diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2015-11-04 09:11:12 -0800 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2015-11-04 09:11:12 -0800 |
| commit | ccc9d4a6d640cbde05d519edeb727881646cf71b (patch) | |
| tree | 736c90b100703501d5e3fa3eccc57a48f70bef14 | |
| parent | 66ef3493d4bb387f5a83915e33dc893102fd1b43 (diff) | |
| parent | 271817a3e92c0455bda5856d87eca244ad67d3a2 (diff) | |
Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Pull crypto update from Herbert Xu:
"API:
- Add support for cipher output IVs in testmgr
- Add missing crypto_ahash_blocksize helper
- Mark authenc and des ciphers as not allowed under FIPS.
Algorithms:
- Add CRC support to 842 compression
- Add keywrap algorithm
- A number of changes to the akcipher interface:
+ Separate functions for setting public/private keys.
+ Use SG lists.
Drivers:
- Add Intel SHA Extension optimised SHA1 and SHA256
- Use dma_map_sg instead of custom functions in crypto drivers
- Add support for STM32 RNG
- Add support for ST RNG
- Add Device Tree support to exynos RNG driver
- Add support for mxs-dcp crypto device on MX6SL
- Add xts(aes) support to caam
- Add ctr(aes) and xts(aes) support to qat
- A large set of fixes from Russell King for the marvell/cesa driver"
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (115 commits)
crypto: asymmetric_keys - Fix unaligned access in x509_get_sig_params()
crypto: akcipher - Don't #include crypto/public_key.h as the contents aren't used
hwrng: exynos - Add Device Tree support
hwrng: exynos - Fix missing configuration after suspend to RAM
hwrng: exynos - Add timeout for waiting on init done
dt-bindings: rng: Describe Exynos4 PRNG bindings
crypto: marvell/cesa - use __le32 for hardware descriptors
crypto: marvell/cesa - fix missing cpu_to_le32() in mv_cesa_dma_add_op()
crypto: marvell/cesa - use memcpy_fromio()/memcpy_toio()
crypto: marvell/cesa - use gfp_t for gfp flags
crypto: marvell/cesa - use dma_addr_t for cur_dma
crypto: marvell/cesa - use readl_relaxed()/writel_relaxed()
crypto: caam - fix indentation of close braces
crypto: caam - only export the state we really need to export
crypto: caam - fix non-block aligned hash calculation
crypto: caam - avoid needlessly saving and restoring caam_hash_ctx
crypto: caam - print errno code when hash registration fails
crypto: marvell/cesa - fix memory leak
crypto: marvell/cesa - fix first-fragment handling in mv_cesa_ahash_dma_last_req()
crypto: marvell/cesa - rearrange handling for sw padded hashes
...
98 files changed, 4070 insertions, 1538 deletions
diff --git a/Documentation/devicetree/bindings/rng/samsung,exynos-rng4.txt b/Documentation/devicetree/bindings/rng/samsung,exynos-rng4.txt new file mode 100644 index 000000000000..4ca8dd4d7e66 --- /dev/null +++ b/Documentation/devicetree/bindings/rng/samsung,exynos-rng4.txt @@ -0,0 +1,17 @@ +Exynos Pseudo Random Number Generator + +Required properties: + +- compatible : Should be "samsung,exynos4-rng". +- reg : Specifies base physical address and size of the registers map. +- clocks : Phandle to clock-controller plus clock-specifier pair. +- clock-names : "secss" as a clock name. + +Example: + + rng@10830400 { + compatible = "samsung,exynos4-rng"; + reg = <0x10830400 0x200>; + clocks = <&clock CLK_SSS>; + clock-names = "secss"; + }; diff --git a/Documentation/devicetree/bindings/rng/st,rng.txt b/Documentation/devicetree/bindings/rng/st,rng.txt new file mode 100644 index 000000000000..35734bc282e9 --- /dev/null +++ b/Documentation/devicetree/bindings/rng/st,rng.txt @@ -0,0 +1,15 @@ +STMicroelectronics HW Random Number Generator +---------------------------------------------- + +Required parameters: +compatible : Should be "st,rng" +reg : Base address and size of IP's register map. +clocks : Phandle to device's clock (See: ../clocks/clock-bindings.txt) + +Example: + +rng@fee80000 { + compatible = "st,rng"; + reg = <0xfee80000 0x1000>; + clocks = <&clk_sysin>; +} diff --git a/Documentation/devicetree/bindings/rng/st,stm32-rng.txt b/Documentation/devicetree/bindings/rng/st,stm32-rng.txt new file mode 100644 index 000000000000..47f04176f93b --- /dev/null +++ b/Documentation/devicetree/bindings/rng/st,stm32-rng.txt @@ -0,0 +1,21 @@ +STMicroelectronics STM32 HW RNG +=============================== + +The STM32 hardware random number generator is a simple fixed purpose IP and +is fully separated from other crypto functions. + +Required properties: + +- compatible : Should be "st,stm32-rng" +- reg : Should be register base and length as documented in the datasheet +- interrupts : The designated IRQ line for the RNG +- clocks : The clock needed to enable the RNG + +Example: + + rng: rng@50060800 { + compatible = "st,stm32-rng"; + reg = <0x50060800 0x400>; + interrupts = <80>; + clocks = <&rcc 0 38>; + }; diff --git a/Documentation/hw_random.txt b/Documentation/hw_random.txt index 026e237bbc87..fce1634907d0 100644 --- a/Documentation/hw_random.txt +++ b/Documentation/hw_random.txt @@ -3,7 +3,7 @@ Introduction: The hw_random framework is software that makes use of a special hardware feature on your CPU or motherboard, a Random Number Generator (RNG). The software has two parts: - a core providing the /dev/hw_random character device and its + a core providing the /dev/hwrng character device and its sysfs support, plus a hardware-specific driver that plugs into that core. @@ -14,7 +14,7 @@ Introduction: http://sourceforge.net/projects/gkernel/ - Those tools use /dev/hw_random to fill the kernel entropy pool, + Those tools use /dev/hwrng to fill the kernel entropy pool, which is used internally and exported by the /dev/urandom and /dev/random special files. @@ -32,13 +32,13 @@ Theory of operation: The rng-tools package uses such tests in "rngd", and lets you run them by hand with a "rngtest" utility. - /dev/hw_random is char device major 10, minor 183. + /dev/hwrng is char device major 10, minor 183. CLASS DEVICE. There is a /sys/class/misc/hw_random node with two unique attributes, "rng_available" and "rng_current". The "rng_available" attribute lists the hardware-specific drivers available, while "rng_current" lists the one which is currently - connected to /dev/hw_random. If your system has more than one + connected to /dev/hwrng. If your system has more than one RNG available, you may change the one used by writing a name from the list in "rng_available" into "rng_current". diff --git a/MAINTAINERS b/MAINTAINERS index dcc8ed6fccde..3036f5a02107 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1529,6 +1529,7 @@ W: http://www.stlinux.com S: Maintained F: arch/arm/mach-sti/ F: arch/arm/boot/dts/sti* +F: drivers/char/hw_random/st-rng.c F: drivers/clocksource/arm_global_timer.c F: drivers/clocksource/clksrc_st_lpc.c F: drivers/i2c/busses/i2c-st.c @@ -6587,6 +6588,13 @@ M: Guenter Roeck <linux@roeck-us.net> S: Maintained F: drivers/net/dsa/mv88e6352.c +MARVELL CRYPTO DRIVER +M: Boris Brezillon <boris.brezillon@free-electrons.com> +M: Arnaud Ebalard <arno@natisbad.org> +F: drivers/crypto/marvell/ +S: Maintained +L: linux-crypto@vger.kernel.org + MARVELL GIGABIT ETHERNET DRIVERS (skge/sky2) M: Mirko Lindner <mlindner@marvell.com> M: Stephen Hemminger <stephen@networkplumber.org> diff --git a/arch/arm/boot/dts/stih407-family.dtsi b/arch/arm/boot/dts/stih407-family.dtsi index ae0527754000..0c24fcb03577 100644 --- a/arch/arm/boot/dts/stih407-family.dtsi +++ b/arch/arm/boot/dts/stih407-family.dtsi @@ -610,5 +610,19 @@ clocks = <&clk_sysin>; st,pwm-num-chan = <4>; }; + + rng10: rng@08a89000 { + compatible = "st,rng"; + reg = <0x08a89000 0x1000>; + clocks = <&clk_sysin>; + status = "okay"; + }; + + rng11: rng@08a8a000 { + compatible = "st,rng"; + reg = <0x08a8a000 0x1000>; + clocks = <&clk_sysin>; + status = "okay"; + }; }; }; diff --git a/arch/arm/boot/dts/stm32f429.dtsi b/arch/arm/boot/dts/stm32f429.dtsi index d78a4815da8f..5e1e234e8c0a 100644 --- a/arch/arm/boot/dts/stm32f429.dtsi +++ b/arch/arm/boot/dts/stm32f429.dtsi @@ -174,6 +174,13 @@ reg = <0x40023800 0x400>; clocks = <&clk_hse>; }; + + rng: rng@50060800 { + compatible = "st,stm32-rng"; + reg = <0x50060800 0x400>; + interrupts = <80>; + clocks = <&rcc 0 38>; + }; }; }; diff --git a/arch/s390/crypto/sha.h b/arch/s390/crypto/sha.h index f4e9dc71675f..10f200790079 100644 --- a/arch/s390/crypto/sha.h +++ b/arch/s390/crypto/sha.h @@ -19,7 +19,7 @@ #include <crypto/sha.h> /* must be big enough for the largest SHA variant */ -#define SHA_MAX_STATE_SIZE 16 +#define SHA_MAX_STATE_SIZE (SHA512_DIGEST_SIZE / 4) #define SHA_MAX_BLOCK_SIZE SHA512_BLOCK_SIZE struct s390_sha_ctx { diff --git a/arch/x86/Makefile b/arch/x86/Makefile index 2dfaa72260b4..4086abca0b32 100644 --- a/arch/x86/Makefile +++ b/arch/x86/Makefile @@ -171,9 +171,11 @@ asinstr += $(call as-instr,pshufb %xmm0$(comma)%xmm0,-DCONFIG_AS_SSSE3=1) asinstr += $(call as-instr,crc32l %eax$(comma)%eax,-DCONFIG_AS_CRC32=1) avx_instr := $(call as-instr,vxorps %ymm0$(comma)%ymm1$(comma)%ymm2,-DCONFIG_AS_AVX=1) avx2_instr :=$(call as-instr,vpbroadcastb %xmm0$(comma)%ymm1,-DCONFIG_AS_AVX2=1) +sha1_ni_instr :=$(call as-instr,sha1msg1 %xmm0$(comma)%xmm1,-DCONFIG_AS_SHA1_NI=1) +sha256_ni_instr :=$(call as-instr,sha256msg1 %xmm0$(comma)%xmm1,-DCONFIG_AS_SHA256_NI=1) -KBUILD_AFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) $(asinstr) $(avx_instr) $(avx2_instr) -KBUILD_CFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) $(asinstr) $(avx_instr) $(avx2_instr) +KBUILD_AFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) $(asinstr) $(avx_instr) $(avx2_instr) $(sha1_ni_instr) $(sha256_ni_instr) +KBUILD_CFLAGS += $(cfi) $(cfi-sigframe) $(cfi-sections) $(asinstr) $(avx_instr) $(avx2_instr) $(sha1_ni_instr) $(sha256_ni_instr) LDFLAGS := -m elf_$(UTS_MACHINE) diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile index 9a2838cf0591..b9b912a44d61 100644 --- a/arch/x86/crypto/Makefile +++ b/arch/x86/crypto/Makefile @@ -5,6 +5,8 @@ avx_supported := $(call as-instr,vpxor %xmm0$(comma)%xmm0$(comma)%xmm0,yes,no) avx2_supported := $(call as-instr,vpgatherdd %ymm0$(comma)(%eax$(comma)%ymm1\ $(comma)4)$(comma)%ymm2,yes,no) +sha1_ni_supported :=$(call as-instr,sha1msg1 %xmm0$(comma)%xmm1,yes,no) +sha256_ni_supported :=$(call as-instr,sha256msg1 %xmm0$(comma)%xmm1,yes,no) obj-$(CONFIG_CRYPTO_GLUE_HELPER_X86) += glue_helper.o @@ -91,9 +93,15 @@ ifeq ($(avx2_supported),yes) sha1-ssse3-y += sha1_avx2_x86_64_asm.o poly1305-x86_64-y += poly1305-avx2-x86_64.o endif +ifeq ($(sha1_ni_supported),yes) +sha1-ssse3-y += sha1_ni_asm.o +endif crc32c-intel-y := crc32c-intel_glue.o crc32c-intel-$(CONFIG_64BIT) += crc32c-pcl-intel-asm_64.o crc32-pclmul-y := crc32-pclmul_asm.o crc32-pclmul_glue.o sha256-ssse3-y := sha256-ssse3-asm.o sha256-avx-asm.o sha256-avx2-asm.o sha256_ssse3_glue.o +ifeq ($(sha256_ni_supported),yes) +sha256-ssse3-y += sha256_ni_asm.o +endif sha512-ssse3-y := sha512-ssse3-asm.o sha512-avx-asm.o sha512-avx2-asm.o sha512_ssse3_glue.o crct10dif-pclmul-y := crct10dif-pcl-asm_64.o crct10dif-pclmul_glue.o diff --git a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S index 225be06edc80..4fe27e074194 100644 --- a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S +++ b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S @@ -330,7 +330,7 @@ ENDPROC(crc_pcl) ## PCLMULQDQ tables ## Table is 128 entries x 2 words (8 bytes) each ################################################################ -.section .rotata, "a", %progbits +.section .rodata, "a", %progbits .align 8 K_table: .long 0x493c7d27, 0x00000001 diff --git a/arch/x86/crypto/sha1_ni_asm.S b/arch/x86/crypto/sha1_ni_asm.S new file mode 100644 index 000000000000..874a651b9e7d --- /dev/null +++ b/arch/x86/crypto/sha1_ni_asm.S @@ -0,0 +1,302 @@ +/* + * Intel SHA Extensions optimized implementation of a SHA-1 update function + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * Contact Information: + * Sean Gulley <sean.m.gulley@intel.com> + * Tim Chen <tim.c.chen@linux.intel.com> + * + * BSD LICENSE + * + * Copyright(c) 2015 Intel Corporation. + * + * 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 the above copyright + * notice, 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 Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS 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. + * + */ + +#include <linux/linkage.h> + +#define DIGEST_PTR %rdi /* 1st arg */ +#define DATA_PTR %rsi /* 2nd arg */ +#define NUM_BLKS %rdx /* 3rd arg */ + +#define RSPSAVE %rax + +/* gcc conversion */ +#define FRAME_SIZE 32 /* space for 2x16 bytes */ + +#define ABCD %xmm0 +#define E0 %xmm1 /* Need two E's b/c they ping pong */ +#define E1 %xmm2 +#define MSG0 %xmm3 +#define MSG1 %xmm4 +#define MSG2 %xmm5 +#define MSG3 %xmm6 +#define SHUF_MASK %xmm7 + + +/* + * Intel SHA Extensions optimized implementation of a SHA-1 update function + * + * The function takes a pointer to the current hash values, a pointer to the + * input data, and a number of 64 byte blocks to process. Once all blocks have + * been processed, the digest pointer is updated with the resulting hash value. + * The function only processes complete blocks, there is no functionality to + * store partial blocks. All message padding and hash value initialization must + * be done outside the update function. + * + * The indented lines in the loop are instructions related to rounds processing. + * The non-indented lines are instructions related to the message schedule. + * + * void sha1_ni_transform(uint32_t *digest, const void *data, + uint32_t numBlocks) + * digest : pointer to digest + * data: pointer to input data + * numBlocks: Number of blocks to process + */ +.text +.align 32 +ENTRY(sha1_ni_transform) + mov %rsp, RSPSAVE + sub $FRAME_SIZE, %rsp + and $~0xF, %rsp + + shl $6, NUM_BLKS /* convert to bytes */ + jz .Ldone_hash + add DATA_PTR, NUM_BLKS /* pointer to end of data */ + + /* load initial hash values */ + pinsrd $3, 1*16(DIGEST_PTR), E0 + movdqu 0*16(DIGEST_PTR), ABCD + pand UPPER_WORD_MASK(%rip), E0 + pshufd $0x1B, ABCD, ABCD + + movdqa PSHUFFLE_BYTE_FLIP_MASK(%rip), SHUF_MASK + +.Lloop0: + /* Save hash values for addition after rounds */ + movdqa E0, (0*16)(%rsp) + movdqa ABCD, (1*16)(%rsp) + + /* Rounds 0-3 */ + movdqu 0*16(DATA_PTR), MSG0 + pshufb SHUF_MASK, MSG0 + paddd MSG0, E0 + movdqa ABCD, E1 + sha1rnds4 $0, E0, ABCD + + /* Rounds 4-7 */ + movdqu 1*16(DATA_PTR), MSG1 + pshufb SHUF_MASK, MSG1 + sha1nexte MSG1, E1 + movdqa ABCD, E0 + sha1rnds4 $0, E1, ABCD + sha1msg1 MSG1, MSG0 + + /* Rounds 8-11 */ + movdqu 2*16(DATA_PTR), MSG2 + pshufb SHUF_MASK, MSG2 + sha1nexte MSG2, E0 + movdqa ABCD, E1 + sha1rnds4 $0, E0, ABCD + sha1msg1 MSG2, MSG1 + pxor MSG2, MSG0 + + /* Rounds 12-15 */ + movdqu 3*16(DATA_PTR), MSG3 + pshufb SHUF_MASK, MSG3 + sha1nexte MSG3, E1 + movdqa ABCD, E0 + sha1msg2 MSG3, MSG0 + sha1rnds4 $0, E1, ABCD + sha1msg1 MSG3, MSG2 + pxor MSG3, MSG1 + + /* Rounds 16-19 */ + sha1nexte MSG0, E0 + movdqa ABCD, E1 + sha1msg2 MSG0, MSG1 + sha1rnds4 $0, E0, ABCD + sha1msg1 MSG0, MSG3 + pxor MSG0, MSG2 + + /* Rounds 20-23 */ + sha1nexte MSG1, E1 + movdqa ABCD, E0 + sha1msg2 MSG1, MSG2 + sha1rnds4 $1, E1, ABCD + sha1msg1 MSG1, MSG0 + pxor MSG1, MSG3 + + /* Rounds 24-27 */ + sha1nexte MSG2, E0 + movdqa ABCD, E1 + sha1msg2 MSG2, MSG3 + sha1rnds4 $1, E0, ABCD + sha1msg1 MSG2, MSG1 + pxor MSG2, MSG0 + + /* Rounds 28-31 */ + sha1nexte MSG3, E1 + movdqa ABCD, E0 + sha1msg2 MSG3, MSG0 + sha1rnds4 $1, E1, ABCD + sha1msg1 MSG3, MSG2 + pxor MSG3, MSG1 + + /* Rounds 32-35 */ + sha1nexte MSG0, E0 + movdqa ABCD, E1 + sha1msg2 MSG0, MSG1 + sha1rnds4 $1, E0, ABCD + sha1msg1 MSG0, MSG3 + pxor MSG0, MSG2 + + /* Rounds 36-39 */ + sha1nexte MSG1, E1 + movdqa ABCD, E0 + sha1msg2 MSG1, MSG2 + sha1rnds4 $1, E1, ABCD + sha1msg1 MSG1, MSG0 + pxor MSG1, MSG3 + + /* Rounds 40-43 */ + sha1nexte MSG2, E0 + movdqa ABCD, E1 + sha1msg2 MSG2, MSG3 + sha1rnds4 $2, E0, ABCD + sha1msg1 MSG2, MSG1 + pxor MSG2, MSG0 + + /* Rounds 44-47 */ + sha1nexte MSG3, E1 + movdqa ABCD, E0 + sha1msg2 MSG3, MSG0 + sha1rnds4 $2, E1, ABCD + sha1msg1 MSG3, MSG2 + pxor MSG3, MSG1 + + /* Rounds 48-51 */ + sha1nexte MSG0, E0 + movdqa ABCD, E1 + sha1msg2 MSG0, MSG1 + sha1rnds4 $2, E0, ABCD + sha1msg1 MSG0, MSG3 + pxor MSG0, MSG2 + + /* Rounds 52-55 */ + sha1nexte MSG1, E1 + movdqa ABCD, E0 + sha1msg2 MSG1, MSG2 + sha1rnds4 $2, E1, ABCD + sha1msg1 MSG1, MSG0 + pxor MSG1, MSG3 + + /* Rounds 56-59 */ + sha1nexte MSG2, E0 + movdqa ABCD, E1 + sha1msg2 MSG2, MSG3 + sha1rnds4 $2, E0, ABCD + sha1msg1 MSG2, MSG1 + pxor MSG2, MSG0 + + /* Rounds 60-63 */ + sha1nexte MSG3, E1 + movdqa ABCD, E0 + sha1msg2 MSG3, MSG0 + sha1rnds4 $3, E1, ABCD + sha1msg1 MSG3, MSG2 + pxor MSG3, MSG1 + + /* Rounds 64-67 */ + sha1nexte MSG0, E0 + movdqa ABCD, E1 + sha1msg2 MSG0, MSG1 + sha1rnds4 $3, E0, ABCD + sha1msg1 MSG0, MSG3 + pxor MSG0, MSG2 + + /* Rounds 68-71 */ + sha1nexte MSG1, E1 + movdqa ABCD, E0 + sha1msg2 MSG1, MSG2 + sha1rnds4 $3, E1, ABCD + pxor MSG1, MSG3 + + /* Rounds 72-75 */ + sha1nexte MSG2, E0 + movdqa ABCD, E1 + sha1msg2 MSG2, MSG3 + sha1rnds4 $3, E0, ABCD + + /* Rounds 76-79 */ + sha1nexte MSG3, E1 + movdqa ABCD, E0 + sha1rnds4 $3, E1, ABCD + + /* Add current hash values with previously saved */ + sha1nexte (0*16)(%rsp), E0 + paddd (1*16)(%rsp), ABCD + + /* Increment data pointer and loop if more to process */ + add $64, DATA_PTR + cmp NUM_BLKS, DATA_PTR + jne .Lloop0 + + /* Write hash values back in the correct order */ + pshufd $0x1B, ABCD, ABCD + movdqu ABCD, 0*16(DIGEST_PTR) + pextrd $3, E0, 1*16(DIGEST_PTR) + +.Ldone_hash: + mov RSPSAVE, %rsp + + ret +ENDPROC(sha1_ni_transform) + +.data + +.align 64 +PSHUFFLE_BYTE_FLIP_MASK: + .octa 0x000102030405060708090a0b0c0d0e0f +UPPER_WORD_MASK: + .octa 0xFFFFFFFF000000000000000000000000 diff --git a/arch/x86/crypto/sha1_ssse3_glue.c b/arch/x86/crypto/sha1_ssse3_glue.c index 00212c32d4db..dd14616b7739 100644 --- a/arch/x86/crypto/sha1_ssse3_glue.c +++ b/arch/x86/crypto/sha1_ssse3_glue.c @@ -31,24 +31,11 @@ #include <crypto/sha1_base.h> #include <asm/fpu/api.h> +typedef void (sha1_transform_fn)(u32 *digest, const char *data, + unsigned int rounds); -asmlinkage void sha1_transform_ssse3(u32 *digest, const char *data, - unsigned int rounds); -#ifdef CONFIG_AS_AVX -asmlinkage void sha1_transform_avx(u32 *digest, const char *data, - unsigned int rounds); -#endif -#ifdef CONFIG_AS_AVX2 -#define SHA1_AVX2_BLOCK_OPTSIZE 4 /* optimal 4*64 bytes of SHA1 blocks */ - -asmlinkage void sha1_transform_avx2(u32 *digest, const char *data, - unsigned int rounds); -#endif - -static void (*sha1_transform_asm)(u32 *, const char *, unsigned int); - -static int sha1_ssse3_update(struct shash_desc *desc, const u8 *data, - unsigned int len) +static int sha1_update(struct shash_desc *desc, const u8 *data, + unsigned int len, sha1_transform_fn *sha1_xform) { struct sha1_state *sctx = shash_desc_ctx(desc); @@ -61,14 +48,14 @@ static int sha1_ssse3_update(struct shash_desc *desc, const u8 *data, kernel_fpu_begin(); sha1_base_do_update(desc, data, len, - (sha1_block_fn *)sha1_transform_asm); + (sha1_block_fn *)sha1_xform); kernel_fpu_end(); return 0; } -static int sha1_ssse3_finup(struct shash_desc *desc, const u8 *data, - unsigned int len, u8 *out) +static int sha1_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out, sha1_transform_fn *sha1_xform) { if (!irq_fpu_usable()) return crypto_sha1_finup(desc, data, len, out); @@ -76,32 +63,37 @@ static int sha1_ssse3_finup(struct shash_desc *desc, const u8 *data, kernel_fpu_begin(); if (len) sha1_base_do_update(desc, data, len, - (sha1_block_fn *)sha1_transform_asm); - sha1_base_do_finalize(desc, (sha1_block_fn *)sha1_transform_asm); + (sha1_block_fn *)sha1_xform); + sha1_base_do_finalize(desc, (sha1_block_fn *)sha1_xform); kernel_fpu_end(); return sha1_base_finish(desc, out); } -/* Add padding and return the message digest. */ -static int sha1_ssse3_final(struct shash_desc *desc, u8 *out) +asmlinkage void sha1_transform_ssse3(u32 *digest, const char *data, + unsigned int rounds); + +static int sha1_ssse3_update(struct shash_desc *desc, const u8 *data, + unsigned int len) { - return sha1_ssse3_finup(desc, NULL, 0, out); + return sha1_update(desc, data, len, + (sha1_transform_fn *) sha1_transform_ssse3); } -#ifdef CONFIG_AS_AVX2 -static void sha1_apply_transform_avx2(u32 *digest, const char *data, - unsigned int rounds) +static int sha1_ssse3_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) { - /* Select the optimal transform based on data block size */ - if (rounds >= SHA1_AVX2_BLOCK_OPTSIZE) - sha1_transform_avx2(digest, data, rounds); - else - sha1_transform_avx(digest, data, rounds); + return sha1_finup(desc, data, len, out, + (sha1_transform_fn *) sha1_transform_ssse3); +} + +/* Add padding and return the message digest. */ +static int sha1_ssse3_final(struct shash_desc *desc, u8 *out) +{ + return sha1_ssse3_finup(desc, NULL, 0, out); } -#endif -static struct shash_alg alg = { +static struct shash_alg sha1_ssse3_alg = { .digestsize = SHA1_DIGEST_SIZE, .init = sha1_base_init, .update = sha1_ssse3_update, @@ -110,7 +102,7 @@ static struct shash_alg alg = { .descsize = sizeof(struct sha1_state), .base = { .cra_name = "sha1", - .cra_driver_name= "sha1-ssse3", + .cra_driver_name = "sha1-ssse3", .cra_priority = 150, .cra_flags = CRYPTO_ALG_TYPE_SHASH, .cra_blocksize = SHA1_BLOCK_SIZE, @@ -118,8 +110,60 @@ static struct shash_alg alg = { } }; +static int register_sha1_ssse3(void) +{ + if (boot_cpu_has(X86_FEATURE_SSSE3)) + return crypto_register_shash(&sha1_ssse3_alg); + return 0; +} + +static void unregister_sha1_ssse3(void) +{ + if (boot_cpu_has(X86_FEATURE_SSSE3)) + crypto_unregister_shash(&sha1_ssse3_alg); +} + #ifdef CONFIG_AS_AVX -static bool __init avx_usable(void) +asmlinkage void sha1_transform_avx(u32 *digest, const char *data, + unsigned int rounds); + +static int sha1_avx_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha1_update(desc, data, len, + (sha1_transform_fn *) sha1_transform_avx); +} + +static int sha1_avx_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha1_finup(desc, data, len, out, + (sha1_transform_fn *) sha1_transform_avx); +} + +static int sha1_avx_final(struct shash_desc *desc, u8 *out) +{ + return sha1_avx_finup(desc, NULL, 0, out); +} + +static struct shash_alg sha1_avx_alg = { + .digestsize = SHA1_DIGEST_SIZE, + .init = sha1_base_init, + .update = sha1_avx_update, + .final = sha1_avx_final, + .finup = sha1_avx_finup, + .descsize = sizeof(struct sha1_state), + .base = { + .cra_name = "sha1", + .cra_driver_name = "sha1-avx", + .cra_priority = 160, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA1_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static bool avx_usable(void) { if (!cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, NULL)) { if (cpu_has_avx) @@ -130,55 +174,197 @@ static bool __init avx_usable(void) return true; } -#ifdef CONFIG_AS_AVX2 -static bool __init avx2_usable(void) +static int register_sha1_avx(void) +{ + if (avx_usable()) + return crypto_register_shash(&sha1_avx_alg); + return 0; +} + +static void unregister_sha1_avx(void) { - if (avx_usable() && cpu_has_avx2 && boot_cpu_has(X86_FEATURE_BMI1) && - boot_cpu_has(X86_FEATURE_BMI2)) + if (avx_usable()) + crypto_unregister_shash(&sha1_avx_alg); +} + +#else /* CONFIG_AS_AVX */ +static inline int register_sha1_avx(void) { return 0; } +static inline void unregister_sha1_avx(void) { } +#endif /* CONFIG_AS_AVX */ + + +#if defined(CONFIG_AS_AVX2) && (CONFIG_AS_AVX) +#define SHA1_AVX2_BLOCK_OPTSIZE 4 /* optimal 4*64 bytes of SHA1 blocks */ + +asmlinkage void sha1_transform_avx2(u32 *digest, const char *data, + unsigned int rounds); + +static bool avx2_usable(void) +{ + if (avx_usable() && boot_cpu_has(X86_FEATURE_AVX2) + && boot_cpu_has(X86_FEATURE_BMI1) + && boot_cpu_has(X86_FEATURE_BMI2)) return true; return false; } + +static void sha1_apply_transform_avx2(u32 *digest, const char *data, + unsigned int rounds) +{ + /* Select the optimal transform based on data block size */ + if (rounds >= SHA1_AVX2_BLOCK_OPTSIZE) + sha1_transform_avx2(digest, data, rounds); + else + sha1_transform_avx(digest, data, rounds); +} + +static int sha1_avx2_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha1_update(desc, data, len, + (sha1_transform_fn *) sha1_apply_transform_avx2); +} + +static int sha1_avx2_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha1_finup(desc, data, len, out, + (sha1_transform_fn *) sha1_apply_transform_avx2); +} + +static int sha1_avx2_final(struct shash_desc *desc, u8 *out) +{ + return sha1_avx2_finup(desc, NULL, 0, out); +} + +static struct shash_alg sha1_avx2_alg = { + .digestsize = SHA1_DIGEST_SIZE, + .init = sha1_base_init, + .update = sha1_avx2_update, + .final = sha1_avx2_final, + .finup = sha1_avx2_finup, + .descsize = sizeof(struct sha1_state), + .base = { + .cra_name = "sha1", + .cra_driver_name = "sha1-avx2", + .cra_priority = 170, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA1_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static int register_sha1_avx2(void) +{ + if (avx2_usable()) + return crypto_register_shash(&sha1_avx2_alg); + return 0; +} + +static void unregister_sha1_avx2(void) +{ + if (avx2_usable()) + crypto_unregister_shash(&sha1_avx2_alg); +} + +#else +static inline int register_sha1_avx2(void) { return 0; } +static inline void unregister_sha1_avx2(void) { } #endif + +#ifdef CONFIG_AS_SHA1_NI +asmlinkage void sha1_ni_transform(u32 *digest, const char *data, + unsigned int rounds); + +static int sha1_ni_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha1_update(desc, data, len, + (sha1_transform_fn *) sha1_ni_transform); +} + +static int sha1_ni_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha1_finup(desc, data, len, out, + (sha1_transform_fn *) sha1_ni_transform); +} + +static int sha1_ni_final(struct shash_desc *desc, u8 *out) +{ + return sha1_ni_finup(desc, NULL, 0, out); +} + +static struct shash_alg sha1_ni_alg = { + .digestsize = SHA1_DIGEST_SIZE, + .init = sha1_base_init, + .update = sha1_ni_update, + .final = sha1_ni_final, + .finup = sha1_ni_finup, + .descsize = sizeof(struct sha1_state), + .base = { + .cra_name = "sha1", + .cra_driver_name = "sha1-ni", + .cra_priority = 250, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA1_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static int register_sha1_ni(void) +{ + if (boot_cpu_has(X86_FEATURE_SHA_NI)) + return crypto_register_shash(&sha1_ni_alg); + return 0; +} + +static void unregister_sha1_ni(void) +{ + if (boot_cpu_has(X86_FEATURE_SHA_NI)) + crypto_unregister_shash(&sha1_ni_alg); +} + +#else +static inline int register_sha1_ni(void) { return 0; } +static inline void unregister_sha1_ni(void) { } #endif static int __init sha1_ssse3_mod_init(void) { - char *algo_name; + if (register_sha1_ssse3()) + goto fail; - /* test for SSSE3 first */ - if (cpu_has_ssse3) { - sha1_transform_asm = sha1_transform_ssse3; - algo_name = "SSSE3"; + if (register_sha1_avx()) { + unregister_sha1_ssse3(); + goto fail; } -#ifdef CONFIG_AS_AVX - /* allow AVX to override SSSE3, it's a little faster */ - if (avx_usable()) { - sha1_transform_asm = sha1_transform_avx; - algo_name = "AVX"; -#ifdef CONFIG_AS_AVX2 - /* allow AVX2 to override AVX, it's a little faster */ - if (avx2_usable()) { - sha1_transform_asm = sha1_apply_transform_avx2; - algo_name = "AVX2"; - } -#endif + if (register_sha1_avx2()) { + unregister_sha1_avx(); + unregister_sha1_ssse3(); + goto fail; } -#endif - if (sha1_transform_asm) { - pr_info("Using %s optimized SHA-1 implementation\n", algo_name); - return crypto_register_shash(&alg); + if (register_sha1_ni()) { + unregister_sha1_avx2(); + unregister_sha1_avx(); + unregister_sha1_ssse3(); + goto fail; } - pr_info("Neither AVX nor AVX2 nor SSSE3 is available/usable.\n"); + return 0; +fail: return -ENODEV; } static void __exit sha1_ssse3_mod_fini(void) { - crypto_unregister_shash(&alg); + unregister_sha1_ni(); + unregister_sha1_avx2(); + unregister_sha1_avx(); + unregister_sha1_ssse3(); } module_init(sha1_ssse3_mod_init); diff --git a/arch/x86/crypto/sha256_ni_asm.S b/arch/x86/crypto/sha256_ni_asm.S new file mode 100644 index 000000000000..748cdf21a938 --- /dev/null +++ b/arch/x86/crypto/sha256_ni_asm.S @@ -0,0 +1,353 @@ +/* + * Intel SHA Extensions optimized implementation of a SHA-256 update function + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * Contact Information: + * Sean Gulley <sean.m.gulley@intel.com> + * Tim Chen <tim.c.chen@linux.intel.com> + * + * BSD LICENSE + * + * Copyright(c) 2015 Intel Corporation. + * + * 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 the above copyright + * notice, 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 Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS 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. + * + */ + +#include <linux/linkage.h> + +#define DIGEST_PTR %rdi /* 1st arg */ +#define DATA_PTR %rsi /* 2nd arg */ +#define NUM_BLKS %rdx /* 3rd arg */ + +#define SHA256CONSTANTS %rax + +#define MSG %xmm0 +#define STATE0 %xmm1 +#define STATE1 %xmm2 +#define MSGTMP0 %xmm3 +#define MSGTMP1 %xmm4 +#define MSGTMP2 %xmm5 +#define MSGTMP3 %xmm6 +#define MSGTMP4 %xmm7 + +#define SHUF_MASK %xmm8 + +#define ABEF_SAVE %xmm9 +#define CDGH_SAVE %xmm10 + +/* + * Intel SHA Extensions optimized implementation of a SHA-256 update function + * + * The function takes a pointer to the current hash values, a pointer to the + * input data, and a number of 64 byte blocks to process. Once all blocks have + * been processed, the digest pointer is updated with the resulting hash value. + * The function only processes complete blocks, there is no functionality to + * store partial blocks. All message padding and hash value initialization must + * be done outside the update function. + * + * The indented lines in the loop are instructions related to rounds processing. + * The non-indented lines are instructions related to the message schedule. + * + * void sha256_ni_transform(uint32_t *digest, const void *data, + uint32_t numBlocks); + * digest : pointer to digest + * data: pointer to input data + * numBlocks: Number of blocks to process + */ + +.text +.align 32 +ENTRY(sha256_ni_transform) + + shl $6, NUM_BLKS /* convert to bytes */ + jz .Ldone_hash + add DATA_PTR, NUM_BLKS /* pointer to end of data */ + + /* + * load initial hash values + * Need to reorder these appropriately + * DCBA, HGFE -> ABEF, CDGH + */ + movdqu 0*16(DIGEST_PTR), STATE0 + movdqu 1*16(DIGEST_PTR), STATE1 + + pshufd $0xB1, STATE0, STATE0 /* CDAB */ + pshufd $0x1B, STATE1, STATE1 /* EFGH */ + movdqa STATE0, MSGTMP4 + palignr $8, STATE1, STATE0 /* ABEF */ + pblendw $0xF0, MSGTMP4, STATE1 /* CDGH */ + + movdqa PSHUFFLE_BYTE_FLIP_MASK(%rip), SHUF_MASK + lea K256(%rip), SHA256CONSTANTS + +.Lloop0: + /* Save hash values for addition after rounds */ + movdqa STATE0, ABEF_SAVE + movdqa STATE1, CDGH_SAVE + + /* Rounds 0-3 */ + movdqu 0*16(DATA_PTR), MSG + pshufb SHUF_MASK, MSG + movdqa MSG, MSGTMP0 + paddd 0*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + + /* Rounds 4-7 */ + movdqu 1*16(DATA_PTR), MSG + pshufb SHUF_MASK, MSG + movdqa MSG, MSGTMP1 + paddd 1*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP1, MSGTMP0 + + /* Rounds 8-11 */ + movdqu 2*16(DATA_PTR), MSG + pshufb SHUF_MASK, MSG + movdqa MSG, MSGTMP2 + paddd 2*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP2, MSGTMP1 + + /* Rounds 12-15 */ + movdqu 3*16(DATA_PTR), MSG + pshufb SHUF_MASK, MSG + movdqa MSG, MSGTMP3 + paddd 3*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP3, MSGTMP4 + palignr $4, MSGTMP2, MSGTMP4 + paddd MSGTMP4, MSGTMP0 + sha256msg2 MSGTMP3, MSGTMP0 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP3, MSGTMP2 + + /* Rounds 16-19 */ + movdqa MSGTMP0, MSG + paddd 4*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP0, MSGTMP4 + palignr $4, MSGTMP3, MSGTMP4 + paddd MSGTMP4, MSGTMP1 + sha256msg2 MSGTMP0, MSGTMP1 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP0, MSGTMP3 + + /* Rounds 20-23 */ + movdqa MSGTMP1, MSG + paddd 5*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP1, MSGTMP4 + palignr $4, MSGTMP0, MSGTMP4 + paddd MSGTMP4, MSGTMP2 + sha256msg2 MSGTMP1, MSGTMP2 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP1, MSGTMP0 + + /* Rounds 24-27 */ + movdqa MSGTMP2, MSG + paddd 6*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP2, MSGTMP4 + palignr $4, MSGTMP1, MSGTMP4 + paddd MSGTMP4, MSGTMP3 + sha256msg2 MSGTMP2, MSGTMP3 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP2, MSGTMP1 + + /* Rounds 28-31 */ + movdqa MSGTMP3, MSG + paddd 7*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP3, MSGTMP4 + palignr $4, MSGTMP2, MSGTMP4 + paddd MSGTMP4, MSGTMP0 + sha256msg2 MSGTMP3, MSGTMP0 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP3, MSGTMP2 + + /* Rounds 32-35 */ + movdqa MSGTMP0, MSG + paddd 8*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP0, MSGTMP4 + palignr $4, MSGTMP3, MSGTMP4 + paddd MSGTMP4, MSGTMP1 + sha256msg2 MSGTMP0, MSGTMP1 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP0, MSGTMP3 + + /* Rounds 36-39 */ + movdqa MSGTMP1, MSG + paddd 9*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP1, MSGTMP4 + palignr $4, MSGTMP0, MSGTMP4 + paddd MSGTMP4, MSGTMP2 + sha256msg2 MSGTMP1, MSGTMP2 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP1, MSGTMP0 + + /* Rounds 40-43 */ + movdqa MSGTMP2, MSG + paddd 10*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP2, MSGTMP4 + palignr $4, MSGTMP1, MSGTMP4 + paddd MSGTMP4, MSGTMP3 + sha256msg2 MSGTMP2, MSGTMP3 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP2, MSGTMP1 + + /* Rounds 44-47 */ + movdqa MSGTMP3, MSG + paddd 11*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP3, MSGTMP4 + palignr $4, MSGTMP2, MSGTMP4 + paddd MSGTMP4, MSGTMP0 + sha256msg2 MSGTMP3, MSGTMP0 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP3, MSGTMP2 + + /* Rounds 48-51 */ + movdqa MSGTMP0, MSG + paddd 12*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP0, MSGTMP4 + palignr $4, MSGTMP3, MSGTMP4 + paddd MSGTMP4, MSGTMP1 + sha256msg2 MSGTMP0, MSGTMP1 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + sha256msg1 MSGTMP0, MSGTMP3 + + /* Rounds 52-55 */ + movdqa MSGTMP1, MSG + paddd 13*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP1, MSGTMP4 + palignr $4, MSGTMP0, MSGTMP4 + paddd MSGTMP4, MSGTMP2 + sha256msg2 MSGTMP1, MSGTMP2 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + + /* Rounds 56-59 */ + movdqa MSGTMP2, MSG + paddd 14*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + movdqa MSGTMP2, MSGTMP4 + palignr $4, MSGTMP1, MSGTMP4 + paddd MSGTMP4, MSGTMP3 + sha256msg2 MSGTMP2, MSGTMP3 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + + /* Rounds 60-63 */ + movdqa MSGTMP3, MSG + paddd 15*16(SHA256CONSTANTS), MSG + sha256rnds2 STATE0, STATE1 + pshufd $0x0E, MSG, MSG + sha256rnds2 STATE1, STATE0 + + /* Add current hash values with previously saved */ + paddd ABEF_SAVE, STATE0 + paddd CDGH_SAVE, STATE1 + + /* Increment data pointer and loop if more to process */ + add $64, DATA_PTR + cmp NUM_BLKS, DATA_PTR + jne .Lloop0 + + /* Write hash values back in the correct order */ + pshufd $0x1B, STATE0, STATE0 /* FEBA */ + pshufd $0xB1, STATE1, STATE1 /* DCHG */ + movdqa STATE0, MSGTMP4 + pblendw $0xF0, STATE1, STATE0 /* DCBA */ + palignr $8, MSGTMP4, STATE1 /* HGFE */ + + movdqu STATE0, 0*16(DIGEST_PTR) + movdqu STATE1, 1*16(DIGEST_PTR) + +.Ldone_hash: + + ret +ENDPROC(sha256_ni_transform) + +.data +.align 64 +K256: + .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 + +PSHUFFLE_BYTE_FLIP_MASK: + .octa 0x0c0d0e0f08090a0b0405060700010203 diff --git a/arch/x86/crypto/sha256_ssse3_glue.c b/arch/x86/crypto/sha256_ssse3_glue.c index 0e0e85aea634..5f4d6086dc59 100644 --- a/arch/x86/crypto/sha256_ssse3_glue.c +++ b/arch/x86/crypto/sha256_ssse3_glue.c @@ -42,19 +42,10 @@ asmlinkage void sha256_transform_ssse3(u32 *digest, const char *data, u64 rounds); -#ifdef CONFIG_AS_AVX -asmlinkage void sha256_transform_avx(u32 *digest, const char *data, - u64 rounds); -#endif -#ifdef CONFIG_AS_AVX2 -asmlinkage void sha256_transform_rorx(u32 *digest, const char *data, - u64 rounds); -#endif - -static void (*sha256_transform_asm)(u32 *, const char *, u64); +typedef void (sha256_transform_fn)(u32 *digest, const char *data, u64 rounds); -static int sha256_ssse3_update(struct shash_desc *desc, const u8 *data, - unsigned int len) +static int sha256_update(struct shash_desc *desc, const u8 *data, + unsigned int len, sha256_transform_fn *sha256_xform) { struct sha256_state *sctx = shash_desc_ctx(desc); @@ -67,14 +58,14 @@ static int sha256_ssse3_update(struct shash_desc *desc, const u8 *data, kernel_fpu_begin(); sha256_base_do_update(desc, data, len, - (sha256_block_fn *)sha256_transform_asm); + (sha256_block_fn *)sha256_xform); kernel_fpu_end(); return 0; } -static int sha256_ssse3_finup(struct shash_desc *desc, const u8 *data, - unsigned int len, u8 *out) +static int sha256_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out, sha256_transform_fn *sha256_xform) { if (!irq_fpu_usable()) return crypto_sha256_finup(desc, data, len, out); @@ -82,20 +73,32 @@ static int sha256_ssse3_finup(struct shash_desc *desc, const u8 *data, kernel_fpu_begin(); if (len) sha256_base_do_update(desc, data, len, - (sha256_block_fn *)sha256_transform_asm); - sha256_base_do_finalize(desc, (sha256_block_fn *)sha256_transform_asm); + (sha256_block_fn *)sha256_xform); + sha256_base_do_finalize(desc, (sha256_block_fn *)sha256_xform); kernel_fpu_end(); return sha256_base_finish(desc, out); } +static int sha256_ssse3_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha256_update(desc, data, len, sha256_transform_ssse3); +} + +static int sha256_ssse3_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha256_finup(desc, data, len, out, sha256_transform_ssse3); +} + /* Add padding and return the message digest. */ static int sha256_ssse3_final(struct shash_desc *desc, u8 *out) { return sha256_ssse3_finup(desc, NULL, 0, out); } -static struct shash_alg algs[] = { { +static struct shash_alg sha256_ssse3_algs[] = { { .digestsize = SHA256_DIGEST_SIZE, .init = sha256_base_init, .update = sha256_ssse3_update, @@ -127,8 +130,75 @@ static struct shash_alg algs[] = { { } } }; +static int register_sha256_ssse3(void) +{ + if (boot_cpu_has(X86_FEATURE_SSSE3)) + return crypto_register_shashes(sha256_ssse3_algs, + ARRAY_SIZE(sha256_ssse3_algs)); + return 0; +} + +static void unregister_sha256_ssse3(void) +{ + if (boot_cpu_has(X86_FEATURE_SSSE3)) + crypto_unregister_shashes(sha256_ssse3_algs, + ARRAY_SIZE(sha256_ssse3_algs)); +} + #ifdef CONFIG_AS_AVX -static bool __init avx_usable(void) +asmlinkage void sha256_transform_avx(u32 *digest, const char *data, + u64 rounds); + +static int sha256_avx_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha256_update(desc, data, len, sha256_transform_avx); +} + +static int sha256_avx_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha256_finup(desc, data, len, out, sha256_transform_avx); +} + +static int sha256_avx_final(struct shash_desc *desc, u8 *out) +{ + return sha256_avx_finup(desc, NULL, 0, out); +} + +static struct shash_alg sha256_avx_algs[] = { { + .digestsize = SHA256_DIGEST_SIZE, + .init = sha256_base_init, + .update = sha256_avx_update, + .final = sha256_avx_final, + .finup = sha256_avx_finup, + .descsize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha256", + .cra_driver_name = "sha256-avx", + .cra_priority = 160, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA256_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}, { + .digestsize = SHA224_DIGEST_SIZE, + .init = sha224_base_init, + .update = sha256_avx_update, + .final = sha256_avx_final, + .finup = sha256_avx_finup, + .descsize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha224", + .cra_driver_name = "sha224-avx", + .cra_priority = 160, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA224_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +} }; + +static bool avx_usable(void) { if (!cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, NULL)) { if (cpu_has_avx) @@ -138,47 +208,216 @@ static bool __init avx_usable(void) return true; } -#endif -static int __init sha256_ssse3_mod_init(void) +static int register_sha256_avx(void) { - /* test for SSSE3 first */ - if (cpu_has_ssse3) - sha256_transform_asm = sha256_transform_ssse3; + if (avx_usable()) + return crypto_register_shashes(sha256_avx_algs, + ARRAY_SIZE(sha256_avx_algs)); + return 0; +} -#ifdef CONFIG_AS_AVX - /* allow AVX to override SSSE3, it's a little faster */ - if (avx_usable()) { -#ifdef CONFIG_AS_AVX2 - if (boot_cpu_has(X86_FEATURE_AVX2) && boot_cpu_has(X86_FEATURE_BMI2)) - sha256_transform_asm = sha256_transform_rorx; - else +static void unregister_sha256_avx(void) +{ + if (avx_usable()) + crypto_unregister_shashes(sha256_avx_algs, + ARRAY_SIZE(sha256_avx_algs)); +} + +#else +static inline int register_sha256_avx(void) { return 0; } +static inline void unregister_sha256_avx(void) { } #endif - sha256_transform_asm = sha256_transform_avx; + +#if defined(CONFIG_AS_AVX2) && defined(CONFIG_AS_AVX) +asmlinkage void sha256_transform_rorx(u32 *digest, const char *data, + u64 rounds); + +static int sha256_avx2_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha256_update(desc, data, len, sha256_transform_rorx); +} + +static int sha256_avx2_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha256_finup(desc, data, len, out, sha256_transform_rorx); +} + +static int sha256_avx2_final(struct shash_desc *desc, u8 *out) +{ + return sha256_avx2_finup(desc, NULL, 0, out); +} + +static struct shash_alg sha256_avx2_algs[] = { { + .digestsize = SHA256_DIGEST_SIZE, + .init = sha256_base_init, + .update = sha256_avx2_update, + .final = sha256_avx2_final, + .finup = sha256_avx2_finup, + .descsize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha256", + .cra_driver_name = "sha256-avx2", + .cra_priority = 170, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA256_BLOCK_SIZE, + .cra_module = THIS_MODULE, } -#endif +}, { + .digestsize = SHA224_DIGEST_SIZE, + .init = sha224_base_init, + .update = sha256_avx2_update, + .final = sha256_avx2_final, + .finup = sha256_avx2_finup, + .descsize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha224", + .cra_driver_name = "sha224-avx2", + .cra_priority = 170, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA224_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +} }; - if (sha256_transform_asm) { -#ifdef CONFIG_AS_AVX - if (sha256_transform_asm == sha256_transform_avx) - pr_info("Using AVX optimized SHA-256 implementation\n"); -#ifdef CONFIG_AS_AVX2 - else if (sha256_transform_asm == sha256_transform_rorx) - pr_info("Using AVX2 optimized SHA-256 implementation\n"); +static bool avx2_usable(void) +{ + if (avx_usable() && boot_cpu_has(X86_FEATURE_AVX2) && + boot_cpu_has(X86_FEATURE_BMI2)) + return true; + + return false; +} + +static int register_sha256_avx2(void) +{ + if (avx2_usable()) + return crypto_register_shashes(sha256_avx2_algs, + ARRAY_SIZE(sha256_avx2_algs)); + return 0; +} + +static void unregister_sha256_avx2(void) +{ + if (avx2_usable()) + crypto_unregister_shashes(sha256_avx2_algs, + ARRAY_SIZE(sha256_avx2_algs)); +} + +#else +static inline int register_sha256_avx2(void) { return 0; } +static inline void unregister_sha256_avx2(void) { } #endif - else + +#ifdef CONFIG_AS_SHA256_NI +asmlinkage void sha256_ni_transform(u32 *digest, const char *data, + u64 rounds); /*unsigned int rounds);*/ + +static int sha256_ni_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha256_update(desc, data, len, sha256_ni_transform); +} + +static int sha256_ni_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha256_finup(desc, data, len, out, sha256_ni_transform); +} + +static int sha256_ni_final(struct shash_desc *desc, u8 *out) +{ + return sha256_ni_finup(desc, NULL, 0, out); +} + +static struct shash_alg sha256_ni_algs[] = { { + .digestsize = SHA256_DIGEST_SIZE, + .init = sha256_base_init, + .update = sha256_ni_update, + .final = sha256_ni_final, + .finup = sha256_ni_finup, + .descsize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha256", + .cra_driver_name = "sha256-ni", + .cra_priority = 250, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA256_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}, { + .digestsize = SHA224_DIGEST_SIZE, + .init = sha224_base_init, + .update = sha256_ni_update, + .final = sha256_ni_final, + .finup = sha256_ni_finup, + .descsize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha224", + .cra_driver_name = "sha224-ni", + .cra_priority = 250, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA224_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +} }; + +static int register_sha256_ni(void) +{ + if (boot_cpu_has(X86_FEATURE_SHA_NI)) + return crypto_register_shashes(sha256_ni_algs, + ARRAY_SIZE(sha256_ni_algs)); + return 0; +} + +static void unregister_sha256_ni(void) +{ + if (boot_cpu_has(X86_FEATURE_SHA_NI)) + crypto_unregister_shashes(sha256_ni_algs, + ARRAY_SIZE(sha256_ni_algs)); +} + +#else +static inline int register_sha256_ni(void) { return 0; } +static inline void unregister_sha256_ni(void) { } #endif - pr_info("Using SSSE3 optimized SHA-256 implementation\n"); - return crypto_register_shashes(algs, ARRAY_SIZE(algs)); + +static int __init sha256_ssse3_mod_init(void) +{ + if (register_sha256_ssse3()) + goto fail; + + if (register_sha256_avx()) { + unregister_sha256_ssse3(); + goto fail; } - pr_info("Neither AVX nor SSSE3 is available/usable.\n"); + if (register_sha256_avx2()) { + unregister_sha256_avx(); + unregister_sha256_ssse3(); + goto fail; + } + + if (register_sha256_ni()) { + unregister_sha256_avx2(); + unregister_sha256_avx(); + unregister_sha256_ssse3(); + goto fail; + } + + return 0; +fail: return -ENODEV; } static void __exit sha256_ssse3_mod_fini(void) { - crypto_unregister_shashes(algs, ARRAY_SIZE(algs)); + unregister_sha256_ni(); + unregister_sha256_avx2(); + unregister_sha256_avx(); + unregister_sha256_ssse3(); } module_init(sha256_ssse3_mod_init); diff --git a/arch/x86/crypto/sha512_ssse3_glue.c b/arch/x86/crypto/sha512_ssse3_glue.c index 0c8c38c101ac..34e5083d6f36 100644 --- a/arch/x86/crypto/sha512_ssse3_glue.c +++ b/arch/x86/crypto/sha512_ssse3_glue.c @@ -41,19 +41,11 @@ asmlinkage void sha512_transform_ssse3(u64 *digest, const char *data, u64 rounds); -#ifdef CONFIG_AS_AVX -asmlinkage void sha512_transform_avx(u64 *digest, const char *data, - u64 rounds); -#endif -#ifdef CONFIG_AS_AVX2 -asmlinkage void sha512_transform_rorx(u64 *digest, const char *data, - u64 rounds); -#endif -static void (*sha512_transform_asm)(u64 *, const char *, u64); +typedef void (sha512_transform_fn)(u64 *digest, const char *data, u64 rounds); -static int sha512_ssse3_update(struct shash_desc *desc, const u8 *data, - unsigned int len) +static int sha512_update(struct shash_desc *desc, const u8 *data, + unsigned int len, sha512_transform_fn *sha512_xform) { struct sha512_state *sctx = shash_desc_ctx(desc); @@ -66,14 +58,14 @@ static int sha512_ssse3_update(struct shash_desc *desc, const u8 *data, kernel_fpu_begin(); sha512_base_do_update(desc, data, len, - (sha512_block_fn *)sha512_transform_asm); + (sha512_block_fn *)sha512_xform); kernel_fpu_end(); return 0; } -static int sha512_ssse3_finup(struct shash_desc *desc, const u8 *data, - unsigned int len, u8 *out) +static int sha512_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out, sha512_transform_fn *sha512_xform) { if (!irq_fpu_usable()) return crypto_sha512_finup(desc, data, len, out); @@ -81,20 +73,32 @@ static int sha512_ssse3_finup(struct shash_desc *desc, const u8 *data, kernel_fpu_begin(); if (len) sha512_base_do_update(desc, data, len, - (sha512_block_fn *)sha512_transform_asm); - sha512_base_do_finalize(desc, (sha512_block_fn *)sha512_transform_asm); + (sha512_block_fn *)sha512_xform); + sha512_base_do_finalize(desc, (sha512_block_fn *)sha512_xform); kernel_fpu_end(); return sha512_base_finish(desc, out); } +static int sha512_ssse3_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha512_update(desc, data, len, sha512_transform_ssse3); +} + +static int sha512_ssse3_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha512_finup(desc, data, len, out, sha512_transform_ssse3); +} + /* Add padding and return the message digest. */ static int sha512_ssse3_final(struct shash_desc *desc, u8 *out) { return sha512_ssse3_finup(desc, NULL, 0, out); } -static struct shash_alg algs[] = { { +static struct shash_alg sha512_ssse3_algs[] = { { .digestsize = SHA512_DIGEST_SIZE, .init = sha512_base_init, .update = sha512_ssse3_update, @@ -126,8 +130,25 @@ static struct shash_alg algs[] = { { } } }; +static int register_sha512_ssse3(void) +{ + if (boot_cpu_has(X86_FEATURE_SSSE3)) + return crypto_register_shashes(sha512_ssse3_algs, + ARRAY_SIZE(sha512_ssse3_algs)); + return 0; +} + +static void unregister_sha512_ssse3(void) +{ + if (boot_cpu_has(X86_FEATURE_SSSE3)) + crypto_unregister_shashes(sha512_ssse3_algs, + ARRAY_SIZE(sha512_ssse3_algs)); +} + #ifdef CONFIG_AS_AVX -static bool __init avx_usable(void) +asmlinkage void sha512_transform_avx(u64 *digest, const char *data, + u64 rounds); +static bool avx_usable(void) { if (!cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, NULL)) { if (cpu_has_avx) @@ -137,47 +158,185 @@ static bool __init avx_usable(void) return true; } -#endif -static int __init sha512_ssse3_mod_init(void) +static int sha512_avx_update(struct shash_desc *desc, const u8 *data, + unsigned int len) { - /* test for SSSE3 first */ - if (cpu_has_ssse3) - sha512_transform_asm = sha512_transform_ssse3; + return sha512_update(desc, data, len, sha512_transform_avx); +} -#ifdef CONFIG_AS_AVX - /* allow AVX to override SSSE3, it's a little faster */ - if (avx_usable()) { -#ifdef CONFIG_AS_AVX2 - if (boot_cpu_has(X86_FEATURE_AVX2)) - sha512_transform_asm = sha512_transform_rorx; - else -#endif - sha512_transform_asm = sha512_transform_avx; +static int sha512_avx_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha512_finup(desc, data, len, out, sha512_transform_avx); +} + +/* Add padding and return the message digest. */ +static int sha512_avx_final(struct shash_desc *desc, u8 *out) +{ + return sha512_avx_finup(desc, NULL, 0, out); +} + +static struct shash_alg sha512_avx_algs[] = { { + .digestsize = SHA512_DIGEST_SIZE, + .init = sha512_base_init, + .update = sha512_avx_update, + .final = sha512_avx_final, + .finup = sha512_avx_finup, + .descsize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha512", + .cra_driver_name = "sha512-avx", + .cra_priority = 160, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA512_BLOCK_SIZE, + .cra_module = THIS_MODULE, } -#endif +}, { + .digestsize = SHA384_DIGEST_SIZE, + .init = sha384_base_init, + .update = sha512_avx_update, + .final = sha512_avx_final, + .finup = sha512_avx_finup, + .descsize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha384", + .cra_driver_name = "sha384-avx", + .cra_priority = 160, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA384_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +} }; - if (sha512_transform_asm) { -#ifdef CONFIG_AS_AVX - if (sha512_transform_asm == sha512_transform_avx) - pr_info("Using AVX optimized SHA-512 implementation\n"); -#ifdef CONFIG_AS_AVX2 - else if (sha512_transform_asm == sha512_transform_rorx) - pr_info("Using AVX2 optimized SHA-512 implementation\n"); +static int register_sha512_avx(void) +{ + if (avx_usable()) + return crypto_register_shashes(sha512_avx_algs, + ARRAY_SIZE(sha512_avx_algs)); + return 0; +} + +static void unregister_sha512_avx(void) +{ + if (avx_usable()) + crypto_unregister_shashes(sha512_avx_algs, + ARRAY_SIZE(sha512_avx_algs)); +} +#else +static inline int register_sha512_avx(void) { return 0; } +static inline void unregister_sha512_avx(void) { } #endif - else + +#if defined(CONFIG_AS_AVX2) && defined(CONFIG_AS_AVX) +asmlinkage void sha512_transform_rorx(u64 *digest, const char *data, + u64 rounds); + +static int sha512_avx2_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + return sha512_update(desc, data, len, sha512_transform_rorx); +} + +static int sha512_avx2_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return sha512_finup(desc, data, len, out, sha512_transform_rorx); +} + +/* Add padding and return the message digest. */ +static int sha512_avx2_final(struct shash_desc *desc, u8 *out) +{ + return sha512_avx2_finup(desc, NULL, 0, out); +} + +static struct shash_alg sha512_avx2_algs[] = { { + .digestsize = SHA512_DIGEST_SIZE, + .init = sha512_base_init, + .update = sha512_avx2_update, + .final = sha512_avx2_final, + .finup = sha512_avx2_finup, + .descsize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha512", + .cra_driver_name = "sha512-avx2", + .cra_priority = 170, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA512_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}, { + .digestsize = SHA384_DIGEST_SIZE, + .init = sha384_base_init, + .update = sha512_avx2_update, + .final = sha512_avx2_final, + .finup = sha512_avx2_finup, + .descsize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha384", + .cra_driver_name = "sha384-avx2", + .cra_priority = 170, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA384_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +} }; + +static bool avx2_usable(void) +{ + if (avx_usable() && boot_cpu_has(X86_FEATURE_AVX2) && + boot_cpu_has(X86_FEATURE_BMI2)) + return true; + + return false; +} + +static int register_sha512_avx2(void) +{ + if (avx2_usable()) + return crypto_register_shashes(sha512_avx2_algs, + ARRAY_SIZE(sha512_avx2_algs)); + return 0; +} + +static void unregister_sha512_avx2(void) +{ + if (avx2_usable()) + crypto_unregister_shashes(sha512_avx2_algs, + ARRAY_SIZE(sha512_avx2_algs)); +} +#else +static inline int register_sha512_avx2(void) { return 0; } +static inline void unregister_sha512_avx2(void) { } #endif - pr_info("Using SSSE3 optimized SHA-512 implementation\n"); - return crypto_register_shashes(algs, ARRAY_SIZE(algs)); + +static int __init sha512_ssse3_mod_init(void) +{ + + if (register_sha512_ssse3()) + goto fail; + + if (register_sha512_avx()) { + unregister_sha512_ssse3(); + goto fail; } - pr_info("Neither AVX nor SSSE3 is available/usable.\n"); + if (register_sha512_avx2()) { + unregister_sha512_avx(); + unregister_sha512_ssse3(); + goto fail; + } + + return 0; +fail: return -ENODEV; } static void __exit sha512_ssse3_mod_fini(void) { - crypto_unregister_shashes(algs, ARRAY_SIZE(algs)); + unregister_sha512_avx2(); + unregister_sha512_avx(); + unregister_sha512_ssse3(); } module_init(sha512_ssse3_mod_init); diff --git a/crypto/Kconfig b/crypto/Kconfig index 48ee3e175dac..7240821137fd 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -348,6 +348,13 @@ config CRYPTO_XTS key size 256, 384 or 512 bits. This implementation currently can't handle a sectorsize which is not a multiple of 16 bytes. +config CRYPTO_KEYWRAP + tristate "Key wrapping support" + select CRYPTO_BLKCIPHER + help + Support for key wrapping (NIST SP800-38F / RFC3394) without + padding. + comment "Hash modes" config CRYPTO_CMAC @@ -597,17 +604,18 @@ config CRYPTO_SHA1 SHA-1 secure hash standard (FIPS 180-1/DFIPS 180-2). config CRYPTO_SHA1_SSSE3 - tristate "SHA1 digest algorithm (SSSE3/AVX/AVX2)" + tristate "SHA1 digest algorithm (SSSE3/AVX/AVX2/SHA-NI)" depends on X86 && 64BIT select CRYPTO_SHA1 select CRYPTO_HASH help SHA-1 secure hash standard (FIPS 180-1/DFIPS 180-2) implemented using Supplemental SSE3 (SSSE3) instructions or Advanced Vector - Extensions (AVX/AVX2), when available. + Extensions (AVX/AVX2) or SHA-NI(SHA Extensions New Instructions), + when available. config CRYPTO_SHA256_SSSE3 - tristate "SHA256 digest algorithm (SSSE3/AVX/AVX2)" + tristate "SHA256 digest algorithm (SSSE3/AVX/AVX2/SHA-NI)" depends on X86 && 64BIT select CRYPTO_SHA256 select CRYPTO_HASH @@ -615,7 +623,8 @@ config CRYPTO_SHA256_SSSE3 SHA-256 secure hash standard (DFIPS 180-2) implemented using Supplemental SSE3 (SSSE3) instructions, or Advanced Vector Extensions version 1 (AVX1), or Advanced Vector Extensions - version 2 (AVX2) instructions, when available. + version 2 (AVX2) instructions, or SHA-NI (SHA Extensions New + Instructions) when available. config CRYPTO_SHA512_SSSE3 tristate "SHA512 digest algorithm (SSSE3/AVX/AVX2)" diff --git a/crypto/Makefile b/crypto/Makefile index e2c59819b236..f7aba923458d 100644 --- a/crypto/Makefile +++ b/crypto/Makefile @@ -31,10 +31,13 @@ obj-$(CONFIG_CRYPTO_HASH2) += crypto_hash.o obj-$(CONFIG_CRYPTO_PCOMP2) += pcompress.o obj-$(CONFIG_CRYPTO_AKCIPHER2) += akcipher.o -$(obj)/rsakey-asn1.o: $(obj)/rsakey-asn1.c $(obj)/rsakey-asn1.h -clean-files += rsakey-asn1.c rsakey-asn1.h +$(obj)/rsapubkey-asn1.o: $(obj)/rsapubkey-asn1.c $(obj)/rsapubkey-asn1.h +$(obj)/rsaprivkey-asn1.o: $(obj)/rsaprivkey-asn1.c $(obj)/rsaprivkey-asn1.h +clean-files += rsapubkey-asn1.c rsapubkey-asn1.h +clean-files += rsaprivkey-asn1.c rsaprivkey-asn1.h -rsa_generic-y := rsakey-asn1.o +rsa_generic-y := rsapubkey-asn1.o +rsa_generic-y += rsaprivkey-asn1.o rsa_generic-y += rsa.o rsa_generic-y += rsa_helper.o obj-$(CONFIG_CRYPTO_RSA) += rsa_generic.o @@ -67,6 +70,7 @@ obj-$(CONFIG_CRYPTO_CTS) += cts.o obj-$(CONFIG_CRYPTO_LRW) += lrw.o obj-$(CONFIG_CRYPTO_XTS) += xts.o obj-$(CONFIG_CRYPTO_CTR) += ctr.o +obj-$(CONFIG_CRYPTO_KEYWRAP) += keywrap.o obj-$(CONFIG_CRYPTO_GCM) += gcm.o obj-$(CONFIG_CRYPTO_CCM) += ccm.o obj-$(CONFIG_CRYPTO_CHACHA20POLY1305) += chacha20poly1305.o diff --git a/crypto/akcipher.c b/crypto/akcipher.c index 528ae6aa9bff..120ec042ec9e 100644 --- a/crypto/akcipher.c +++ b/crypto/akcipher.c @@ -21,7 +21,6 @@ #include <linux/cryptouser.h> #include <net/netlink.h> #include <crypto/akcipher.h> -#include <crypto/public_key.h> #include "internal.h" #ifdef CONFIG_NET diff --git a/crypto/asymmetric_keys/pkcs7_verify.c b/crypto/asymmetric_keys/pkcs7_verify.c index d20c0b4b880e..325575caf6b4 100644 --- a/crypto/asymmetric_keys/pkcs7_verify.c +++ b/crypto/asymmetric_keys/pkcs7_verify.c @@ -49,11 +49,12 @@ static int pkcs7_digest(struct pkcs7_message *pkcs7, sinfo->sig.digest_size = digest_size = crypto_shash_digestsize(tfm); ret = -ENOMEM; - digest = kzalloc(digest_size + desc_size, GFP_KERNEL); + digest = kzalloc(ALIGN(digest_size, __alignof__(*desc)) + desc_size, + GFP_KERNEL); if (!digest) goto error_no_desc; - desc = digest + digest_size; + desc = PTR_ALIGN(digest + digest_size, __alignof__(*desc)); desc->tfm = tfm; desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP; diff --git a/crypto/asymmetric_keys/x509_cert_parser.c b/crypto/asymmetric_keys/x509_cert_parser.c index af71878dc15b..3000ea3b6687 100644 --- a/crypto/asymmetric_keys/x509_cert_parser.c +++ b/crypto/asymmetric_keys/x509_cert_parser.c @@ -546,9 +546,9 @@ int x509_decode_time(time64_t *_t, size_t hdrlen, if (year < 1970 || mon < 1 || mon > 12 || day < 1 || day > mon_len || - hour < 0 || hour > 23 || - min < 0 || min > 59 || - sec < 0 || sec > 59) + hour > 23 || + min > 59 || + sec > 59) goto invalid_time; *_t = mktime64(year, mon, day, hour, min, sec); diff --git a/crypto/asymmetric_keys/x509_public_key.c b/crypto/asymmetric_keys/x509_public_key.c index 197096632412..68c3c40501ab 100644 --- a/crypto/asymmetric_keys/x509_public_key.c +++ b/crypto/asymmetric_keys/x509_public_key.c @@ -194,14 +194,15 @@ int x509_get_sig_params(struct x509_certificate *cert) * digest storage space. */ ret = -ENOMEM; - digest = kzalloc(digest_size + desc_size, GFP_KERNEL); + digest = kzalloc(ALIGN(digest_size, __alignof__(*desc)) + desc_size, + GFP_KERNEL); if (!digest) goto error; cert->sig.digest = digest; cert->sig.digest_size = digest_size; - desc = digest + digest_size; + desc = PTR_ALIGN(digest + digest_size, __alignof__(*desc)); desc->tfm = tfm; desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP; diff --git a/crypto/jitterentropy-kcapi.c b/crypto/jitterentropy-kcapi.c index ceea83d13168..597cedd3531c 100644 --- a/crypto/jitterentropy-kcapi.c +++ b/crypto/jitterentropy-kcapi.c @@ -98,10 +98,6 @@ void jent_get_nstime(__u64 *out) * If random_get_entropy does not return a value (which is possible on, * for example, MIPS), invoke __getnstimeofday * hoping that there are timers we can work with. - * - * The list of available timers can be obtained from - * /sys/devices/system/clocksource/clocksource0/available_clocksource - * and are registered with clocksource_register() */ if ((0 == tmp) && (0 == __getnstimeofday(&ts))) { diff --git a/crypto/keywrap.c b/crypto/keywrap.c new file mode 100644 index 000000000000..b1d106ce55f3 --- /dev/null +++ b/crypto/keywrap.c @@ -0,0 +1,419 @@ +/* + * Key Wrapping: RFC3394 / NIST SP800-38F + * + * Copyright (C) 2015, Stephan Mueller <smueller@chronox.de> + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, and the entire permission notice in its entirety, + * including the disclaimer of warranties. + * 2. 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. + * 3. The name of the author may not be used to endorse or promote + * products derived from this software without specific prior + * written permission. + * + * ALTERNATIVELY, this product may be distributed under the terms of + * the GNU General Public License, in which case the provisions of the GPL2 + * are required INSTEAD OF the above restrictions. (This clause is + * necessary due to a potential bad interaction between the GPL and + * the restrictions contained in a BSD-style copyright.) + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES + * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, ALL OF + * WHICH ARE HEREBY DISCLAIMED. IN NO EVENT SHALL THE AUTHOR 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 NOT ADVISED OF THE POSSIBILITY OF SUCH + * DAMAGE. + */ + +/* + * Note for using key wrapping: + * + * * The result of the encryption operation is the ciphertext starting + * with the 2nd semiblock. The first semiblock is provided as the IV. + * The IV used to start the encryption operation is the default IV. + * + * * The input for the decryption is the first semiblock handed in as an + * IV. The ciphertext is the data starting with the 2nd semiblock. The + * return code of the decryption operation will be EBADMSG in case an + * integrity error occurs. + * + * To obtain the full result of an encryption as expected by SP800-38F, the + * caller must allocate a buffer of plaintext + 8 bytes: + * + * unsigned int datalen = ptlen + crypto_skcipher_ivsize(tfm); + * u8 data[datalen]; + * u8 *iv = data; + * u8 *pt = data + crypto_skcipher_ivsize(tfm); + * <ensure that pt contains the plaintext of size ptlen> + * sg_init_one(&sg, ptdata, ptlen); + * skcipher_request_set_crypt(req, &sg, &sg, ptlen, iv); + * + * ==> After encryption, data now contains full KW result as per SP800-38F. + * + * In case of decryption, ciphertext now already has the expected length + * and must be segmented appropriately: + * + * unsigned int datalen = CTLEN; + * u8 data[datalen]; + * <ensure that data contains full ciphertext> + * u8 *iv = data; + * u8 *ct = data + crypto_skcipher_ivsize(tfm); + * unsigned int ctlen = datalen - crypto_skcipher_ivsize(tfm); + * sg_init_one(&sg, ctdata, ctlen); + * skcipher_request_set_crypt(req, &sg, &sg, ptlen, iv); + * + * ==> After decryption (which hopefully does not return EBADMSG), the ct + * pointer now points to the plaintext of size ctlen. + * + * Note 2: KWP is not implemented as this would defy in-place operation. + * If somebody wants to wrap non-aligned data, he should simply pad + * the input with zeros to fill it up to the 8 byte boundary. + */ + +#include <linux/module.h> +#include <linux/crypto.h> +#include <linux/scatterlist.h> +#include <crypto/scatterwalk.h> +#include <crypto/internal/skcipher.h> + +struct crypto_kw_ctx { + struct crypto_cipher *child; +}; + +struct crypto_kw_block { +#define SEMIBSIZE 8 + u8 A[SEMIBSIZE]; + u8 R[SEMIBSIZE]; +}; + +/* convert 64 bit integer into its string representation */ +static inline void crypto_kw_cpu_to_be64(u64 val, u8 *buf) +{ + __be64 *a = (__be64 *)buf; + + *a = cpu_to_be64(val); +} + +/* + * Fast forward the SGL to the "end" length minus SEMIBSIZE. + * The start in the SGL defined by the fast-forward is returned with + * the walk variable + */ +static void crypto_kw_scatterlist_ff(struct scatter_walk *walk, + struct scatterlist *sg, + unsigned int end) +{ + unsigned int skip = 0; + + /* The caller should only operate on full SEMIBLOCKs. */ + BUG_ON(end < SEMIBSIZE); + + skip = end - SEMIBSIZE; + while (sg) { + if (sg->length > skip) { + scatterwalk_start(walk, sg); + scatterwalk_advance(walk, skip); + break; + } else + skip -= sg->length; + + sg = sg_next(sg); + } +} + +static int crypto_kw_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_blkcipher *tfm = desc->tfm; + struct crypto_kw_ctx *ctx = crypto_blkcipher_ctx(tfm); + struct crypto_cipher *child = ctx->child; + + unsigned long alignmask = max_t(unsigned long, SEMIBSIZE, + crypto_cipher_alignmask(child)); + unsigned int i; + + u8 blockbuf[sizeof(struct crypto_kw_block) + alignmask]; + struct crypto_kw_block *block = (struct crypto_kw_block *) + PTR_ALIGN(blockbuf + 0, alignmask + 1); + + u64 t = 6 * ((nbytes) >> 3); + struct scatterlist *lsrc, *ldst; + int ret = 0; + + /* + * Require at least 2 semiblocks (note, the 3rd semiblock that is + * required by SP800-38F is the IV. + */ + if (nbytes < (2 * SEMIBSIZE) || nbytes % SEMIBSIZE) + return -EINVAL; + + /* Place the IV into block A */ + memcpy(block->A, desc->info, SEMIBSIZE); + + /* + * src scatterlist is read-only. dst scatterlist is r/w. During the + * first loop, lsrc points to src and ldst to dst. For any + * subsequent round, the code operates on dst only. + */ + lsrc = src; + ldst = dst; + + for (i = 0; i < 6; i++) { + u8 tbe_buffer[SEMIBSIZE + alignmask]; + /* alignment for the crypto_xor and the _to_be64 operation */ + u8 *tbe = PTR_ALIGN(tbe_buffer + 0, alignmask + 1); + unsigned int tmp_nbytes = nbytes; + struct scatter_walk src_walk, dst_walk; + + while (tmp_nbytes) { + /* move pointer by tmp_nbytes in the SGL */ + crypto_kw_scatterlist_ff(&src_walk, lsrc, tmp_nbytes); + /* get the source block */ + scatterwalk_copychunks(block->R, &src_walk, SEMIBSIZE, + false); + + /* perform KW operation: get counter as byte string */ + crypto_kw_cpu_to_be64(t, tbe); + /* perform KW operation: modify IV with counter */ + crypto_xor(block->A, tbe, SEMIBSIZE); + t--; + /* perform KW operation: decrypt block */ + crypto_cipher_decrypt_one(child, (u8*)block, + (u8*)block); + + /* move pointer by tmp_nbytes in the SGL */ + crypto_kw_scatterlist_ff(&dst_walk, ldst, tmp_nbytes); + /* Copy block->R into place */ + scatterwalk_copychunks(block->R, &dst_walk, SEMIBSIZE, + true); + + tmp_nbytes -= SEMIBSIZE; + } + + /* we now start to operate on the dst SGL only */ + lsrc = dst; + ldst = dst; + } + + /* Perform authentication check */ + if (crypto_memneq("\xA6\xA6\xA6\xA6\xA6\xA6\xA6\xA6", block->A, + SEMIBSIZE)) + ret = -EBADMSG; + + memzero_explicit(&block, sizeof(struct crypto_kw_block)); + + return ret; +} + +static int crypto_kw_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_blkcipher *tfm = desc->tfm; + struct crypto_kw_ctx *ctx = crypto_blkcipher_ctx(tfm); + struct crypto_cipher *child = ctx->child; + + unsigned long alignmask = max_t(unsigned long, SEMIBSIZE, + crypto_cipher_alignmask(child)); + unsigned int i; + + u8 blockbuf[sizeof(struct crypto_kw_block) + alignmask]; + struct crypto_kw_block *block = (struct crypto_kw_block *) + PTR_ALIGN(blockbuf + 0, alignmask + 1); + + u64 t = 1; + struct scatterlist *lsrc, *ldst; + + /* + * Require at least 2 semiblocks (note, the 3rd semiblock that is + * required by SP800-38F is the IV that occupies the first semiblock. + * This means that the dst memory must be one semiblock larger than src. + * Also ensure that the given data is aligned to semiblock. + */ + if (nbytes < (2 * SEMIBSIZE) || nbytes % SEMIBSIZE) + return -EINVAL; + + /* + * Place the predefined IV into block A -- for encrypt, the caller + * does not need to provide an IV, but he needs to fetch the final IV. + */ + memcpy(block->A, "\xA6\xA6\xA6\xA6\xA6\xA6\xA6\xA6", SEMIBSIZE); + + /* + * src scatterlist is read-only. dst scatterlist is r/w. During the + * first loop, lsrc points to src and ldst to dst. For any + * subsequent round, the code operates on dst only. + */ + lsrc = src; + ldst = dst; + + for (i = 0; i < 6; i++) { + u8 tbe_buffer[SEMIBSIZE + alignmask]; + u8 *tbe = PTR_ALIGN(tbe_buffer + 0, alignmask + 1); + unsigned int tmp_nbytes = nbytes; + struct scatter_walk src_walk, dst_walk; + + scatterwalk_start(&src_walk, lsrc); + scatterwalk_start(&dst_walk, ldst); + + while (tmp_nbytes) { + /* get the source block */ + scatterwalk_copychunks(block->R, &src_walk, SEMIBSIZE, + false); + + /* perform KW operation: encrypt block */ + crypto_cipher_encrypt_one(child, (u8 *)block, + (u8 *)block); + /* perform KW operation: get counter as byte string */ + crypto_kw_cpu_to_be64(t, tbe); + /* perform KW operation: modify IV with counter */ + crypto_xor(block->A, tbe, SEMIBSIZE); + t++; + + /* Copy block->R into place */ + scatterwalk_copychunks(block->R, &dst_walk, SEMIBSIZE, + true); + + tmp_nbytes -= SEMIBSIZE; + } + + /* we now start to operate on the dst SGL only */ + lsrc = dst; + ldst = dst; + } + + /* establish the IV for the caller to pick up */ + memcpy(desc->info, block->A, SEMIBSIZE); + + memzero_explicit(&block, sizeof(struct crypto_kw_block)); + + return 0; +} + +static int crypto_kw_setkey(struct crypto_tfm *parent, const u8 *key, + unsigned int keylen) +{ + struct crypto_kw_ctx *ctx = crypto_tfm_ctx(parent); + struct crypto_cipher *child = ctx->child; + int err; + + crypto_cipher_clear_flags(child, CRYPTO_TFM_REQ_MASK); + crypto_cipher_set_flags(child, crypto_tfm_get_flags(parent) & + CRYPTO_TFM_REQ_MASK); + err = crypto_cipher_setkey(child, key, keylen); + crypto_tfm_set_flags(parent, crypto_cipher_get_flags(child) & + CRYPTO_TFM_RES_MASK); + return err; +} + +static int crypto_kw_init_tfm(struct crypto_tfm *tfm) +{ + struct crypto_instance *inst = crypto_tfm_alg_instance(tfm); + struct crypto_spawn *spawn = crypto_instance_ctx(inst); + struct crypto_kw_ctx *ctx = crypto_tfm_ctx(tfm); + struct crypto_cipher *cipher; + + cipher = crypto_spawn_cipher(spawn); + if (IS_ERR(cipher)) + return PTR_ERR(cipher); + + ctx->child = cipher; + return 0; +} + +static void crypto_kw_exit_tfm(struct crypto_tfm *tfm) +{ + struct crypto_kw_ctx *ctx = crypto_tfm_ctx(tfm); + + crypto_free_cipher(ctx->child); +} + +static struct crypto_instance *crypto_kw_alloc(struct rtattr **tb) +{ + struct crypto_instance *inst = NULL; + struct crypto_alg *alg = NULL; + int err; + + err = crypto_check_attr_type(tb, CRYPTO_ALG_TYPE_BLKCIPHER); + if (err) + return ERR_PTR(err); + + alg = crypto_get_attr_alg(tb, CRYPTO_ALG_TYPE_CIPHER, + CRYPTO_ALG_TYPE_MASK); + if (IS_ERR(alg)) + return ERR_CAST(alg); + + inst = ERR_PTR(-EINVAL); + /* Section 5.1 requirement for KW */ + if (alg->cra_blocksize != sizeof(struct crypto_kw_block)) + goto err; + + inst = crypto_alloc_instance("kw", alg); + if (IS_ERR(inst)) + goto err; + + inst->alg.cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER; + inst->alg.cra_priority = alg->cra_priority; + inst->alg.cra_blocksize = SEMIBSIZE; + inst->alg.cra_alignmask = 0; + inst->alg.cra_type = &crypto_blkcipher_type; + inst->alg.cra_blkcipher.ivsize = SEMIBSIZE; + inst->alg.cra_blkcipher.min_keysize = alg->cra_cipher.cia_min_keysize; + inst->alg.cra_blkcipher.max_keysize = alg->cra_cipher.cia_max_keysize; + + inst->alg.cra_ctxsize = sizeof(struct crypto_kw_ctx); + + inst->alg.cra_init = crypto_kw_init_tfm; + inst->alg.cra_exit = crypto_kw_exit_tfm; + + inst->alg.cra_blkcipher.setkey = crypto_kw_setkey; + inst->alg.cra_blkcipher.encrypt = crypto_kw_encrypt; + inst->alg.cra_blkcipher.decrypt = crypto_kw_decrypt; + +err: + crypto_mod_put(alg); + return inst; +} + +static void crypto_kw_free(struct crypto_instance *inst) +{ + crypto_drop_spawn(crypto_instance_ctx(inst)); + kfree(inst); +} + +static struct crypto_template crypto_kw_tmpl = { + .name = "kw", + .alloc = crypto_kw_alloc, + .free = crypto_kw_free, + .module = THIS_MODULE, +}; + +static int __init crypto_kw_init(void) +{ + return crypto_register_template(&crypto_kw_tmpl); +} + +static void __exit crypto_kw_exit(void) +{ + crypto_unregister_template(&crypto_kw_tmpl); +} + +module_init(crypto_kw_init); +module_exit(crypto_kw_exit); + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_AUTHOR("Stephan Mueller <smueller@chronox.de>"); +MODULE_DESCRIPTION("Key Wrapping (RFC3394 / NIST SP800-38F)"); +MODULE_ALIAS_CRYPTO("kw"); diff --git a/crypto/rsa.c b/crypto/rsa.c index 466003e1a8cf..1093e041db03 100644 --- a/crypto/rsa.c +++ b/crypto/rsa.c @@ -97,24 +97,21 @@ static int rsa_enc(struct akcipher_request *req) goto err_free_c; } - m = mpi_read_raw_data(req->src, req->src_len); - if (!m) { - ret = -ENOMEM; + ret = -ENOMEM; + m = mpi_read_raw_from_sgl(req->src, req->src_len); + if (!m) goto err_free_c; - } ret = _rsa_enc(pkey, c, m); if (ret) goto err_free_m; - ret = mpi_read_buffer(c, req->dst, req->dst_len, &req->dst_len, &sign); + ret = mpi_write_to_sgl(c, req->dst, &req->dst_len, &sign); if (ret) goto err_free_m; - if (sign < 0) { + if (sign < 0) ret = -EBADMSG; - goto err_free_m; - } err_free_m: mpi_free(m); @@ -145,25 +142,21 @@ static int rsa_dec(struct akcipher_request *req) goto err_free_m; } - c = mpi_read_raw_data(req->src, req->src_len); - if (!c) { - ret = -ENOMEM; + ret = -ENOMEM; + c = mpi_read_raw_from_sgl(req->src, req->src_len); + if (!c) goto err_free_m; - } ret = _rsa_dec(pkey, m, c); if (ret) goto err_free_c; - ret = mpi_read_buffer(m, req->dst, req->dst_len, &req->dst_len, &sign); + ret = mpi_write_to_sgl(m, req->dst, &req->dst_len, &sign); if (ret) goto err_free_c; - if (sign < 0) { + if (sign < 0) ret = -EBADMSG; - goto err_free_c; - } - err_free_c: mpi_free(c); err_free_m: @@ -193,24 +186,21 @@ static int rsa_sign(struct akcipher_request *req) goto err_free_s; } - m = mpi_read_raw_data(req->src, req->src_len); - if (!m) { - ret = -ENOMEM; + ret = -ENOMEM; + m = mpi_read_raw_from_sgl(req->src, req->src_len); + if (!m) goto err_free_s; - } ret = _rsa_sign(pkey, s, m); if (ret) goto err_free_m; - ret = mpi_read_buffer(s, req->dst, req->dst_len, &req->dst_len, &sign); + ret = mpi_write_to_sgl(s, req->dst, &req->dst_len, &sign); if (ret) goto err_free_m; - if (sign < 0) { + if (sign < 0) ret = -EBADMSG; - goto err_free_m; - } err_free_m: mpi_free(m); @@ -241,7 +231,8 @@ static int rsa_verify(struct akcipher_request *req) goto err_free_m; } - s = mpi_read_raw_data(req->src, req->src_len); + ret = -ENOMEM; + s = mpi_read_raw_from_sgl(req->src, req->src_len); if (!s) { ret = -ENOMEM; goto err_free_m; @@ -251,14 +242,12 @@ static int rsa_verify(struct akcipher_request *req) if (ret) goto err_free_s; - ret = mpi_read_buffer(m, req->dst, req->dst_len, &req->dst_len, &sign); + ret = mpi_write_to_sgl(m, req->dst, &req->dst_len, &sign); if (ret) goto err_free_s; - if (sign < 0) { + if (sign < 0) ret = -EBADMSG; - goto err_free_s; - } err_free_s: mpi_free(s); @@ -282,13 +271,13 @@ static int rsa_check_key_length(unsigned int len) return -EINVAL; } -static int rsa_setkey(struct crypto_akcipher *tfm, const void *key, - unsigned int keylen) +static int rsa_set_pub_key(struct crypto_akcipher *tfm, const void *key, + unsigned int keylen) { struct rsa_key *pkey = akcipher_tfm_ctx(tfm); int ret; - ret = rsa_parse_key(pkey, key, keylen); + ret = rsa_parse_pub_key(pkey, key, keylen); if (ret) return ret; @@ -299,6 +288,30 @@ static int rsa_setkey(struct crypto_akcipher *tfm, const void *key, return ret; } +static int rsa_set_priv_key(struct crypto_akcipher *tfm, const void *key, + unsigned int keylen) +{ + struct rsa_key *pkey = akcipher_tfm_ctx(tfm); + int ret; + + ret = rsa_parse_priv_key(pkey, key, keylen); + if (ret) + return ret; + + if (rsa_check_key_length(mpi_get_size(pkey->n) << 3)) { + rsa_free_key(pkey); + ret = -EINVAL; + } + return ret; +} + +static int rsa_max_size(struct crypto_akcipher *tfm) +{ + struct rsa_key *pkey = akcipher_tfm_ctx(tfm); + + return pkey->n ? mpi_get_size(pkey->n) : -EINVAL; +} + static void rsa_exit_tfm(struct crypto_akcipher *tfm) { struct rsa_key *pkey = akcipher_tfm_ctx(tfm); @@ -311,7 +324,9 @@ static struct akcipher_alg rsa = { .decrypt = rsa_dec, .sign = rsa_sign, .verify = rsa_verify, - .setkey = rsa_setkey, + .set_priv_key = rsa_set_priv_key, + .set_pub_key = rsa_set_pub_key, + .max_size = rsa_max_size, .exit = rsa_exit_tfm, .base = { .cra_name = "rsa", diff --git a/crypto/rsa_helper.c b/crypto/rsa_helper.c index 8d96ce969b44..d226f48d0907 100644 --- a/crypto/rsa_helper.c +++ b/crypto/rsa_helper.c @@ -15,7 +15,8 @@ #include <linux/err.h> #include <linux/fips.h> #include <crypto/internal/rsa.h> -#include "rsakey-asn1.h" +#include "rsapubkey-asn1.h" +#include "rsaprivkey-asn1.h" int rsa_get_n(void *context, size_t hdrlen, unsigned char tag, const void *value, size_t vlen) @@ -94,8 +95,8 @@ void rsa_free_key(struct rsa_key *key) EXPORT_SYMBOL_GPL(rsa_free_key); /** - * rsa_parse_key() - extracts an rsa key from BER encoded buffer - * and stores it in the provided struct rsa_key + * rsa_parse_pub_key() - extracts an rsa public key from BER encoded buffer + * and stores it in the provided struct rsa_key * * @rsa_key: struct rsa_key key representation * @key: key in BER format @@ -103,13 +104,13 @@ EXPORT_SYMBOL_GPL(rsa_free_key); * * Return: 0 on success or error code in case of error */ -int rsa_parse_key(struct rsa_key *rsa_key, const void *key, - unsigned int key_len) +int rsa_parse_pub_key(struct rsa_key *rsa_key, const void *key, + unsigned int key_len) { int ret; free_mpis(rsa_key); - ret = asn1_ber_decoder(&rsakey_decoder, rsa_key, key, key_len); + ret = asn1_ber_decoder(&rsapubkey_decoder, rsa_key, key, key_len); if (ret < 0) goto error; @@ -118,4 +119,31 @@ error: free_mpis(rsa_key); return ret; } -EXPORT_SYMBOL_GPL(rsa_parse_key); +EXPORT_SYMBOL_GPL(rsa_parse_pub_key); + +/** + * rsa_parse_pub_key() - extracts an rsa private key from BER encoded buffer + * and stores it in the provided struct rsa_key + * + * @rsa_key: struct rsa_key key representation + * @key: key in BER format + * @key_len: length of key + * + * Return: 0 on success or error code in case of error + */ +int rsa_parse_priv_key(struct rsa_key *rsa_key, const void *key, + unsigned int key_len) +{ + int ret; + + free_mpis(rsa_key); + ret = asn1_ber_decoder(&rsaprivkey_decoder, rsa_key, key, key_len); + if (ret < 0) + goto error; + + return 0; +error: + free_mpis(rsa_key); + return ret; +} +EXPORT_SYMBOL_GPL(rsa_parse_priv_key); diff --git a/crypto/rsakey.asn1 b/crypto/rsakey.asn1 deleted file mode 100644 index 3c7b5df7b428..000000000000 --- a/crypto/rsakey.asn1 +++ /dev/null @@ -1,5 +0,0 @@ -RsaKey ::= SEQUENCE { - n INTEGER ({ rsa_get_n }), - e INTEGER ({ rsa_get_e }), - d INTEGER ({ rsa_get_d }) -} diff --git a/crypto/rsaprivkey.asn1 b/crypto/rsaprivkey.asn1 new file mode 100644 index 000000000000..731aea5edb0c --- /dev/null +++ b/crypto/rsaprivkey.asn1 @@ -0,0 +1,11 @@ +RsaPrivKey ::= SEQUENCE { + version INTEGER, + n INTEGER ({ rsa_get_n }), + e INTEGER ({ rsa_get_e }), + d INTEGER ({ rsa_get_d }), + prime1 INTEGER, + prime2 INTEGER, + exponent1 INTEGER, + exponent2 INTEGER, + coefficient INTEGER +} diff --git a/crypto/rsapubkey.asn1 b/crypto/rsapubkey.asn1 new file mode 100644 index 000000000000..725498e461d2 --- /dev/null +++ b/crypto/rsapubkey.asn1 @@ -0,0 +1,4 @@ +RsaPubKey ::= SEQUENCE { + n INTEGER ({ rsa_get_n }), + e INTEGER ({ rsa_get_e }) +} diff --git a/crypto/skcipher.c b/crypto/skcipher.c index dd5fc1bf6447..7591928be7ca 100644 --- a/crypto/skcipher.c +++ b/crypto/skcipher.c @@ -91,7 +91,7 @@ static void crypto_exit_skcipher_ops_blkcipher(struct crypto_tfm *tfm) crypto_free_blkcipher(*ctx); } -int crypto_init_skcipher_ops_blkcipher(struct crypto_tfm *tfm) +static int crypto_init_skcipher_ops_blkcipher(struct crypto_tfm *tfm) { struct crypto_alg *calg = tfm->__crt_alg; struct crypto_skcipher *skcipher = __crypto_skcipher_cast(tfm); @@ -182,7 +182,7 @@ static void crypto_exit_skcipher_ops_ablkcipher(struct crypto_tfm *tfm) crypto_free_ablkcipher(*ctx); } -int crypto_init_skcipher_ops_ablkcipher(struct crypto_tfm *tfm) +static int crypto_init_skcipher_ops_ablkcipher(struct crypto_tfm *tfm) { struct crypto_alg *calg = tfm->__crt_alg; struct crypto_skcipher *skcipher = __crypto_skcipher_cast(tfm); diff --git a/crypto/tcrypt.c b/crypto/tcrypt.c index 2b00b617daab..46a4a757d478 100644 --- a/crypto/tcrypt.c +++ b/crypto/tcrypt.c @@ -48,6 +48,8 @@ #define ENCRYPT 1 #define DECRYPT 0 +#define MAX_DIGEST_SIZE 64 + /* * return a string with the driver name */ @@ -950,7 +952,7 @@ static void test_ahash_speed(const char *algo, unsigned int secs, struct tcrypt_result tresult; struct ahash_request *req; struct crypto_ahash *tfm; - static char output[1024]; + char *output; int i, ret; tfm = crypto_alloc_ahash(algo, 0, 0); @@ -963,9 +965,9 @@ static void test_ahash_speed(const char *algo, unsigned int secs, printk(KERN_INFO "\ntesting speed of async %s (%s)\n", algo, get_driver_name(crypto_ahash, tfm)); - if (crypto_ahash_digestsize(tfm) > sizeof(output)) { - pr_err("digestsize(%u) > outputbuffer(%zu)\n", - crypto_ahash_digestsize(tfm), sizeof(output)); + if (crypto_ahash_digestsize(tfm) > MAX_DIGEST_SIZE) { + pr_err("digestsize(%u) > %d\n", crypto_ahash_digestsize(tfm), + MAX_DIGEST_SIZE); goto out; } @@ -980,6 +982,10 @@ static void test_ahash_speed(const char *algo, unsigned int secs, ahash_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG, tcrypt_complete, &tresult); + output = kmalloc(MAX_DIGEST_SIZE, GFP_KERNEL); + if (!output) + goto out_nomem; + for (i = 0; speed[i].blen != 0; i++) { if (speed[i].blen > TVMEMSIZE * PAGE_SIZE) { pr_err("template (%u) too big for tvmem (%lu)\n", @@ -1006,6 +1012,9 @@ static void test_ahash_speed(const char *algo, unsigned int secs, } } + kfree(output); + +out_nomem: ahash_request_free(req); out: diff --git a/crypto/testmgr.c b/crypto/testmgr.c index fa18753f5c34..ae8c57fd8bc7 100644 --- a/crypto/testmgr.c +++ b/crypto/testmgr.c @@ -1034,12 +1034,22 @@ static int __test_skcipher(struct crypto_skcipher *tfm, int enc, q = data; if (memcmp(q, template[i].result, template[i].rlen)) { - pr_err("alg: skcipher%s: Test %d failed on %s for %s\n", + pr_err("alg: skcipher%s: Test %d failed (invalid result) on %s for %s\n", d, j, e, algo); hexdump(q, template[i].rlen); ret = -EINVAL; goto out; } + + if (template[i].iv_out && + memcmp(iv, template[i].iv_out, + crypto_skcipher_ivsize(tfm))) { + pr_err("alg: skcipher%s: Test %d failed (invalid output IV) on %s for %s\n", + d, j, e, algo); + hexdump(iv, crypto_skcipher_ivsize(tfm)); + ret = -EINVAL; + goto out; + } } j = 0; @@ -1845,34 +1855,34 @@ static int do_test_rsa(struct crypto_akcipher *tfm, struct tcrypt_result result; unsigned int out_len_max, out_len = 0; int err = -ENOMEM; + struct scatterlist src, dst, src_tab[2]; req = akcipher_request_alloc(tfm, GFP_KERNEL); if (!req) return err; init_completion(&result.completion); - err = crypto_akcipher_setkey(tfm, vecs->key, vecs->key_len); - if (err) - goto free_req; - akcipher_request_set_crypt(req, vecs->m, outbuf_enc, vecs->m_size, - out_len); - /* expect this to fail, and update the required buf len */ - crypto_akcipher_encrypt(req); - out_len = req->dst_len; - if (!out_len) { - err = -EINVAL; + if (vecs->public_key_vec) + err = crypto_akcipher_set_pub_key(tfm, vecs->key, + vecs->key_len); + else + err = crypto_akcipher_set_priv_key(tfm, vecs->key, + vecs->key_len); + if (err) goto free_req; - } - out_len_max = out_len; - err = -ENOMEM; + out_len_max = crypto_akcipher_maxsize(tfm); outbuf_enc = kzalloc(out_len_max, GFP_KERNEL); if (!outbuf_enc) goto free_req; - akcipher_request_set_crypt(req, vecs->m, outbuf_enc, vecs->m_size, - out_len); + sg_init_table(src_tab, 2); + sg_set_buf(&src_tab[0], vecs->m, 8); + sg_set_buf(&src_tab[1], vecs->m + 8, vecs->m_size - 8); + sg_init_one(&dst, outbuf_enc, out_len_max); + akcipher_request_set_crypt(req, src_tab, &dst, vecs->m_size, + out_len_max); akcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG, tcrypt_complete, &result); @@ -1882,13 +1892,13 @@ static int do_test_rsa(struct crypto_akcipher *tfm, pr_err("alg: rsa: encrypt test failed. err %d\n", err); goto free_all; } - if (out_len != vecs->c_size) { + if (req->dst_len != vecs->c_size) { pr_err("alg: rsa: encrypt test failed. Invalid output len\n"); err = -EINVAL; goto free_all; } /* verify that encrypted message is equal to expected */ - if (memcmp(vecs->c, outbuf_enc, vecs->c_size)) { + if (memcmp(vecs->c, sg_virt(req->dst), vecs->c_size)) { pr_err("alg: rsa: encrypt test failed. Invalid output\n"); err = -EINVAL; goto free_all; @@ -1903,9 +1913,10 @@ static int do_test_rsa(struct crypto_akcipher *tfm, err = -ENOMEM; goto free_all; } + sg_init_one(&src, vecs->c, vecs->c_size); + sg_init_one(&dst, outbuf_dec, out_len_max); init_completion(&result.completion); - akcipher_request_set_crypt(req, outbuf_enc, outbuf_dec, vecs->c_size, - out_len); + akcipher_request_set_crypt(req, &src, &dst, vecs->c_size, out_len_max); /* Run RSA decrypt - m = c^d mod n;*/ err = wait_async_op(&result, crypto_akcipher_decrypt(req)); @@ -2080,7 +2091,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(md5),ecb(cipher_null))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2096,7 +2106,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha1),cbc(aes))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2110,7 +2119,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha1),cbc(des))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2124,7 +2132,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha1),cbc(des3_ede))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2138,7 +2145,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha1),ecb(cipher_null))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2158,7 +2164,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha224),cbc(des))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2172,7 +2177,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha224),cbc(des3_ede))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2186,7 +2190,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha256),cbc(aes))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2200,7 +2203,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha256),cbc(des))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2214,7 +2216,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha256),cbc(des3_ede))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2228,7 +2229,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha384),cbc(des))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2242,7 +2242,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha384),cbc(des3_ede))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2256,7 +2255,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha512),cbc(aes))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2270,7 +2268,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha512),cbc(des))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -2284,7 +2281,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha512),cbc(des3_ede))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = { .enc = { @@ -3011,7 +3007,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "ecb(des)", .test = alg_test_skcipher, - .fips_allowed = 1, .suite = { .cipher = { .enc = { @@ -3292,6 +3287,22 @@ static const struct alg_test_desc alg_test_descs[] = { .fips_allowed = 1, .test = alg_test_null, }, { + .alg = "kw(aes)", + .test = alg_test_skcipher, + .fips_allowed = 1, + .suite = { + .cipher = { + .enc = { + .vecs = aes_kw_enc_tv_template, + .count = ARRAY_SIZE(aes_kw_enc_tv_template) + }, + .dec = { + .vecs = aes_kw_dec_tv_template, + .count = ARRAY_SIZE(aes_kw_dec_tv_template) + } + } + } + }, { .alg = "lrw(aes)", .test = alg_test_skcipher, .suite = { diff --git a/crypto/testmgr.h b/crypto/testmgr.h index 64b8a8082645..da0a8fd765f4 100644 --- a/crypto/testmgr.h +++ b/crypto/testmgr.h @@ -67,6 +67,7 @@ struct hash_testvec { struct cipher_testvec { char *key; char *iv; + char *iv_out; char *input; char *result; unsigned short tap[MAX_TAP]; @@ -149,7 +150,8 @@ static struct akcipher_testvec rsa_tv_template[] = { { #ifndef CONFIG_CRYPTO_FIPS .key = - "\x30\x81\x88" /* sequence of 136 bytes */ + "\x30\x81\x9A" /* sequence of 154 bytes */ + "\x02\x01\x01" /* version - integer of 1 byte */ "\x02\x41" /* modulus - integer of 65 bytes */ "\x00\xAA\x36\xAB\xCE\x88\xAC\xFD\xFF\x55\x52\x3C\x7F\xC4\x52\x3F" "\x90\xEF\xA0\x0D\xF3\x77\x4A\x25\x9F\x2E\x62\xB4\xC5\xD9\x9C\xB5" @@ -161,19 +163,25 @@ static struct akcipher_testvec rsa_tv_template[] = { "\x0A\x03\x37\x48\x62\x64\x87\x69\x5F\x5F\x30\xBC\x38\xB9\x8B\x44" "\xC2\xCD\x2D\xFF\x43\x40\x98\xCD\x20\xD8\xA1\x38\xD0\x90\xBF\x64" "\x79\x7C\x3F\xA7\xA2\xCD\xCB\x3C\xD1\xE0\xBD\xBA\x26\x54\xB4\xF9" - "\xDF\x8E\x8A\xE5\x9D\x73\x3D\x9F\x33\xB3\x01\x62\x4A\xFD\x1D\x51", + "\xDF\x8E\x8A\xE5\x9D\x73\x3D\x9F\x33\xB3\x01\x62\x4A\xFD\x1D\x51" + "\x02\x01\x00" /* prime1 - integer of 1 byte */ + "\x02\x01\x00" /* prime2 - integer of 1 byte */ + "\x02\x01\x00" /* exponent1 - integer of 1 byte */ + "\x02\x01\x00" /* exponent2 - integer of 1 byte */ + "\x02\x01\x00", /* coefficient - integer of 1 byte */ .m = "\x54\x85\x9b\x34\x2c\x49\xea\x2a", .c = "\x63\x1c\xcd\x7b\xe1\x7e\xe4\xde\xc9\xa8\x89\xa1\x74\xcb\x3c\x63" "\x7d\x24\xec\x83\xc3\x15\xe4\x7f\x73\x05\x34\xd1\xec\x22\xbb\x8a" "\x5e\x32\x39\x6d\xc1\x1d\x7d\x50\x3b\x9f\x7a\xad\xf0\x2e\x25\x53" "\x9f\x6e\xbd\x4c\x55\x84\x0c\x9b\xcf\x1a\x4b\x51\x1e\x9e\x0c\x06", - .key_len = 139, + .key_len = 157, .m_size = 8, .c_size = 64, }, { .key = - "\x30\x82\x01\x0B" /* sequence of 267 bytes */ + "\x30\x82\x01\x1D" /* sequence of 285 bytes */ + "\x02\x01\x01" /* version - integer of 1 byte */ "\x02\x81\x81" /* modulus - integer of 129 bytes */ "\x00\xBB\xF8\x2F\x09\x06\x82\xCE\x9C\x23\x38\xAC\x2B\x9D\xA8\x71" "\xF7\x36\x8D\x07\xEE\xD4\x10\x43\xA4\x40\xD6\xB6\xF0\x74\x54\xF5" @@ -194,8 +202,13 @@ static struct akcipher_testvec rsa_tv_template[] = { "\x44\xE5\x6A\xAF\x68\xC5\x6C\x09\x2C\xD3\x8D\xC3\xBE\xF5\xD2\x0A" "\x93\x99\x26\xED\x4F\x74\xA1\x3E\xDD\xFB\xE1\xA1\xCE\xCC\x48\x94" "\xAF\x94\x28\xC2\xB7\xB8\x88\x3F\xE4\x46\x3A\x4B\xC8\x5B\x1C\xB3" - "\xC1", - .key_len = 271, + "\xC1" + "\x02\x01\x00" /* prime1 - integer of 1 byte */ + "\x02\x01\x00" /* prime2 - integer of 1 byte */ + "\x02\x01\x00" /* exponent1 - integer of 1 byte */ + "\x02\x01\x00" /* exponent2 - integer of 1 byte */ + "\x02\x01\x00", /* coefficient - integer of 1 byte */ + .key_len = 289, .m = "\x54\x85\x9b\x34\x2c\x49\xea\x2a", .c = "\x74\x1b\x55\xac\x47\xb5\x08\x0a\x6e\x2b\x2d\xf7\x94\xb8\x8a\x95" @@ -211,7 +224,8 @@ static struct akcipher_testvec rsa_tv_template[] = { }, { #endif .key = - "\x30\x82\x02\x0D" /* sequence of 525 bytes */ + "\x30\x82\x02\x1F" /* sequence of 543 bytes */ + "\x02\x01\x01" /* version - integer of 1 byte */ "\x02\x82\x01\x00" /* modulus - integer of 256 bytes */ "\xDB\x10\x1A\xC2\xA3\xF1\xDC\xFF\x13\x6B\xED\x44\xDF\xF0\x02\x6D" "\x13\xC7\x88\xDA\x70\x6B\x54\xF1\xE8\x27\xDC\xC3\x0F\x99\x6A\xFA" @@ -246,8 +260,13 @@ static struct akcipher_testvec rsa_tv_template[] = { "\x77\xAF\x51\x27\x5B\x5E\x69\xB8\x81\xE6\x11\xC5\x43\x23\x81\x04" "\x62\xFF\xE9\x46\xB8\xD8\x44\xDB\xA5\xCC\x31\x54\x34\xCE\x3E\x82" "\xD6\xBF\x7A\x0B\x64\x21\x6D\x88\x7E\x5B\x45\x12\x1E\x63\x8D\x49" - "\xA7\x1D\xD9\x1E\x06\xCD\xE8\xBA\x2C\x8C\x69\x32\xEA\xBE\x60\x71", - .key_len = 529, + "\xA7\x1D\xD9\x1E\x06\xCD\xE8\xBA\x2C\x8C\x69\x32\xEA\xBE\x60\x71" + "\x02\x01\x00" /* prime1 - integer of 1 byte */ + "\x02\x01\x00" /* prime2 - integer of 1 byte */ + "\x02\x01\x00" /* exponent1 - integer of 1 byte */ + "\x02\x01\x00" /* exponent2 - integer of 1 byte */ + "\x02\x01\x00", /* coefficient - integer of 1 byte */ + .key_len = 547, .m = "\x54\x85\x9b\x34\x2c\x49\xea\x2a", .c = "\xb2\x97\x76\xb4\xae\x3e\x38\x3c\x7e\x64\x1f\xcc\xa2\x7f\xf6\xbe" @@ -23814,6 +23833,46 @@ static struct aead_testvec rfc7539esp_dec_tv_template[] = { }; /* + * All key wrapping test vectors taken from + * http://csrc.nist.gov/groups/STM/cavp/documents/mac/kwtestvectors.zip + * + * Note: as documented in keywrap.c, the ivout for encryption is the first + * semiblock of the ciphertext from the test vector. For decryption, iv is + * the first semiblock of the ciphertext. + */ +static struct cipher_testvec aes_kw_enc_tv_template[] = { + { + .key = "\x75\x75\xda\x3a\x93\x60\x7c\xc2" + "\xbf\xd8\xce\xc7\xaa\xdf\xd9\xa6", + .klen = 16, + .input = "\x42\x13\x6d\x3c\x38\x4a\x3e\xea" + "\xc9\x5a\x06\x6f\xd2\x8f\xed\x3f", + .ilen = 16, + .result = "\xf6\x85\x94\x81\x6f\x64\xca\xa3" + "\xf5\x6f\xab\xea\x25\x48\xf5\xfb", + .rlen = 16, + .iv_out = "\x03\x1f\x6b\xd7\xe6\x1e\x64\x3d", + }, +}; + +static struct cipher_testvec aes_kw_dec_tv_template[] = { + { + .key = "\x80\xaa\x99\x73\x27\xa4\x80\x6b" + "\x6a\x7a\x41\xa5\x2b\x86\xc3\x71" + "\x03\x86\xf9\x32\x78\x6e\xf7\x96" + "\x76\xfa\xfb\x90\xb8\x26\x3c\x5f", + .klen = 32, + .input = "\xd3\x3d\x3d\x97\x7b\xf0\xa9\x15" + "\x59\xf9\x9c\x8a\xcd\x29\x3d\x43", + .ilen = 16, + .result = "\x0a\x25\x6b\xa7\x5c\xfa\x03\xaa" + "\xa0\x2b\xa9\x42\x03\xf1\x5b\xaa", + .rlen = 16, + .iv = "\x42\x3c\x96\x0d\x8a\x2a\xc4\xc1", + }, +}; + +/* * ANSI X9.31 Continuous Pseudo-Random Number Generator (AES mode) * test vectors, taken from Appendix B.2.9 and B.2.10: * http://csrc.nist.gov/groups/STM/cavp/documents/rng/RNGVS.pdf diff --git a/drivers/char/hw_random/Kconfig b/drivers/char/hw_random/Kconfig index f48cf11c655e..dbf22719462f 100644 --- a/drivers/char/hw_random/Kconfig +++ b/drivers/char/hw_random/Kconfig @@ -10,7 +10,7 @@ menuconfig HW_RANDOM To compile this driver as a module, choose M here: the module will be called rng-core. This provides a device - that's usually called /dev/hw_random, and which exposes one + that's usually called /dev/hwrng, and which exposes one of possibly several hardware random number generators. These hardware random number generators do not feed directly @@ -346,6 +346,16 @@ config HW_RANDOM_MSM If unsure, say Y. +config HW_RANDOM_ST + tristate "ST Microelectronics HW Random Number Generator support" + depends on HW_RANDOM && ARCH_STI + ---help--- + This driver provides kernel-side support for the Random Number + Generator hardware found on STi series of SoCs. + + To compile this driver as a module, choose M here: the + module will be called st-rng. + config HW_RANDOM_XGENE tristate "APM X-Gene True Random Number Generator (TRNG) support" depends on HW_RANDOM && ARCH_XGENE @@ -359,6 +369,18 @@ config HW_RANDOM_XGENE If unsure, say Y. +config HW_RANDOM_STM32 + tristate "STMicroelectronics STM32 random number generator" + depends on HW_RANDOM && (ARCH_STM32 || COMPILE_TEST) + help + This driver provides kernel-side support for the Random Number + Generator hardware found on STM32 microcontrollers. + + To compile this driver as a module, choose M here: the + module will be called stm32-rng. + + If unsure, say N. + endif # HW_RANDOM config UML_RANDOM diff --git a/drivers/char/hw_random/Makefile b/drivers/char/hw_random/Makefile index 055bb01510ad..5ad397635128 100644 --- a/drivers/char/hw_random/Makefile +++ b/drivers/char/hw_random/Makefile @@ -30,4 +30,6 @@ obj-$(CONFIG_HW_RANDOM_TPM) += tpm-rng.o obj-$(CONFIG_HW_RANDOM_BCM2835) += bcm2835-rng.o obj-$(CONFIG_HW_RANDOM_IPROC_RNG200) += iproc-rng200.o obj-$(CONFIG_HW_RANDOM_MSM) += msm-rng.o +obj-$(CONFIG_HW_RANDOM_ST) += st-rng.o obj-$(CONFIG_HW_RANDOM_XGENE) += xgene-rng.o +obj-$(CONFIG_HW_RANDOM_STM32) += stm32-rng.o diff --git a/drivers/char/hw_random/core.c b/drivers/char/hw_random/core.c index 5643b65cee20..6f497aa1b276 100644 --- a/drivers/char/hw_random/core.c +++ b/drivers/char/hw_random/core.c @@ -323,7 +323,7 @@ static ssize_t hwrng_attr_current_store(struct device *dev, return -ERESTARTSYS; err = -ENODEV; list_for_each_entry(rng, &rng_list, list) { - if (strcmp(rng->name, buf) == 0) { + if (sysfs_streq(rng->name, buf)) { err = 0; if (rng != current_rng) err = set_current_rng(rng); diff --git a/drivers/char/hw_random/exynos-rng.c b/drivers/char/hw_random/exynos-rng.c index dc4701fd814f..30cf4623184f 100644 --- a/drivers/char/hw_random/exynos-rng.c +++ b/drivers/char/hw_random/exynos-rng.c @@ -53,15 +53,11 @@ static void exynos_rng_writel(struct exynos_rng *rng, u32 val, u32 offset) __raw_writel(val, rng->mem + offset); } -static int exynos_init(struct hwrng *rng) +static int exynos_rng_configure(struct exynos_rng *exynos_rng) { - struct exynos_rng *exynos_rng = container_of(rng, - struct exynos_rng, rng); int i; int ret = 0; - pm_runtime_get_sync(exynos_rng->dev); - for (i = 0 ; i < 5 ; i++) exynos_rng_writel(exynos_rng, jiffies, EXYNOS_PRNG_SEED_OFFSET + 4*i); @@ -70,6 +66,17 @@ static int exynos_init(struct hwrng *rng) & SEED_SETTING_DONE)) ret = -EIO; + return ret; +} + +static int exynos_init(struct hwrng *rng) +{ + struct exynos_rng *exynos_rng = container_of(rng, + struct exynos_rng, rng); + int ret = 0; + + pm_runtime_get_sync(exynos_rng->dev); + ret = exynos_rng_configure(exynos_rng); pm_runtime_put_noidle(exynos_rng->dev); return ret; @@ -81,21 +88,24 @@ static int exynos_read(struct hwrng *rng, void *buf, struct exynos_rng *exynos_rng = container_of(rng, struct exynos_rng, rng); u32 *data = buf; + int retry = 100; pm_runtime_get_sync(exynos_rng->dev); exynos_rng_writel(exynos_rng, PRNG_START, 0); while (!(exynos_rng_readl(exynos_rng, - EXYNOS_PRNG_STATUS_OFFSET) & PRNG_DONE)) + EXYNOS_PRNG_STATUS_OFFSET) & PRNG_DONE) && --retry) cpu_relax(); + if (!retry) + return -ETIMEDOUT; exynos_rng_writel(exynos_rng, PRNG_DONE, EXYNOS_PRNG_STATUS_OFFSET); *data = exynos_rng_readl(exynos_rng, EXYNOS_PRNG_OUT1_OFFSET); pm_runtime_mark_last_busy(exynos_rng->dev); - pm_runtime_autosuspend(exynos_rng->dev); + pm_runtime_put_sync_autosuspend(exynos_rng->dev); return 4; } @@ -152,15 +162,45 @@ static int exynos_rng_runtime_resume(struct device *dev) return clk_prepare_enable(exynos_rng->clk); } + +static int exynos_rng_suspend(struct device *dev) +{ + return pm_runtime_force_suspend(dev); +} + +static int exynos_rng_resume(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct exynos_rng *exynos_rng = platform_get_drvdata(pdev); + int ret; + + ret = pm_runtime_force_resume(dev); + if (ret) + return ret; + + return exynos_rng_configure(exynos_rng); +} #endif -static UNIVERSAL_DEV_PM_OPS(exynos_rng_pm_ops, exynos_rng_runtime_suspend, - exynos_rng_runtime_resume, NULL); +static const struct dev_pm_ops exynos_rng_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(exynos_rng_suspend, exynos_rng_resume) + SET_RUNTIME_PM_OPS(exynos_rng_runtime_suspend, + exynos_rng_runtime_resume, NULL) +}; + +static const struct of_device_id exynos_rng_dt_match[] = { + { + .compatible = "samsung,exynos4-rng", + }, + { }, +}; +MODULE_DEVICE_TABLE(of, exynos_rng_dt_match); static struct platform_driver exynos_rng_driver = { .driver = { .name = "exynos-rng", .pm = &exynos_rng_pm_ops, + .of_match_table = exynos_rng_dt_match, }, .probe = exynos_rng_probe, }; diff --git a/drivers/char/hw_random/mxc-rnga.c b/drivers/char/hw_random/mxc-rnga.c index 6cbb72ec6013..467362262651 100644 --- a/drivers/char/hw_random/mxc-rnga.c +++ b/drivers/char/hw_random/mxc-rnga.c @@ -141,12 +141,11 @@ static void mxc_rnga_cleanup(struct hwrng *rng) static int __init mxc_rnga_probe(struct platform_device *pdev) { - int err = -ENODEV; + int err; struct resource *res; struct mxc_rng *mxc_rng; - mxc_rng = devm_kzalloc(&pdev->dev, sizeof(struct mxc_rng), - GFP_KERNEL); + mxc_rng = devm_kzalloc(&pdev->dev, sizeof(*mxc_rng), GFP_KERNEL); if (!mxc_rng) return -ENOMEM; @@ -160,13 +159,12 @@ static int __init mxc_rnga_probe(struct platform_device *pdev) mxc_rng->clk = devm_clk_get(&pdev->dev, NULL); if (IS_ERR(mxc_rng->clk)) { dev_err(&pdev->dev, "Could not get rng_clk!\n"); - err = PTR_ERR(mxc_rng->clk); - goto out; + return PTR_ERR(mxc_rng->clk); } err = clk_prepare_enable(mxc_rng->clk); if (err) - goto out; + return err; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); mxc_rng->mem = devm_ioremap_resource(&pdev->dev, res); @@ -181,14 +179,10 @@ static int __init mxc_rnga_probe(struct platform_device *pdev) goto err_ioremap; } - dev_info(&pdev->dev, "MXC RNGA Registered.\n"); - return 0; err_ioremap: clk_disable_unprepare(mxc_rng->clk); - -out: return err; } diff --git a/drivers/char/hw_random/octeon-rng.c b/drivers/char/hw_random/octeon-rng.c index 6234a4a19b56..8c78aa090492 100644 --- a/drivers/char/hw_random/octeon-rng.c +++ b/drivers/char/hw_random/octeon-rng.c @@ -96,7 +96,7 @@ static int octeon_rng_probe(struct platform_device *pdev) rng->ops = ops; platform_set_drvdata(pdev, &rng->ops); - ret = hwrng_register(&rng->ops); + ret = devm_hwrng_register(&pdev->dev, &rng->ops); if (ret) return -ENOENT; @@ -105,21 +105,11 @@ static int octeon_rng_probe(struct platform_device *pdev) return 0; } -static int octeon_rng_remove(struct platform_device *pdev) -{ - struct hwrng *rng = platform_get_drvdata(pdev); - - hwrng_unregister(rng); - - return 0; -} - static struct platform_driver octeon_rng_driver = { .driver = { .name = "octeon_rng", }, .probe = octeon_rng_probe, - .remove = octeon_rng_remove, }; module_platform_driver(octeon_rng_driver); diff --git a/drivers/char/hw_random/pasemi-rng.c b/drivers/char/hw_random/pasemi-rng.c index 51cb1d5cc489..699b7259f5d7 100644 --- a/drivers/char/hw_random/pasemi-rng.c +++ b/drivers/char/hw_random/pasemi-rng.c @@ -138,6 +138,7 @@ static const struct of_device_id rng_match[] = { { .compatible = "pasemi,pwrficient-rng", }, { }, }; +MODULE_DEVICE_TABLE(of, rng_match); static struct platform_driver rng_driver = { .driver = { diff --git a/drivers/char/hw_random/ppc4xx-rng.c b/drivers/char/hw_random/ppc4xx-rng.c index b2cfda0fa93e..c0db4387d2e2 100644 --- a/drivers/char/hw_random/ppc4xx-rng.c +++ b/drivers/char/hw_random/ppc4xx-rng.c @@ -129,6 +129,7 @@ static const struct of_device_id ppc4xx_rng_match[] = { { .compatible = "amcc,ppc440epx-rng", }, {}, }; +MODULE_DEVICE_TABLE(of, ppc4xx_rng_match); static struct platform_driver ppc4xx_rng_driver = { .driver = { diff --git a/drivers/char/hw_random/st-rng.c b/drivers/char/hw_random/st-rng.c new file mode 100644 index 000000000000..1d35363d23c5 --- /dev/null +++ b/drivers/char/hw_random/st-rng.c @@ -0,0 +1,151 @@ +/* + * ST Random Number Generator Driver ST's Platforms + * + * Author: Pankaj Dev: <pankaj.dev@st.com> + * Lee Jones <lee.jones@linaro.org> + * + * Copyright (C) 2015 STMicroelectronics (R&D) Limited + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/clk.h> +#include <linux/delay.h> +#include <linux/hw_random.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/platform_device.h> +#include <linux/slab.h> + +/* Registers */ +#define ST_RNG_STATUS_REG 0x20 +#define ST_RNG_DATA_REG 0x24 + +/* Registers fields */ +#define ST_RNG_STATUS_BAD_SEQUENCE BIT(0) +#define ST_RNG_STATUS_BAD_ALTERNANCE BIT(1) +#define ST_RNG_STATUS_FIFO_FULL BIT(5) + +#define ST_RNG_SAMPLE_SIZE 2 /* 2 Byte (16bit) samples */ +#define ST_RNG_FIFO_DEPTH 4 +#define ST_RNG_FIFO_SIZE (ST_RNG_FIFO_DEPTH * ST_RNG_SAMPLE_SIZE) + +/* + * Samples are documented to be available every 0.667us, so in theory + * the 4 sample deep FIFO should take 2.668us to fill. However, during + * thorough testing, it became apparent that filling the FIFO actually + * takes closer to 12us. We then multiply by 2 in order to account for + * the lack of udelay()'s reliability, suggested by Russell King. + */ +#define ST_RNG_FILL_FIFO_TIMEOUT (12 * 2) + +struct st_rng_data { + void __iomem *base; + struct clk *clk; + struct hwrng ops; +}; + +static int st_rng_read(struct hwrng *rng, void *data, size_t max, bool wait) +{ + struct st_rng_data *ddata = (struct st_rng_data *)rng->priv; + u32 status; + int i; + + if (max < sizeof(u16)) + return -EINVAL; + + /* Wait until FIFO is full - max 4uS*/ + for (i = 0; i < ST_RNG_FILL_FIFO_TIMEOUT; i++) { + status = readl_relaxed(ddata->base + ST_RNG_STATUS_REG); + if (status & ST_RNG_STATUS_FIFO_FULL) + break; + udelay(1); + } + + if (i == ST_RNG_FILL_FIFO_TIMEOUT) + return 0; + + for (i = 0; i < ST_RNG_FIFO_SIZE && i < max; i += 2) + *(u16 *)(data + i) = + readl_relaxed(ddata->base + ST_RNG_DATA_REG); + + return i; /* No of bytes read */ +} + +static int st_rng_probe(struct platform_device *pdev) +{ + struct st_rng_data *ddata; + struct resource *res; + struct clk *clk; + void __iomem *base; + int ret; + + ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL); + if (!ddata) + return -ENOMEM; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + base = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(base)) + return PTR_ERR(base); + + clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(clk)) + return PTR_ERR(clk); + + ret = clk_prepare_enable(clk); + if (ret) + return ret; + + ddata->ops.priv = (unsigned long)ddata; + ddata->ops.read = st_rng_read; + ddata->ops.name = pdev->name; + ddata->base = base; + ddata->clk = clk; + + dev_set_drvdata(&pdev->dev, ddata); + + ret = hwrng_register(&ddata->ops); + if (ret) { + dev_err(&pdev->dev, "Failed to register HW RNG\n"); + return ret; + } + + dev_info(&pdev->dev, "Successfully registered HW RNG\n"); + + return 0; +} + +static int st_rng_remove(struct platform_device *pdev) +{ + struct st_rng_data *ddata = dev_get_drvdata(&pdev->dev); + + hwrng_unregister(&ddata->ops); + + clk_disable_unprepare(ddata->clk); + + return 0; +} + +static const struct of_device_id st_rng_match[] = { + { .compatible = "st,rng" }, + {}, +}; +MODULE_DEVICE_TABLE(of, st_rng_match); + +static struct platform_driver st_rng_driver = { + .driver = { + .name = "st-hwrandom", + .of_match_table = of_match_ptr(st_rng_match), + }, + .probe = st_rng_probe, + .remove = st_rng_remove +}; + +module_platform_driver(st_rng_driver); + +MODULE_AUTHOR("Pankaj Dev <pankaj.dev@st.com>"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/char/hw_random/stm32-rng.c b/drivers/char/hw_random/stm32-rng.c new file mode 100644 index 000000000000..92a810648bd0 --- /dev/null +++ b/drivers/char/hw_random/stm32-rng.c @@ -0,0 +1,202 @@ +/* + * Copyright (c) 2015, Daniel Thompson + * + * This file 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. + * + * This file is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/clk.h> +#include <linux/delay.h> +#include <linux/hw_random.h> +#include <linux/io.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/of_address.h> +#include <linux/of_platform.h> +#include <linux/pm_runtime.h> +#include <linux/slab.h> + +#define RNG_CR 0x00 +#define RNG_CR_RNGEN BIT(2) + +#define RNG_SR 0x04 +#define RNG_SR_SEIS BIT(6) +#define RNG_SR_CEIS BIT(5) +#define RNG_SR_DRDY BIT(0) + +#define RNG_DR 0x08 + +/* + * It takes 40 cycles @ 48MHz to generate each random number (e.g. <1us). + * At the time of writing STM32 parts max out at ~200MHz meaning a timeout + * of 500 leaves us a very comfortable margin for error. The loop to which + * the timeout applies takes at least 4 instructions per iteration so the + * timeout is enough to take us up to multi-GHz parts! + */ +#define RNG_TIMEOUT 500 + +struct stm32_rng_private { + struct hwrng rng; + void __iomem *base; + struct clk *clk; +}; + +static int stm32_rng_read(struct hwrng *rng, void *data, size_t max, bool wait) +{ + struct stm32_rng_private *priv = + container_of(rng, struct stm32_rng_private, rng); + u32 sr; + int retval = 0; + + pm_runtime_get_sync((struct device *) priv->rng.priv); + + while (max > sizeof(u32)) { + sr = readl_relaxed(priv->base + RNG_SR); + if (!sr && wait) { + unsigned int timeout = RNG_TIMEOUT; + + do { + cpu_relax(); + sr = readl_relaxed(priv->base + RNG_SR); + } while (!sr && --timeout); + } + + /* If error detected or data not ready... */ + if (sr != RNG_SR_DRDY) + break; + + *(u32 *)data = readl_relaxed(priv->base + RNG_DR); + + retval += sizeof(u32); + data += sizeof(u32); + max -= sizeof(u32); + } + + if (WARN_ONCE(sr & (RNG_SR_SEIS | RNG_SR_CEIS), + "bad RNG status - %x\n", sr)) + writel_relaxed(0, priv->base + RNG_SR); + + pm_runtime_mark_last_busy((struct device *) priv->rng.priv); + pm_runtime_put_sync_autosuspend((struct device *) priv->rng.priv); + + return retval || !wait ? retval : -EIO; +} + +static int stm32_rng_init(struct hwrng *rng) +{ + struct stm32_rng_private *priv = + container_of(rng, struct stm32_rng_private, rng); + int err; + + err = clk_prepare_enable(priv->clk); + if (err) + return err; + + writel_relaxed(RNG_CR_RNGEN, priv->base + RNG_CR); + + /* clear error indicators */ + writel_relaxed(0, priv->base + RNG_SR); + + return 0; +} + +static void stm32_rng_cleanup(struct hwrng *rng) +{ + struct stm32_rng_private *priv = + container_of(rng, struct stm32_rng_private, rng); + + writel_relaxed(0, priv->base + RNG_CR); + clk_disable_unprepare(priv->clk); +} + +static int stm32_rng_probe(struct platform_device *ofdev) +{ + struct device *dev = &ofdev->dev; + struct device_node *np = ofdev->dev.of_node; + struct stm32_rng_private *priv; + struct resource res; + int err; + + priv = devm_kzalloc(dev, sizeof(struct stm32_rng_private), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + err = of_address_to_resource(np, 0, &res); + if (err) + return err; + + priv->base = devm_ioremap_resource(dev, &res); + if (IS_ERR(priv->base)) + return PTR_ERR(priv->base); + + priv->clk = devm_clk_get(&ofdev->dev, NULL); + if (IS_ERR(priv->clk)) + return PTR_ERR(priv->clk); + + dev_set_drvdata(dev, priv); + + priv->rng.name = dev_driver_string(dev), +#ifndef CONFIG_PM + priv->rng.init = stm32_rng_init, + priv->rng.cleanup = stm32_rng_cleanup, +#endif + priv->rng.read = stm32_rng_read, + priv->rng.priv = (unsigned long) dev; + + pm_runtime_set_autosuspend_delay(dev, 100); + pm_runtime_use_autosuspend(dev); + pm_runtime_enable(dev); + + return devm_hwrng_register(dev, &priv->rng); +} + +#ifdef CONFIG_PM +static int stm32_rng_runtime_suspend(struct device *dev) +{ + struct stm32_rng_private *priv = dev_get_drvdata(dev); + + stm32_rng_cleanup(&priv->rng); + + return 0; +} + +static int stm32_rng_runtime_resume(struct device *dev) +{ + struct stm32_rng_private *priv = dev_get_drvdata(dev); + + return stm32_rng_init(&priv->rng); +} +#endif + +static UNIVERSAL_DEV_PM_OPS(stm32_rng_pm_ops, stm32_rng_runtime_suspend, + stm32_rng_runtime_resume, NULL); + +static const struct of_device_id stm32_rng_match[] = { + { + .compatible = "st,stm32-rng", + }, + {}, +}; +MODULE_DEVICE_TABLE(of, stm32_rng_match); + +static struct platform_driver stm32_rng_driver = { + .driver = { + .name = "stm32-rng", + .pm = &stm32_rng_pm_ops, + .of_match_table = stm32_rng_match, + }, + .probe = stm32_rng_probe, +}; + +module_platform_driver(stm32_rng_driver); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Daniel Thompson <daniel.thompson@linaro.org>"); +MODULE_DESCRIPTION("STMicroelectronics STM32 RNG device driver"); diff --git a/drivers/crypto/Kconfig b/drivers/crypto/Kconfig index d234719065a5..2569e043317e 100644 --- a/drivers/crypto/Kconfig +++ b/drivers/crypto/Kconfig @@ -420,7 +420,7 @@ config CRYPTO_DEV_CCP bool "Support for AMD Cryptographic Coprocessor" depends on ((X86 && PCI) || (ARM64 && (OF_ADDRESS || ACPI))) && HAS_IOMEM help - The AMD Cryptographic Coprocessor provides hardware support + The AMD Cryptographic Coprocessor provides hardware offload support for encryption, hashing and related operations. if CRYPTO_DEV_CCP @@ -429,7 +429,8 @@ endif config CRYPTO_DEV_MXS_DCP tristate "Support for Freescale MXS DCP" - depends on ARCH_MXS + depends on (ARCH_MXS || ARCH_MXC) + select STMP_DEVICE select CRYPTO_CBC select CRYPTO_ECB select CRYPTO_AES diff --git a/drivers/crypto/amcc/crypto4xx_core.c b/drivers/crypto/amcc/crypto4xx_core.c index 192a8fa325c1..58a630e55d5d 100644 --- a/drivers/crypto/amcc/crypto4xx_core.c +++ b/drivers/crypto/amcc/crypto4xx_core.c @@ -740,26 +740,6 @@ void crypto4xx_return_pd(struct crypto4xx_device *dev, pd_uinfo->state = PD_ENTRY_FREE; } -/* - * derive number of elements in scatterlist - * Shamlessly copy from talitos.c - */ -static int get_sg_count(struct scatterlist *sg_list, int nbytes) -{ - struct scatterlist *sg = sg_list; - int sg_nents = 0; - - while (nbytes) { - sg_nents++; - if (sg->length > nbytes) - break; - nbytes -= sg->length; - sg = sg_next(sg); - } - - return sg_nents; -} - static u32 get_next_gd(u32 current) { if (current != PPC4XX_LAST_GD) @@ -800,7 +780,7 @@ u32 crypto4xx_build_pd(struct crypto_async_request *req, u32 gd_idx = 0; /* figure how many gd is needed */ - num_gd = get_sg_count(src, datalen); + num_gd = sg_nents_for_len(src, datalen); if (num_gd == 1) num_gd = 0; @@ -1284,6 +1264,7 @@ static const struct of_device_id crypto4xx_match[] = { { .compatible = "amcc,ppc4xx-crypto",}, { }, }; +MODULE_DEVICE_TABLE(of, crypto4xx_match); static struct platform_driver crypto4xx_driver = { .driver = { diff --git a/drivers/crypto/atmel-aes.c b/drivers/crypto/atmel-aes.c index 0f9a9dc06a83..fb16d812c8f5 100644 --- a/drivers/crypto/atmel-aes.c +++ b/drivers/crypto/atmel-aes.c @@ -260,7 +260,11 @@ static struct atmel_aes_dev *atmel_aes_find_dev(struct atmel_aes_ctx *ctx) static int atmel_aes_hw_init(struct atmel_aes_dev *dd) { - clk_prepare_enable(dd->iclk); + int err; + + err = clk_prepare_enable(dd->iclk); + if (err) + return err; if (!(dd->flags & AES_FLAGS_INIT)) { atmel_aes_write(dd, AES_CR, AES_CR_SWRST); @@ -1320,7 +1324,6 @@ static int atmel_aes_probe(struct platform_device *pdev) struct crypto_platform_data *pdata; struct device *dev = &pdev->dev; struct resource *aes_res; - unsigned long aes_phys_size; int err; pdata = pdev->dev.platform_data; @@ -1337,7 +1340,7 @@ static int atmel_aes_probe(struct platform_device *pdev) goto aes_dd_err; } - aes_dd = kzalloc(sizeof(struct atmel_aes_dev), GFP_KERNEL); + aes_dd = devm_kzalloc(&pdev->dev, sizeof(*aes_dd), GFP_KERNEL); if (aes_dd == NULL) { dev_err(dev, "unable to alloc data struct.\n"); err = -ENOMEM; @@ -1368,36 +1371,35 @@ static int atmel_aes_probe(struct platform_device *pdev) goto res_err; } aes_dd->phys_base = aes_res->start; - aes_phys_size = resource_size(aes_res); /* Get the IRQ */ aes_dd->irq = platform_get_irq(pdev, 0); if (aes_dd->irq < 0) { dev_err(dev, "no IRQ resource info\n"); err = aes_dd->irq; - goto aes_irq_err; + goto res_err; } - err = request_irq(aes_dd->irq, atmel_aes_irq, IRQF_SHARED, "atmel-aes", - aes_dd); + err = devm_request_irq(&pdev->dev, aes_dd->irq, atmel_aes_irq, + IRQF_SHARED, "atmel-aes", aes_dd); if (err) { dev_err(dev, "unable to request aes irq.\n"); - goto aes_irq_err; + goto res_err; } /* Initializing the clock */ - aes_dd->iclk = clk_get(&pdev->dev, "aes_clk"); + aes_dd->iclk = devm_clk_get(&pdev->dev, "aes_clk"); if (IS_ERR(aes_dd->iclk)) { dev_err(dev, "clock initialization failed.\n"); err = PTR_ERR(aes_dd->iclk); - goto clk_err; + goto res_err; } - aes_dd->io_base = ioremap(aes_dd->phys_base, aes_phys_size); + aes_dd->io_base = devm_ioremap_resource(&pdev->dev, aes_res); if (!aes_dd->io_base) { dev_err(dev, "can't ioremap\n"); err = -ENOMEM; - goto aes_io_err; + goto res_err; } atmel_aes_hw_version_init(aes_dd); @@ -1434,17 +1436,9 @@ err_algs: err_aes_dma: atmel_aes_buff_cleanup(aes_dd); err_aes_buff: - iounmap(aes_dd->io_base); -aes_io_err: - clk_put(aes_dd->iclk); -clk_err: - free_irq(aes_dd->irq, aes_dd); -aes_irq_err: res_err: tasklet_kill(&aes_dd->done_task); tasklet_kill(&aes_dd->queue_task); - kfree(aes_dd); - aes_dd = NULL; aes_dd_err: dev_err(dev, "initialization failed.\n"); @@ -1469,16 +1463,6 @@ static int atmel_aes_remove(struct platform_device *pdev) atmel_aes_dma_cleanup(aes_dd); - iounmap(aes_dd->io_base); - - clk_put(aes_dd->iclk); - - if (aes_dd->irq > 0) - free_irq(aes_dd->irq, aes_dd); - - kfree(aes_dd); - aes_dd = NULL; - return 0; } diff --git a/drivers/crypto/atmel-sha.c b/drivers/crypto/atmel-sha.c index 5b35433c5399..660d8c06540b 100644 --- a/drivers/crypto/atmel-sha.c +++ b/drivers/crypto/atmel-sha.c @@ -794,7 +794,11 @@ static void atmel_sha_finish_req(struct ahash_request *req, int err) static int atmel_sha_hw_init(struct atmel_sha_dev *dd) { - clk_prepare_enable(dd->iclk); + int err; + + err = clk_prepare_enable(dd->iclk); + if (err) + return err; if (!(SHA_FLAGS_INIT & dd->flags)) { atmel_sha_write(dd, SHA_CR, SHA_CR_SWRST); @@ -1345,11 +1349,9 @@ static int atmel_sha_probe(struct platform_device *pdev) struct crypto_platform_data *pdata; struct device *dev = &pdev->dev; struct resource *sha_res; - unsigned long sha_phys_size; int err; - sha_dd = devm_kzalloc(&pdev->dev, sizeof(struct atmel_sha_dev), - GFP_KERNEL); + sha_dd = devm_kzalloc(&pdev->dev, sizeof(*sha_dd), GFP_KERNEL); if (sha_dd == NULL) { dev_err(dev, "unable to alloc data struct.\n"); err = -ENOMEM; @@ -1378,7 +1380,6 @@ static int atmel_sha_probe(struct platform_device *pdev) goto res_err; } sha_dd->phys_base = sha_res->start; - sha_phys_size = resource_size(sha_res); /* Get the IRQ */ sha_dd->irq = platform_get_irq(pdev, 0); @@ -1388,26 +1389,26 @@ static int atmel_sha_probe(struct platform_device *pdev) goto res_err; } - err = request_irq(sha_dd->irq, atmel_sha_irq, IRQF_SHARED, "atmel-sha", - sha_dd); + err = devm_request_irq(&pdev->dev, sha_dd->irq, atmel_sha_irq, + IRQF_SHARED, "atmel-sha", sha_dd); if (err) { dev_err(dev, "unable to request sha irq.\n"); goto res_err; } /* Initializing the clock */ - sha_dd->iclk = clk_get(&pdev->dev, "sha_clk"); + sha_dd->iclk = devm_clk_get(&pdev->dev, "sha_clk"); if (IS_ERR(sha_dd->iclk)) { dev_err(dev, "clock initialization failed.\n"); err = PTR_ERR(sha_dd->iclk); - goto clk_err; + goto res_err; } - sha_dd->io_base = ioremap(sha_dd->phys_base, sha_phys_size); + sha_dd->io_base = devm_ioremap_resource(&pdev->dev, sha_res); if (!sha_dd->io_base) { dev_err(dev, "can't ioremap\n"); err = -ENOMEM; - goto sha_io_err; + goto res_err; } atmel_sha_hw_version_init(sha_dd); @@ -1421,12 +1422,12 @@ static int atmel_sha_probe(struct platform_device *pdev) if (IS_ERR(pdata)) { dev_err(&pdev->dev, "platform data not available\n"); err = PTR_ERR(pdata); - goto err_pdata; + goto res_err; } } if (!pdata->dma_slave) { err = -ENXIO; - goto err_pdata; + goto res_err; } err = atmel_sha_dma_init(sha_dd, pdata); if (err) @@ -1457,12 +1458,6 @@ err_algs: if (sha_dd->caps.has_dma) atmel_sha_dma_cleanup(sha_dd); err_sha_dma: -err_pdata: - iounmap(sha_dd->io_base); -sha_io_err: - clk_put(sha_dd->iclk); -clk_err: - free_irq(sha_dd->irq, sha_dd); res_err: tasklet_kill(&sha_dd->done_task); sha_dd_err: diff --git a/drivers/crypto/atmel-tdes.c b/drivers/crypto/atmel-tdes.c index ca2999709eb4..2c7a628d0375 100644 --- a/drivers/crypto/atmel-tdes.c +++ b/drivers/crypto/atmel-tdes.c @@ -218,7 +218,11 @@ static struct atmel_tdes_dev *atmel_tdes_find_dev(struct atmel_tdes_ctx *ctx) static int atmel_tdes_hw_init(struct atmel_tdes_dev *dd) { - clk_prepare_enable(dd->iclk); + int err; + + err = clk_prepare_enable(dd->iclk); + if (err) + return err; if (!(dd->flags & TDES_FLAGS_INIT)) { atmel_tdes_write(dd, TDES_CR, TDES_CR_SWRST); @@ -1355,7 +1359,6 @@ static int atmel_tdes_probe(struct platform_device *pdev) struct crypto_platform_data *pdata; struct device *dev = &pdev->dev; struct resource *tdes_res; - unsigned long tdes_phys_size; int err; tdes_dd = devm_kmalloc(&pdev->dev, sizeof(*tdes_dd), GFP_KERNEL); @@ -1389,7 +1392,6 @@ static int atmel_tdes_probe(struct platform_device *pdev) goto res_err; } tdes_dd->phys_base = tdes_res->start; - tdes_phys_size = resource_size(tdes_res); /* Get the IRQ */ tdes_dd->irq = platform_get_irq(pdev, 0); @@ -1399,26 +1401,26 @@ static int atmel_tdes_probe(struct platform_device *pdev) goto res_err; } - err = request_irq(tdes_dd->irq, atmel_tdes_irq, IRQF_SHARED, - "atmel-tdes", tdes_dd); + err = devm_request_irq(&pdev->dev, tdes_dd->irq, atmel_tdes_irq, + IRQF_SHARED, "atmel-tdes", tdes_dd); if (err) { dev_err(dev, "unable to request tdes irq.\n"); - goto tdes_irq_err; + goto res_err; } /* Initializing the clock */ - tdes_dd->iclk = clk_get(&pdev->dev, "tdes_clk"); + tdes_dd->iclk = devm_clk_get(&pdev->dev, "tdes_clk"); if (IS_ERR(tdes_dd->iclk)) { dev_err(dev, "clock initialization failed.\n"); err = PTR_ERR(tdes_dd->iclk); - goto clk_err; + goto res_err; } - tdes_dd->io_base = ioremap(tdes_dd->phys_base, tdes_phys_size); + tdes_dd->io_base = devm_ioremap_resource(&pdev->dev, tdes_res); if (!tdes_dd->io_base) { dev_err(dev, "can't ioremap\n"); err = -ENOMEM; - goto tdes_io_err; + goto res_err; } atmel_tdes_hw_version_init(tdes_dd); @@ -1474,12 +1476,6 @@ err_tdes_dma: err_pdata: atmel_tdes_buff_cleanup(tdes_dd); err_tdes_buff: - iounmap(tdes_dd->io_base); -tdes_io_err: - clk_put(tdes_dd->iclk); -clk_err: - free_irq(tdes_dd->irq, tdes_dd); -tdes_irq_err: res_err: tasklet_kill(&tdes_dd->done_task); tasklet_kill(&tdes_dd->queue_task); @@ -1510,13 +1506,6 @@ static int atmel_tdes_remove(struct platform_device *pdev) atmel_tdes_buff_cleanup(tdes_dd); - iounmap(tdes_dd->io_base); - - clk_put(tdes_dd->iclk); - - if (tdes_dd->irq >= 0) - free_irq(tdes_dd->irq, tdes_dd); - return 0; } diff --git a/drivers/crypto/bfin_crc.c b/drivers/crypto/bfin_crc.c index 2f0b3337505d..95b73968cf72 100644 --- a/drivers/crypto/bfin_crc.c +++ b/drivers/crypto/bfin_crc.c @@ -96,26 +96,6 @@ struct bfin_crypto_crc_ctx { u32 key; }; - -/* - * derive number of elements in scatterlist - */ -static int sg_count(struct scatterlist *sg_list) -{ - struct scatterlist *sg = sg_list; - int sg_nents = 1; - - if (sg_list == NULL) - return 0; - - while (!sg_is_last(sg)) { - sg_nents++; - sg = sg_next(sg); - } - - return sg_nents; -} - /* * get element in scatter list by given index */ @@ -160,7 +140,7 @@ static int bfin_crypto_crc_init(struct ahash_request *req) } spin_unlock_bh(&crc_list.lock); - if (sg_count(req->src) > CRC_MAX_DMA_DESC) { + if (sg_nents(req->src) > CRC_MAX_DMA_DESC) { dev_dbg(ctx->crc->dev, "init: requested sg list is too big > %d\n", CRC_MAX_DMA_DESC); return -EINVAL; @@ -376,7 +356,8 @@ static int bfin_crypto_crc_handle_queue(struct bfin_crypto_crc *crc, ctx->sg = req->src; /* Chop crc buffer size to multiple of 32 bit */ - nsg = ctx->sg_nents = sg_count(ctx->sg); + nsg = sg_nents(ctx->sg); + ctx->sg_nents = nsg; ctx->sg_buflen = ctx->buflast_len + req->nbytes; ctx->bufnext_len = ctx->sg_buflen % 4; ctx->sg_buflen &= ~0x3; diff --git a/drivers/crypto/caam/caamalg.c b/drivers/crypto/caam/caamalg.c index ba79d638f782..ea8189f4b021 100644 --- a/drivers/crypto/caam/caamalg.c +++ b/drivers/crypto/caam/caamalg.c @@ -1705,14 +1705,131 @@ static int ablkcipher_setkey(struct crypto_ablkcipher *ablkcipher, return ret; } +static int xts_ablkcipher_setkey(struct crypto_ablkcipher *ablkcipher, + const u8 *key, unsigned int keylen) +{ + struct caam_ctx *ctx = crypto_ablkcipher_ctx(ablkcipher); + struct device *jrdev = ctx->jrdev; + u32 *key_jump_cmd, *desc; + __be64 sector_size = cpu_to_be64(512); + + if (keylen != 2 * AES_MIN_KEY_SIZE && keylen != 2 * AES_MAX_KEY_SIZE) { + crypto_ablkcipher_set_flags(ablkcipher, + CRYPTO_TFM_RES_BAD_KEY_LEN); + dev_err(jrdev, "key size mismatch\n"); + return -EINVAL; + } + + memcpy(ctx->key, key, keylen); + ctx->key_dma = dma_map_single(jrdev, ctx->key, keylen, DMA_TO_DEVICE); + if (dma_mapping_error(jrdev, ctx->key_dma)) { + dev_err(jrdev, "unable to map key i/o memory\n"); + return -ENOMEM; + } + ctx->enckeylen = keylen; + + /* xts_ablkcipher_encrypt shared descriptor */ + desc = ctx->sh_desc_enc; + init_sh_desc(desc, HDR_SHARE_SERIAL | HDR_SAVECTX); + /* Skip if already shared */ + key_jump_cmd = append_jump(desc, JUMP_JSL | JUMP_TEST_ALL | + JUMP_COND_SHRD); + + /* Load class1 keys only */ + append_key_as_imm(desc, (void *)ctx->key, ctx->enckeylen, + ctx->enckeylen, CLASS_1 | KEY_DEST_CLASS_REG); + + /* Load sector size with index 40 bytes (0x28) */ + append_cmd(desc, CMD_LOAD | IMMEDIATE | LDST_SRCDST_BYTE_CONTEXT | + LDST_CLASS_1_CCB | (0x28 << LDST_OFFSET_SHIFT) | 8); + append_data(desc, (void *)§or_size, 8); + + set_jump_tgt_here(desc, key_jump_cmd); + + /* + * create sequence for loading the sector index + * Upper 8B of IV - will be used as sector index + * Lower 8B of IV - will be discarded + */ + append_cmd(desc, CMD_SEQ_LOAD | LDST_SRCDST_BYTE_CONTEXT | + LDST_CLASS_1_CCB | (0x20 << LDST_OFFSET_SHIFT) | 8); + append_seq_fifo_load(desc, 8, FIFOLD_CLASS_SKIP); + + /* Load operation */ + append_operation(desc, ctx->class1_alg_type | OP_ALG_AS_INITFINAL | + OP_ALG_ENCRYPT); + + /* Perform operation */ + ablkcipher_append_src_dst(desc); + + ctx->sh_desc_enc_dma = dma_map_single(jrdev, desc, desc_bytes(desc), + DMA_TO_DEVICE); + if (dma_mapping_error(jrdev, ctx->sh_desc_enc_dma)) { + dev_err(jrdev, "unable to map shared descriptor\n"); + return -ENOMEM; + } +#ifdef DEBUG + print_hex_dump(KERN_ERR, + "xts ablkcipher enc shdesc@" __stringify(__LINE__) ": ", + DUMP_PREFIX_ADDRESS, 16, 4, desc, desc_bytes(desc), 1); +#endif + + /* xts_ablkcipher_decrypt shared descriptor */ + desc = ctx->sh_desc_dec; + + init_sh_desc(desc, HDR_SHARE_SERIAL | HDR_SAVECTX); + /* Skip if already shared */ + key_jump_cmd = append_jump(desc, JUMP_JSL | JUMP_TEST_ALL | + JUMP_COND_SHRD); + + /* Load class1 key only */ + append_key_as_imm(desc, (void *)ctx->key, ctx->enckeylen, + ctx->enckeylen, CLASS_1 | KEY_DEST_CLASS_REG); + + /* Load sector size with index 40 bytes (0x28) */ + append_cmd(desc, CMD_LOAD | IMMEDIATE | LDST_SRCDST_BYTE_CONTEXT | + LDST_CLASS_1_CCB | (0x28 << LDST_OFFSET_SHIFT) | 8); + append_data(desc, (void *)§or_size, 8); + + set_jump_tgt_here(desc, key_jump_cmd); + + /* + * create sequence for loading the sector index + * Upper 8B of IV - will be used as sector index + * Lower 8B of IV - will be discarded + */ + append_cmd(desc, CMD_SEQ_LOAD | LDST_SRCDST_BYTE_CONTEXT | + LDST_CLASS_1_CCB | (0x20 << LDST_OFFSET_SHIFT) | 8); + append_seq_fifo_load(desc, 8, FIFOLD_CLASS_SKIP); + + /* Load operation */ + append_dec_op1(desc, ctx->class1_alg_type); + + /* Perform operation */ + ablkcipher_append_src_dst(desc); + + ctx->sh_desc_dec_dma = dma_map_single(jrdev, desc, desc_bytes(desc), + DMA_TO_DEVICE); + if (dma_mapping_error(jrdev, ctx->sh_desc_dec_dma)) { + dma_unmap_single(jrdev, ctx->sh_desc_enc_dma, + desc_bytes(ctx->sh_desc_enc), DMA_TO_DEVICE); + dev_err(jrdev, "unable to map shared descriptor\n"); + return -ENOMEM; + } +#ifdef DEBUG + print_hex_dump(KERN_ERR, + "xts ablkcipher dec shdesc@" __stringify(__LINE__) ": ", + DUMP_PREFIX_ADDRESS, 16, 4, desc, desc_bytes(desc), 1); +#endif + + return 0; +} + /* * aead_edesc - s/w-extended aead descriptor * @assoc_nents: number of segments in associated data (SPI+Seq) scatterlist - * @assoc_chained: if source is chained * @src_nents: number of segments in input scatterlist - * @src_chained: if source is chained * @dst_nents: number of segments in output scatterlist - * @dst_chained: if destination is chained * @iv_dma: dma address of iv for checking continuity and link table * @desc: h/w descriptor (variable length; must not exceed MAX_CAAM_DESCSIZE) * @sec4_sg_bytes: length of dma mapped sec4_sg space @@ -1721,11 +1838,8 @@ static int ablkcipher_setkey(struct crypto_ablkcipher *ablkcipher, */ struct aead_edesc { int assoc_nents; - bool assoc_chained; int src_nents; - bool src_chained; int dst_nents; - bool dst_chained; dma_addr_t iv_dma; int sec4_sg_bytes; dma_addr_t sec4_sg_dma; @@ -1736,9 +1850,7 @@ struct aead_edesc { /* * ablkcipher_edesc - s/w-extended ablkcipher descriptor * @src_nents: number of segments in input scatterlist - * @src_chained: if source is chained * @dst_nents: number of segments in output scatterlist - * @dst_chained: if destination is chained * @iv_dma: dma address of iv for checking continuity and link table * @desc: h/w descriptor (variable length; must not exceed MAX_CAAM_DESCSIZE) * @sec4_sg_bytes: length of dma mapped sec4_sg space @@ -1747,9 +1859,7 @@ struct aead_edesc { */ struct ablkcipher_edesc { int src_nents; - bool src_chained; int dst_nents; - bool dst_chained; dma_addr_t iv_dma; int sec4_sg_bytes; dma_addr_t sec4_sg_dma; @@ -1759,18 +1869,15 @@ struct ablkcipher_edesc { static void caam_unmap(struct device *dev, struct scatterlist *src, struct scatterlist *dst, int src_nents, - bool src_chained, int dst_nents, bool dst_chained, + int dst_nents, dma_addr_t iv_dma, int ivsize, dma_addr_t sec4_sg_dma, int sec4_sg_bytes) { if (dst != src) { - dma_unmap_sg_chained(dev, src, src_nents ? : 1, DMA_TO_DEVICE, - src_chained); - dma_unmap_sg_chained(dev, dst, dst_nents ? : 1, DMA_FROM_DEVICE, - dst_chained); + dma_unmap_sg(dev, src, src_nents ? : 1, DMA_TO_DEVICE); + dma_unmap_sg(dev, dst, dst_nents ? : 1, DMA_FROM_DEVICE); } else { - dma_unmap_sg_chained(dev, src, src_nents ? : 1, - DMA_BIDIRECTIONAL, src_chained); + dma_unmap_sg(dev, src, src_nents ? : 1, DMA_BIDIRECTIONAL); } if (iv_dma) @@ -1785,8 +1892,7 @@ static void aead_unmap(struct device *dev, struct aead_request *req) { caam_unmap(dev, req->src, req->dst, - edesc->src_nents, edesc->src_chained, edesc->dst_nents, - edesc->dst_chained, 0, 0, + edesc->src_nents, edesc->dst_nents, 0, 0, edesc->sec4_sg_dma, edesc->sec4_sg_bytes); } @@ -1798,8 +1904,8 @@ static void ablkcipher_unmap(struct device *dev, int ivsize = crypto_ablkcipher_ivsize(ablkcipher); caam_unmap(dev, req->src, req->dst, - edesc->src_nents, edesc->src_chained, edesc->dst_nents, - edesc->dst_chained, edesc->iv_dma, ivsize, + edesc->src_nents, edesc->dst_nents, + edesc->iv_dma, ivsize, edesc->sec4_sg_dma, edesc->sec4_sg_bytes); } @@ -2169,22 +2275,18 @@ static struct aead_edesc *aead_edesc_alloc(struct aead_request *req, struct aead_edesc *edesc; int sgc; bool all_contig = true; - bool src_chained = false, dst_chained = false; int sec4_sg_index, sec4_sg_len = 0, sec4_sg_bytes; unsigned int authsize = ctx->authsize; if (unlikely(req->dst != req->src)) { - src_nents = sg_count(req->src, req->assoclen + req->cryptlen, - &src_chained); + src_nents = sg_count(req->src, req->assoclen + req->cryptlen); dst_nents = sg_count(req->dst, req->assoclen + req->cryptlen + - (encrypt ? authsize : (-authsize)), - &dst_chained); + (encrypt ? authsize : (-authsize))); } else { src_nents = sg_count(req->src, req->assoclen + req->cryptlen + - (encrypt ? authsize : 0), - &src_chained); + (encrypt ? authsize : 0)); } /* Check if data are contiguous. */ @@ -2207,37 +2309,35 @@ static struct aead_edesc *aead_edesc_alloc(struct aead_request *req, } if (likely(req->src == req->dst)) { - sgc = dma_map_sg_chained(jrdev, req->src, src_nents ? : 1, - DMA_BIDIRECTIONAL, src_chained); + sgc = dma_map_sg(jrdev, req->src, src_nents ? : 1, + DMA_BIDIRECTIONAL); if (unlikely(!sgc)) { dev_err(jrdev, "unable to map source\n"); kfree(edesc); return ERR_PTR(-ENOMEM); } } else { - sgc = dma_map_sg_chained(jrdev, req->src, src_nents ? : 1, - DMA_TO_DEVICE, src_chained); + sgc = dma_map_sg(jrdev, req->src, src_nents ? : 1, + DMA_TO_DEVICE); if (unlikely(!sgc)) { dev_err(jrdev, "unable to map source\n"); kfree(edesc); return ERR_PTR(-ENOMEM); } - sgc = dma_map_sg_chained(jrdev, req->dst, dst_nents ? : 1, - DMA_FROM_DEVICE, dst_chained); + sgc = dma_map_sg(jrdev, req->dst, dst_nents ? : 1, + DMA_FROM_DEVICE); if (unlikely(!sgc)) { dev_err(jrdev, "unable to map destination\n"); - dma_unmap_sg_chained(jrdev, req->src, src_nents ? : 1, - DMA_TO_DEVICE, src_chained); + dma_unmap_sg(jrdev, req->src, src_nents ? : 1, + DMA_TO_DEVICE); kfree(edesc); return ERR_PTR(-ENOMEM); } } edesc->src_nents = src_nents; - edesc->src_chained = src_chained; edesc->dst_nents = dst_nents; - edesc->dst_chained = dst_chained; edesc->sec4_sg = (void *)edesc + sizeof(struct aead_edesc) + desc_bytes; *all_contig_ptr = all_contig; @@ -2467,22 +2567,21 @@ static struct ablkcipher_edesc *ablkcipher_edesc_alloc(struct ablkcipher_request bool iv_contig = false; int sgc; int ivsize = crypto_ablkcipher_ivsize(ablkcipher); - bool src_chained = false, dst_chained = false; int sec4_sg_index; - src_nents = sg_count(req->src, req->nbytes, &src_chained); + src_nents = sg_count(req->src, req->nbytes); if (req->dst != req->src) - dst_nents = sg_count(req->dst, req->nbytes, &dst_chained); + dst_nents = sg_count(req->dst, req->nbytes); if (likely(req->src == req->dst)) { - sgc = dma_map_sg_chained(jrdev, req->src, src_nents ? : 1, - DMA_BIDIRECTIONAL, src_chained); + sgc = dma_map_sg(jrdev, req->src, src_nents ? : 1, + DMA_BIDIRECTIONAL); } else { - sgc = dma_map_sg_chained(jrdev, req->src, src_nents ? : 1, - DMA_TO_DEVICE, src_chained); - sgc = dma_map_sg_chained(jrdev, req->dst, dst_nents ? : 1, - DMA_FROM_DEVICE, dst_chained); + sgc = dma_map_sg(jrdev, req->src, src_nents ? : 1, + DMA_TO_DEVICE); + sgc = dma_map_sg(jrdev, req->dst, dst_nents ? : 1, + DMA_FROM_DEVICE); } iv_dma = dma_map_single(jrdev, req->info, ivsize, DMA_TO_DEVICE); @@ -2511,9 +2610,7 @@ static struct ablkcipher_edesc *ablkcipher_edesc_alloc(struct ablkcipher_request } edesc->src_nents = src_nents; - edesc->src_chained = src_chained; edesc->dst_nents = dst_nents; - edesc->dst_chained = dst_chained; edesc->sec4_sg_bytes = sec4_sg_bytes; edesc->sec4_sg = (void *)edesc + sizeof(struct ablkcipher_edesc) + desc_bytes; @@ -2646,22 +2743,21 @@ static struct ablkcipher_edesc *ablkcipher_giv_edesc_alloc( bool iv_contig = false; int sgc; int ivsize = crypto_ablkcipher_ivsize(ablkcipher); - bool src_chained = false, dst_chained = false; int sec4_sg_index; - src_nents = sg_count(req->src, req->nbytes, &src_chained); + src_nents = sg_count(req->src, req->nbytes); if (unlikely(req->dst != req->src)) - dst_nents = sg_count(req->dst, req->nbytes, &dst_chained); + dst_nents = sg_count(req->dst, req->nbytes); if (likely(req->src == req->dst)) { - sgc = dma_map_sg_chained(jrdev, req->src, src_nents ? : 1, - DMA_BIDIRECTIONAL, src_chained); + sgc = dma_map_sg(jrdev, req->src, src_nents ? : 1, + DMA_BIDIRECTIONAL); } else { - sgc = dma_map_sg_chained(jrdev, req->src, src_nents ? : 1, - DMA_TO_DEVICE, src_chained); - sgc = dma_map_sg_chained(jrdev, req->dst, dst_nents ? : 1, - DMA_FROM_DEVICE, dst_chained); + sgc = dma_map_sg(jrdev, req->src, src_nents ? : 1, + DMA_TO_DEVICE); + sgc = dma_map_sg(jrdev, req->dst, dst_nents ? : 1, + DMA_FROM_DEVICE); } /* @@ -2690,9 +2786,7 @@ static struct ablkcipher_edesc *ablkcipher_giv_edesc_alloc( } edesc->src_nents = src_nents; - edesc->src_chained = src_chained; edesc->dst_nents = dst_nents; - edesc->dst_chained = dst_chained; edesc->sec4_sg_bytes = sec4_sg_bytes; edesc->sec4_sg = (void *)edesc + sizeof(struct ablkcipher_edesc) + desc_bytes; @@ -2871,7 +2965,23 @@ static struct caam_alg_template driver_algs[] = { .ivsize = CTR_RFC3686_IV_SIZE, }, .class1_alg_type = OP_ALG_ALGSEL_AES | OP_ALG_AAI_CTR_MOD128, - } + }, + { + .name = "xts(aes)", + .driver_name = "xts-aes-caam", + .blocksize = AES_BLOCK_SIZE, + .type = CRYPTO_ALG_TYPE_ABLKCIPHER, + .template_ablkcipher = { + .setkey = xts_ablkcipher_setkey, + .encrypt = ablkcipher_encrypt, + .decrypt = ablkcipher_decrypt, + .geniv = "eseqiv", + .min_keysize = 2 * AES_MIN_KEY_SIZE, + .max_keysize = 2 * AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + }, + .class1_alg_type = OP_ALG_ALGSEL_AES | OP_ALG_AAI_XTS, + }, }; static struct caam_aead_alg driver_aeads[] = { diff --git a/drivers/crypto/caam/caamhash.c b/drivers/crypto/caam/caamhash.c index 94433b9fc200..49106ea42887 100644 --- a/drivers/crypto/caam/caamhash.c +++ b/drivers/crypto/caam/caamhash.c @@ -134,6 +134,15 @@ struct caam_hash_state { int current_buf; }; +struct caam_export_state { + u8 buf[CAAM_MAX_HASH_BLOCK_SIZE]; + u8 caam_ctx[MAX_CTX_LEN]; + int buflen; + int (*update)(struct ahash_request *req); + int (*final)(struct ahash_request *req); + int (*finup)(struct ahash_request *req); +}; + /* Common job descriptor seq in/out ptr routines */ /* Map state->caam_ctx, and append seq_out_ptr command that points to it */ @@ -181,10 +190,9 @@ static inline dma_addr_t buf_map_to_sec4_sg(struct device *jrdev, /* Map req->src and put it in link table */ static inline void src_map_to_sec4_sg(struct device *jrdev, struct scatterlist *src, int src_nents, - struct sec4_sg_entry *sec4_sg, - bool chained) + struct sec4_sg_entry *sec4_sg) { - dma_map_sg_chained(jrdev, src, src_nents, DMA_TO_DEVICE, chained); + dma_map_sg(jrdev, src, src_nents, DMA_TO_DEVICE); sg_to_sec4_sg_last(src, src_nents, sec4_sg, 0); } @@ -585,7 +593,6 @@ badkey: * ahash_edesc - s/w-extended ahash descriptor * @dst_dma: physical mapped address of req->result * @sec4_sg_dma: physical mapped address of h/w link table - * @chained: if source is chained * @src_nents: number of segments in input scatterlist * @sec4_sg_bytes: length of dma mapped sec4_sg space * @sec4_sg: pointer to h/w link table @@ -594,7 +601,6 @@ badkey: struct ahash_edesc { dma_addr_t dst_dma; dma_addr_t sec4_sg_dma; - bool chained; int src_nents; int sec4_sg_bytes; struct sec4_sg_entry *sec4_sg; @@ -606,8 +612,7 @@ static inline void ahash_unmap(struct device *dev, struct ahash_request *req, int dst_len) { if (edesc->src_nents) - dma_unmap_sg_chained(dev, req->src, edesc->src_nents, - DMA_TO_DEVICE, edesc->chained); + dma_unmap_sg(dev, req->src, edesc->src_nents, DMA_TO_DEVICE); if (edesc->dst_dma) dma_unmap_single(dev, edesc->dst_dma, dst_len, DMA_FROM_DEVICE); @@ -788,7 +793,6 @@ static int ahash_update_ctx(struct ahash_request *req) dma_addr_t ptr = ctx->sh_desc_update_dma; int src_nents, sec4_sg_bytes, sec4_sg_src_index; struct ahash_edesc *edesc; - bool chained = false; int ret = 0; int sh_len; @@ -797,8 +801,8 @@ static int ahash_update_ctx(struct ahash_request *req) to_hash = in_len - *next_buflen; if (to_hash) { - src_nents = __sg_count(req->src, req->nbytes - (*next_buflen), - &chained); + src_nents = sg_nents_for_len(req->src, + req->nbytes - (*next_buflen)); sec4_sg_src_index = 1 + (*buflen ? 1 : 0); sec4_sg_bytes = (sec4_sg_src_index + src_nents) * sizeof(struct sec4_sg_entry); @@ -816,7 +820,6 @@ static int ahash_update_ctx(struct ahash_request *req) } edesc->src_nents = src_nents; - edesc->chained = chained; edesc->sec4_sg_bytes = sec4_sg_bytes; edesc->sec4_sg = (void *)edesc + sizeof(struct ahash_edesc) + DESC_JOB_IO_LEN; @@ -829,12 +832,11 @@ static int ahash_update_ctx(struct ahash_request *req) state->buf_dma = try_buf_map_to_sec4_sg(jrdev, edesc->sec4_sg + 1, buf, state->buf_dma, - *next_buflen, *buflen); + *buflen, last_buflen); if (src_nents) { src_map_to_sec4_sg(jrdev, req->src, src_nents, - edesc->sec4_sg + sec4_sg_src_index, - chained); + edesc->sec4_sg + sec4_sg_src_index); if (*next_buflen) scatterwalk_map_and_copy(next_buf, req->src, to_hash - *buflen, @@ -996,11 +998,10 @@ static int ahash_finup_ctx(struct ahash_request *req) int src_nents; int digestsize = crypto_ahash_digestsize(ahash); struct ahash_edesc *edesc; - bool chained = false; int ret = 0; int sh_len; - src_nents = __sg_count(req->src, req->nbytes, &chained); + src_nents = sg_nents_for_len(req->src, req->nbytes); sec4_sg_src_index = 1 + (buflen ? 1 : 0); sec4_sg_bytes = (sec4_sg_src_index + src_nents) * sizeof(struct sec4_sg_entry); @@ -1018,7 +1019,6 @@ static int ahash_finup_ctx(struct ahash_request *req) init_job_desc_shared(desc, ptr, sh_len, HDR_SHARE_DEFER | HDR_REVERSE); edesc->src_nents = src_nents; - edesc->chained = chained; edesc->sec4_sg_bytes = sec4_sg_bytes; edesc->sec4_sg = (void *)edesc + sizeof(struct ahash_edesc) + DESC_JOB_IO_LEN; @@ -1033,7 +1033,7 @@ static int ahash_finup_ctx(struct ahash_request *req) last_buflen); src_map_to_sec4_sg(jrdev, req->src, src_nents, edesc->sec4_sg + - sec4_sg_src_index, chained); + sec4_sg_src_index); edesc->sec4_sg_dma = dma_map_single(jrdev, edesc->sec4_sg, sec4_sg_bytes, DMA_TO_DEVICE); @@ -1081,14 +1081,12 @@ static int ahash_digest(struct ahash_request *req) int src_nents, sec4_sg_bytes; dma_addr_t src_dma; struct ahash_edesc *edesc; - bool chained = false; int ret = 0; u32 options; int sh_len; - src_nents = sg_count(req->src, req->nbytes, &chained); - dma_map_sg_chained(jrdev, req->src, src_nents ? : 1, DMA_TO_DEVICE, - chained); + src_nents = sg_count(req->src, req->nbytes); + dma_map_sg(jrdev, req->src, src_nents ? : 1, DMA_TO_DEVICE); sec4_sg_bytes = src_nents * sizeof(struct sec4_sg_entry); /* allocate space for base edesc and hw desc commands, link tables */ @@ -1102,7 +1100,6 @@ static int ahash_digest(struct ahash_request *req) DESC_JOB_IO_LEN; edesc->sec4_sg_bytes = sec4_sg_bytes; edesc->src_nents = src_nents; - edesc->chained = chained; sh_len = desc_len(sh_desc); desc = edesc->hw_desc; @@ -1228,7 +1225,6 @@ static int ahash_update_no_ctx(struct ahash_request *req) struct ahash_edesc *edesc; u32 *desc, *sh_desc = ctx->sh_desc_update_first; dma_addr_t ptr = ctx->sh_desc_update_first_dma; - bool chained = false; int ret = 0; int sh_len; @@ -1236,8 +1232,8 @@ static int ahash_update_no_ctx(struct ahash_request *req) to_hash = in_len - *next_buflen; if (to_hash) { - src_nents = __sg_count(req->src, req->nbytes - (*next_buflen), - &chained); + src_nents = sg_nents_for_len(req->src, + req->nbytes - (*next_buflen)); sec4_sg_bytes = (1 + src_nents) * sizeof(struct sec4_sg_entry); @@ -1254,7 +1250,6 @@ static int ahash_update_no_ctx(struct ahash_request *req) } edesc->src_nents = src_nents; - edesc->chained = chained; edesc->sec4_sg_bytes = sec4_sg_bytes; edesc->sec4_sg = (void *)edesc + sizeof(struct ahash_edesc) + DESC_JOB_IO_LEN; @@ -1263,7 +1258,7 @@ static int ahash_update_no_ctx(struct ahash_request *req) state->buf_dma = buf_map_to_sec4_sg(jrdev, edesc->sec4_sg, buf, *buflen); src_map_to_sec4_sg(jrdev, req->src, src_nents, - edesc->sec4_sg + 1, chained); + edesc->sec4_sg + 1); if (*next_buflen) { scatterwalk_map_and_copy(next_buf, req->src, to_hash - *buflen, @@ -1343,11 +1338,10 @@ static int ahash_finup_no_ctx(struct ahash_request *req) int sec4_sg_bytes, sec4_sg_src_index, src_nents; int digestsize = crypto_ahash_digestsize(ahash); struct ahash_edesc *edesc; - bool chained = false; int sh_len; int ret = 0; - src_nents = __sg_count(req->src, req->nbytes, &chained); + src_nents = sg_nents_for_len(req->src, req->nbytes); sec4_sg_src_index = 2; sec4_sg_bytes = (sec4_sg_src_index + src_nents) * sizeof(struct sec4_sg_entry); @@ -1365,7 +1359,6 @@ static int ahash_finup_no_ctx(struct ahash_request *req) init_job_desc_shared(desc, ptr, sh_len, HDR_SHARE_DEFER | HDR_REVERSE); edesc->src_nents = src_nents; - edesc->chained = chained; edesc->sec4_sg_bytes = sec4_sg_bytes; edesc->sec4_sg = (void *)edesc + sizeof(struct ahash_edesc) + DESC_JOB_IO_LEN; @@ -1374,8 +1367,7 @@ static int ahash_finup_no_ctx(struct ahash_request *req) state->buf_dma, buflen, last_buflen); - src_map_to_sec4_sg(jrdev, req->src, src_nents, edesc->sec4_sg + 1, - chained); + src_map_to_sec4_sg(jrdev, req->src, src_nents, edesc->sec4_sg + 1); edesc->sec4_sg_dma = dma_map_single(jrdev, edesc->sec4_sg, sec4_sg_bytes, DMA_TO_DEVICE); @@ -1429,7 +1421,6 @@ static int ahash_update_first(struct ahash_request *req) dma_addr_t src_dma; u32 options; struct ahash_edesc *edesc; - bool chained = false; int ret = 0; int sh_len; @@ -1438,10 +1429,8 @@ static int ahash_update_first(struct ahash_request *req) to_hash = req->nbytes - *next_buflen; if (to_hash) { - src_nents = sg_count(req->src, req->nbytes - (*next_buflen), - &chained); - dma_map_sg_chained(jrdev, req->src, src_nents ? : 1, - DMA_TO_DEVICE, chained); + src_nents = sg_count(req->src, req->nbytes - (*next_buflen)); + dma_map_sg(jrdev, req->src, src_nents ? : 1, DMA_TO_DEVICE); sec4_sg_bytes = src_nents * sizeof(struct sec4_sg_entry); /* @@ -1457,7 +1446,6 @@ static int ahash_update_first(struct ahash_request *req) } edesc->src_nents = src_nents; - edesc->chained = chained; edesc->sec4_sg_bytes = sec4_sg_bytes; edesc->sec4_sg = (void *)edesc + sizeof(struct ahash_edesc) + DESC_JOB_IO_LEN; @@ -1574,25 +1562,42 @@ static int ahash_final(struct ahash_request *req) static int ahash_export(struct ahash_request *req, void *out) { - struct crypto_ahash *ahash = crypto_ahash_reqtfm(req); - struct caam_hash_ctx *ctx = crypto_ahash_ctx(ahash); struct caam_hash_state *state = ahash_request_ctx(req); + struct caam_export_state *export = out; + int len; + u8 *buf; + + if (state->current_buf) { + buf = state->buf_1; + len = state->buflen_1; + } else { + buf = state->buf_0; + len = state->buflen_1; + } + + memcpy(export->buf, buf, len); + memcpy(export->caam_ctx, state->caam_ctx, sizeof(export->caam_ctx)); + export->buflen = len; + export->update = state->update; + export->final = state->final; + export->finup = state->finup; - memcpy(out, ctx, sizeof(struct caam_hash_ctx)); - memcpy(out + sizeof(struct caam_hash_ctx), state, - sizeof(struct caam_hash_state)); return 0; } static int ahash_import(struct ahash_request *req, const void *in) { - struct crypto_ahash *ahash = crypto_ahash_reqtfm(req); - struct caam_hash_ctx *ctx = crypto_ahash_ctx(ahash); struct caam_hash_state *state = ahash_request_ctx(req); + const struct caam_export_state *export = in; + + memset(state, 0, sizeof(*state)); + memcpy(state->buf_0, export->buf, export->buflen); + memcpy(state->caam_ctx, export->caam_ctx, sizeof(state->caam_ctx)); + state->buflen_0 = export->buflen; + state->update = export->update; + state->final = export->final; + state->finup = export->finup; - memcpy(ctx, in, sizeof(struct caam_hash_ctx)); - memcpy(state, in + sizeof(struct caam_hash_ctx), - sizeof(struct caam_hash_state)); return 0; } @@ -1626,8 +1631,9 @@ static struct caam_hash_template driver_hash[] = { .setkey = ahash_setkey, .halg = { .digestsize = SHA1_DIGEST_SIZE, - }, + .statesize = sizeof(struct caam_export_state), }, + }, .alg_type = OP_ALG_ALGSEL_SHA1, .alg_op = OP_ALG_ALGSEL_SHA1 | OP_ALG_AAI_HMAC, }, { @@ -1647,8 +1653,9 @@ static struct caam_hash_template driver_hash[] = { .setkey = ahash_setkey, .halg = { .digestsize = SHA224_DIGEST_SIZE, - }, + .statesize = sizeof(struct caam_export_state), }, + }, .alg_type = OP_ALG_ALGSEL_SHA224, .alg_op = OP_ALG_ALGSEL_SHA224 | OP_ALG_AAI_HMAC, }, { @@ -1668,8 +1675,9 @@ static struct caam_hash_template driver_hash[] = { .setkey = ahash_setkey, .halg = { .digestsize = SHA256_DIGEST_SIZE, - }, + .statesize = sizeof(struct caam_export_state), }, + }, .alg_type = OP_ALG_ALGSEL_SHA256, .alg_op = OP_ALG_ALGSEL_SHA256 | OP_ALG_AAI_HMAC, }, { @@ -1689,8 +1697,9 @@ static struct caam_hash_template driver_hash[] = { .setkey = ahash_setkey, .halg = { .digestsize = SHA384_DIGEST_SIZE, - }, + .statesize = sizeof(struct caam_export_state), }, + }, .alg_type = OP_ALG_ALGSEL_SHA384, .alg_op = OP_ALG_ALGSEL_SHA384 | OP_ALG_AAI_HMAC, }, { @@ -1710,8 +1719,9 @@ static struct caam_hash_template driver_hash[] = { .setkey = ahash_setkey, .halg = { .digestsize = SHA512_DIGEST_SIZE, - }, + .statesize = sizeof(struct caam_export_state), }, + }, .alg_type = OP_ALG_ALGSEL_SHA512, .alg_op = OP_ALG_ALGSEL_SHA512 | OP_ALG_AAI_HMAC, }, { @@ -1731,8 +1741,9 @@ static struct caam_hash_template driver_hash[] = { .setkey = ahash_setkey, .halg = { .digestsize = MD5_DIGEST_SIZE, - }, + .statesize = sizeof(struct caam_export_state), }, + }, .alg_type = OP_ALG_ALGSEL_MD5, .alg_op = OP_ALG_ALGSEL_MD5 | OP_ALG_AAI_HMAC, }, @@ -1952,8 +1963,9 @@ static int __init caam_algapi_hash_init(void) err = crypto_register_ahash(&t_alg->ahash_alg); if (err) { - pr_warn("%s alg registration failed\n", - t_alg->ahash_alg.halg.base.cra_driver_name); + pr_warn("%s alg registration failed: %d\n", + t_alg->ahash_alg.halg.base.cra_driver_name, + err); kfree(t_alg); } else list_add_tail(&t_alg->entry, &hash_list); @@ -1968,8 +1980,9 @@ static int __init caam_algapi_hash_init(void) err = crypto_register_ahash(&t_alg->ahash_alg); if (err) { - pr_warn("%s alg registration failed\n", - t_alg->ahash_alg.halg.base.cra_driver_name); + pr_warn("%s alg registration failed: %d\n", + t_alg->ahash_alg.halg.base.cra_driver_name, + err); kfree(t_alg); } else list_add_tail(&t_alg->entry, &hash_list); diff --git a/drivers/crypto/caam/desc.h b/drivers/crypto/caam/desc.h index 983d663ef671..1e93c6af2275 100644 --- a/drivers/crypto/caam/desc.h +++ b/drivers/crypto/caam/desc.h @@ -1492,7 +1492,6 @@ struct sec4_sg_entry { #define JUMP_JSL (1 << JUMP_JSL_SHIFT) #define JUMP_TYPE_SHIFT 22 -#define JUMP_TYPE_MASK (0x03 << JUMP_TYPE_SHIFT) #define JUMP_TYPE_LOCAL (0x00 << JUMP_TYPE_SHIFT) #define JUMP_TYPE_NONLOCAL (0x01 << JUMP_TYPE_SHIFT) #define JUMP_TYPE_HALT (0x02 << JUMP_TYPE_SHIFT) diff --git a/drivers/crypto/caam/sg_sw_sec4.h b/drivers/crypto/caam/sg_sw_sec4.h index 18cd6d1f5870..12ec6616e89d 100644 --- a/drivers/crypto/caam/sg_sw_sec4.h +++ b/drivers/crypto/caam/sg_sw_sec4.h @@ -69,81 +69,13 @@ static inline struct sec4_sg_entry *sg_to_sec4_sg_len( return sec4_sg_ptr - 1; } -/* count number of elements in scatterlist */ -static inline int __sg_count(struct scatterlist *sg_list, int nbytes, - bool *chained) -{ - struct scatterlist *sg = sg_list; - int sg_nents = 0; - - while (nbytes > 0) { - sg_nents++; - nbytes -= sg->length; - if (!sg_is_last(sg) && (sg + 1)->length == 0) - *chained = true; - sg = sg_next(sg); - } - - return sg_nents; -} - /* derive number of elements in scatterlist, but return 0 for 1 */ -static inline int sg_count(struct scatterlist *sg_list, int nbytes, - bool *chained) +static inline int sg_count(struct scatterlist *sg_list, int nbytes) { - int sg_nents = __sg_count(sg_list, nbytes, chained); + int sg_nents = sg_nents_for_len(sg_list, nbytes); if (likely(sg_nents == 1)) return 0; return sg_nents; } - -static inline void dma_unmap_sg_chained( - struct device *dev, struct scatterlist *sg, unsigned int nents, - enum dma_data_direction dir, bool chained) -{ - if (unlikely(chained)) { - int i; - struct scatterlist *tsg = sg; - - /* - * Use a local copy of the sg pointer to avoid moving the - * head of the list pointed to by sg as we walk the list. - */ - for (i = 0; i < nents; i++) { - dma_unmap_sg(dev, tsg, 1, dir); - tsg = sg_next(tsg); - } - } else if (nents) { - dma_unmap_sg(dev, sg, nents, dir); - } -} - -static inline int dma_map_sg_chained( - struct device *dev, struct scatterlist *sg, unsigned int nents, - enum dma_data_direction dir, bool chained) -{ - if (unlikely(chained)) { - int i; - struct scatterlist *tsg = sg; - - /* - * Use a local copy of the sg pointer to avoid moving the - * head of the list pointed to by sg as we walk the list. - */ - for (i = 0; i < nents; i++) { - if (!dma_map_sg(dev, tsg, 1, dir)) { - dma_unmap_sg_chained(dev, sg, i, dir, - chained); - nents = 0; - break; - } - - tsg = sg_next(tsg); - } - } else - nents = dma_map_sg(dev, sg, nents, dir); - - return nents; -} diff --git a/drivers/crypto/ccp/Kconfig b/drivers/crypto/ccp/Kconfig index ae38f6b6cc10..3cd8481065f8 100644 --- a/drivers/crypto/ccp/Kconfig +++ b/drivers/crypto/ccp/Kconfig @@ -5,12 +5,12 @@ config CRYPTO_DEV_CCP_DD select HW_RANDOM help Provides the interface to use the AMD Cryptographic Coprocessor - which can be used to accelerate or offload encryption operations - such as SHA, AES and more. If you choose 'M' here, this module - will be called ccp. + which can be used to offload encryption operations such as SHA, + AES and more. If you choose 'M' here, this module will be called + ccp. config CRYPTO_DEV_CCP_CRYPTO - tristate "Encryption and hashing acceleration support" + tristate "Encryption and hashing offload support" depends on CRYPTO_DEV_CCP_DD default m select CRYPTO_HASH @@ -18,6 +18,5 @@ config CRYPTO_DEV_CCP_CRYPTO select CRYPTO_AUTHENC help Support for using the cryptographic API with the AMD Cryptographic - Coprocessor. This module supports acceleration and offload of SHA - and AES algorithms. If you choose 'M' here, this module will be - called ccp_crypto. + Coprocessor. This module supports offload of SHA and AES algorithms. + If you choose 'M' here, this module will be called ccp_crypto. diff --git a/drivers/crypto/ccp/ccp-crypto-aes-cmac.c b/drivers/crypto/ccp/ccp-crypto-aes-cmac.c index ea7e8446956a..d89f20c04266 100644 --- a/drivers/crypto/ccp/ccp-crypto-aes-cmac.c +++ b/drivers/crypto/ccp/ccp-crypto-aes-cmac.c @@ -118,10 +118,19 @@ static int ccp_do_cmac_update(struct ahash_request *req, unsigned int nbytes, if (rctx->buf_count) { sg_init_one(&rctx->buf_sg, rctx->buf, rctx->buf_count); sg = ccp_crypto_sg_table_add(&rctx->data_sg, &rctx->buf_sg); + if (!sg) { + ret = -EINVAL; + goto e_free; + } } - if (nbytes) + if (nbytes) { sg = ccp_crypto_sg_table_add(&rctx->data_sg, req->src); + if (!sg) { + ret = -EINVAL; + goto e_free; + } + } if (need_pad) { int pad_length = block_size - (len & (block_size - 1)); @@ -132,6 +141,10 @@ static int ccp_do_cmac_update(struct ahash_request *req, unsigned int nbytes, rctx->pad[0] = 0x80; sg_init_one(&rctx->pad_sg, rctx->pad, pad_length); sg = ccp_crypto_sg_table_add(&rctx->data_sg, &rctx->pad_sg); + if (!sg) { + ret = -EINVAL; + goto e_free; + } } if (sg) { sg_mark_end(sg); @@ -163,6 +176,11 @@ static int ccp_do_cmac_update(struct ahash_request *req, unsigned int nbytes, ret = ccp_crypto_enqueue_request(&req->base, &rctx->cmd); return ret; + +e_free: + sg_free_table(&rctx->data_sg); + + return ret; } static int ccp_aes_cmac_init(struct ahash_request *req) diff --git a/drivers/crypto/ccp/ccp-crypto-main.c b/drivers/crypto/ccp/ccp-crypto-main.c index bdec01ec608f..e0380e59c361 100644 --- a/drivers/crypto/ccp/ccp-crypto-main.c +++ b/drivers/crypto/ccp/ccp-crypto-main.c @@ -305,14 +305,16 @@ struct scatterlist *ccp_crypto_sg_table_add(struct sg_table *table, for (sg = table->sgl; sg; sg = sg_next(sg)) if (!sg_page(sg)) break; - BUG_ON(!sg); + if (WARN_ON(!sg)) + return NULL; for (; sg && sg_add; sg = sg_next(sg), sg_add = sg_next(sg_add)) { sg_set_page(sg, sg_page(sg_add), sg_add->length, sg_add->offset); sg_last = sg; } - BUG_ON(sg_add); + if (WARN_ON(sg_add)) + return NULL; return sg_last; } diff --git a/drivers/crypto/ccp/ccp-crypto-sha.c b/drivers/crypto/ccp/ccp-crypto-sha.c index 507b34e0cc19..d14b3f28e010 100644 --- a/drivers/crypto/ccp/ccp-crypto-sha.c +++ b/drivers/crypto/ccp/ccp-crypto-sha.c @@ -107,7 +107,15 @@ static int ccp_do_sha_update(struct ahash_request *req, unsigned int nbytes, sg_init_one(&rctx->buf_sg, rctx->buf, rctx->buf_count); sg = ccp_crypto_sg_table_add(&rctx->data_sg, &rctx->buf_sg); + if (!sg) { + ret = -EINVAL; + goto e_free; + } sg = ccp_crypto_sg_table_add(&rctx->data_sg, req->src); + if (!sg) { + ret = -EINVAL; + goto e_free; + } sg_mark_end(sg); sg = rctx->data_sg.sgl; @@ -142,6 +150,11 @@ static int ccp_do_sha_update(struct ahash_request *req, unsigned int nbytes, ret = ccp_crypto_enqueue_request(&req->base, &rctx->cmd); return ret; + +e_free: + sg_free_table(&rctx->data_sg); + + return ret; } static int ccp_sha_init(struct ahash_request *req) diff --git a/drivers/crypto/ccp/ccp-ops.c b/drivers/crypto/ccp/ccp-ops.c index d09c6c4af4aa..c6e883b296a9 100644 --- a/drivers/crypto/ccp/ccp-ops.c +++ b/drivers/crypto/ccp/ccp-ops.c @@ -611,15 +611,16 @@ static void ccp_get_dm_area(struct ccp_dm_workarea *wa, unsigned int wa_offset, 1); } -static void ccp_reverse_set_dm_area(struct ccp_dm_workarea *wa, - struct scatterlist *sg, - unsigned int len, unsigned int se_len, - bool sign_extend) +static int ccp_reverse_set_dm_area(struct ccp_dm_workarea *wa, + struct scatterlist *sg, + unsigned int len, unsigned int se_len, + bool sign_extend) { unsigned int nbytes, sg_offset, dm_offset, ksb_len, i; u8 buffer[CCP_REVERSE_BUF_SIZE]; - BUG_ON(se_len > sizeof(buffer)); + if (WARN_ON(se_len > sizeof(buffer))) + return -EINVAL; sg_offset = len; dm_offset = 0; @@ -642,6 +643,8 @@ static void ccp_reverse_set_dm_area(struct ccp_dm_workarea *wa, se_len - ksb_len); } } + + return 0; } static void ccp_reverse_get_dm_area(struct ccp_dm_workarea *wa, @@ -1606,8 +1609,10 @@ static int ccp_run_rsa_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) if (ret) goto e_ksb; - ccp_reverse_set_dm_area(&exp, rsa->exp, rsa->exp_len, CCP_KSB_BYTES, - false); + ret = ccp_reverse_set_dm_area(&exp, rsa->exp, rsa->exp_len, + CCP_KSB_BYTES, false); + if (ret) + goto e_exp; ret = ccp_copy_to_ksb(cmd_q, &exp, op.jobid, op.ksb_key, CCP_PASSTHRU_BYTESWAP_NOOP); if (ret) { @@ -1623,11 +1628,15 @@ static int ccp_run_rsa_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) if (ret) goto e_exp; - ccp_reverse_set_dm_area(&src, rsa->mod, rsa->mod_len, CCP_KSB_BYTES, - false); + ret = ccp_reverse_set_dm_area(&src, rsa->mod, rsa->mod_len, + CCP_KSB_BYTES, false); + if (ret) + goto e_src; src.address += o_len; /* Adjust the address for the copy operation */ - ccp_reverse_set_dm_area(&src, rsa->src, rsa->src_len, CCP_KSB_BYTES, - false); + ret = ccp_reverse_set_dm_area(&src, rsa->src, rsa->src_len, + CCP_KSB_BYTES, false); + if (ret) + goto e_src; src.address -= o_len; /* Reset the address to original value */ /* Prepare the output area for the operation */ @@ -1841,21 +1850,27 @@ static int ccp_run_ecc_mm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) save = src.address; /* Copy the ECC modulus */ - ccp_reverse_set_dm_area(&src, ecc->mod, ecc->mod_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->mod, ecc->mod_len, + CCP_ECC_OPERAND_SIZE, false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; /* Copy the first operand */ - ccp_reverse_set_dm_area(&src, ecc->u.mm.operand_1, - ecc->u.mm.operand_1_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->u.mm.operand_1, + ecc->u.mm.operand_1_len, + CCP_ECC_OPERAND_SIZE, false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; if (ecc->function != CCP_ECC_FUNCTION_MINV_384BIT) { /* Copy the second operand */ - ccp_reverse_set_dm_area(&src, ecc->u.mm.operand_2, - ecc->u.mm.operand_2_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->u.mm.operand_2, + ecc->u.mm.operand_2_len, + CCP_ECC_OPERAND_SIZE, false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; } @@ -1960,18 +1975,24 @@ static int ccp_run_ecc_pm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) save = src.address; /* Copy the ECC modulus */ - ccp_reverse_set_dm_area(&src, ecc->mod, ecc->mod_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->mod, ecc->mod_len, + CCP_ECC_OPERAND_SIZE, false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; /* Copy the first point X and Y coordinate */ - ccp_reverse_set_dm_area(&src, ecc->u.pm.point_1.x, - ecc->u.pm.point_1.x_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->u.pm.point_1.x, + ecc->u.pm.point_1.x_len, + CCP_ECC_OPERAND_SIZE, false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; - ccp_reverse_set_dm_area(&src, ecc->u.pm.point_1.y, - ecc->u.pm.point_1.y_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->u.pm.point_1.y, + ecc->u.pm.point_1.y_len, + CCP_ECC_OPERAND_SIZE, false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; /* Set the first point Z coordianate to 1 */ @@ -1980,13 +2001,17 @@ static int ccp_run_ecc_pm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) if (ecc->function == CCP_ECC_FUNCTION_PADD_384BIT) { /* Copy the second point X and Y coordinate */ - ccp_reverse_set_dm_area(&src, ecc->u.pm.point_2.x, - ecc->u.pm.point_2.x_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->u.pm.point_2.x, + ecc->u.pm.point_2.x_len, + CCP_ECC_OPERAND_SIZE, false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; - ccp_reverse_set_dm_area(&src, ecc->u.pm.point_2.y, - ecc->u.pm.point_2.y_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->u.pm.point_2.y, + ecc->u.pm.point_2.y_len, + CCP_ECC_OPERAND_SIZE, false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; /* Set the second point Z coordianate to 1 */ @@ -1994,16 +2019,21 @@ static int ccp_run_ecc_pm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) src.address += CCP_ECC_OPERAND_SIZE; } else { /* Copy the Domain "a" parameter */ - ccp_reverse_set_dm_area(&src, ecc->u.pm.domain_a, - ecc->u.pm.domain_a_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->u.pm.domain_a, + ecc->u.pm.domain_a_len, + CCP_ECC_OPERAND_SIZE, false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; if (ecc->function == CCP_ECC_FUNCTION_PMUL_384BIT) { /* Copy the scalar value */ - ccp_reverse_set_dm_area(&src, ecc->u.pm.scalar, - ecc->u.pm.scalar_len, - CCP_ECC_OPERAND_SIZE, false); + ret = ccp_reverse_set_dm_area(&src, ecc->u.pm.scalar, + ecc->u.pm.scalar_len, + CCP_ECC_OPERAND_SIZE, + false); + if (ret) + goto e_src; src.address += CCP_ECC_OPERAND_SIZE; } } diff --git a/drivers/crypto/ccp/ccp-pci.c b/drivers/crypto/ccp/ccp-pci.c index af190d4795a8..6ade02f04f91 100644 --- a/drivers/crypto/ccp/ccp-pci.c +++ b/drivers/crypto/ccp/ccp-pci.c @@ -319,7 +319,7 @@ static const struct pci_device_id ccp_pci_table[] = { MODULE_DEVICE_TABLE(pci, ccp_pci_table); static struct pci_driver ccp_pci_driver = { - .name = "AMD Cryptographic Coprocessor", + .name = "ccp", .id_table = ccp_pci_table, .probe = ccp_pci_probe, .remove = ccp_pci_remove, diff --git a/drivers/crypto/ccp/ccp-platform.c b/drivers/crypto/ccp/ccp-platform.c index bb241c3ab6b9..8b923b7e9389 100644 --- a/drivers/crypto/ccp/ccp-platform.c +++ b/drivers/crypto/ccp/ccp-platform.c @@ -29,7 +29,6 @@ #include "ccp-dev.h" struct ccp_platform { - int use_acpi; int coherent; }; @@ -95,7 +94,6 @@ static int ccp_platform_probe(struct platform_device *pdev) struct ccp_device *ccp; struct ccp_platform *ccp_platform; struct device *dev = &pdev->dev; - struct acpi_device *adev = ACPI_COMPANION(dev); struct resource *ior; int ret; @@ -112,8 +110,6 @@ static int ccp_platform_probe(struct platform_device *pdev) ccp->get_irq = ccp_get_irqs; ccp->free_irq = ccp_free_irqs; - ccp_platform->use_acpi = (!adev || acpi_disabled) ? 0 : 1; - ior = ccp_find_mmio_area(ccp); ccp->io_map = devm_ioremap_resource(dev, ior); if (IS_ERR(ccp->io_map)) { @@ -229,7 +225,7 @@ MODULE_DEVICE_TABLE(of, ccp_of_match); static struct platform_driver ccp_platform_driver = { .driver = { - .name = "AMD Cryptographic Coprocessor", + .name = "ccp", #ifdef CONFIG_ACPI .acpi_match_table = ccp_acpi_match, #endif diff --git a/drivers/crypto/marvell/cesa.h b/drivers/crypto/marvell/cesa.h index bc2a55bc35e4..bd985e72520b 100644 --- a/drivers/crypto/marvell/cesa.h +++ b/drivers/crypto/marvell/cesa.h @@ -174,19 +174,19 @@ #define CESA_SA_DESC_MAC_DATA(offset) \ cpu_to_le32(CESA_SA_DATA_SRAM_OFFSET + (offset)) -#define CESA_SA_DESC_MAC_DATA_MSK GENMASK(15, 0) +#define CESA_SA_DESC_MAC_DATA_MSK cpu_to_le32(GENMASK(15, 0)) #define CESA_SA_DESC_MAC_TOTAL_LEN(total_len) cpu_to_le32((total_len) << 16) -#define CESA_SA_DESC_MAC_TOTAL_LEN_MSK GENMASK(31, 16) +#define CESA_SA_DESC_MAC_TOTAL_LEN_MSK cpu_to_le32(GENMASK(31, 16)) #define CESA_SA_DESC_MAC_SRC_TOTAL_LEN_MAX 0xffff #define CESA_SA_DESC_MAC_DIGEST(offset) \ cpu_to_le32(CESA_SA_MAC_DIG_SRAM_OFFSET + (offset)) -#define CESA_SA_DESC_MAC_DIGEST_MSK GENMASK(15, 0) +#define CESA_SA_DESC_MAC_DIGEST_MSK cpu_to_le32(GENMASK(15, 0)) #define CESA_SA_DESC_MAC_FRAG_LEN(frag_len) cpu_to_le32((frag_len) << 16) -#define CESA_SA_DESC_MAC_FRAG_LEN_MSK GENMASK(31, 16) +#define CESA_SA_DESC_MAC_FRAG_LEN_MSK cpu_to_le32(GENMASK(31, 16)) #define CESA_SA_DESC_MAC_IV(offset) \ cpu_to_le32((CESA_SA_MAC_IIV_SRAM_OFFSET + (offset)) | \ @@ -219,14 +219,14 @@ * to be executed. */ struct mv_cesa_sec_accel_desc { - u32 config; - u32 enc_p; - u32 enc_len; - u32 enc_key_p; - u32 enc_iv; - u32 mac_src_p; - u32 mac_digest; - u32 mac_iv; + __le32 config; + __le32 enc_p; + __le32 enc_len; + __le32 enc_key_p; + __le32 enc_iv; + __le32 mac_src_p; + __le32 mac_digest; + __le32 mac_iv; }; /** @@ -293,11 +293,13 @@ struct mv_cesa_op_ctx { * operation. */ struct mv_cesa_tdma_desc { - u32 byte_cnt; - u32 src; - u32 dst; - u32 next_dma; - u32 cur_dma; + __le32 byte_cnt; + __le32 src; + __le32 dst; + __le32 next_dma; + + /* Software state */ + dma_addr_t cur_dma; struct mv_cesa_tdma_desc *next; union { struct mv_cesa_op_ctx *op; @@ -612,7 +614,8 @@ struct mv_cesa_ahash_req { u64 len; int src_nents; bool last_req; - __be32 state[8]; + bool algo_le; + u32 state[8]; }; /* CESA functions */ @@ -626,7 +629,7 @@ static inline void mv_cesa_update_op_cfg(struct mv_cesa_op_ctx *op, op->desc.config |= cpu_to_le32(cfg); } -static inline u32 mv_cesa_get_op_cfg(struct mv_cesa_op_ctx *op) +static inline u32 mv_cesa_get_op_cfg(const struct mv_cesa_op_ctx *op) { return le32_to_cpu(op->desc.config); } @@ -676,7 +679,7 @@ static inline void mv_cesa_set_int_mask(struct mv_cesa_engine *engine, if (int_mask == engine->int_mask) return; - writel(int_mask, engine->regs + CESA_SA_INT_MSK); + writel_relaxed(int_mask, engine->regs + CESA_SA_INT_MSK); engine->int_mask = int_mask; } @@ -685,6 +688,12 @@ static inline u32 mv_cesa_get_int_mask(struct mv_cesa_engine *engine) return engine->int_mask; } +static inline bool mv_cesa_mac_op_is_first_frag(const struct mv_cesa_op_ctx *op) +{ + return (mv_cesa_get_op_cfg(op) & CESA_SA_DESC_CFG_FRAG_MSK) == + CESA_SA_DESC_CFG_FIRST_FRAG; +} + int mv_cesa_queue_req(struct crypto_async_request *req); /* @@ -789,10 +798,8 @@ int mv_cesa_dma_add_data_transfer(struct mv_cesa_tdma_chain *chain, dma_addr_t dst, dma_addr_t src, u32 size, u32 flags, gfp_t gfp_flags); -int mv_cesa_dma_add_dummy_launch(struct mv_cesa_tdma_chain *chain, - u32 flags); - -int mv_cesa_dma_add_dummy_end(struct mv_cesa_tdma_chain *chain, u32 flags); +int mv_cesa_dma_add_dummy_launch(struct mv_cesa_tdma_chain *chain, gfp_t flags); +int mv_cesa_dma_add_dummy_end(struct mv_cesa_tdma_chain *chain, gfp_t flags); int mv_cesa_dma_add_op_transfers(struct mv_cesa_tdma_chain *chain, struct mv_cesa_dma_iter *dma_iter, diff --git a/drivers/crypto/marvell/cipher.c b/drivers/crypto/marvell/cipher.c index 3df2f4e7adb2..6edae64bb387 100644 --- a/drivers/crypto/marvell/cipher.c +++ b/drivers/crypto/marvell/cipher.c @@ -98,14 +98,14 @@ static void mv_cesa_ablkcipher_std_step(struct ablkcipher_request *req) /* FIXME: only update enc_len field */ if (!sreq->skip_ctx) { - memcpy(engine->sram, &sreq->op, sizeof(sreq->op)); + memcpy_toio(engine->sram, &sreq->op, sizeof(sreq->op)); sreq->skip_ctx = true; } else { - memcpy(engine->sram, &sreq->op, sizeof(sreq->op.desc)); + memcpy_toio(engine->sram, &sreq->op, sizeof(sreq->op.desc)); } mv_cesa_set_int_mask(engine, CESA_SA_INT_ACCEL0_DONE); - writel(CESA_SA_CFG_PARA_DIS, engine->regs + CESA_SA_CFG); + writel_relaxed(CESA_SA_CFG_PARA_DIS, engine->regs + CESA_SA_CFG); writel(CESA_SA_CMD_EN_CESA_SA_ACCL0, engine->regs + CESA_SA_CMD); } @@ -145,8 +145,9 @@ static int mv_cesa_ablkcipher_process(struct crypto_async_request *req, if (ret) return ret; - memcpy(ablkreq->info, engine->sram + CESA_SA_CRYPT_IV_SRAM_OFFSET, - crypto_ablkcipher_ivsize(crypto_ablkcipher_reqtfm(ablkreq))); + memcpy_fromio(ablkreq->info, + engine->sram + CESA_SA_CRYPT_IV_SRAM_OFFSET, + crypto_ablkcipher_ivsize(crypto_ablkcipher_reqtfm(ablkreq))); return 0; } @@ -181,7 +182,7 @@ mv_cesa_ablkcipher_std_prepare(struct ablkcipher_request *req) sreq->size = 0; sreq->offset = 0; mv_cesa_adjust_op(engine, &sreq->op); - memcpy(engine->sram, &sreq->op, sizeof(sreq->op)); + memcpy_toio(engine->sram, &sreq->op, sizeof(sreq->op)); } static inline void mv_cesa_ablkcipher_prepare(struct crypto_async_request *req, diff --git a/drivers/crypto/marvell/hash.c b/drivers/crypto/marvell/hash.c index e8d0d7128137..6ec55b4a087b 100644 --- a/drivers/crypto/marvell/hash.c +++ b/drivers/crypto/marvell/hash.c @@ -27,10 +27,10 @@ mv_cesa_ahash_req_iter_init(struct mv_cesa_ahash_dma_iter *iter, struct ahash_request *req) { struct mv_cesa_ahash_req *creq = ahash_request_ctx(req); - unsigned int len = req->nbytes; + unsigned int len = req->nbytes + creq->cache_ptr; if (!creq->last_req) - len = (len + creq->cache_ptr) & ~CESA_HASH_BLOCK_SIZE_MSK; + len &= ~CESA_HASH_BLOCK_SIZE_MSK; mv_cesa_req_dma_iter_init(&iter->base, len); mv_cesa_sg_dma_iter_init(&iter->src, req->src, DMA_TO_DEVICE); @@ -179,7 +179,6 @@ static int mv_cesa_ahash_pad_len(struct mv_cesa_ahash_req *creq) static int mv_cesa_ahash_pad_req(struct mv_cesa_ahash_req *creq, u8 *buf) { - __be64 bits = cpu_to_be64(creq->len << 3); unsigned int index, padlen; buf[0] = 0x80; @@ -187,7 +186,14 @@ static int mv_cesa_ahash_pad_req(struct mv_cesa_ahash_req *creq, u8 *buf) index = creq->len & CESA_HASH_BLOCK_SIZE_MSK; padlen = mv_cesa_ahash_pad_len(creq); memset(buf + 1, 0, padlen - 1); - memcpy(buf + padlen, &bits, sizeof(bits)); + + if (creq->algo_le) { + __le64 bits = cpu_to_le64(creq->len << 3); + memcpy(buf + padlen, &bits, sizeof(bits)); + } else { + __be64 bits = cpu_to_be64(creq->len << 3); + memcpy(buf + padlen, &bits, sizeof(bits)); + } return padlen + 8; } @@ -203,8 +209,8 @@ static void mv_cesa_ahash_std_step(struct ahash_request *req) size_t len; if (creq->cache_ptr) - memcpy(engine->sram + CESA_SA_DATA_SRAM_OFFSET, creq->cache, - creq->cache_ptr); + memcpy_toio(engine->sram + CESA_SA_DATA_SRAM_OFFSET, + creq->cache, creq->cache_ptr); len = min_t(size_t, req->nbytes + creq->cache_ptr - sreq->offset, CESA_SA_SRAM_PAYLOAD_SIZE); @@ -245,10 +251,10 @@ static void mv_cesa_ahash_std_step(struct ahash_request *req) if (len + trailerlen > CESA_SA_SRAM_PAYLOAD_SIZE) { len &= CESA_HASH_BLOCK_SIZE_MSK; new_cache_ptr = 64 - trailerlen; - memcpy(creq->cache, - engine->sram + - CESA_SA_DATA_SRAM_OFFSET + len, - new_cache_ptr); + memcpy_fromio(creq->cache, + engine->sram + + CESA_SA_DATA_SRAM_OFFSET + len, + new_cache_ptr); } else { len += mv_cesa_ahash_pad_req(creq, engine->sram + len + @@ -266,7 +272,7 @@ static void mv_cesa_ahash_std_step(struct ahash_request *req) mv_cesa_update_op_cfg(op, frag_mode, CESA_SA_DESC_CFG_FRAG_MSK); /* FIXME: only update enc_len field */ - memcpy(engine->sram, op, sizeof(*op)); + memcpy_toio(engine->sram, op, sizeof(*op)); if (frag_mode == CESA_SA_DESC_CFG_FIRST_FRAG) mv_cesa_update_op_cfg(op, CESA_SA_DESC_CFG_MID_FRAG, @@ -275,7 +281,7 @@ static void mv_cesa_ahash_std_step(struct ahash_request *req) creq->cache_ptr = new_cache_ptr; mv_cesa_set_int_mask(engine, CESA_SA_INT_ACCEL0_DONE); - writel(CESA_SA_CFG_PARA_DIS, engine->regs + CESA_SA_CFG); + writel_relaxed(CESA_SA_CFG_PARA_DIS, engine->regs + CESA_SA_CFG); writel(CESA_SA_CMD_EN_CESA_SA_ACCL0, engine->regs + CESA_SA_CMD); } @@ -306,7 +312,7 @@ static void mv_cesa_ahash_std_prepare(struct ahash_request *req) sreq->offset = 0; mv_cesa_adjust_op(engine, &creq->op_tmpl); - memcpy(engine->sram, &creq->op_tmpl, sizeof(creq->op_tmpl)); + memcpy_toio(engine->sram, &creq->op_tmpl, sizeof(creq->op_tmpl)); } static void mv_cesa_ahash_step(struct crypto_async_request *req) @@ -338,7 +344,7 @@ static int mv_cesa_ahash_process(struct crypto_async_request *req, u32 status) digsize = crypto_ahash_digestsize(crypto_ahash_reqtfm(ahashreq)); for (i = 0; i < digsize / 4; i++) - creq->state[i] = readl(engine->regs + CESA_IVDIG(i)); + creq->state[i] = readl_relaxed(engine->regs + CESA_IVDIG(i)); if (creq->cache_ptr) sg_pcopy_to_buffer(ahashreq->src, creq->src_nents, @@ -347,18 +353,21 @@ static int mv_cesa_ahash_process(struct crypto_async_request *req, u32 status) ahashreq->nbytes - creq->cache_ptr); if (creq->last_req) { - for (i = 0; i < digsize / 4; i++) { - /* - * Hardware provides MD5 digest in a different - * endianness than SHA-1 and SHA-256 ones. - */ - if (digsize == MD5_DIGEST_SIZE) - creq->state[i] = cpu_to_le32(creq->state[i]); - else - creq->state[i] = cpu_to_be32(creq->state[i]); - } + /* + * Hardware's MD5 digest is in little endian format, but + * SHA in big endian format + */ + if (creq->algo_le) { + __le32 *result = (void *)ahashreq->result; + + for (i = 0; i < digsize / 4; i++) + result[i] = cpu_to_le32(creq->state[i]); + } else { + __be32 *result = (void *)ahashreq->result; - memcpy(ahashreq->result, creq->state, digsize); + for (i = 0; i < digsize / 4; i++) + result[i] = cpu_to_be32(creq->state[i]); + } } return ret; @@ -381,8 +390,7 @@ static void mv_cesa_ahash_prepare(struct crypto_async_request *req, digsize = crypto_ahash_digestsize(crypto_ahash_reqtfm(ahashreq)); for (i = 0; i < digsize / 4; i++) - writel(creq->state[i], - engine->regs + CESA_IVDIG(i)); + writel_relaxed(creq->state[i], engine->regs + CESA_IVDIG(i)); } static void mv_cesa_ahash_req_cleanup(struct crypto_async_request *req) @@ -404,7 +412,7 @@ static const struct mv_cesa_req_ops mv_cesa_ahash_req_ops = { }; static int mv_cesa_ahash_init(struct ahash_request *req, - struct mv_cesa_op_ctx *tmpl) + struct mv_cesa_op_ctx *tmpl, bool algo_le) { struct mv_cesa_ahash_req *creq = ahash_request_ctx(req); @@ -418,6 +426,7 @@ static int mv_cesa_ahash_init(struct ahash_request *req, mv_cesa_set_mac_op_frag_len(tmpl, 0); creq->op_tmpl = *tmpl; creq->len = 0; + creq->algo_le = algo_le; return 0; } @@ -462,145 +471,114 @@ static int mv_cesa_ahash_cache_req(struct ahash_request *req, bool *cached) } static struct mv_cesa_op_ctx * -mv_cesa_ahash_dma_add_cache(struct mv_cesa_tdma_chain *chain, - struct mv_cesa_ahash_dma_iter *dma_iter, - struct mv_cesa_ahash_req *creq, - gfp_t flags) +mv_cesa_dma_add_frag(struct mv_cesa_tdma_chain *chain, + struct mv_cesa_op_ctx *tmpl, unsigned int frag_len, + gfp_t flags) { - struct mv_cesa_ahash_dma_req *ahashdreq = &creq->req.dma; - struct mv_cesa_op_ctx *op = NULL; + struct mv_cesa_op_ctx *op; int ret; - if (!creq->cache_ptr) - return NULL; + op = mv_cesa_dma_add_op(chain, tmpl, false, flags); + if (IS_ERR(op)) + return op; - ret = mv_cesa_dma_add_data_transfer(chain, - CESA_SA_DATA_SRAM_OFFSET, - ahashdreq->cache_dma, - creq->cache_ptr, - CESA_TDMA_DST_IN_SRAM, - flags); + /* Set the operation block fragment length. */ + mv_cesa_set_mac_op_frag_len(op, frag_len); + + /* Append dummy desc to launch operation */ + ret = mv_cesa_dma_add_dummy_launch(chain, flags); if (ret) return ERR_PTR(ret); - if (!dma_iter->base.op_len) { - op = mv_cesa_dma_add_op(chain, &creq->op_tmpl, false, flags); - if (IS_ERR(op)) - return op; - - mv_cesa_set_mac_op_frag_len(op, creq->cache_ptr); - - /* Add dummy desc to launch crypto operation */ - ret = mv_cesa_dma_add_dummy_launch(chain, flags); - if (ret) - return ERR_PTR(ret); - } + if (mv_cesa_mac_op_is_first_frag(tmpl)) + mv_cesa_update_op_cfg(tmpl, + CESA_SA_DESC_CFG_MID_FRAG, + CESA_SA_DESC_CFG_FRAG_MSK); return op; } -static struct mv_cesa_op_ctx * -mv_cesa_ahash_dma_add_data(struct mv_cesa_tdma_chain *chain, - struct mv_cesa_ahash_dma_iter *dma_iter, - struct mv_cesa_ahash_req *creq, - gfp_t flags) +static int +mv_cesa_ahash_dma_add_cache(struct mv_cesa_tdma_chain *chain, + struct mv_cesa_ahash_dma_iter *dma_iter, + struct mv_cesa_ahash_req *creq, + gfp_t flags) { - struct mv_cesa_op_ctx *op; - int ret; - - op = mv_cesa_dma_add_op(chain, &creq->op_tmpl, false, flags); - if (IS_ERR(op)) - return op; - - mv_cesa_set_mac_op_frag_len(op, dma_iter->base.op_len); - - if ((mv_cesa_get_op_cfg(&creq->op_tmpl) & CESA_SA_DESC_CFG_FRAG_MSK) == - CESA_SA_DESC_CFG_FIRST_FRAG) - mv_cesa_update_op_cfg(&creq->op_tmpl, - CESA_SA_DESC_CFG_MID_FRAG, - CESA_SA_DESC_CFG_FRAG_MSK); - - /* Add input transfers */ - ret = mv_cesa_dma_add_op_transfers(chain, &dma_iter->base, - &dma_iter->src, flags); - if (ret) - return ERR_PTR(ret); + struct mv_cesa_ahash_dma_req *ahashdreq = &creq->req.dma; - /* Add dummy desc to launch crypto operation */ - ret = mv_cesa_dma_add_dummy_launch(chain, flags); - if (ret) - return ERR_PTR(ret); + if (!creq->cache_ptr) + return 0; - return op; + return mv_cesa_dma_add_data_transfer(chain, + CESA_SA_DATA_SRAM_OFFSET, + ahashdreq->cache_dma, + creq->cache_ptr, + CESA_TDMA_DST_IN_SRAM, + flags); } static struct mv_cesa_op_ctx * mv_cesa_ahash_dma_last_req(struct mv_cesa_tdma_chain *chain, struct mv_cesa_ahash_dma_iter *dma_iter, struct mv_cesa_ahash_req *creq, - struct mv_cesa_op_ctx *op, - gfp_t flags) + unsigned int frag_len, gfp_t flags) { struct mv_cesa_ahash_dma_req *ahashdreq = &creq->req.dma; unsigned int len, trailerlen, padoff = 0; + struct mv_cesa_op_ctx *op; int ret; - if (!creq->last_req) - return op; - - if (op && creq->len <= CESA_SA_DESC_MAC_SRC_TOTAL_LEN_MAX) { - u32 frag = CESA_SA_DESC_CFG_NOT_FRAG; - - if ((mv_cesa_get_op_cfg(op) & CESA_SA_DESC_CFG_FRAG_MSK) != - CESA_SA_DESC_CFG_FIRST_FRAG) - frag = CESA_SA_DESC_CFG_LAST_FRAG; + /* + * If the transfer is smaller than our maximum length, and we have + * some data outstanding, we can ask the engine to finish the hash. + */ + if (creq->len <= CESA_SA_DESC_MAC_SRC_TOTAL_LEN_MAX && frag_len) { + op = mv_cesa_dma_add_frag(chain, &creq->op_tmpl, frag_len, + flags); + if (IS_ERR(op)) + return op; - mv_cesa_update_op_cfg(op, frag, CESA_SA_DESC_CFG_FRAG_MSK); + mv_cesa_set_mac_op_total_len(op, creq->len); + mv_cesa_update_op_cfg(op, mv_cesa_mac_op_is_first_frag(op) ? + CESA_SA_DESC_CFG_NOT_FRAG : + CESA_SA_DESC_CFG_LAST_FRAG, + CESA_SA_DESC_CFG_FRAG_MSK); return op; } + /* + * The request is longer than the engine can handle, or we have + * no data outstanding. Manually generate the padding, adding it + * as a "mid" fragment. + */ ret = mv_cesa_ahash_dma_alloc_padding(ahashdreq, flags); if (ret) return ERR_PTR(ret); trailerlen = mv_cesa_ahash_pad_req(creq, ahashdreq->padding); - if (op) { - len = min(CESA_SA_SRAM_PAYLOAD_SIZE - dma_iter->base.op_len, - trailerlen); - if (len) { - ret = mv_cesa_dma_add_data_transfer(chain, + len = min(CESA_SA_SRAM_PAYLOAD_SIZE - frag_len, trailerlen); + if (len) { + ret = mv_cesa_dma_add_data_transfer(chain, CESA_SA_DATA_SRAM_OFFSET + - dma_iter->base.op_len, + frag_len, ahashdreq->padding_dma, len, CESA_TDMA_DST_IN_SRAM, flags); - if (ret) - return ERR_PTR(ret); - - mv_cesa_update_op_cfg(op, CESA_SA_DESC_CFG_MID_FRAG, - CESA_SA_DESC_CFG_FRAG_MSK); - mv_cesa_set_mac_op_frag_len(op, - dma_iter->base.op_len + len); - padoff += len; - } - } - - if (padoff >= trailerlen) - return op; + if (ret) + return ERR_PTR(ret); - if ((mv_cesa_get_op_cfg(&creq->op_tmpl) & CESA_SA_DESC_CFG_FRAG_MSK) != - CESA_SA_DESC_CFG_FIRST_FRAG) - mv_cesa_update_op_cfg(&creq->op_tmpl, - CESA_SA_DESC_CFG_MID_FRAG, - CESA_SA_DESC_CFG_FRAG_MSK); + op = mv_cesa_dma_add_frag(chain, &creq->op_tmpl, frag_len + len, + flags); + if (IS_ERR(op)) + return op; - op = mv_cesa_dma_add_op(chain, &creq->op_tmpl, false, flags); - if (IS_ERR(op)) - return op; + if (len == trailerlen) + return op; - mv_cesa_set_mac_op_frag_len(op, trailerlen - padoff); + padoff += len; + } ret = mv_cesa_dma_add_data_transfer(chain, CESA_SA_DATA_SRAM_OFFSET, @@ -612,12 +590,8 @@ mv_cesa_ahash_dma_last_req(struct mv_cesa_tdma_chain *chain, if (ret) return ERR_PTR(ret); - /* Add dummy desc to launch crypto operation */ - ret = mv_cesa_dma_add_dummy_launch(chain, flags); - if (ret) - return ERR_PTR(ret); - - return op; + return mv_cesa_dma_add_frag(chain, &creq->op_tmpl, trailerlen - padoff, + flags); } static int mv_cesa_ahash_dma_req_init(struct ahash_request *req) @@ -627,9 +601,9 @@ static int mv_cesa_ahash_dma_req_init(struct ahash_request *req) GFP_KERNEL : GFP_ATOMIC; struct mv_cesa_ahash_dma_req *ahashdreq = &creq->req.dma; struct mv_cesa_tdma_req *dreq = &ahashdreq->base; - struct mv_cesa_tdma_chain chain; struct mv_cesa_ahash_dma_iter iter; struct mv_cesa_op_ctx *op = NULL; + unsigned int frag_len; int ret; dreq->chain.first = NULL; @@ -644,29 +618,59 @@ static int mv_cesa_ahash_dma_req_init(struct ahash_request *req) } } - mv_cesa_tdma_desc_iter_init(&chain); + mv_cesa_tdma_desc_iter_init(&dreq->chain); mv_cesa_ahash_req_iter_init(&iter, req); - op = mv_cesa_ahash_dma_add_cache(&chain, &iter, - creq, flags); - if (IS_ERR(op)) { - ret = PTR_ERR(op); + /* + * Add the cache (left-over data from a previous block) first. + * This will never overflow the SRAM size. + */ + ret = mv_cesa_ahash_dma_add_cache(&dreq->chain, &iter, creq, flags); + if (ret) goto err_free_tdma; - } - do { - if (!iter.base.op_len) - break; + if (iter.src.sg) { + /* + * Add all the new data, inserting an operation block and + * launch command between each full SRAM block-worth of + * data. We intentionally do not add the final op block. + */ + while (true) { + ret = mv_cesa_dma_add_op_transfers(&dreq->chain, + &iter.base, + &iter.src, flags); + if (ret) + goto err_free_tdma; + + frag_len = iter.base.op_len; - op = mv_cesa_ahash_dma_add_data(&chain, &iter, - creq, flags); - if (IS_ERR(op)) { - ret = PTR_ERR(op); - goto err_free_tdma; + if (!mv_cesa_ahash_req_iter_next_op(&iter)) + break; + + op = mv_cesa_dma_add_frag(&dreq->chain, &creq->op_tmpl, + frag_len, flags); + if (IS_ERR(op)) { + ret = PTR_ERR(op); + goto err_free_tdma; + } } - } while (mv_cesa_ahash_req_iter_next_op(&iter)); + } else { + /* Account for the data that was in the cache. */ + frag_len = iter.base.op_len; + } + + /* + * At this point, frag_len indicates whether we have any data + * outstanding which needs an operation. Queue up the final + * operation, which depends whether this is the final request. + */ + if (creq->last_req) + op = mv_cesa_ahash_dma_last_req(&dreq->chain, &iter, creq, + frag_len, flags); + else if (frag_len) + op = mv_cesa_dma_add_frag(&dreq->chain, &creq->op_tmpl, + frag_len, flags); - op = mv_cesa_ahash_dma_last_req(&chain, &iter, creq, op, flags); if (IS_ERR(op)) { ret = PTR_ERR(op); goto err_free_tdma; @@ -674,7 +678,7 @@ static int mv_cesa_ahash_dma_req_init(struct ahash_request *req) if (op) { /* Add dummy desc to wait for crypto operation end */ - ret = mv_cesa_dma_add_dummy_end(&chain, flags); + ret = mv_cesa_dma_add_dummy_end(&dreq->chain, flags); if (ret) goto err_free_tdma; } @@ -685,8 +689,6 @@ static int mv_cesa_ahash_dma_req_init(struct ahash_request *req) else creq->cache_ptr = 0; - dreq->chain = chain; - return 0; err_free_tdma: @@ -795,47 +797,50 @@ static int mv_cesa_ahash_finup(struct ahash_request *req) return ret; } -static int mv_cesa_md5_init(struct ahash_request *req) +static int mv_cesa_ahash_export(struct ahash_request *req, void *hash, + u64 *len, void *cache) { - struct mv_cesa_op_ctx tmpl; - - mv_cesa_set_op_cfg(&tmpl, CESA_SA_DESC_CFG_MACM_MD5); - - mv_cesa_ahash_init(req, &tmpl); - - return 0; -} - -static int mv_cesa_md5_export(struct ahash_request *req, void *out) -{ - struct md5_state *out_state = out; struct crypto_ahash *ahash = crypto_ahash_reqtfm(req); struct mv_cesa_ahash_req *creq = ahash_request_ctx(req); unsigned int digsize = crypto_ahash_digestsize(ahash); + unsigned int blocksize; + + blocksize = crypto_ahash_blocksize(ahash); - out_state->byte_count = creq->len; - memcpy(out_state->hash, creq->state, digsize); - memset(out_state->block, 0, sizeof(out_state->block)); + *len = creq->len; + memcpy(hash, creq->state, digsize); + memset(cache, 0, blocksize); if (creq->cache) - memcpy(out_state->block, creq->cache, creq->cache_ptr); + memcpy(cache, creq->cache, creq->cache_ptr); return 0; } -static int mv_cesa_md5_import(struct ahash_request *req, const void *in) +static int mv_cesa_ahash_import(struct ahash_request *req, const void *hash, + u64 len, const void *cache) { - const struct md5_state *in_state = in; struct crypto_ahash *ahash = crypto_ahash_reqtfm(req); struct mv_cesa_ahash_req *creq = ahash_request_ctx(req); unsigned int digsize = crypto_ahash_digestsize(ahash); + unsigned int blocksize; unsigned int cache_ptr; int ret; - creq->len = in_state->byte_count; - memcpy(creq->state, in_state->hash, digsize); + ret = crypto_ahash_init(req); + if (ret) + return ret; + + blocksize = crypto_ahash_blocksize(ahash); + if (len >= blocksize) + mv_cesa_update_op_cfg(&creq->op_tmpl, + CESA_SA_DESC_CFG_MID_FRAG, + CESA_SA_DESC_CFG_FRAG_MSK); + + creq->len = len; + memcpy(creq->state, hash, digsize); creq->cache_ptr = 0; - cache_ptr = creq->len % sizeof(in_state->block); + cache_ptr = do_div(len, blocksize); if (!cache_ptr) return 0; @@ -843,12 +848,39 @@ static int mv_cesa_md5_import(struct ahash_request *req, const void *in) if (ret) return ret; - memcpy(creq->cache, in_state->block, cache_ptr); + memcpy(creq->cache, cache, cache_ptr); creq->cache_ptr = cache_ptr; return 0; } +static int mv_cesa_md5_init(struct ahash_request *req) +{ + struct mv_cesa_op_ctx tmpl = { }; + + mv_cesa_set_op_cfg(&tmpl, CESA_SA_DESC_CFG_MACM_MD5); + + mv_cesa_ahash_init(req, &tmpl, true); + + return 0; +} + +static int mv_cesa_md5_export(struct ahash_request *req, void *out) +{ + struct md5_state *out_state = out; + + return mv_cesa_ahash_export(req, out_state->hash, + &out_state->byte_count, out_state->block); +} + +static int mv_cesa_md5_import(struct ahash_request *req, const void *in) +{ + const struct md5_state *in_state = in; + + return mv_cesa_ahash_import(req, in_state->hash, in_state->byte_count, + in_state->block); +} + static int mv_cesa_md5_digest(struct ahash_request *req) { int ret; @@ -870,6 +902,7 @@ struct ahash_alg mv_md5_alg = { .import = mv_cesa_md5_import, .halg = { .digestsize = MD5_DIGEST_SIZE, + .statesize = sizeof(struct md5_state), .base = { .cra_name = "md5", .cra_driver_name = "mv-md5", @@ -886,11 +919,11 @@ struct ahash_alg mv_md5_alg = { static int mv_cesa_sha1_init(struct ahash_request *req) { - struct mv_cesa_op_ctx tmpl; + struct mv_cesa_op_ctx tmpl = { }; mv_cesa_set_op_cfg(&tmpl, CESA_SA_DESC_CFG_MACM_SHA1); - mv_cesa_ahash_init(req, &tmpl); + mv_cesa_ahash_init(req, &tmpl, false); return 0; } @@ -898,44 +931,17 @@ static int mv_cesa_sha1_init(struct ahash_request *req) static int mv_cesa_sha1_export(struct ahash_request *req, void *out) { struct sha1_state *out_state = out; - struct crypto_ahash *ahash = crypto_ahash_reqtfm(req); - struct mv_cesa_ahash_req *creq = ahash_request_ctx(req); - unsigned int digsize = crypto_ahash_digestsize(ahash); - - out_state->count = creq->len; - memcpy(out_state->state, creq->state, digsize); - memset(out_state->buffer, 0, sizeof(out_state->buffer)); - if (creq->cache) - memcpy(out_state->buffer, creq->cache, creq->cache_ptr); - return 0; + return mv_cesa_ahash_export(req, out_state->state, &out_state->count, + out_state->buffer); } static int mv_cesa_sha1_import(struct ahash_request *req, const void *in) { const struct sha1_state *in_state = in; - struct crypto_ahash *ahash = crypto_ahash_reqtfm(req); - struct mv_cesa_ahash_req *creq = ahash_request_ctx(req); - unsigned int digsize = crypto_ahash_digestsize(ahash); - unsigned int cache_ptr; - int ret; - creq->len = in_state->count; - memcpy(creq->state, in_state->state, digsize); - creq->cache_ptr = 0; - - cache_ptr = creq->len % SHA1_BLOCK_SIZE; - if (!cache_ptr) - return 0; - - ret = mv_cesa_ahash_alloc_cache(req); - if (ret) - return ret; - - memcpy(creq->cache, in_state->buffer, cache_ptr); - creq->cache_ptr = cache_ptr; - - return 0; + return mv_cesa_ahash_import(req, in_state->state, in_state->count, + in_state->buffer); } static int mv_cesa_sha1_digest(struct ahash_request *req) @@ -959,6 +965,7 @@ struct ahash_alg mv_sha1_alg = { .import = mv_cesa_sha1_import, .halg = { .digestsize = SHA1_DIGEST_SIZE, + .statesize = sizeof(struct sha1_state), .base = { .cra_name = "sha1", .cra_driver_name = "mv-sha1", @@ -975,11 +982,11 @@ struct ahash_alg mv_sha1_alg = { static int mv_cesa_sha256_init(struct ahash_request *req) { - struct mv_cesa_op_ctx tmpl; + struct mv_cesa_op_ctx tmpl = { }; mv_cesa_set_op_cfg(&tmpl, CESA_SA_DESC_CFG_MACM_SHA256); - mv_cesa_ahash_init(req, &tmpl); + mv_cesa_ahash_init(req, &tmpl, false); return 0; } @@ -998,44 +1005,17 @@ static int mv_cesa_sha256_digest(struct ahash_request *req) static int mv_cesa_sha256_export(struct ahash_request *req, void *out) { struct sha256_state *out_state = out; - struct crypto_ahash *ahash = crypto_ahash_reqtfm(req); - struct mv_cesa_ahash_req *creq = ahash_request_ctx(req); - unsigned int ds = crypto_ahash_digestsize(ahash); - out_state->count = creq->len; - memcpy(out_state->state, creq->state, ds); - memset(out_state->buf, 0, sizeof(out_state->buf)); - if (creq->cache) - memcpy(out_state->buf, creq->cache, creq->cache_ptr); - - return 0; + return mv_cesa_ahash_export(req, out_state->state, &out_state->count, + out_state->buf); } static int mv_cesa_sha256_import(struct ahash_request *req, const void *in) { const struct sha256_state *in_state = in; - struct crypto_ahash *ahash = crypto_ahash_reqtfm(req); - struct mv_cesa_ahash_req *creq = ahash_request_ctx(req); - unsigned int digsize = crypto_ahash_digestsize(ahash); - unsigned int cache_ptr; - int ret; - - creq->len = in_state->count; - memcpy(creq->state, in_state->state, digsize); - creq->cache_ptr = 0; - - cache_ptr = creq->len % SHA256_BLOCK_SIZE; - if (!cache_ptr) - return 0; - - ret = mv_cesa_ahash_alloc_cache(req); - if (ret) - return ret; - - memcpy(creq->cache, in_state->buf, cache_ptr); - creq->cache_ptr = cache_ptr; - return 0; + return mv_cesa_ahash_import(req, in_state->state, in_state->count, + in_state->buf); } struct ahash_alg mv_sha256_alg = { @@ -1048,6 +1028,7 @@ struct ahash_alg mv_sha256_alg = { .import = mv_cesa_sha256_import, .halg = { .digestsize = SHA256_DIGEST_SIZE, + .statesize = sizeof(struct sha256_state), .base = { .cra_name = "sha256", .cra_driver_name = "mv-sha256", @@ -1231,12 +1212,12 @@ static int mv_cesa_ahmac_cra_init(struct crypto_tfm *tfm) static int mv_cesa_ahmac_md5_init(struct ahash_request *req) { struct mv_cesa_hmac_ctx *ctx = crypto_tfm_ctx(req->base.tfm); - struct mv_cesa_op_ctx tmpl; + struct mv_cesa_op_ctx tmpl = { }; mv_cesa_set_op_cfg(&tmpl, CESA_SA_DESC_CFG_MACM_HMAC_MD5); memcpy(tmpl.ctx.hash.iv, ctx->iv, sizeof(ctx->iv)); - mv_cesa_ahash_init(req, &tmpl); + mv_cesa_ahash_init(req, &tmpl, true); return 0; } @@ -1301,12 +1282,12 @@ struct ahash_alg mv_ahmac_md5_alg = { static int mv_cesa_ahmac_sha1_init(struct ahash_request *req) { struct mv_cesa_hmac_ctx *ctx = crypto_tfm_ctx(req->base.tfm); - struct mv_cesa_op_ctx tmpl; + struct mv_cesa_op_ctx tmpl = { }; mv_cesa_set_op_cfg(&tmpl, CESA_SA_DESC_CFG_MACM_HMAC_SHA1); memcpy(tmpl.ctx.hash.iv, ctx->iv, sizeof(ctx->iv)); - mv_cesa_ahash_init(req, &tmpl); + mv_cesa_ahash_init(req, &tmpl, false); return 0; } @@ -1391,12 +1372,12 @@ static int mv_cesa_ahmac_sha256_setkey(struct crypto_ahash *tfm, const u8 *key, static int mv_cesa_ahmac_sha256_init(struct ahash_request *req) { struct mv_cesa_hmac_ctx *ctx = crypto_tfm_ctx(req->base.tfm); - struct mv_cesa_op_ctx tmpl; + struct mv_cesa_op_ctx tmpl = { }; mv_cesa_set_op_cfg(&tmpl, CESA_SA_DESC_CFG_MACM_HMAC_SHA256); memcpy(tmpl.ctx.hash.iv, ctx->iv, sizeof(ctx->iv)); - mv_cesa_ahash_init(req, &tmpl); + mv_cesa_ahash_init(req, &tmpl, false); return 0; } diff --git a/drivers/crypto/marvell/tdma.c b/drivers/crypto/marvell/tdma.c index 64a366c50174..76427981275b 100644 --- a/drivers/crypto/marvell/tdma.c +++ b/drivers/crypto/marvell/tdma.c @@ -41,18 +41,18 @@ void mv_cesa_dma_step(struct mv_cesa_tdma_req *dreq) { struct mv_cesa_engine *engine = dreq->base.engine; - writel(0, engine->regs + CESA_SA_CFG); + writel_relaxed(0, engine->regs + CESA_SA_CFG); mv_cesa_set_int_mask(engine, CESA_SA_INT_ACC0_IDMA_DONE); - writel(CESA_TDMA_DST_BURST_128B | CESA_TDMA_SRC_BURST_128B | - CESA_TDMA_NO_BYTE_SWAP | CESA_TDMA_EN, - engine->regs + CESA_TDMA_CONTROL); - - writel(CESA_SA_CFG_ACT_CH0_IDMA | CESA_SA_CFG_MULTI_PKT | - CESA_SA_CFG_CH0_W_IDMA | CESA_SA_CFG_PARA_DIS, - engine->regs + CESA_SA_CFG); - writel(dreq->chain.first->cur_dma, - engine->regs + CESA_TDMA_NEXT_ADDR); + writel_relaxed(CESA_TDMA_DST_BURST_128B | CESA_TDMA_SRC_BURST_128B | + CESA_TDMA_NO_BYTE_SWAP | CESA_TDMA_EN, + engine->regs + CESA_TDMA_CONTROL); + + writel_relaxed(CESA_SA_CFG_ACT_CH0_IDMA | CESA_SA_CFG_MULTI_PKT | + CESA_SA_CFG_CH0_W_IDMA | CESA_SA_CFG_PARA_DIS, + engine->regs + CESA_SA_CFG); + writel_relaxed(dreq->chain.first->cur_dma, + engine->regs + CESA_TDMA_NEXT_ADDR); writel(CESA_SA_CMD_EN_CESA_SA_ACCL0, engine->regs + CESA_SA_CMD); } @@ -69,7 +69,7 @@ void mv_cesa_dma_cleanup(struct mv_cesa_tdma_req *dreq) tdma = tdma->next; dma_pool_free(cesa_dev->dma->tdma_desc_pool, old_tdma, - le32_to_cpu(old_tdma->cur_dma)); + old_tdma->cur_dma); } dreq->chain.first = NULL; @@ -105,9 +105,9 @@ mv_cesa_dma_add_desc(struct mv_cesa_tdma_chain *chain, gfp_t flags) return ERR_PTR(-ENOMEM); memset(new_tdma, 0, sizeof(*new_tdma)); - new_tdma->cur_dma = cpu_to_le32(dma_handle); + new_tdma->cur_dma = dma_handle; if (chain->last) { - chain->last->next_dma = new_tdma->cur_dma; + chain->last->next_dma = cpu_to_le32(dma_handle); chain->last->next = new_tdma; } else { chain->first = new_tdma; @@ -126,6 +126,7 @@ struct mv_cesa_op_ctx *mv_cesa_dma_add_op(struct mv_cesa_tdma_chain *chain, struct mv_cesa_tdma_desc *tdma; struct mv_cesa_op_ctx *op; dma_addr_t dma_handle; + unsigned int size; tdma = mv_cesa_dma_add_desc(chain, flags); if (IS_ERR(tdma)) @@ -137,10 +138,12 @@ struct mv_cesa_op_ctx *mv_cesa_dma_add_op(struct mv_cesa_tdma_chain *chain, *op = *op_templ; + size = skip_ctx ? sizeof(op->desc) : sizeof(*op); + tdma = chain->last; tdma->op = op; - tdma->byte_cnt = (skip_ctx ? sizeof(op->desc) : sizeof(*op)) | BIT(31); - tdma->src = dma_handle; + tdma->byte_cnt = cpu_to_le32(size | BIT(31)); + tdma->src = cpu_to_le32(dma_handle); tdma->flags = CESA_TDMA_DST_IN_SRAM | CESA_TDMA_OP; return op; @@ -156,7 +159,7 @@ int mv_cesa_dma_add_data_transfer(struct mv_cesa_tdma_chain *chain, if (IS_ERR(tdma)) return PTR_ERR(tdma); - tdma->byte_cnt = size | BIT(31); + tdma->byte_cnt = cpu_to_le32(size | BIT(31)); tdma->src = src; tdma->dst = dst; @@ -166,8 +169,7 @@ int mv_cesa_dma_add_data_transfer(struct mv_cesa_tdma_chain *chain, return 0; } -int mv_cesa_dma_add_dummy_launch(struct mv_cesa_tdma_chain *chain, - u32 flags) +int mv_cesa_dma_add_dummy_launch(struct mv_cesa_tdma_chain *chain, gfp_t flags) { struct mv_cesa_tdma_desc *tdma; @@ -178,7 +180,7 @@ int mv_cesa_dma_add_dummy_launch(struct mv_cesa_tdma_chain *chain, return 0; } -int mv_cesa_dma_add_dummy_end(struct mv_cesa_tdma_chain *chain, u32 flags) +int mv_cesa_dma_add_dummy_end(struct mv_cesa_tdma_chain *chain, gfp_t flags) { struct mv_cesa_tdma_desc *tdma; @@ -186,7 +188,7 @@ int mv_cesa_dma_add_dummy_end(struct mv_cesa_tdma_chain *chain, u32 flags) if (IS_ERR(tdma)) return PTR_ERR(tdma); - tdma->byte_cnt = BIT(31); + tdma->byte_cnt = cpu_to_le32(BIT(31)); return 0; } diff --git a/drivers/crypto/n2_core.c b/drivers/crypto/n2_core.c index 2e8dab9d4263..5450880abb7b 100644 --- a/drivers/crypto/n2_core.c +++ b/drivers/crypto/n2_core.c @@ -34,7 +34,7 @@ #define DRV_MODULE_VERSION "0.2" #define DRV_MODULE_RELDATE "July 28, 2011" -static char version[] = +static const char version[] = DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n"; MODULE_AUTHOR("David S. Miller (davem@davemloft.net)"); diff --git a/drivers/crypto/nx/nx-842-powernv.c b/drivers/crypto/nx/nx-842-powernv.c index 3750e13d8721..9ef51fafdbff 100644 --- a/drivers/crypto/nx/nx-842-powernv.c +++ b/drivers/crypto/nx/nx-842-powernv.c @@ -491,7 +491,7 @@ static int nx842_powernv_compress(const unsigned char *in, unsigned int inlen, void *wmem) { return nx842_powernv_function(in, inlen, out, outlenp, - wmem, CCW_FC_842_COMP_NOCRC); + wmem, CCW_FC_842_COMP_CRC); } /** @@ -519,7 +519,7 @@ static int nx842_powernv_decompress(const unsigned char *in, unsigned int inlen, void *wmem) { return nx842_powernv_function(in, inlen, out, outlenp, - wmem, CCW_FC_842_DECOMP_NOCRC); + wmem, CCW_FC_842_DECOMP_CRC); } static int __init nx842_powernv_probe(struct device_node *dn) diff --git a/drivers/crypto/nx/nx-842-pseries.c b/drivers/crypto/nx/nx-842-pseries.c index f4cbde03c6ad..cddc6d8b55d9 100644 --- a/drivers/crypto/nx/nx-842-pseries.c +++ b/drivers/crypto/nx/nx-842-pseries.c @@ -234,6 +234,10 @@ static int nx842_validate_result(struct device *dev, dev_dbg(dev, "%s: Out of space in output buffer\n", __func__); return -ENOSPC; + case 65: /* Calculated CRC doesn't match the passed value */ + dev_dbg(dev, "%s: CRC mismatch for decompression\n", + __func__); + return -EINVAL; case 66: /* Input data contains an illegal template field */ case 67: /* Template indicates data past the end of the input stream */ dev_dbg(dev, "%s: Bad data for decompression (code:%d)\n", @@ -324,7 +328,7 @@ static int nx842_pseries_compress(const unsigned char *in, unsigned int inlen, slout.entries = (struct nx842_slentry *)workmem->slout; /* Init operation */ - op.flags = NX842_OP_COMPRESS; + op.flags = NX842_OP_COMPRESS_CRC; csbcpb = &workmem->csbcpb; memset(csbcpb, 0, sizeof(*csbcpb)); op.csbcpb = nx842_get_pa(csbcpb); @@ -457,7 +461,7 @@ static int nx842_pseries_decompress(const unsigned char *in, unsigned int inlen, slout.entries = (struct nx842_slentry *)workmem->slout; /* Init operation */ - op.flags = NX842_OP_DECOMPRESS; + op.flags = NX842_OP_DECOMPRESS_CRC; csbcpb = &workmem->csbcpb; memset(csbcpb, 0, sizeof(*csbcpb)); op.csbcpb = nx842_get_pa(csbcpb); diff --git a/drivers/crypto/picoxcell_crypto.c b/drivers/crypto/picoxcell_crypto.c index da36de26a4dc..615da961c4d8 100644 --- a/drivers/crypto/picoxcell_crypto.c +++ b/drivers/crypto/picoxcell_crypto.c @@ -1591,6 +1591,7 @@ static const struct of_device_id spacc_of_id_table[] = { { .compatible = "picochip,spacc-l2" }, {} }; +MODULE_DEVICE_TABLE(of, spacc_of_id_table); #endif /* CONFIG_OF */ static bool spacc_is_compatible(struct platform_device *pdev, diff --git a/drivers/crypto/qat/qat_common/Makefile b/drivers/crypto/qat/qat_common/Makefile index df20a9de1c58..9e9e196c6d51 100644 --- a/drivers/crypto/qat/qat_common/Makefile +++ b/drivers/crypto/qat/qat_common/Makefile @@ -1,5 +1,10 @@ -$(obj)/qat_rsakey-asn1.o: $(obj)/qat_rsakey-asn1.c $(obj)/qat_rsakey-asn1.h -clean-files += qat_rsakey-asn1.c qat_rsakey-asn1.h +$(obj)/qat_rsapubkey-asn1.o: $(obj)/qat_rsapubkey-asn1.c \ + $(obj)/qat_rsapubkey-asn1.h +$(obj)/qat_rsaprivkey-asn1.o: $(obj)/qat_rsaprivkey-asn1.c \ + $(obj)/qat_rsaprivkey-asn1.h + +clean-files += qat_rsapubkey-asn1.c qat_rsapubkey-asn1.h +clean-files += qat_rsaprivkey-asn1.c qat_rsapvivkey-asn1.h obj-$(CONFIG_CRYPTO_DEV_QAT) += intel_qat.o intel_qat-objs := adf_cfg.o \ @@ -13,7 +18,8 @@ intel_qat-objs := adf_cfg.o \ adf_hw_arbiter.o \ qat_crypto.o \ qat_algs.o \ - qat_rsakey-asn1.o \ + qat_rsapubkey-asn1.o \ + qat_rsaprivkey-asn1.o \ qat_asym_algs.o \ qat_uclo.o \ qat_hal.o diff --git a/drivers/crypto/qat/qat_common/adf_common_drv.h b/drivers/crypto/qat/qat_common/adf_common_drv.h index 7836dffc3d47..3f76bd495bcb 100644 --- a/drivers/crypto/qat/qat_common/adf_common_drv.h +++ b/drivers/crypto/qat/qat_common/adf_common_drv.h @@ -163,10 +163,8 @@ struct qat_crypto_instance *qat_crypto_get_instance_node(int node); void qat_crypto_put_instance(struct qat_crypto_instance *inst); void qat_alg_callback(void *resp); void qat_alg_asym_callback(void *resp); -int qat_algs_init(void); -void qat_algs_exit(void); int qat_algs_register(void); -int qat_algs_unregister(void); +void qat_algs_unregister(void); int qat_asym_algs_register(void); void qat_asym_algs_unregister(void); diff --git a/drivers/crypto/qat/qat_common/adf_ctl_drv.c b/drivers/crypto/qat/qat_common/adf_ctl_drv.c index cd8a12af8ec5..03856ad280b9 100644 --- a/drivers/crypto/qat/qat_common/adf_ctl_drv.c +++ b/drivers/crypto/qat/qat_common/adf_ctl_drv.c @@ -463,9 +463,6 @@ static int __init adf_register_ctl_device_driver(void) { mutex_init(&adf_ctl_lock); - if (qat_algs_init()) - goto err_algs_init; - if (adf_chr_drv_create()) goto err_chr_dev; @@ -482,8 +479,6 @@ err_crypto_register: err_aer: adf_chr_drv_destroy(); err_chr_dev: - qat_algs_exit(); -err_algs_init: mutex_destroy(&adf_ctl_lock); return -EFAULT; } @@ -493,7 +488,6 @@ static void __exit adf_unregister_ctl_device_driver(void) adf_chr_drv_destroy(); adf_exit_aer(); qat_crypto_unregister(); - qat_algs_exit(); adf_clean_vf_map(false); mutex_destroy(&adf_ctl_lock); } diff --git a/drivers/crypto/qat/qat_common/adf_init.c b/drivers/crypto/qat/qat_common/adf_init.c index ac37a89965ac..d873eeecc363 100644 --- a/drivers/crypto/qat/qat_common/adf_init.c +++ b/drivers/crypto/qat/qat_common/adf_init.c @@ -272,12 +272,10 @@ int adf_dev_stop(struct adf_accel_dev *accel_dev) clear_bit(ADF_STATUS_STARTING, &accel_dev->status); clear_bit(ADF_STATUS_STARTED, &accel_dev->status); - if (!list_empty(&accel_dev->crypto_list) && qat_algs_unregister()) - dev_err(&GET_DEV(accel_dev), - "Failed to unregister crypto algs\n"); - - if (!list_empty(&accel_dev->crypto_list)) + if (!list_empty(&accel_dev->crypto_list)) { + qat_algs_unregister(); qat_asym_algs_unregister(); + } list_for_each(list_itr, &service_table) { service = list_entry(list_itr, struct service_hndl, list); diff --git a/drivers/crypto/qat/qat_common/adf_sriov.c b/drivers/crypto/qat/qat_common/adf_sriov.c index 2f77a4a8cecb..1117a8b58280 100644 --- a/drivers/crypto/qat/qat_common/adf_sriov.c +++ b/drivers/crypto/qat/qat_common/adf_sriov.c @@ -244,11 +244,8 @@ int adf_sriov_configure(struct pci_dev *pdev, int numvfs) return -EFAULT; } - if (!iommu_present(&pci_bus_type)) { - dev_err(&pdev->dev, - "IOMMU must be enabled for SR-IOV to work\n"); - return -EINVAL; - } + if (!iommu_present(&pci_bus_type)) + dev_warn(&pdev->dev, "IOMMU should be enabled for SR-IOV to work correctly\n"); if (accel_dev->pf.vf_info) { dev_info(&pdev->dev, "Already enabled for this device\n"); diff --git a/drivers/crypto/qat/qat_common/qat_algs.c b/drivers/crypto/qat/qat_common/qat_algs.c index 2bd913aceaeb..59e4c3af15ed 100644 --- a/drivers/crypto/qat/qat_common/qat_algs.c +++ b/drivers/crypto/qat/qat_common/qat_algs.c @@ -62,13 +62,13 @@ #include "icp_qat_fw.h" #include "icp_qat_fw_la.h" -#define QAT_AES_HW_CONFIG_CBC_ENC(alg) \ - ICP_QAT_HW_CIPHER_CONFIG_BUILD(ICP_QAT_HW_CIPHER_CBC_MODE, alg, \ +#define QAT_AES_HW_CONFIG_ENC(alg, mode) \ + ICP_QAT_HW_CIPHER_CONFIG_BUILD(mode, alg, \ ICP_QAT_HW_CIPHER_NO_CONVERT, \ ICP_QAT_HW_CIPHER_ENCRYPT) -#define QAT_AES_HW_CONFIG_CBC_DEC(alg) \ - ICP_QAT_HW_CIPHER_CONFIG_BUILD(ICP_QAT_HW_CIPHER_CBC_MODE, alg, \ +#define QAT_AES_HW_CONFIG_DEC(alg, mode) \ + ICP_QAT_HW_CIPHER_CONFIG_BUILD(mode, alg, \ ICP_QAT_HW_CIPHER_KEY_CONVERT, \ ICP_QAT_HW_CIPHER_DECRYPT) @@ -271,7 +271,8 @@ static void qat_alg_init_common_hdr(struct icp_qat_fw_comn_req_hdr *header) static int qat_alg_aead_init_enc_session(struct crypto_aead *aead_tfm, int alg, - struct crypto_authenc_keys *keys) + struct crypto_authenc_keys *keys, + int mode) { struct qat_alg_aead_ctx *ctx = crypto_aead_ctx(aead_tfm); unsigned int digestsize = crypto_aead_authsize(aead_tfm); @@ -288,7 +289,7 @@ static int qat_alg_aead_init_enc_session(struct crypto_aead *aead_tfm, struct icp_qat_fw_auth_cd_ctrl_hdr *hash_cd_ctrl = ptr; /* CD setup */ - cipher->aes.cipher_config.val = QAT_AES_HW_CONFIG_CBC_ENC(alg); + cipher->aes.cipher_config.val = QAT_AES_HW_CONFIG_ENC(alg, mode); memcpy(cipher->aes.key, keys->enckey, keys->enckeylen); hash->sha.inner_setup.auth_config.config = ICP_QAT_HW_AUTH_CONFIG_BUILD(ICP_QAT_HW_AUTH_MODE1, @@ -351,7 +352,8 @@ static int qat_alg_aead_init_enc_session(struct crypto_aead *aead_tfm, static int qat_alg_aead_init_dec_session(struct crypto_aead *aead_tfm, int alg, - struct crypto_authenc_keys *keys) + struct crypto_authenc_keys *keys, + int mode) { struct qat_alg_aead_ctx *ctx = crypto_aead_ctx(aead_tfm); unsigned int digestsize = crypto_aead_authsize(aead_tfm); @@ -373,7 +375,7 @@ static int qat_alg_aead_init_dec_session(struct crypto_aead *aead_tfm, sizeof(struct icp_qat_fw_la_cipher_req_params)); /* CD setup */ - cipher->aes.cipher_config.val = QAT_AES_HW_CONFIG_CBC_DEC(alg); + cipher->aes.cipher_config.val = QAT_AES_HW_CONFIG_DEC(alg, mode); memcpy(cipher->aes.key, keys->enckey, keys->enckeylen); hash->sha.inner_setup.auth_config.config = ICP_QAT_HW_AUTH_CONFIG_BUILD(ICP_QAT_HW_AUTH_MODE1, @@ -464,7 +466,7 @@ static void qat_alg_ablkcipher_init_com(struct qat_alg_ablkcipher_ctx *ctx, static void qat_alg_ablkcipher_init_enc(struct qat_alg_ablkcipher_ctx *ctx, int alg, const uint8_t *key, - unsigned int keylen) + unsigned int keylen, int mode) { struct icp_qat_hw_cipher_algo_blk *enc_cd = ctx->enc_cd; struct icp_qat_fw_la_bulk_req *req = &ctx->enc_fw_req; @@ -472,12 +474,12 @@ static void qat_alg_ablkcipher_init_enc(struct qat_alg_ablkcipher_ctx *ctx, qat_alg_ablkcipher_init_com(ctx, req, enc_cd, key, keylen); cd_pars->u.s.content_desc_addr = ctx->enc_cd_paddr; - enc_cd->aes.cipher_config.val = QAT_AES_HW_CONFIG_CBC_ENC(alg); + enc_cd->aes.cipher_config.val = QAT_AES_HW_CONFIG_ENC(alg, mode); } static void qat_alg_ablkcipher_init_dec(struct qat_alg_ablkcipher_ctx *ctx, int alg, const uint8_t *key, - unsigned int keylen) + unsigned int keylen, int mode) { struct icp_qat_hw_cipher_algo_blk *dec_cd = ctx->dec_cd; struct icp_qat_fw_la_bulk_req *req = &ctx->dec_fw_req; @@ -485,29 +487,48 @@ static void qat_alg_ablkcipher_init_dec(struct qat_alg_ablkcipher_ctx *ctx, qat_alg_ablkcipher_init_com(ctx, req, dec_cd, key, keylen); cd_pars->u.s.content_desc_addr = ctx->dec_cd_paddr; - dec_cd->aes.cipher_config.val = QAT_AES_HW_CONFIG_CBC_DEC(alg); + + if (mode != ICP_QAT_HW_CIPHER_CTR_MODE) + dec_cd->aes.cipher_config.val = + QAT_AES_HW_CONFIG_DEC(alg, mode); + else + dec_cd->aes.cipher_config.val = + QAT_AES_HW_CONFIG_ENC(alg, mode); } -static int qat_alg_validate_key(int key_len, int *alg) +static int qat_alg_validate_key(int key_len, int *alg, int mode) { - switch (key_len) { - case AES_KEYSIZE_128: - *alg = ICP_QAT_HW_CIPHER_ALGO_AES128; - break; - case AES_KEYSIZE_192: - *alg = ICP_QAT_HW_CIPHER_ALGO_AES192; - break; - case AES_KEYSIZE_256: - *alg = ICP_QAT_HW_CIPHER_ALGO_AES256; - break; - default: - return -EINVAL; + if (mode != ICP_QAT_HW_CIPHER_XTS_MODE) { + switch (key_len) { + case AES_KEYSIZE_128: + *alg = ICP_QAT_HW_CIPHER_ALGO_AES128; + break; + case AES_KEYSIZE_192: + *alg = ICP_QAT_HW_CIPHER_ALGO_AES192; + break; + case AES_KEYSIZE_256: + *alg = ICP_QAT_HW_CIPHER_ALGO_AES256; + break; + default: + return -EINVAL; + } + } else { + switch (key_len) { + case AES_KEYSIZE_128 << 1: + *alg = ICP_QAT_HW_CIPHER_ALGO_AES128; + break; + case AES_KEYSIZE_256 << 1: + *alg = ICP_QAT_HW_CIPHER_ALGO_AES256; + break; + default: + return -EINVAL; + } } return 0; } -static int qat_alg_aead_init_sessions(struct crypto_aead *tfm, - const uint8_t *key, unsigned int keylen) +static int qat_alg_aead_init_sessions(struct crypto_aead *tfm, const u8 *key, + unsigned int keylen, int mode) { struct crypto_authenc_keys keys; int alg; @@ -515,13 +536,13 @@ static int qat_alg_aead_init_sessions(struct crypto_aead *tfm, if (crypto_authenc_extractkeys(&keys, key, keylen)) goto bad_key; - if (qat_alg_validate_key(keys.enckeylen, &alg)) + if (qat_alg_validate_key(keys.enckeylen, &alg, mode)) goto bad_key; - if (qat_alg_aead_init_enc_session(tfm, alg, &keys)) + if (qat_alg_aead_init_enc_session(tfm, alg, &keys, mode)) goto error; - if (qat_alg_aead_init_dec_session(tfm, alg, &keys)) + if (qat_alg_aead_init_dec_session(tfm, alg, &keys, mode)) goto error; return 0; @@ -534,15 +555,16 @@ error: static int qat_alg_ablkcipher_init_sessions(struct qat_alg_ablkcipher_ctx *ctx, const uint8_t *key, - unsigned int keylen) + unsigned int keylen, + int mode) { int alg; - if (qat_alg_validate_key(keylen, &alg)) + if (qat_alg_validate_key(keylen, &alg, mode)) goto bad_key; - qat_alg_ablkcipher_init_enc(ctx, alg, key, keylen); - qat_alg_ablkcipher_init_dec(ctx, alg, key, keylen); + qat_alg_ablkcipher_init_enc(ctx, alg, key, keylen, mode); + qat_alg_ablkcipher_init_dec(ctx, alg, key, keylen, mode); return 0; bad_key: crypto_tfm_set_flags(ctx->tfm, CRYPTO_TFM_RES_BAD_KEY_LEN); @@ -586,7 +608,8 @@ static int qat_alg_aead_setkey(struct crypto_aead *tfm, const uint8_t *key, goto out_free_enc; } } - if (qat_alg_aead_init_sessions(tfm, key, keylen)) + if (qat_alg_aead_init_sessions(tfm, key, keylen, + ICP_QAT_HW_CIPHER_CBC_MODE)) goto out_free_all; return 0; @@ -876,8 +899,8 @@ static int qat_alg_aead_enc(struct aead_request *areq) } static int qat_alg_ablkcipher_setkey(struct crypto_ablkcipher *tfm, - const uint8_t *key, - unsigned int keylen) + const u8 *key, unsigned int keylen, + int mode) { struct qat_alg_ablkcipher_ctx *ctx = crypto_ablkcipher_ctx(tfm); struct device *dev; @@ -918,7 +941,7 @@ static int qat_alg_ablkcipher_setkey(struct crypto_ablkcipher *tfm, } } spin_unlock(&ctx->lock); - if (qat_alg_ablkcipher_init_sessions(ctx, key, keylen)) + if (qat_alg_ablkcipher_init_sessions(ctx, key, keylen, mode)) goto out_free_all; return 0; @@ -936,6 +959,27 @@ out_free_enc: return -ENOMEM; } +static int qat_alg_ablkcipher_cbc_setkey(struct crypto_ablkcipher *tfm, + const u8 *key, unsigned int keylen) +{ + return qat_alg_ablkcipher_setkey(tfm, key, keylen, + ICP_QAT_HW_CIPHER_CBC_MODE); +} + +static int qat_alg_ablkcipher_ctr_setkey(struct crypto_ablkcipher *tfm, + const u8 *key, unsigned int keylen) +{ + return qat_alg_ablkcipher_setkey(tfm, key, keylen, + ICP_QAT_HW_CIPHER_CTR_MODE); +} + +static int qat_alg_ablkcipher_xts_setkey(struct crypto_ablkcipher *tfm, + const u8 *key, unsigned int keylen) +{ + return qat_alg_ablkcipher_setkey(tfm, key, keylen, + ICP_QAT_HW_CIPHER_XTS_MODE); +} + static int qat_alg_ablkcipher_encrypt(struct ablkcipher_request *req) { struct crypto_ablkcipher *atfm = crypto_ablkcipher_reqtfm(req); @@ -1171,7 +1215,51 @@ static struct crypto_alg qat_algs[] = { { .cra_exit = qat_alg_ablkcipher_exit, .cra_u = { .ablkcipher = { - .setkey = qat_alg_ablkcipher_setkey, + .setkey = qat_alg_ablkcipher_cbc_setkey, + .decrypt = qat_alg_ablkcipher_decrypt, + .encrypt = qat_alg_ablkcipher_encrypt, + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + }, + }, +}, { + .cra_name = "ctr(aes)", + .cra_driver_name = "qat_aes_ctr", + .cra_priority = 4001, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct qat_alg_ablkcipher_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_init = qat_alg_ablkcipher_init, + .cra_exit = qat_alg_ablkcipher_exit, + .cra_u = { + .ablkcipher = { + .setkey = qat_alg_ablkcipher_ctr_setkey, + .decrypt = qat_alg_ablkcipher_decrypt, + .encrypt = qat_alg_ablkcipher_encrypt, + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + }, + }, +}, { + .cra_name = "xts(aes)", + .cra_driver_name = "qat_aes_xts", + .cra_priority = 4001, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER | CRYPTO_ALG_ASYNC, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct qat_alg_ablkcipher_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_init = qat_alg_ablkcipher_init, + .cra_exit = qat_alg_ablkcipher_exit, + .cra_u = { + .ablkcipher = { + .setkey = qat_alg_ablkcipher_xts_setkey, .decrypt = qat_alg_ablkcipher_decrypt, .encrypt = qat_alg_ablkcipher_encrypt, .min_keysize = AES_MIN_KEY_SIZE, @@ -1212,7 +1300,7 @@ unreg_algs: goto unlock; } -int qat_algs_unregister(void) +void qat_algs_unregister(void) { mutex_lock(&algs_lock); if (--active_devs != 0) @@ -1223,14 +1311,4 @@ int qat_algs_unregister(void) unlock: mutex_unlock(&algs_lock); - return 0; -} - -int qat_algs_init(void) -{ - return 0; -} - -void qat_algs_exit(void) -{ } diff --git a/drivers/crypto/qat/qat_common/qat_asym_algs.c b/drivers/crypto/qat/qat_common/qat_asym_algs.c index e87f51023ba4..51c594fdacdc 100644 --- a/drivers/crypto/qat/qat_common/qat_asym_algs.c +++ b/drivers/crypto/qat/qat_common/qat_asym_algs.c @@ -51,7 +51,9 @@ #include <crypto/akcipher.h> #include <linux/dma-mapping.h> #include <linux/fips.h> -#include "qat_rsakey-asn1.h" +#include <crypto/scatterwalk.h> +#include "qat_rsapubkey-asn1.h" +#include "qat_rsaprivkey-asn1.h" #include "icp_qat_fw_pke.h" #include "adf_accel_devices.h" #include "adf_transport.h" @@ -106,6 +108,7 @@ struct qat_rsa_request { dma_addr_t phy_in; dma_addr_t phy_out; char *src_align; + char *dst_align; struct icp_qat_fw_pke_request req; struct qat_rsa_ctx *ctx; int err; @@ -118,7 +121,6 @@ static void qat_rsa_cb(struct icp_qat_fw_pke_resp *resp) struct device *dev = &GET_DEV(req->ctx->inst->accel_dev); int err = ICP_QAT_FW_PKE_RESP_PKE_STAT_GET( resp->pke_resp_hdr.comn_resp_flags); - char *ptr = areq->dst; err = (err == ICP_QAT_FW_COMN_STATUS_FLAG_OK) ? 0 : -EINVAL; @@ -129,24 +131,44 @@ static void qat_rsa_cb(struct icp_qat_fw_pke_resp *resp) dma_unmap_single(dev, req->in.enc.m, req->ctx->key_sz, DMA_TO_DEVICE); - dma_unmap_single(dev, req->out.enc.c, req->ctx->key_sz, - DMA_FROM_DEVICE); + areq->dst_len = req->ctx->key_sz; + if (req->dst_align) { + char *ptr = req->dst_align; + + while (!(*ptr) && areq->dst_len) { + areq->dst_len--; + ptr++; + } + + if (areq->dst_len != req->ctx->key_sz) + memmove(req->dst_align, ptr, areq->dst_len); + + scatterwalk_map_and_copy(req->dst_align, areq->dst, 0, + areq->dst_len, 1); + + dma_free_coherent(dev, req->ctx->key_sz, req->dst_align, + req->out.enc.c); + } else { + char *ptr = sg_virt(areq->dst); + + while (!(*ptr) && areq->dst_len) { + areq->dst_len--; + ptr++; + } + + if (sg_virt(areq->dst) != ptr && areq->dst_len) + memmove(sg_virt(areq->dst), ptr, areq->dst_len); + + dma_unmap_single(dev, req->out.enc.c, req->ctx->key_sz, + DMA_FROM_DEVICE); + } + dma_unmap_single(dev, req->phy_in, sizeof(struct qat_rsa_input_params), DMA_TO_DEVICE); dma_unmap_single(dev, req->phy_out, sizeof(struct qat_rsa_output_params), DMA_TO_DEVICE); - areq->dst_len = req->ctx->key_sz; - /* Need to set the corect length of the output */ - while (!(*ptr) && areq->dst_len) { - areq->dst_len--; - ptr++; - } - - if (areq->dst_len != req->ctx->key_sz) - memmove(areq->dst, ptr, areq->dst_len); - akcipher_request_complete(areq, err); } @@ -255,8 +277,16 @@ static int qat_rsa_enc(struct akcipher_request *req) * same as modulo n so in case it is different we need to allocate a * new buf and copy src data. * In other case we just need to map the user provided buffer. + * Also need to make sure that it is in contiguous buffer. */ - if (req->src_len < ctx->key_sz) { + if (sg_is_last(req->src) && req->src_len == ctx->key_sz) { + qat_req->src_align = NULL; + qat_req->in.enc.m = dma_map_single(dev, sg_virt(req->src), + req->src_len, DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, qat_req->in.enc.m))) + return ret; + + } else { int shift = ctx->key_sz - req->src_len; qat_req->src_align = dma_zalloc_coherent(dev, ctx->key_sz, @@ -265,29 +295,39 @@ static int qat_rsa_enc(struct akcipher_request *req) if (unlikely(!qat_req->src_align)) return ret; - memcpy(qat_req->src_align + shift, req->src, req->src_len); + scatterwalk_map_and_copy(qat_req->src_align + shift, req->src, + 0, req->src_len, 0); + } + if (sg_is_last(req->dst) && req->dst_len == ctx->key_sz) { + qat_req->dst_align = NULL; + qat_req->out.enc.c = dma_map_single(dev, sg_virt(req->dst), + req->dst_len, + DMA_FROM_DEVICE); + + if (unlikely(dma_mapping_error(dev, qat_req->out.enc.c))) + goto unmap_src; + } else { - qat_req->src_align = NULL; - qat_req->in.enc.m = dma_map_single(dev, req->src, req->src_len, - DMA_TO_DEVICE); + qat_req->dst_align = dma_zalloc_coherent(dev, ctx->key_sz, + &qat_req->out.enc.c, + GFP_KERNEL); + if (unlikely(!qat_req->dst_align)) + goto unmap_src; + } qat_req->in.in_tab[3] = 0; - qat_req->out.enc.c = dma_map_single(dev, req->dst, req->dst_len, - DMA_FROM_DEVICE); qat_req->out.out_tab[1] = 0; qat_req->phy_in = dma_map_single(dev, &qat_req->in.enc.m, sizeof(struct qat_rsa_input_params), DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, qat_req->phy_in))) + goto unmap_dst; + qat_req->phy_out = dma_map_single(dev, &qat_req->out.enc.c, sizeof(struct qat_rsa_output_params), - DMA_TO_DEVICE); - - if (unlikely((!qat_req->src_align && - dma_mapping_error(dev, qat_req->in.enc.m)) || - dma_mapping_error(dev, qat_req->out.enc.c) || - dma_mapping_error(dev, qat_req->phy_in) || - dma_mapping_error(dev, qat_req->phy_out))) - goto unmap; + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, qat_req->phy_out))) + goto unmap_in_params; msg->pke_mid.src_data_addr = qat_req->phy_in; msg->pke_mid.dest_data_addr = qat_req->phy_out; @@ -300,7 +340,7 @@ static int qat_rsa_enc(struct akcipher_request *req) if (!ret) return -EINPROGRESS; -unmap: +unmap_src: if (qat_req->src_align) dma_free_coherent(dev, ctx->key_sz, qat_req->src_align, qat_req->in.enc.m); @@ -308,9 +348,15 @@ unmap: if (!dma_mapping_error(dev, qat_req->in.enc.m)) dma_unmap_single(dev, qat_req->in.enc.m, ctx->key_sz, DMA_TO_DEVICE); - if (!dma_mapping_error(dev, qat_req->out.enc.c)) - dma_unmap_single(dev, qat_req->out.enc.c, ctx->key_sz, - DMA_FROM_DEVICE); +unmap_dst: + if (qat_req->dst_align) + dma_free_coherent(dev, ctx->key_sz, qat_req->dst_align, + qat_req->out.enc.c); + else + if (!dma_mapping_error(dev, qat_req->out.enc.c)) + dma_unmap_single(dev, qat_req->out.enc.c, ctx->key_sz, + DMA_FROM_DEVICE); +unmap_in_params: if (!dma_mapping_error(dev, qat_req->phy_in)) dma_unmap_single(dev, qat_req->phy_in, sizeof(struct qat_rsa_input_params), @@ -362,8 +408,16 @@ static int qat_rsa_dec(struct akcipher_request *req) * same as modulo n so in case it is different we need to allocate a * new buf and copy src data. * In other case we just need to map the user provided buffer. + * Also need to make sure that it is in contiguous buffer. */ - if (req->src_len < ctx->key_sz) { + if (sg_is_last(req->src) && req->src_len == ctx->key_sz) { + qat_req->src_align = NULL; + qat_req->in.dec.c = dma_map_single(dev, sg_virt(req->src), + req->dst_len, DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, qat_req->in.dec.c))) + return ret; + + } else { int shift = ctx->key_sz - req->src_len; qat_req->src_align = dma_zalloc_coherent(dev, ctx->key_sz, @@ -372,29 +426,40 @@ static int qat_rsa_dec(struct akcipher_request *req) if (unlikely(!qat_req->src_align)) return ret; - memcpy(qat_req->src_align + shift, req->src, req->src_len); + scatterwalk_map_and_copy(qat_req->src_align + shift, req->src, + 0, req->src_len, 0); + } + if (sg_is_last(req->dst) && req->dst_len == ctx->key_sz) { + qat_req->dst_align = NULL; + qat_req->out.dec.m = dma_map_single(dev, sg_virt(req->dst), + req->dst_len, + DMA_FROM_DEVICE); + + if (unlikely(dma_mapping_error(dev, qat_req->out.dec.m))) + goto unmap_src; + } else { - qat_req->src_align = NULL; - qat_req->in.dec.c = dma_map_single(dev, req->src, req->src_len, - DMA_TO_DEVICE); + qat_req->dst_align = dma_zalloc_coherent(dev, ctx->key_sz, + &qat_req->out.dec.m, + GFP_KERNEL); + if (unlikely(!qat_req->dst_align)) + goto unmap_src; + } + qat_req->in.in_tab[3] = 0; - qat_req->out.dec.m = dma_map_single(dev, req->dst, req->dst_len, - DMA_FROM_DEVICE); qat_req->out.out_tab[1] = 0; qat_req->phy_in = dma_map_single(dev, &qat_req->in.dec.c, sizeof(struct qat_rsa_input_params), DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, qat_req->phy_in))) + goto unmap_dst; + qat_req->phy_out = dma_map_single(dev, &qat_req->out.dec.m, sizeof(struct qat_rsa_output_params), - DMA_TO_DEVICE); - - if (unlikely((!qat_req->src_align && - dma_mapping_error(dev, qat_req->in.dec.c)) || - dma_mapping_error(dev, qat_req->out.dec.m) || - dma_mapping_error(dev, qat_req->phy_in) || - dma_mapping_error(dev, qat_req->phy_out))) - goto unmap; + DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, qat_req->phy_out))) + goto unmap_in_params; msg->pke_mid.src_data_addr = qat_req->phy_in; msg->pke_mid.dest_data_addr = qat_req->phy_out; @@ -407,7 +472,7 @@ static int qat_rsa_dec(struct akcipher_request *req) if (!ret) return -EINPROGRESS; -unmap: +unmap_src: if (qat_req->src_align) dma_free_coherent(dev, ctx->key_sz, qat_req->src_align, qat_req->in.dec.c); @@ -415,9 +480,15 @@ unmap: if (!dma_mapping_error(dev, qat_req->in.dec.c)) dma_unmap_single(dev, qat_req->in.dec.c, ctx->key_sz, DMA_TO_DEVICE); - if (!dma_mapping_error(dev, qat_req->out.dec.m)) - dma_unmap_single(dev, qat_req->out.dec.m, ctx->key_sz, - DMA_FROM_DEVICE); +unmap_dst: + if (qat_req->dst_align) + dma_free_coherent(dev, ctx->key_sz, qat_req->dst_align, + qat_req->out.dec.m); + else + if (!dma_mapping_error(dev, qat_req->out.dec.m)) + dma_unmap_single(dev, qat_req->out.dec.m, ctx->key_sz, + DMA_FROM_DEVICE); +unmap_in_params: if (!dma_mapping_error(dev, qat_req->phy_in)) dma_unmap_single(dev, qat_req->phy_in, sizeof(struct qat_rsa_input_params), @@ -531,7 +602,7 @@ err: } static int qat_rsa_setkey(struct crypto_akcipher *tfm, const void *key, - unsigned int keylen) + unsigned int keylen, bool private) { struct qat_rsa_ctx *ctx = akcipher_tfm_ctx(tfm); struct device *dev = &GET_DEV(ctx->inst->accel_dev); @@ -550,7 +621,13 @@ static int qat_rsa_setkey(struct crypto_akcipher *tfm, const void *key, ctx->n = NULL; ctx->e = NULL; ctx->d = NULL; - ret = asn1_ber_decoder(&qat_rsakey_decoder, ctx, key, keylen); + + if (private) + ret = asn1_ber_decoder(&qat_rsaprivkey_decoder, ctx, key, + keylen); + else + ret = asn1_ber_decoder(&qat_rsapubkey_decoder, ctx, key, + keylen); if (ret < 0) goto free; @@ -559,6 +636,11 @@ static int qat_rsa_setkey(struct crypto_akcipher *tfm, const void *key, ret = -EINVAL; goto free; } + if (private && !ctx->d) { + /* invalid private key provided */ + ret = -EINVAL; + goto free; + } return 0; free: @@ -579,6 +661,25 @@ free: return ret; } +static int qat_rsa_setpubkey(struct crypto_akcipher *tfm, const void *key, + unsigned int keylen) +{ + return qat_rsa_setkey(tfm, key, keylen, false); +} + +static int qat_rsa_setprivkey(struct crypto_akcipher *tfm, const void *key, + unsigned int keylen) +{ + return qat_rsa_setkey(tfm, key, keylen, true); +} + +static int qat_rsa_max_size(struct crypto_akcipher *tfm) +{ + struct qat_rsa_ctx *ctx = akcipher_tfm_ctx(tfm); + + return (ctx->n) ? ctx->key_sz : -EINVAL; +} + static int qat_rsa_init_tfm(struct crypto_akcipher *tfm) { struct qat_rsa_ctx *ctx = akcipher_tfm_ctx(tfm); @@ -617,7 +718,9 @@ static struct akcipher_alg rsa = { .decrypt = qat_rsa_dec, .sign = qat_rsa_dec, .verify = qat_rsa_enc, - .setkey = qat_rsa_setkey, + .set_pub_key = qat_rsa_setpubkey, + .set_priv_key = qat_rsa_setprivkey, + .max_size = qat_rsa_max_size, .init = qat_rsa_init_tfm, .exit = qat_rsa_exit_tfm, .reqsize = sizeof(struct qat_rsa_request) + 64, diff --git a/drivers/crypto/qat/qat_common/qat_crypto.c b/drivers/crypto/qat/qat_common/qat_crypto.c index 07c2f9f9d1fc..9cab15497f04 100644 --- a/drivers/crypto/qat/qat_common/qat_crypto.c +++ b/drivers/crypto/qat/qat_common/qat_crypto.c @@ -60,8 +60,8 @@ static struct service_hndl qat_crypto; void qat_crypto_put_instance(struct qat_crypto_instance *inst) { - if (atomic_sub_return(1, &inst->refctr) == 0) - adf_dev_put(inst->accel_dev); + atomic_dec(&inst->refctr); + adf_dev_put(inst->accel_dev); } static int qat_crypto_free_instances(struct adf_accel_dev *accel_dev) @@ -97,49 +97,66 @@ static int qat_crypto_free_instances(struct adf_accel_dev *accel_dev) struct qat_crypto_instance *qat_crypto_get_instance_node(int node) { struct adf_accel_dev *accel_dev = NULL; - struct qat_crypto_instance *inst_best = NULL; + struct qat_crypto_instance *inst = NULL; struct list_head *itr; unsigned long best = ~0; list_for_each(itr, adf_devmgr_get_head()) { - accel_dev = list_entry(itr, struct adf_accel_dev, list); + struct adf_accel_dev *tmp_dev; + unsigned long ctr; + + tmp_dev = list_entry(itr, struct adf_accel_dev, list); + + if ((node == dev_to_node(&GET_DEV(tmp_dev)) || + dev_to_node(&GET_DEV(tmp_dev)) < 0) && + adf_dev_started(tmp_dev) && + !list_empty(&tmp_dev->crypto_list)) { + ctr = atomic_read(&tmp_dev->ref_count); + if (best > ctr) { + accel_dev = tmp_dev; + best = ctr; + } + } + } + if (!accel_dev) + pr_info("QAT: Could not find a device on node %d\n", node); + + /* Get any started device */ + list_for_each(itr, adf_devmgr_get_head()) { + struct adf_accel_dev *tmp_dev; - if ((node == dev_to_node(&GET_DEV(accel_dev)) || - dev_to_node(&GET_DEV(accel_dev)) < 0) && - adf_dev_started(accel_dev) && - !list_empty(&accel_dev->crypto_list)) + tmp_dev = list_entry(itr, struct adf_accel_dev, list); + + if (adf_dev_started(tmp_dev) && + !list_empty(&tmp_dev->crypto_list)) { + accel_dev = tmp_dev; break; - accel_dev = NULL; - } - if (!accel_dev) { - pr_err("QAT: Could not find a device on node %d\n", node); - accel_dev = adf_devmgr_get_first(); + } } - if (!accel_dev || !adf_dev_started(accel_dev)) + + if (!accel_dev) return NULL; + best = ~0; list_for_each(itr, &accel_dev->crypto_list) { - struct qat_crypto_instance *inst; - unsigned long cur; - - inst = list_entry(itr, struct qat_crypto_instance, list); - cur = atomic_read(&inst->refctr); - if (best > cur) { - inst_best = inst; - best = cur; + struct qat_crypto_instance *tmp_inst; + unsigned long ctr; + + tmp_inst = list_entry(itr, struct qat_crypto_instance, list); + ctr = atomic_read(&tmp_inst->refctr); + if (best > ctr) { + inst = tmp_inst; + best = ctr; } } - if (inst_best) { - if (atomic_add_return(1, &inst_best->refctr) == 1) { - if (adf_dev_get(accel_dev)) { - atomic_dec(&inst_best->refctr); - dev_err(&GET_DEV(accel_dev), - "Could not increment dev refctr\n"); - return NULL; - } + if (inst) { + if (adf_dev_get(accel_dev)) { + dev_err(&GET_DEV(accel_dev), "Could not increment dev refctr\n"); + return NULL; } + atomic_inc(&inst->refctr); } - return inst_best; + return inst; } static int qat_crypto_create_instances(struct adf_accel_dev *accel_dev) diff --git a/drivers/crypto/qat/qat_common/qat_hal.c b/drivers/crypto/qat/qat_common/qat_hal.c index 8e711d1c3084..380e761801a7 100644 --- a/drivers/crypto/qat/qat_common/qat_hal.c +++ b/drivers/crypto/qat/qat_common/qat_hal.c @@ -1034,7 +1034,7 @@ static int qat_hal_concat_micro_code(uint64_t *micro_inst, unsigned int inst_num, unsigned int size, unsigned int addr, unsigned int *value) { - int i, val_indx; + int i; unsigned int cur_value; const uint64_t *inst_arr; int fixup_offset; @@ -1042,8 +1042,7 @@ static int qat_hal_concat_micro_code(uint64_t *micro_inst, int orig_num; orig_num = inst_num; - val_indx = 0; - cur_value = value[val_indx++]; + cur_value = value[0]; inst_arr = inst_4b; usize = ARRAY_SIZE(inst_4b); fixup_offset = inst_num; diff --git a/drivers/crypto/qat/qat_common/qat_rsakey.asn1 b/drivers/crypto/qat/qat_common/qat_rsakey.asn1 deleted file mode 100644 index 97b0e02b600a..000000000000 --- a/drivers/crypto/qat/qat_common/qat_rsakey.asn1 +++ /dev/null @@ -1,5 +0,0 @@ -RsaKey ::= SEQUENCE { - n INTEGER ({ qat_rsa_get_n }), - e INTEGER ({ qat_rsa_get_e }), - d INTEGER ({ qat_rsa_get_d }) -} diff --git a/drivers/crypto/qat/qat_common/qat_rsaprivkey.asn1 b/drivers/crypto/qat/qat_common/qat_rsaprivkey.asn1 new file mode 100644 index 000000000000..f0066adb79b8 --- /dev/null +++ b/drivers/crypto/qat/qat_common/qat_rsaprivkey.asn1 @@ -0,0 +1,11 @@ +RsaPrivKey ::= SEQUENCE { + version INTEGER, + n INTEGER ({ qat_rsa_get_n }), + e INTEGER ({ qat_rsa_get_e }), + d INTEGER ({ qat_rsa_get_d }), + prime1 INTEGER, + prime2 INTEGER, + exponent1 INTEGER, + exponent2 INTEGER, + coefficient INTEGER +} diff --git a/drivers/crypto/qat/qat_common/qat_rsapubkey.asn1 b/drivers/crypto/qat/qat_common/qat_rsapubkey.asn1 new file mode 100644 index 000000000000..bd667b31a21a --- /dev/null +++ b/drivers/crypto/qat/qat_common/qat_rsapubkey.asn1 @@ -0,0 +1,4 @@ +RsaPubKey ::= SEQUENCE { + n INTEGER ({ qat_rsa_get_n }), + e INTEGER ({ qat_rsa_get_e }) +} diff --git a/drivers/crypto/qce/ablkcipher.c b/drivers/crypto/qce/ablkcipher.c index ad592de475a4..2c0d63d48747 100644 --- a/drivers/crypto/qce/ablkcipher.c +++ b/drivers/crypto/qce/ablkcipher.c @@ -44,10 +44,8 @@ static void qce_ablkcipher_done(void *data) error); if (diff_dst) - qce_unmapsg(qce->dev, rctx->src_sg, rctx->src_nents, dir_src, - rctx->dst_chained); - qce_unmapsg(qce->dev, rctx->dst_sg, rctx->dst_nents, dir_dst, - rctx->dst_chained); + dma_unmap_sg(qce->dev, rctx->src_sg, rctx->src_nents, dir_src); + dma_unmap_sg(qce->dev, rctx->dst_sg, rctx->dst_nents, dir_dst); sg_free_table(&rctx->dst_tbl); @@ -80,15 +78,11 @@ qce_ablkcipher_async_req_handle(struct crypto_async_request *async_req) dir_src = diff_dst ? DMA_TO_DEVICE : DMA_BIDIRECTIONAL; dir_dst = diff_dst ? DMA_FROM_DEVICE : DMA_BIDIRECTIONAL; - rctx->src_nents = qce_countsg(req->src, req->nbytes, - &rctx->src_chained); - if (diff_dst) { - rctx->dst_nents = qce_countsg(req->dst, req->nbytes, - &rctx->dst_chained); - } else { + rctx->src_nents = sg_nents_for_len(req->src, req->nbytes); + if (diff_dst) + rctx->dst_nents = sg_nents_for_len(req->dst, req->nbytes); + else rctx->dst_nents = rctx->src_nents; - rctx->dst_chained = rctx->src_chained; - } rctx->dst_nents += 1; @@ -116,14 +110,12 @@ qce_ablkcipher_async_req_handle(struct crypto_async_request *async_req) sg_mark_end(sg); rctx->dst_sg = rctx->dst_tbl.sgl; - ret = qce_mapsg(qce->dev, rctx->dst_sg, rctx->dst_nents, dir_dst, - rctx->dst_chained); + ret = dma_map_sg(qce->dev, rctx->dst_sg, rctx->dst_nents, dir_dst); if (ret < 0) goto error_free; if (diff_dst) { - ret = qce_mapsg(qce->dev, req->src, rctx->src_nents, dir_src, - rctx->src_chained); + ret = dma_map_sg(qce->dev, req->src, rctx->src_nents, dir_src); if (ret < 0) goto error_unmap_dst; rctx->src_sg = req->src; @@ -149,11 +141,9 @@ error_terminate: qce_dma_terminate_all(&qce->dma); error_unmap_src: if (diff_dst) - qce_unmapsg(qce->dev, req->src, rctx->src_nents, dir_src, - rctx->src_chained); + dma_unmap_sg(qce->dev, req->src, rctx->src_nents, dir_src); error_unmap_dst: - qce_unmapsg(qce->dev, rctx->dst_sg, rctx->dst_nents, dir_dst, - rctx->dst_chained); + dma_unmap_sg(qce->dev, rctx->dst_sg, rctx->dst_nents, dir_dst); error_free: sg_free_table(&rctx->dst_tbl); return ret; diff --git a/drivers/crypto/qce/cipher.h b/drivers/crypto/qce/cipher.h index d5757cfcda2d..5c6a5f8633e5 100644 --- a/drivers/crypto/qce/cipher.h +++ b/drivers/crypto/qce/cipher.h @@ -32,8 +32,6 @@ struct qce_cipher_ctx { * @ivsize: IV size * @src_nents: source entries * @dst_nents: destination entries - * @src_chained: is source chained - * @dst_chained: is destination chained * @result_sg: scatterlist used for result buffer * @dst_tbl: destination sg table * @dst_sg: destination sg pointer table beginning @@ -47,8 +45,6 @@ struct qce_cipher_reqctx { unsigned int ivsize; int src_nents; int dst_nents; - bool src_chained; - bool dst_chained; struct scatterlist result_sg; struct sg_table dst_tbl; struct scatterlist *dst_sg; diff --git a/drivers/crypto/qce/dma.c b/drivers/crypto/qce/dma.c index 378cb768647f..4797e795c9b9 100644 --- a/drivers/crypto/qce/dma.c +++ b/drivers/crypto/qce/dma.c @@ -54,58 +54,6 @@ void qce_dma_release(struct qce_dma_data *dma) kfree(dma->result_buf); } -int qce_mapsg(struct device *dev, struct scatterlist *sg, int nents, - enum dma_data_direction dir, bool chained) -{ - int err; - - if (chained) { - while (sg) { - err = dma_map_sg(dev, sg, 1, dir); - if (!err) - return -EFAULT; - sg = sg_next(sg); - } - } else { - err = dma_map_sg(dev, sg, nents, dir); - if (!err) - return -EFAULT; - } - - return nents; -} - -void qce_unmapsg(struct device *dev, struct scatterlist *sg, int nents, - enum dma_data_direction dir, bool chained) -{ - if (chained) - while (sg) { - dma_unmap_sg(dev, sg, 1, dir); - sg = sg_next(sg); - } - else - dma_unmap_sg(dev, sg, nents, dir); -} - -int qce_countsg(struct scatterlist *sglist, int nbytes, bool *chained) -{ - struct scatterlist *sg = sglist; - int nents = 0; - - if (chained) - *chained = false; - - while (nbytes > 0 && sg) { - nents++; - nbytes -= sg->length; - if (!sg_is_last(sg) && (sg + 1)->length == 0 && chained) - *chained = true; - sg = sg_next(sg); - } - - return nents; -} - struct scatterlist * qce_sgtable_add(struct sg_table *sgt, struct scatterlist *new_sgl) { diff --git a/drivers/crypto/qce/dma.h b/drivers/crypto/qce/dma.h index 65bedb81de0b..130235d17bb4 100644 --- a/drivers/crypto/qce/dma.h +++ b/drivers/crypto/qce/dma.h @@ -49,11 +49,6 @@ int qce_dma_prep_sgs(struct qce_dma_data *dma, struct scatterlist *sg_in, dma_async_tx_callback cb, void *cb_param); void qce_dma_issue_pending(struct qce_dma_data *dma); int qce_dma_terminate_all(struct qce_dma_data *dma); -int qce_countsg(struct scatterlist *sg_list, int nbytes, bool *chained); -void qce_unmapsg(struct device *dev, struct scatterlist *sg, int nents, - enum dma_data_direction dir, bool chained); -int qce_mapsg(struct device *dev, struct scatterlist *sg, int nents, - enum dma_data_direction dir, bool chained); struct scatterlist * qce_sgtable_add(struct sg_table *sgt, struct scatterlist *sg_add); diff --git a/drivers/crypto/qce/sha.c b/drivers/crypto/qce/sha.c index be2f5049256a..0c9973ec80eb 100644 --- a/drivers/crypto/qce/sha.c +++ b/drivers/crypto/qce/sha.c @@ -51,9 +51,8 @@ static void qce_ahash_done(void *data) if (error) dev_dbg(qce->dev, "ahash dma termination error (%d)\n", error); - qce_unmapsg(qce->dev, req->src, rctx->src_nents, DMA_TO_DEVICE, - rctx->src_chained); - qce_unmapsg(qce->dev, &rctx->result_sg, 1, DMA_FROM_DEVICE, 0); + dma_unmap_sg(qce->dev, req->src, rctx->src_nents, DMA_TO_DEVICE); + dma_unmap_sg(qce->dev, &rctx->result_sg, 1, DMA_FROM_DEVICE); memcpy(rctx->digest, result->auth_iv, digestsize); if (req->result) @@ -92,16 +91,14 @@ static int qce_ahash_async_req_handle(struct crypto_async_request *async_req) rctx->authklen = AES_KEYSIZE_128; } - rctx->src_nents = qce_countsg(req->src, req->nbytes, - &rctx->src_chained); - ret = qce_mapsg(qce->dev, req->src, rctx->src_nents, DMA_TO_DEVICE, - rctx->src_chained); + rctx->src_nents = sg_nents_for_len(req->src, req->nbytes); + ret = dma_map_sg(qce->dev, req->src, rctx->src_nents, DMA_TO_DEVICE); if (ret < 0) return ret; sg_init_one(&rctx->result_sg, qce->dma.result_buf, QCE_RESULT_BUF_SZ); - ret = qce_mapsg(qce->dev, &rctx->result_sg, 1, DMA_FROM_DEVICE, 0); + ret = dma_map_sg(qce->dev, &rctx->result_sg, 1, DMA_FROM_DEVICE); if (ret < 0) goto error_unmap_src; @@ -121,10 +118,9 @@ static int qce_ahash_async_req_handle(struct crypto_async_request *async_req) error_terminate: qce_dma_terminate_all(&qce->dma); error_unmap_dst: - qce_unmapsg(qce->dev, &rctx->result_sg, 1, DMA_FROM_DEVICE, 0); + dma_unmap_sg(qce->dev, &rctx->result_sg, 1, DMA_FROM_DEVICE); error_unmap_src: - qce_unmapsg(qce->dev, req->src, rctx->src_nents, DMA_TO_DEVICE, - rctx->src_chained); + dma_unmap_sg(qce->dev, req->src, rctx->src_nents, DMA_TO_DEVICE); return ret; } diff --git a/drivers/crypto/qce/sha.h b/drivers/crypto/qce/sha.h index 286f0d5397f3..236bb5e9ae75 100644 --- a/drivers/crypto/qce/sha.h +++ b/drivers/crypto/qce/sha.h @@ -36,7 +36,6 @@ struct qce_sha_ctx { * @flags: operation flags * @src_orig: original request sg list * @nbytes_orig: original request number of bytes - * @src_chained: is source scatterlist chained * @src_nents: source number of entries * @byte_count: byte count * @count: save count in states during update, import and export @@ -55,7 +54,6 @@ struct qce_sha_reqctx { unsigned long flags; struct scatterlist *src_orig; unsigned int nbytes_orig; - bool src_chained; int src_nents; __be32 byte_count[2]; u64 count; diff --git a/drivers/crypto/sahara.c b/drivers/crypto/sahara.c index 820dc3acb28c..f68c24a98277 100644 --- a/drivers/crypto/sahara.c +++ b/drivers/crypto/sahara.c @@ -173,7 +173,6 @@ struct sahara_aes_reqctx { * @sg_in_idx: number of hw links * @in_sg: scatterlist for input data * @in_sg_chain: scatterlists for chained input data - * @in_sg_chained: specifies if chained scatterlists are used or not * @total: total number of bytes for transfer * @last: is this the last block * @first: is this the first block @@ -191,7 +190,6 @@ struct sahara_sha_reqctx { unsigned int sg_in_idx; struct scatterlist *in_sg; struct scatterlist in_sg_chain[2]; - bool in_sg_chained; size_t total; unsigned int last; unsigned int first; @@ -274,31 +272,7 @@ static u32 sahara_aes_data_link_hdr(struct sahara_dev *dev) SAHARA_HDR_CHA_SKHA | SAHARA_HDR_PARITY_BIT; } -static int sahara_sg_length(struct scatterlist *sg, - unsigned int total) -{ - int sg_nb; - unsigned int len; - struct scatterlist *sg_list; - - sg_nb = 0; - sg_list = sg; - - while (total) { - len = min(sg_list->length, total); - - sg_nb++; - total -= len; - - sg_list = sg_next(sg_list); - if (!sg_list) - total = 0; - } - - return sg_nb; -} - -static char *sahara_err_src[16] = { +static const char *sahara_err_src[16] = { "No error", "Header error", "Descriptor length error", @@ -317,14 +291,14 @@ static char *sahara_err_src[16] = { "DMA error" }; -static char *sahara_err_dmasize[4] = { +static const char *sahara_err_dmasize[4] = { "Byte transfer", "Half-word transfer", "Word transfer", "Reserved" }; -static char *sahara_err_dmasrc[8] = { +static const char *sahara_err_dmasrc[8] = { "No error", "AHB bus error", "Internal IP bus error", @@ -335,7 +309,7 @@ static char *sahara_err_dmasrc[8] = { "DMA HW error" }; -static char *sahara_cha_errsrc[12] = { +static const char *sahara_cha_errsrc[12] = { "Input buffer non-empty", "Illegal address", "Illegal mode", @@ -350,7 +324,7 @@ static char *sahara_cha_errsrc[12] = { "Reserved" }; -static char *sahara_cha_err[4] = { "No error", "SKHA", "MDHA", "RNG" }; +static const char *sahara_cha_err[4] = { "No error", "SKHA", "MDHA", "RNG" }; static void sahara_decode_error(struct sahara_dev *dev, unsigned int error) { @@ -380,7 +354,7 @@ static void sahara_decode_error(struct sahara_dev *dev, unsigned int error) dev_err(dev->device, "\n"); } -static char *sahara_state[4] = { "Idle", "Busy", "Error", "HW Fault" }; +static const char *sahara_state[4] = { "Idle", "Busy", "Error", "HW Fault" }; static void sahara_decode_status(struct sahara_dev *dev, unsigned int status) { @@ -502,8 +476,8 @@ static int sahara_hw_descriptor_create(struct sahara_dev *dev) idx++; } - dev->nb_in_sg = sahara_sg_length(dev->in_sg, dev->total); - dev->nb_out_sg = sahara_sg_length(dev->out_sg, dev->total); + dev->nb_in_sg = sg_nents_for_len(dev->in_sg, dev->total); + dev->nb_out_sg = sg_nents_for_len(dev->out_sg, dev->total); if ((dev->nb_in_sg + dev->nb_out_sg) > SAHARA_MAX_HW_LINK) { dev_err(dev->device, "not enough hw links (%d)\n", dev->nb_in_sg + dev->nb_out_sg); @@ -818,45 +792,26 @@ static int sahara_sha_hw_links_create(struct sahara_dev *dev, dev->in_sg = rctx->in_sg; - dev->nb_in_sg = sahara_sg_length(dev->in_sg, rctx->total); + dev->nb_in_sg = sg_nents_for_len(dev->in_sg, rctx->total); if ((dev->nb_in_sg) > SAHARA_MAX_HW_LINK) { dev_err(dev->device, "not enough hw links (%d)\n", dev->nb_in_sg + dev->nb_out_sg); return -EINVAL; } - if (rctx->in_sg_chained) { - i = start; - sg = dev->in_sg; - while (sg) { - ret = dma_map_sg(dev->device, sg, 1, - DMA_TO_DEVICE); - if (!ret) - return -EFAULT; - - dev->hw_link[i]->len = sg->length; - dev->hw_link[i]->p = sg->dma_address; + sg = dev->in_sg; + ret = dma_map_sg(dev->device, dev->in_sg, dev->nb_in_sg, DMA_TO_DEVICE); + if (!ret) + return -EFAULT; + + for (i = start; i < dev->nb_in_sg + start; i++) { + dev->hw_link[i]->len = sg->length; + dev->hw_link[i]->p = sg->dma_address; + if (i == (dev->nb_in_sg + start - 1)) { + dev->hw_link[i]->next = 0; + } else { dev->hw_link[i]->next = dev->hw_phys_link[i + 1]; sg = sg_next(sg); - i += 1; - } - dev->hw_link[i-1]->next = 0; - } else { - sg = dev->in_sg; - ret = dma_map_sg(dev->device, dev->in_sg, dev->nb_in_sg, - DMA_TO_DEVICE); - if (!ret) - return -EFAULT; - - for (i = start; i < dev->nb_in_sg + start; i++) { - dev->hw_link[i]->len = sg->length; - dev->hw_link[i]->p = sg->dma_address; - if (i == (dev->nb_in_sg + start - 1)) { - dev->hw_link[i]->next = 0; - } else { - dev->hw_link[i]->next = dev->hw_phys_link[i + 1]; - sg = sg_next(sg); - } } } @@ -1004,7 +959,6 @@ static int sahara_sha_prepare_request(struct ahash_request *req) rctx->total = req->nbytes + rctx->buf_cnt; rctx->in_sg = rctx->in_sg_chain; - rctx->in_sg_chained = true; req->src = rctx->in_sg_chain; /* only data from previous operation */ } else if (rctx->buf_cnt) { @@ -1015,13 +969,11 @@ static int sahara_sha_prepare_request(struct ahash_request *req) /* buf was copied into rembuf above */ sg_init_one(rctx->in_sg, rctx->rembuf, rctx->buf_cnt); rctx->total = rctx->buf_cnt; - rctx->in_sg_chained = false; /* no data from previous operation */ } else { rctx->in_sg = req->src; rctx->total = req->nbytes; req->src = rctx->in_sg; - rctx->in_sg_chained = false; } /* on next call, we only have the remaining data in the buffer */ @@ -1030,23 +982,6 @@ static int sahara_sha_prepare_request(struct ahash_request *req) return -EINPROGRESS; } -static void sahara_sha_unmap_sg(struct sahara_dev *dev, - struct sahara_sha_reqctx *rctx) -{ - struct scatterlist *sg; - - if (rctx->in_sg_chained) { - sg = dev->in_sg; - while (sg) { - dma_unmap_sg(dev->device, sg, 1, DMA_TO_DEVICE); - sg = sg_next(sg); - } - } else { - dma_unmap_sg(dev->device, dev->in_sg, dev->nb_in_sg, - DMA_TO_DEVICE); - } -} - static int sahara_sha_process(struct ahash_request *req) { struct sahara_dev *dev = dev_ptr; @@ -1086,7 +1021,8 @@ static int sahara_sha_process(struct ahash_request *req) } if (rctx->sg_in_idx) - sahara_sha_unmap_sg(dev, rctx); + dma_unmap_sg(dev->device, dev->in_sg, dev->nb_in_sg, + DMA_TO_DEVICE); memcpy(rctx->context, dev->context_base, rctx->context_size); diff --git a/drivers/crypto/talitos.c b/drivers/crypto/talitos.c index 3b20a1bce703..46f531e19ccf 100644 --- a/drivers/crypto/talitos.c +++ b/drivers/crypto/talitos.c @@ -857,8 +857,6 @@ badkey: * talitos_edesc - s/w-extended descriptor * @src_nents: number of segments in input scatterlist * @dst_nents: number of segments in output scatterlist - * @src_chained: whether src is chained or not - * @dst_chained: whether dst is chained or not * @icv_ool: whether ICV is out-of-line * @iv_dma: dma address of iv for checking continuity and link table * @dma_len: length of dma mapped link_tbl space @@ -874,8 +872,6 @@ badkey: struct talitos_edesc { int src_nents; int dst_nents; - bool src_chained; - bool dst_chained; bool icv_ool; dma_addr_t iv_dma; int dma_len; @@ -887,29 +883,6 @@ struct talitos_edesc { }; }; -static int talitos_map_sg(struct device *dev, struct scatterlist *sg, - unsigned int nents, enum dma_data_direction dir, - bool chained) -{ - if (unlikely(chained)) - while (sg) { - dma_map_sg(dev, sg, 1, dir); - sg = sg_next(sg); - } - else - dma_map_sg(dev, sg, nents, dir); - return nents; -} - -static void talitos_unmap_sg_chain(struct device *dev, struct scatterlist *sg, - enum dma_data_direction dir) -{ - while (sg) { - dma_unmap_sg(dev, sg, 1, dir); - sg = sg_next(sg); - } -} - static void talitos_sg_unmap(struct device *dev, struct talitos_edesc *edesc, struct scatterlist *src, @@ -919,24 +892,13 @@ static void talitos_sg_unmap(struct device *dev, unsigned int dst_nents = edesc->dst_nents ? : 1; if (src != dst) { - if (edesc->src_chained) - talitos_unmap_sg_chain(dev, src, DMA_TO_DEVICE); - else - dma_unmap_sg(dev, src, src_nents, DMA_TO_DEVICE); + dma_unmap_sg(dev, src, src_nents, DMA_TO_DEVICE); if (dst) { - if (edesc->dst_chained) - talitos_unmap_sg_chain(dev, dst, - DMA_FROM_DEVICE); - else - dma_unmap_sg(dev, dst, dst_nents, - DMA_FROM_DEVICE); + dma_unmap_sg(dev, dst, dst_nents, DMA_FROM_DEVICE); } } else - if (edesc->src_chained) - talitos_unmap_sg_chain(dev, src, DMA_BIDIRECTIONAL); - else - dma_unmap_sg(dev, src, src_nents, DMA_BIDIRECTIONAL); + dma_unmap_sg(dev, src, src_nents, DMA_BIDIRECTIONAL); } static void ipsec_esp_unmap(struct device *dev, @@ -1118,10 +1080,9 @@ static int ipsec_esp(struct talitos_edesc *edesc, struct aead_request *areq, map_single_talitos_ptr(dev, &desc->ptr[0], ctx->authkeylen, &ctx->key, DMA_TO_DEVICE); - sg_count = talitos_map_sg(dev, areq->src, edesc->src_nents ?: 1, - (areq->src == areq->dst) ? DMA_BIDIRECTIONAL - : DMA_TO_DEVICE, - edesc->src_chained); + sg_count = dma_map_sg(dev, areq->src, edesc->src_nents ?: 1, + (areq->src == areq->dst) ? DMA_BIDIRECTIONAL + : DMA_TO_DEVICE); /* hmac data */ desc->ptr[1].len = cpu_to_be16(areq->assoclen); @@ -1185,9 +1146,8 @@ static int ipsec_esp(struct talitos_edesc *edesc, struct aead_request *areq, desc->ptr[5].j_extent = authsize; if (areq->src != areq->dst) - sg_count = talitos_map_sg(dev, areq->dst, - edesc->dst_nents ? : 1, - DMA_FROM_DEVICE, edesc->dst_chained); + sg_count = dma_map_sg(dev, areq->dst, edesc->dst_nents ? : 1, + DMA_FROM_DEVICE); edesc->icv_ool = false; @@ -1234,26 +1194,6 @@ static int ipsec_esp(struct talitos_edesc *edesc, struct aead_request *areq, } /* - * derive number of elements in scatterlist - */ -static int sg_count(struct scatterlist *sg_list, int nbytes, bool *chained) -{ - struct scatterlist *sg = sg_list; - int sg_nents = 0; - - *chained = false; - while (nbytes > 0 && sg) { - sg_nents++; - nbytes -= sg->length; - if (!sg_is_last(sg) && (sg + 1)->length == 0) - *chained = true; - sg = sg_next(sg); - } - - return sg_nents; -} - -/* * allocate and map the extended descriptor */ static struct talitos_edesc *talitos_edesc_alloc(struct device *dev, @@ -1270,7 +1210,6 @@ static struct talitos_edesc *talitos_edesc_alloc(struct device *dev, { struct talitos_edesc *edesc; int src_nents, dst_nents, alloc_len, dma_len; - bool src_chained = false, dst_chained = false; dma_addr_t iv_dma = 0; gfp_t flags = cryptoflags & CRYPTO_TFM_REQ_MAY_SLEEP ? GFP_KERNEL : GFP_ATOMIC; @@ -1287,18 +1226,16 @@ static struct talitos_edesc *talitos_edesc_alloc(struct device *dev, iv_dma = dma_map_single(dev, iv, ivsize, DMA_TO_DEVICE); if (!dst || dst == src) { - src_nents = sg_count(src, assoclen + cryptlen + authsize, - &src_chained); + src_nents = sg_nents_for_len(src, + assoclen + cryptlen + authsize); src_nents = (src_nents == 1) ? 0 : src_nents; dst_nents = dst ? src_nents : 0; } else { /* dst && dst != src*/ - src_nents = sg_count(src, assoclen + cryptlen + - (encrypt ? 0 : authsize), - &src_chained); + src_nents = sg_nents_for_len(src, assoclen + cryptlen + + (encrypt ? 0 : authsize)); src_nents = (src_nents == 1) ? 0 : src_nents; - dst_nents = sg_count(dst, assoclen + cryptlen + - (encrypt ? authsize : 0), - &dst_chained); + dst_nents = sg_nents_for_len(dst, assoclen + cryptlen + + (encrypt ? authsize : 0)); dst_nents = (dst_nents == 1) ? 0 : dst_nents; } @@ -1332,8 +1269,6 @@ static struct talitos_edesc *talitos_edesc_alloc(struct device *dev, edesc->src_nents = src_nents; edesc->dst_nents = dst_nents; - edesc->src_chained = src_chained; - edesc->dst_chained = dst_chained; edesc->iv_dma = iv_dma; edesc->dma_len = dma_len; if (dma_len) @@ -1518,8 +1453,7 @@ int map_sg_in_talitos_ptr(struct device *dev, struct scatterlist *src, } else { to_talitos_ptr_extent_clear(ptr, is_sec1); - sg_count = talitos_map_sg(dev, src, edesc->src_nents ? : 1, dir, - edesc->src_chained); + sg_count = dma_map_sg(dev, src, edesc->src_nents ? : 1, dir); if (sg_count == 1) { to_talitos_ptr(ptr, sg_dma_address(src), is_sec1); @@ -1552,8 +1486,7 @@ void map_sg_out_talitos_ptr(struct device *dev, struct scatterlist *dst, bool is_sec1 = has_ftr_sec1(priv); if (dir != DMA_NONE) - sg_count = talitos_map_sg(dev, dst, edesc->dst_nents ? : 1, - dir, edesc->dst_chained); + sg_count = dma_map_sg(dev, dst, edesc->dst_nents ? : 1, dir); to_talitos_ptr_len(ptr, len, is_sec1); @@ -1897,12 +1830,11 @@ static int ahash_process_req(struct ahash_request *areq, unsigned int nbytes) unsigned int nbytes_to_hash; unsigned int to_hash_later; unsigned int nsg; - bool chained; if (!req_ctx->last && (nbytes + req_ctx->nbuf <= blocksize)) { /* Buffer up to one whole block */ sg_copy_to_buffer(areq->src, - sg_count(areq->src, nbytes, &chained), + sg_nents_for_len(areq->src, nbytes), req_ctx->buf + req_ctx->nbuf, nbytes); req_ctx->nbuf += nbytes; return 0; @@ -1935,7 +1867,7 @@ static int ahash_process_req(struct ahash_request *areq, unsigned int nbytes) req_ctx->psrc = areq->src; if (to_hash_later) { - int nents = sg_count(areq->src, nbytes, &chained); + int nents = sg_nents_for_len(areq->src, nbytes); sg_pcopy_to_buffer(areq->src, nents, req_ctx->bufnext, to_hash_later, diff --git a/drivers/crypto/ux500/cryp/cryp_core.c b/drivers/crypto/ux500/cryp/cryp_core.c index fded0a5cfcd7..4c243c1ffc7f 100644 --- a/drivers/crypto/ux500/cryp/cryp_core.c +++ b/drivers/crypto/ux500/cryp/cryp_core.c @@ -1414,7 +1414,7 @@ static int ux500_cryp_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; dev_dbg(dev, "[%s]", __func__); - device_data = kzalloc(sizeof(struct cryp_device_data), GFP_ATOMIC); + device_data = devm_kzalloc(dev, sizeof(*device_data), GFP_ATOMIC); if (!device_data) { dev_err(dev, "[%s]: kzalloc() failed!", __func__); ret = -ENOMEM; @@ -1435,23 +1435,15 @@ static int ux500_cryp_probe(struct platform_device *pdev) dev_err(dev, "[%s]: platform_get_resource() failed", __func__); ret = -ENODEV; - goto out_kfree; - } - - res = request_mem_region(res->start, resource_size(res), pdev->name); - if (res == NULL) { - dev_err(dev, "[%s]: request_mem_region() failed", - __func__); - ret = -EBUSY; - goto out_kfree; + goto out; } device_data->phybase = res->start; - device_data->base = ioremap(res->start, resource_size(res)); + device_data->base = devm_ioremap_resource(dev, res); if (!device_data->base) { dev_err(dev, "[%s]: ioremap failed!", __func__); ret = -ENOMEM; - goto out_free_mem; + goto out; } spin_lock_init(&device_data->ctx_lock); @@ -1463,11 +1455,11 @@ static int ux500_cryp_probe(struct platform_device *pdev) dev_err(dev, "[%s]: could not get cryp regulator", __func__); ret = PTR_ERR(device_data->pwr_regulator); device_data->pwr_regulator = NULL; - goto out_unmap; + goto out; } /* Enable the clk for CRYP hardware block */ - device_data->clk = clk_get(&pdev->dev, NULL); + device_data->clk = devm_clk_get(&pdev->dev, NULL); if (IS_ERR(device_data->clk)) { dev_err(dev, "[%s]: clk_get() failed!", __func__); ret = PTR_ERR(device_data->clk); @@ -1477,7 +1469,7 @@ static int ux500_cryp_probe(struct platform_device *pdev) ret = clk_prepare(device_data->clk); if (ret) { dev_err(dev, "[%s]: clk_prepare() failed!", __func__); - goto out_clk; + goto out_regulator; } /* Enable device power (and clock) */ @@ -1510,11 +1502,8 @@ static int ux500_cryp_probe(struct platform_device *pdev) goto out_power; } - ret = request_irq(res_irq->start, - cryp_interrupt_handler, - 0, - "cryp1", - device_data); + ret = devm_request_irq(&pdev->dev, res_irq->start, + cryp_interrupt_handler, 0, "cryp1", device_data); if (ret) { dev_err(dev, "[%s]: Unable to request IRQ", __func__); goto out_power; @@ -1550,28 +1539,15 @@ out_power: out_clk_unprepare: clk_unprepare(device_data->clk); -out_clk: - clk_put(device_data->clk); - out_regulator: regulator_put(device_data->pwr_regulator); -out_unmap: - iounmap(device_data->base); - -out_free_mem: - release_mem_region(res->start, resource_size(res)); - -out_kfree: - kfree(device_data); out: return ret; } static int ux500_cryp_remove(struct platform_device *pdev) { - struct resource *res = NULL; - struct resource *res_irq = NULL; struct cryp_device_data *device_data; dev_dbg(&pdev->dev, "[%s]", __func__); @@ -1607,37 +1583,18 @@ static int ux500_cryp_remove(struct platform_device *pdev) if (list_empty(&driver_data.device_list.k_list)) cryp_algs_unregister_all(); - res_irq = platform_get_resource(pdev, IORESOURCE_IRQ, 0); - if (!res_irq) - dev_err(&pdev->dev, "[%s]: IORESOURCE_IRQ, unavailable", - __func__); - else { - disable_irq(res_irq->start); - free_irq(res_irq->start, device_data); - } - if (cryp_disable_power(&pdev->dev, device_data, false)) dev_err(&pdev->dev, "[%s]: cryp_disable_power() failed", __func__); clk_unprepare(device_data->clk); - clk_put(device_data->clk); regulator_put(device_data->pwr_regulator); - iounmap(device_data->base); - - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (res) - release_mem_region(res->start, resource_size(res)); - - kfree(device_data); - return 0; } static void ux500_cryp_shutdown(struct platform_device *pdev) { - struct resource *res_irq = NULL; struct cryp_device_data *device_data; dev_dbg(&pdev->dev, "[%s]", __func__); @@ -1673,15 +1630,6 @@ static void ux500_cryp_shutdown(struct platform_device *pdev) if (list_empty(&driver_data.device_list.k_list)) cryp_algs_unregister_all(); - res_irq = platform_get_resource(pdev, IORESOURCE_IRQ, 0); - if (!res_irq) - dev_err(&pdev->dev, "[%s]: IORESOURCE_IRQ, unavailable", - __func__); - else { - disable_irq(res_irq->start); - free_irq(res_irq->start, device_data); - } - if (cryp_disable_power(&pdev->dev, device_data, false)) dev_err(&pdev->dev, "[%s]: cryp_disable_power() failed", __func__); @@ -1777,6 +1725,7 @@ static const struct of_device_id ux500_cryp_match[] = { { .compatible = "stericsson,ux500-cryp" }, { }, }; +MODULE_DEVICE_TABLE(of, ux500_cryp_match); static struct platform_driver cryp_driver = { .probe = ux500_cryp_probe, diff --git a/drivers/crypto/ux500/hash/hash_core.c b/drivers/crypto/ux500/hash/hash_core.c index 5f5f360628fc..f47d112041b2 100644 --- a/drivers/crypto/ux500/hash/hash_core.c +++ b/drivers/crypto/ux500/hash/hash_core.c @@ -1657,7 +1657,7 @@ static int ux500_hash_probe(struct platform_device *pdev) struct hash_device_data *device_data; struct device *dev = &pdev->dev; - device_data = kzalloc(sizeof(*device_data), GFP_ATOMIC); + device_data = devm_kzalloc(dev, sizeof(*device_data), GFP_ATOMIC); if (!device_data) { ret = -ENOMEM; goto out; @@ -1670,22 +1670,15 @@ static int ux500_hash_probe(struct platform_device *pdev) if (!res) { dev_dbg(dev, "%s: platform_get_resource() failed!\n", __func__); ret = -ENODEV; - goto out_kfree; - } - - res = request_mem_region(res->start, resource_size(res), pdev->name); - if (res == NULL) { - dev_dbg(dev, "%s: request_mem_region() failed!\n", __func__); - ret = -EBUSY; - goto out_kfree; + goto out; } device_data->phybase = res->start; - device_data->base = ioremap(res->start, resource_size(res)); + device_data->base = devm_ioremap_resource(dev, res); if (!device_data->base) { dev_err(dev, "%s: ioremap() failed!\n", __func__); ret = -ENOMEM; - goto out_free_mem; + goto out; } spin_lock_init(&device_data->ctx_lock); spin_lock_init(&device_data->power_state_lock); @@ -1696,11 +1689,11 @@ static int ux500_hash_probe(struct platform_device *pdev) dev_err(dev, "%s: regulator_get() failed!\n", __func__); ret = PTR_ERR(device_data->regulator); device_data->regulator = NULL; - goto out_unmap; + goto out; } /* Enable the clock for HASH1 hardware block */ - device_data->clk = clk_get(dev, NULL); + device_data->clk = devm_clk_get(dev, NULL); if (IS_ERR(device_data->clk)) { dev_err(dev, "%s: clk_get() failed!\n", __func__); ret = PTR_ERR(device_data->clk); @@ -1710,7 +1703,7 @@ static int ux500_hash_probe(struct platform_device *pdev) ret = clk_prepare(device_data->clk); if (ret) { dev_err(dev, "%s: clk_prepare() failed!\n", __func__); - goto out_clk; + goto out_regulator; } /* Enable device power (and clock) */ @@ -1752,20 +1745,9 @@ out_power: out_clk_unprepare: clk_unprepare(device_data->clk); -out_clk: - clk_put(device_data->clk); - out_regulator: regulator_put(device_data->regulator); -out_unmap: - iounmap(device_data->base); - -out_free_mem: - release_mem_region(res->start, resource_size(res)); - -out_kfree: - kfree(device_data); out: return ret; } @@ -1776,7 +1758,6 @@ out: */ static int ux500_hash_remove(struct platform_device *pdev) { - struct resource *res; struct hash_device_data *device_data; struct device *dev = &pdev->dev; @@ -1816,17 +1797,8 @@ static int ux500_hash_remove(struct platform_device *pdev) __func__); clk_unprepare(device_data->clk); - clk_put(device_data->clk); regulator_put(device_data->regulator); - iounmap(device_data->base); - - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (res) - release_mem_region(res->start, resource_size(res)); - - kfree(device_data); - return 0; } @@ -1836,7 +1808,6 @@ static int ux500_hash_remove(struct platform_device *pdev) */ static void ux500_hash_shutdown(struct platform_device *pdev) { - struct resource *res = NULL; struct hash_device_data *device_data; device_data = platform_get_drvdata(pdev); @@ -1870,12 +1841,6 @@ static void ux500_hash_shutdown(struct platform_device *pdev) if (list_empty(&driver_data.device_list.k_list)) ahash_algs_unregister_all(device_data); - iounmap(device_data->base); - - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (res) - release_mem_region(res->start, resource_size(res)); - if (hash_disable_power(device_data, false)) dev_err(&pdev->dev, "%s: hash_disable_power() failed\n", __func__); @@ -1958,6 +1923,7 @@ static const struct of_device_id ux500_hash_match[] = { { .compatible = "stericsson,ux500-hash" }, { }, }; +MODULE_DEVICE_TABLE(of, ux500_hash_match); static struct platform_driver hash_driver = { .probe = ux500_hash_probe, diff --git a/include/crypto/akcipher.h b/include/crypto/akcipher.h index 69d163e39101..45cd5b328040 100644 --- a/include/crypto/akcipher.h +++ b/include/crypto/akcipher.h @@ -18,21 +18,21 @@ * struct akcipher_request - public key request * * @base: Common attributes for async crypto requests - * @src: Pointer to memory containing the input parameters - * The format of the parameter(s) is expeted to be Octet String - * @dst: Pointer to memory whare the result will be stored - * @src_len: Size of the input parameter + * @src: Source data + * @dst: Destination data + * @src_len: Size of the input buffer * @dst_len: Size of the output buffer. It needs to be at leaset * as big as the expected result depending on the operation * After operation it will be updated with the acctual size of the - * result. In case of error, where the dst_len was insufficient, + * result. + * In case of error where the dst sgl size was insufficient, * it will be updated to the size required for the operation. * @__ctx: Start of private context data */ struct akcipher_request { struct crypto_async_request base; - void *src; - void *dst; + struct scatterlist *src; + struct scatterlist *dst; unsigned int src_len; unsigned int dst_len; void *__ctx[] CRYPTO_MINALIGN_ATTR; @@ -67,8 +67,13 @@ struct crypto_akcipher { * algorithm. In case of error, where the dst_len was insufficient, * the req->dst_len will be updated to the size required for the * operation - * @setkey: Function invokes the algorithm specific set key function, which - * knows how to decode and interpret the BER encoded key + * @set_pub_key: Function invokes the algorithm specific set public key + * function, which knows how to decode and interpret + * the BER encoded public key + * @set_priv_key: Function invokes the algorithm specific set private key + * function, which knows how to decode and interpret + * the BER encoded private key + * @max_size: Function returns dest buffer size reqired for a given key. * @init: Initialize the cryptographic transformation object. * This function is used to initialize the cryptographic * transformation object. This function is called only once at @@ -89,8 +94,11 @@ struct akcipher_alg { int (*verify)(struct akcipher_request *req); int (*encrypt)(struct akcipher_request *req); int (*decrypt)(struct akcipher_request *req); - int (*setkey)(struct crypto_akcipher *tfm, const void *key, - unsigned int keylen); + int (*set_pub_key)(struct crypto_akcipher *tfm, const void *key, + unsigned int keylen); + int (*set_priv_key)(struct crypto_akcipher *tfm, const void *key, + unsigned int keylen); + int (*max_size)(struct crypto_akcipher *tfm); int (*init)(struct crypto_akcipher *tfm); void (*exit)(struct crypto_akcipher *tfm); @@ -229,14 +237,14 @@ static inline void akcipher_request_set_callback(struct akcipher_request *req, * Sets parameters required by crypto operation * * @req: public key request - * @src: ptr to input parameter - * @dst: ptr of output parameter - * @src_len: size of the input buffer - * @dst_len: size of the output buffer. It will be updated by the - * implementation to reflect the acctual size of the result + * @src: ptr to input scatter list + * @dst: ptr to output scatter list + * @src_len: size of the src input scatter list to be processed + * @dst_len: size of the dst output scatter list */ static inline void akcipher_request_set_crypt(struct akcipher_request *req, - void *src, void *dst, + struct scatterlist *src, + struct scatterlist *dst, unsigned int src_len, unsigned int dst_len) { @@ -247,6 +255,22 @@ static inline void akcipher_request_set_crypt(struct akcipher_request *req, } /** + * crypto_akcipher_maxsize() -- Get len for output buffer + * + * Function returns the dest buffer size required for a given key + * + * @tfm: AKCIPHER tfm handle allocated with crypto_alloc_akcipher() + * + * Return: minimum len for output buffer or error code in key hasn't been set + */ +static inline int crypto_akcipher_maxsize(struct crypto_akcipher *tfm) +{ + struct akcipher_alg *alg = crypto_akcipher_alg(tfm); + + return alg->max_size(tfm); +} + +/** * crypto_akcipher_encrypt() -- Invoke public key encrypt operation * * Function invokes the specific public key encrypt operation for a given @@ -319,22 +343,44 @@ static inline int crypto_akcipher_verify(struct akcipher_request *req) } /** - * crypto_akcipher_setkey() -- Invoke public key setkey operation + * crypto_akcipher_set_pub_key() -- Invoke set public key operation + * + * Function invokes the algorithm specific set key function, which knows + * how to decode and interpret the encoded key + * + * @tfm: tfm handle + * @key: BER encoded public key + * @keylen: length of the key + * + * Return: zero on success; error code in case of error + */ +static inline int crypto_akcipher_set_pub_key(struct crypto_akcipher *tfm, + const void *key, + unsigned int keylen) +{ + struct akcipher_alg *alg = crypto_akcipher_alg(tfm); + + return alg->set_pub_key(tfm, key, keylen); +} + +/** + * crypto_akcipher_set_priv_key() -- Invoke set private key operation * * Function invokes the algorithm specific set key function, which knows * how to decode and interpret the encoded key * * @tfm: tfm handle - * @key: BER encoded private or public key + * @key: BER encoded private key * @keylen: length of the key * * Return: zero on success; error code in case of error */ -static inline int crypto_akcipher_setkey(struct crypto_akcipher *tfm, void *key, - unsigned int keylen) +static inline int crypto_akcipher_set_priv_key(struct crypto_akcipher *tfm, + const void *key, + unsigned int keylen) { struct akcipher_alg *alg = crypto_akcipher_alg(tfm); - return alg->setkey(tfm, key, keylen); + return alg->set_priv_key(tfm, key, keylen); } #endif diff --git a/include/crypto/hash.h b/include/crypto/hash.h index 8e920b44c0ac..3d69c93d50e8 100644 --- a/include/crypto/hash.h +++ b/include/crypto/hash.h @@ -264,6 +264,20 @@ static inline unsigned int crypto_ahash_alignmask( return crypto_tfm_alg_alignmask(crypto_ahash_tfm(tfm)); } +/** + * crypto_ahash_blocksize() - obtain block size for cipher + * @tfm: cipher handle + * + * The block size for the message digest cipher referenced with the cipher + * handle is returned. + * + * Return: block size of cipher + */ +static inline unsigned int crypto_ahash_blocksize(struct crypto_ahash *tfm) +{ + return crypto_tfm_alg_blocksize(crypto_ahash_tfm(tfm)); +} + static inline struct hash_alg_common *__crypto_hash_alg_common( struct crypto_alg *alg) { diff --git a/include/crypto/internal/rsa.h b/include/crypto/internal/rsa.h index a8c86365439f..f997e2d29b5a 100644 --- a/include/crypto/internal/rsa.h +++ b/include/crypto/internal/rsa.h @@ -20,8 +20,11 @@ struct rsa_key { MPI d; }; -int rsa_parse_key(struct rsa_key *rsa_key, const void *key, - unsigned int key_len); +int rsa_parse_pub_key(struct rsa_key *rsa_key, const void *key, + unsigned int key_len); + +int rsa_parse_priv_key(struct rsa_key *rsa_key, const void *key, + unsigned int key_len); void rsa_free_key(struct rsa_key *rsa_key); #endif diff --git a/include/linux/mpi.h b/include/linux/mpi.h index 641b7d6fd096..3a5abe95affd 100644 --- a/include/linux/mpi.h +++ b/include/linux/mpi.h @@ -31,12 +31,7 @@ #define G10_MPI_H #include <linux/types.h> - -/* DSI defines */ - -#define SHA1_DIGEST_LENGTH 20 - -/*end of DSI defines */ +#include <linux/scatterlist.h> #define BYTES_PER_MPI_LIMB (BITS_PER_LONG / 8) #define BITS_PER_MPI_LIMB BITS_PER_LONG @@ -78,6 +73,7 @@ void mpi_swap(MPI a, MPI b); MPI do_encode_md(const void *sha_buffer, unsigned nbits); MPI mpi_read_raw_data(const void *xbuffer, size_t nbytes); MPI mpi_read_from_buffer(const void *buffer, unsigned *ret_nread); +MPI mpi_read_raw_from_sgl(struct scatterlist *sgl, unsigned int len); int mpi_fromstr(MPI val, const char *str); u32 mpi_get_keyid(MPI a, u32 *keyid); void *mpi_get_buffer(MPI a, unsigned *nbytes, int *sign); @@ -85,6 +81,8 @@ int mpi_read_buffer(MPI a, uint8_t *buf, unsigned buf_len, unsigned *nbytes, int *sign); void *mpi_get_secure_buffer(MPI a, unsigned *nbytes, int *sign); int mpi_set_buffer(MPI a, const void *buffer, unsigned nbytes, int sign); +int mpi_write_to_sgl(MPI a, struct scatterlist *sg, unsigned *nbytes, + int *sign); #define log_mpidump g10_log_mpidump diff --git a/lib/842/842.h b/lib/842/842.h index 7c200030acf7..e0a122bc1cdb 100644 --- a/lib/842/842.h +++ b/lib/842/842.h @@ -76,6 +76,7 @@ #include <linux/module.h> #include <linux/kernel.h> #include <linux/bitops.h> +#include <linux/crc32.h> #include <asm/unaligned.h> #include <linux/sw842.h> @@ -98,6 +99,7 @@ #define I2_BITS (8) #define I4_BITS (9) #define I8_BITS (8) +#define CRC_BITS (32) #define REPEAT_BITS_MAX (0x3f) #define SHORT_DATA_BITS_MAX (0x7) diff --git a/lib/842/842_compress.c b/lib/842/842_compress.c index 7ce68948e68c..4051339bdfbd 100644 --- a/lib/842/842_compress.c +++ b/lib/842/842_compress.c @@ -490,6 +490,7 @@ int sw842_compress(const u8 *in, unsigned int ilen, int ret; u64 last, next, pad, total; u8 repeat_count = 0; + u32 crc; BUILD_BUG_ON(sizeof(*p) > SW842_MEM_COMPRESS); @@ -580,6 +581,18 @@ skip_comp: if (ret) return ret; + /* + * crc(0:31) is appended to target data starting with the next + * bit after End of stream template. + * nx842 calculates CRC for data in big-endian format. So doing + * same here so that sw842 decompression can be used for both + * compressed data. + */ + crc = crc32_be(0, in, ilen); + ret = add_bits(p, crc, CRC_BITS); + if (ret) + return ret; + if (p->bit) { p->out++; p->olen--; diff --git a/lib/842/842_decompress.c b/lib/842/842_decompress.c index 5446ff0c9ba0..8881dad2a6a0 100644 --- a/lib/842/842_decompress.c +++ b/lib/842/842_decompress.c @@ -285,6 +285,7 @@ int sw842_decompress(const u8 *in, unsigned int ilen, struct sw842_param p; int ret; u64 op, rep, tmp, bytes, total; + u64 crc; p.in = (u8 *)in; p.bit = 0; @@ -375,6 +376,22 @@ int sw842_decompress(const u8 *in, unsigned int ilen, } } while (op != OP_END); + /* + * crc(0:31) is saved in compressed data starting with the + * next bit after End of stream template. + */ + ret = next_bits(&p, &crc, CRC_BITS); + if (ret) + return ret; + + /* + * Validate CRC saved in compressed data. + */ + if (crc != (u64)crc32_be(0, out, total - p.olen)) { + pr_debug("CRC mismatch for decompression\n"); + return -EINVAL; + } + if (unlikely((total - p.olen) > UINT_MAX)) return -ENOSPC; diff --git a/lib/mpi/mpicoder.c b/lib/mpi/mpicoder.c index 95c52a95259e..c7e0a705eecf 100644 --- a/lib/mpi/mpicoder.c +++ b/lib/mpi/mpicoder.c @@ -319,3 +319,202 @@ int mpi_set_buffer(MPI a, const void *xbuffer, unsigned nbytes, int sign) return 0; } EXPORT_SYMBOL_GPL(mpi_set_buffer); + +/** + * mpi_write_to_sgl() - Funnction exports MPI to an sgl (msb first) + * + * This function works in the same way as the mpi_read_buffer, but it + * takes an sgl instead of u8 * buf. + * + * @a: a multi precision integer + * @sgl: scatterlist to write to. Needs to be at least + * mpi_get_size(a) long. + * @nbytes: in/out param - it has the be set to the maximum number of + * bytes that can be written to sgl. This has to be at least + * the size of the integer a. On return it receives the actual + * length of the data written. + * @sign: if not NULL, it will be set to the sign of a. + * + * Return: 0 on success or error code in case of error + */ +int mpi_write_to_sgl(MPI a, struct scatterlist *sgl, unsigned *nbytes, + int *sign) +{ + u8 *p, *p2; + mpi_limb_t alimb, alimb2; + unsigned int n = mpi_get_size(a); + int i, x, y = 0, lzeros = 0, buf_len; + + if (!nbytes || *nbytes < n) + return -EINVAL; + + if (sign) + *sign = a->sign; + + p = (void *)&a->d[a->nlimbs] - 1; + + for (i = a->nlimbs * sizeof(alimb) - 1; i >= 0; i--, p--) { + if (!*p) + lzeros++; + else + break; + } + + *nbytes = n - lzeros; + buf_len = sgl->length; + p2 = sg_virt(sgl); + + for (i = a->nlimbs - 1; i >= 0; i--) { + alimb = a->d[i]; + p = (u8 *)&alimb2; +#if BYTES_PER_MPI_LIMB == 4 + *p++ = alimb >> 24; + *p++ = alimb >> 16; + *p++ = alimb >> 8; + *p++ = alimb; +#elif BYTES_PER_MPI_LIMB == 8 + *p++ = alimb >> 56; + *p++ = alimb >> 48; + *p++ = alimb >> 40; + *p++ = alimb >> 32; + *p++ = alimb >> 24; + *p++ = alimb >> 16; + *p++ = alimb >> 8; + *p++ = alimb; +#else +#error please implement for this limb size. +#endif + if (lzeros > 0) { + if (lzeros >= sizeof(alimb)) { + p -= sizeof(alimb); + continue; + } else { + mpi_limb_t *limb1 = (void *)p - sizeof(alimb); + mpi_limb_t *limb2 = (void *)p - sizeof(alimb) + + lzeros; + *limb1 = *limb2; + p -= lzeros; + y = lzeros; + } + lzeros -= sizeof(alimb); + } + + p = p - (sizeof(alimb) - y); + + for (x = 0; x < sizeof(alimb) - y; x++) { + if (!buf_len) { + sgl = sg_next(sgl); + if (!sgl) + return -EINVAL; + buf_len = sgl->length; + p2 = sg_virt(sgl); + } + *p2++ = *p++; + buf_len--; + } + y = 0; + } + return 0; +} +EXPORT_SYMBOL_GPL(mpi_write_to_sgl); + +/* + * mpi_read_raw_from_sgl() - Function allocates an MPI and populates it with + * data from the sgl + * + * This function works in the same way as the mpi_read_raw_data, but it + * takes an sgl instead of void * buffer. i.e. it allocates + * a new MPI and reads the content of the sgl to the MPI. + * + * @sgl: scatterlist to read from + * @len: number of bytes to read + * + * Return: Pointer to a new MPI or NULL on error + */ +MPI mpi_read_raw_from_sgl(struct scatterlist *sgl, unsigned int len) +{ + struct scatterlist *sg; + int x, i, j, z, lzeros, ents; + unsigned int nbits, nlimbs, nbytes; + mpi_limb_t a; + MPI val = NULL; + + lzeros = 0; + ents = sg_nents(sgl); + + for_each_sg(sgl, sg, ents, i) { + const u8 *buff = sg_virt(sg); + int len = sg->length; + + while (len && !*buff) { + lzeros++; + len--; + buff++; + } + + if (len && *buff) + break; + + ents--; + lzeros = 0; + } + + sgl = sg; + + if (!ents) + nbytes = 0; + else + nbytes = len - lzeros; + + nbits = nbytes * 8; + if (nbits > MAX_EXTERN_MPI_BITS) { + pr_info("MPI: mpi too large (%u bits)\n", nbits); + return NULL; + } + + if (nbytes > 0) + nbits -= count_leading_zeros(*(u8 *)(sg_virt(sgl) + lzeros)); + else + nbits = 0; + + nlimbs = DIV_ROUND_UP(nbytes, BYTES_PER_MPI_LIMB); + val = mpi_alloc(nlimbs); + if (!val) + return NULL; + + val->nbits = nbits; + val->sign = 0; + val->nlimbs = nlimbs; + + if (nbytes == 0) + return val; + + j = nlimbs - 1; + a = 0; + z = 0; + x = BYTES_PER_MPI_LIMB - nbytes % BYTES_PER_MPI_LIMB; + x %= BYTES_PER_MPI_LIMB; + + for_each_sg(sgl, sg, ents, i) { + const u8 *buffer = sg_virt(sg) + lzeros; + int len = sg->length - lzeros; + int buf_shift = x; + + if (sg_is_last(sg) && (len % BYTES_PER_MPI_LIMB)) + len += BYTES_PER_MPI_LIMB - (len % BYTES_PER_MPI_LIMB); + + for (; x < len + buf_shift; x++) { + a <<= 8; + a |= *buffer++; + if (((z + x + 1) % BYTES_PER_MPI_LIMB) == 0) { + val->d[j--] = a; + a = 0; + } + } + z += x; + x = 0; + lzeros = 0; + } + return val; +} +EXPORT_SYMBOL_GPL(mpi_read_raw_from_sgl); |