linux/drivers/crypto/s5p-sss.c

2330 lines
58 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
//
// Cryptographic API.
//
// Support for Samsung S5PV210 and Exynos HW acceleration.
//
// Copyright (C) 2011 NetUP Inc. All rights reserved.
// Copyright (c) 2017 Samsung Electronics Co., Ltd. All rights reserved.
//
// Hash part based on omap-sham.c driver.
#include <linux/clk.h>
#include <linux/crypto.h>
#include <linux/dma-mapping.h>
#include <linux/err.h>
#include <linux/errno.h>
#include <linux/init.h>
#include <linux/interrupt.h>
#include <linux/io.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/of.h>
#include <linux/platform_device.h>
#include <linux/scatterlist.h>
#include <crypto/ctr.h>
#include <crypto/aes.h>
#include <crypto/algapi.h>
#include <crypto/scatterwalk.h>
#include <crypto/hash.h>
#include <crypto/md5.h>
#include <crypto/sha.h>
#include <crypto/internal/hash.h>
#define _SBF(s, v) ((v) << (s))
/* Feed control registers */
#define SSS_REG_FCINTSTAT 0x0000
#define SSS_FCINTSTAT_HPARTINT BIT(7)
#define SSS_FCINTSTAT_HDONEINT BIT(5)
#define SSS_FCINTSTAT_BRDMAINT BIT(3)
#define SSS_FCINTSTAT_BTDMAINT BIT(2)
#define SSS_FCINTSTAT_HRDMAINT BIT(1)
#define SSS_FCINTSTAT_PKDMAINT BIT(0)
#define SSS_REG_FCINTENSET 0x0004
#define SSS_FCINTENSET_HPARTINTENSET BIT(7)
#define SSS_FCINTENSET_HDONEINTENSET BIT(5)
#define SSS_FCINTENSET_BRDMAINTENSET BIT(3)
#define SSS_FCINTENSET_BTDMAINTENSET BIT(2)
#define SSS_FCINTENSET_HRDMAINTENSET BIT(1)
#define SSS_FCINTENSET_PKDMAINTENSET BIT(0)
#define SSS_REG_FCINTENCLR 0x0008
#define SSS_FCINTENCLR_HPARTINTENCLR BIT(7)
#define SSS_FCINTENCLR_HDONEINTENCLR BIT(5)
#define SSS_FCINTENCLR_BRDMAINTENCLR BIT(3)
#define SSS_FCINTENCLR_BTDMAINTENCLR BIT(2)
#define SSS_FCINTENCLR_HRDMAINTENCLR BIT(1)
#define SSS_FCINTENCLR_PKDMAINTENCLR BIT(0)
#define SSS_REG_FCINTPEND 0x000C
#define SSS_FCINTPEND_HPARTINTP BIT(7)
#define SSS_FCINTPEND_HDONEINTP BIT(5)
#define SSS_FCINTPEND_BRDMAINTP BIT(3)
#define SSS_FCINTPEND_BTDMAINTP BIT(2)
#define SSS_FCINTPEND_HRDMAINTP BIT(1)
#define SSS_FCINTPEND_PKDMAINTP BIT(0)
#define SSS_REG_FCFIFOSTAT 0x0010
#define SSS_FCFIFOSTAT_BRFIFOFUL BIT(7)
#define SSS_FCFIFOSTAT_BRFIFOEMP BIT(6)
#define SSS_FCFIFOSTAT_BTFIFOFUL BIT(5)
#define SSS_FCFIFOSTAT_BTFIFOEMP BIT(4)
#define SSS_FCFIFOSTAT_HRFIFOFUL BIT(3)
#define SSS_FCFIFOSTAT_HRFIFOEMP BIT(2)
#define SSS_FCFIFOSTAT_PKFIFOFUL BIT(1)
#define SSS_FCFIFOSTAT_PKFIFOEMP BIT(0)
#define SSS_REG_FCFIFOCTRL 0x0014
#define SSS_FCFIFOCTRL_DESSEL BIT(2)
#define SSS_HASHIN_INDEPENDENT _SBF(0, 0x00)
#define SSS_HASHIN_CIPHER_INPUT _SBF(0, 0x01)
#define SSS_HASHIN_CIPHER_OUTPUT _SBF(0, 0x02)
#define SSS_HASHIN_MASK _SBF(0, 0x03)
#define SSS_REG_FCBRDMAS 0x0020
#define SSS_REG_FCBRDMAL 0x0024
#define SSS_REG_FCBRDMAC 0x0028
#define SSS_FCBRDMAC_BYTESWAP BIT(1)
#define SSS_FCBRDMAC_FLUSH BIT(0)
#define SSS_REG_FCBTDMAS 0x0030
#define SSS_REG_FCBTDMAL 0x0034
#define SSS_REG_FCBTDMAC 0x0038
#define SSS_FCBTDMAC_BYTESWAP BIT(1)
#define SSS_FCBTDMAC_FLUSH BIT(0)
#define SSS_REG_FCHRDMAS 0x0040
#define SSS_REG_FCHRDMAL 0x0044
#define SSS_REG_FCHRDMAC 0x0048
#define SSS_FCHRDMAC_BYTESWAP BIT(1)
#define SSS_FCHRDMAC_FLUSH BIT(0)
#define SSS_REG_FCPKDMAS 0x0050
#define SSS_REG_FCPKDMAL 0x0054
#define SSS_REG_FCPKDMAC 0x0058
#define SSS_FCPKDMAC_BYTESWAP BIT(3)
#define SSS_FCPKDMAC_DESCEND BIT(2)
#define SSS_FCPKDMAC_TRANSMIT BIT(1)
#define SSS_FCPKDMAC_FLUSH BIT(0)
#define SSS_REG_FCPKDMAO 0x005C
/* AES registers */
#define SSS_REG_AES_CONTROL 0x00
#define SSS_AES_BYTESWAP_DI BIT(11)
#define SSS_AES_BYTESWAP_DO BIT(10)
#define SSS_AES_BYTESWAP_IV BIT(9)
#define SSS_AES_BYTESWAP_CNT BIT(8)
#define SSS_AES_BYTESWAP_KEY BIT(7)
#define SSS_AES_KEY_CHANGE_MODE BIT(6)
#define SSS_AES_KEY_SIZE_128 _SBF(4, 0x00)
#define SSS_AES_KEY_SIZE_192 _SBF(4, 0x01)
#define SSS_AES_KEY_SIZE_256 _SBF(4, 0x02)
#define SSS_AES_FIFO_MODE BIT(3)
#define SSS_AES_CHAIN_MODE_ECB _SBF(1, 0x00)
#define SSS_AES_CHAIN_MODE_CBC _SBF(1, 0x01)
#define SSS_AES_CHAIN_MODE_CTR _SBF(1, 0x02)
#define SSS_AES_MODE_DECRYPT BIT(0)
#define SSS_REG_AES_STATUS 0x04
#define SSS_AES_BUSY BIT(2)
#define SSS_AES_INPUT_READY BIT(1)
#define SSS_AES_OUTPUT_READY BIT(0)
#define SSS_REG_AES_IN_DATA(s) (0x10 + (s << 2))
#define SSS_REG_AES_OUT_DATA(s) (0x20 + (s << 2))
#define SSS_REG_AES_IV_DATA(s) (0x30 + (s << 2))
#define SSS_REG_AES_CNT_DATA(s) (0x40 + (s << 2))
#define SSS_REG_AES_KEY_DATA(s) (0x80 + (s << 2))
#define SSS_REG(dev, reg) ((dev)->ioaddr + (SSS_REG_##reg))
#define SSS_READ(dev, reg) __raw_readl(SSS_REG(dev, reg))
#define SSS_WRITE(dev, reg, val) __raw_writel((val), SSS_REG(dev, reg))
#define SSS_AES_REG(dev, reg) ((dev)->aes_ioaddr + SSS_REG_##reg)
#define SSS_AES_WRITE(dev, reg, val) __raw_writel((val), \
SSS_AES_REG(dev, reg))
/* HW engine modes */
#define FLAGS_AES_DECRYPT BIT(0)
#define FLAGS_AES_MODE_MASK _SBF(1, 0x03)
#define FLAGS_AES_CBC _SBF(1, 0x01)
#define FLAGS_AES_CTR _SBF(1, 0x02)
#define AES_KEY_LEN 16
#define CRYPTO_QUEUE_LEN 1
/* HASH registers */
#define SSS_REG_HASH_CTRL 0x00
#define SSS_HASH_USER_IV_EN BIT(5)
#define SSS_HASH_INIT_BIT BIT(4)
#define SSS_HASH_ENGINE_SHA1 _SBF(1, 0x00)
#define SSS_HASH_ENGINE_MD5 _SBF(1, 0x01)
#define SSS_HASH_ENGINE_SHA256 _SBF(1, 0x02)
#define SSS_HASH_ENGINE_MASK _SBF(1, 0x03)
#define SSS_REG_HASH_CTRL_PAUSE 0x04
#define SSS_HASH_PAUSE BIT(0)
#define SSS_REG_HASH_CTRL_FIFO 0x08
#define SSS_HASH_FIFO_MODE_DMA BIT(0)
#define SSS_HASH_FIFO_MODE_CPU 0
#define SSS_REG_HASH_CTRL_SWAP 0x0C
#define SSS_HASH_BYTESWAP_DI BIT(3)
#define SSS_HASH_BYTESWAP_DO BIT(2)
#define SSS_HASH_BYTESWAP_IV BIT(1)
#define SSS_HASH_BYTESWAP_KEY BIT(0)
#define SSS_REG_HASH_STATUS 0x10
#define SSS_HASH_STATUS_MSG_DONE BIT(6)
#define SSS_HASH_STATUS_PARTIAL_DONE BIT(4)
#define SSS_HASH_STATUS_BUFFER_READY BIT(0)
#define SSS_REG_HASH_MSG_SIZE_LOW 0x20
#define SSS_REG_HASH_MSG_SIZE_HIGH 0x24
#define SSS_REG_HASH_PRE_MSG_SIZE_LOW 0x28
#define SSS_REG_HASH_PRE_MSG_SIZE_HIGH 0x2C
#define SSS_REG_HASH_IV(s) (0xB0 + ((s) << 2))
#define SSS_REG_HASH_OUT(s) (0x100 + ((s) << 2))
#define HASH_BLOCK_SIZE 64
#define HASH_REG_SIZEOF 4
#define HASH_MD5_MAX_REG (MD5_DIGEST_SIZE / HASH_REG_SIZEOF)
#define HASH_SHA1_MAX_REG (SHA1_DIGEST_SIZE / HASH_REG_SIZEOF)
#define HASH_SHA256_MAX_REG (SHA256_DIGEST_SIZE / HASH_REG_SIZEOF)
/*
* HASH bit numbers, used by device, setting in dev->hash_flags with
* functions set_bit(), clear_bit() or tested with test_bit() or BIT(),
* to keep HASH state BUSY or FREE, or to signal state from irq_handler
* to hash_tasklet. SGS keep track of allocated memory for scatterlist
*/
#define HASH_FLAGS_BUSY 0
#define HASH_FLAGS_FINAL 1
#define HASH_FLAGS_DMA_ACTIVE 2
#define HASH_FLAGS_OUTPUT_READY 3
#define HASH_FLAGS_DMA_READY 4
#define HASH_FLAGS_SGS_COPIED 5
#define HASH_FLAGS_SGS_ALLOCED 6
/* HASH HW constants */
#define BUFLEN HASH_BLOCK_SIZE
#define SSS_HASH_DMA_LEN_ALIGN 8
#define SSS_HASH_DMA_ALIGN_MASK (SSS_HASH_DMA_LEN_ALIGN - 1)
#define SSS_HASH_QUEUE_LENGTH 10
/**
* struct samsung_aes_variant - platform specific SSS driver data
* @aes_offset: AES register offset from SSS module's base.
* @hash_offset: HASH register offset from SSS module's base.
*
* Specifies platform specific configuration of SSS module.
* Note: A structure for driver specific platform data is used for future
* expansion of its usage.
*/
struct samsung_aes_variant {
unsigned int aes_offset;
unsigned int hash_offset;
};
struct s5p_aes_reqctx {
unsigned long mode;
};
struct s5p_aes_ctx {
struct s5p_aes_dev *dev;
uint8_t aes_key[AES_MAX_KEY_SIZE];
uint8_t nonce[CTR_RFC3686_NONCE_SIZE];
int keylen;
};
/**
* struct s5p_aes_dev - Crypto device state container
* @dev: Associated device
* @clk: Clock for accessing hardware
* @ioaddr: Mapped IO memory region
* @aes_ioaddr: Per-varian offset for AES block IO memory
* @irq_fc: Feed control interrupt line
* @req: Crypto request currently handled by the device
* @ctx: Configuration for currently handled crypto request
* @sg_src: Scatter list with source data for currently handled block
* in device. This is DMA-mapped into device.
* @sg_dst: Scatter list with destination data for currently handled block
* in device. This is DMA-mapped into device.
* @sg_src_cpy: In case of unaligned access, copied scatter list
* with source data.
* @sg_dst_cpy: In case of unaligned access, copied scatter list
* with destination data.
* @tasklet: New request scheduling jib
* @queue: Crypto queue
* @busy: Indicates whether the device is currently handling some request
* thus it uses some of the fields from this state, like:
* req, ctx, sg_src/dst (and copies). This essentially
* protects against concurrent access to these fields.
* @lock: Lock for protecting both access to device hardware registers
* and fields related to current request (including the busy field).
* @res: Resources for hash.
* @io_hash_base: Per-variant offset for HASH block IO memory.
* @hash_lock: Lock for protecting hash_req, hash_queue and hash_flags
* variable.
* @hash_flags: Flags for current HASH op.
* @hash_queue: Async hash queue.
* @hash_tasklet: New HASH request scheduling job.
* @xmit_buf: Buffer for current HASH request transfer into SSS block.
* @hash_req: Current request sending to SSS HASH block.
* @hash_sg_iter: Scatterlist transferred through DMA into SSS HASH block.
* @hash_sg_cnt: Counter for hash_sg_iter.
*
* @use_hash: true if HASH algs enabled
*/
struct s5p_aes_dev {
struct device *dev;
struct clk *clk;
void __iomem *ioaddr;
void __iomem *aes_ioaddr;
int irq_fc;
struct ablkcipher_request *req;
struct s5p_aes_ctx *ctx;
struct scatterlist *sg_src;
struct scatterlist *sg_dst;
struct scatterlist *sg_src_cpy;
struct scatterlist *sg_dst_cpy;
struct tasklet_struct tasklet;
struct crypto_queue queue;
bool busy;
spinlock_t lock;
struct resource *res;
void __iomem *io_hash_base;
spinlock_t hash_lock; /* protect hash_ vars */
unsigned long hash_flags;
struct crypto_queue hash_queue;
struct tasklet_struct hash_tasklet;
u8 xmit_buf[BUFLEN];
struct ahash_request *hash_req;
struct scatterlist *hash_sg_iter;
unsigned int hash_sg_cnt;
bool use_hash;
};
/**
* struct s5p_hash_reqctx - HASH request context
* @dd: Associated device
* @op_update: Current request operation (OP_UPDATE or OP_FINAL)
* @digcnt: Number of bytes processed by HW (without buffer[] ones)
* @digest: Digest message or IV for partial result
* @nregs: Number of HW registers for digest or IV read/write
* @engine: Bits for selecting type of HASH in SSS block
* @sg: sg for DMA transfer
* @sg_len: Length of sg for DMA transfer
* @sgl[]: sg for joining buffer and req->src scatterlist
* @skip: Skip offset in req->src for current op
* @total: Total number of bytes for current request
* @finup: Keep state for finup or final.
* @error: Keep track of error.
* @bufcnt: Number of bytes holded in buffer[]
* @buffer[]: For byte(s) from end of req->src in UPDATE op
*/
struct s5p_hash_reqctx {
struct s5p_aes_dev *dd;
bool op_update;
u64 digcnt;
u8 digest[SHA256_DIGEST_SIZE];
unsigned int nregs; /* digest_size / sizeof(reg) */
u32 engine;
struct scatterlist *sg;
unsigned int sg_len;
struct scatterlist sgl[2];
unsigned int skip;
unsigned int total;
bool finup;
bool error;
u32 bufcnt;
u8 buffer[0];
};
/**
* struct s5p_hash_ctx - HASH transformation context
* @dd: Associated device
* @flags: Bits for algorithm HASH.
* @fallback: Software transformation for zero message or size < BUFLEN.
*/
struct s5p_hash_ctx {
struct s5p_aes_dev *dd;
unsigned long flags;
struct crypto_shash *fallback;
};
static const struct samsung_aes_variant s5p_aes_data = {
.aes_offset = 0x4000,
.hash_offset = 0x6000,
};
static const struct samsung_aes_variant exynos_aes_data = {
.aes_offset = 0x200,
.hash_offset = 0x400,
};
static const struct of_device_id s5p_sss_dt_match[] = {
{
.compatible = "samsung,s5pv210-secss",
.data = &s5p_aes_data,
},
{
.compatible = "samsung,exynos4210-secss",
.data = &exynos_aes_data,
},
{ },
};
MODULE_DEVICE_TABLE(of, s5p_sss_dt_match);
static inline const struct samsung_aes_variant *find_s5p_sss_version
(const struct platform_device *pdev)
{
if (IS_ENABLED(CONFIG_OF) && (pdev->dev.of_node)) {
const struct of_device_id *match;
match = of_match_node(s5p_sss_dt_match,
pdev->dev.of_node);
return (const struct samsung_aes_variant *)match->data;
}
return (const struct samsung_aes_variant *)
platform_get_device_id(pdev)->driver_data;
}
static struct s5p_aes_dev *s5p_dev;
static void s5p_set_dma_indata(struct s5p_aes_dev *dev,
const struct scatterlist *sg)
{
SSS_WRITE(dev, FCBRDMAS, sg_dma_address(sg));
SSS_WRITE(dev, FCBRDMAL, sg_dma_len(sg));
}
static void s5p_set_dma_outdata(struct s5p_aes_dev *dev,
const struct scatterlist *sg)
{
SSS_WRITE(dev, FCBTDMAS, sg_dma_address(sg));
SSS_WRITE(dev, FCBTDMAL, sg_dma_len(sg));
}
static void s5p_free_sg_cpy(struct s5p_aes_dev *dev, struct scatterlist **sg)
{
int len;
if (!*sg)
return;
len = ALIGN(dev->req->nbytes, AES_BLOCK_SIZE);
free_pages((unsigned long)sg_virt(*sg), get_order(len));
kfree(*sg);
*sg = NULL;
}
static void s5p_sg_copy_buf(void *buf, struct scatterlist *sg,
unsigned int nbytes, int out)
{
struct scatter_walk walk;
if (!nbytes)
return;
scatterwalk_start(&walk, sg);
scatterwalk_copychunks(buf, &walk, nbytes, out);
scatterwalk_done(&walk, out, 0);
}
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
static void s5p_sg_done(struct s5p_aes_dev *dev)
{
if (dev->sg_dst_cpy) {
dev_dbg(dev->dev,
"Copying %d bytes of output data back to original place\n",
dev->req->nbytes);
s5p_sg_copy_buf(sg_virt(dev->sg_dst_cpy), dev->req->dst,
dev->req->nbytes, 1);
}
s5p_free_sg_cpy(dev, &dev->sg_src_cpy);
s5p_free_sg_cpy(dev, &dev->sg_dst_cpy);
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
}
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
/* Calls the completion. Cannot be called with dev->lock hold. */
static void s5p_aes_complete(struct s5p_aes_dev *dev, int err)
{
dev->req->base.complete(&dev->req->base, err);
}
static void s5p_unset_outdata(struct s5p_aes_dev *dev)
{
dma_unmap_sg(dev->dev, dev->sg_dst, 1, DMA_FROM_DEVICE);
}
static void s5p_unset_indata(struct s5p_aes_dev *dev)
{
dma_unmap_sg(dev->dev, dev->sg_src, 1, DMA_TO_DEVICE);
}
static int s5p_make_sg_cpy(struct s5p_aes_dev *dev, struct scatterlist *src,
struct scatterlist **dst)
{
void *pages;
int len;
*dst = kmalloc(sizeof(**dst), GFP_ATOMIC);
if (!*dst)
return -ENOMEM;
len = ALIGN(dev->req->nbytes, AES_BLOCK_SIZE);
pages = (void *)__get_free_pages(GFP_ATOMIC, get_order(len));
if (!pages) {
kfree(*dst);
*dst = NULL;
return -ENOMEM;
}
s5p_sg_copy_buf(pages, src, dev->req->nbytes, 0);
sg_init_table(*dst, 1);
sg_set_buf(*dst, pages, len);
return 0;
}
static int s5p_set_outdata(struct s5p_aes_dev *dev, struct scatterlist *sg)
{
int err;
if (!sg->length) {
err = -EINVAL;
goto exit;
}
err = dma_map_sg(dev->dev, sg, 1, DMA_FROM_DEVICE);
if (!err) {
err = -ENOMEM;
goto exit;
}
dev->sg_dst = sg;
err = 0;
exit:
return err;
}
static int s5p_set_indata(struct s5p_aes_dev *dev, struct scatterlist *sg)
{
int err;
if (!sg->length) {
err = -EINVAL;
goto exit;
}
err = dma_map_sg(dev->dev, sg, 1, DMA_TO_DEVICE);
if (!err) {
err = -ENOMEM;
goto exit;
}
dev->sg_src = sg;
err = 0;
exit:
return err;
}
crypto: s5p-sss - Fix missed interrupts when working with 8 kB blocks The tcrypt testing module on Exynos5422-based Odroid XU3/4 board failed on testing 8 kB size blocks: $ sudo modprobe tcrypt sec=1 mode=500 testing speed of async ecb(aes) (ecb-aes-s5p) encryption test 0 (128 bit key, 16 byte blocks): 21971 operations in 1 seconds (351536 bytes) test 1 (128 bit key, 64 byte blocks): 21731 operations in 1 seconds (1390784 bytes) test 2 (128 bit key, 256 byte blocks): 21932 operations in 1 seconds (5614592 bytes) test 3 (128 bit key, 1024 byte blocks): 21685 operations in 1 seconds (22205440 bytes) test 4 (128 bit key, 8192 byte blocks): This was caused by a race issue of missed BRDMA_DONE ("Block cipher Receiving DMA") interrupt. Device starts processing the data in DMA mode immediately after setting length of DMA block: receiving (FCBRDMAL) or transmitting (FCBTDMAL). The driver sets these lengths from interrupt handler through s5p_set_dma_indata() function (or xxx_setdata()). However the interrupt handler was first dealing with receive buffer (dma-unmap old, dma-map new, set receive block length which starts the operation), then with transmit buffer and finally was clearing pending interrupts (FCINTPEND). Because of the time window between setting receive buffer length and clearing pending interrupts, the operation on receive buffer could end already and driver would miss new interrupt. User manual for Exynos5422 confirms in example code that setting DMA block lengths should be the last operation. The tcrypt hang could be also observed in following blocked-task dmesg: INFO: task modprobe:258 blocked for more than 120 seconds. Not tainted 4.6.0-rc4-next-20160419-00005-g9eac8b7b7753-dirty #42 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. modprobe D c06b09d8 0 258 256 0x00000000 [<c06b09d8>] (__schedule) from [<c06b0f24>] (schedule+0x40/0xac) [<c06b0f24>] (schedule) from [<c06b49f8>] (schedule_timeout+0x124/0x178) [<c06b49f8>] (schedule_timeout) from [<c06b17fc>] (wait_for_common+0xb8/0x144) [<c06b17fc>] (wait_for_common) from [<bf0013b8>] (test_acipher_speed+0x49c/0x740 [tcrypt]) [<bf0013b8>] (test_acipher_speed [tcrypt]) from [<bf003e8c>] (do_test+0x2240/0x30ec [tcrypt]) [<bf003e8c>] (do_test [tcrypt]) from [<bf008048>] (tcrypt_mod_init+0x48/0xa4 [tcrypt]) [<bf008048>] (tcrypt_mod_init [tcrypt]) from [<c010177c>] (do_one_initcall+0x3c/0x16c) [<c010177c>] (do_one_initcall) from [<c0191ff0>] (do_init_module+0x5c/0x1ac) [<c0191ff0>] (do_init_module) from [<c0185610>] (load_module+0x1a30/0x1d08) [<c0185610>] (load_module) from [<c0185ab0>] (SyS_finit_module+0x8c/0x98) [<c0185ab0>] (SyS_finit_module) from [<c01078c0>] (ret_fast_syscall+0x0/0x3c) Fixes: a49e490c7a8a ("crypto: s5p-sss - add S5PV210 advanced crypto engine support") Cc: <stable@vger.kernel.org> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-04-22 20:15:23 +08:00
/*
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
* Returns -ERRNO on error (mapping of new data failed).
* On success returns:
* - 0 if there is no more data,
* - 1 if new transmitting (output) data is ready and its address+length
* have to be written to device (by calling s5p_set_dma_outdata()).
crypto: s5p-sss - Fix missed interrupts when working with 8 kB blocks The tcrypt testing module on Exynos5422-based Odroid XU3/4 board failed on testing 8 kB size blocks: $ sudo modprobe tcrypt sec=1 mode=500 testing speed of async ecb(aes) (ecb-aes-s5p) encryption test 0 (128 bit key, 16 byte blocks): 21971 operations in 1 seconds (351536 bytes) test 1 (128 bit key, 64 byte blocks): 21731 operations in 1 seconds (1390784 bytes) test 2 (128 bit key, 256 byte blocks): 21932 operations in 1 seconds (5614592 bytes) test 3 (128 bit key, 1024 byte blocks): 21685 operations in 1 seconds (22205440 bytes) test 4 (128 bit key, 8192 byte blocks): This was caused by a race issue of missed BRDMA_DONE ("Block cipher Receiving DMA") interrupt. Device starts processing the data in DMA mode immediately after setting length of DMA block: receiving (FCBRDMAL) or transmitting (FCBTDMAL). The driver sets these lengths from interrupt handler through s5p_set_dma_indata() function (or xxx_setdata()). However the interrupt handler was first dealing with receive buffer (dma-unmap old, dma-map new, set receive block length which starts the operation), then with transmit buffer and finally was clearing pending interrupts (FCINTPEND). Because of the time window between setting receive buffer length and clearing pending interrupts, the operation on receive buffer could end already and driver would miss new interrupt. User manual for Exynos5422 confirms in example code that setting DMA block lengths should be the last operation. The tcrypt hang could be also observed in following blocked-task dmesg: INFO: task modprobe:258 blocked for more than 120 seconds. Not tainted 4.6.0-rc4-next-20160419-00005-g9eac8b7b7753-dirty #42 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. modprobe D c06b09d8 0 258 256 0x00000000 [<c06b09d8>] (__schedule) from [<c06b0f24>] (schedule+0x40/0xac) [<c06b0f24>] (schedule) from [<c06b49f8>] (schedule_timeout+0x124/0x178) [<c06b49f8>] (schedule_timeout) from [<c06b17fc>] (wait_for_common+0xb8/0x144) [<c06b17fc>] (wait_for_common) from [<bf0013b8>] (test_acipher_speed+0x49c/0x740 [tcrypt]) [<bf0013b8>] (test_acipher_speed [tcrypt]) from [<bf003e8c>] (do_test+0x2240/0x30ec [tcrypt]) [<bf003e8c>] (do_test [tcrypt]) from [<bf008048>] (tcrypt_mod_init+0x48/0xa4 [tcrypt]) [<bf008048>] (tcrypt_mod_init [tcrypt]) from [<c010177c>] (do_one_initcall+0x3c/0x16c) [<c010177c>] (do_one_initcall) from [<c0191ff0>] (do_init_module+0x5c/0x1ac) [<c0191ff0>] (do_init_module) from [<c0185610>] (load_module+0x1a30/0x1d08) [<c0185610>] (load_module) from [<c0185ab0>] (SyS_finit_module+0x8c/0x98) [<c0185ab0>] (SyS_finit_module) from [<c01078c0>] (ret_fast_syscall+0x0/0x3c) Fixes: a49e490c7a8a ("crypto: s5p-sss - add S5PV210 advanced crypto engine support") Cc: <stable@vger.kernel.org> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-04-22 20:15:23 +08:00
*/
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
static int s5p_aes_tx(struct s5p_aes_dev *dev)
{
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
int ret = 0;
s5p_unset_outdata(dev);
if (!sg_is_last(dev->sg_dst)) {
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
ret = s5p_set_outdata(dev, sg_next(dev->sg_dst));
if (!ret)
ret = 1;
}
crypto: s5p-sss - Fix missed interrupts when working with 8 kB blocks The tcrypt testing module on Exynos5422-based Odroid XU3/4 board failed on testing 8 kB size blocks: $ sudo modprobe tcrypt sec=1 mode=500 testing speed of async ecb(aes) (ecb-aes-s5p) encryption test 0 (128 bit key, 16 byte blocks): 21971 operations in 1 seconds (351536 bytes) test 1 (128 bit key, 64 byte blocks): 21731 operations in 1 seconds (1390784 bytes) test 2 (128 bit key, 256 byte blocks): 21932 operations in 1 seconds (5614592 bytes) test 3 (128 bit key, 1024 byte blocks): 21685 operations in 1 seconds (22205440 bytes) test 4 (128 bit key, 8192 byte blocks): This was caused by a race issue of missed BRDMA_DONE ("Block cipher Receiving DMA") interrupt. Device starts processing the data in DMA mode immediately after setting length of DMA block: receiving (FCBRDMAL) or transmitting (FCBTDMAL). The driver sets these lengths from interrupt handler through s5p_set_dma_indata() function (or xxx_setdata()). However the interrupt handler was first dealing with receive buffer (dma-unmap old, dma-map new, set receive block length which starts the operation), then with transmit buffer and finally was clearing pending interrupts (FCINTPEND). Because of the time window between setting receive buffer length and clearing pending interrupts, the operation on receive buffer could end already and driver would miss new interrupt. User manual for Exynos5422 confirms in example code that setting DMA block lengths should be the last operation. The tcrypt hang could be also observed in following blocked-task dmesg: INFO: task modprobe:258 blocked for more than 120 seconds. Not tainted 4.6.0-rc4-next-20160419-00005-g9eac8b7b7753-dirty #42 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. modprobe D c06b09d8 0 258 256 0x00000000 [<c06b09d8>] (__schedule) from [<c06b0f24>] (schedule+0x40/0xac) [<c06b0f24>] (schedule) from [<c06b49f8>] (schedule_timeout+0x124/0x178) [<c06b49f8>] (schedule_timeout) from [<c06b17fc>] (wait_for_common+0xb8/0x144) [<c06b17fc>] (wait_for_common) from [<bf0013b8>] (test_acipher_speed+0x49c/0x740 [tcrypt]) [<bf0013b8>] (test_acipher_speed [tcrypt]) from [<bf003e8c>] (do_test+0x2240/0x30ec [tcrypt]) [<bf003e8c>] (do_test [tcrypt]) from [<bf008048>] (tcrypt_mod_init+0x48/0xa4 [tcrypt]) [<bf008048>] (tcrypt_mod_init [tcrypt]) from [<c010177c>] (do_one_initcall+0x3c/0x16c) [<c010177c>] (do_one_initcall) from [<c0191ff0>] (do_init_module+0x5c/0x1ac) [<c0191ff0>] (do_init_module) from [<c0185610>] (load_module+0x1a30/0x1d08) [<c0185610>] (load_module) from [<c0185ab0>] (SyS_finit_module+0x8c/0x98) [<c0185ab0>] (SyS_finit_module) from [<c01078c0>] (ret_fast_syscall+0x0/0x3c) Fixes: a49e490c7a8a ("crypto: s5p-sss - add S5PV210 advanced crypto engine support") Cc: <stable@vger.kernel.org> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-04-22 20:15:23 +08:00
return ret;
}
crypto: s5p-sss - Fix missed interrupts when working with 8 kB blocks The tcrypt testing module on Exynos5422-based Odroid XU3/4 board failed on testing 8 kB size blocks: $ sudo modprobe tcrypt sec=1 mode=500 testing speed of async ecb(aes) (ecb-aes-s5p) encryption test 0 (128 bit key, 16 byte blocks): 21971 operations in 1 seconds (351536 bytes) test 1 (128 bit key, 64 byte blocks): 21731 operations in 1 seconds (1390784 bytes) test 2 (128 bit key, 256 byte blocks): 21932 operations in 1 seconds (5614592 bytes) test 3 (128 bit key, 1024 byte blocks): 21685 operations in 1 seconds (22205440 bytes) test 4 (128 bit key, 8192 byte blocks): This was caused by a race issue of missed BRDMA_DONE ("Block cipher Receiving DMA") interrupt. Device starts processing the data in DMA mode immediately after setting length of DMA block: receiving (FCBRDMAL) or transmitting (FCBTDMAL). The driver sets these lengths from interrupt handler through s5p_set_dma_indata() function (or xxx_setdata()). However the interrupt handler was first dealing with receive buffer (dma-unmap old, dma-map new, set receive block length which starts the operation), then with transmit buffer and finally was clearing pending interrupts (FCINTPEND). Because of the time window between setting receive buffer length and clearing pending interrupts, the operation on receive buffer could end already and driver would miss new interrupt. User manual for Exynos5422 confirms in example code that setting DMA block lengths should be the last operation. The tcrypt hang could be also observed in following blocked-task dmesg: INFO: task modprobe:258 blocked for more than 120 seconds. Not tainted 4.6.0-rc4-next-20160419-00005-g9eac8b7b7753-dirty #42 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. modprobe D c06b09d8 0 258 256 0x00000000 [<c06b09d8>] (__schedule) from [<c06b0f24>] (schedule+0x40/0xac) [<c06b0f24>] (schedule) from [<c06b49f8>] (schedule_timeout+0x124/0x178) [<c06b49f8>] (schedule_timeout) from [<c06b17fc>] (wait_for_common+0xb8/0x144) [<c06b17fc>] (wait_for_common) from [<bf0013b8>] (test_acipher_speed+0x49c/0x740 [tcrypt]) [<bf0013b8>] (test_acipher_speed [tcrypt]) from [<bf003e8c>] (do_test+0x2240/0x30ec [tcrypt]) [<bf003e8c>] (do_test [tcrypt]) from [<bf008048>] (tcrypt_mod_init+0x48/0xa4 [tcrypt]) [<bf008048>] (tcrypt_mod_init [tcrypt]) from [<c010177c>] (do_one_initcall+0x3c/0x16c) [<c010177c>] (do_one_initcall) from [<c0191ff0>] (do_init_module+0x5c/0x1ac) [<c0191ff0>] (do_init_module) from [<c0185610>] (load_module+0x1a30/0x1d08) [<c0185610>] (load_module) from [<c0185ab0>] (SyS_finit_module+0x8c/0x98) [<c0185ab0>] (SyS_finit_module) from [<c01078c0>] (ret_fast_syscall+0x0/0x3c) Fixes: a49e490c7a8a ("crypto: s5p-sss - add S5PV210 advanced crypto engine support") Cc: <stable@vger.kernel.org> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-04-22 20:15:23 +08:00
/*
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
* Returns -ERRNO on error (mapping of new data failed).
* On success returns:
* - 0 if there is no more data,
* - 1 if new receiving (input) data is ready and its address+length
* have to be written to device (by calling s5p_set_dma_indata()).
crypto: s5p-sss - Fix missed interrupts when working with 8 kB blocks The tcrypt testing module on Exynos5422-based Odroid XU3/4 board failed on testing 8 kB size blocks: $ sudo modprobe tcrypt sec=1 mode=500 testing speed of async ecb(aes) (ecb-aes-s5p) encryption test 0 (128 bit key, 16 byte blocks): 21971 operations in 1 seconds (351536 bytes) test 1 (128 bit key, 64 byte blocks): 21731 operations in 1 seconds (1390784 bytes) test 2 (128 bit key, 256 byte blocks): 21932 operations in 1 seconds (5614592 bytes) test 3 (128 bit key, 1024 byte blocks): 21685 operations in 1 seconds (22205440 bytes) test 4 (128 bit key, 8192 byte blocks): This was caused by a race issue of missed BRDMA_DONE ("Block cipher Receiving DMA") interrupt. Device starts processing the data in DMA mode immediately after setting length of DMA block: receiving (FCBRDMAL) or transmitting (FCBTDMAL). The driver sets these lengths from interrupt handler through s5p_set_dma_indata() function (or xxx_setdata()). However the interrupt handler was first dealing with receive buffer (dma-unmap old, dma-map new, set receive block length which starts the operation), then with transmit buffer and finally was clearing pending interrupts (FCINTPEND). Because of the time window between setting receive buffer length and clearing pending interrupts, the operation on receive buffer could end already and driver would miss new interrupt. User manual for Exynos5422 confirms in example code that setting DMA block lengths should be the last operation. The tcrypt hang could be also observed in following blocked-task dmesg: INFO: task modprobe:258 blocked for more than 120 seconds. Not tainted 4.6.0-rc4-next-20160419-00005-g9eac8b7b7753-dirty #42 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. modprobe D c06b09d8 0 258 256 0x00000000 [<c06b09d8>] (__schedule) from [<c06b0f24>] (schedule+0x40/0xac) [<c06b0f24>] (schedule) from [<c06b49f8>] (schedule_timeout+0x124/0x178) [<c06b49f8>] (schedule_timeout) from [<c06b17fc>] (wait_for_common+0xb8/0x144) [<c06b17fc>] (wait_for_common) from [<bf0013b8>] (test_acipher_speed+0x49c/0x740 [tcrypt]) [<bf0013b8>] (test_acipher_speed [tcrypt]) from [<bf003e8c>] (do_test+0x2240/0x30ec [tcrypt]) [<bf003e8c>] (do_test [tcrypt]) from [<bf008048>] (tcrypt_mod_init+0x48/0xa4 [tcrypt]) [<bf008048>] (tcrypt_mod_init [tcrypt]) from [<c010177c>] (do_one_initcall+0x3c/0x16c) [<c010177c>] (do_one_initcall) from [<c0191ff0>] (do_init_module+0x5c/0x1ac) [<c0191ff0>] (do_init_module) from [<c0185610>] (load_module+0x1a30/0x1d08) [<c0185610>] (load_module) from [<c0185ab0>] (SyS_finit_module+0x8c/0x98) [<c0185ab0>] (SyS_finit_module) from [<c01078c0>] (ret_fast_syscall+0x0/0x3c) Fixes: a49e490c7a8a ("crypto: s5p-sss - add S5PV210 advanced crypto engine support") Cc: <stable@vger.kernel.org> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-04-22 20:15:23 +08:00
*/
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
static int s5p_aes_rx(struct s5p_aes_dev *dev/*, bool *set_dma*/)
{
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
int ret = 0;
s5p_unset_indata(dev);
if (!sg_is_last(dev->sg_src)) {
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
ret = s5p_set_indata(dev, sg_next(dev->sg_src));
if (!ret)
ret = 1;
}
crypto: s5p-sss - Fix missed interrupts when working with 8 kB blocks The tcrypt testing module on Exynos5422-based Odroid XU3/4 board failed on testing 8 kB size blocks: $ sudo modprobe tcrypt sec=1 mode=500 testing speed of async ecb(aes) (ecb-aes-s5p) encryption test 0 (128 bit key, 16 byte blocks): 21971 operations in 1 seconds (351536 bytes) test 1 (128 bit key, 64 byte blocks): 21731 operations in 1 seconds (1390784 bytes) test 2 (128 bit key, 256 byte blocks): 21932 operations in 1 seconds (5614592 bytes) test 3 (128 bit key, 1024 byte blocks): 21685 operations in 1 seconds (22205440 bytes) test 4 (128 bit key, 8192 byte blocks): This was caused by a race issue of missed BRDMA_DONE ("Block cipher Receiving DMA") interrupt. Device starts processing the data in DMA mode immediately after setting length of DMA block: receiving (FCBRDMAL) or transmitting (FCBTDMAL). The driver sets these lengths from interrupt handler through s5p_set_dma_indata() function (or xxx_setdata()). However the interrupt handler was first dealing with receive buffer (dma-unmap old, dma-map new, set receive block length which starts the operation), then with transmit buffer and finally was clearing pending interrupts (FCINTPEND). Because of the time window between setting receive buffer length and clearing pending interrupts, the operation on receive buffer could end already and driver would miss new interrupt. User manual for Exynos5422 confirms in example code that setting DMA block lengths should be the last operation. The tcrypt hang could be also observed in following blocked-task dmesg: INFO: task modprobe:258 blocked for more than 120 seconds. Not tainted 4.6.0-rc4-next-20160419-00005-g9eac8b7b7753-dirty #42 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. modprobe D c06b09d8 0 258 256 0x00000000 [<c06b09d8>] (__schedule) from [<c06b0f24>] (schedule+0x40/0xac) [<c06b0f24>] (schedule) from [<c06b49f8>] (schedule_timeout+0x124/0x178) [<c06b49f8>] (schedule_timeout) from [<c06b17fc>] (wait_for_common+0xb8/0x144) [<c06b17fc>] (wait_for_common) from [<bf0013b8>] (test_acipher_speed+0x49c/0x740 [tcrypt]) [<bf0013b8>] (test_acipher_speed [tcrypt]) from [<bf003e8c>] (do_test+0x2240/0x30ec [tcrypt]) [<bf003e8c>] (do_test [tcrypt]) from [<bf008048>] (tcrypt_mod_init+0x48/0xa4 [tcrypt]) [<bf008048>] (tcrypt_mod_init [tcrypt]) from [<c010177c>] (do_one_initcall+0x3c/0x16c) [<c010177c>] (do_one_initcall) from [<c0191ff0>] (do_init_module+0x5c/0x1ac) [<c0191ff0>] (do_init_module) from [<c0185610>] (load_module+0x1a30/0x1d08) [<c0185610>] (load_module) from [<c0185ab0>] (SyS_finit_module+0x8c/0x98) [<c0185ab0>] (SyS_finit_module) from [<c01078c0>] (ret_fast_syscall+0x0/0x3c) Fixes: a49e490c7a8a ("crypto: s5p-sss - add S5PV210 advanced crypto engine support") Cc: <stable@vger.kernel.org> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-04-22 20:15:23 +08:00
return ret;
}
static inline u32 s5p_hash_read(struct s5p_aes_dev *dd, u32 offset)
{
return __raw_readl(dd->io_hash_base + offset);
}
static inline void s5p_hash_write(struct s5p_aes_dev *dd,
u32 offset, u32 value)
{
__raw_writel(value, dd->io_hash_base + offset);
}
/**
* s5p_set_dma_hashdata() - start DMA with sg
* @dev: device
* @sg: scatterlist ready to DMA transmit
*/
static void s5p_set_dma_hashdata(struct s5p_aes_dev *dev,
const struct scatterlist *sg)
{
dev->hash_sg_cnt--;
SSS_WRITE(dev, FCHRDMAS, sg_dma_address(sg));
SSS_WRITE(dev, FCHRDMAL, sg_dma_len(sg)); /* DMA starts */
}
/**
* s5p_hash_rx() - get next hash_sg_iter
* @dev: device
*
* Return:
* 2 if there is no more data and it is UPDATE op
* 1 if new receiving (input) data is ready and can be written to device
* 0 if there is no more data and it is FINAL op
*/
static int s5p_hash_rx(struct s5p_aes_dev *dev)
{
if (dev->hash_sg_cnt > 0) {
dev->hash_sg_iter = sg_next(dev->hash_sg_iter);
return 1;
}
set_bit(HASH_FLAGS_DMA_READY, &dev->hash_flags);
if (test_bit(HASH_FLAGS_FINAL, &dev->hash_flags))
return 0;
return 2;
}
static irqreturn_t s5p_aes_interrupt(int irq, void *dev_id)
{
struct platform_device *pdev = dev_id;
struct s5p_aes_dev *dev = platform_get_drvdata(pdev);
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
int err_dma_tx = 0;
int err_dma_rx = 0;
int err_dma_hx = 0;
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
bool tx_end = false;
bool hx_end = false;
unsigned long flags;
uint32_t status;
u32 st_bits;
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
int err;
spin_lock_irqsave(&dev->lock, flags);
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
/*
* Handle rx or tx interrupt. If there is still data (scatterlist did not
* reach end), then map next scatterlist entry.
* In case of such mapping error, s5p_aes_complete() should be called.
*
* If there is no more data in tx scatter list, call s5p_aes_complete()
* and schedule new tasklet.
*
* Handle hx interrupt. If there is still data map next entry.
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
*/
status = SSS_READ(dev, FCINTSTAT);
if (status & SSS_FCINTSTAT_BRDMAINT)
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
err_dma_rx = s5p_aes_rx(dev);
if (status & SSS_FCINTSTAT_BTDMAINT) {
if (sg_is_last(dev->sg_dst))
tx_end = true;
err_dma_tx = s5p_aes_tx(dev);
}
if (status & SSS_FCINTSTAT_HRDMAINT)
err_dma_hx = s5p_hash_rx(dev);
st_bits = status & (SSS_FCINTSTAT_BRDMAINT | SSS_FCINTSTAT_BTDMAINT |
SSS_FCINTSTAT_HRDMAINT);
/* clear DMA bits */
SSS_WRITE(dev, FCINTPEND, st_bits);
/* clear HASH irq bits */
if (status & (SSS_FCINTSTAT_HDONEINT | SSS_FCINTSTAT_HPARTINT)) {
/* cannot have both HPART and HDONE */
if (status & SSS_FCINTSTAT_HPARTINT)
st_bits = SSS_HASH_STATUS_PARTIAL_DONE;
if (status & SSS_FCINTSTAT_HDONEINT)
st_bits = SSS_HASH_STATUS_MSG_DONE;
set_bit(HASH_FLAGS_OUTPUT_READY, &dev->hash_flags);
s5p_hash_write(dev, SSS_REG_HASH_STATUS, st_bits);
hx_end = true;
/* when DONE or PART, do not handle HASH DMA */
err_dma_hx = 0;
}
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
if (err_dma_rx < 0) {
err = err_dma_rx;
goto error;
}
if (err_dma_tx < 0) {
err = err_dma_tx;
goto error;
}
if (tx_end) {
s5p_sg_done(dev);
if (err_dma_hx == 1)
s5p_set_dma_hashdata(dev, dev->hash_sg_iter);
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
spin_unlock_irqrestore(&dev->lock, flags);
s5p_aes_complete(dev, 0);
/* Device is still busy */
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
tasklet_schedule(&dev->tasklet);
} else {
/*
* Writing length of DMA block (either receiving or
* transmitting) will start the operation immediately, so this
* should be done at the end (even after clearing pending
* interrupts to not miss the interrupt).
*/
if (err_dma_tx == 1)
s5p_set_dma_outdata(dev, dev->sg_dst);
if (err_dma_rx == 1)
s5p_set_dma_indata(dev, dev->sg_src);
if (err_dma_hx == 1)
s5p_set_dma_hashdata(dev, dev->hash_sg_iter);
crypto: s5p-sss - Fix missed interrupts when working with 8 kB blocks The tcrypt testing module on Exynos5422-based Odroid XU3/4 board failed on testing 8 kB size blocks: $ sudo modprobe tcrypt sec=1 mode=500 testing speed of async ecb(aes) (ecb-aes-s5p) encryption test 0 (128 bit key, 16 byte blocks): 21971 operations in 1 seconds (351536 bytes) test 1 (128 bit key, 64 byte blocks): 21731 operations in 1 seconds (1390784 bytes) test 2 (128 bit key, 256 byte blocks): 21932 operations in 1 seconds (5614592 bytes) test 3 (128 bit key, 1024 byte blocks): 21685 operations in 1 seconds (22205440 bytes) test 4 (128 bit key, 8192 byte blocks): This was caused by a race issue of missed BRDMA_DONE ("Block cipher Receiving DMA") interrupt. Device starts processing the data in DMA mode immediately after setting length of DMA block: receiving (FCBRDMAL) or transmitting (FCBTDMAL). The driver sets these lengths from interrupt handler through s5p_set_dma_indata() function (or xxx_setdata()). However the interrupt handler was first dealing with receive buffer (dma-unmap old, dma-map new, set receive block length which starts the operation), then with transmit buffer and finally was clearing pending interrupts (FCINTPEND). Because of the time window between setting receive buffer length and clearing pending interrupts, the operation on receive buffer could end already and driver would miss new interrupt. User manual for Exynos5422 confirms in example code that setting DMA block lengths should be the last operation. The tcrypt hang could be also observed in following blocked-task dmesg: INFO: task modprobe:258 blocked for more than 120 seconds. Not tainted 4.6.0-rc4-next-20160419-00005-g9eac8b7b7753-dirty #42 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. modprobe D c06b09d8 0 258 256 0x00000000 [<c06b09d8>] (__schedule) from [<c06b0f24>] (schedule+0x40/0xac) [<c06b0f24>] (schedule) from [<c06b49f8>] (schedule_timeout+0x124/0x178) [<c06b49f8>] (schedule_timeout) from [<c06b17fc>] (wait_for_common+0xb8/0x144) [<c06b17fc>] (wait_for_common) from [<bf0013b8>] (test_acipher_speed+0x49c/0x740 [tcrypt]) [<bf0013b8>] (test_acipher_speed [tcrypt]) from [<bf003e8c>] (do_test+0x2240/0x30ec [tcrypt]) [<bf003e8c>] (do_test [tcrypt]) from [<bf008048>] (tcrypt_mod_init+0x48/0xa4 [tcrypt]) [<bf008048>] (tcrypt_mod_init [tcrypt]) from [<c010177c>] (do_one_initcall+0x3c/0x16c) [<c010177c>] (do_one_initcall) from [<c0191ff0>] (do_init_module+0x5c/0x1ac) [<c0191ff0>] (do_init_module) from [<c0185610>] (load_module+0x1a30/0x1d08) [<c0185610>] (load_module) from [<c0185ab0>] (SyS_finit_module+0x8c/0x98) [<c0185ab0>] (SyS_finit_module) from [<c01078c0>] (ret_fast_syscall+0x0/0x3c) Fixes: a49e490c7a8a ("crypto: s5p-sss - add S5PV210 advanced crypto engine support") Cc: <stable@vger.kernel.org> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2016-04-22 20:15:23 +08:00
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
spin_unlock_irqrestore(&dev->lock, flags);
}
goto hash_irq_end;
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
error:
s5p_sg_done(dev);
dev->busy = false;
if (err_dma_hx == 1)
s5p_set_dma_hashdata(dev, dev->hash_sg_iter);
spin_unlock_irqrestore(&dev->lock, flags);
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
s5p_aes_complete(dev, err);
hash_irq_end:
/*
* Note about else if:
* when hash_sg_iter reaches end and its UPDATE op,
* issue SSS_HASH_PAUSE and wait for HPART irq
*/
if (hx_end)
tasklet_schedule(&dev->hash_tasklet);
else if (err_dma_hx == 2)
s5p_hash_write(dev, SSS_REG_HASH_CTRL_PAUSE,
SSS_HASH_PAUSE);
return IRQ_HANDLED;
}
/**
* s5p_hash_read_msg() - read message or IV from HW
* @req: AHASH request
*/
static void s5p_hash_read_msg(struct ahash_request *req)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
struct s5p_aes_dev *dd = ctx->dd;
u32 *hash = (u32 *)ctx->digest;
unsigned int i;
for (i = 0; i < ctx->nregs; i++)
hash[i] = s5p_hash_read(dd, SSS_REG_HASH_OUT(i));
}
/**
* s5p_hash_write_ctx_iv() - write IV for next partial/finup op.
* @dd: device
* @ctx: request context
*/
static void s5p_hash_write_ctx_iv(struct s5p_aes_dev *dd,
const struct s5p_hash_reqctx *ctx)
{
const u32 *hash = (const u32 *)ctx->digest;
unsigned int i;
for (i = 0; i < ctx->nregs; i++)
s5p_hash_write(dd, SSS_REG_HASH_IV(i), hash[i]);
}
/**
* s5p_hash_write_iv() - write IV for next partial/finup op.
* @req: AHASH request
*/
static void s5p_hash_write_iv(struct ahash_request *req)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
s5p_hash_write_ctx_iv(ctx->dd, ctx);
}
/**
* s5p_hash_copy_result() - copy digest into req->result
* @req: AHASH request
*/
static void s5p_hash_copy_result(struct ahash_request *req)
{
const struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
if (!req->result)
return;
memcpy(req->result, ctx->digest, ctx->nregs * HASH_REG_SIZEOF);
}
/**
* s5p_hash_dma_flush() - flush HASH DMA
* @dev: secss device
*/
static void s5p_hash_dma_flush(struct s5p_aes_dev *dev)
{
SSS_WRITE(dev, FCHRDMAC, SSS_FCHRDMAC_FLUSH);
}
/**
* s5p_hash_dma_enable() - enable DMA mode for HASH
* @dev: secss device
*
* enable DMA mode for HASH
*/
static void s5p_hash_dma_enable(struct s5p_aes_dev *dev)
{
s5p_hash_write(dev, SSS_REG_HASH_CTRL_FIFO, SSS_HASH_FIFO_MODE_DMA);
}
/**
* s5p_hash_irq_disable() - disable irq HASH signals
* @dev: secss device
* @flags: bitfield with irq's to be disabled
*/
static void s5p_hash_irq_disable(struct s5p_aes_dev *dev, u32 flags)
{
SSS_WRITE(dev, FCINTENCLR, flags);
}
/**
* s5p_hash_irq_enable() - enable irq signals
* @dev: secss device
* @flags: bitfield with irq's to be enabled
*/
static void s5p_hash_irq_enable(struct s5p_aes_dev *dev, int flags)
{
SSS_WRITE(dev, FCINTENSET, flags);
}
/**
* s5p_hash_set_flow() - set flow inside SecSS AES/DES with/without HASH
* @dev: secss device
* @hashflow: HASH stream flow with/without crypto AES/DES
*/
static void s5p_hash_set_flow(struct s5p_aes_dev *dev, u32 hashflow)
{
unsigned long flags;
u32 flow;
spin_lock_irqsave(&dev->lock, flags);
flow = SSS_READ(dev, FCFIFOCTRL);
flow &= ~SSS_HASHIN_MASK;
flow |= hashflow;
SSS_WRITE(dev, FCFIFOCTRL, flow);
spin_unlock_irqrestore(&dev->lock, flags);
}
/**
* s5p_ahash_dma_init() - enable DMA and set HASH flow inside SecSS
* @dev: secss device
* @hashflow: HASH stream flow with/without AES/DES
*
* flush HASH DMA and enable DMA, set HASH stream flow inside SecSS HW,
* enable HASH irq's HRDMA, HDONE, HPART
*/
static void s5p_ahash_dma_init(struct s5p_aes_dev *dev, u32 hashflow)
{
s5p_hash_irq_disable(dev, SSS_FCINTENCLR_HRDMAINTENCLR |
SSS_FCINTENCLR_HDONEINTENCLR |
SSS_FCINTENCLR_HPARTINTENCLR);
s5p_hash_dma_flush(dev);
s5p_hash_dma_enable(dev);
s5p_hash_set_flow(dev, hashflow & SSS_HASHIN_MASK);
s5p_hash_irq_enable(dev, SSS_FCINTENSET_HRDMAINTENSET |
SSS_FCINTENSET_HDONEINTENSET |
SSS_FCINTENSET_HPARTINTENSET);
}
/**
* s5p_hash_write_ctrl() - prepare HASH block in SecSS for processing
* @dd: secss device
* @length: length for request
* @final: true if final op
*
* Prepare SSS HASH block for processing bytes in DMA mode. If it is called
* after previous updates, fill up IV words. For final, calculate and set
* lengths for HASH so SecSS can finalize hash. For partial, set SSS HASH
* length as 2^63 so it will be never reached and set to zero prelow and
* prehigh.
*
* This function does not start DMA transfer.
*/
static void s5p_hash_write_ctrl(struct s5p_aes_dev *dd, size_t length,
bool final)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(dd->hash_req);
u32 prelow, prehigh, low, high;
u32 configflags, swapflags;
u64 tmplen;
configflags = ctx->engine | SSS_HASH_INIT_BIT;
if (likely(ctx->digcnt)) {
s5p_hash_write_ctx_iv(dd, ctx);
configflags |= SSS_HASH_USER_IV_EN;
}
if (final) {
/* number of bytes for last part */
low = length;
high = 0;
/* total number of bits prev hashed */
tmplen = ctx->digcnt * 8;
prelow = (u32)tmplen;
prehigh = (u32)(tmplen >> 32);
} else {
prelow = 0;
prehigh = 0;
low = 0;
high = BIT(31);
}
swapflags = SSS_HASH_BYTESWAP_DI | SSS_HASH_BYTESWAP_DO |
SSS_HASH_BYTESWAP_IV | SSS_HASH_BYTESWAP_KEY;
s5p_hash_write(dd, SSS_REG_HASH_MSG_SIZE_LOW, low);
s5p_hash_write(dd, SSS_REG_HASH_MSG_SIZE_HIGH, high);
s5p_hash_write(dd, SSS_REG_HASH_PRE_MSG_SIZE_LOW, prelow);
s5p_hash_write(dd, SSS_REG_HASH_PRE_MSG_SIZE_HIGH, prehigh);
s5p_hash_write(dd, SSS_REG_HASH_CTRL_SWAP, swapflags);
s5p_hash_write(dd, SSS_REG_HASH_CTRL, configflags);
}
/**
* s5p_hash_xmit_dma() - start DMA hash processing
* @dd: secss device
* @length: length for request
* @final: true if final op
*
* Update digcnt here, as it is needed for finup/final op.
*/
static int s5p_hash_xmit_dma(struct s5p_aes_dev *dd, size_t length,
bool final)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(dd->hash_req);
unsigned int cnt;
cnt = dma_map_sg(dd->dev, ctx->sg, ctx->sg_len, DMA_TO_DEVICE);
if (!cnt) {
dev_err(dd->dev, "dma_map_sg error\n");
ctx->error = true;
return -EINVAL;
}
set_bit(HASH_FLAGS_DMA_ACTIVE, &dd->hash_flags);
dd->hash_sg_iter = ctx->sg;
dd->hash_sg_cnt = cnt;
s5p_hash_write_ctrl(dd, length, final);
ctx->digcnt += length;
ctx->total -= length;
/* catch last interrupt */
if (final)
set_bit(HASH_FLAGS_FINAL, &dd->hash_flags);
s5p_set_dma_hashdata(dd, dd->hash_sg_iter); /* DMA starts */
return -EINPROGRESS;
}
/**
* s5p_hash_copy_sgs() - copy request's bytes into new buffer
* @ctx: request context
* @sg: source scatterlist request
* @new_len: number of bytes to process from sg
*
* Allocate new buffer, copy data for HASH into it. If there was xmit_buf
* filled, copy it first, then copy data from sg into it. Prepare one sgl[0]
* with allocated buffer.
*
* Set bit in dd->hash_flag so we can free it after irq ends processing.
*/
static int s5p_hash_copy_sgs(struct s5p_hash_reqctx *ctx,
struct scatterlist *sg, unsigned int new_len)
{
unsigned int pages, len;
void *buf;
len = new_len + ctx->bufcnt;
pages = get_order(len);
buf = (void *)__get_free_pages(GFP_ATOMIC, pages);
if (!buf) {
dev_err(ctx->dd->dev, "alloc pages for unaligned case.\n");
ctx->error = true;
return -ENOMEM;
}
if (ctx->bufcnt)
memcpy(buf, ctx->dd->xmit_buf, ctx->bufcnt);
scatterwalk_map_and_copy(buf + ctx->bufcnt, sg, ctx->skip,
new_len, 0);
sg_init_table(ctx->sgl, 1);
sg_set_buf(ctx->sgl, buf, len);
ctx->sg = ctx->sgl;
ctx->sg_len = 1;
ctx->bufcnt = 0;
ctx->skip = 0;
set_bit(HASH_FLAGS_SGS_COPIED, &ctx->dd->hash_flags);
return 0;
}
/**
* s5p_hash_copy_sg_lists() - copy sg list and make fixes in copy
* @ctx: request context
* @sg: source scatterlist request
* @new_len: number of bytes to process from sg
*
* Allocate new scatterlist table, copy data for HASH into it. If there was
* xmit_buf filled, prepare it first, then copy page, length and offset from
* source sg into it, adjusting begin and/or end for skip offset and
* hash_later value.
*
* Resulting sg table will be assigned to ctx->sg. Set flag so we can free
* it after irq ends processing.
*/
static int s5p_hash_copy_sg_lists(struct s5p_hash_reqctx *ctx,
struct scatterlist *sg, unsigned int new_len)
{
unsigned int skip = ctx->skip, n = sg_nents(sg);
struct scatterlist *tmp;
unsigned int len;
if (ctx->bufcnt)
n++;
ctx->sg = kmalloc_array(n, sizeof(*sg), GFP_KERNEL);
if (!ctx->sg) {
ctx->error = true;
return -ENOMEM;
}
sg_init_table(ctx->sg, n);
tmp = ctx->sg;
ctx->sg_len = 0;
if (ctx->bufcnt) {
sg_set_buf(tmp, ctx->dd->xmit_buf, ctx->bufcnt);
tmp = sg_next(tmp);
ctx->sg_len++;
}
while (sg && skip >= sg->length) {
skip -= sg->length;
sg = sg_next(sg);
}
while (sg && new_len) {
len = sg->length - skip;
if (new_len < len)
len = new_len;
new_len -= len;
sg_set_page(tmp, sg_page(sg), len, sg->offset + skip);
skip = 0;
if (new_len <= 0)
sg_mark_end(tmp);
tmp = sg_next(tmp);
ctx->sg_len++;
sg = sg_next(sg);
}
set_bit(HASH_FLAGS_SGS_ALLOCED, &ctx->dd->hash_flags);
return 0;
}
/**
* s5p_hash_prepare_sgs() - prepare sg for processing
* @ctx: request context
* @sg: source scatterlist request
* @nbytes: number of bytes to process from sg
* @final: final flag
*
* Check two conditions: (1) if buffers in sg have len aligned data, and (2)
* sg table have good aligned elements (list_ok). If one of this checks fails,
* then either (1) allocates new buffer for data with s5p_hash_copy_sgs, copy
* data into this buffer and prepare request in sgl, or (2) allocates new sg
* table and prepare sg elements.
*
* For digest or finup all conditions can be good, and we may not need any
* fixes.
*/
static int s5p_hash_prepare_sgs(struct s5p_hash_reqctx *ctx,
struct scatterlist *sg,
unsigned int new_len, bool final)
{
unsigned int skip = ctx->skip, nbytes = new_len, n = 0;
bool aligned = true, list_ok = true;
struct scatterlist *sg_tmp = sg;
if (!sg || !sg->length || !new_len)
return 0;
if (skip || !final)
list_ok = false;
while (nbytes > 0 && sg_tmp) {
n++;
if (skip >= sg_tmp->length) {
skip -= sg_tmp->length;
if (!sg_tmp->length) {
aligned = false;
break;
}
} else {
if (!IS_ALIGNED(sg_tmp->length - skip, BUFLEN)) {
aligned = false;
break;
}
if (nbytes < sg_tmp->length - skip) {
list_ok = false;
break;
}
nbytes -= sg_tmp->length - skip;
skip = 0;
}
sg_tmp = sg_next(sg_tmp);
}
if (!aligned)
return s5p_hash_copy_sgs(ctx, sg, new_len);
else if (!list_ok)
return s5p_hash_copy_sg_lists(ctx, sg, new_len);
/*
* Have aligned data from previous operation and/or current
* Note: will enter here only if (digest or finup) and aligned
*/
if (ctx->bufcnt) {
ctx->sg_len = n;
sg_init_table(ctx->sgl, 2);
sg_set_buf(ctx->sgl, ctx->dd->xmit_buf, ctx->bufcnt);
sg_chain(ctx->sgl, 2, sg);
ctx->sg = ctx->sgl;
ctx->sg_len++;
} else {
ctx->sg = sg;
ctx->sg_len = n;
}
return 0;
}
/**
* s5p_hash_prepare_request() - prepare request for processing
* @req: AHASH request
* @update: true if UPDATE op
*
* Note 1: we can have update flag _and_ final flag at the same time.
* Note 2: we enter here when digcnt > BUFLEN (=HASH_BLOCK_SIZE) or
* either req->nbytes or ctx->bufcnt + req->nbytes is > BUFLEN or
* we have final op
*/
static int s5p_hash_prepare_request(struct ahash_request *req, bool update)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
bool final = ctx->finup;
int xmit_len, hash_later, nbytes;
int ret;
if (update)
nbytes = req->nbytes;
else
nbytes = 0;
ctx->total = nbytes + ctx->bufcnt;
if (!ctx->total)
return 0;
if (nbytes && (!IS_ALIGNED(ctx->bufcnt, BUFLEN))) {
/* bytes left from previous request, so fill up to BUFLEN */
int len = BUFLEN - ctx->bufcnt % BUFLEN;
if (len > nbytes)
len = nbytes;
scatterwalk_map_and_copy(ctx->buffer + ctx->bufcnt, req->src,
0, len, 0);
ctx->bufcnt += len;
nbytes -= len;
ctx->skip = len;
} else {
ctx->skip = 0;
}
if (ctx->bufcnt)
memcpy(ctx->dd->xmit_buf, ctx->buffer, ctx->bufcnt);
xmit_len = ctx->total;
if (final) {
hash_later = 0;
} else {
if (IS_ALIGNED(xmit_len, BUFLEN))
xmit_len -= BUFLEN;
else
xmit_len -= xmit_len & (BUFLEN - 1);
hash_later = ctx->total - xmit_len;
/* copy hash_later bytes from end of req->src */
/* previous bytes are in xmit_buf, so no overwrite */
scatterwalk_map_and_copy(ctx->buffer, req->src,
req->nbytes - hash_later,
hash_later, 0);
}
if (xmit_len > BUFLEN) {
ret = s5p_hash_prepare_sgs(ctx, req->src, nbytes - hash_later,
final);
if (ret)
return ret;
} else {
/* have buffered data only */
if (unlikely(!ctx->bufcnt)) {
/* first update didn't fill up buffer */
scatterwalk_map_and_copy(ctx->dd->xmit_buf, req->src,
0, xmit_len, 0);
}
sg_init_table(ctx->sgl, 1);
sg_set_buf(ctx->sgl, ctx->dd->xmit_buf, xmit_len);
ctx->sg = ctx->sgl;
ctx->sg_len = 1;
}
ctx->bufcnt = hash_later;
if (!final)
ctx->total = xmit_len;
return 0;
}
/**
* s5p_hash_update_dma_stop() - unmap DMA
* @dd: secss device
*
* Unmap scatterlist ctx->sg.
*/
static void s5p_hash_update_dma_stop(struct s5p_aes_dev *dd)
{
const struct s5p_hash_reqctx *ctx = ahash_request_ctx(dd->hash_req);
dma_unmap_sg(dd->dev, ctx->sg, ctx->sg_len, DMA_TO_DEVICE);
clear_bit(HASH_FLAGS_DMA_ACTIVE, &dd->hash_flags);
}
/**
* s5p_hash_finish() - copy calculated digest to crypto layer
* @req: AHASH request
*/
static void s5p_hash_finish(struct ahash_request *req)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
struct s5p_aes_dev *dd = ctx->dd;
if (ctx->digcnt)
s5p_hash_copy_result(req);
dev_dbg(dd->dev, "hash_finish digcnt: %lld\n", ctx->digcnt);
}
/**
* s5p_hash_finish_req() - finish request
* @req: AHASH request
* @err: error
*/
static void s5p_hash_finish_req(struct ahash_request *req, int err)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
struct s5p_aes_dev *dd = ctx->dd;
unsigned long flags;
if (test_bit(HASH_FLAGS_SGS_COPIED, &dd->hash_flags))
free_pages((unsigned long)sg_virt(ctx->sg),
get_order(ctx->sg->length));
if (test_bit(HASH_FLAGS_SGS_ALLOCED, &dd->hash_flags))
kfree(ctx->sg);
ctx->sg = NULL;
dd->hash_flags &= ~(BIT(HASH_FLAGS_SGS_ALLOCED) |
BIT(HASH_FLAGS_SGS_COPIED));
if (!err && !ctx->error) {
s5p_hash_read_msg(req);
if (test_bit(HASH_FLAGS_FINAL, &dd->hash_flags))
s5p_hash_finish(req);
} else {
ctx->error = true;
}
spin_lock_irqsave(&dd->hash_lock, flags);
dd->hash_flags &= ~(BIT(HASH_FLAGS_BUSY) | BIT(HASH_FLAGS_FINAL) |
BIT(HASH_FLAGS_DMA_READY) |
BIT(HASH_FLAGS_OUTPUT_READY));
spin_unlock_irqrestore(&dd->hash_lock, flags);
if (req->base.complete)
req->base.complete(&req->base, err);
}
/**
* s5p_hash_handle_queue() - handle hash queue
* @dd: device s5p_aes_dev
* @req: AHASH request
*
* If req!=NULL enqueue it on dd->queue, if FLAGS_BUSY is not set on the
* device then processes the first request from the dd->queue
*
* Returns: see s5p_hash_final below.
*/
static int s5p_hash_handle_queue(struct s5p_aes_dev *dd,
struct ahash_request *req)
{
struct crypto_async_request *async_req, *backlog;
struct s5p_hash_reqctx *ctx;
unsigned long flags;
int err = 0, ret = 0;
retry:
spin_lock_irqsave(&dd->hash_lock, flags);
if (req)
ret = ahash_enqueue_request(&dd->hash_queue, req);
if (test_bit(HASH_FLAGS_BUSY, &dd->hash_flags)) {
spin_unlock_irqrestore(&dd->hash_lock, flags);
return ret;
}
backlog = crypto_get_backlog(&dd->hash_queue);
async_req = crypto_dequeue_request(&dd->hash_queue);
if (async_req)
set_bit(HASH_FLAGS_BUSY, &dd->hash_flags);
spin_unlock_irqrestore(&dd->hash_lock, flags);
if (!async_req)
return ret;
if (backlog)
backlog->complete(backlog, -EINPROGRESS);
req = ahash_request_cast(async_req);
dd->hash_req = req;
ctx = ahash_request_ctx(req);
err = s5p_hash_prepare_request(req, ctx->op_update);
if (err || !ctx->total)
goto out;
dev_dbg(dd->dev, "handling new req, op_update: %u, nbytes: %d\n",
ctx->op_update, req->nbytes);
s5p_ahash_dma_init(dd, SSS_HASHIN_INDEPENDENT);
if (ctx->digcnt)
s5p_hash_write_iv(req); /* restore hash IV */
if (ctx->op_update) { /* HASH_OP_UPDATE */
err = s5p_hash_xmit_dma(dd, ctx->total, ctx->finup);
if (err != -EINPROGRESS && ctx->finup && !ctx->error)
/* no final() after finup() */
err = s5p_hash_xmit_dma(dd, ctx->total, true);
} else { /* HASH_OP_FINAL */
err = s5p_hash_xmit_dma(dd, ctx->total, true);
}
out:
if (err != -EINPROGRESS) {
/* hash_tasklet_cb will not finish it, so do it here */
s5p_hash_finish_req(req, err);
req = NULL;
/*
* Execute next request immediately if there is anything
* in queue.
*/
goto retry;
}
return ret;
}
/**
* s5p_hash_tasklet_cb() - hash tasklet
* @data: ptr to s5p_aes_dev
*/
static void s5p_hash_tasklet_cb(unsigned long data)
{
struct s5p_aes_dev *dd = (struct s5p_aes_dev *)data;
if (!test_bit(HASH_FLAGS_BUSY, &dd->hash_flags)) {
s5p_hash_handle_queue(dd, NULL);
return;
}
if (test_bit(HASH_FLAGS_DMA_READY, &dd->hash_flags)) {
if (test_and_clear_bit(HASH_FLAGS_DMA_ACTIVE,
&dd->hash_flags)) {
s5p_hash_update_dma_stop(dd);
}
if (test_and_clear_bit(HASH_FLAGS_OUTPUT_READY,
&dd->hash_flags)) {
/* hash or semi-hash ready */
clear_bit(HASH_FLAGS_DMA_READY, &dd->hash_flags);
goto finish;
}
}
return;
finish:
/* finish curent request */
s5p_hash_finish_req(dd->hash_req, 0);
/* If we are not busy, process next req */
if (!test_bit(HASH_FLAGS_BUSY, &dd->hash_flags))
s5p_hash_handle_queue(dd, NULL);
}
/**
* s5p_hash_enqueue() - enqueue request
* @req: AHASH request
* @op: operation UPDATE (true) or FINAL (false)
*
* Returns: see s5p_hash_final below.
*/
static int s5p_hash_enqueue(struct ahash_request *req, bool op)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
struct s5p_hash_ctx *tctx = crypto_tfm_ctx(req->base.tfm);
ctx->op_update = op;
return s5p_hash_handle_queue(tctx->dd, req);
}
/**
* s5p_hash_update() - process the hash input data
* @req: AHASH request
*
* If request will fit in buffer, copy it and return immediately
* else enqueue it with OP_UPDATE.
*
* Returns: see s5p_hash_final below.
*/
static int s5p_hash_update(struct ahash_request *req)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
if (!req->nbytes)
return 0;
if (ctx->bufcnt + req->nbytes <= BUFLEN) {
scatterwalk_map_and_copy(ctx->buffer + ctx->bufcnt, req->src,
0, req->nbytes, 0);
ctx->bufcnt += req->nbytes;
return 0;
}
return s5p_hash_enqueue(req, true); /* HASH_OP_UPDATE */
}
/**
* s5p_hash_shash_digest() - calculate shash digest
* @tfm: crypto transformation
* @flags: tfm flags
* @data: input data
* @len: length of data
* @out: output buffer
*/
static int s5p_hash_shash_digest(struct crypto_shash *tfm, u32 flags,
const u8 *data, unsigned int len, u8 *out)
{
SHASH_DESC_ON_STACK(shash, tfm);
shash->tfm = tfm;
shash->flags = flags & ~CRYPTO_TFM_REQ_MAY_SLEEP;
return crypto_shash_digest(shash, data, len, out);
}
/**
* s5p_hash_final_shash() - calculate shash digest
* @req: AHASH request
*/
static int s5p_hash_final_shash(struct ahash_request *req)
{
struct s5p_hash_ctx *tctx = crypto_tfm_ctx(req->base.tfm);
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
return s5p_hash_shash_digest(tctx->fallback, req->base.flags,
ctx->buffer, ctx->bufcnt, req->result);
}
/**
* s5p_hash_final() - close up hash and calculate digest
* @req: AHASH request
*
* Note: in final req->src do not have any data, and req->nbytes can be
* non-zero.
*
* If there were no input data processed yet and the buffered hash data is
* less than BUFLEN (64) then calculate the final hash immediately by using
* SW algorithm fallback.
*
* Otherwise enqueues the current AHASH request with OP_FINAL operation op
* and finalize hash message in HW. Note that if digcnt!=0 then there were
* previous update op, so there are always some buffered bytes in ctx->buffer,
* which means that ctx->bufcnt!=0
*
* Returns:
* 0 if the request has been processed immediately,
* -EINPROGRESS if the operation has been queued for later execution or is set
* to processing by HW,
* -EBUSY if queue is full and request should be resubmitted later,
* other negative values denotes an error.
*/
static int s5p_hash_final(struct ahash_request *req)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
ctx->finup = true;
if (ctx->error)
return -EINVAL; /* uncompleted hash is not needed */
if (!ctx->digcnt && ctx->bufcnt < BUFLEN)
return s5p_hash_final_shash(req);
return s5p_hash_enqueue(req, false); /* HASH_OP_FINAL */
}
/**
* s5p_hash_finup() - process last req->src and calculate digest
* @req: AHASH request containing the last update data
*
* Return values: see s5p_hash_final above.
*/
static int s5p_hash_finup(struct ahash_request *req)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
int err1, err2;
ctx->finup = true;
err1 = s5p_hash_update(req);
if (err1 == -EINPROGRESS || err1 == -EBUSY)
return err1;
/*
* final() has to be always called to cleanup resources even if
* update() failed, except EINPROGRESS or calculate digest for small
* size
*/
err2 = s5p_hash_final(req);
return err1 ?: err2;
}
/**
* s5p_hash_init() - initialize AHASH request contex
* @req: AHASH request
*
* Init async hash request context.
*/
static int s5p_hash_init(struct ahash_request *req)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
struct s5p_hash_ctx *tctx = crypto_ahash_ctx(tfm);
ctx->dd = tctx->dd;
ctx->error = false;
ctx->finup = false;
ctx->bufcnt = 0;
ctx->digcnt = 0;
ctx->total = 0;
ctx->skip = 0;
dev_dbg(tctx->dd->dev, "init: digest size: %d\n",
crypto_ahash_digestsize(tfm));
switch (crypto_ahash_digestsize(tfm)) {
case MD5_DIGEST_SIZE:
ctx->engine = SSS_HASH_ENGINE_MD5;
ctx->nregs = HASH_MD5_MAX_REG;
break;
case SHA1_DIGEST_SIZE:
ctx->engine = SSS_HASH_ENGINE_SHA1;
ctx->nregs = HASH_SHA1_MAX_REG;
break;
case SHA256_DIGEST_SIZE:
ctx->engine = SSS_HASH_ENGINE_SHA256;
ctx->nregs = HASH_SHA256_MAX_REG;
break;
default:
ctx->error = true;
return -EINVAL;
}
return 0;
}
/**
* s5p_hash_digest - calculate digest from req->src
* @req: AHASH request
*
* Return values: see s5p_hash_final above.
*/
static int s5p_hash_digest(struct ahash_request *req)
{
return s5p_hash_init(req) ?: s5p_hash_finup(req);
}
/**
* s5p_hash_cra_init_alg - init crypto alg transformation
* @tfm: crypto transformation
*/
static int s5p_hash_cra_init_alg(struct crypto_tfm *tfm)
{
struct s5p_hash_ctx *tctx = crypto_tfm_ctx(tfm);
const char *alg_name = crypto_tfm_alg_name(tfm);
tctx->dd = s5p_dev;
/* Allocate a fallback and abort if it failed. */
tctx->fallback = crypto_alloc_shash(alg_name, 0,
CRYPTO_ALG_NEED_FALLBACK);
if (IS_ERR(tctx->fallback)) {
pr_err("fallback alloc fails for '%s'\n", alg_name);
return PTR_ERR(tctx->fallback);
}
crypto_ahash_set_reqsize(__crypto_ahash_cast(tfm),
sizeof(struct s5p_hash_reqctx) + BUFLEN);
return 0;
}
/**
* s5p_hash_cra_init - init crypto tfm
* @tfm: crypto transformation
*/
static int s5p_hash_cra_init(struct crypto_tfm *tfm)
{
return s5p_hash_cra_init_alg(tfm);
}
/**
* s5p_hash_cra_exit - exit crypto tfm
* @tfm: crypto transformation
*
* free allocated fallback
*/
static void s5p_hash_cra_exit(struct crypto_tfm *tfm)
{
struct s5p_hash_ctx *tctx = crypto_tfm_ctx(tfm);
crypto_free_shash(tctx->fallback);
tctx->fallback = NULL;
}
/**
* s5p_hash_export - export hash state
* @req: AHASH request
* @out: buffer for exported state
*/
static int s5p_hash_export(struct ahash_request *req, void *out)
{
const struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
memcpy(out, ctx, sizeof(*ctx) + ctx->bufcnt);
return 0;
}
/**
* s5p_hash_import - import hash state
* @req: AHASH request
* @in: buffer with state to be imported from
*/
static int s5p_hash_import(struct ahash_request *req, const void *in)
{
struct s5p_hash_reqctx *ctx = ahash_request_ctx(req);
struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
struct s5p_hash_ctx *tctx = crypto_ahash_ctx(tfm);
const struct s5p_hash_reqctx *ctx_in = in;
memcpy(ctx, in, sizeof(*ctx) + BUFLEN);
if (ctx_in->bufcnt > BUFLEN) {
ctx->error = true;
return -EINVAL;
}
ctx->dd = tctx->dd;
ctx->error = false;
return 0;
}
static struct ahash_alg algs_sha1_md5_sha256[] = {
{
.init = s5p_hash_init,
.update = s5p_hash_update,
.final = s5p_hash_final,
.finup = s5p_hash_finup,
.digest = s5p_hash_digest,
.export = s5p_hash_export,
.import = s5p_hash_import,
.halg.statesize = sizeof(struct s5p_hash_reqctx) + BUFLEN,
.halg.digestsize = SHA1_DIGEST_SIZE,
.halg.base = {
.cra_name = "sha1",
.cra_driver_name = "exynos-sha1",
.cra_priority = 100,
.cra_flags = CRYPTO_ALG_KERN_DRIVER_ONLY |
CRYPTO_ALG_ASYNC |
CRYPTO_ALG_NEED_FALLBACK,
.cra_blocksize = HASH_BLOCK_SIZE,
.cra_ctxsize = sizeof(struct s5p_hash_ctx),
.cra_alignmask = SSS_HASH_DMA_ALIGN_MASK,
.cra_module = THIS_MODULE,
.cra_init = s5p_hash_cra_init,
.cra_exit = s5p_hash_cra_exit,
}
},
{
.init = s5p_hash_init,
.update = s5p_hash_update,
.final = s5p_hash_final,
.finup = s5p_hash_finup,
.digest = s5p_hash_digest,
.export = s5p_hash_export,
.import = s5p_hash_import,
.halg.statesize = sizeof(struct s5p_hash_reqctx) + BUFLEN,
.halg.digestsize = MD5_DIGEST_SIZE,
.halg.base = {
.cra_name = "md5",
.cra_driver_name = "exynos-md5",
.cra_priority = 100,
.cra_flags = CRYPTO_ALG_KERN_DRIVER_ONLY |
CRYPTO_ALG_ASYNC |
CRYPTO_ALG_NEED_FALLBACK,
.cra_blocksize = HASH_BLOCK_SIZE,
.cra_ctxsize = sizeof(struct s5p_hash_ctx),
.cra_alignmask = SSS_HASH_DMA_ALIGN_MASK,
.cra_module = THIS_MODULE,
.cra_init = s5p_hash_cra_init,
.cra_exit = s5p_hash_cra_exit,
}
},
{
.init = s5p_hash_init,
.update = s5p_hash_update,
.final = s5p_hash_final,
.finup = s5p_hash_finup,
.digest = s5p_hash_digest,
.export = s5p_hash_export,
.import = s5p_hash_import,
.halg.statesize = sizeof(struct s5p_hash_reqctx) + BUFLEN,
.halg.digestsize = SHA256_DIGEST_SIZE,
.halg.base = {
.cra_name = "sha256",
.cra_driver_name = "exynos-sha256",
.cra_priority = 100,
.cra_flags = CRYPTO_ALG_KERN_DRIVER_ONLY |
CRYPTO_ALG_ASYNC |
CRYPTO_ALG_NEED_FALLBACK,
.cra_blocksize = HASH_BLOCK_SIZE,
.cra_ctxsize = sizeof(struct s5p_hash_ctx),
.cra_alignmask = SSS_HASH_DMA_ALIGN_MASK,
.cra_module = THIS_MODULE,
.cra_init = s5p_hash_cra_init,
.cra_exit = s5p_hash_cra_exit,
}
}
};
static void s5p_set_aes(struct s5p_aes_dev *dev,
const uint8_t *key, const uint8_t *iv,
unsigned int keylen)
{
void __iomem *keystart;
if (iv)
memcpy_toio(dev->aes_ioaddr + SSS_REG_AES_IV_DATA(0), iv, 0x10);
if (keylen == AES_KEYSIZE_256)
keystart = dev->aes_ioaddr + SSS_REG_AES_KEY_DATA(0);
else if (keylen == AES_KEYSIZE_192)
keystart = dev->aes_ioaddr + SSS_REG_AES_KEY_DATA(2);
else
keystart = dev->aes_ioaddr + SSS_REG_AES_KEY_DATA(4);
memcpy_toio(keystart, key, keylen);
}
static bool s5p_is_sg_aligned(struct scatterlist *sg)
{
while (sg) {
if (!IS_ALIGNED(sg->length, AES_BLOCK_SIZE))
return false;
sg = sg_next(sg);
}
return true;
}
static int s5p_set_indata_start(struct s5p_aes_dev *dev,
struct ablkcipher_request *req)
{
struct scatterlist *sg;
int err;
dev->sg_src_cpy = NULL;
sg = req->src;
if (!s5p_is_sg_aligned(sg)) {
dev_dbg(dev->dev,
"At least one unaligned source scatter list, making a copy\n");
err = s5p_make_sg_cpy(dev, sg, &dev->sg_src_cpy);
if (err)
return err;
sg = dev->sg_src_cpy;
}
err = s5p_set_indata(dev, sg);
if (err) {
s5p_free_sg_cpy(dev, &dev->sg_src_cpy);
return err;
}
return 0;
}
static int s5p_set_outdata_start(struct s5p_aes_dev *dev,
struct ablkcipher_request *req)
{
struct scatterlist *sg;
int err;
dev->sg_dst_cpy = NULL;
sg = req->dst;
if (!s5p_is_sg_aligned(sg)) {
dev_dbg(dev->dev,
"At least one unaligned dest scatter list, making a copy\n");
err = s5p_make_sg_cpy(dev, sg, &dev->sg_dst_cpy);
if (err)
return err;
sg = dev->sg_dst_cpy;
}
err = s5p_set_outdata(dev, sg);
if (err) {
s5p_free_sg_cpy(dev, &dev->sg_dst_cpy);
return err;
}
return 0;
}
static void s5p_aes_crypt_start(struct s5p_aes_dev *dev, unsigned long mode)
{
struct ablkcipher_request *req = dev->req;
uint32_t aes_control;
unsigned long flags;
int err;
u8 *iv;
aes_control = SSS_AES_KEY_CHANGE_MODE;
if (mode & FLAGS_AES_DECRYPT)
aes_control |= SSS_AES_MODE_DECRYPT;
if ((mode & FLAGS_AES_MODE_MASK) == FLAGS_AES_CBC) {
aes_control |= SSS_AES_CHAIN_MODE_CBC;
iv = req->info;
} else if ((mode & FLAGS_AES_MODE_MASK) == FLAGS_AES_CTR) {
aes_control |= SSS_AES_CHAIN_MODE_CTR;
iv = req->info;
} else {
iv = NULL; /* AES_ECB */
}
if (dev->ctx->keylen == AES_KEYSIZE_192)
aes_control |= SSS_AES_KEY_SIZE_192;
else if (dev->ctx->keylen == AES_KEYSIZE_256)
aes_control |= SSS_AES_KEY_SIZE_256;
aes_control |= SSS_AES_FIFO_MODE;
/* as a variant it is possible to use byte swapping on DMA side */
aes_control |= SSS_AES_BYTESWAP_DI
| SSS_AES_BYTESWAP_DO
| SSS_AES_BYTESWAP_IV
| SSS_AES_BYTESWAP_KEY
| SSS_AES_BYTESWAP_CNT;
spin_lock_irqsave(&dev->lock, flags);
SSS_WRITE(dev, FCINTENCLR,
SSS_FCINTENCLR_BTDMAINTENCLR | SSS_FCINTENCLR_BRDMAINTENCLR);
SSS_WRITE(dev, FCFIFOCTRL, 0x00);
err = s5p_set_indata_start(dev, req);
if (err)
goto indata_error;
err = s5p_set_outdata_start(dev, req);
if (err)
goto outdata_error;
SSS_AES_WRITE(dev, AES_CONTROL, aes_control);
s5p_set_aes(dev, dev->ctx->aes_key, iv, dev->ctx->keylen);
s5p_set_dma_indata(dev, dev->sg_src);
s5p_set_dma_outdata(dev, dev->sg_dst);
SSS_WRITE(dev, FCINTENSET,
SSS_FCINTENSET_BTDMAINTENSET | SSS_FCINTENSET_BRDMAINTENSET);
spin_unlock_irqrestore(&dev->lock, flags);
return;
outdata_error:
s5p_unset_indata(dev);
indata_error:
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
s5p_sg_done(dev);
dev->busy = false;
spin_unlock_irqrestore(&dev->lock, flags);
crypto: s5p-sss - Fix spinlock recursion on LRW(AES) Running TCRYPT with LRW compiled causes spinlock recursion: testing speed of async lrw(aes) (lrw(ecb-aes-s5p)) encryption tcrypt: test 0 (256 bit key, 16 byte blocks): 19007 operations in 1 seconds (304112 bytes) tcrypt: test 1 (256 bit key, 64 byte blocks): 15753 operations in 1 seconds (1008192 bytes) tcrypt: test 2 (256 bit key, 256 byte blocks): 14293 operations in 1 seconds (3659008 bytes) tcrypt: test 3 (256 bit key, 1024 byte blocks): 11906 operations in 1 seconds (12191744 bytes) tcrypt: test 4 (256 bit key, 8192 byte blocks): BUG: spinlock recursion on CPU#1, irq/84-10830000/89  lock: 0xeea99a68, .magic: dead4ead, .owner: irq/84-10830000/89, .owner_cpu: 1 CPU: 1 PID: 89 Comm: irq/84-10830000 Not tainted 4.11.0-rc1-00001-g897ca6d0800d #559 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [<c010e1ec>] (unwind_backtrace) from [<c010ae1c>] (show_stack+0x10/0x14) [<c010ae1c>] (show_stack) from [<c03449c0>] (dump_stack+0x78/0x8c) [<c03449c0>] (dump_stack) from [<c015de68>] (do_raw_spin_lock+0x11c/0x120) [<c015de68>] (do_raw_spin_lock) from [<c0720110>] (_raw_spin_lock_irqsave+0x20/0x28) [<c0720110>] (_raw_spin_lock_irqsave) from [<c0572ca0>] (s5p_aes_crypt+0x2c/0xb4) [<c0572ca0>] (s5p_aes_crypt) from [<bf1d8aa4>] (do_encrypt+0x78/0xb0 [lrw]) [<bf1d8aa4>] (do_encrypt [lrw]) from [<bf1d8b00>] (encrypt_done+0x24/0x54 [lrw]) [<bf1d8b00>] (encrypt_done [lrw]) from [<c05732a0>] (s5p_aes_complete+0x60/0xcc) [<c05732a0>] (s5p_aes_complete) from [<c0573440>] (s5p_aes_interrupt+0x134/0x1a0) [<c0573440>] (s5p_aes_interrupt) from [<c01667c4>] (irq_thread_fn+0x1c/0x54) [<c01667c4>] (irq_thread_fn) from [<c0166a98>] (irq_thread+0x12c/0x1e0) [<c0166a98>] (irq_thread) from [<c0136a28>] (kthread+0x108/0x138) [<c0136a28>] (kthread) from [<c0107778>] (ret_from_fork+0x14/0x3c) Interrupt handling routine was calling req->base.complete() under spinlock. In most cases this wasn't fatal but when combined with some of the cipher modes (like LRW) this caused recursion - starting the new encryption (s5p_aes_crypt()) while still holding the spinlock from previous round (s5p_aes_complete()). Beside that, the s5p_aes_interrupt() error handling path could execute two completions in case of error for RX and TX blocks. Rewrite the interrupt handling routine and the completion by: 1. Splitting the operations on scatterlist copies from s5p_aes_complete() into separate s5p_sg_done(). This still should be done under lock. The s5p_aes_complete() now only calls req->base.complete() and it has to be called outside of lock. 2. Moving the s5p_aes_complete() out of spinlock critical sections. In interrupt service routine s5p_aes_interrupts(), it appeared in few places, including error paths inside other functions called from ISR. This code was not so obvious to read so simplify it by putting the s5p_aes_complete() only within ISR level. Reported-by: Nathan Royce <nroycea+kernel@gmail.com> Cc: <stable@vger.kernel.org> # v4.10.x: 07de4bc88c crypto: s5p-sss - Fix completing Cc: <stable@vger.kernel.org> # v4.10.x Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2017-03-09 05:14:20 +08:00
s5p_aes_complete(dev, err);
}
static void s5p_tasklet_cb(unsigned long data)
{
struct s5p_aes_dev *dev = (struct s5p_aes_dev *)data;
struct crypto_async_request *async_req, *backlog;
struct s5p_aes_reqctx *reqctx;
unsigned long flags;
spin_lock_irqsave(&dev->lock, flags);
backlog = crypto_get_backlog(&dev->queue);
async_req = crypto_dequeue_request(&dev->queue);
if (!async_req) {
dev->busy = false;
spin_unlock_irqrestore(&dev->lock, flags);
return;
}
spin_unlock_irqrestore(&dev->lock, flags);
if (backlog)
backlog->complete(backlog, -EINPROGRESS);
dev->req = ablkcipher_request_cast(async_req);
dev->ctx = crypto_tfm_ctx(dev->req->base.tfm);
reqctx = ablkcipher_request_ctx(dev->req);
s5p_aes_crypt_start(dev, reqctx->mode);
}
static int s5p_aes_handle_req(struct s5p_aes_dev *dev,
struct ablkcipher_request *req)
{
unsigned long flags;
int err;
spin_lock_irqsave(&dev->lock, flags);
err = ablkcipher_enqueue_request(&dev->queue, req);
if (dev->busy) {
spin_unlock_irqrestore(&dev->lock, flags);
goto exit;
}
dev->busy = true;
spin_unlock_irqrestore(&dev->lock, flags);
tasklet_schedule(&dev->tasklet);
exit:
return err;
}
static int s5p_aes_crypt(struct ablkcipher_request *req, unsigned long mode)
{
struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
struct s5p_aes_reqctx *reqctx = ablkcipher_request_ctx(req);
struct s5p_aes_ctx *ctx = crypto_ablkcipher_ctx(tfm);
struct s5p_aes_dev *dev = ctx->dev;
if (!IS_ALIGNED(req->nbytes, AES_BLOCK_SIZE)) {
dev_err(dev->dev, "request size is not exact amount of AES blocks\n");
return -EINVAL;
}
reqctx->mode = mode;
return s5p_aes_handle_req(dev, req);
}
static int s5p_aes_setkey(struct crypto_ablkcipher *cipher,
const uint8_t *key, unsigned int keylen)
{
struct crypto_tfm *tfm = crypto_ablkcipher_tfm(cipher);
struct s5p_aes_ctx *ctx = crypto_tfm_ctx(tfm);
if (keylen != AES_KEYSIZE_128 &&
keylen != AES_KEYSIZE_192 &&
keylen != AES_KEYSIZE_256)
return -EINVAL;
memcpy(ctx->aes_key, key, keylen);
ctx->keylen = keylen;
return 0;
}
static int s5p_aes_ecb_encrypt(struct ablkcipher_request *req)
{
return s5p_aes_crypt(req, 0);
}
static int s5p_aes_ecb_decrypt(struct ablkcipher_request *req)
{
return s5p_aes_crypt(req, FLAGS_AES_DECRYPT);
}
static int s5p_aes_cbc_encrypt(struct ablkcipher_request *req)
{
return s5p_aes_crypt(req, FLAGS_AES_CBC);
}
static int s5p_aes_cbc_decrypt(struct ablkcipher_request *req)
{
return s5p_aes_crypt(req, FLAGS_AES_DECRYPT | FLAGS_AES_CBC);
}
static int s5p_aes_cra_init(struct crypto_tfm *tfm)
{
struct s5p_aes_ctx *ctx = crypto_tfm_ctx(tfm);
ctx->dev = s5p_dev;
tfm->crt_ablkcipher.reqsize = sizeof(struct s5p_aes_reqctx);
return 0;
}
static struct crypto_alg algs[] = {
{
.cra_name = "ecb(aes)",
.cra_driver_name = "ecb-aes-s5p",
.cra_priority = 100,
.cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER |
CRYPTO_ALG_ASYNC |
CRYPTO_ALG_KERN_DRIVER_ONLY,
.cra_blocksize = AES_BLOCK_SIZE,
.cra_ctxsize = sizeof(struct s5p_aes_ctx),
.cra_alignmask = 0x0f,
.cra_type = &crypto_ablkcipher_type,
.cra_module = THIS_MODULE,
.cra_init = s5p_aes_cra_init,
.cra_u.ablkcipher = {
.min_keysize = AES_MIN_KEY_SIZE,
.max_keysize = AES_MAX_KEY_SIZE,
.setkey = s5p_aes_setkey,
.encrypt = s5p_aes_ecb_encrypt,
.decrypt = s5p_aes_ecb_decrypt,
}
},
{
.cra_name = "cbc(aes)",
.cra_driver_name = "cbc-aes-s5p",
.cra_priority = 100,
.cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER |
CRYPTO_ALG_ASYNC |
CRYPTO_ALG_KERN_DRIVER_ONLY,
.cra_blocksize = AES_BLOCK_SIZE,
.cra_ctxsize = sizeof(struct s5p_aes_ctx),
.cra_alignmask = 0x0f,
.cra_type = &crypto_ablkcipher_type,
.cra_module = THIS_MODULE,
.cra_init = s5p_aes_cra_init,
.cra_u.ablkcipher = {
.min_keysize = AES_MIN_KEY_SIZE,
.max_keysize = AES_MAX_KEY_SIZE,
.ivsize = AES_BLOCK_SIZE,
.setkey = s5p_aes_setkey,
.encrypt = s5p_aes_cbc_encrypt,
.decrypt = s5p_aes_cbc_decrypt,
}
},
};
static int s5p_aes_probe(struct platform_device *pdev)
{
struct device *dev = &pdev->dev;
int i, j, err = -ENODEV;
const struct samsung_aes_variant *variant;
struct s5p_aes_dev *pdata;
struct resource *res;
unsigned int hash_i;
if (s5p_dev)
return -EEXIST;
pdata = devm_kzalloc(dev, sizeof(*pdata), GFP_KERNEL);
if (!pdata)
return -ENOMEM;
variant = find_s5p_sss_version(pdev);
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
/*
* Note: HASH and PRNG uses the same registers in secss, avoid
* overwrite each other. This will drop HASH when CONFIG_EXYNOS_RNG
* is enabled in config. We need larger size for HASH registers in
* secss, current describe only AES/DES
*/
if (IS_ENABLED(CONFIG_CRYPTO_DEV_EXYNOS_HASH)) {
if (variant == &exynos_aes_data) {
res->end += 0x300;
pdata->use_hash = true;
}
}
pdata->res = res;
pdata->ioaddr = devm_ioremap_resource(&pdev->dev, res);
if (IS_ERR(pdata->ioaddr)) {
if (!pdata->use_hash)
return PTR_ERR(pdata->ioaddr);
/* try AES without HASH */
res->end -= 0x300;
pdata->use_hash = false;
pdata->ioaddr = devm_ioremap_resource(&pdev->dev, res);
if (IS_ERR(pdata->ioaddr))
return PTR_ERR(pdata->ioaddr);
}
pdata->clk = devm_clk_get(dev, "secss");
if (IS_ERR(pdata->clk)) {
dev_err(dev, "failed to find secss clock source\n");
return -ENOENT;
}
err = clk_prepare_enable(pdata->clk);
if (err < 0) {
dev_err(dev, "Enabling SSS clk failed, err %d\n", err);
return err;
}
spin_lock_init(&pdata->lock);
spin_lock_init(&pdata->hash_lock);
pdata->aes_ioaddr = pdata->ioaddr + variant->aes_offset;
pdata->io_hash_base = pdata->ioaddr + variant->hash_offset;
pdata->irq_fc = platform_get_irq(pdev, 0);
if (pdata->irq_fc < 0) {
err = pdata->irq_fc;
dev_warn(dev, "feed control interrupt is not available.\n");
goto err_irq;
}
err = devm_request_threaded_irq(dev, pdata->irq_fc, NULL,
s5p_aes_interrupt, IRQF_ONESHOT,
pdev->name, pdev);
if (err < 0) {
dev_warn(dev, "feed control interrupt is not available.\n");
goto err_irq;
}
pdata->busy = false;
pdata->dev = dev;
platform_set_drvdata(pdev, pdata);
s5p_dev = pdata;
tasklet_init(&pdata->tasklet, s5p_tasklet_cb, (unsigned long)pdata);
crypto_init_queue(&pdata->queue, CRYPTO_QUEUE_LEN);
for (i = 0; i < ARRAY_SIZE(algs); i++) {
err = crypto_register_alg(&algs[i]);
if (err)
goto err_algs;
}
if (pdata->use_hash) {
tasklet_init(&pdata->hash_tasklet, s5p_hash_tasklet_cb,
(unsigned long)pdata);
crypto_init_queue(&pdata->hash_queue, SSS_HASH_QUEUE_LENGTH);
for (hash_i = 0; hash_i < ARRAY_SIZE(algs_sha1_md5_sha256);
hash_i++) {
struct ahash_alg *alg;
alg = &algs_sha1_md5_sha256[hash_i];
err = crypto_register_ahash(alg);
if (err) {
dev_err(dev, "can't register '%s': %d\n",
alg->halg.base.cra_driver_name, err);
goto err_hash;
}
}
}
dev_info(dev, "s5p-sss driver registered\n");
return 0;
err_hash:
for (j = hash_i - 1; j >= 0; j--)
crypto_unregister_ahash(&algs_sha1_md5_sha256[j]);
tasklet_kill(&pdata->hash_tasklet);
res->end -= 0x300;
err_algs:
if (i < ARRAY_SIZE(algs))
dev_err(dev, "can't register '%s': %d\n", algs[i].cra_name,
err);
for (j = 0; j < i; j++)
crypto_unregister_alg(&algs[j]);
tasklet_kill(&pdata->tasklet);
err_irq:
clk_disable_unprepare(pdata->clk);
s5p_dev = NULL;
return err;
}
static int s5p_aes_remove(struct platform_device *pdev)
{
struct s5p_aes_dev *pdata = platform_get_drvdata(pdev);
int i;
if (!pdata)
return -ENODEV;
for (i = 0; i < ARRAY_SIZE(algs); i++)
crypto_unregister_alg(&algs[i]);
tasklet_kill(&pdata->tasklet);
if (pdata->use_hash) {
for (i = ARRAY_SIZE(algs_sha1_md5_sha256) - 1; i >= 0; i--)
crypto_unregister_ahash(&algs_sha1_md5_sha256[i]);
pdata->res->end -= 0x300;
tasklet_kill(&pdata->hash_tasklet);
pdata->use_hash = false;
}
clk_disable_unprepare(pdata->clk);
s5p_dev = NULL;
return 0;
}
static struct platform_driver s5p_aes_crypto = {
.probe = s5p_aes_probe,
.remove = s5p_aes_remove,
.driver = {
.name = "s5p-secss",
.of_match_table = s5p_sss_dt_match,
},
};
module_platform_driver(s5p_aes_crypto);
MODULE_DESCRIPTION("S5PV210 AES hw acceleration support.");
MODULE_LICENSE("GPL v2");
MODULE_AUTHOR("Vladimir Zapolskiy <vzapolskiy@gmail.com>");
MODULE_AUTHOR("Kamil Konieczny <k.konieczny@partner.samsung.com>");