* [PATCH 2/2] crypto: powerpc: Add POWER8 optimised crc32c
2016-06-30 22:19 [PATCH 1/2] powerpc: define FUNC_START/FUNC_END Anton Blanchard
@ 2016-06-30 22:19 ` Anton Blanchard
2016-07-01 9:32 ` Herbert Xu
` (2 more replies)
0 siblings, 3 replies; 5+ messages in thread
From: Anton Blanchard @ 2016-06-30 22:19 UTC (permalink / raw)
To: benh, paulus, mpe, herbert, davem; +Cc: linuxppc-dev, linux-crypto
From: Anton Blanchard <anton@samba.org>
Use the vector polynomial multiply-sum instructions in POWER8 to
speed up crc32c.
This is just over 41x faster than the slice-by-8 method that it
replaces. Measurements on a 4.1 GHz POWER8 show it sustaining
52 GiB/sec.
A simple btrfs write performance test:
dd if=/dev/zero of=/mnt/tmpfile bs=1M count=4096
sync
is over 3.7x faster.
Signed-off-by: Anton Blanchard <anton@samba.org>
---
arch/powerpc/crypto/Makefile | 2 +
arch/powerpc/crypto/crc32c-vpmsum_asm.S | 1553 ++++++++++++++++++++++++++++++
arch/powerpc/crypto/crc32c-vpmsum_glue.c | 167 ++++
arch/powerpc/include/asm/ppc-opcode.h | 12 +
crypto/Kconfig | 11 +
5 files changed, 1745 insertions(+)
create mode 100644 arch/powerpc/crypto/crc32c-vpmsum_asm.S
create mode 100644 arch/powerpc/crypto/crc32c-vpmsum_glue.c
diff --git a/arch/powerpc/crypto/Makefile b/arch/powerpc/crypto/Makefile
index 9c221b6..7998c17 100644
--- a/arch/powerpc/crypto/Makefile
+++ b/arch/powerpc/crypto/Makefile
@@ -9,9 +9,11 @@ obj-$(CONFIG_CRYPTO_MD5_PPC) += md5-ppc.o
obj-$(CONFIG_CRYPTO_SHA1_PPC) += sha1-powerpc.o
obj-$(CONFIG_CRYPTO_SHA1_PPC_SPE) += sha1-ppc-spe.o
obj-$(CONFIG_CRYPTO_SHA256_PPC_SPE) += sha256-ppc-spe.o
+obj-$(CONFIG_CRYPT_CRC32C_VPMSUM) += crc32c-vpmsum.o
aes-ppc-spe-y := aes-spe-core.o aes-spe-keys.o aes-tab-4k.o aes-spe-modes.o aes-spe-glue.o
md5-ppc-y := md5-asm.o md5-glue.o
sha1-powerpc-y := sha1-powerpc-asm.o sha1.o
sha1-ppc-spe-y := sha1-spe-asm.o sha1-spe-glue.o
sha256-ppc-spe-y := sha256-spe-asm.o sha256-spe-glue.o
+crc32c-vpmsum-y := crc32c-vpmsum_asm.o crc32c-vpmsum_glue.o
diff --git a/arch/powerpc/crypto/crc32c-vpmsum_asm.S b/arch/powerpc/crypto/crc32c-vpmsum_asm.S
new file mode 100644
index 0000000..dc640b2
--- /dev/null
+++ b/arch/powerpc/crypto/crc32c-vpmsum_asm.S
@@ -0,0 +1,1553 @@
+/*
+ * Calculate the checksum of data that is 16 byte aligned and a multiple of
+ * 16 bytes.
+ *
+ * The first step is to reduce it to 1024 bits. We do this in 8 parallel
+ * chunks in order to mask the latency of the vpmsum instructions. If we
+ * have more than 32 kB of data to checksum we repeat this step multiple
+ * times, passing in the previous 1024 bits.
+ *
+ * The next step is to reduce the 1024 bits to 64 bits. This step adds
+ * 32 bits of 0s to the end - this matches what a CRC does. We just
+ * calculate constants that land the data in this 32 bits.
+ *
+ * We then use fixed point Barrett reduction to compute a mod n over GF(2)
+ * for n = CRC using POWER8 instructions. We use x = 32.
+ *
+ * http://en.wikipedia.org/wiki/Barrett_reduction
+ *
+ * Copyright (C) 2015 Anton Blanchard <anton@au.ibm.com>, IBM
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <asm/ppc_asm.h>
+#include <asm/ppc-opcode.h>
+
+ .section .rodata
+.balign 16
+
+.byteswap_constant:
+ /* byte reverse permute constant */
+ .octa 0x0F0E0D0C0B0A09080706050403020100
+
+#define MAX_SIZE 32768
+.constants:
+
+ /* Reduce 262144 kbits to 1024 bits */
+ /* x^261120 mod p(x)` << 1, x^261184 mod p(x)` << 1 */
+ .octa 0x00000000b6ca9e20000000009c37c408
+
+ /* x^260096 mod p(x)` << 1, x^260160 mod p(x)` << 1 */
+ .octa 0x00000000350249a800000001b51df26c
+
+ /* x^259072 mod p(x)` << 1, x^259136 mod p(x)` << 1 */
+ .octa 0x00000001862dac54000000000724b9d0
+
+ /* x^258048 mod p(x)` << 1, x^258112 mod p(x)` << 1 */
+ .octa 0x00000001d87fb48c00000001c00532fe
+
+ /* x^257024 mod p(x)` << 1, x^257088 mod p(x)` << 1 */
+ .octa 0x00000001f39b699e00000000f05a9362
+
+ /* x^256000 mod p(x)` << 1, x^256064 mod p(x)` << 1 */
+ .octa 0x0000000101da11b400000001e1007970
+
+ /* x^254976 mod p(x)` << 1, x^255040 mod p(x)` << 1 */
+ .octa 0x00000001cab571e000000000a57366ee
+
+ /* x^253952 mod p(x)` << 1, x^254016 mod p(x)` << 1 */
+ .octa 0x00000000c7020cfe0000000192011284
+
+ /* x^252928 mod p(x)` << 1, x^252992 mod p(x)` << 1 */
+ .octa 0x00000000cdaed1ae0000000162716d9a
+
+ /* x^251904 mod p(x)` << 1, x^251968 mod p(x)` << 1 */
+ .octa 0x00000001e804effc00000000cd97ecde
+
+ /* x^250880 mod p(x)` << 1, x^250944 mod p(x)` << 1 */
+ .octa 0x0000000077c3ea3a0000000058812bc0
+
+ /* x^249856 mod p(x)` << 1, x^249920 mod p(x)` << 1 */
+ .octa 0x0000000068df31b40000000088b8c12e
+
+ /* x^248832 mod p(x)` << 1, x^248896 mod p(x)` << 1 */
+ .octa 0x00000000b059b6c200000001230b234c
+
+ /* x^247808 mod p(x)` << 1, x^247872 mod p(x)` << 1 */
+ .octa 0x0000000145fb8ed800000001120b416e
+
+ /* x^246784 mod p(x)` << 1, x^246848 mod p(x)` << 1 */
+ .octa 0x00000000cbc0916800000001974aecb0
+
+ /* x^245760 mod p(x)` << 1, x^245824 mod p(x)` << 1 */
+ .octa 0x000000005ceeedc2000000008ee3f226
+
+ /* x^244736 mod p(x)` << 1, x^244800 mod p(x)` << 1 */
+ .octa 0x0000000047d74e8600000001089aba9a
+
+ /* x^243712 mod p(x)` << 1, x^243776 mod p(x)` << 1 */
+ .octa 0x00000001407e9e220000000065113872
+
+ /* x^242688 mod p(x)` << 1, x^242752 mod p(x)` << 1 */
+ .octa 0x00000001da967bda000000005c07ec10
+
+ /* x^241664 mod p(x)` << 1, x^241728 mod p(x)` << 1 */
+ .octa 0x000000006c8983680000000187590924
+
+ /* x^240640 mod p(x)` << 1, x^240704 mod p(x)` << 1 */
+ .octa 0x00000000f2d14c9800000000e35da7c6
+
+ /* x^239616 mod p(x)` << 1, x^239680 mod p(x)` << 1 */
+ .octa 0x00000001993c6ad4000000000415855a
+
+ /* x^238592 mod p(x)` << 1, x^238656 mod p(x)` << 1 */
+ .octa 0x000000014683d1ac0000000073617758
+
+ /* x^237568 mod p(x)` << 1, x^237632 mod p(x)` << 1 */
+ .octa 0x00000001a7c93e6c0000000176021d28
+
+ /* x^236544 mod p(x)` << 1, x^236608 mod p(x)` << 1 */
+ .octa 0x000000010211e90a00000001c358fd0a
+
+ /* x^235520 mod p(x)` << 1, x^235584 mod p(x)` << 1 */
+ .octa 0x000000001119403e00000001ff7a2c18
+
+ /* x^234496 mod p(x)` << 1, x^234560 mod p(x)` << 1 */
+ .octa 0x000000001c3261aa00000000f2d9f7e4
+
+ /* x^233472 mod p(x)` << 1, x^233536 mod p(x)` << 1 */
+ .octa 0x000000014e37a634000000016cf1f9c8
+
+ /* x^232448 mod p(x)` << 1, x^232512 mod p(x)` << 1 */
+ .octa 0x0000000073786c0c000000010af9279a
+
+ /* x^231424 mod p(x)` << 1, x^231488 mod p(x)` << 1 */
+ .octa 0x000000011dc037f80000000004f101e8
+
+ /* x^230400 mod p(x)` << 1, x^230464 mod p(x)` << 1 */
+ .octa 0x0000000031433dfc0000000070bcf184
+
+ /* x^229376 mod p(x)` << 1, x^229440 mod p(x)` << 1 */
+ .octa 0x000000009cde8348000000000a8de642
+
+ /* x^228352 mod p(x)` << 1, x^228416 mod p(x)` << 1 */
+ .octa 0x0000000038d3c2a60000000062ea130c
+
+ /* x^227328 mod p(x)` << 1, x^227392 mod p(x)` << 1 */
+ .octa 0x000000011b25f26000000001eb31cbb2
+
+ /* x^226304 mod p(x)` << 1, x^226368 mod p(x)` << 1 */
+ .octa 0x000000001629e6f00000000170783448
+
+ /* x^225280 mod p(x)` << 1, x^225344 mod p(x)` << 1 */
+ .octa 0x0000000160838b4c00000001a684b4c6
+
+ /* x^224256 mod p(x)` << 1, x^224320 mod p(x)` << 1 */
+ .octa 0x000000007a44011c00000000253ca5b4
+
+ /* x^223232 mod p(x)` << 1, x^223296 mod p(x)` << 1 */
+ .octa 0x00000000226f417a0000000057b4b1e2
+
+ /* x^222208 mod p(x)` << 1, x^222272 mod p(x)` << 1 */
+ .octa 0x0000000045eb2eb400000000b6bd084c
+
+ /* x^221184 mod p(x)` << 1, x^221248 mod p(x)` << 1 */
+ .octa 0x000000014459d70c0000000123c2d592
+
+ /* x^220160 mod p(x)` << 1, x^220224 mod p(x)` << 1 */
+ .octa 0x00000001d406ed8200000000159dafce
+
+ /* x^219136 mod p(x)` << 1, x^219200 mod p(x)` << 1 */
+ .octa 0x0000000160c8e1a80000000127e1a64e
+
+ /* x^218112 mod p(x)` << 1, x^218176 mod p(x)` << 1 */
+ .octa 0x0000000027ba80980000000056860754
+
+ /* x^217088 mod p(x)` << 1, x^217152 mod p(x)` << 1 */
+ .octa 0x000000006d92d01800000001e661aae8
+
+ /* x^216064 mod p(x)` << 1, x^216128 mod p(x)` << 1 */
+ .octa 0x000000012ed7e3f200000000f82c6166
+
+ /* x^215040 mod p(x)` << 1, x^215104 mod p(x)` << 1 */
+ .octa 0x000000002dc8778800000000c4f9c7ae
+
+ /* x^214016 mod p(x)` << 1, x^214080 mod p(x)` << 1 */
+ .octa 0x0000000018240bb80000000074203d20
+
+ /* x^212992 mod p(x)` << 1, x^213056 mod p(x)` << 1 */
+ .octa 0x000000001ad381580000000198173052
+
+ /* x^211968 mod p(x)` << 1, x^212032 mod p(x)` << 1 */
+ .octa 0x00000001396b78f200000001ce8aba54
+
+ /* x^210944 mod p(x)` << 1, x^211008 mod p(x)` << 1 */
+ .octa 0x000000011a68133400000001850d5d94
+
+ /* x^209920 mod p(x)` << 1, x^209984 mod p(x)` << 1 */
+ .octa 0x000000012104732e00000001d609239c
+
+ /* x^208896 mod p(x)` << 1, x^208960 mod p(x)` << 1 */
+ .octa 0x00000000a140d90c000000001595f048
+
+ /* x^207872 mod p(x)` << 1, x^207936 mod p(x)` << 1 */
+ .octa 0x00000001b7215eda0000000042ccee08
+
+ /* x^206848 mod p(x)` << 1, x^206912 mod p(x)` << 1 */
+ .octa 0x00000001aaf1df3c000000010a389d74
+
+ /* x^205824 mod p(x)` << 1, x^205888 mod p(x)` << 1 */
+ .octa 0x0000000029d15b8a000000012a840da6
+
+ /* x^204800 mod p(x)` << 1, x^204864 mod p(x)` << 1 */
+ .octa 0x00000000f1a96922000000001d181c0c
+
+ /* x^203776 mod p(x)` << 1, x^203840 mod p(x)` << 1 */
+ .octa 0x00000001ac80d03c0000000068b7d1f6
+
+ /* x^202752 mod p(x)` << 1, x^202816 mod p(x)` << 1 */
+ .octa 0x000000000f11d56a000000005b0f14fc
+
+ /* x^201728 mod p(x)` << 1, x^201792 mod p(x)` << 1 */
+ .octa 0x00000001f1c022a20000000179e9e730
+
+ /* x^200704 mod p(x)` << 1, x^200768 mod p(x)` << 1 */
+ .octa 0x0000000173d00ae200000001ce1368d6
+
+ /* x^199680 mod p(x)` << 1, x^199744 mod p(x)` << 1 */
+ .octa 0x00000001d4ffe4ac0000000112c3a84c
+
+ /* x^198656 mod p(x)` << 1, x^198720 mod p(x)` << 1 */
+ .octa 0x000000016edc5ae400000000de940fee
+
+ /* x^197632 mod p(x)` << 1, x^197696 mod p(x)` << 1 */
+ .octa 0x00000001f1a0214000000000fe896b7e
+
+ /* x^196608 mod p(x)` << 1, x^196672 mod p(x)` << 1 */
+ .octa 0x00000000ca0b28a000000001f797431c
+
+ /* x^195584 mod p(x)` << 1, x^195648 mod p(x)` << 1 */
+ .octa 0x00000001928e30a20000000053e989ba
+
+ /* x^194560 mod p(x)` << 1, x^194624 mod p(x)` << 1 */
+ .octa 0x0000000097b1b002000000003920cd16
+
+ /* x^193536 mod p(x)` << 1, x^193600 mod p(x)` << 1 */
+ .octa 0x00000000b15bf90600000001e6f579b8
+
+ /* x^192512 mod p(x)` << 1, x^192576 mod p(x)` << 1 */
+ .octa 0x00000000411c5d52000000007493cb0a
+
+ /* x^191488 mod p(x)` << 1, x^191552 mod p(x)` << 1 */
+ .octa 0x00000001c36f330000000001bdd376d8
+
+ /* x^190464 mod p(x)` << 1, x^190528 mod p(x)` << 1 */
+ .octa 0x00000001119227e0000000016badfee6
+
+ /* x^189440 mod p(x)` << 1, x^189504 mod p(x)` << 1 */
+ .octa 0x00000000114d47020000000071de5c58
+
+ /* x^188416 mod p(x)` << 1, x^188480 mod p(x)` << 1 */
+ .octa 0x00000000458b5b9800000000453f317c
+
+ /* x^187392 mod p(x)` << 1, x^187456 mod p(x)` << 1 */
+ .octa 0x000000012e31fb8e0000000121675cce
+
+ /* x^186368 mod p(x)` << 1, x^186432 mod p(x)` << 1 */
+ .octa 0x000000005cf619d800000001f409ee92
+
+ /* x^185344 mod p(x)` << 1, x^185408 mod p(x)` << 1 */
+ .octa 0x0000000063f4d8b200000000f36b9c88
+
+ /* x^184320 mod p(x)` << 1, x^184384 mod p(x)` << 1 */
+ .octa 0x000000004138dc8a0000000036b398f4
+
+ /* x^183296 mod p(x)` << 1, x^183360 mod p(x)` << 1 */
+ .octa 0x00000001d29ee8e000000001748f9adc
+
+ /* x^182272 mod p(x)` << 1, x^182336 mod p(x)` << 1 */
+ .octa 0x000000006a08ace800000001be94ec00
+
+ /* x^181248 mod p(x)` << 1, x^181312 mod p(x)` << 1 */
+ .octa 0x0000000127d4201000000000b74370d6
+
+ /* x^180224 mod p(x)` << 1, x^180288 mod p(x)` << 1 */
+ .octa 0x0000000019d76b6200000001174d0b98
+
+ /* x^179200 mod p(x)` << 1, x^179264 mod p(x)` << 1 */
+ .octa 0x00000001b1471f6e00000000befc06a4
+
+ /* x^178176 mod p(x)` << 1, x^178240 mod p(x)` << 1 */
+ .octa 0x00000001f64c19cc00000001ae125288
+
+ /* x^177152 mod p(x)` << 1, x^177216 mod p(x)` << 1 */
+ .octa 0x00000000003c0ea00000000095c19b34
+
+ /* x^176128 mod p(x)` << 1, x^176192 mod p(x)` << 1 */
+ .octa 0x000000014d73abf600000001a78496f2
+
+ /* x^175104 mod p(x)` << 1, x^175168 mod p(x)` << 1 */
+ .octa 0x00000001620eb84400000001ac5390a0
+
+ /* x^174080 mod p(x)` << 1, x^174144 mod p(x)` << 1 */
+ .octa 0x0000000147655048000000002a80ed6e
+
+ /* x^173056 mod p(x)` << 1, x^173120 mod p(x)` << 1 */
+ .octa 0x0000000067b5077e00000001fa9b0128
+
+ /* x^172032 mod p(x)` << 1, x^172096 mod p(x)` << 1 */
+ .octa 0x0000000010ffe20600000001ea94929e
+
+ /* x^171008 mod p(x)` << 1, x^171072 mod p(x)` << 1 */
+ .octa 0x000000000fee8f1e0000000125f4305c
+
+ /* x^169984 mod p(x)` << 1, x^170048 mod p(x)` << 1 */
+ .octa 0x00000001da26fbae00000001471e2002
+
+ /* x^168960 mod p(x)` << 1, x^169024 mod p(x)` << 1 */
+ .octa 0x00000001b3a8bd880000000132d2253a
+
+ /* x^167936 mod p(x)` << 1, x^168000 mod p(x)` << 1 */
+ .octa 0x00000000e8f3898e00000000f26b3592
+
+ /* x^166912 mod p(x)` << 1, x^166976 mod p(x)` << 1 */
+ .octa 0x00000000b0d0d28c00000000bc8b67b0
+
+ /* x^165888 mod p(x)` << 1, x^165952 mod p(x)` << 1 */
+ .octa 0x0000000030f2a798000000013a826ef2
+
+ /* x^164864 mod p(x)` << 1, x^164928 mod p(x)` << 1 */
+ .octa 0x000000000fba10020000000081482c84
+
+ /* x^163840 mod p(x)` << 1, x^163904 mod p(x)` << 1 */
+ .octa 0x00000000bdb9bd7200000000e77307c2
+
+ /* x^162816 mod p(x)` << 1, x^162880 mod p(x)` << 1 */
+ .octa 0x0000000075d3bf5a00000000d4a07ec8
+
+ /* x^161792 mod p(x)` << 1, x^161856 mod p(x)` << 1 */
+ .octa 0x00000000ef1f98a00000000017102100
+
+ /* x^160768 mod p(x)` << 1, x^160832 mod p(x)` << 1 */
+ .octa 0x00000000689c760200000000db406486
+
+ /* x^159744 mod p(x)` << 1, x^159808 mod p(x)` << 1 */
+ .octa 0x000000016d5fa5fe0000000192db7f88
+
+ /* x^158720 mod p(x)` << 1, x^158784 mod p(x)` << 1 */
+ .octa 0x00000001d0d2b9ca000000018bf67b1e
+
+ /* x^157696 mod p(x)` << 1, x^157760 mod p(x)` << 1 */
+ .octa 0x0000000041e7b470000000007c09163e
+
+ /* x^156672 mod p(x)` << 1, x^156736 mod p(x)` << 1 */
+ .octa 0x00000001cbb6495e000000000adac060
+
+ /* x^155648 mod p(x)` << 1, x^155712 mod p(x)` << 1 */
+ .octa 0x000000010052a0b000000000bd8316ae
+
+ /* x^154624 mod p(x)` << 1, x^154688 mod p(x)` << 1 */
+ .octa 0x00000001d8effb5c000000019f09ab54
+
+ /* x^153600 mod p(x)` << 1, x^153664 mod p(x)` << 1 */
+ .octa 0x00000001d969853c0000000125155542
+
+ /* x^152576 mod p(x)` << 1, x^152640 mod p(x)` << 1 */
+ .octa 0x00000000523ccce2000000018fdb5882
+
+ /* x^151552 mod p(x)` << 1, x^151616 mod p(x)` << 1 */
+ .octa 0x000000001e2436bc00000000e794b3f4
+
+ /* x^150528 mod p(x)` << 1, x^150592 mod p(x)` << 1 */
+ .octa 0x00000000ddd1c3a2000000016f9bb022
+
+ /* x^149504 mod p(x)` << 1, x^149568 mod p(x)` << 1 */
+ .octa 0x0000000019fcfe3800000000290c9978
+
+ /* x^148480 mod p(x)` << 1, x^148544 mod p(x)` << 1 */
+ .octa 0x00000001ce95db640000000083c0f350
+
+ /* x^147456 mod p(x)` << 1, x^147520 mod p(x)` << 1 */
+ .octa 0x00000000af5828060000000173ea6628
+
+ /* x^146432 mod p(x)` << 1, x^146496 mod p(x)` << 1 */
+ .octa 0x00000001006388f600000001c8b4e00a
+
+ /* x^145408 mod p(x)` << 1, x^145472 mod p(x)` << 1 */
+ .octa 0x0000000179eca00a00000000de95d6aa
+
+ /* x^144384 mod p(x)` << 1, x^144448 mod p(x)` << 1 */
+ .octa 0x0000000122410a6a000000010b7f7248
+
+ /* x^143360 mod p(x)` << 1, x^143424 mod p(x)` << 1 */
+ .octa 0x000000004288e87c00000001326e3a06
+
+ /* x^142336 mod p(x)` << 1, x^142400 mod p(x)` << 1 */
+ .octa 0x000000016c5490da00000000bb62c2e6
+
+ /* x^141312 mod p(x)` << 1, x^141376 mod p(x)` << 1 */
+ .octa 0x00000000d1c71f6e0000000156a4b2c2
+
+ /* x^140288 mod p(x)` << 1, x^140352 mod p(x)` << 1 */
+ .octa 0x00000001b4ce08a6000000011dfe763a
+
+ /* x^139264 mod p(x)` << 1, x^139328 mod p(x)` << 1 */
+ .octa 0x00000001466ba60c000000007bcca8e2
+
+ /* x^138240 mod p(x)` << 1, x^138304 mod p(x)` << 1 */
+ .octa 0x00000001f6c488a40000000186118faa
+
+ /* x^137216 mod p(x)` << 1, x^137280 mod p(x)` << 1 */
+ .octa 0x000000013bfb06820000000111a65a88
+
+ /* x^136192 mod p(x)` << 1, x^136256 mod p(x)` << 1 */
+ .octa 0x00000000690e9e54000000003565e1c4
+
+ /* x^135168 mod p(x)` << 1, x^135232 mod p(x)` << 1 */
+ .octa 0x00000000281346b6000000012ed02a82
+
+ /* x^134144 mod p(x)` << 1, x^134208 mod p(x)` << 1 */
+ .octa 0x000000015646402400000000c486ecfc
+
+ /* x^133120 mod p(x)` << 1, x^133184 mod p(x)` << 1 */
+ .octa 0x000000016063a8dc0000000001b951b2
+
+ /* x^132096 mod p(x)` << 1, x^132160 mod p(x)` << 1 */
+ .octa 0x0000000116a663620000000048143916
+
+ /* x^131072 mod p(x)` << 1, x^131136 mod p(x)` << 1 */
+ .octa 0x000000017e8aa4d200000001dc2ae124
+
+ /* x^130048 mod p(x)` << 1, x^130112 mod p(x)` << 1 */
+ .octa 0x00000001728eb10c00000001416c58d6
+
+ /* x^129024 mod p(x)` << 1, x^129088 mod p(x)` << 1 */
+ .octa 0x00000001b08fd7fa00000000a479744a
+
+ /* x^128000 mod p(x)` << 1, x^128064 mod p(x)` << 1 */
+ .octa 0x00000001092a16e80000000096ca3a26
+
+ /* x^126976 mod p(x)` << 1, x^127040 mod p(x)` << 1 */
+ .octa 0x00000000a505637c00000000ff223d4e
+
+ /* x^125952 mod p(x)` << 1, x^126016 mod p(x)` << 1 */
+ .octa 0x00000000d94869b2000000010e84da42
+
+ /* x^124928 mod p(x)` << 1, x^124992 mod p(x)` << 1 */
+ .octa 0x00000001c8b203ae00000001b61ba3d0
+
+ /* x^123904 mod p(x)` << 1, x^123968 mod p(x)` << 1 */
+ .octa 0x000000005704aea000000000680f2de8
+
+ /* x^122880 mod p(x)` << 1, x^122944 mod p(x)` << 1 */
+ .octa 0x000000012e295fa2000000008772a9a8
+
+ /* x^121856 mod p(x)` << 1, x^121920 mod p(x)` << 1 */
+ .octa 0x000000011d0908bc0000000155f295bc
+
+ /* x^120832 mod p(x)` << 1, x^120896 mod p(x)` << 1 */
+ .octa 0x0000000193ed97ea00000000595f9282
+
+ /* x^119808 mod p(x)` << 1, x^119872 mod p(x)` << 1 */
+ .octa 0x000000013a0f1c520000000164b1c25a
+
+ /* x^118784 mod p(x)` << 1, x^118848 mod p(x)` << 1 */
+ .octa 0x000000010c2c40c000000000fbd67c50
+
+ /* x^117760 mod p(x)` << 1, x^117824 mod p(x)` << 1 */
+ .octa 0x00000000ff6fac3e0000000096076268
+
+ /* x^116736 mod p(x)` << 1, x^116800 mod p(x)` << 1 */
+ .octa 0x000000017b3609c000000001d288e4cc
+
+ /* x^115712 mod p(x)` << 1, x^115776 mod p(x)` << 1 */
+ .octa 0x0000000088c8c92200000001eaac1bdc
+
+ /* x^114688 mod p(x)` << 1, x^114752 mod p(x)` << 1 */
+ .octa 0x00000001751baae600000001f1ea39e2
+
+ /* x^113664 mod p(x)` << 1, x^113728 mod p(x)` << 1 */
+ .octa 0x000000010795297200000001eb6506fc
+
+ /* x^112640 mod p(x)` << 1, x^112704 mod p(x)` << 1 */
+ .octa 0x0000000162b00abe000000010f806ffe
+
+ /* x^111616 mod p(x)` << 1, x^111680 mod p(x)` << 1 */
+ .octa 0x000000000d7b404c000000010408481e
+
+ /* x^110592 mod p(x)` << 1, x^110656 mod p(x)` << 1 */
+ .octa 0x00000000763b13d40000000188260534
+
+ /* x^109568 mod p(x)` << 1, x^109632 mod p(x)` << 1 */
+ .octa 0x00000000f6dc22d80000000058fc73e0
+
+ /* x^108544 mod p(x)` << 1, x^108608 mod p(x)` << 1 */
+ .octa 0x000000007daae06000000000391c59b8
+
+ /* x^107520 mod p(x)` << 1, x^107584 mod p(x)` << 1 */
+ .octa 0x000000013359ab7c000000018b638400
+
+ /* x^106496 mod p(x)` << 1, x^106560 mod p(x)` << 1 */
+ .octa 0x000000008add438a000000011738f5c4
+
+ /* x^105472 mod p(x)` << 1, x^105536 mod p(x)` << 1 */
+ .octa 0x00000001edbefdea000000008cf7c6da
+
+ /* x^104448 mod p(x)` << 1, x^104512 mod p(x)` << 1 */
+ .octa 0x000000004104e0f800000001ef97fb16
+
+ /* x^103424 mod p(x)` << 1, x^103488 mod p(x)` << 1 */
+ .octa 0x00000000b48a82220000000102130e20
+
+ /* x^102400 mod p(x)` << 1, x^102464 mod p(x)` << 1 */
+ .octa 0x00000001bcb4684400000000db968898
+
+ /* x^101376 mod p(x)` << 1, x^101440 mod p(x)` << 1 */
+ .octa 0x000000013293ce0a00000000b5047b5e
+
+ /* x^100352 mod p(x)` << 1, x^100416 mod p(x)` << 1 */
+ .octa 0x00000001710d0844000000010b90fdb2
+
+ /* x^99328 mod p(x)` << 1, x^99392 mod p(x)` << 1 */
+ .octa 0x0000000117907f6e000000004834a32e
+
+ /* x^98304 mod p(x)` << 1, x^98368 mod p(x)` << 1 */
+ .octa 0x0000000087ddf93e0000000059c8f2b0
+
+ /* x^97280 mod p(x)` << 1, x^97344 mod p(x)` << 1 */
+ .octa 0x000000005970e9b00000000122cec508
+
+ /* x^96256 mod p(x)` << 1, x^96320 mod p(x)` << 1 */
+ .octa 0x0000000185b2b7d0000000000a330cda
+
+ /* x^95232 mod p(x)` << 1, x^95296 mod p(x)` << 1 */
+ .octa 0x00000001dcee0efc000000014a47148c
+
+ /* x^94208 mod p(x)` << 1, x^94272 mod p(x)` << 1 */
+ .octa 0x0000000030da27220000000042c61cb8
+
+ /* x^93184 mod p(x)` << 1, x^93248 mod p(x)` << 1 */
+ .octa 0x000000012f925a180000000012fe6960
+
+ /* x^92160 mod p(x)` << 1, x^92224 mod p(x)` << 1 */
+ .octa 0x00000000dd2e357c00000000dbda2c20
+
+ /* x^91136 mod p(x)` << 1, x^91200 mod p(x)` << 1 */
+ .octa 0x00000000071c80de000000011122410c
+
+ /* x^90112 mod p(x)` << 1, x^90176 mod p(x)` << 1 */
+ .octa 0x000000011513140a00000000977b2070
+
+ /* x^89088 mod p(x)` << 1, x^89152 mod p(x)` << 1 */
+ .octa 0x00000001df876e8e000000014050438e
+
+ /* x^88064 mod p(x)` << 1, x^88128 mod p(x)` << 1 */
+ .octa 0x000000015f81d6ce0000000147c840e8
+
+ /* x^87040 mod p(x)` << 1, x^87104 mod p(x)` << 1 */
+ .octa 0x000000019dd94dbe00000001cc7c88ce
+
+ /* x^86016 mod p(x)` << 1, x^86080 mod p(x)` << 1 */
+ .octa 0x00000001373d206e00000001476b35a4
+
+ /* x^84992 mod p(x)` << 1, x^85056 mod p(x)` << 1 */
+ .octa 0x00000000668ccade000000013d52d508
+
+ /* x^83968 mod p(x)` << 1, x^84032 mod p(x)` << 1 */
+ .octa 0x00000001b192d268000000008e4be32e
+
+ /* x^82944 mod p(x)` << 1, x^83008 mod p(x)` << 1 */
+ .octa 0x00000000e30f3a7800000000024120fe
+
+ /* x^81920 mod p(x)` << 1, x^81984 mod p(x)` << 1 */
+ .octa 0x000000010ef1f7bc00000000ddecddb4
+
+ /* x^80896 mod p(x)` << 1, x^80960 mod p(x)` << 1 */
+ .octa 0x00000001f5ac738000000000d4d403bc
+
+ /* x^79872 mod p(x)` << 1, x^79936 mod p(x)` << 1 */
+ .octa 0x000000011822ea7000000001734b89aa
+
+ /* x^78848 mod p(x)` << 1, x^78912 mod p(x)` << 1 */
+ .octa 0x00000000c3a33848000000010e7a58d6
+
+ /* x^77824 mod p(x)` << 1, x^77888 mod p(x)` << 1 */
+ .octa 0x00000001bd151c2400000001f9f04e9c
+
+ /* x^76800 mod p(x)` << 1, x^76864 mod p(x)` << 1 */
+ .octa 0x0000000056002d7600000000b692225e
+
+ /* x^75776 mod p(x)` << 1, x^75840 mod p(x)` << 1 */
+ .octa 0x000000014657c4f4000000019b8d3f3e
+
+ /* x^74752 mod p(x)` << 1, x^74816 mod p(x)` << 1 */
+ .octa 0x0000000113742d7c00000001a874f11e
+
+ /* x^73728 mod p(x)` << 1, x^73792 mod p(x)` << 1 */
+ .octa 0x000000019c5920ba000000010d5a4254
+
+ /* x^72704 mod p(x)` << 1, x^72768 mod p(x)` << 1 */
+ .octa 0x000000005216d2d600000000bbb2f5d6
+
+ /* x^71680 mod p(x)` << 1, x^71744 mod p(x)` << 1 */
+ .octa 0x0000000136f5ad8a0000000179cc0e36
+
+ /* x^70656 mod p(x)` << 1, x^70720 mod p(x)` << 1 */
+ .octa 0x000000018b07beb600000001dca1da4a
+
+ /* x^69632 mod p(x)` << 1, x^69696 mod p(x)` << 1 */
+ .octa 0x00000000db1e93b000000000feb1a192
+
+ /* x^68608 mod p(x)` << 1, x^68672 mod p(x)` << 1 */
+ .octa 0x000000000b96fa3a00000000d1eeedd6
+
+ /* x^67584 mod p(x)` << 1, x^67648 mod p(x)` << 1 */
+ .octa 0x00000001d9968af0000000008fad9bb4
+
+ /* x^66560 mod p(x)` << 1, x^66624 mod p(x)` << 1 */
+ .octa 0x000000000e4a77a200000001884938e4
+
+ /* x^65536 mod p(x)` << 1, x^65600 mod p(x)` << 1 */
+ .octa 0x00000000508c2ac800000001bc2e9bc0
+
+ /* x^64512 mod p(x)` << 1, x^64576 mod p(x)` << 1 */
+ .octa 0x0000000021572a8000000001f9658a68
+
+ /* x^63488 mod p(x)` << 1, x^63552 mod p(x)` << 1 */
+ .octa 0x00000001b859daf2000000001b9224fc
+
+ /* x^62464 mod p(x)` << 1, x^62528 mod p(x)` << 1 */
+ .octa 0x000000016f7884740000000055b2fb84
+
+ /* x^61440 mod p(x)` << 1, x^61504 mod p(x)` << 1 */
+ .octa 0x00000001b438810e000000018b090348
+
+ /* x^60416 mod p(x)` << 1, x^60480 mod p(x)` << 1 */
+ .octa 0x0000000095ddc6f2000000011ccbd5ea
+
+ /* x^59392 mod p(x)` << 1, x^59456 mod p(x)` << 1 */
+ .octa 0x00000001d977c20c0000000007ae47f8
+
+ /* x^58368 mod p(x)` << 1, x^58432 mod p(x)` << 1 */
+ .octa 0x00000000ebedb99a0000000172acbec0
+
+ /* x^57344 mod p(x)` << 1, x^57408 mod p(x)` << 1 */
+ .octa 0x00000001df9e9e9200000001c6e3ff20
+
+ /* x^56320 mod p(x)` << 1, x^56384 mod p(x)` << 1 */
+ .octa 0x00000001a4a3f95200000000e1b38744
+
+ /* x^55296 mod p(x)` << 1, x^55360 mod p(x)` << 1 */
+ .octa 0x00000000e2f5122000000000791585b2
+
+ /* x^54272 mod p(x)` << 1, x^54336 mod p(x)` << 1 */
+ .octa 0x000000004aa01f3e00000000ac53b894
+
+ /* x^53248 mod p(x)` << 1, x^53312 mod p(x)` << 1 */
+ .octa 0x00000000b3e90a5800000001ed5f2cf4
+
+ /* x^52224 mod p(x)` << 1, x^52288 mod p(x)` << 1 */
+ .octa 0x000000000c9ca2aa00000001df48b2e0
+
+ /* x^51200 mod p(x)` << 1, x^51264 mod p(x)` << 1 */
+ .octa 0x000000015168231600000000049c1c62
+
+ /* x^50176 mod p(x)` << 1, x^50240 mod p(x)` << 1 */
+ .octa 0x0000000036fce78c000000017c460c12
+
+ /* x^49152 mod p(x)` << 1, x^49216 mod p(x)` << 1 */
+ .octa 0x000000009037dc10000000015be4da7e
+
+ /* x^48128 mod p(x)` << 1, x^48192 mod p(x)` << 1 */
+ .octa 0x00000000d3298582000000010f38f668
+
+ /* x^47104 mod p(x)` << 1, x^47168 mod p(x)` << 1 */
+ .octa 0x00000001b42e8ad60000000039f40a00
+
+ /* x^46080 mod p(x)` << 1, x^46144 mod p(x)` << 1 */
+ .octa 0x00000000142a983800000000bd4c10c4
+
+ /* x^45056 mod p(x)` << 1, x^45120 mod p(x)` << 1 */
+ .octa 0x0000000109c7f1900000000042db1d98
+
+ /* x^44032 mod p(x)` << 1, x^44096 mod p(x)` << 1 */
+ .octa 0x0000000056ff931000000001c905bae6
+
+ /* x^43008 mod p(x)` << 1, x^43072 mod p(x)` << 1 */
+ .octa 0x00000001594513aa00000000069d40ea
+
+ /* x^41984 mod p(x)` << 1, x^42048 mod p(x)` << 1 */
+ .octa 0x00000001e3b5b1e8000000008e4fbad0
+
+ /* x^40960 mod p(x)` << 1, x^41024 mod p(x)` << 1 */
+ .octa 0x000000011dd5fc080000000047bedd46
+
+ /* x^39936 mod p(x)` << 1, x^40000 mod p(x)` << 1 */
+ .octa 0x00000001675f0cc20000000026396bf8
+
+ /* x^38912 mod p(x)` << 1, x^38976 mod p(x)` << 1 */
+ .octa 0x00000000d1c8dd4400000000379beb92
+
+ /* x^37888 mod p(x)` << 1, x^37952 mod p(x)` << 1 */
+ .octa 0x0000000115ebd3d8000000000abae54a
+
+ /* x^36864 mod p(x)` << 1, x^36928 mod p(x)` << 1 */
+ .octa 0x00000001ecbd0dac0000000007e6a128
+
+ /* x^35840 mod p(x)` << 1, x^35904 mod p(x)` << 1 */
+ .octa 0x00000000cdf67af2000000000ade29d2
+
+ /* x^34816 mod p(x)` << 1, x^34880 mod p(x)` << 1 */
+ .octa 0x000000004c01ff4c00000000f974c45c
+
+ /* x^33792 mod p(x)` << 1, x^33856 mod p(x)` << 1 */
+ .octa 0x00000000f2d8657e00000000e77ac60a
+
+ /* x^32768 mod p(x)` << 1, x^32832 mod p(x)` << 1 */
+ .octa 0x000000006bae74c40000000145895816
+
+ /* x^31744 mod p(x)` << 1, x^31808 mod p(x)` << 1 */
+ .octa 0x0000000152af8aa00000000038e362be
+
+ /* x^30720 mod p(x)` << 1, x^30784 mod p(x)` << 1 */
+ .octa 0x0000000004663802000000007f991a64
+
+ /* x^29696 mod p(x)` << 1, x^29760 mod p(x)` << 1 */
+ .octa 0x00000001ab2f5afc00000000fa366d3a
+
+ /* x^28672 mod p(x)` << 1, x^28736 mod p(x)` << 1 */
+ .octa 0x0000000074a4ebd400000001a2bb34f0
+
+ /* x^27648 mod p(x)` << 1, x^27712 mod p(x)` << 1 */
+ .octa 0x00000001d7ab3a4c0000000028a9981e
+
+ /* x^26624 mod p(x)` << 1, x^26688 mod p(x)` << 1 */
+ .octa 0x00000001a8da60c600000001dbc672be
+
+ /* x^25600 mod p(x)` << 1, x^25664 mod p(x)` << 1 */
+ .octa 0x000000013cf6382000000000b04d77f6
+
+ /* x^24576 mod p(x)` << 1, x^24640 mod p(x)` << 1 */
+ .octa 0x00000000bec12e1e0000000124400d96
+
+ /* x^23552 mod p(x)` << 1, x^23616 mod p(x)` << 1 */
+ .octa 0x00000001c6368010000000014ca4b414
+
+ /* x^22528 mod p(x)` << 1, x^22592 mod p(x)` << 1 */
+ .octa 0x00000001e6e78758000000012fe2c938
+
+ /* x^21504 mod p(x)` << 1, x^21568 mod p(x)` << 1 */
+ .octa 0x000000008d7f2b3c00000001faed01e6
+
+ /* x^20480 mod p(x)` << 1, x^20544 mod p(x)` << 1 */
+ .octa 0x000000016b4a156e000000007e80ecfe
+
+ /* x^19456 mod p(x)` << 1, x^19520 mod p(x)` << 1 */
+ .octa 0x00000001c63cfeb60000000098daee94
+
+ /* x^18432 mod p(x)` << 1, x^18496 mod p(x)` << 1 */
+ .octa 0x000000015f902670000000010a04edea
+
+ /* x^17408 mod p(x)` << 1, x^17472 mod p(x)` << 1 */
+ .octa 0x00000001cd5de11e00000001c00b4524
+
+ /* x^16384 mod p(x)` << 1, x^16448 mod p(x)` << 1 */
+ .octa 0x000000001acaec540000000170296550
+
+ /* x^15360 mod p(x)` << 1, x^15424 mod p(x)` << 1 */
+ .octa 0x000000002bd0ca780000000181afaa48
+
+ /* x^14336 mod p(x)` << 1, x^14400 mod p(x)` << 1 */
+ .octa 0x0000000032d63d5c0000000185a31ffa
+
+ /* x^13312 mod p(x)` << 1, x^13376 mod p(x)` << 1 */
+ .octa 0x000000001c6d4e4c000000002469f608
+
+ /* x^12288 mod p(x)` << 1, x^12352 mod p(x)` << 1 */
+ .octa 0x0000000106a60b92000000006980102a
+
+ /* x^11264 mod p(x)` << 1, x^11328 mod p(x)` << 1 */
+ .octa 0x00000000d3855e120000000111ea9ca8
+
+ /* x^10240 mod p(x)` << 1, x^10304 mod p(x)` << 1 */
+ .octa 0x00000000e312563600000001bd1d29ce
+
+ /* x^9216 mod p(x)` << 1, x^9280 mod p(x)` << 1 */
+ .octa 0x000000009e8f7ea400000001b34b9580
+
+ /* x^8192 mod p(x)` << 1, x^8256 mod p(x)` << 1 */
+ .octa 0x00000001c82e562c000000003076054e
+
+ /* x^7168 mod p(x)` << 1, x^7232 mod p(x)` << 1 */
+ .octa 0x00000000ca9f09ce000000012a608ea4
+
+ /* x^6144 mod p(x)` << 1, x^6208 mod p(x)` << 1 */
+ .octa 0x00000000c63764e600000000784d05fe
+
+ /* x^5120 mod p(x)` << 1, x^5184 mod p(x)` << 1 */
+ .octa 0x0000000168d2e49e000000016ef0d82a
+
+ /* x^4096 mod p(x)` << 1, x^4160 mod p(x)` << 1 */
+ .octa 0x00000000e986c1480000000075bda454
+
+ /* x^3072 mod p(x)` << 1, x^3136 mod p(x)` << 1 */
+ .octa 0x00000000cfb65894000000003dc0a1c4
+
+ /* x^2048 mod p(x)` << 1, x^2112 mod p(x)` << 1 */
+ .octa 0x0000000111cadee400000000e9a5d8be
+
+ /* x^1024 mod p(x)` << 1, x^1088 mod p(x)` << 1 */
+ .octa 0x0000000171fb63ce00000001609bc4b4
+
+.short_constants:
+
+ /* Reduce final 1024-2048 bits to 64 bits, shifting 32 bits to include the trailing 32 bits of zeros */
+ /* x^1952 mod p(x)`, x^1984 mod p(x)`, x^2016 mod p(x)`, x^2048 mod p(x)` */
+ .octa 0x7fec2963e5bf80485cf015c388e56f72
+
+ /* x^1824 mod p(x)`, x^1856 mod p(x)`, x^1888 mod p(x)`, x^1920 mod p(x)` */
+ .octa 0x38e888d4844752a9963a18920246e2e6
+
+ /* x^1696 mod p(x)`, x^1728 mod p(x)`, x^1760 mod p(x)`, x^1792 mod p(x)` */
+ .octa 0x42316c00730206ad419a441956993a31
+
+ /* x^1568 mod p(x)`, x^1600 mod p(x)`, x^1632 mod p(x)`, x^1664 mod p(x)` */
+ .octa 0x543d5c543e65ddf9924752ba2b830011
+
+ /* x^1440 mod p(x)`, x^1472 mod p(x)`, x^1504 mod p(x)`, x^1536 mod p(x)` */
+ .octa 0x78e87aaf56767c9255bd7f9518e4a304
+
+ /* x^1312 mod p(x)`, x^1344 mod p(x)`, x^1376 mod p(x)`, x^1408 mod p(x)` */
+ .octa 0x8f68fcec1903da7f6d76739fe0553f1e
+
+ /* x^1184 mod p(x)`, x^1216 mod p(x)`, x^1248 mod p(x)`, x^1280 mod p(x)` */
+ .octa 0x3f4840246791d588c133722b1fe0b5c3
+
+ /* x^1056 mod p(x)`, x^1088 mod p(x)`, x^1120 mod p(x)`, x^1152 mod p(x)` */
+ .octa 0x34c96751b04de25a64b67ee0e55ef1f3
+
+ /* x^928 mod p(x)`, x^960 mod p(x)`, x^992 mod p(x)`, x^1024 mod p(x)` */
+ .octa 0x156c8e180b4a395b069db049b8fdb1e7
+
+ /* x^800 mod p(x)`, x^832 mod p(x)`, x^864 mod p(x)`, x^896 mod p(x)` */
+ .octa 0xe0b99ccbe661f7bea11bfaf3c9e90b9e
+
+ /* x^672 mod p(x)`, x^704 mod p(x)`, x^736 mod p(x)`, x^768 mod p(x)` */
+ .octa 0x041d37768cd75659817cdc5119b29a35
+
+ /* x^544 mod p(x)`, x^576 mod p(x)`, x^608 mod p(x)`, x^640 mod p(x)` */
+ .octa 0x3a0777818cfaa9651ce9d94b36c41f1c
+
+ /* x^416 mod p(x)`, x^448 mod p(x)`, x^480 mod p(x)`, x^512 mod p(x)` */
+ .octa 0x0e148e8252377a554f256efcb82be955
+
+ /* x^288 mod p(x)`, x^320 mod p(x)`, x^352 mod p(x)`, x^384 mod p(x)` */
+ .octa 0x9c25531d19e65ddeec1631edb2dea967
+
+ /* x^160 mod p(x)`, x^192 mod p(x)`, x^224 mod p(x)`, x^256 mod p(x)` */
+ .octa 0x790606ff9957c0a65d27e147510ac59a
+
+ /* x^32 mod p(x)`, x^64 mod p(x)`, x^96 mod p(x)`, x^128 mod p(x)` */
+ .octa 0x82f63b786ea2d55ca66805eb18b8ea18
+
+
+.barrett_constants:
+ /* 33 bit reflected Barrett constant m - (4^32)/n */
+ .octa 0x000000000000000000000000dea713f1 /* x^64 div p(x)` */
+ /* 33 bit reflected Barrett constant n */
+ .octa 0x00000000000000000000000105ec76f1
+
+ .text
+
+#if defined(__BIG_ENDIAN__)
+#define BYTESWAP_DATA
+#else
+#undef BYTESWAP_DATA
+#endif
+
+#define off16 r25
+#define off32 r26
+#define off48 r27
+#define off64 r28
+#define off80 r29
+#define off96 r30
+#define off112 r31
+
+#define const1 v24
+#define const2 v25
+
+#define byteswap v26
+#define mask_32bit v27
+#define mask_64bit v28
+#define zeroes v29
+
+#ifdef BYTESWAP_DATA
+#define VPERM(A, B, C, D) vperm A, B, C, D
+#else
+#define VPERM(A, B, C, D)
+#endif
+
+/* unsigned int __crc32c_vpmsum(unsigned int crc, void *p, unsigned long len) */
+FUNC_START(__crc32c_vpmsum)
+ std r31,-8(r1)
+ std r30,-16(r1)
+ std r29,-24(r1)
+ std r28,-32(r1)
+ std r27,-40(r1)
+ std r26,-48(r1)
+ std r25,-56(r1)
+
+ li off16,16
+ li off32,32
+ li off48,48
+ li off64,64
+ li off80,80
+ li off96,96
+ li off112,112
+ li r0,0
+
+ /* Enough room for saving 10 non volatile VMX registers */
+ subi r6,r1,56+10*16
+ subi r7,r1,56+2*16
+
+ stvx v20,0,r6
+ stvx v21,off16,r6
+ stvx v22,off32,r6
+ stvx v23,off48,r6
+ stvx v24,off64,r6
+ stvx v25,off80,r6
+ stvx v26,off96,r6
+ stvx v27,off112,r6
+ stvx v28,0,r7
+ stvx v29,off16,r7
+
+ mr r10,r3
+
+ vxor zeroes,zeroes,zeroes
+ vspltisw v0,-1
+
+ vsldoi mask_32bit,zeroes,v0,4
+ vsldoi mask_64bit,zeroes,v0,8
+
+ /* Get the initial value into v8 */
+ vxor v8,v8,v8
+ MTVRD(v8, R3)
+ vsldoi v8,zeroes,v8,8 /* shift into bottom 32 bits */
+
+#ifdef BYTESWAP_DATA
+ addis r3,r2,.byteswap_constant@toc@ha
+ addi r3,r3,.byteswap_constant@toc@l
+
+ lvx byteswap,0,r3
+ addi r3,r3,16
+#endif
+
+ cmpdi r5,256
+ blt .Lshort
+
+ rldicr r6,r5,0,56
+
+ /* Checksum in blocks of MAX_SIZE */
+1: lis r7,MAX_SIZE@h
+ ori r7,r7,MAX_SIZE@l
+ mr r9,r7
+ cmpd r6,r7
+ bgt 2f
+ mr r7,r6
+2: subf r6,r7,r6
+
+ /* our main loop does 128 bytes at a time */
+ srdi r7,r7,7
+
+ /*
+ * Work out the offset into the constants table to start at. Each
+ * constant is 16 bytes, and it is used against 128 bytes of input
+ * data - 128 / 16 = 8
+ */
+ sldi r8,r7,4
+ srdi r9,r9,3
+ subf r8,r8,r9
+
+ /* We reduce our final 128 bytes in a separate step */
+ addi r7,r7,-1
+ mtctr r7
+
+ addis r3,r2,.constants@toc@ha
+ addi r3,r3,.constants@toc@l
+
+ /* Find the start of our constants */
+ add r3,r3,r8
+
+ /* zero v0-v7 which will contain our checksums */
+ vxor v0,v0,v0
+ vxor v1,v1,v1
+ vxor v2,v2,v2
+ vxor v3,v3,v3
+ vxor v4,v4,v4
+ vxor v5,v5,v5
+ vxor v6,v6,v6
+ vxor v7,v7,v7
+
+ lvx const1,0,r3
+
+ /*
+ * If we are looping back to consume more data we use the values
+ * already in v16-v23.
+ */
+ cmpdi r0,1
+ beq 2f
+
+ /* First warm up pass */
+ lvx v16,0,r4
+ lvx v17,off16,r4
+ VPERM(v16,v16,v16,byteswap)
+ VPERM(v17,v17,v17,byteswap)
+ lvx v18,off32,r4
+ lvx v19,off48,r4
+ VPERM(v18,v18,v18,byteswap)
+ VPERM(v19,v19,v19,byteswap)
+ lvx v20,off64,r4
+ lvx v21,off80,r4
+ VPERM(v20,v20,v20,byteswap)
+ VPERM(v21,v21,v21,byteswap)
+ lvx v22,off96,r4
+ lvx v23,off112,r4
+ VPERM(v22,v22,v22,byteswap)
+ VPERM(v23,v23,v23,byteswap)
+ addi r4,r4,8*16
+
+ /* xor in initial value */
+ vxor v16,v16,v8
+
+2: bdz .Lfirst_warm_up_done
+
+ addi r3,r3,16
+ lvx const2,0,r3
+
+ /* Second warm up pass */
+ VPMSUMD(v8,v16,const1)
+ lvx v16,0,r4
+ VPERM(v16,v16,v16,byteswap)
+ ori r2,r2,0
+
+ VPMSUMD(v9,v17,const1)
+ lvx v17,off16,r4
+ VPERM(v17,v17,v17,byteswap)
+ ori r2,r2,0
+
+ VPMSUMD(v10,v18,const1)
+ lvx v18,off32,r4
+ VPERM(v18,v18,v18,byteswap)
+ ori r2,r2,0
+
+ VPMSUMD(v11,v19,const1)
+ lvx v19,off48,r4
+ VPERM(v19,v19,v19,byteswap)
+ ori r2,r2,0
+
+ VPMSUMD(v12,v20,const1)
+ lvx v20,off64,r4
+ VPERM(v20,v20,v20,byteswap)
+ ori r2,r2,0
+
+ VPMSUMD(v13,v21,const1)
+ lvx v21,off80,r4
+ VPERM(v21,v21,v21,byteswap)
+ ori r2,r2,0
+
+ VPMSUMD(v14,v22,const1)
+ lvx v22,off96,r4
+ VPERM(v22,v22,v22,byteswap)
+ ori r2,r2,0
+
+ VPMSUMD(v15,v23,const1)
+ lvx v23,off112,r4
+ VPERM(v23,v23,v23,byteswap)
+
+ addi r4,r4,8*16
+
+ bdz .Lfirst_cool_down
+
+ /*
+ * main loop. We modulo schedule it such that it takes three iterations
+ * to complete - first iteration load, second iteration vpmsum, third
+ * iteration xor.
+ */
+ .balign 16
+4: lvx const1,0,r3
+ addi r3,r3,16
+ ori r2,r2,0
+
+ vxor v0,v0,v8
+ VPMSUMD(v8,v16,const2)
+ lvx v16,0,r4
+ VPERM(v16,v16,v16,byteswap)
+ ori r2,r2,0
+
+ vxor v1,v1,v9
+ VPMSUMD(v9,v17,const2)
+ lvx v17,off16,r4
+ VPERM(v17,v17,v17,byteswap)
+ ori r2,r2,0
+
+ vxor v2,v2,v10
+ VPMSUMD(v10,v18,const2)
+ lvx v18,off32,r4
+ VPERM(v18,v18,v18,byteswap)
+ ori r2,r2,0
+
+ vxor v3,v3,v11
+ VPMSUMD(v11,v19,const2)
+ lvx v19,off48,r4
+ VPERM(v19,v19,v19,byteswap)
+ lvx const2,0,r3
+ ori r2,r2,0
+
+ vxor v4,v4,v12
+ VPMSUMD(v12,v20,const1)
+ lvx v20,off64,r4
+ VPERM(v20,v20,v20,byteswap)
+ ori r2,r2,0
+
+ vxor v5,v5,v13
+ VPMSUMD(v13,v21,const1)
+ lvx v21,off80,r4
+ VPERM(v21,v21,v21,byteswap)
+ ori r2,r2,0
+
+ vxor v6,v6,v14
+ VPMSUMD(v14,v22,const1)
+ lvx v22,off96,r4
+ VPERM(v22,v22,v22,byteswap)
+ ori r2,r2,0
+
+ vxor v7,v7,v15
+ VPMSUMD(v15,v23,const1)
+ lvx v23,off112,r4
+ VPERM(v23,v23,v23,byteswap)
+
+ addi r4,r4,8*16
+
+ bdnz 4b
+
+.Lfirst_cool_down:
+ /* First cool down pass */
+ lvx const1,0,r3
+ addi r3,r3,16
+
+ vxor v0,v0,v8
+ VPMSUMD(v8,v16,const1)
+ ori r2,r2,0
+
+ vxor v1,v1,v9
+ VPMSUMD(v9,v17,const1)
+ ori r2,r2,0
+
+ vxor v2,v2,v10
+ VPMSUMD(v10,v18,const1)
+ ori r2,r2,0
+
+ vxor v3,v3,v11
+ VPMSUMD(v11,v19,const1)
+ ori r2,r2,0
+
+ vxor v4,v4,v12
+ VPMSUMD(v12,v20,const1)
+ ori r2,r2,0
+
+ vxor v5,v5,v13
+ VPMSUMD(v13,v21,const1)
+ ori r2,r2,0
+
+ vxor v6,v6,v14
+ VPMSUMD(v14,v22,const1)
+ ori r2,r2,0
+
+ vxor v7,v7,v15
+ VPMSUMD(v15,v23,const1)
+ ori r2,r2,0
+
+.Lsecond_cool_down:
+ /* Second cool down pass */
+ vxor v0,v0,v8
+ vxor v1,v1,v9
+ vxor v2,v2,v10
+ vxor v3,v3,v11
+ vxor v4,v4,v12
+ vxor v5,v5,v13
+ vxor v6,v6,v14
+ vxor v7,v7,v15
+
+ /*
+ * vpmsumd produces a 96 bit result in the least significant bits
+ * of the register. Since we are bit reflected we have to shift it
+ * left 32 bits so it occupies the least significant bits in the
+ * bit reflected domain.
+ */
+ vsldoi v0,v0,zeroes,4
+ vsldoi v1,v1,zeroes,4
+ vsldoi v2,v2,zeroes,4
+ vsldoi v3,v3,zeroes,4
+ vsldoi v4,v4,zeroes,4
+ vsldoi v5,v5,zeroes,4
+ vsldoi v6,v6,zeroes,4
+ vsldoi v7,v7,zeroes,4
+
+ /* xor with last 1024 bits */
+ lvx v8,0,r4
+ lvx v9,off16,r4
+ VPERM(v8,v8,v8,byteswap)
+ VPERM(v9,v9,v9,byteswap)
+ lvx v10,off32,r4
+ lvx v11,off48,r4
+ VPERM(v10,v10,v10,byteswap)
+ VPERM(v11,v11,v11,byteswap)
+ lvx v12,off64,r4
+ lvx v13,off80,r4
+ VPERM(v12,v12,v12,byteswap)
+ VPERM(v13,v13,v13,byteswap)
+ lvx v14,off96,r4
+ lvx v15,off112,r4
+ VPERM(v14,v14,v14,byteswap)
+ VPERM(v15,v15,v15,byteswap)
+
+ addi r4,r4,8*16
+
+ vxor v16,v0,v8
+ vxor v17,v1,v9
+ vxor v18,v2,v10
+ vxor v19,v3,v11
+ vxor v20,v4,v12
+ vxor v21,v5,v13
+ vxor v22,v6,v14
+ vxor v23,v7,v15
+
+ li r0,1
+ cmpdi r6,0
+ addi r6,r6,128
+ bne 1b
+
+ /* Work out how many bytes we have left */
+ andi. r5,r5,127
+
+ /* Calculate where in the constant table we need to start */
+ subfic r6,r5,128
+ add r3,r3,r6
+
+ /* How many 16 byte chunks are in the tail */
+ srdi r7,r5,4
+ mtctr r7
+
+ /*
+ * Reduce the previously calculated 1024 bits to 64 bits, shifting
+ * 32 bits to include the trailing 32 bits of zeros
+ */
+ lvx v0,0,r3
+ lvx v1,off16,r3
+ lvx v2,off32,r3
+ lvx v3,off48,r3
+ lvx v4,off64,r3
+ lvx v5,off80,r3
+ lvx v6,off96,r3
+ lvx v7,off112,r3
+ addi r3,r3,8*16
+
+ VPMSUMW(v0,v16,v0)
+ VPMSUMW(v1,v17,v1)
+ VPMSUMW(v2,v18,v2)
+ VPMSUMW(v3,v19,v3)
+ VPMSUMW(v4,v20,v4)
+ VPMSUMW(v5,v21,v5)
+ VPMSUMW(v6,v22,v6)
+ VPMSUMW(v7,v23,v7)
+
+ /* Now reduce the tail (0 - 112 bytes) */
+ cmpdi r7,0
+ beq 1f
+
+ lvx v16,0,r4
+ lvx v17,0,r3
+ VPERM(v16,v16,v16,byteswap)
+ VPMSUMW(v16,v16,v17)
+ vxor v0,v0,v16
+ bdz 1f
+
+ lvx v16,off16,r4
+ lvx v17,off16,r3
+ VPERM(v16,v16,v16,byteswap)
+ VPMSUMW(v16,v16,v17)
+ vxor v0,v0,v16
+ bdz 1f
+
+ lvx v16,off32,r4
+ lvx v17,off32,r3
+ VPERM(v16,v16,v16,byteswap)
+ VPMSUMW(v16,v16,v17)
+ vxor v0,v0,v16
+ bdz 1f
+
+ lvx v16,off48,r4
+ lvx v17,off48,r3
+ VPERM(v16,v16,v16,byteswap)
+ VPMSUMW(v16,v16,v17)
+ vxor v0,v0,v16
+ bdz 1f
+
+ lvx v16,off64,r4
+ lvx v17,off64,r3
+ VPERM(v16,v16,v16,byteswap)
+ VPMSUMW(v16,v16,v17)
+ vxor v0,v0,v16
+ bdz 1f
+
+ lvx v16,off80,r4
+ lvx v17,off80,r3
+ VPERM(v16,v16,v16,byteswap)
+ VPMSUMW(v16,v16,v17)
+ vxor v0,v0,v16
+ bdz 1f
+
+ lvx v16,off96,r4
+ lvx v17,off96,r3
+ VPERM(v16,v16,v16,byteswap)
+ VPMSUMW(v16,v16,v17)
+ vxor v0,v0,v16
+
+ /* Now xor all the parallel chunks together */
+1: vxor v0,v0,v1
+ vxor v2,v2,v3
+ vxor v4,v4,v5
+ vxor v6,v6,v7
+
+ vxor v0,v0,v2
+ vxor v4,v4,v6
+
+ vxor v0,v0,v4
+
+.Lbarrett_reduction:
+ /* Barrett constants */
+ addis r3,r2,.barrett_constants@toc@ha
+ addi r3,r3,.barrett_constants@toc@l
+
+ lvx const1,0,r3
+ lvx const2,off16,r3
+
+ vsldoi v1,v0,v0,8
+ vxor v0,v0,v1 /* xor two 64 bit results together */
+
+ /* shift left one bit */
+ vspltisb v1,1
+ vsl v0,v0,v1
+
+ vand v0,v0,mask_64bit
+
+ /*
+ * The reflected version of Barrett reduction. Instead of bit
+ * reflecting our data (which is expensive to do), we bit reflect our
+ * constants and our algorithm, which means the intermediate data in
+ * our vector registers goes from 0-63 instead of 63-0. We can reflect
+ * the algorithm because we don't carry in mod 2 arithmetic.
+ */
+ vand v1,v0,mask_32bit /* bottom 32 bits of a */
+ VPMSUMD(v1,v1,const1) /* ma */
+ vand v1,v1,mask_32bit /* bottom 32bits of ma */
+ VPMSUMD(v1,v1,const2) /* qn */
+ vxor v0,v0,v1 /* a - qn, subtraction is xor in GF(2) */
+
+ /*
+ * Since we are bit reflected, the result (ie the low 32 bits) is in
+ * the high 32 bits. We just need to shift it left 4 bytes
+ * V0 [ 0 1 X 3 ]
+ * V0 [ 0 X 2 3 ]
+ */
+ vsldoi v0,v0,zeroes,4 /* shift result into top 64 bits of */
+
+ /* Get it into r3 */
+ MFVRD(R3, v0)
+
+.Lout:
+ subi r6,r1,56+10*16
+ subi r7,r1,56+2*16
+
+ lvx v20,0,r6
+ lvx v21,off16,r6
+ lvx v22,off32,r6
+ lvx v23,off48,r6
+ lvx v24,off64,r6
+ lvx v25,off80,r6
+ lvx v26,off96,r6
+ lvx v27,off112,r6
+ lvx v28,0,r7
+ lvx v29,off16,r7
+
+ ld r31,-8(r1)
+ ld r30,-16(r1)
+ ld r29,-24(r1)
+ ld r28,-32(r1)
+ ld r27,-40(r1)
+ ld r26,-48(r1)
+ ld r25,-56(r1)
+
+ blr
+
+.Lfirst_warm_up_done:
+ lvx const1,0,r3
+ addi r3,r3,16
+
+ VPMSUMD(v8,v16,const1)
+ VPMSUMD(v9,v17,const1)
+ VPMSUMD(v10,v18,const1)
+ VPMSUMD(v11,v19,const1)
+ VPMSUMD(v12,v20,const1)
+ VPMSUMD(v13,v21,const1)
+ VPMSUMD(v14,v22,const1)
+ VPMSUMD(v15,v23,const1)
+
+ b .Lsecond_cool_down
+
+.Lshort:
+ cmpdi r5,0
+ beq .Lzero
+
+ addis r3,r2,.short_constants@toc@ha
+ addi r3,r3,.short_constants@toc@l
+
+ /* Calculate where in the constant table we need to start */
+ subfic r6,r5,256
+ add r3,r3,r6
+
+ /* How many 16 byte chunks? */
+ srdi r7,r5,4
+ mtctr r7
+
+ vxor v19,v19,v19
+ vxor v20,v20,v20
+
+ lvx v0,0,r4
+ lvx v16,0,r3
+ VPERM(v0,v0,v16,byteswap)
+ vxor v0,v0,v8 /* xor in initial value */
+ VPMSUMW(v0,v0,v16)
+ bdz .Lv0
+
+ lvx v1,off16,r4
+ lvx v17,off16,r3
+ VPERM(v1,v1,v17,byteswap)
+ VPMSUMW(v1,v1,v17)
+ bdz .Lv1
+
+ lvx v2,off32,r4
+ lvx v16,off32,r3
+ VPERM(v2,v2,v16,byteswap)
+ VPMSUMW(v2,v2,v16)
+ bdz .Lv2
+
+ lvx v3,off48,r4
+ lvx v17,off48,r3
+ VPERM(v3,v3,v17,byteswap)
+ VPMSUMW(v3,v3,v17)
+ bdz .Lv3
+
+ lvx v4,off64,r4
+ lvx v16,off64,r3
+ VPERM(v4,v4,v16,byteswap)
+ VPMSUMW(v4,v4,v16)
+ bdz .Lv4
+
+ lvx v5,off80,r4
+ lvx v17,off80,r3
+ VPERM(v5,v5,v17,byteswap)
+ VPMSUMW(v5,v5,v17)
+ bdz .Lv5
+
+ lvx v6,off96,r4
+ lvx v16,off96,r3
+ VPERM(v6,v6,v16,byteswap)
+ VPMSUMW(v6,v6,v16)
+ bdz .Lv6
+
+ lvx v7,off112,r4
+ lvx v17,off112,r3
+ VPERM(v7,v7,v17,byteswap)
+ VPMSUMW(v7,v7,v17)
+ bdz .Lv7
+
+ addi r3,r3,128
+ addi r4,r4,128
+
+ lvx v8,0,r4
+ lvx v16,0,r3
+ VPERM(v8,v8,v16,byteswap)
+ VPMSUMW(v8,v8,v16)
+ bdz .Lv8
+
+ lvx v9,off16,r4
+ lvx v17,off16,r3
+ VPERM(v9,v9,v17,byteswap)
+ VPMSUMW(v9,v9,v17)
+ bdz .Lv9
+
+ lvx v10,off32,r4
+ lvx v16,off32,r3
+ VPERM(v10,v10,v16,byteswap)
+ VPMSUMW(v10,v10,v16)
+ bdz .Lv10
+
+ lvx v11,off48,r4
+ lvx v17,off48,r3
+ VPERM(v11,v11,v17,byteswap)
+ VPMSUMW(v11,v11,v17)
+ bdz .Lv11
+
+ lvx v12,off64,r4
+ lvx v16,off64,r3
+ VPERM(v12,v12,v16,byteswap)
+ VPMSUMW(v12,v12,v16)
+ bdz .Lv12
+
+ lvx v13,off80,r4
+ lvx v17,off80,r3
+ VPERM(v13,v13,v17,byteswap)
+ VPMSUMW(v13,v13,v17)
+ bdz .Lv13
+
+ lvx v14,off96,r4
+ lvx v16,off96,r3
+ VPERM(v14,v14,v16,byteswap)
+ VPMSUMW(v14,v14,v16)
+ bdz .Lv14
+
+ lvx v15,off112,r4
+ lvx v17,off112,r3
+ VPERM(v15,v15,v17,byteswap)
+ VPMSUMW(v15,v15,v17)
+
+.Lv15: vxor v19,v19,v15
+.Lv14: vxor v20,v20,v14
+.Lv13: vxor v19,v19,v13
+.Lv12: vxor v20,v20,v12
+.Lv11: vxor v19,v19,v11
+.Lv10: vxor v20,v20,v10
+.Lv9: vxor v19,v19,v9
+.Lv8: vxor v20,v20,v8
+.Lv7: vxor v19,v19,v7
+.Lv6: vxor v20,v20,v6
+.Lv5: vxor v19,v19,v5
+.Lv4: vxor v20,v20,v4
+.Lv3: vxor v19,v19,v3
+.Lv2: vxor v20,v20,v2
+.Lv1: vxor v19,v19,v1
+.Lv0: vxor v20,v20,v0
+
+ vxor v0,v19,v20
+
+ b .Lbarrett_reduction
+
+.Lzero:
+ mr r3,r10
+ b .Lout
+
+FUNC_END(__crc32_vpmsum)
diff --git a/arch/powerpc/crypto/crc32c-vpmsum_glue.c b/arch/powerpc/crypto/crc32c-vpmsum_glue.c
new file mode 100644
index 0000000..bfe3d37
--- /dev/null
+++ b/arch/powerpc/crypto/crc32c-vpmsum_glue.c
@@ -0,0 +1,167 @@
+#include <linux/crc32.h>
+#include <crypto/internal/hash.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <asm/switch_to.h>
+
+#define CHKSUM_BLOCK_SIZE 1
+#define CHKSUM_DIGEST_SIZE 4
+
+#define VMX_ALIGN 16
+#define VMX_ALIGN_MASK (VMX_ALIGN-1)
+
+#define VECTOR_BREAKPOINT 512
+
+u32 __crc32c_vpmsum(u32 crc, unsigned char const *p, size_t len);
+
+static u32 crc32c_vpmsum(u32 crc, unsigned char const *p, size_t len)
+{
+ unsigned int prealign;
+ unsigned int tail;
+
+ if (len < (VECTOR_BREAKPOINT + VMX_ALIGN) || in_interrupt())
+ return __crc32c_le(crc, p, len);
+
+ if ((unsigned long)p & VMX_ALIGN_MASK) {
+ prealign = VMX_ALIGN - ((unsigned long)p & VMX_ALIGN_MASK);
+ crc = __crc32c_le(crc, p, prealign);
+ len -= prealign;
+ p += prealign;
+ }
+
+ if (len & ~VMX_ALIGN_MASK) {
+ pagefault_disable();
+ enable_kernel_altivec();
+ crc = __crc32c_vpmsum(crc, p, len & ~VMX_ALIGN_MASK);
+ pagefault_enable();
+ }
+
+ tail = len & VMX_ALIGN_MASK;
+ if (tail) {
+ p += len & ~VMX_ALIGN_MASK;
+ crc = __crc32c_le(crc, p, tail);
+ }
+
+ return crc;
+}
+
+static int crc32c_vpmsum_cra_init(struct crypto_tfm *tfm)
+{
+ u32 *key = crypto_tfm_ctx(tfm);
+
+ *key = 0;
+
+ return 0;
+}
+
+/*
+ * Setting the seed allows arbitrary accumulators and flexible XOR policy
+ * If your algorithm starts with ~0, then XOR with ~0 before you set
+ * the seed.
+ */
+static int crc32c_vpmsum_setkey(struct crypto_shash *hash, const u8 *key,
+ unsigned int keylen)
+{
+ u32 *mctx = crypto_shash_ctx(hash);
+
+ if (keylen != sizeof(u32)) {
+ crypto_shash_set_flags(hash, CRYPTO_TFM_RES_BAD_KEY_LEN);
+ return -EINVAL;
+ }
+ *mctx = le32_to_cpup((__le32 *)key);
+ return 0;
+}
+
+static int crc32c_vpmsum_init(struct shash_desc *desc)
+{
+ u32 *mctx = crypto_shash_ctx(desc->tfm);
+ u32 *crcp = shash_desc_ctx(desc);
+
+ *crcp = *mctx;
+
+ return 0;
+}
+
+static int crc32c_vpmsum_update(struct shash_desc *desc, const u8 *data,
+ unsigned int len)
+{
+ u32 *crcp = shash_desc_ctx(desc);
+
+ *crcp = crc32c_vpmsum(*crcp, data, len);
+
+ return 0;
+}
+
+static int __crc32c_vpmsum_finup(u32 *crcp, const u8 *data, unsigned int len,
+ u8 *out)
+{
+ *(__le32 *)out = ~cpu_to_le32(crc32c_vpmsum(*crcp, data, len));
+
+ return 0;
+}
+
+static int crc32c_vpmsum_finup(struct shash_desc *desc, const u8 *data,
+ unsigned int len, u8 *out)
+{
+ return __crc32c_vpmsum_finup(shash_desc_ctx(desc), data, len, out);
+}
+
+static int crc32c_vpmsum_final(struct shash_desc *desc, u8 *out)
+{
+ u32 *crcp = shash_desc_ctx(desc);
+
+ *(__le32 *)out = ~cpu_to_le32p(crcp);
+
+ return 0;
+}
+
+static int crc32c_vpmsum_digest(struct shash_desc *desc, const u8 *data,
+ unsigned int len, u8 *out)
+{
+ return __crc32c_vpmsum_finup(crypto_shash_ctx(desc->tfm), data, len,
+ out);
+}
+
+static struct shash_alg alg = {
+ .setkey = crc32c_vpmsum_setkey,
+ .init = crc32c_vpmsum_init,
+ .update = crc32c_vpmsum_update,
+ .final = crc32c_vpmsum_final,
+ .finup = crc32c_vpmsum_finup,
+ .digest = crc32c_vpmsum_digest,
+ .descsize = sizeof(u32),
+ .digestsize = CHKSUM_DIGEST_SIZE,
+ .base = {
+ .cra_name = "crc32c",
+ .cra_driver_name = "crc32c-vpmsum",
+ .cra_priority = 200,
+ .cra_blocksize = CHKSUM_BLOCK_SIZE,
+ .cra_ctxsize = sizeof(u32),
+ .cra_module = THIS_MODULE,
+ .cra_init = crc32c_vpmsum_cra_init,
+ }
+};
+
+static int __init crc32c_vpmsum_mod_init(void)
+{
+ if (!cpu_has_feature(CPU_FTR_ARCH_207S))
+ return -ENODEV;
+
+ return crypto_register_shash(&alg);
+}
+
+static void __exit crc32c_vpmsum_mod_fini(void)
+{
+ crypto_unregister_shash(&alg);
+}
+
+module_init(crc32c_vpmsum_mod_init);
+module_exit(crc32c_vpmsum_mod_fini);
+
+MODULE_AUTHOR("Anton Blanchard <anton@samba.org>");
+MODULE_DESCRIPTION("CRC32C using vector polynomial multiply-sum instructions");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_CRYPTO("crc32c");
+MODULE_ALIAS_CRYPTO("crc32c-vpmsum");
diff --git a/arch/powerpc/include/asm/ppc-opcode.h b/arch/powerpc/include/asm/ppc-opcode.h
index 1d035c1..49cd876 100644
--- a/arch/powerpc/include/asm/ppc-opcode.h
+++ b/arch/powerpc/include/asm/ppc-opcode.h
@@ -174,6 +174,8 @@
#define PPC_INST_MFSPR_DSCR_USER_MASK 0xfc1fffff
#define PPC_INST_MTSPR_DSCR_USER 0x7c0303a6
#define PPC_INST_MTSPR_DSCR_USER_MASK 0xfc1fffff
+#define PPC_INST_MFVSRD 0x7c000066
+#define PPC_INST_MTVSRD 0x7c000166
#define PPC_INST_SLBFEE 0x7c0007a7
#define PPC_INST_STRING 0x7c00042a
@@ -188,6 +190,8 @@
#define PPC_INST_WAIT 0x7c00007c
#define PPC_INST_TLBIVAX 0x7c000624
#define PPC_INST_TLBSRX_DOT 0x7c0006a5
+#define PPC_INST_VPMSUMW 0x10000488
+#define PPC_INST_VPMSUMD 0x100004c8
#define PPC_INST_XXLOR 0xf0000510
#define PPC_INST_XXSWAPD 0xf0000250
#define PPC_INST_XVCPSGNDP 0xf0000780
@@ -359,6 +363,14 @@
VSX_XX1((s), a, b))
#define LXVD2X(s, a, b) stringify_in_c(.long PPC_INST_LXVD2X | \
VSX_XX1((s), a, b))
+#define MFVRD(a, t) stringify_in_c(.long PPC_INST_MFVSRD | \
+ VSX_XX1((t)+32, a, R0))
+#define MTVRD(t, a) stringify_in_c(.long PPC_INST_MTVSRD | \
+ VSX_XX1((t)+32, a, R0))
+#define VPMSUMW(t, a, b) stringify_in_c(.long PPC_INST_VPMSUMW | \
+ VSX_XX3((t), a, b))
+#define VPMSUMD(t, a, b) stringify_in_c(.long PPC_INST_VPMSUMD | \
+ VSX_XX3((t), a, b))
#define XXLOR(t, a, b) stringify_in_c(.long PPC_INST_XXLOR | \
VSX_XX3((t), a, b))
#define XXSWAPD(t, a) stringify_in_c(.long PPC_INST_XXSWAPD | \
diff --git a/crypto/Kconfig b/crypto/Kconfig
index 1d33beb..caa0314 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -414,6 +414,17 @@ config CRYPTO_CRC32C_INTEL
gain performance compared with software implementation.
Module will be crc32c-intel.
+config CRYPT_CRC32C_VPMSUM
+ tristate "CRC32c CRC algorithm (powerpc64)"
+ depends on PPC64
+ select CRYPTO_HASH
+ select CRC32
+ help
+ CRC32c algorithm implemented using vector polynomial multiply-sum
+ (vpmsum) instructions, introduced in POWER8. Enable on POWER8
+ and newer processors for improved performance.
+
+
config CRYPTO_CRC32C_SPARC64
tristate "CRC32c CRC algorithm (SPARC64)"
depends on SPARC64
--
2.7.4
^ permalink raw reply related [flat|nested] 5+ messages in thread