	10,21,6
+	addc	24,24,12
+	adde	25,25,10
+
+	b	.Lfe51_reduce
+SYM_FUNC_END(x25519_fe51_sqr)
+
+.align	5
+SYM_FUNC_START(x25519_fe51_mul121666)
+
+	stdu	1,-144(1)
+	std	21,56(1)
+	std	22,64(1)
+	std	23,72(1)
+	std	24,80(1)
+	std	25,88(1)
+	std	26,96(1)
+	std	27,104(1)
+	std	28,112(1)
+	std	29,120(1)
+	std	30,128(1)
+	std	31,136(1)
+
+	lis	6,1
+	ori	6,6,56130
+	ld	7,0(4)
+	ld	8,8(4)
+	ld	9,16(4)
+	ld	10,24(4)
+	ld	11,32(4)
+
+	mulld	22,7,6
+	mulhdu	23,7,6
+	mulld	24,8,6
+	mulhdu	25,8,6
+	mulld	26,9,6
+	mulhdu	27,9,6
+	mulld	28,10,6
+	mulhdu	29,10,6
+	mulld	30,11,6
+	mulhdu	31,11,6
+
+	b	.Lfe51_reduce
+SYM_FUNC_END(x25519_fe51_mul121666)
+
+.align	5
+SYM_FUNC_START(x25519_fe51_sqr_times)
+
+	stdu	1,-144(1)
+	std	21,56(1)
+	std	22,64(1)
+	std	23,72(1)
+	std	24,80(1)
+	std	25,88(1)
+	std	26,96(1)
+	std	27,104(1)
+	std	28,112(1)
+	std	29,120(1)
+	std	30,128(1)
+	std	31,136(1)
+
+	ld	7,0(4)
+	ld	8,8(4)
+	ld	9,16(4)
+	ld	10,24(4)
+	ld	11,32(4)
+
+	mtctr	5
+
+.Lsqr_times_loop:
+	add	6,7,7
+	mulli	21,11,19
+
+	mulld	22,7,7
+	mulhdu	23,7,7
+	mulld	24,8,6
+	mulhdu	25,8,6
+	mulld	26,9,6
+	mulhdu	27,9,6
+	mulld	28,10,6
+	mulhdu	29,10,6
+	mulld	30,11,6
+	mulhdu	31,11,6
+	add	6,8,8
+	mulld	12,11,21
+	mulhdu	11,11,21
+	addc	28,28,12
+	adde	29,29,11
+
+	mulli	5,10,19
+
+	mulld	12,8,8
+	mulhdu	11,8,8
+	addc	26,26,12
+	adde	27,27,11
+	mulld	12,9,6
+	mulhdu	11,9,6
+	addc	28,28,12
+	adde	29,29,11
+	mulld	12,10,6
+	mulhdu	11,10,6
+	addc	30,30,12
+	adde	31,31,11
+	mulld	12,21,6
+	mulhdu	11,21,6
+	add	6,10,10
+	addc	22,22,12
+	adde	23,23,11
+	mulld	12,10,5
+	mulhdu	10,10,5
+	addc	24,24,12
+	adde	25,25,10
+	mulld	12,6,21
+	mulhdu	10,6,21
+	add	6,9,9
+	addc	26,26,12
+	adde	27,27,10
+
+	mulld	12,9,9
+	mulhdu	10,9,9
+	addc	30,30,12
+	adde	31,31,10
+	mulld	12,5,6
+	mulhdu	10,5,6
+	addc	22,22,12
+	adde	23,23,10
+	mulld	12,21,6
+	mulhdu	10,21,6
+	addc	24,24,12
+	adde	25,25,10
+
+	# fe51_reduce
+	li	0,-1
+	srdi	0,0,13
+
+	srdi	12,26,51
+	and	9,26,0
+	insrdi	12,27,51,0
+	srdi	21,22,51
+	and	7,22,0
+	insrdi	21,23,51,0
+	addc	28,28,12
+	addze	29,29
+	addc	24,24,21
+	addze	25,25
+
+	srdi	12,28,51
+	and	10,28,0
+	insrdi	12,29,51,0
+	srdi	21,24,51
+	and	8,24,0
+	insrdi	21,25,51,0
+	addc	30,30,12
+	addze	31,31
+	add	9,9,21
+
+	srdi	12,30,51
+	and	11,30,0
+	insrdi	12,31,51,0
+	mulli	12,12,19
+
+	add	7,7,12
+
+	srdi	21,9,51
+	and	9,9,0
+	add	10,10,21
+
+	srdi	12,7,51
+	and	7,7,0
+	add	8,8,12
+
+	bdnz	.Lsqr_times_loop
+
+	std	9,16(3)
+	std	10,24(3)
+	std	11,32(3)
+	std	7,0(3)
+	std	8,8(3)
+
+	ld	21,56(1)
+	ld	22,64(1)
+	ld	23,72(1)
+	ld	24,80(1)
+	ld	25,88(1)
+	ld	26,96(1)
+	ld	27,104(1)
+	ld	28,112(1)
+	ld	29,120(1)
+	ld	30,128(1)
+	ld	31,136(1)
+	addi	1,1,144
+	blr
+SYM_FUNC_END(x25519_fe51_sqr_times)
+
+.align	5
+SYM_FUNC_START(x25519_fe51_frombytes)
+
+	li	12, -1
+	srdi	12, 12, 13	# 0x7ffffffffffff
+
+	ld	5, 0(4)
+	ld	6, 8(4)
+	ld	7, 16(4)
+	ld	8, 24(4)
+
+	srdi	10, 5, 51
+	and	5, 5, 12	# h0
+
+	sldi	11, 6, 13
+	or	11, 10, 11	# h1t
+	srdi	10, 6, 38
+	and	6, 11, 12	# h1
+
+	sldi	11, 7, 26
+	or	10, 10, 11	# h2t
+
+	srdi	11, 7, 25
+	and	7, 10, 12	# h2
+	sldi	10, 8, 39
+	or	11, 11, 10	# h3t
+
+	srdi	9, 8, 12
+	and	8, 11, 12	# h3
+	and	9, 9, 12	# h4
+
+	std	5, 0(3)
+	std	6, 8(3)
+	std	7, 16(3)
+	std	8, 24(3)
+	std	9, 32(3)
+
+	blr
+SYM_FUNC_END(x25519_fe51_frombytes)
+
+.align	5
+SYM_FUNC_START(x25519_fe51_tobytes)
+
+	ld	5, 0(4)
+	ld	6, 8(4)
+	ld	7, 16(4)
+	ld	8, 24(4)
+	ld	9, 32(4)
+
+	li	12, -1
+	srdi	12, 12, 13	# 0x7ffffffffffff
+
+	# Full reducuction
+	addi	10, 5, 19
+	srdi	10, 10, 51
+	add	10, 10, 6
+	srdi	10, 10, 51
+	add	10, 10, 7
+	srdi	10, 10, 51
+	add	10, 10, 8
+	srdi	10, 10, 51
+	add	10, 10, 9
+	srdi	10, 10, 51
+
+	mulli	10, 10, 19
+	add	5, 5, 10
+	srdi	11, 5, 51
+	add	6, 6, 11
+	srdi	11, 6, 51
+	add	7, 7, 11
+	srdi	11, 7, 51
+	add	8, 8, 11
+	srdi	11, 8, 51
+	add	9, 9, 11
+
+	and	5, 5, 12
+	and	6, 6, 12
+	and	7, 7, 12
+	and	8, 8, 12
+	and	9, 9, 12
+
+	sldi	10, 6, 51
+	or	5, 5, 10	# s0
+
+	srdi	11, 6, 13
+	sldi	10, 7, 38
+	or	6, 11, 10	# s1
+
+	srdi	11, 7, 26
+	sldi	10, 8, 25
+	or	7, 11, 10	# s2
+
+	srdi	11, 8, 39
+	sldi	10, 9, 12
+	or	8, 11, 10	# s4
+
+	std	5, 0(3)
+	std	6, 8(3)
+	std	7, 16(3)
+	std	8, 24(3)
+
+	blr
+SYM_FUNC_END(x25519_fe51_tobytes)
+
+.align	5
+SYM_FUNC_START(x25519_cswap)
+
+	li	7, 5
+	neg	6, 5
+	mtctr	7
+
+.Lswap_loop:
+	ld	8, 0(3)
+	ld	9, 0(4)
+	xor	10, 8, 9
+	and	10, 10, 6
+	xor	11, 8, 10
+	xor	12, 9, 10
+	std	11, 0(3)
+	addi	3, 3, 8
+	std	12, 0(4)
+	addi	4, 4, 8
+	bdnz	.Lswap_loop
+
+	blr
+SYM_FUNC_END(x25519_cswap)
diff --git a/lib/crypto/powerpc/curve25519.h b/lib/crypto/powerpc/curve25519.h
new file mode 100644
index 000000000..dee6234c4
--- /dev/null
+++ b/lib/crypto/powerpc/curve25519.h
@@ -0,0 +1,186 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright 2024- IBM Corp.
+ *
+ * X25519 scalar multiplication with 51 bits limbs for PPC64le.
+ *   Based on RFC7748 and AArch64 optimized implementation for X25519
+ *     - Algorithm 1 Scalar multiplication of a variable point
+ */
+
+#include <linux/types.h>
+#include <linux/jump_label.h>
+#include <linux/kernel.h>
+
+#include <linux/cpufeature.h>
+#include <linux/processor.h>
+
+typedef uint64_t fe51[5];
+
+asmlinkage void x25519_fe51_mul(fe51 h, const fe51 f, const fe51 g);
+asmlinkage void x25519_fe51_sqr(fe51 h, const fe51 f);
+asmlinkage void x25519_fe51_mul121666(fe51 h, fe51 f);
+asmlinkage void x25519_fe51_sqr_times(fe51 h, const fe51 f, int n);
+asmlinkage void x25519_fe51_frombytes(fe51 h, const uint8_t *s);
+asmlinkage void x25519_fe51_tobytes(uint8_t *s, const fe51 h);
+asmlinkage void x25519_cswap(fe51 p, fe51 q, unsigned int bit);
+
+#define fmul x25519_fe51_mul
+#define fsqr x25519_fe51_sqr
+#define fmul121666 x25519_fe51_mul121666
+#define fe51_tobytes x25519_fe51_tobytes
+
+static void fadd(fe51 h, const fe51 f, const fe51 g)
+{
+	h[0] = f[0] + g[0];
+	h[1] = f[1] + g[1];
+	h[2] = f[2] + g[2];
+	h[3] = f[3] + g[3];
+	h[4] = f[4] + g[4];
+}
+
+/*
+ * Prime = 2 ** 255 - 19, 255 bits
+ *    (0x7fffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffed)
+ *
+ * Prime in 5 51-bit limbs
+ */
+static fe51 prime51 = { 0x7ffffffffffed, 0x7ffffffffffff, 0x7ffffffffffff, 0x7ffffffffffff, 0x7ffffffffffff};
+
+static void fsub(fe51 h, const fe51 f, const fe51 g)
+{
+	h[0] = (f[0] + ((prime51[0] * 2))) - g[0];
+	h[1] = (f[1] + ((prime51[1] * 2))) - g[1];
+	h[2] = (f[2] + ((prime51[2] * 2))) - g[2];
+	h[3] = (f[3] + ((prime51[3] * 2))) - g[3];
+	h[4] = (f[4] + ((prime51[4] * 2))) - g[4];
+}
+
+static void fe51_frombytes(fe51 h, const uint8_t *s)
+{
+	/*
+	 * Make sure 64-bit aligned.
+	 */
+	unsigned char sbuf[32+8];
+	unsigned char *sb = PTR_ALIGN((void *)sbuf, 8);
+
+	memcpy(sb, s, 32);
+	x25519_fe51_frombytes(h, sb);
+}
+
+static void finv(fe51 o, const fe51 i)
+{
+	fe51 a0, b, c, t00;
+
+	fsqr(a0, i);
+	x25519_fe51_sqr_times(t00, a0, 2);
+
+	fmul(b, t00, i);
+	fmul(a0, b, a0);
+
+	fsqr(t00, a0);
+
+	fmul(b, t00, b);
+	x25519_fe51_sqr_times(t00, b, 5);
+
+	fmul(b, t00, b);
+	x25519_fe51_sqr_times(t00, b, 10);
+
+	fmul(c, t00, b);
+	x25519_fe51_sqr_times(t00, c, 20);
+
+	fmul(t00, t00, c);
+	x25519_fe51_sqr_times(t00, t00, 10);
+
+	fmul(b, t00, b);
+	x25519_fe51_sqr_times(t00, b, 50);
+
+	fmul(c, t00, b);
+	x25519_fe51_sqr_times(t00, c, 100);
+
+	fmul(t00, t00, c);
+	x25519_fe51_sqr_times(t00, t00, 50);
+
+	fmul(t00, t00, b);
+	x25519_fe51_sqr_times(t00, t00, 5);
+
+	fmul(o, t00, a0);
+}
+
+static void curve25519_fe51(uint8_t out[32], const uint8_t scalar[32],
+			    const uint8_t point[32])
+{
+	fe51 x1, x2, z2, x3, z3;
+	uint8_t s[32];
+	unsigned int swap = 0;
+	int i;
+
+	memcpy(s, scalar, 32);
+	s[0]  &= 0xf8;
+	s[31] &= 0x7f;
+	s[31] |= 0x40;
+	fe51_frombytes(x1, point);
+
+	z2[0] = z2[1] = z2[2] = z2[3] = z2[4] = 0;
+	x3[0] = x1[0];
+	x3[1] = x1[1];
+	x3[2] = x1[2];
+	x3[3] = x1[3];
+	x3[4] = x1[4];
+
+	x2[0] = z3[0] = 1;
+	x2[1] = z3[1] = 0;
+	x2[2] = z3[2] = 0;
+	x2[3] = z3[3] = 0;
+	x2[4] = z3[4] = 0;
+
+	for (i = 254; i >= 0; --i) {
+		unsigned int k_t = 1 & (s[i / 8] >> (i & 7));
+		fe51 a, b, c, d, e;
+		fe51 da, cb, aa, bb;
+		fe51 dacb_p, dacb_m;
+
+		swap ^= k_t;
+		x25519_cswap(x2, x3, swap);
+		x25519_cswap(z2, z3, swap);
+		swap = k_t;
+
+		fsub(b, x2, z2);		// B = x_2 - z_2
+		fadd(a, x2, z2);		// A = x_2 + z_2
+		fsub(d, x3, z3);		// D = x_3 - z_3
+		fadd(c, x3, z3);		// C = x_3 + z_3
+
+		fsqr(bb, b);			// BB = B^2
+		fsqr(aa, a);			// AA = A^2
+		fmul(da, d, a);			// DA = D * A
+		fmul(cb, c, b);			// CB = C * B
+
+		fsub(e, aa, bb);		// E = AA - BB
+		fmul(x2, aa, bb);		// x2 = AA * BB
+		fadd(dacb_p, da, cb);		// DA + CB
+		fsub(dacb_m, da, cb);		// DA - CB
+
+		fmul121666(z3, e);		// 121666 * E
+		fsqr(z2, dacb_m);		// (DA - CB)^2
+		fsqr(x3, dacb_p);		// x3 = (DA + CB)^2
+		fadd(b, bb, z3);		// BB + 121666 * E
+		fmul(z3, x1, z2);		// z3 = x1 * (DA - CB)^2
+		fmul(z2, e, b);		// z2 = e * (BB + (DA + CB)^2)
+	}
+
+	finv(z2, z2);
+	fmul(x2, x2, z2);
+	fe51_tobytes(out, x2);
+}
+
+static void curve25519_arch(u8 mypublic[CURVE25519_KEY_SIZE],
+			    const u8 secret[CURVE25519_KEY_SIZE],
+			    const u8 basepoint[CURVE25519_KEY_SIZE])
+{
+	curve25519_fe51(mypublic, secret, basepoint);
+}
+
+static void curve25519_base_arch(u8 pub[CURVE25519_KEY_SIZE],
+				 const u8 secret[CURVE25519_KEY_SIZE])
+{
+	curve25519_fe51(pub, secret, curve25519_base_point);
+}
diff --git a/lib/crypto/powerpc/gf128hash.h b/lib/crypto/powerpc/gf128hash.h
new file mode 100644
index 000000000..629cd325d
--- /dev/null
+++ b/lib/crypto/powerpc/gf128hash.h
@@ -0,0 +1,109 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * GHASH routines supporting VMX instructions on the Power 8
+ *
+ * Copyright (C) 2015, 2019 International Business Machines Inc.
+ * Copyright (C) 2014 - 2018 Linaro Ltd.
+ * Copyright 2026 Google LLC
+ */
+
+#include <asm/simd.h>
+#include <asm/switch_to.h>
+#include <linux/cpufeature.h>
+#include <linux/jump_label.h>
+#include <linux/preempt.h>
+#include <linux/uaccess.h>
+
+static __ro_after_init DEFINE_STATIC_KEY_FALSE(have_vec_crypto);
+
+void gcm_init_p8(u64 htable[4][2], const u8 h[16]);
+void gcm_gmult_p8(u8 Xi[16], const u64 htable[4][2]);
+void gcm_ghash_p8(u8 Xi[16], const u64 htable[4][2], const u8 *in, size_t len);
+
+#define ghash_preparekey_arch ghash_preparekey_arch
+static void ghash_preparekey_arch(struct ghash_key *key,
+				  const u8 raw_key[GHASH_BLOCK_SIZE])
+{
+	ghash_key_to_polyval(raw_key, &key->h);
+
+	if (static_branch_likely(&have_vec_crypto) && likely(may_use_simd())) {
+		preempt_disable();
+		pagefault_disable();
+		enable_kernel_vsx();
+		gcm_init_p8(key->htable, raw_key);
+		disable_kernel_vsx();
+		pagefault_enable();
+		preempt_enable();
+	} else {
+		/* This reproduces gcm_init_p8() on both LE and BE systems. */
+		key->htable[0][0] = 0;
+		key->htable[0][1] = 0xc200000000000000;
+
+		key->htable[1][0] = 0;
+		key->htable[1][1] = le64_to_cpu(key->h.lo);
+
+		key->htable[2][0] = le64_to_cpu(key->h.lo);
+		key->htable[2][1] = le64_to_cpu(key->h.hi);
+
+		key->htable[3][0] = le64_to_cpu(key->h.hi);
+		key->htable[3][1] = 0;
+	}
+}
+
+#define ghash_mul_arch ghash_mul_arch
+static void ghash_mul_arch(struct polyval_elem *acc,
+			   const struct ghash_key *key)
+{
+	if (static_branch_likely(&have_vec_crypto) && likely(may_use_simd())) {
+		u8 ghash_acc[GHASH_BLOCK_SIZE];
+
+		polyval_acc_to_ghash(acc, ghash_acc);
+
+		preempt_disable();
+		pagefault_disable();
+		enable_kernel_vsx();
+		gcm_gmult_p8(ghash_acc, key->htable);
+		disable_kernel_vsx();
+		pagefault_enable();
+		preempt_enable();
+
+		ghash_acc_to_polyval(ghash_acc, acc);
+		memzero_explicit(ghash_acc, sizeof(ghash_acc));
+	} else {
+		polyval_mul_generic(acc, &key->h);
+	}
+}
+
+#define ghash_blocks_arch ghash_blocks_arch
+static void ghash_blocks_arch(struct polyval_elem *acc,
+			      const struct ghash_key *key,
+			      const u8 *data, size_t nblocks)
+{
+	if (static_branch_likely(&have_vec_crypto) && likely(may_use_simd())) {
+		u8 ghash_acc[GHASH_BLOCK_SIZE];
+
+		polyval_acc_to_ghash(acc, ghash_acc);
+
+		preempt_disable();
+		pagefault_disable();
+		enable_kernel_vsx();
+		gcm_ghash_p8(ghash_acc, key->htable, data,
+			     nblocks * GHASH_BLOCK_SIZE);
+		disable_kernel_vsx();
+		pagefault_enable();
+		preempt_enable();
+
+		ghash_acc_to_polyval(ghash_acc, acc);
+		memzero_explicit(ghash_acc, sizeof(ghash_acc));
+	} else {
+		ghash_blocks_generic(acc, &key->h, data, nblocks);
+	}
+}
+
+#define gf128hash_mod_init_arch gf128hash_mod_init_arch
+static void gf128hash_mod_init_arch(void)
+{
+	if (cpu_has_feature(CPU_FTR_ARCH_207S) &&
+	    (cur_cpu_spec->cpu_user_features2 & PPC_FEATURE2_VEC_CRYPTO))
+		static_branch_enable(&have_vec_crypto);
+}
diff --git a/lib/crypto/powerpc/ghashp8-ppc.pl b/lib/crypto/powerpc/ghashp8-ppc.pl
new file mode 100644
index 000000000..7c38eedc0
--- /dev/null
+++ b/lib/crypto/powerpc/ghashp8-ppc.pl
@@ -0,0 +1,244 @@
+#!/usr/bin/env perl
+# SPDX-License-Identifier: GPL-2.0
+
+# This code is taken from the OpenSSL project but the author (Andy Polyakov)
+# has relicensed it under the GPLv2. Therefore 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 original headers, including the original license headers, are
+# included below for completeness.
+
+# ====================================================================
+# Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
+# project. The module is, however, dual licensed under OpenSSL and
+# CRYPTOGAMS licenses depending on where you obtain it. For further
+# details see https://www.openssl.org/~appro/cryptogams/.
+# ====================================================================
+#
+# GHASH for PowerISA v2.07.
+#
+# July 2014
+#
+# Accurate performance measurements are problematic, because it's
+# always virtualized setup with possibly throttled processor.
+# Relative comparison is therefore more informative. This initial
+# version is ~2.1x slower than hardware-assisted AES-128-CTR, ~12x
+# faster than "4-bit" integer-only compiler-generated 64-bit code.
+# "Initial version" means that there is room for futher improvement.
+
+$flavour=shift;
+$output =shift;
+
+if ($flavour =~ /64/) {
+	$SIZE_T=8;
+	$LRSAVE=2*$SIZE_T;
+	$STU="stdu";
+	$POP="ld";
+	$PUSH="std";
+} elsif ($flavour =~ /32/) {
+	$SIZE_T=4;
+	$LRSAVE=$SIZE_T;
+	$STU="stwu";
+	$POP="lwz";
+	$PUSH="stw";
+} else { die "nonsense $flavour"; }
+
+$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
+( $xlate="${dir}ppc-xlate.pl" and -f $xlate ) or
+( $xlate="${dir}../../perlasm/ppc-xlate.pl" and -f $xlate) or
+( $xlate="${dir}../../../arch/powerpc/crypto/ppc-xlate.pl" and -f $xlate) or
+die "can't locate ppc-xlate.pl";
+
+open STDOUT,"| $^X $xlate $flavour $output" || die "can't call $xlate: $!";
+
+my ($Xip,$Htbl,$inp,$len)=map("r$_",(3..6));	# argument block
+
+my ($Xl,$Xm,$Xh,$IN)=map("v$_",(0..3));
+my ($zero,$t0,$t1,$t2,$xC2,$H,$Hh,$Hl,$lemask)=map("v$_",(4..12));
+my $vrsave="r12";
+
+$code=<<___;
+.machine	"any"
+
+.text
+
+.globl	.gcm_init_p8
+	lis		r0,0xfff0
+	li		r8,0x10
+	mfspr		$vrsave,256
+	li		r9,0x20
+	mtspr		256,r0
+	li		r10,0x30
+	lvx_u		$H,0,r4			# load H
+	le?xor		r7,r7,r7
+	le?addi		r7,r7,0x8		# need a vperm start with 08
+	le?lvsr		5,0,r7
+	le?vspltisb	6,0x0f
+	le?vxor		5,5,6			# set a b-endian mask
+	le?vperm	$H,$H,$H,5
+
+	vspltisb	$xC2,-16		# 0xf0
+	vspltisb	$t0,1			# one
+	vaddubm		$xC2,$xC2,$xC2		# 0xe0
+	vxor		$zero,$zero,$zero
+	vor		$xC2,$xC2,$t0		# 0xe1
+	vsldoi		$xC2,$xC2,$zero,15	# 0xe1...
+	vsldoi		$t1,$zero,$t0,1		# ...1
+	vaddubm		$xC2,$xC2,$xC2		# 0xc2...
+	vspltisb	$t2,7
+	vor		$xC2,$xC2,$t1		# 0xc2....01
+	vspltb		$t1,$H,0		# most significant byte
+	vsl		$H,$H,$t0		# H<<=1
+	vsrab		$t1,$t1,$t2		# broadcast carry bit
+	vand		$t1,$t1,$xC2
+	vxor		$H,$H,$t1		# twisted H
+
+	vsldoi		$H,$H,$H,8		# twist even more ...
+	vsldoi		$xC2,$zero,$xC2,8	# 0xc2.0
+	vsldoi		$Hl,$zero,$H,8		# ... and split
+	vsldoi		$Hh,$H,$zero,8
+
+	stvx_u		$xC2,0,r3		# save pre-computed table
+	stvx_u		$Hl,r8,r3
+	stvx_u		$H, r9,r3
+	stvx_u		$Hh,r10,r3
+
+	mtspr		256,$vrsave
+	blr
+	.long		0
+	.byte		0,12,0x14,0,0,0,2,0
+	.long		0
+.size	.gcm_init_p8,.-.gcm_init_p8
+
+.globl	.gcm_gmult_p8
+	lis		r0,0xfff8
+	li		r8,0x10
+	mfspr		$vrsave,256
+	li		r9,0x20
+	mtspr		256,r0
+	li		r10,0x30
+	lvx_u		$IN,0,$Xip		# load Xi
+
+	lvx_u		$Hl,r8,$Htbl		# load pre-computed table
+	 le?lvsl	$lemask,r0,r0
+	lvx_u		$H, r9,$Htbl
+	 le?vspltisb	$t0,0x07
+	lvx_u		$Hh,r10,$Htbl
+	 le?vxor	$lemask,$lemask,$t0
+	lvx_u		$xC2,0,$Htbl
+	 le?vperm	$IN,$IN,$IN,$lemask
+	vxor		$zero,$zero,$zero
+
+	vpmsumd		$Xl,$IN,$Hl		# H.lo·Xi.lo
+	vpmsumd		$Xm,$IN,$H		# H.hi·Xi.lo+H.lo·Xi.hi
+	vpmsumd		$Xh,$IN,$Hh		# H.hi·Xi.hi
+
+	vpmsumd		$t2,$Xl,$xC2		# 1st phase
+
+	vsldoi		$t0,$Xm,$zero,8
+	vsldoi		$t1,$zero,$Xm,8
+	vxor		$Xl,$Xl,$t0
+	vxor		$Xh,$Xh,$t1
+
+	vsldoi		$Xl,$Xl,$Xl,8
+	vxor		$Xl,$Xl,$t2
+
+	vsldoi		$t1,$Xl,$Xl,8		# 2nd phase
+	vpmsumd		$Xl,$Xl,$xC2
+	vxor		$t1,$t1,$Xh
+	vxor		$Xl,$Xl,$t1
+
+	le?vperm	$Xl,$Xl,$Xl,$lemask
+	stvx_u		$Xl,0,$Xip		# write out Xi
+
+	mtspr		256,$vrsave
+	blr
+	.long		0
+	.byte		0,12,0x14,0,0,0,2,0
+	.long		0
+.size	.gcm_gmult_p8,.-.gcm_gmult_p8
+
+.globl	.gcm_ghash_p8
+	lis		r0,0xfff8
+	li		r8,0x10
+	mfspr		$vrsave,256
+	li		r9,0x20
+	mtspr		256,r0
+	li		r10,0x30
+	lvx_u		$Xl,0,$Xip		# load Xi
+
+	lvx_u		$Hl,r8,$Htbl		# load pre-computed table
+	 le?lvsl	$lemask,r0,r0
+	lvx_u		$H, r9,$Htbl
+	 le?vspltisb	$t0,0x07
+	lvx_u		$Hh,r10,$Htbl
+	 le?vxor	$lemask,$lemask,$t0
+	lvx_u		$xC2,0,$Htbl
+	 le?vperm	$Xl,$Xl,$Xl,$lemask
+	vxor		$zero,$zero,$zero
+
+	lvx_u		$IN,0,$inp
+	addi		$inp,$inp,16
+	subi		$len,$len,16
+	 le?vperm	$IN,$IN,$IN,$lemask
+	vxor		$IN,$IN,$Xl
+	b		Loop
+
+.align	5
+Loop:
+	 subic		$len,$len,16
+	vpmsumd		$Xl,$IN,$Hl		# H.lo·Xi.lo
+	 subfe.		r0,r0,r0		# borrow?-1:0
+	vpmsumd		$Xm,$IN,$H		# H.hi·Xi.lo+H.lo·Xi.hi
+	 and		r0,r0,$len
+	vpmsumd		$Xh,$IN,$Hh		# H.hi·Xi.hi
+	 add		$inp,$inp,r0
+
+	vpmsumd		$t2,$Xl,$xC2		# 1st phase
+
+	vsldoi		$t0,$Xm,$zero,8
+	vsldoi		$t1,$zero,$Xm,8
+	vxor		$Xl,$Xl,$t0
+	vxor		$Xh,$Xh,$t1
+
+	vsldoi		$Xl,$Xl,$Xl,8
+	vxor		$Xl,$Xl,$t2
+	 lvx_u		$IN,0,$inp
+	 addi		$inp,$inp,16
+
+	vsldoi		$t1,$Xl,$Xl,8		# 2nd phase
+	vpmsumd		$Xl,$Xl,$xC2
+	 le?vperm	$IN,$IN,$IN,$lemask
+	vxor		$t1,$t1,$Xh
+	vxor		$IN,$IN,$t1
+	vxor		$IN,$IN,$Xl
+	beq		Loop			# did $len-=16 borrow?
+
+	vxor		$Xl,$Xl,$t1
+	le?vperm	$Xl,$Xl,$Xl,$lemask
+	stvx_u		$Xl,0,$Xip		# write out Xi
+
+	mtspr		256,$vrsave
+	blr
+	.long		0
+	.byte		0,12,0x14,0,0,0,4,0
+	.long		0
+.size	.gcm_ghash_p8,.-.gcm_ghash_p8
+
+.asciz  "GHASH for PowerISA 2.07, CRYPTOGAMS by <appro\@openssl.org>"
+.align  2
+___
+
+foreach (split("\n",$code)) {
+	if ($flavour =~ /le$/o) {	# little-endian
+	    s/le\?//o		or
+	    s/be\?/#be#/o;
+	} else {
+	    s/le\?/#le#/o	or
+	    s/be\?//o;
+	}
+	print $_,"\n";
+}
+
+close STDOUT; # enforce flush
diff --git a/lib/crypto/powerpc/poly1305-p10le_64.S b/lib/crypto/powerpc/poly1305-p10le_64.S
new file mode 100644
index 000000000..a3c1987f1
--- /dev/null
+++ b/lib/crypto/powerpc/poly1305-p10le_64.S
@@ -0,0 +1,1075 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+#
+# Accelerated poly1305 implementation for ppc64le.
+#
+# Copyright 2023- IBM Corp. All rights reserved
+#
+#===================================================================================
+# Written by Danny Tsen <dtsen@us.ibm.com>
+#
+# Poly1305 - this version mainly using vector/VSX/Scalar
+#  - 26 bits limbs
+#  - Handle multiple 64 byte blcok.
+#
+# Block size 16 bytes
+# key = (r, s)
+# clamp r &= 0x0FFFFFFC0FFFFFFC 0x0FFFFFFC0FFFFFFF
+# p = 2^130 - 5
+# a += m
+# a = (r + a) % p
+# a += s
+#
+# Improve performance by breaking down polynominal to the sum of products with
+#     h4 = m1 * r⁴ + m2 * r³ + m3 * r² + m4 * r
+#
+#  07/22/21 - this revison based on the above sum of products.  Setup r^4, r^3, r^2, r and s3, s2, s1, s0
+#             to 9 vectors for multiplications.
+#
+# setup r^4, r^3, r^2, r vectors
+#    vs    [r^1, r^3, r^2, r^4]
+#    vs0 = [r0,.....]
+#    vs1 = [r1,.....]
+#    vs2 = [r2,.....]
+#    vs3 = [r3,.....]
+#    vs4 = [r4,.....]
+#    vs5 = [r1*5,...]
+#    vs6 = [r2*5,...]
+#    vs7 = [r2*5,...]
+#    vs8 = [r4*5,...]
+#
+#  Each word in a vector consists a member of a "r/s" in [a * r/s].
+#
+# r0, r4*5, r3*5, r2*5, r1*5;
+# r1, r0,   r4*5, r3*5, r2*5;
+# r2, r1,   r0,   r4*5, r3*5;
+# r3, r2,   r1,   r0,   r4*5;
+# r4, r3,   r2,   r1,   r0  ;
+#
+#
+# poly1305_p10le_4blocks( uint8_t *k, uint32_t mlen, uint8_t *m)
+#  k = 32 bytes key
+#  r3 = k (r, s)
+#  r4 = mlen
+#  r5 = m
+#
+#include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/asm-compat.h>
+#include <linux/linkage.h>
+
+.machine "any"
+
+.text
+
+.macro	SAVE_GPR GPR OFFSET FRAME
+	std	\GPR,\OFFSET(\FRAME)
+.endm
+
+.macro	SAVE_VRS VRS OFFSET FRAME
+	li	16, \OFFSET
+	stvx	\VRS, 16, \FRAME
+.endm
+
+.macro	SAVE_VSX VSX OFFSET FRAME
+	li	16, \OFFSET
+	stxvx	\VSX, 16, \FRAME
+.endm
+
+.macro	RESTORE_GPR GPR OFFSET FRAME
+	ld	\GPR,\OFFSET(\FRAME)
+.endm
+
+.macro	RESTORE_VRS VRS OFFSET FRAME
+	li	16, \OFFSET
+	lvx	\VRS, 16, \FRAME
+.endm
+
+.macro	RESTORE_VSX VSX OFFSET FRAME
+	li	16, \OFFSET
+	lxvx	\VSX, 16, \FRAME
+.endm
+
+.macro SAVE_REGS
+	mflr 0
+	std 0, 16(1)
+	stdu 1,-752(1)
+
+	SAVE_GPR 14, 112, 1
+	SAVE_GPR 15, 120, 1
+	SAVE_GPR 16, 128, 1
+	SAVE_GPR 17, 136, 1
+	SAVE_GPR 18, 144, 1
+	SAVE_GPR 19, 152, 1
+	SAVE_GPR 20, 160, 1
+	SAVE_GPR 21, 168, 1
+	SAVE_GPR 22, 176, 1
+	SAVE_GPR 23, 184, 1
+	SAVE_GPR 24, 192, 1
+	SAVE_GPR 25, 200, 1
+	SAVE_GPR 26, 208, 1
+	SAVE_GPR 27, 216, 1
+	SAVE_GPR 28, 224, 1
+	SAVE_GPR 29, 232, 1
+	SAVE_GPR 30, 240, 1
+	SAVE_GPR 31, 248, 1
+
+	addi	9, 1, 256
+	SAVE_VRS 20, 0, 9
+	SAVE_VRS 21, 16, 9
+	SAVE_VRS 22, 32, 9
+	SAVE_VRS 23, 48, 9
+	SAVE_VRS 24, 64, 9
+	SAVE_VRS 25, 80, 9
+	SAVE_VRS 26, 96, 9
+	SAVE_VRS 27, 112, 9
+	SAVE_VRS 28, 128, 9
+	SAVE_VRS 29, 144, 9
+	SAVE_VRS 30, 160, 9
+	SAVE_VRS 31, 176, 9
+
+	SAVE_VSX 14, 192, 9
+	SAVE_VSX 15, 208, 9
+	SAVE_VSX 16, 224, 9
+	SAVE_VSX 17, 240, 9
+	SAVE_VSX 18, 256, 9
+	SAVE_VSX 19, 272, 9
+	SAVE_VSX 20, 288, 9
+	SAVE_VSX 21, 304, 9
+	SAVE_VSX 22, 320, 9
+	SAVE_VSX 23, 336, 9
+	SAVE_VSX 24, 352, 9
+	SAVE_VSX 25, 368, 9
+	SAVE_VSX 26, 384, 9
+	SAVE_VSX 27, 400, 9
+	SAVE_VSX 28, 416, 9
+	SAVE_VSX 29, 432, 9
+	SAVE_VSX 30, 448, 9
+	SAVE_VSX 31, 464, 9
+.endm # SAVE_REGS
+
+.macro RESTORE_REGS
+	addi	9, 1, 256
+	RESTORE_VRS 20, 0, 9
+	RESTORE_VRS 21, 16, 9
+	RESTORE_VRS 22, 32, 9
+	RESTORE_VRS 23, 48, 9
+	RESTORE_VRS 24, 64, 9
+	RESTORE_VRS 25, 80, 9
+	RESTORE_VRS 26, 96, 9
+	RESTORE_VRS 27, 112, 9
+	RESTORE_VRS 28, 128, 9
+	RESTORE_VRS 29, 144, 9
+	RESTORE_VRS 30, 160, 9
+	RESTORE_VRS 31, 176, 9
+
+	RESTORE_VSX 14, 192, 9
+	RESTORE_VSX 15, 208, 9
+	RESTORE_VSX 16, 224, 9
+	RESTORE_VSX 17, 240, 9
+	RESTORE_VSX 18, 256, 9
+	RESTORE_VSX 19, 272, 9
+	RESTORE_VSX 20, 288, 9
+	RESTORE_VSX 21, 304, 9
+	RESTORE_VSX 22, 320, 9
+	RESTORE_VSX 23, 336, 9
+	RESTORE_VSX 24, 352, 9
+	RESTORE_VSX 25, 368, 9
+	RESTORE_VSX 26, 384, 9
+	RESTORE_VSX 27, 400, 9
+	RESTORE_VSX 28, 416, 9
+	RESTORE_VSX 29, 432, 9
+	RESTORE_VSX 30, 448, 9
+	RESTORE_VSX 31, 464, 9
+
+	RESTORE_GPR 14, 112, 1
+	RESTORE_GPR 15, 120, 1
+	RESTORE_GPR 16, 128, 1
+	RESTORE_GPR 17, 136, 1
+	RESTORE_GPR 18, 144, 1
+	RESTORE_GPR 19, 152, 1
+	RESTORE_GPR 20, 160, 1
+	RESTORE_GPR 21, 168, 1
+	RESTORE_GPR 22, 176, 1
+	RESTORE_GPR 23, 184, 1
+	RESTORE_GPR 24, 192, 1
+	RESTORE_GPR 25, 200, 1
+	RESTORE_GPR 26, 208, 1
+	RESTORE_GPR 27, 216, 1
+	RESTORE_GPR 28, 224, 1
+	RESTORE_GPR 29, 232, 1
+	RESTORE_GPR 30, 240, 1
+	RESTORE_GPR 31, 248, 1
+
+	addi    1, 1, 752
+	ld 0, 16(1)
+	mtlr 0
+.endm # RESTORE_REGS
+
+#
+# p[0] = a0*r0 + a1*r4*5 + a2*r3*5 + a3*r2*5 + a4*r1*5;
+# p[1] = a0*r1 + a1*r0   + a2*r4*5 + a3*r3*5 + a4*r2*5;
+# p[2] = a0*r2 + a1*r1   + a2*r0   + a3*r4*5 + a4*r3*5;
+# p[3] = a0*r3 + a1*r2   + a2*r1   + a3*r0   + a4*r4*5;
+# p[4] = a0*r4 + a1*r3   + a2*r2   + a3*r1   + a4*r0  ;
+#
+#    [r^2, r^3, r^1, r^4]
+#    [m3,  m2,  m4,  m1]
+#
+# multiply odd and even words
+.macro mul_odd
+	vmulouw	14, 4, 26
+	vmulouw	10, 5, 3
+	vmulouw	11, 6, 2
+	vmulouw	12, 7, 1
+	vmulouw	13, 8, 0
+	vmulouw	15, 4, 27
+	vaddudm	14, 14, 10
+	vaddudm	14, 14, 11
+	vmulouw	10, 5, 26
+	vmulouw	11, 6, 3
+	vaddudm	14, 14, 12
+	vaddudm	14, 14, 13	# x0
+	vaddudm	15, 15, 10
+	vaddudm	15, 15, 11
+	vmulouw	12, 7, 2
+	vmulouw	13, 8, 1
+	vaddudm	15, 15, 12
+	vaddudm	15, 15, 13	# x1
+	vmulouw	16, 4, 28
+	vmulouw	10, 5, 27
+	vmulouw	11, 6, 26
+	vaddudm	16, 16, 10
+	vaddudm	16, 16, 11
+	vmulouw	12, 7, 3
+	vmulouw	13, 8, 2
+	vaddudm	16, 16, 12
+	vaddudm	16, 16, 13	# x2
+	vmulouw	17, 4, 29
+	vmulouw	10, 5, 28
+	vmulouw	11, 6, 27
+	vaddudm	17, 17, 10
+	vaddudm	17, 17, 11
+	vmulouw	12, 7, 26
+	vmulouw	13, 8, 3
+	vaddudm	17, 17, 12
+	vaddudm	17, 17, 13	# x3
+	vmulouw	18, 4, 30
+	vmulouw	10, 5, 29
+	vmulouw	11, 6, 28
+	vaddudm	18, 18, 10
+	vaddudm	18, 18, 11
+	vmulouw	12, 7, 27
+	vmulouw	13, 8, 26
+	vaddudm	18, 18, 12
+	vaddudm	18, 18, 13	# x4
+.endm
+
+.macro mul_even
+	vmuleuw	9, 4, 26
+	vmuleuw	10, 5, 3
+	vmuleuw	11, 6, 2
+	vmuleuw	12, 7, 1
+	vmuleuw	13, 8, 0
+	vaddudm	14, 14, 9
+	vaddudm	14, 14, 10
+	vaddudm	14, 14, 11
+	vaddudm	14, 14, 12
+	vaddudm	14, 14, 13	# x0
+
+	vmuleuw	9, 4, 27
+	vmuleuw	10, 5, 26
+	vmuleuw	11, 6, 3
+	vmuleuw	12, 7, 2
+	vmuleuw	13, 8, 1
+	vaddudm	15, 15, 9
+	vaddudm	15, 15, 10
+	vaddudm	15, 15, 11
+	vaddudm	15, 15, 12
+	vaddudm	15, 15, 13	# x1
+
+	vmuleuw	9, 4, 28
+	vmuleuw	10, 5, 27
+	vmuleuw	11, 6, 26
+	vmuleuw	12, 7, 3
+	vmuleuw	13, 8, 2
+	vaddudm	16, 16, 9
+	vaddudm	16, 16, 10
+	vaddudm	16, 16, 11
+	vaddudm	16, 16, 12
+	vaddudm	16, 16, 13	# x2
+
+	vmuleuw	9, 4, 29
+	vmuleuw	10, 5, 28
+	vmuleuw	11, 6, 27
+	vmuleuw	12, 7, 26
+	vmuleuw	13, 8, 3
+	vaddudm	17, 17, 9
+	vaddudm	17, 17, 10
+	vaddudm	17, 17, 11
+	vaddudm	17, 17, 12
+	vaddudm	17, 17, 13	# x3
+
+	vmuleuw	9, 4, 30
+	vmuleuw	10, 5, 29
+	vmuleuw	11, 6, 28
+	vmuleuw	12, 7, 27
+	vmuleuw	13, 8, 26
+	vaddudm	18, 18, 9
+	vaddudm	18, 18, 10
+	vaddudm	18, 18, 11
+	vaddudm	18, 18, 12
+	vaddudm	18, 18, 13	# x4
+.endm
+
+#
+# poly1305_setup_r
+#
+# setup r^4, r^3, r^2, r vectors
+#    [r, r^3, r^2, r^4]
+#    vs0 = [r0,...]
+#    vs1 = [r1,...]
+#    vs2 = [r2,...]
+#    vs3 = [r3,...]
+#    vs4 = [r4,...]
+#    vs5 = [r4*5,...]
+#    vs6 = [r3*5,...]
+#    vs7 = [r2*5,...]
+#    vs8 = [r1*5,...]
+#
+# r0, r4*5, r3*5, r2*5, r1*5;
+# r1, r0,   r4*5, r3*5, r2*5;
+# r2, r1,   r0,   r4*5, r3*5;
+# r3, r2,   r1,   r0,   r4*5;
+# r4, r3,   r2,   r1,   r0  ;
+#
+.macro poly1305_setup_r
+
+	# save r
+	xxlor	26, 58, 58
+	xxlor	27, 59, 59
+	xxlor	28, 60, 60
+	xxlor	29, 61, 61
+	xxlor	30, 62, 62
+
+	xxlxor	31, 31, 31
+
+#    [r, r^3, r^2, r^4]
+	# compute r^2
+	vmr	4, 26
+	vmr	5, 27
+	vmr	6, 28
+	vmr	7, 29
+	vmr	8, 30
+	bl	do_mul		# r^2 r^1
+	xxpermdi 58, 58, 36, 0x3		# r0
+	xxpermdi 59, 59, 37, 0x3		# r1
+	xxpermdi 60, 60, 38, 0x3		# r2
+	xxpermdi 61, 61, 39, 0x3		# r3
+	xxpermdi 62, 62, 40, 0x3		# r4
+	xxpermdi 36, 36, 36, 0x3
+	xxpermdi 37, 37, 37, 0x3
+	xxpermdi 38, 38, 38, 0x3
+	xxpermdi 39, 39, 39, 0x3
+	xxpermdi 40, 40, 40, 0x3
+	vspltisb 13, 2
+	vsld	9, 27, 13
+	vsld	10, 28, 13
+	vsld	11, 29, 13
+	vsld	12, 30, 13
+	vaddudm	0, 9, 27
+	vaddudm	1, 10, 28
+	vaddudm	2, 11, 29
+	vaddudm	3, 12, 30
+
+	bl	do_mul		# r^4 r^3
+	vmrgow	26, 26, 4
+	vmrgow	27, 27, 5
+	vmrgow	28, 28, 6
+	vmrgow	29, 29, 7
+	vmrgow	30, 30, 8
+	vspltisb 13, 2
+	vsld	9, 27, 13
+	vsld	10, 28, 13
+	vsld	11, 29, 13
+	vsld	12, 30, 13
+	vaddudm	0, 9, 27
+	vaddudm	1, 10, 28
+	vaddudm	2, 11, 29
+	vaddudm	3, 12, 30
+
+	# r^2 r^4
+	xxlor	0, 58, 58
+	xxlor	1, 59, 59
+	xxlor	2, 60, 60
+	xxlor	3, 61, 61
+	xxlor	4, 62, 62
+	xxlor	5, 32, 32
+	xxlor	6, 33, 33
+	xxlor	7, 34, 34
+	xxlor	8, 35, 35
+
+	vspltw	9, 26, 3
+	vspltw	10, 26, 2
+	vmrgow	26, 10, 9
+	vspltw	9, 27, 3
+	vspltw	10, 27, 2
+	vmrgow	27, 10, 9
+	vspltw	9, 28, 3
+	vspltw	10, 28, 2
+	vmrgow	28, 10, 9
+	vspltw	9, 29, 3
+	vspltw	10, 29, 2
+	vmrgow	29, 10, 9
+	vspltw	9, 30, 3
+	vspltw	10, 30, 2
+	vmrgow	30, 10, 9
+
+	vsld	9, 27, 13
+	vsld	10, 28, 13
+	vsld	11, 29, 13
+	vsld	12, 30, 13
+	vaddudm	0, 9, 27
+	vaddudm	1, 10, 28
+	vaddudm	2, 11, 29
+	vaddudm	3, 12, 30
+.endm
+
+SYM_FUNC_START_LOCAL(do_mul)
+	mul_odd
+
+	# do reduction ( h %= p )
+	# carry reduction
+	vspltisb 9, 2
+	vsrd	10, 14, 31
+	vsrd	11, 17, 31
+	vand	7, 17, 25
+	vand	4, 14, 25
+	vaddudm	18, 18, 11
+	vsrd	12, 18, 31
+	vaddudm	15, 15, 10
+
+	vsrd	11, 15, 31
+	vand	8, 18, 25
+	vand	5, 15, 25
+	vaddudm	4, 4, 12
+	vsld	10, 12, 9
+	vaddudm	6, 16, 11
+
+	vsrd	13, 6, 31
+	vand	6, 6, 25
+	vaddudm	4, 4, 10
+	vsrd	10, 4, 31
+	vaddudm	7, 7, 13
+
+	vsrd	11, 7, 31
+	vand	7, 7, 25
+	vand	4, 4, 25
+	vaddudm	5, 5, 10
+	vaddudm	8, 8, 11
+	blr
+SYM_FUNC_END(do_mul)
+
+#
+# init key
+#
+.macro do_poly1305_init
+	addis	10, 2, rmask@toc@ha
+	addi	10, 10, rmask@toc@l
+
+	ld	11, 0(10)
+	ld	12, 8(10)
+
+	li	14, 16
+	li	15, 32
+	addis	10, 2, cnum@toc@ha
+	addi	10, 10, cnum@toc@l
+	lvx	25, 0, 10	# v25 - mask
+	lvx	31, 14, 10	# v31 = 1a
+	lvx	19, 15, 10	# v19 = 1 << 24
+	lxv	24, 48(10)	# vs24
+	lxv	25, 64(10)	# vs25
+
+	# initialize
+	# load key from r3 to vectors
+	ld	9, 24(3)
+	ld	10, 32(3)
+	and.	9, 9, 11
+	and.	10, 10, 12
+
+	# break 26 bits
+	extrdi	14, 9, 26, 38
+	extrdi	15, 9, 26, 12
+	extrdi	16, 9, 12, 0
+	mtvsrdd	58, 0, 14
+	insrdi	16, 10, 14, 38
+	mtvsrdd	59, 0, 15
+	extrdi	17, 10, 26, 24
+	mtvsrdd	60, 0, 16
+	extrdi	18, 10, 24, 0
+	mtvsrdd	61, 0, 17
+	mtvsrdd	62, 0, 18
+
+	# r1 = r1 * 5, r2 = r2 * 5, r3 = r3 * 5, r4 = r4 * 5
+	li	9, 5
+	mtvsrdd	36, 0, 9
+	vmulouw	0, 27, 4		# v0 = rr0
+	vmulouw	1, 28, 4		# v1 = rr1
+	vmulouw	2, 29, 4		# v2 = rr2
+	vmulouw	3, 30, 4		# v3 = rr3
+.endm
+
+#
+# poly1305_p10le_4blocks( uint8_t *k, uint32_t mlen, uint8_t *m)
+#  k = 32 bytes key
+#  r3 = k (r, s)
+#  r4 = mlen
+#  r5 = m
+#
+SYM_FUNC_START(poly1305_p10le_4blocks)
+.align 5
+	cmpdi	5, 64
+	blt	Out_no_poly1305
+
+	SAVE_REGS
+
+	do_poly1305_init
+
+	li	21, 0	# counter to message
+
+	poly1305_setup_r
+
+	# load previous H state
+	# break/convert r6 to 26 bits
+	ld	9, 0(3)
+	ld	10, 8(3)
+	ld	19, 16(3)
+	sldi	19, 19, 24
+	mtvsrdd	41, 0, 19
+	extrdi	14, 9, 26, 38
+	extrdi	15, 9, 26, 12
+	extrdi	16, 9, 12, 0
+	mtvsrdd	36, 0, 14
+	insrdi	16, 10, 14, 38
+	mtvsrdd	37, 0, 15
+	extrdi	17, 10, 26, 24
+	mtvsrdd	38, 0, 16
+	extrdi	18, 10, 24, 0
+	mtvsrdd	39, 0, 17
+	mtvsrdd	40, 0, 18
+	vor	8, 8, 9
+
+	# input m1 m2
+	add	20, 4, 21
+	xxlor	49, 24, 24
+	xxlor	50, 25, 25
+	lxvw4x	43, 0, 20
+	addi	17, 20, 16
+	lxvw4x	44, 0, 17
+	vperm	14, 11, 12, 17
+	vperm	15, 11, 12, 18
+	vand	9, 14, 25	# a0
+	vsrd	10, 14, 31	# >> 26
+	vsrd	11, 10, 31	# 12 bits left
+	vand	10, 10, 25	# a1
+	vspltisb 13, 12
+	vand	16, 15, 25
+	vsld	12, 16, 13
+	vor	11, 11, 12
+	vand	11, 11, 25	# a2
+	vspltisb 13, 14
+	vsrd	12, 15, 13	# >> 14
+	vsrd	13, 12, 31	# >> 26, a4
+	vand	12, 12, 25	# a3
+
+	vaddudm	20, 4, 9
+	vaddudm	21, 5, 10
+	vaddudm	22, 6, 11
+	vaddudm	23, 7, 12
+	vaddudm	24, 8, 13
+
+	# m3 m4
+	addi	17, 17, 16
+	lxvw4x	43, 0, 17
+	addi	17, 17, 16
+	lxvw4x	44, 0, 17
+	vperm	14, 11, 12, 17
+	vperm	15, 11, 12, 18
+	vand	9, 14, 25	# a0
+	vsrd	10, 14, 31	# >> 26
+	vsrd	11, 10, 31	# 12 bits left
+	vand	10, 10, 25	# a1
+	vspltisb 13, 12
+	vand	16, 15, 25
+	vsld	12, 16, 13
+	vspltisb 13, 14
+	vor	11, 11, 12
+	vand	11, 11, 25	# a2
+	vsrd	12, 15, 13	# >> 14
+	vsrd	13, 12, 31	# >> 26, a4
+	vand	12, 12, 25	# a3
+
+	# Smash 4 message blocks into 5 vectors of [m4,  m2,  m3,  m1]
+	vmrgow	4, 9, 20
+	vmrgow	5, 10, 21
+	vmrgow	6, 11, 22
+	vmrgow	7, 12, 23
+	vmrgow	8, 13, 24
+	vaddudm	8, 8, 19
+
+	addi	5, 5, -64	# len -= 64
+	addi	21, 21, 64	# offset += 64
+
+	li      9, 64
+	divdu   31, 5, 9
+
+	cmpdi	31, 0
+	ble	Skip_block_loop
+
+	mtctr	31
+
+# h4 =   m1 * r⁴ + m2 * r³ + m3 * r² + m4 * r
+# Rewrite the polynominal sum of product as follows,
+# h1 = (h0 + m1) * r^2,	h2 = (h0 + m2) * r^2
+# h3 = (h1 + m3) * r^2,	h4 = (h2 + m4) * r^2  --> (h0 + m1) r*4 + (h3 + m3) r^2, (h0 + m2) r^4 + (h0 + m4) r^2
+#  .... Repeat
+# h5 = (h3 + m5) * r^2,	h6 = (h4 + m6) * r^2  -->
+# h7 = (h5 + m7) * r^2,	h8 = (h6 + m8) * r^1  --> m5 * r^4 + m6 * r^3 + m7 * r^2 + m8 * r
+#
+loop_4blocks:
+
+	# Multiply odd words and even words
+	mul_odd
+	mul_even
+	# carry reduction
+	vspltisb 9, 2
+	vsrd	10, 14, 31
+	vsrd	11, 17, 31
+	vand	7, 17, 25
+	vand	4, 14, 25
+	vaddudm	18, 18, 11
+	vsrd	12, 18, 31
+	vaddudm	15, 15, 10
+
+	vsrd	11, 15, 31
+	vand	8, 18, 25
+	vand	5, 15, 25
+	vaddudm	4, 4, 12
+	vsld	10, 12, 9
+	vaddudm	6, 16, 11
+
+	vsrd	13, 6, 31
+	vand	6, 6, 25
+	vaddudm	4, 4, 10
+	vsrd	10, 4, 31
+	vaddudm	7, 7, 13
+
+	vsrd	11, 7, 31
+	vand	7, 7, 25
+	vand	4, 4, 25
+	vaddudm	5, 5, 10
+	vaddudm	8, 8, 11
+
+	# input m1  m2  m3  m4
+	add	20, 4, 21
+	xxlor	49, 24, 24
+	xxlor	50, 25, 25
+	lxvw4x	43, 0, 20
+	addi	17, 20, 16
+	lxvw4x	44, 0, 17
+	vperm	14, 11, 12, 17
+	vperm	15, 11, 12, 18
+	addi	17, 17, 16
+	lxvw4x	43, 0, 17
+	addi	17, 17, 16
+	lxvw4x	44, 0, 17
+	vperm	17, 11, 12, 17
+	vperm	18, 11, 12, 18
+
+	vand	20, 14, 25	# a0
+	vand	9, 17, 25	# a0
+	vsrd	21, 14, 31	# >> 26
+	vsrd	22, 21, 31	# 12 bits left
+	vsrd	10, 17, 31	# >> 26
+	vsrd	11, 10, 31	# 12 bits left
+
+	vand	21, 21, 25	# a1
+	vand	10, 10, 25	# a1
+
+	vspltisb 13, 12
+	vand	16, 15, 25
+	vsld	23, 16, 13
+	vor	22, 22, 23
+	vand	22, 22, 25	# a2
+	vand	16, 18, 25
+	vsld	12, 16, 13
+	vor	11, 11, 12
+	vand	11, 11, 25	# a2
+	vspltisb 13, 14
+	vsrd	23, 15, 13	# >> 14
+	vsrd	24, 23, 31	# >> 26, a4
+	vand	23, 23, 25	# a3
+	vsrd	12, 18, 13	# >> 14
+	vsrd	13, 12, 31	# >> 26, a4
+	vand	12, 12, 25	# a3
+
+	vaddudm	4, 4, 20
+	vaddudm	5, 5, 21
+	vaddudm	6, 6, 22
+	vaddudm	7, 7, 23
+	vaddudm	8, 8, 24
+
+	# Smash 4 message blocks into 5 vectors of [m4,  m2,  m3,  m1]
+	vmrgow	4, 9, 4
+	vmrgow	5, 10, 5
+	vmrgow	6, 11, 6
+	vmrgow	7, 12, 7
+	vmrgow	8, 13, 8
+	vaddudm	8, 8, 19
+
+	addi	5, 5, -64	# len -= 64
+	addi	21, 21, 64	# offset += 64
+
+	bdnz	loop_4blocks
+
+Skip_block_loop:
+	xxlor	58, 0, 0
+	xxlor	59, 1, 1
+	xxlor	60, 2, 2
+	xxlor	61, 3, 3
+	xxlor	62, 4, 4
+	xxlor	32, 5, 5
+	xxlor	33, 6, 6
+	xxlor	34, 7, 7
+	xxlor	35, 8, 8
+
+	# Multiply odd words and even words
+	mul_odd
+	mul_even
+
+	# Sum the products.
+	xxpermdi 41, 31, 46, 0
+	xxpermdi 42, 31, 47, 0
+	vaddudm	4, 14, 9
+	xxpermdi 36, 31, 36, 3
+	vaddudm	5, 15, 10
+	xxpermdi 37, 31, 37, 3
+	xxpermdi 43, 31, 48, 0
+	vaddudm	6, 16, 11
+	xxpermdi 38, 31, 38, 3
+	xxpermdi 44, 31, 49, 0
+	vaddudm	7, 17, 12
+	xxpermdi 39, 31, 39, 3
+	xxpermdi 45, 31, 50, 0
+	vaddudm	8, 18, 13
+	xxpermdi 40, 31, 40, 3
+
+	# carry reduction
+	vspltisb 9, 2
+	vsrd	10, 4, 31
+	vsrd	11, 7, 31
+	vand	7, 7, 25
+	vand	4, 4, 25
+	vaddudm	8, 8, 11
+	vsrd	12, 8, 31
+	vaddudm	5, 5, 10
+
+	vsrd	11, 5, 31
+	vand	8, 8, 25
+	vand	5, 5, 25
+	vaddudm	4, 4, 12
+	vsld	10, 12, 9
+	vaddudm	6, 6, 11
+
+	vsrd	13, 6, 31
+	vand	6, 6, 25
+	vaddudm	4, 4, 10
+	vsrd	10, 4, 31
+	vaddudm	7, 7, 13
+
+	vsrd	11, 7, 31
+	vand	7, 7, 25
+	vand	4, 4, 25
+	vaddudm	5, 5, 10
+	vsrd	10, 5, 31
+	vand	5, 5, 25
+	vaddudm	6, 6, 10
+	vaddudm	8, 8, 11
+
+	b	do_final_update
+
+do_final_update:
+	# combine 26 bit limbs
+	# v4, v5, v6, v7 and v8 are 26 bit vectors
+	vsld	5, 5, 31
+	vor	20, 4, 5
+	vspltisb 11, 12
+	vsrd	12, 6, 11
+	vsld	6, 6, 31
+	vsld	6, 6, 31
+	vor	20, 20, 6
+	vspltisb 11, 14
+	vsld	7, 7, 11
+	vor	21, 7, 12
+	mfvsrld	16, 40		# save last 2 bytes
+	vsld	8, 8, 11
+	vsld	8, 8, 31
+	vor	21, 21, 8
+	mfvsrld	17, 52
+	mfvsrld	19, 53
+	srdi	16, 16, 24
+
+	std	17, 0(3)
+	std	19, 8(3)
+	stw	16, 16(3)
+
+Out_loop:
+	li	3, 0
+
+	RESTORE_REGS
+
+	blr
+
+Out_no_poly1305:
+	li	3, 0
+	blr
+SYM_FUNC_END(poly1305_p10le_4blocks)
+
+#
+# =======================================================================
+# The following functions implement 64 x 64 bits multiplication poly1305.
+#
+SYM_FUNC_START_LOCAL(Poly1305_init_64)
+	#  mask 0x0FFFFFFC0FFFFFFC
+	#  mask 0x0FFFFFFC0FFFFFFF
+	addis	10, 2, rmask@toc@ha
+	addi	10, 10, rmask@toc@l
+	ld	11, 0(10)
+	ld	12, 8(10)
+
+	# initialize
+	# load key from r3
+	ld	9, 24(3)
+	ld	10, 32(3)
+	and.	9, 9, 11	# cramp mask r0
+	and.	10, 10, 12	# cramp mask r1
+
+        srdi    21, 10, 2
+        add     19, 21, 10      # s1: r19 - (r1 >> 2) *5
+
+        # setup r and s
+        li      25, 0
+	mtvsrdd 32+0, 9, 19	# r0, s1
+	mtvsrdd 32+1, 10, 9	# r1, r0
+	mtvsrdd 32+2, 19, 25	# s1
+	mtvsrdd 32+3, 9, 25	# r0
+
+	blr
+SYM_FUNC_END(Poly1305_init_64)
+
+# Poly1305_mult
+# v6 = (h0, h1), v8 = h2
+# v0 = (r0, s1), v1 = (r1, r0), v2 = s1, v3 = r0
+#
+# Output: v7, v10, v11
+#
+SYM_FUNC_START_LOCAL(Poly1305_mult)
+	#
+	#	d0 = h0 * r0 + h1 * s1
+	vmsumudm	7, 6, 0, 9		# h0 * r0, h1 * s1
+
+	#	d1 = h0 * r1 + h1 * r0 + h2 * s1
+	vmsumudm	11, 6, 1, 9		# h0 * r1, h1 * r0
+	vmsumudm	10, 8, 2, 11		# d1 += h2 * s1
+
+	#       d2 = r0
+	vmsumudm	11, 8, 3, 9		# d2 = h2 * r0
+	blr
+SYM_FUNC_END(Poly1305_mult)
+
+#
+# carry reduction
+# h %=p
+#
+# Input: v7, v10, v11
+# Output: r27, r28, r29
+#
+SYM_FUNC_START_LOCAL(Carry_reduction)
+	mfvsrld	27, 32+7
+	mfvsrld	28, 32+10
+	mfvsrld	29, 32+11
+	mfvsrd	20, 32+7	# h0.h
+	mfvsrd	21, 32+10	# h1.h
+
+	addc	28, 28, 20
+	adde	29, 29, 21
+	srdi	22, 29, 0x2
+	sldi	23, 22, 0x2
+	add	23, 23, 22	# (h2 & 3) * 5
+	addc	27, 27, 23	# h0
+	addze	28, 28		# h1
+	andi.	29, 29, 0x3	# h2
+	blr
+SYM_FUNC_END(Carry_reduction)
+
+#
+# poly1305 multiplication
+# h *= r, h %= p
+#	d0 = h0 * r0 + h1 * s1
+#	d1 = h0 * r1 + h1 * r0 + h2 * s1
+#       d2 = h0 * r0
+#
+#
+# unsigned int poly1305_test_64s(unisgned char *state, const byte *src, size_t len, highbit)
+#   - no highbit if final leftover block (highbit = 0)
+#
+SYM_FUNC_START(poly1305_64s)
+	cmpdi	5, 0
+	ble	Out_no_poly1305_64
+
+	mflr 0
+	std 0, 16(1)
+	stdu 1,-400(1)
+
+	SAVE_GPR 14, 112, 1
+	SAVE_GPR 15, 120, 1
+	SAVE_GPR 16, 128, 1
+	SAVE_GPR 17, 136, 1
+	SAVE_GPR 18, 144, 1
+	SAVE_GPR 19, 152, 1
+	SAVE_GPR 20, 160, 1
+	SAVE_GPR 21, 168, 1
+	SAVE_GPR 22, 176, 1
+	SAVE_GPR 23, 184, 1
+	SAVE_GPR 24, 192, 1
+	SAVE_GPR 25, 200, 1
+	SAVE_GPR 26, 208, 1
+	SAVE_GPR 27, 216, 1
+	SAVE_GPR 28, 224, 1
+	SAVE_GPR 29, 232, 1
+	SAVE_GPR 30, 240, 1
+	SAVE_GPR 31, 248, 1
+
+	# Init poly1305
+	bl Poly1305_init_64
+
+	li 25, 0			# offset to inp and outp
+
+	add 11, 25, 4
+
+	# load h
+	# h0, h1, h2?
+        ld	27, 0(3)
+        ld	28, 8(3)
+        lwz	29, 16(3)
+
+        li      30, 16
+        divdu   31, 5, 30
+
+        mtctr   31
+
+        mr      24, 6		# highbit
+
+Loop_block_64:
+	vxor	9, 9, 9
+
+	ld	20, 0(11)
+	ld	21, 8(11)
+	addi	11, 11, 16
+
+	addc	27, 27, 20
+	adde	28, 28, 21
+	adde	29, 29, 24
+
+	li	22, 0
+	mtvsrdd	32+6, 27, 28	# h0, h1
+	mtvsrdd	32+8, 29, 22	# h2
+
+	bl	Poly1305_mult
+
+	bl	Carry_reduction
+
+	bdnz	Loop_block_64
+
+	std	27, 0(3)
+	std	28, 8(3)
+	stw	29, 16(3)
+
+	li	3, 0
+
+	RESTORE_GPR 14, 112, 1
+	RESTORE_GPR 15, 120, 1
+	RESTORE_GPR 16, 128, 1
+	RESTORE_GPR 17, 136, 1
+	RESTORE_GPR 18, 144, 1
+	RESTORE_GPR 19, 152, 1
+	RESTORE_GPR 20, 160, 1
+	RESTORE_GPR 21, 168, 1
+	RESTORE_GPR 22, 176, 1
+	RESTORE_GPR 23, 184, 1
+	RESTORE_GPR 24, 192, 1
+	RESTORE_GPR 25, 200, 1
+	RESTORE_GPR 26, 208, 1
+	RESTORE_GPR 27, 216, 1
+	RESTORE_GPR 28, 224, 1
+	RESTORE_GPR 29, 232, 1
+	RESTORE_GPR 30, 240, 1
+	RESTORE_GPR 31, 248, 1
+
+	addi    1, 1, 400
+	ld 0, 16(1)
+	mtlr 0
+
+	blr
+
+Out_no_poly1305_64:
+	li	3, 0
+	blr
+SYM_FUNC_END(poly1305_64s)
+
+#
+# Input: r3 = h, r4 = s, r5 = mac
+# mac = h + s
+#
+SYM_FUNC_START(poly1305_emit_64)
+	ld	10, 0(3)
+	ld	11, 8(3)
+	ld	12, 16(3)
+
+	# compare modulus
+	# h + 5 + (-p)
+	mr	6, 10
+	mr	7, 11
+	mr	8, 12
+	addic.	6, 6, 5
+	addze	7, 7
+	addze	8, 8
+	srdi	9, 8, 2		# overflow?
+	cmpdi	9, 0
+	beq	Skip_h64
+	mr	10, 6
+	mr	11, 7
+	mr	12, 8
+
+Skip_h64:
+	ld	6, 0(4)
+	ld	7, 8(4)
+	addc	10, 10, 6
+	adde	11, 11, 7
+	addze	12, 12
+
+	std	10, 0(5)
+	std	11, 8(5)
+	blr
+SYM_FUNC_END(poly1305_emit_64)
+
+SYM_DATA_START_LOCAL(RMASK)
+.align 5
+rmask:
+.byte	0xff, 0xff, 0xff, 0x0f, 0xfc, 0xff, 0xff, 0x0f, 0xfc, 0xff, 0xff, 0x0f, 0xfc, 0xff, 0xff, 0x0f
+cnum:
+.long	0x03ffffff, 0x00000000, 0x03ffffff, 0x00000000
+.long	0x1a, 0x00, 0x1a, 0x00
+.long	0x01000000, 0x01000000, 0x01000000, 0x01000000
+.long	0x00010203, 0x04050607, 0x10111213, 0x14151617
+.long	0x08090a0b, 0x0c0d0e0f, 0x18191a1b, 0x1c1d1e1f
+SYM_DATA_END(RMASK)
diff --git a/lib/crypto/powerpc/poly1305.h b/lib/crypto/powerpc/poly1305.h
new file mode 100644
index 000000000..b8ed098a0
--- /dev/null
+++ b/lib/crypto/powerpc/poly1305.h
@@ -0,0 +1,74 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Poly1305 authenticator algorithm, RFC7539.
+ *
+ * Copyright 2023- IBM Corp. All rights reserved.
+ */
+#include <asm/switch_to.h>
+#include <linux/cpufeature.h>
+#include <linux/jump_label.h>
+#include <linux/kernel.h>
+#include <linux/unaligned.h>
+
+asmlinkage void poly1305_p10le_4blocks(struct poly1305_block_state *state, const u8 *m, u32 mlen);
+asmlinkage void poly1305_64s(struct poly1305_block_state *state, const u8 *m, u32 mlen, int highbit);
+asmlinkage void poly1305_emit_64(const struct poly1305_state *state, const u32 nonce[4], u8 digest[POLY1305_DIGEST_SIZE]);
+
+static __ro_after_init DEFINE_STATIC_KEY_FALSE(have_p10);
+
+static void vsx_begin(void)
+{
+	preempt_disable();
+	enable_kernel_vsx();
+}
+
+static void vsx_end(void)
+{
+	disable_kernel_vsx();
+	preempt_enable();
+}
+
+static void poly1305_block_init(struct poly1305_block_state *dctx,
+				const u8 raw_key[POLY1305_BLOCK_SIZE])
+{
+	if (!static_key_enabled(&have_p10))
+		return poly1305_block_init_generic(dctx, raw_key);
+
+	dctx->h = (struct poly1305_state){};
+	dctx->core_r.key.r64[0] = get_unaligned_le64(raw_key + 0);
+	dctx->core_r.key.r64[1] = get_unaligned_le64(raw_key + 8);
+}
+
+static void poly1305_blocks(struct poly1305_block_state *state, const u8 *src,
+			    unsigned int len, u32 padbit)
+{
+	if (!static_key_enabled(&have_p10))
+		return poly1305_blocks_generic(state, src, len, padbit);
+	vsx_begin();
+	if (len >= POLY1305_BLOCK_SIZE * 4) {
+		poly1305_p10le_4blocks(state, src, len);
+		src += len - (len % (POLY1305_BLOCK_SIZE * 4));
+		len %= POLY1305_BLOCK_SIZE * 4;
+	}
+	while (len >= POLY1305_BLOCK_SIZE) {
+		poly1305_64s(state, src, POLY1305_BLOCK_SIZE, padbit);
+		len -= POLY1305_BLOCK_SIZE;
+		src += POLY1305_BLOCK_SIZE;
+	}
+	vsx_end();
+}
+
+static void poly1305_emit(const struct poly1305_state *state,
+			  u8 digest[POLY1305_DIGEST_SIZE], const u32 nonce[4])
+{
+	if (!static_key_enabled(&have_p10))
+		return poly1305_emit_generic(state, digest, nonce);
+	poly1305_emit_64(state, nonce, digest);
+}
+
+#define poly1305_mod_init_arch poly1305_mod_init_arch
+static void poly1305_mod_init_arch(void)
+{
+	if (cpu_has_feature(CPU_FTR_ARCH_31))
+		static_branch_enable(&have_p10);
+}
diff --git a/lib/crypto/powerpc/sha1-powerpc-asm.S b/lib/crypto/powerpc/sha1-powerpc-asm.S
new file mode 100644
index 000000000..f0d5ed557
--- /dev/null
+++ b/lib/crypto/powerpc/sha1-powerpc-asm.S
@@ -0,0 +1,188 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * SHA-1 implementation for PowerPC.
+ *
+ * Copyright (C) 2005 Paul Mackerras <paulus@samba.org>
+ */
+
+#include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
+#include <asm/asm-compat.h>
+
+#ifdef __BIG_ENDIAN__
+#define LWZ(rt, d, ra)	\
+	lwz	rt,d(ra)
+#else
+#define LWZ(rt, d, ra)	\
+	li	rt,d;	\
+	lwbrx	rt,rt,ra
+#endif
+
+/*
+ * We roll the registers for T, A, B, C, D, E around on each
+ * iteration; T on iteration t is A on iteration t+1, and so on.
+ * We use registers 7 - 12 for this.
+ */
+#define RT(t)	((((t)+5)%6)+7)
+#define RA(t)	((((t)+4)%6)+7)
+#define RB(t)	((((t)+3)%6)+7)
+#define RC(t)	((((t)+2)%6)+7)
+#define RD(t)	((((t)+1)%6)+7)
+#define RE(t)	((((t)+0)%6)+7)
+
+/* We use registers 16 - 31 for the W values */
+#define W(t)	(((t)%16)+16)
+
+#define LOADW(t)				\
+	LWZ(W(t),(t)*4,r4)
+
+#define STEPD0_LOAD(t)				\
+	andc	r0,RD(t),RB(t);		\
+	and	r6,RB(t),RC(t);		\
+	rotlwi	RT(t),RA(t),5;			\
+	or	r6,r6,r0;			\
+	add	r0,RE(t),r15;			\
+	add	RT(t),RT(t),r6;		\
+	add	r14,r0,W(t);			\
+	LWZ(W((t)+4),((t)+4)*4,r4);	\
+	rotlwi	RB(t),RB(t),30;			\
+	add	RT(t),RT(t),r14
+
+#define STEPD0_UPDATE(t)			\
+	and	r6,RB(t),RC(t);		\
+	andc	r0,RD(t),RB(t);		\
+	rotlwi	RT(t),RA(t),5;			\
+	rotlwi	RB(t),RB(t),30;			\
+	or	r6,r6,r0;			\
+	add	r0,RE(t),r15;			\
+	xor	r5,W((t)+4-3),W((t)+4-8);		\
+	add	RT(t),RT(t),r6;		\
+	xor	W((t)+4),W((t)+4-16),W((t)+4-14);	\
+	add	r0,r0,W(t);			\
+	xor	W((t)+4),W((t)+4),r5;			\
+	add	RT(t),RT(t),r0;		\
+	rotlwi	W((t)+4),W((t)+4),1
+
+#define STEPD1(t)				\
+	xor	r6,RB(t),RC(t);		\
+	rotlwi	RT(t),RA(t),5;			\
+	rotlwi	RB(t),RB(t),30;			\
+	xor	r6,r6,RD(t);			\
+	add	r0,RE(t),r15;			\
+	add	RT(t),RT(t),r6;		\
+	add	r0,r0,W(t);			\
+	add	RT(t),RT(t),r0
+
+#define STEPD1_UPDATE(t)				\
+	xor	r6,RB(t),RC(t);		\
+	rotlwi	RT(t),RA(t),5;			\
+	rotlwi	RB(t),RB(t),30;			\
+	xor	r6,r6,RD(t);			\
+	add	r0,RE(t),r15;			\
+	xor	r5,W((t)+4-3),W((t)+4-8);		\
+	add	RT(t),RT(t),r6;		\
+	xor	W((t)+4),W((t)+4-16),W((t)+4-14);	\
+	add	r0,r0,W(t);			\
+	xor	W((t)+4),W((t)+4),r5;			\
+	add	RT(t),RT(t),r0;		\
+	rotlwi	W((t)+4),W((t)+4),1
+
+#define STEPD2_UPDATE(t)			\
+	and	r6,RB(t),RC(t);		\
+	and	r0,RB(t),RD(t);		\
+	rotlwi	RT(t),RA(t),5;			\
+	or	r6,r6,r0;			\
+	rotlwi	RB(t),RB(t),30;			\
+	and	r0,RC(t),RD(t);		\
+	xor	r5,W((t)+4-3),W((t)+4-8);	\
+	or	r6,r6,r0;			\
+	xor	W((t)+4),W((t)+4-16),W((t)+4-14);	\
+	add	r0,RE(t),r15;			\
+	add	RT(t),RT(t),r6;		\
+	add	r0,r0,W(t);			\
+	xor	W((t)+4),W((t)+4),r5;		\
+	add	RT(t),RT(t),r0;		\
+	rotlwi	W((t)+4),W((t)+4),1
+
+#define STEP0LD4(t)				\
+	STEPD0_LOAD(t);				\
+	STEPD0_LOAD((t)+1);			\
+	STEPD0_LOAD((t)+2);			\
+	STEPD0_LOAD((t)+3)
+
+#define STEPUP4(t, fn)				\
+	STEP##fn##_UPDATE(t);			\
+	STEP##fn##_UPDATE((t)+1);		\
+	STEP##fn##_UPDATE((t)+2);		\
+	STEP##fn##_UPDATE((t)+3)
+
+#define STEPUP20(t, fn)				\
+	STEPUP4(t, fn);				\
+	STEPUP4((t)+4, fn);			\
+	STEPUP4((t)+8, fn);			\
+	STEPUP4((t)+12, fn);			\
+	STEPUP4((t)+16, fn)
+
+_GLOBAL(powerpc_sha_transform)
+	PPC_STLU r1,-INT_FRAME_SIZE(r1)
+	SAVE_GPRS(14, 31, r1)
+
+	/* Load up A - E */
+	lwz	RA(0),0(r3)	/* A */
+	lwz	RB(0),4(r3)	/* B */
+	lwz	RC(0),8(r3)	/* C */
+	lwz	RD(0),12(r3)	/* D */
+	lwz	RE(0),16(r3)	/* E */
+
+	LOADW(0)
+	LOADW(1)
+	LOADW(2)
+	LOADW(3)
+
+	lis	r15,0x5a82	/* K0-19 */
+	ori	r15,r15,0x7999
+	STEP0LD4(0)
+	STEP0LD4(4)
+	STEP0LD4(8)
+	STEPUP4(12, D0)
+	STEPUP4(16, D0)
+
+	lis	r15,0x6ed9	/* K20-39 */
+	ori	r15,r15,0xeba1
+	STEPUP20(20, D1)
+
+	lis	r15,0x8f1b	/* K40-59 */
+	ori	r15,r15,0xbcdc
+	STEPUP20(40, D2)
+
+	lis	r15,0xca62	/* K60-79 */
+	ori	r15,r15,0xc1d6
+	STEPUP4(60, D1)
+	STEPUP4(64, D1)
+	STEPUP4(68, D1)
+	STEPUP4(72, D1)
+	lwz	r20,16(r3)
+	STEPD1(76)
+	lwz	r19,12(r3)
+	STEPD1(77)
+	lwz	r18,8(r3)
+	STEPD1(78)
+	lwz	r17,4(r3)
+	STEPD1(79)
+
+	lwz	r16,0(r3)
+	add	r20,RE(80),r20
+	add	RD(0),RD(80),r19
+	add	RC(0),RC(80),r18
+	add	RB(0),RB(80),r17
+	add	RA(0),RA(80),r16
+	mr	RE(0),r20
+	stw	RA(0),0(r3)
+	stw	RB(0),4(r3)
+	stw	RC(0),8(r3)
+	stw	RD(0),12(r3)
+	stw	RE(0),16(r3)
+
+	REST_GPRS(14, 31, r1)
+	addi	r1,r1,INT_FRAME_SIZE
+	blr
diff --git a/lib/crypto/powerpc/sha1-spe-asm.S b/lib/crypto/powerpc/sha1-spe-asm.S
new file mode 100644
index 000000000..0f447523b
--- /dev/null
+++ b/lib/crypto/powerpc/sha1-spe-asm.S
@@ -0,0 +1,294 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Fast SHA-1 implementation for SPE instruction set (PPC)
+ *
+ * This code makes use of the SPE SIMD instruction set as defined in
+ * http://cache.freescale.com/files/32bit/doc/ref_manual/SPEPIM.pdf
+ * Implementation is based on optimization guide notes from
+ * http://cache.freescale.com/files/32bit/doc/app_note/AN2665.pdf
+ *
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ */
+
+#include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
+
+#define rHP	r3	/* pointer to hash value			*/
+#define rWP	r4	/* pointer to input				*/
+#define rKP	r5	/* pointer to constants				*/
+
+#define rW0	r14	/* 64 bit round words				*/
+#define rW1	r15
+#define rW2	r16
+#define rW3	r17
+#define rW4	r18
+#define rW5	r19
+#define rW6	r20
+#define rW7	r21
+
+#define rH0	r6	/* 32 bit hash values 				*/
+#define rH1	r7
+#define rH2	r8
+#define rH3	r9
+#define rH4	r10
+
+#define rT0	r22	/* 64 bit temporary				*/
+#define rT1	r0	/* 32 bit temporaries				*/
+#define rT2	r11
+#define rT3	r12
+
+#define rK	r23	/* 64 bit constant in volatile register		*/
+
+#define LOAD_K01
+
+#define LOAD_K11 \
+	evlwwsplat	rK,0(rKP);
+
+#define LOAD_K21 \
+	evlwwsplat	rK,4(rKP);
+
+#define LOAD_K31 \
+	evlwwsplat	rK,8(rKP);
+
+#define LOAD_K41 \
+	evlwwsplat	rK,12(rKP);
+
+#define INITIALIZE \
+	stwu		r1,-128(r1);	/* create stack frame		*/ \
+	evstdw		r14,8(r1);	/* We must save non volatile	*/ \
+	evstdw		r15,16(r1);	/* registers. Take the chance	*/ \
+	evstdw		r16,24(r1);	/* and save the SPE part too	*/ \
+	evstdw		r17,32(r1);					   \
+	evstdw		r18,40(r1);					   \
+	evstdw		r19,48(r1);					   \
+	evstdw		r20,56(r1);					   \
+	evstdw		r21,64(r1);					   \
+	evstdw		r22,72(r1);					   \
+	evstdw		r23,80(r1);
+
+
+#define FINALIZE \
+	evldw		r14,8(r1);	/* restore SPE registers	*/ \
+	evldw		r15,16(r1);					   \
+	evldw		r16,24(r1);					   \
+	evldw		r17,32(r1);					   \
+	evldw		r18,40(r1);					   \
+	evldw		r19,48(r1);					   \
+	evldw		r20,56(r1);					   \
+	evldw		r21,64(r1);					   \
+	evldw		r22,72(r1);					   \
+	evldw		r23,80(r1);					   \
+	xor		r0,r0,r0;					   \
+	stw		r0,8(r1);	/* Delete sensitive data	*/ \
+	stw		r0,16(r1);	/* that we might have pushed	*/ \
+	stw		r0,24(r1);	/* from other context that runs	*/ \
+	stw		r0,32(r1);	/* the same code. Assume that	*/ \
+	stw		r0,40(r1);	/* the lower part of the GPRs	*/ \
+	stw		r0,48(r1);	/* were already overwritten on	*/ \
+	stw		r0,56(r1);	/* the way down to here		*/ \
+	stw		r0,64(r1);					   \
+	stw		r0,72(r1);					   \
+	stw		r0,80(r1);					   \
+	addi		r1,r1,128;	/* cleanup stack frame		*/
+
+#ifdef __BIG_ENDIAN__
+#define LOAD_DATA(reg, off) \
+	lwz		reg,off(rWP);	/* load data			*/
+#define NEXT_BLOCK \
+	addi		rWP,rWP,64;	/* increment per block		*/
+#else
+#define LOAD_DATA(reg, off) \
+	lwbrx		reg,0,rWP;	/* load data			*/ \
+	addi		rWP,rWP,4;	/* increment per word		*/
+#define NEXT_BLOCK			/* nothing to do		*/
+#endif
+
+#define	R_00_15(a, b, c, d, e, w0, w1, k, off) \
+	LOAD_DATA(w0, off)		/* 1: W				*/ \
+	and		rT2,b,c;	/* 1: F' = B and C 		*/ \
+	LOAD_K##k##1							   \
+	andc		rT1,d,b;	/* 1: F" = ~B and D 		*/ \
+	rotrwi		rT0,a,27;	/* 1: A' = A rotl 5		*/ \
+	or		rT2,rT2,rT1;	/* 1: F = F' or F"		*/ \
+	add		e,e,rT0;	/* 1: E = E + A'		*/ \
+	rotrwi		b,b,2;		/* 1: B = B rotl 30		*/ \
+	add		e,e,w0;		/* 1: E = E + W			*/ \
+	LOAD_DATA(w1, off+4)		/* 2: W				*/ \
+	add		e,e,rT2;	/* 1: E = E + F			*/ \
+	and		rT1,a,b;	/* 2: F' = B and C 		*/ \
+	add		e,e,rK;		/* 1: E = E + K			*/ \
+	andc		rT2,c,a;	/* 2: F" = ~B and D 		*/ \
+	add		d,d,rK;		/* 2: E = E + K			*/ \
+	or		rT2,rT2,rT1;	/* 2: F = F' or F"		*/ \
+	rotrwi		rT0,e,27;	/* 2: A' = A rotl 5		*/ \
+	add		d,d,w1;		/* 2: E = E + W			*/ \
+	rotrwi		a,a,2;		/* 2: B = B rotl 30		*/ \
+	add		d,d,rT0;	/* 2: E = E + A'		*/ \
+	evmergelo	w1,w1,w0;	/*    mix W[0]/W[1]		*/ \
+	add		d,d,rT2		/* 2: E = E + F			*/
+
+#define R_16_19(a, b, c, d, e, w0, w1, w4, w6, w7, k) \
+	and		rT2,b,c;	/* 1: F' = B and C 		*/ \
+	evmergelohi	rT0,w7,w6;	/*    W[-3]			*/ \
+	andc		rT1,d,b;	/* 1: F" = ~B and D 		*/ \
+	evxor		w0,w0,rT0;	/*    W = W[-16] xor W[-3]	*/ \
+	or		rT1,rT1,rT2;	/* 1: F = F' or F"		*/ \
+	evxor		w0,w0,w4;	/*    W = W xor W[-8]		*/ \
+	add		e,e,rT1;	/* 1: E = E + F			*/ \
+	evxor		w0,w0,w1;	/*    W = W xor W[-14]		*/ \
+	rotrwi		rT2,a,27;	/* 1: A' = A rotl 5		*/ \
+	evrlwi		w0,w0,1;	/*    W = W rotl 1		*/ \
+	add		e,e,rT2;	/* 1: E = E + A'		*/ \
+	evaddw		rT0,w0,rK;	/*    WK = W + K		*/ \
+	rotrwi		b,b,2;		/* 1: B = B rotl 30		*/ \
+	LOAD_K##k##1							   \
+	evmergehi	rT1,rT1,rT0;	/*    WK1/WK2			*/ \
+	add		e,e,rT0;	/* 1: E = E + WK		*/ \
+	add		d,d,rT1;	/* 2: E = E + WK		*/ \
+	and		rT2,a,b;	/* 2: F' = B and C 		*/ \
+	andc		rT1,c,a;	/* 2: F" = ~B and D 		*/ \
+	rotrwi		rT0,e,27;	/* 2: A' = A rotl 5		*/ \
+	or		rT1,rT1,rT2;	/* 2: F = F' or F"		*/ \
+	add		d,d,rT0;	/* 2: E = E + A'		*/ \
+	rotrwi		a,a,2;		/* 2: B = B rotl 30		*/ \
+	add		d,d,rT1		/* 2: E = E + F			*/
+
+#define R_20_39(a, b, c, d, e, w0, w1, w4, w6, w7, k) \
+	evmergelohi	rT0,w7,w6;	/*    W[-3]			*/ \
+	xor		rT2,b,c;	/* 1: F' = B xor C		*/ \
+	evxor		w0,w0,rT0;	/*    W = W[-16] xor W[-3]	*/ \
+	xor		rT2,rT2,d;	/* 1: F = F' xor D		*/ \
+	evxor		w0,w0,w4;	/*    W = W xor W[-8]		*/ \
+	add		e,e,rT2;	/* 1: E = E + F			*/ \
+	evxor		w0,w0,w1;	/*    W = W xor W[-14]		*/ \
+	rotrwi		rT2,a,27;	/* 1: A' = A rotl 5		*/ \
+	evrlwi		w0,w0,1;	/*    W = W rotl 1		*/ \
+	add		e,e,rT2;	/* 1: E = E + A'		*/ \
+	evaddw		rT0,w0,rK;	/*    WK = W + K		*/ \
+	rotrwi		b,b,2;		/* 1: B = B rotl 30		*/ \
+	LOAD_K##k##1							   \
+	evmergehi	rT1,rT1,rT0;	/*    WK1/WK2			*/ \
+	add		e,e,rT0;	/* 1: E = E + WK		*/ \
+	xor		rT2,a,b;	/* 2: F' = B xor C		*/ \
+	add		d,d,rT1;	/* 2: E = E + WK		*/ \
+	xor		rT2,rT2,c;	/* 2: F = F' xor D		*/ \
+	rotrwi		rT0,e,27;	/* 2: A' = A rotl 5		*/ \
+	add		d,d,rT2;	/* 2: E = E + F			*/ \
+	rotrwi		a,a,2;		/* 2: B = B rotl 30		*/ \
+	add		d,d,rT0		/* 2: E = E + A'		*/
+
+#define R_40_59(a, b, c, d, e, w0, w1, w4, w6, w7, k) \
+	and		rT2,b,c;	/* 1: F' = B and C		*/ \
+	evmergelohi	rT0,w7,w6;	/*    W[-3]			*/ \
+	or		rT1,b,c;	/* 1: F" = B or C		*/ \
+	evxor		w0,w0,rT0;	/*    W = W[-16] xor W[-3]	*/ \
+	and		rT1,d,rT1;	/* 1: F" = F" and D		*/ \
+	evxor		w0,w0,w4;	/*    W = W xor W[-8]		*/ \
+	or		rT2,rT2,rT1;	/* 1: F = F' or F"		*/ \
+	evxor		w0,w0,w1;	/*    W = W xor W[-14]		*/ \
+	add		e,e,rT2;	/* 1: E = E + F			*/ \
+	evrlwi		w0,w0,1;	/*    W = W rotl 1		*/ \
+	rotrwi		rT2,a,27;	/* 1: A' = A rotl 5		*/ \
+	evaddw		rT0,w0,rK;	/*    WK = W + K		*/ \
+	add		e,e,rT2;	/* 1: E = E + A'		*/ \
+	LOAD_K##k##1							   \
+	evmergehi	rT1,rT1,rT0;	/*    WK1/WK2			*/ \
+	rotrwi		b,b,2;		/* 1: B = B rotl 30		*/ \
+	add		e,e,rT0;	/* 1: E = E + WK		*/ \
+	and		rT2,a,b;	/* 2: F' = B and C		*/ \
+	or		rT0,a,b;	/* 2: F" = B or C		*/ \
+	add		d,d,rT1;	/* 2: E = E + WK		*/ \
+	and		rT0,c,rT0;	/* 2: F" = F" and D		*/ \
+	rotrwi		a,a,2;		/* 2: B = B rotl 30		*/ \
+	or		rT2,rT2,rT0;	/* 2: F = F' or F"		*/ \
+	rotrwi		rT0,e,27;	/* 2: A' = A rotl 5		*/ \
+	add		d,d,rT2;	/* 2: E = E + F			*/ \
+	add		d,d,rT0		/* 2: E = E + A'		*/
+
+#define R_60_79(a, b, c, d, e, w0, w1, w4, w6, w7, k) \
+	R_20_39(a, b, c, d, e, w0, w1, w4, w6, w7, k)
+
+_GLOBAL(ppc_spe_sha1_transform)
+	INITIALIZE
+
+	lwz		rH0,0(rHP)
+	lwz		rH1,4(rHP)
+	mtctr		r5
+	lwz		rH2,8(rHP)
+	lis		rKP,PPC_SPE_SHA1_K@h
+	lwz		rH3,12(rHP)
+	ori		rKP,rKP,PPC_SPE_SHA1_K@l
+	lwz		rH4,16(rHP)
+
+ppc_spe_sha1_main:
+	R_00_15(rH0, rH1, rH2, rH3, rH4, rW1, rW0, 1, 0)
+	R_00_15(rH3, rH4, rH0, rH1, rH2, rW2, rW1, 0, 8)
+	R_00_15(rH1, rH2, rH3, rH4, rH0, rW3, rW2, 0, 16)
+	R_00_15(rH4, rH0, rH1, rH2, rH3, rW4, rW3, 0, 24)
+	R_00_15(rH2, rH3, rH4, rH0, rH1, rW5, rW4, 0, 32)
+	R_00_15(rH0, rH1, rH2, rH3, rH4, rW6, rW5, 0, 40)
+	R_00_15(rH3, rH4, rH0, rH1, rH2, rT3, rW6, 0, 48)
+	R_00_15(rH1, rH2, rH3, rH4, rH0, rT3, rW7, 0, 56)
+
+	R_16_19(rH4, rH0, rH1, rH2, rH3, rW0, rW1, rW4, rW6, rW7, 0)
+	R_16_19(rH2, rH3, rH4, rH0, rH1, rW1, rW2, rW5, rW7, rW0, 2)
+
+	R_20_39(rH0, rH1, rH2, rH3, rH4, rW2, rW3, rW6, rW0, rW1, 0)
+	R_20_39(rH3, rH4, rH0, rH1, rH2, rW3, rW4, rW7, rW1, rW2, 0)
+	R_20_39(rH1, rH2, rH3, rH4, rH0, rW4, rW5, rW0, rW2, rW3, 0)
+	R_20_39(rH4, rH0, rH1, rH2, rH3, rW5, rW6, rW1, rW3, rW4, 0)
+	R_20_39(rH2, rH3, rH4, rH0, rH1, rW6, rW7, rW2, rW4, rW5, 0)
+	R_20_39(rH0, rH1, rH2, rH3, rH4, rW7, rW0, rW3, rW5, rW6, 0)
+	R_20_39(rH3, rH4, rH0, rH1, rH2, rW0, rW1, rW4, rW6, rW7, 0)
+	R_20_39(rH1, rH2, rH3, rH4, rH0, rW1, rW2, rW5, rW7, rW0, 0)
+	R_20_39(rH4, rH0, rH1, rH2, rH3, rW2, rW3, rW6, rW0, rW1, 0)
+	R_20_39(rH2, rH3, rH4, rH0, rH1, rW3, rW4, rW7, rW1, rW2, 3)
+
+	R_40_59(rH0, rH1, rH2, rH3, rH4, rW4, rW5, rW0, rW2, rW3, 0)
+	R_40_59(rH3, rH4, rH0, rH1, rH2, rW5, rW6, rW1, rW3, rW4, 0)
+	R_40_59(rH1, rH2, rH3, rH4, rH0, rW6, rW7, rW2, rW4, rW5, 0)
+	R_40_59(rH4, rH0, rH1, rH2, rH3, rW7, rW0, rW3, rW5, rW6, 0)
+	R_40_59(rH2, rH3, rH4, rH0, rH1, rW0, rW1, rW4, rW6, rW7, 0)
+	R_40_59(rH0, rH1, rH2, rH3, rH4, rW1, rW2, rW5, rW7, rW0, 0)
+	R_40_59(rH3, rH4, rH0, rH1, rH2, rW2, rW3, rW6, rW0, rW1, 0)
+	R_40_59(rH1, rH2, rH3, rH4, rH0, rW3, rW4, rW7, rW1, rW2, 0)
+	R_40_59(rH4, rH0, rH1, rH2, rH3, rW4, rW5, rW0, rW2, rW3, 0)
+	R_40_59(rH2, rH3, rH4, rH0, rH1, rW5, rW6, rW1, rW3, rW4, 4)
+
+	R_60_79(rH0, rH1, rH2, rH3, rH4, rW6, rW7, rW2, rW4, rW5, 0)
+	R_60_79(rH3, rH4, rH0, rH1, rH2, rW7, rW0, rW3, rW5, rW6, 0)
+	R_60_79(rH1, rH2, rH3, rH4, rH0, rW0, rW1, rW4, rW6, rW7, 0)
+	R_60_79(rH4, rH0, rH1, rH2, rH3, rW1, rW2, rW5, rW7, rW0, 0)
+	R_60_79(rH2, rH3, rH4, rH0, rH1, rW2, rW3, rW6, rW0, rW1, 0)
+	R_60_79(rH0, rH1, rH2, rH3, rH4, rW3, rW4, rW7, rW1, rW2, 0)
+	R_60_79(rH3, rH4, rH0, rH1, rH2, rW4, rW5, rW0, rW2, rW3, 0)
+	lwz		rT3,0(rHP)
+	R_60_79(rH1, rH2, rH3, rH4, rH0, rW5, rW6, rW1, rW3, rW4, 0)
+	lwz		rW1,4(rHP)
+	R_60_79(rH4, rH0, rH1, rH2, rH3, rW6, rW7, rW2, rW4, rW5, 0)
+	lwz		rW2,8(rHP)
+	R_60_79(rH2, rH3, rH4, rH0, rH1, rW7, rW0, rW3, rW5, rW6, 0)
+	lwz		rW3,12(rHP)
+	NEXT_BLOCK
+	lwz		rW4,16(rHP)
+
+	add		rH0,rH0,rT3
+	stw		rH0,0(rHP)
+	add		rH1,rH1,rW1
+	stw		rH1,4(rHP)
+	add		rH2,rH2,rW2
+	stw		rH2,8(rHP)
+	add		rH3,rH3,rW3
+	stw		rH3,12(rHP)
+	add		rH4,rH4,rW4
+	stw		rH4,16(rHP)
+
+	bdnz		ppc_spe_sha1_main
+
+	FINALIZE
+	blr
+
+.data
+.align 4
+PPC_SPE_SHA1_K:
+	.long 0x5A827999,0x6ED9EBA1,0x8F1BBCDC,0xCA62C1D6
diff --git a/lib/crypto/powerpc/sha1.h b/lib/crypto/powerpc/sha1.h
new file mode 100644
index 000000000..e2c010f03
--- /dev/null
+++ b/lib/crypto/powerpc/sha1.h
@@ -0,0 +1,67 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * SHA-1 optimized for PowerPC
+ *
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ */
+
+#include <asm/switch_to.h>
+#include <linux/preempt.h>
+
+#ifdef CONFIG_SPE
+/*
+ * MAX_BYTES defines the number of bytes that are allowed to be processed
+ * between preempt_disable() and preempt_enable(). SHA1 takes ~1000
+ * operations per 64 bytes. e500 cores can issue two arithmetic instructions
+ * per clock cycle using one 32/64 bit unit (SU1) and one 32 bit unit (SU2).
+ * Thus 2KB of input data will need an estimated maximum of 18,000 cycles.
+ * Headroom for cache misses included. Even with the low end model clocked
+ * at 667 MHz this equals to a critical time window of less than 27us.
+ *
+ */
+#define MAX_BYTES 2048
+
+asmlinkage void ppc_spe_sha1_transform(struct sha1_block_state *state,
+				       const u8 *data, u32 nblocks);
+
+static void spe_begin(void)
+{
+	/* We just start SPE operations and will save SPE registers later. */
+	preempt_disable();
+	enable_kernel_spe();
+}
+
+static void spe_end(void)
+{
+	disable_kernel_spe();
+	/* reenable preemption */
+	preempt_enable();
+}
+
+static void sha1_blocks(struct sha1_block_state *state,
+			const u8 *data, size_t nblocks)
+{
+	do {
+		u32 unit = min_t(size_t, nblocks, MAX_BYTES / SHA1_BLOCK_SIZE);
+
+		spe_begin();
+		ppc_spe_sha1_transform(state, data, unit);
+		spe_end();
+
+		data += unit * SHA1_BLOCK_SIZE;
+		nblocks -= unit;
+	} while (nblocks);
+}
+#else /* CONFIG_SPE */
+asmlinkage void powerpc_sha_transform(struct sha1_block_state *state,
+				      const u8 data[SHA1_BLOCK_SIZE]);
+
+static void sha1_blocks(struct sha1_block_state *state,
+			const u8 *data, size_t nblocks)
+{
+	do {
+		powerpc_sha_transform(state, data);
+		data += SHA1_BLOCK_SIZE;
+	} while (--nblocks);
+}
+#endif /* !CONFIG_SPE */
diff --git a/lib/crypto/powerpc/sha256-spe-asm.S b/lib/crypto/powerpc/sha256-spe-asm.S
new file mode 100644
index 000000000..cd99d71da
--- /dev/null
+++ b/lib/crypto/powerpc/sha256-spe-asm.S
@@ -0,0 +1,318 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Fast SHA-256 implementation for SPE instruction set (PPC)
+ *
+ * This code makes use of the SPE SIMD instruction set as defined in
+ * http://cache.freescale.com/files/32bit/doc/ref_manual/SPEPIM.pdf
+ * Implementation is based on optimization guide notes from
+ * http://cache.freescale.com/files/32bit/doc/app_note/AN2665.pdf
+ *
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ */
+
+#include <asm/ppc_asm.h>
+#include <asm/asm-offsets.h>
+
+#define rHP	r3	/* pointer to hash values in memory		*/
+#define rKP	r24	/* pointer to round constants			*/
+#define rWP	r4	/* pointer to input data			*/
+
+#define rH0	r5	/* 8 32 bit hash values in 8 registers		*/
+#define rH1	r6
+#define rH2	r7
+#define rH3	r8
+#define rH4	r9
+#define rH5	r10
+#define rH6	r11
+#define rH7	r12
+
+#define rW0	r14	/* 64 bit registers. 16 words in 8 registers	*/
+#define rW1	r15
+#define rW2	r16
+#define rW3	r17
+#define rW4	r18
+#define rW5	r19
+#define rW6	r20
+#define rW7	r21
+
+#define rT0	r22	/* 64 bit temporaries 				*/
+#define rT1	r23
+#define rT2	r0	/* 32 bit temporaries				*/
+#define rT3	r25
+
+#define CMP_KN_LOOP
+#define CMP_KC_LOOP \
+	cmpwi		rT1,0;
+
+#define INITIALIZE \
+	stwu		r1,-128(r1);	/* create stack frame		*/ \
+	evstdw		r14,8(r1);	/* We must save non volatile	*/ \
+	evstdw		r15,16(r1);	/* registers. Take the chance	*/ \
+	evstdw		r16,24(r1);	/* and save the SPE part too	*/ \
+	evstdw		r17,32(r1);					   \
+	evstdw		r18,40(r1);					   \
+	evstdw		r19,48(r1);					   \
+	evstdw		r20,56(r1);					   \
+	evstdw		r21,64(r1);					   \
+	evstdw		r22,72(r1);					   \
+	evstdw		r23,80(r1);					   \
+	stw		r24,88(r1);	/* save normal registers	*/ \
+	stw		r25,92(r1);
+
+
+#define FINALIZE \
+	evldw		r14,8(r1);	/* restore SPE registers	*/ \
+	evldw		r15,16(r1);					   \
+	evldw		r16,24(r1);					   \
+	evldw		r17,32(r1);					   \
+	evldw		r18,40(r1);					   \
+	evldw		r19,48(r1);					   \
+	evldw		r20,56(r1);					   \
+	evldw		r21,64(r1);					   \
+	evldw		r22,72(r1);					   \
+	evldw		r23,80(r1);					   \
+	lwz		r24,88(r1);	/* restore normal registers	*/ \
+	lwz		r25,92(r1);					   \
+	xor		r0,r0,r0;					   \
+	stw		r0,8(r1);	/* Delete sensitive data	*/ \
+	stw		r0,16(r1);	/* that we might have pushed	*/ \
+	stw		r0,24(r1);	/* from other context that runs	*/ \
+	stw		r0,32(r1);	/* the same code. Assume that	*/ \
+	stw		r0,40(r1);	/* the lower part of the GPRs	*/ \
+	stw		r0,48(r1);	/* was already overwritten on	*/ \
+	stw		r0,56(r1);	/* the way down to here		*/ \
+	stw		r0,64(r1);					   \
+	stw		r0,72(r1);					   \
+	stw		r0,80(r1);					   \
+	addi		r1,r1,128;	/* cleanup stack frame		*/
+
+#ifdef __BIG_ENDIAN__
+#define LOAD_DATA(reg, off) \
+	lwz		reg,off(rWP);	/* load data			*/
+#define NEXT_BLOCK \
+	addi		rWP,rWP,64;	/* increment per block		*/
+#else
+#define LOAD_DATA(reg, off) \
+	lwbrx		reg,0,rWP; 	/* load data			*/ \
+	addi		rWP,rWP,4;	/* increment per word		*/
+#define NEXT_BLOCK			/* nothing to do		*/
+#endif
+
+#define R_LOAD_W(a, b, c, d, e, f, g, h, w, off) \
+	LOAD_DATA(w, off)		/* 1: W				*/ \
+	rotrwi		rT0,e,6;	/* 1: S1 = e rotr 6		*/ \
+	rotrwi		rT1,e,11;	/* 1: S1' = e rotr 11		*/ \
+	rotrwi		rT2,e,25;	/* 1: S1" = e rotr 25		*/ \
+	xor		rT0,rT0,rT1;	/* 1: S1 = S1 xor S1'		*/ \
+	and		rT3,e,f;	/* 1: ch = e and f		*/ \
+	xor		rT0,rT0,rT2;	/* 1: S1 = S1 xor S1"		*/ \
+	andc		rT1,g,e;	/* 1: ch' = ~e and g		*/ \
+	lwz		rT2,off(rKP);	/* 1: K				*/ \
+	xor		rT3,rT3,rT1;	/* 1: ch = ch xor ch'		*/ \
+	add		h,h,rT0;	/* 1: temp1 = h + S1		*/ \
+	add		rT3,rT3,w;	/* 1: temp1' = ch + w		*/ \
+	rotrwi		rT0,a,2;	/* 1: S0 = a rotr 2		*/ \
+	add		h,h,rT3;	/* 1: temp1 = temp1 + temp1'	*/ \
+	rotrwi		rT1,a,13;	/* 1: S0' = a rotr 13		*/ \
+	add		h,h,rT2;	/* 1: temp1 = temp1 + K		*/ \
+	rotrwi		rT3,a,22;	/* 1: S0" = a rotr 22		*/ \
+	xor		rT0,rT0,rT1;	/* 1: S0 = S0 xor S0'		*/ \
+	add		d,d,h;		/* 1: d = d + temp1		*/ \
+	xor		rT3,rT0,rT3;	/* 1: S0 = S0 xor S0"		*/ \
+	evmergelo	w,w,w;		/*    shift W			*/ \
+	or		rT2,a,b;	/* 1: maj = a or b		*/ \
+	and		rT1,a,b;	/* 1: maj' = a and b		*/ \
+	and		rT2,rT2,c;	/* 1: maj = maj and c		*/ \
+	LOAD_DATA(w, off+4)		/* 2: W				*/ \
+	or		rT2,rT1,rT2;	/* 1: maj = maj or maj'		*/ \
+	rotrwi		rT0,d,6;	/* 2: S1 = e rotr 6		*/ \
+	add		rT3,rT3,rT2;	/* 1: temp2 = S0 + maj		*/ \
+	rotrwi		rT1,d,11;	/* 2: S1' = e rotr 11		*/ \
+	add		h,h,rT3;	/* 1: h = temp1 + temp2		*/ \
+	rotrwi		rT2,d,25;	/* 2: S1" = e rotr 25		*/ \
+	xor		rT0,rT0,rT1;	/* 2: S1 = S1 xor S1'		*/ \
+	and		rT3,d,e;	/* 2: ch = e and f		*/ \
+	xor		rT0,rT0,rT2;	/* 2: S1 = S1 xor S1"		*/ \
+	andc		rT1,f,d;	/* 2: ch' = ~e and g		*/ \
+	lwz		rT2,off+4(rKP);	/* 2: K				*/ \
+	xor		rT3,rT3,rT1;	/* 2: ch = ch xor ch'		*/ \
+	add		g,g,rT0;	/* 2: temp1 = h + S1		*/ \
+	add		rT3,rT3,w;	/* 2: temp1' = ch + w		*/ \
+	rotrwi		rT0,h,2;	/* 2: S0 = a rotr 2		*/ \
+	add		g,g,rT3;	/* 2: temp1 = temp1 + temp1'	*/ \
+	rotrwi		rT1,h,13;	/* 2: S0' = a rotr 13		*/ \
+	add		g,g,rT2;	/* 2: temp1 = temp1 + K		*/ \
+	rotrwi		rT3,h,22;	/* 2: S0" = a rotr 22		*/ \
+	xor		rT0,rT0,rT1;	/* 2: S0 = S0 xor S0'		*/ \
+	or		rT2,h,a;	/* 2: maj = a or b		*/ \
+	xor		rT3,rT0,rT3;	/* 2: S0 = S0 xor S0"		*/ \
+	and		rT1,h,a;	/* 2: maj' = a and b		*/ \
+	and		rT2,rT2,b;	/* 2: maj = maj and c		*/ \
+	add		c,c,g;		/* 2: d = d + temp1		*/ \
+	or		rT2,rT1,rT2;	/* 2: maj = maj or maj'		*/ \
+	add		rT3,rT3,rT2;	/* 2: temp2 = S0 + maj		*/ \
+	add		g,g,rT3		/* 2: h = temp1 + temp2		*/
+
+#define R_CALC_W(a, b, c, d, e, f, g, h, w0, w1, w4, w5, w7, k, off) \
+	rotrwi		rT2,e,6;	/* 1: S1 = e rotr 6		*/ \
+	evmergelohi	rT0,w0,w1;	/*    w[-15]			*/ \
+	rotrwi		rT3,e,11;	/* 1: S1' = e rotr 11		*/ \
+	evsrwiu		rT1,rT0,3;	/*    s0 = w[-15] >> 3		*/ \
+	xor		rT2,rT2,rT3;	/* 1: S1 = S1 xor S1'		*/ \
+	evrlwi		rT0,rT0,25;	/*    s0' = w[-15] rotr	7	*/ \
+	rotrwi		rT3,e,25;	/* 1: S1' = e rotr 25		*/ \
+	evxor		rT1,rT1,rT0;	/*    s0 = s0 xor s0'		*/ \
+	xor		rT2,rT2,rT3;	/* 1: S1 = S1 xor S1'		*/ \
+	evrlwi		rT0,rT0,21;	/*    s0' = w[-15] rotr 18	*/ \
+	add		h,h,rT2;	/* 1: temp1 = h + S1		*/ \
+	evxor		rT0,rT0,rT1;	/*    s0 = s0 xor s0'		*/ \
+	and		rT2,e,f;	/* 1: ch = e and f		*/ \
+	evaddw		w0,w0,rT0;	/*    w = w[-16] + s0		*/ \
+	andc		rT3,g,e;	/* 1: ch' = ~e and g		*/ \
+	evsrwiu		rT0,w7,10;	/*    s1 = w[-2] >> 10		*/ \
+	xor		rT2,rT2,rT3;	/* 1: ch = ch xor ch'		*/ \
+	evrlwi		rT1,w7,15;	/*    s1' = w[-2] rotr 17	*/ \
+	add		h,h,rT2;	/* 1: temp1 = temp1 + ch	*/ \
+	evxor		rT0,rT0,rT1;	/*    s1 = s1 xor s1'		*/ \
+	rotrwi		rT2,a,2;	/* 1: S0 = a rotr 2		*/ \
+	evrlwi		rT1,w7,13;	/*    s1' = w[-2] rotr 19	*/ \
+	rotrwi		rT3,a,13;	/* 1: S0' = a rotr 13		*/ \
+	evxor		rT0,rT0,rT1;	/*    s1 = s1 xor s1'		*/ \
+	xor		rT2,rT2,rT3;	/* 1: S0 = S0 xor S0'		*/ \
+	evldw		rT1,off(rKP);	/*    k				*/ \
+	rotrwi		rT3,a,22;	/* 1: S0' = a rotr 22		*/ \
+	evaddw		w0,w0,rT0;	/*    w = w + s1		*/ \
+	xor		rT2,rT2,rT3;	/* 1: S0 = S0 xor S0'		*/ \
+	evmergelohi	rT0,w4,w5;	/*    w[-7]			*/ \
+	and		rT3,a,b;	/* 1: maj = a and b		*/ \
+	evaddw		w0,w0,rT0;	/*    w = w + w[-7]		*/ \
+	CMP_K##k##_LOOP							   \
+	add		rT2,rT2,rT3;	/* 1: temp2 = S0 + maj		*/ \
+	evaddw		rT1,rT1,w0;	/*    wk = w + k		*/ \
+	xor		rT3,a,b;	/* 1: maj = a xor b		*/ \
+	evmergehi	rT0,rT1,rT1;	/*    wk1/wk2			*/ \
+	and		rT3,rT3,c;	/* 1: maj = maj and c		*/ \
+	add		h,h,rT0;	/* 1: temp1 = temp1 + wk	*/ \
+	add		rT2,rT2,rT3;	/* 1: temp2 = temp2 + maj	*/ \
+	add		g,g,rT1;	/* 2: temp1 = temp1 + wk	*/ \
+	add		d,d,h;		/* 1: d = d + temp1		*/ \
+	rotrwi		rT0,d,6;	/* 2: S1 = e rotr 6		*/ \
+	add		h,h,rT2;	/* 1: h = temp1 + temp2		*/ \
+	rotrwi		rT1,d,11;	/* 2: S1' = e rotr 11		*/ \
+	rotrwi		rT2,d,25;	/* 2: S" = e rotr 25		*/ \
+	xor		rT0,rT0,rT1;	/* 2: S1 = S1 xor S1'		*/ \
+	and		rT3,d,e;	/* 2: ch = e and f		*/ \
+	xor		rT0,rT0,rT2;	/* 2: S1 = S1 xor S1"		*/ \
+	andc		rT1,f,d;	/* 2: ch' = ~e and g		*/ \
+	add		g,g,rT0;	/* 2: temp1 = h + S1		*/ \
+	xor		rT3,rT3,rT1;	/* 2: ch = ch xor ch'		*/ \
+	rotrwi		rT0,h,2;	/* 2: S0 = a rotr 2		*/ \
+	add		g,g,rT3;	/* 2: temp1 = temp1 + ch	*/ \
+	rotrwi		rT1,h,13;	/* 2: S0' = a rotr 13		*/ \
+	rotrwi		rT3,h,22;	/* 2: S0" = a rotr 22		*/ \
+	xor		rT0,rT0,rT1;	/* 2: S0 = S0 xor S0'		*/ \
+	or		rT2,h,a;	/* 2: maj = a or b		*/ \
+	and		rT1,h,a;	/* 2: maj' = a and b		*/ \
+	and		rT2,rT2,b;	/* 2: maj = maj and c		*/ \
+	xor		rT3,rT0,rT3;	/* 2: S0 = S0 xor S0"		*/ \
+	or		rT2,rT1,rT2;	/* 2: maj = maj or maj'		*/ \
+	add		c,c,g;		/* 2: d = d + temp1		*/ \
+	add		rT3,rT3,rT2;	/* 2: temp2 = S0 + maj		*/ \
+	add		g,g,rT3		/* 2: h = temp1 + temp2		*/
+
+_GLOBAL(ppc_spe_sha256_transform)
+	INITIALIZE
+
+	mtctr		r5
+	lwz		rH0,0(rHP)
+	lwz		rH1,4(rHP)
+	lwz		rH2,8(rHP)
+	lwz		rH3,12(rHP)
+	lwz		rH4,16(rHP)
+	lwz		rH5,20(rHP)
+	lwz		rH6,24(rHP)
+	lwz		rH7,28(rHP)
+
+ppc_spe_sha256_main:
+	lis		rKP,PPC_SPE_SHA256_K@ha
+	addi		rKP,rKP,PPC_SPE_SHA256_K@l
+
+	R_LOAD_W(rH0, rH1, rH2, rH3, rH4, rH5, rH6, rH7, rW0, 0)
+	R_LOAD_W(rH6, rH7, rH0, rH1, rH2, rH3, rH4, rH5, rW1, 8)
+	R_LOAD_W(rH4, rH5, rH6, rH7, rH0, rH1, rH2, rH3, rW2, 16)
+	R_LOAD_W(rH2, rH3, rH4, rH5, rH6, rH7, rH0, rH1, rW3, 24)
+	R_LOAD_W(rH0, rH1, rH2, rH3, rH4, rH5, rH6, rH7, rW4, 32)
+	R_LOAD_W(rH6, rH7, rH0, rH1, rH2, rH3, rH4, rH5, rW5, 40)
+	R_LOAD_W(rH4, rH5, rH6, rH7, rH0, rH1, rH2, rH3, rW6, 48)
+	R_LOAD_W(rH2, rH3, rH4, rH5, rH6, rH7, rH0, rH1, rW7, 56)
+ppc_spe_sha256_16_rounds:
+	addi		rKP,rKP,64
+	R_CALC_W(rH0, rH1, rH2, rH3, rH4, rH5, rH6, rH7,
+		 rW0, rW1, rW4, rW5, rW7, N, 0)
+	R_CALC_W(rH6, rH7, rH0, rH1, rH2, rH3, rH4, rH5,
+		 rW1, rW2, rW5, rW6, rW0, N, 8)
+	R_CALC_W(rH4, rH5, rH6, rH7, rH0, rH1, rH2, rH3,
+		 rW2, rW3, rW6, rW7, rW1, N, 16)
+	R_CALC_W(rH2, rH3, rH4, rH5, rH6, rH7, rH0, rH1,
+		 rW3, rW4, rW7, rW0, rW2, N, 24)
+	R_CALC_W(rH0, rH1, rH2, rH3, rH4, rH5, rH6, rH7,
+		 rW4, rW5, rW0, rW1, rW3, N, 32)
+	R_CALC_W(rH6, rH7, rH0, rH1, rH2, rH3, rH4, rH5,
+		 rW5, rW6, rW1, rW2, rW4, N, 40)
+	R_CALC_W(rH4, rH5, rH6, rH7, rH0, rH1, rH2, rH3,
+		 rW6, rW7, rW2, rW3, rW5, N, 48)
+	R_CALC_W(rH2, rH3, rH4, rH5, rH6, rH7, rH0, rH1,
+		 rW7, rW0, rW3, rW4, rW6, C, 56)
+	bt		gt,ppc_spe_sha256_16_rounds
+
+	lwz		rW0,0(rHP)
+	NEXT_BLOCK
+	lwz		rW1,4(rHP)
+	lwz		rW2,8(rHP)
+	lwz		rW3,12(rHP)
+	lwz		rW4,16(rHP)
+	lwz		rW5,20(rHP)
+	lwz		rW6,24(rHP)
+	lwz		rW7,28(rHP)
+
+	add		rH0,rH0,rW0
+	stw		rH0,0(rHP)
+	add		rH1,rH1,rW1
+	stw		rH1,4(rHP)
+	add		rH2,rH2,rW2
+	stw		rH2,8(rHP)
+	add		rH3,rH3,rW3
+	stw		rH3,12(rHP)
+	add		rH4,rH4,rW4
+	stw		rH4,16(rHP)
+	add		rH5,rH5,rW5
+	stw		rH5,20(rHP)
+	add		rH6,rH6,rW6
+	stw		rH6,24(rHP)
+	add		rH7,rH7,rW7
+	stw		rH7,28(rHP)
+
+	bdnz		ppc_spe_sha256_main
+
+	FINALIZE
+	blr
+
+.data
+.align 5
+PPC_SPE_SHA256_K:
+	.long 0x428a2f98,0x71374491,0xb5c0fbcf,0xe9b5dba5
+	.long 0x3956c25b,0x59f111f1,0x923f82a4,0xab1c5ed5
+	.long 0xd807aa98,0x12835b01,0x243185be,0x550c7dc3
+	.long 0x72be5d74,0x80deb1fe,0x9bdc06a7,0xc19bf174
+	.long 0xe49b69c1,0xefbe4786,0x0fc19dc6,0x240ca1cc
+	.long 0x2de92c6f,0x4a7484aa,0x5cb0a9dc,0x76f988da
+	.long 0x983e5152,0xa831c66d,0xb00327c8,0xbf597fc7
+	.long 0xc6e00bf3,0xd5a79147,0x06ca6351,0x14292967
+	.long 0x27b70a85,0x2e1b2138,0x4d2c6dfc,0x53380d13
+	.long 0x650a7354,0x766a0abb,0x81c2c92e,0x92722c85
+	.long 0xa2bfe8a1,0xa81a664b,0xc24b8b70,0xc76c51a3
+	.long 0xd192e819,0xd6990624,0xf40e3585,0x106aa070
+	.long 0x19a4c116,0x1e376c08,0x2748774c,0x34b0bcb5
+	.long 0x391c0cb3,0x4ed8aa4a,0x5b9cca4f,0x682e6ff3
+	.long 0x748f82ee,0x78a5636f,0x84c87814,0x8cc70208
+	.long 0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2
diff --git a/lib/crypto/powerpc/sha256.h b/lib/crypto/powerpc/sha256.h
new file mode 100644
index 000000000..50d355441
--- /dev/null
+++ b/lib/crypto/powerpc/sha256.h
@@ -0,0 +1,58 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * SHA-256 Secure Hash Algorithm, SPE optimized
+ *
+ * Based on generic implementation. The assembler module takes care
+ * about the SPE registers so it can run from interrupt context.
+ *
+ * Copyright (c) 2015 Markus Stockhausen <stockhausen@collogia.de>
+ */
+
+#include <asm/switch_to.h>
+#include <linux/preempt.h>
+
+/*
+ * MAX_BYTES defines the number of bytes that are allowed to be processed
+ * between preempt_disable() and preempt_enable(). SHA256 takes ~2,000
+ * operations per 64 bytes. e500 cores can issue two arithmetic instructions
+ * per clock cycle using one 32/64 bit unit (SU1) and one 32 bit unit (SU2).
+ * Thus 1KB of input data will need an estimated maximum of 18,000 cycles.
+ * Headroom for cache misses included. Even with the low end model clocked
+ * at 667 MHz this equals to a critical time window of less than 27us.
+ *
+ */
+#define MAX_BYTES 1024
+
+extern void ppc_spe_sha256_transform(struct sha256_block_state *state,
+				     const u8 *src, u32 blocks);
+
+static void spe_begin(void)
+{
+	/* We just start SPE operations and will save SPE registers later. */
+	preempt_disable();
+	enable_kernel_spe();
+}
+
+static void spe_end(void)
+{
+	disable_kernel_spe();
+	/* reenable preemption */
+	preempt_enable();
+}
+
+static void sha256_blocks(struct sha256_block_state *state,
+			  const u8 *data, size_t nblocks)
+{
+	do {
+		/* cut input data into smaller blocks */
+		u32 unit = min_t(size_t, nblocks,
+				 MAX_BYTES / SHA256_BLOCK_SIZE);
+
+		spe_begin();
+		ppc_spe_sha256_transform(state, data, unit);
+		spe_end();
+
+		data += unit * SHA256_BLOCK_SIZE;
+		nblocks -= unit;
+	} while (nblocks);
+}