/*
 * Copyright 2016 Broadcom
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License, version 2, as
 * published by the Free Software Foundation (the "GPL").
 *
 * This program is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License version 2 (GPLv2) for more details.
 *
 * You should have received a copy of the GNU General Public License
 * version 2 (GPLv2) along with this source code.
 */

#include <linux/err.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/interrupt.h>
#include <linux/platform_device.h>
#include <linux/scatterlist.h>
#include <linux/crypto.h>
#include <linux/kthread.h>
#include <linux/rtnetlink.h>
#include <linux/sched.h>
#include <linux/of_address.h>
#include <linux/of_device.h>
#include <linux/io.h>
#include <linux/bitops.h>

#include <crypto/algapi.h>
#include <crypto/aead.h>
#include <crypto/internal/aead.h>
#include <crypto/aes.h>
#include <crypto/des.h>
#include <crypto/hmac.h>
#include <crypto/sha.h>
#include <crypto/md5.h>
#include <crypto/authenc.h>
#include <crypto/skcipher.h>
#include <crypto/hash.h>
#include <crypto/sha3.h>

#include "util.h"
#include "cipher.h"
#include "spu.h"
#include "spum.h"
#include "spu2.h"

/* ================= Device Structure ================== */

struct device_private iproc_priv;

/* ==================== Parameters ===================== */

int flow_debug_logging;
module_param(flow_debug_logging, int, 0644);
MODULE_PARM_DESC(flow_debug_logging, "Enable Flow Debug Logging");

int packet_debug_logging;
module_param(packet_debug_logging, int, 0644);
MODULE_PARM_DESC(packet_debug_logging, "Enable Packet Debug Logging");

int debug_logging_sleep;
module_param(debug_logging_sleep, int, 0644);
MODULE_PARM_DESC(debug_logging_sleep, "Packet Debug Logging Sleep");

/*
 * The value of these module parameters is used to set the priority for each
 * algo type when this driver registers algos with the kernel crypto API.
 * To use a priority other than the default, set the priority in the insmod or
 * modprobe. Changing the module priority after init time has no effect.
 *
 * The default priorities are chosen to be lower (less preferred) than ARMv8 CE
 * algos, but more preferred than generic software algos.
 */
static int cipher_pri = 150;
module_param(cipher_pri, int, 0644);
MODULE_PARM_DESC(cipher_pri, "Priority for cipher algos");

static int hash_pri = 100;
module_param(hash_pri, int, 0644);
MODULE_PARM_DESC(hash_pri, "Priority for hash algos");

static int aead_pri = 150;
module_param(aead_pri, int, 0644);
MODULE_PARM_DESC(aead_pri, "Priority for AEAD algos");

/* A type 3 BCM header, expected to precede the SPU header for SPU-M.
 * Bits 3 and 4 in the first byte encode the channel number (the dma ringset).
 * 0x60 - ring 0
 * 0x68 - ring 1
 * 0x70 - ring 2
 * 0x78 - ring 3
 */
char BCMHEADER[] = { 0x60, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x28 };
/*
 * Some SPU hw does not use BCM header on SPU messages. So BCM_HDR_LEN
 * is set dynamically after reading SPU type from device tree.
 */
#define BCM_HDR_LEN  iproc_priv.bcm_hdr_len

/* min and max time to sleep before retrying when mbox queue is full. usec */
#define MBOX_SLEEP_MIN  800
#define MBOX_SLEEP_MAX 1000

/**
 * select_channel() - Select a SPU channel to handle a crypto request. Selects
 * channel in round robin order.
 *
 * Return:  channel index
 */
static u8 select_channel(void)
{
        u8 chan_idx = atomic_inc_return(&iproc_priv.next_chan);

        return chan_idx % iproc_priv.spu.num_chan;
}

/**
 * spu_ablkcipher_rx_sg_create() - Build up the scatterlist of buffers used to
 * receive a SPU response message for an ablkcipher request. Includes buffers to
 * catch SPU message headers and the response data.
 * @mssg:       mailbox message containing the receive sg
 * @rctx:       crypto request context
 * @rx_frag_num: number of scatterlist elements required to hold the
 *              SPU response message
 * @chunksize:  Number of bytes of response data expected
 * @stat_pad_len: Number of bytes required to pad the STAT field to
 *              a 4-byte boundary
 *
 * The scatterlist that gets allocated here is freed in spu_chunk_cleanup()
 * when the request completes, whether the request is handled successfully or
 * there is an error.
 *
 * Returns:
 *   0 if successful
 *   < 0 if an error
 */
static int
spu_ablkcipher_rx_sg_create(struct brcm_message *mssg,
                            struct iproc_reqctx_s *rctx,
                            u8 rx_frag_num,
                            unsigned int chunksize, u32 stat_pad_len)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct scatterlist *sg; /* used to build sgs in mbox message */
        struct iproc_ctx_s *ctx = rctx->ctx;
        u32 datalen;            /* Number of bytes of response data expected */

        mssg->spu.dst = kcalloc(rx_frag_num, sizeof(struct scatterlist),
                                rctx->gfp);
        if (!mssg->spu.dst)
                return -ENOMEM;

        sg = mssg->spu.dst;
        sg_init_table(sg, rx_frag_num);
        /* Space for SPU message header */
        sg_set_buf(sg++, rctx->msg_buf.spu_resp_hdr, ctx->spu_resp_hdr_len);

        /* If XTS tweak in payload, add buffer to receive encrypted tweak */
        if ((ctx->cipher.mode == CIPHER_MODE_XTS) &&
            spu->spu_xts_tweak_in_payload())
                sg_set_buf(sg++, rctx->msg_buf.c.supdt_tweak,
                           SPU_XTS_TWEAK_SIZE);

        /* Copy in each dst sg entry from request, up to chunksize */
        datalen = spu_msg_sg_add(&sg, &rctx->dst_sg, &rctx->dst_skip,
                                 rctx->dst_nents, chunksize);
        if (datalen < chunksize) {
                pr_err("%s(): failed to copy dst sg to mbox msg. chunksize %u, datalen %u",
                       __func__, chunksize, datalen);
                return -EFAULT;
        }

        if (ctx->cipher.alg == CIPHER_ALG_RC4)
                /* Add buffer to catch 260-byte SUPDT field for RC4 */
                sg_set_buf(sg++, rctx->msg_buf.c.supdt_tweak, SPU_SUPDT_LEN);

        if (stat_pad_len)
                sg_set_buf(sg++, rctx->msg_buf.rx_stat_pad, stat_pad_len);

        memset(rctx->msg_buf.rx_stat, 0, SPU_RX_STATUS_LEN);
        sg_set_buf(sg, rctx->msg_buf.rx_stat, spu->spu_rx_status_len());

        return 0;
}

/**
 * spu_ablkcipher_tx_sg_create() - Build up the scatterlist of buffers used to
 * send a SPU request message for an ablkcipher request. Includes SPU message
 * headers and the request data.
 * @mssg:       mailbox message containing the transmit sg
 * @rctx:       crypto request context
 * @tx_frag_num: number of scatterlist elements required to construct the
 *              SPU request message
 * @chunksize:  Number of bytes of request data
 * @pad_len:    Number of pad bytes
 *
 * The scatterlist that gets allocated here is freed in spu_chunk_cleanup()
 * when the request completes, whether the request is handled successfully or
 * there is an error.
 *
 * Returns:
 *   0 if successful
 *   < 0 if an error
 */
static int
spu_ablkcipher_tx_sg_create(struct brcm_message *mssg,
                            struct iproc_reqctx_s *rctx,
                            u8 tx_frag_num, unsigned int chunksize, u32 pad_len)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct scatterlist *sg; /* used to build sgs in mbox message */
        struct iproc_ctx_s *ctx = rctx->ctx;
        u32 datalen;            /* Number of bytes of response data expected */
        u32 stat_len;

        mssg->spu.src = kcalloc(tx_frag_num, sizeof(struct scatterlist),
                                rctx->gfp);
        if (unlikely(!mssg->spu.src))
                return -ENOMEM;

        sg = mssg->spu.src;
        sg_init_table(sg, tx_frag_num);

        sg_set_buf(sg++, rctx->msg_buf.bcm_spu_req_hdr,
                   BCM_HDR_LEN + ctx->spu_req_hdr_len);

        /* if XTS tweak in payload, copy from IV (where crypto API puts it) */
        if ((ctx->cipher.mode == CIPHER_MODE_XTS) &&
            spu->spu_xts_tweak_in_payload())
                sg_set_buf(sg++, rctx->msg_buf.iv_ctr, SPU_XTS_TWEAK_SIZE);

        /* Copy in each src sg entry from request, up to chunksize */
        datalen = spu_msg_sg_add(&sg, &rctx->src_sg, &rctx->src_skip,
                                 rctx->src_nents, chunksize);
        if (unlikely(datalen < chunksize)) {
                pr_err("%s(): failed to copy src sg to mbox msg",
                       __func__);
                return -EFAULT;
        }

        if (pad_len)
                sg_set_buf(sg++, rctx->msg_buf.spu_req_pad, pad_len);

        stat_len = spu->spu_tx_status_len();
        if (stat_len) {
                memset(rctx->msg_buf.tx_stat, 0, stat_len);
                sg_set_buf(sg, rctx->msg_buf.tx_stat, stat_len);
        }
        return 0;
}

static int mailbox_send_message(struct brcm_message *mssg, u32 flags,
                                u8 chan_idx)
{
        int err;
        int retry_cnt = 0;
        struct device *dev = &(iproc_priv.pdev->dev);

        err = mbox_send_message(iproc_priv.mbox[chan_idx], mssg);
        if (flags & CRYPTO_TFM_REQ_MAY_SLEEP) {
                while ((err == -ENOBUFS) && (retry_cnt < SPU_MB_RETRY_MAX)) {
                        /*
                         * Mailbox queue is full. Since MAY_SLEEP is set, assume
                         * not in atomic context and we can wait and try again.
                         */
                        retry_cnt++;
                        usleep_range(MBOX_SLEEP_MIN, MBOX_SLEEP_MAX);
                        err = mbox_send_message(iproc_priv.mbox[chan_idx],
                                                mssg);
                        atomic_inc(&iproc_priv.mb_no_spc);
                }
        }
        if (err < 0) {
                atomic_inc(&iproc_priv.mb_send_fail);
                return err;
        }

        /* Check error returned by mailbox controller */
        err = mssg->error;
        if (unlikely(err < 0)) {
                dev_err(dev, "message error %d", err);
                /* Signal txdone for mailbox channel */
        }

        /* Signal txdone for mailbox channel */
        mbox_client_txdone(iproc_priv.mbox[chan_idx], err);
        return err;
}

/**
 * handle_ablkcipher_req() - Submit as much of a block cipher request as fits in
 * a single SPU request message, starting at the current position in the request
 * data.
 * @rctx:       Crypto request context
 *
 * This may be called on the crypto API thread, or, when a request is so large
 * it must be broken into multiple SPU messages, on the thread used to invoke
 * the response callback. When requests are broken into multiple SPU
 * messages, we assume subsequent messages depend on previous results, and
 * thus always wait for previous results before submitting the next message.
 * Because requests are submitted in lock step like this, there is no need
 * to synchronize access to request data structures.
 *
 * Return: -EINPROGRESS: request has been accepted and result will be returned
 *                       asynchronously
 *         Any other value indicates an error
 */
static int handle_ablkcipher_req(struct iproc_reqctx_s *rctx)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct crypto_async_request *areq = rctx->parent;
        struct ablkcipher_request *req =
            container_of(areq, struct ablkcipher_request, base);
        struct iproc_ctx_s *ctx = rctx->ctx;
        struct spu_cipher_parms cipher_parms;
        int err = 0;
        unsigned int chunksize = 0;     /* Num bytes of request to submit */
        int remaining = 0;      /* Bytes of request still to process */
        int chunk_start;        /* Beginning of data for current SPU msg */

        /* IV or ctr value to use in this SPU msg */
        u8 local_iv_ctr[MAX_IV_SIZE];
        u32 stat_pad_len;       /* num bytes to align status field */
        u32 pad_len;            /* total length of all padding */
        bool update_key = false;
        struct brcm_message *mssg;      /* mailbox message */

        /* number of entries in src and dst sg in mailbox message. */
        u8 rx_frag_num = 2;     /* response header and STATUS */
        u8 tx_frag_num = 1;     /* request header */

        flow_log("%s\n", __func__);

        cipher_parms.alg = ctx->cipher.alg;
        cipher_parms.mode = ctx->cipher.mode;
        cipher_parms.type = ctx->cipher_type;
        cipher_parms.key_len = ctx->enckeylen;
        cipher_parms.key_buf = ctx->enckey;
        cipher_parms.iv_buf = local_iv_ctr;
        cipher_parms.iv_len = rctx->iv_ctr_len;

        mssg = &rctx->mb_mssg;
        chunk_start = rctx->src_sent;
        remaining = rctx->total_todo - chunk_start;

        /* determine the chunk we are breaking off and update the indexes */
        if ((ctx->max_payload != SPU_MAX_PAYLOAD_INF) &&
            (remaining > ctx->max_payload))
                chunksize = ctx->max_payload;
        else
                chunksize = remaining;

        rctx->src_sent += chunksize;
        rctx->total_sent = rctx->src_sent;

        /* Count number of sg entries to be included in this request */
        rctx->src_nents = spu_sg_count(rctx->src_sg, rctx->src_skip, chunksize);
        rctx->dst_nents = spu_sg_count(rctx->dst_sg, rctx->dst_skip, chunksize);

        if ((ctx->cipher.mode == CIPHER_MODE_CBC) &&
            rctx->is_encrypt && chunk_start)
                /*
                 * Encrypting non-first first chunk. Copy last block of
                 * previous result to IV for this chunk.
                 */
                sg_copy_part_to_buf(req->dst, rctx->msg_buf.iv_ctr,
                                    rctx->iv_ctr_len,
                                    chunk_start - rctx->iv_ctr_len);

        if (rctx->iv_ctr_len) {
                /* get our local copy of the iv */
                __builtin_memcpy(local_iv_ctr, rctx->msg_buf.iv_ctr,
                                 rctx->iv_ctr_len);

                /* generate the next IV if possible */
                if ((ctx->cipher.mode == CIPHER_MODE_CBC) &&
                    !rctx->is_encrypt) {
                        /*
                         * CBC Decrypt: next IV is the last ciphertext block in
                         * this chunk
                         */
                        sg_copy_part_to_buf(req->src, rctx->msg_buf.iv_ctr,
                                            rctx->iv_ctr_len,
                                            rctx->src_sent - rctx->iv_ctr_len);
                } else if (ctx->cipher.mode == CIPHER_MODE_CTR) {
                        /*
                         * The SPU hardware increments the counter once for
                         * each AES block of 16 bytes. So update the counter
                         * for the next chunk, if there is one. Note that for
                         * this chunk, the counter has already been copied to
                         * local_iv_ctr. We can assume a block size of 16,
                         * because we only support CTR mode for AES, not for
                         * any other cipher alg.
                         */
                        add_to_ctr(rctx->msg_buf.iv_ctr, chunksize >> 4);
                }
        }

        if (ctx->cipher.alg == CIPHER_ALG_RC4) {
                rx_frag_num++;
                if (chunk_start) {
                        /*
                         * for non-first RC4 chunks, use SUPDT from previous
                         * response as key for this chunk.
                         */
                        cipher_parms.key_buf = rctx->msg_buf.c.supdt_tweak;
                        update_key = true;
                        cipher_parms.type = CIPHER_TYPE_UPDT;
                } else if (!rctx->is_encrypt) {
                        /*
                         * First RC4 chunk. For decrypt, key in pre-built msg
                         * header may have been changed if encrypt required
                         * multiple chunks. So revert the key to the
                         * ctx->enckey value.
                         */
                        update_key = true;
                        cipher_parms.type = CIPHER_TYPE_INIT;
                }
        }

        if (ctx->max_payload == SPU_MAX_PAYLOAD_INF)
                flow_log("max_payload infinite\n");
        else
                flow_log("max_payload %u\n", ctx->max_payload);

        flow_log("sent:%u start:%u remains:%u size:%u\n",
                 rctx->src_sent, chunk_start, remaining, chunksize);

        /* Copy SPU header template created at setkey time */
        memcpy(rctx->msg_buf.bcm_spu_req_hdr, ctx->bcm_spu_req_hdr,
               sizeof(rctx->msg_buf.bcm_spu_req_hdr));

        /*
         * Pass SUPDT field as key. Key field in finish() call is only used
         * when update_key has been set above for RC4. Will be ignored in
         * all other cases.
         */
        spu->spu_cipher_req_finish(rctx->msg_buf.bcm_spu_req_hdr + BCM_HDR_LEN,
                                   ctx->spu_req_hdr_len, !(rctx->is_encrypt),
                                   &cipher_parms, update_key, chunksize);

        atomic64_add(chunksize, &iproc_priv.bytes_out);

        stat_pad_len = spu->spu_wordalign_padlen(chunksize);
        if (stat_pad_len)
                rx_frag_num++;
        pad_len = stat_pad_len;
        if (pad_len) {
                tx_frag_num++;
                spu->spu_request_pad(rctx->msg_buf.spu_req_pad, 0,
                                     0, ctx->auth.alg, ctx->auth.mode,
                                     rctx->total_sent, stat_pad_len);
        }

        spu->spu_dump_msg_hdr(rctx->msg_buf.bcm_spu_req_hdr + BCM_HDR_LEN,
                              ctx->spu_req_hdr_len);
        packet_log("payload:\n");
        dump_sg(rctx->src_sg, rctx->src_skip, chunksize);
        packet_dump("   pad: ", rctx->msg_buf.spu_req_pad, pad_len);

        /*
         * Build mailbox message containing SPU request msg and rx buffers
         * to catch response message
         */
        memset(mssg, 0, sizeof(*mssg));
        mssg->type = BRCM_MESSAGE_SPU;
        mssg->ctx = rctx;       /* Will be returned in response */

        /* Create rx scatterlist to catch result */
        rx_frag_num += rctx->dst_nents;

        if ((ctx->cipher.mode == CIPHER_MODE_XTS) &&
            spu->spu_xts_tweak_in_payload())
                rx_frag_num++;  /* extra sg to insert tweak */

        err = spu_ablkcipher_rx_sg_create(mssg, rctx, rx_frag_num, chunksize,
                                          stat_pad_len);
        if (err)
                return err;

        /* Create tx scatterlist containing SPU request message */
        tx_frag_num += rctx->src_nents;
        if (spu->spu_tx_status_len())
                tx_frag_num++;

        if ((ctx->cipher.mode == CIPHER_MODE_XTS) &&
            spu->spu_xts_tweak_in_payload())
                tx_frag_num++;  /* extra sg to insert tweak */

        err = spu_ablkcipher_tx_sg_create(mssg, rctx, tx_frag_num, chunksize,
                                          pad_len);
        if (err)
                return err;

        err = mailbox_send_message(mssg, req->base.flags, rctx->chan_idx);
        if (unlikely(err < 0))
                return err;

        return -EINPROGRESS;
}

/**
 * handle_ablkcipher_resp() - Process a block cipher SPU response. Updates the
 * total received count for the request and updates global stats.
 * @rctx:       Crypto request context
 */
static void handle_ablkcipher_resp(struct iproc_reqctx_s *rctx)
{
        struct spu_hw *spu = &iproc_priv.spu;
#ifdef DEBUG
        struct crypto_async_request *areq = rctx->parent;
        struct ablkcipher_request *req = ablkcipher_request_cast(areq);
#endif
        struct iproc_ctx_s *ctx = rctx->ctx;
        u32 payload_len;

        /* See how much data was returned */
        payload_len = spu->spu_payload_length(rctx->msg_buf.spu_resp_hdr);

        /*
         * In XTS mode, the first SPU_XTS_TWEAK_SIZE bytes may be the
         * encrypted tweak ("i") value; we don't count those.
         */
        if ((ctx->cipher.mode == CIPHER_MODE_XTS) &&
            spu->spu_xts_tweak_in_payload() &&
            (payload_len >= SPU_XTS_TWEAK_SIZE))
                payload_len -= SPU_XTS_TWEAK_SIZE;

        atomic64_add(payload_len, &iproc_priv.bytes_in);

        flow_log("%s() offset: %u, bd_len: %u BD:\n",
                 __func__, rctx->total_received, payload_len);

        dump_sg(req->dst, rctx->total_received, payload_len);
        if (ctx->cipher.alg == CIPHER_ALG_RC4)
                packet_dump("  supdt ", rctx->msg_buf.c.supdt_tweak,
                            SPU_SUPDT_LEN);

        rctx->total_received += payload_len;
        if (rctx->total_received == rctx->total_todo) {
                atomic_inc(&iproc_priv.op_counts[SPU_OP_CIPHER]);
                atomic_inc(
                   &iproc_priv.cipher_cnt[ctx->cipher.alg][ctx->cipher.mode]);
        }
}

/**
 * spu_ahash_rx_sg_create() - Build up the scatterlist of buffers used to
 * receive a SPU response message for an ahash request.
 * @mssg:       mailbox message containing the receive sg
 * @rctx:       crypto request context
 * @rx_frag_num: number of scatterlist elements required to hold the
 *              SPU response message
 * @digestsize: length of hash digest, in bytes
 * @stat_pad_len: Number of bytes required to pad the STAT field to
 *              a 4-byte boundary
 *
 * The scatterlist that gets allocated here is freed in spu_chunk_cleanup()
 * when the request completes, whether the request is handled successfully or
 * there is an error.
 *
 * Return:
 *   0 if successful
 *   < 0 if an error
 */
static int
spu_ahash_rx_sg_create(struct brcm_message *mssg,
                       struct iproc_reqctx_s *rctx,
                       u8 rx_frag_num, unsigned int digestsize,
                       u32 stat_pad_len)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct scatterlist *sg; /* used to build sgs in mbox message */
        struct iproc_ctx_s *ctx = rctx->ctx;

        mssg->spu.dst = kcalloc(rx_frag_num, sizeof(struct scatterlist),
                                rctx->gfp);
        if (!mssg->spu.dst)
                return -ENOMEM;

        sg = mssg->spu.dst;
        sg_init_table(sg, rx_frag_num);
        /* Space for SPU message header */
        sg_set_buf(sg++, rctx->msg_buf.spu_resp_hdr, ctx->spu_resp_hdr_len);

        /* Space for digest */
        sg_set_buf(sg++, rctx->msg_buf.digest, digestsize);

        if (stat_pad_len)
                sg_set_buf(sg++, rctx->msg_buf.rx_stat_pad, stat_pad_len);

        memset(rctx->msg_buf.rx_stat, 0, SPU_RX_STATUS_LEN);
        sg_set_buf(sg, rctx->msg_buf.rx_stat, spu->spu_rx_status_len());
        return 0;
}

/**
 * spu_ahash_tx_sg_create() -  Build up the scatterlist of buffers used to send
 * a SPU request message for an ahash request. Includes SPU message headers and
 * the request data.
 * @mssg:       mailbox message containing the transmit sg
 * @rctx:       crypto request context
 * @tx_frag_num: number of scatterlist elements required to construct the
 *              SPU request message
 * @spu_hdr_len: length in bytes of SPU message header
 * @hash_carry_len: Number of bytes of data carried over from previous req
 * @new_data_len: Number of bytes of new request data
 * @pad_len:    Number of pad bytes
 *
 * The scatterlist that gets allocated here is freed in spu_chunk_cleanup()
 * when the request completes, whether the request is handled successfully or
 * there is an error.
 *
 * Return:
 *   0 if successful
 *   < 0 if an error
 */
static int
spu_ahash_tx_sg_create(struct brcm_message *mssg,
                       struct iproc_reqctx_s *rctx,
                       u8 tx_frag_num,
                       u32 spu_hdr_len,
                       unsigned int hash_carry_len,
                       unsigned int new_data_len, u32 pad_len)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct scatterlist *sg; /* used to build sgs in mbox message */
        u32 datalen;            /* Number of bytes of response data expected */
        u32 stat_len;

        mssg->spu.src = kcalloc(tx_frag_num, sizeof(struct scatterlist),
                                rctx->gfp);
        if (!mssg->spu.src)
                return -ENOMEM;

        sg = mssg->spu.src;
        sg_init_table(sg, tx_frag_num);

        sg_set_buf(sg++, rctx->msg_buf.bcm_spu_req_hdr,
                   BCM_HDR_LEN + spu_hdr_len);

        if (hash_carry_len)
                sg_set_buf(sg++, rctx->hash_carry, hash_carry_len);

        if (new_data_len) {
                /* Copy in each src sg entry from request, up to chunksize */
                datalen = spu_msg_sg_add(&sg, &rctx->src_sg, &rctx->src_skip,
                                         rctx->src_nents, new_data_len);
                if (datalen < new_data_len) {
                        pr_err("%s(): failed to copy src sg to mbox msg",
                               __func__);
                        return -EFAULT;
                }
        }

        if (pad_len)
                sg_set_buf(sg++, rctx->msg_buf.spu_req_pad, pad_len);

        stat_len = spu->spu_tx_status_len();
        if (stat_len) {
                memset(rctx->msg_buf.tx_stat, 0, stat_len);
                sg_set_buf(sg, rctx->msg_buf.tx_stat, stat_len);
        }

        return 0;
}

/**
 * handle_ahash_req() - Process an asynchronous hash request from the crypto
 * API.
 * @rctx:  Crypto request context
 *
 * Builds a SPU request message embedded in a mailbox message and submits the
 * mailbox message on a selected mailbox channel. The SPU request message is
 * constructed as a scatterlist, including entries from the crypto API's
 * src scatterlist to avoid copying the data to be hashed. This function is
 * called either on the thread from the crypto API, or, in the case that the
 * crypto API request is too large to fit in a single SPU request message,
 * on the thread that invokes the receive callback with a response message.
 * Because some operations require the response from one chunk before the next
 * chunk can be submitted, we always wait for the response for the previous
 * chunk before submitting the next chunk. Because requests are submitted in
 * lock step like this, there is no need to synchronize access to request data
 * structures.
 *
 * Return:
 *   -EINPROGRESS: request has been submitted to SPU and response will be
 *                 returned asynchronously
 *   -EAGAIN:      non-final request included a small amount of data, which for
 *                 efficiency we did not submit to the SPU, but instead stored
 *                 to be submitted to the SPU with the next part of the request
 *   other:        an error code
 */
static int handle_ahash_req(struct iproc_reqctx_s *rctx)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct crypto_async_request *areq = rctx->parent;
        struct ahash_request *req = ahash_request_cast(areq);
        struct crypto_ahash *ahash = crypto_ahash_reqtfm(req);
        struct crypto_tfm *tfm = crypto_ahash_tfm(ahash);
        unsigned int blocksize = crypto_tfm_alg_blocksize(tfm);
        struct iproc_ctx_s *ctx = rctx->ctx;

        /* number of bytes still to be hashed in this req */
        unsigned int nbytes_to_hash = 0;
        int err = 0;
        unsigned int chunksize = 0;     /* length of hash carry + new data */
        /*
         * length of new data, not from hash carry, to be submitted in
         * this hw request
         */
        unsigned int new_data_len;

        unsigned int chunk_start = 0;
        u32 db_size;     /* Length of data field, incl gcm and hash padding */
        int pad_len = 0; /* total pad len, including gcm, hash, stat padding */
        u32 data_pad_len = 0;   /* length of GCM/CCM padding */
        u32 stat_pad_len = 0;   /* length of padding to align STATUS word */
        struct brcm_message *mssg;      /* mailbox message */
        struct spu_request_opts req_opts;
        struct spu_cipher_parms cipher_parms;
        struct spu_hash_parms hash_parms;
        struct spu_aead_parms aead_parms;
        unsigned int local_nbuf;
        u32 spu_hdr_len;
        unsigned int digestsize;
        u16 rem = 0;

        /*
         * number of entries in src and dst sg. Always includes SPU msg header.
         * rx always includes a buffer to catch digest and STATUS.
         */
        u8 rx_frag_num = 3;
        u8 tx_frag_num = 1;

        flow_log("total_todo %u, total_sent %u\n",
                 rctx->total_todo, rctx->total_sent);

        memset(&req_opts, 0, sizeof(req_opts));
        memset(&cipher_parms, 0, sizeof(cipher_parms));
        memset(&hash_parms, 0, sizeof(hash_parms));
        memset(&aead_parms, 0, sizeof(aead_parms));

        req_opts.bd_suppress = true;
        hash_parms.alg = ctx->auth.alg;
        hash_parms.mode = ctx->auth.mode;
        hash_parms.type = HASH_TYPE_NONE;
        hash_parms.key_buf = (u8 *)ctx->authkey;
        hash_parms.key_len = ctx->authkeylen;

        /*
         * For hash algorithms below assignment looks bit odd but
         * it's needed for AES-XCBC and AES-CMAC hash algorithms
         * to differentiate between 128, 192, 256 bit key values.
         * Based on the key values, hash algorithm is selected.
         * For example for 128 bit key, hash algorithm is AES-128.
         */
        cipher_parms.type = ctx->cipher_type;

        mssg = &rctx->mb_mssg;
        chunk_start = rctx->src_sent;

        /*
         * Compute the amount remaining to hash. This may include data
         * carried over from previous requests.
         */
        nbytes_to_hash = rctx->total_todo - rctx->total_sent;
        chunksize = nbytes_to_hash;
        if ((ctx->max_payload != SPU_MAX_PAYLOAD_INF) &&
            (chunksize > ctx->max_payload))
                chunksize = ctx->max_payload;

        /*
         * If this is not a final request and the request data is not a multiple
         * of a full block, then simply park the extra data and prefix it to the
         * data for the next request.
         */
        if (!rctx->is_final) {
                u8 *dest = rctx->hash_carry + rctx->hash_carry_len;
                u16 new_len;  /* len of data to add to hash carry */

                rem = chunksize % blocksize;   /* remainder */
                if (rem) {
                        /* chunksize not a multiple of blocksize */
                        chunksize -= rem;
                        if (chunksize == 0) {
                                /* Don't have a full block to submit to hw */
                                new_len = rem - rctx->hash_carry_len;
                                sg_copy_part_to_buf(req->src, dest, new_len,
                                                    rctx->src_sent);
                                rctx->hash_carry_len = rem;
                                flow_log("Exiting with hash carry len: %u\n",
                                         rctx->hash_carry_len);
                                packet_dump("  buf: ",
                                            rctx->hash_carry,
                                            rctx->hash_carry_len);
                                return -EAGAIN;
                        }
                }
        }

        /* if we have hash carry, then prefix it to the data in this request */
        local_nbuf = rctx->hash_carry_len;
        rctx->hash_carry_len = 0;
        if (local_nbuf)
                tx_frag_num++;
        new_data_len = chunksize - local_nbuf;

        /* Count number of sg entries to be used in this request */
        rctx->src_nents = spu_sg_count(rctx->src_sg, rctx->src_skip,
                                       new_data_len);

        /* AES hashing keeps key size in type field, so need to copy it here */
        if (hash_parms.alg == HASH_ALG_AES)
                hash_parms.type = (enum hash_type)cipher_parms.type;
        else
                hash_parms.type = spu->spu_hash_type(rctx->total_sent);

        digestsize = spu->spu_digest_size(ctx->digestsize, ctx->auth.alg,
                                          hash_parms.type);
        hash_parms.digestsize = digestsize;

        /* update the indexes */
        rctx->total_sent += chunksize;
        /* if you sent a prebuf then that wasn't from this req->src */
        rctx->src_sent += new_data_len;

        if ((rctx->total_sent == rctx->total_todo) && rctx->is_final)
                hash_parms.pad_len = spu->spu_hash_pad_len(hash_parms.alg,
                                                           hash_parms.mode,
                                                           chunksize,
                                                           blocksize);

        /*
         * If a non-first chunk, then include the digest returned from the
         * previous chunk so that hw can add to it (except for AES types).
         */
        if ((hash_parms.type == HASH_TYPE_UPDT) &&
            (hash_parms.alg != HASH_ALG_AES)) {
                hash_parms.key_buf = rctx->incr_hash;
                hash_parms.key_len = digestsize;
        }

        atomic64_add(chunksize, &iproc_priv.bytes_out);

        flow_log("%s() final: %u nbuf: %u ",
                 __func__, rctx->is_final, local_nbuf);

        if (ctx->max_payload == SPU_MAX_PAYLOAD_INF)
                flow_log("max_payload infinite\n");
        else
                flow_log("max_payload %u\n", ctx->max_payload);

        flow_log("chunk_start: %u chunk_size: %u\n", chunk_start, chunksize);

        /* Prepend SPU header with type 3 BCM header */
        memcpy(rctx->msg_buf.bcm_spu_req_hdr, BCMHEADER, BCM_HDR_LEN);

        hash_parms.prebuf_len = local_nbuf;
        spu_hdr_len = spu->spu_create_request(rctx->msg_buf.bcm_spu_req_hdr +
                                              BCM_HDR_LEN,
                                              &req_opts, &cipher_parms,
                                              &hash_parms, &aead_parms,
                                              new_data_len);

        if (spu_hdr_len == 0) {
                pr_err("Failed to create SPU request header\n");
                return -EFAULT;
        }

        /*
         * Determine total length of padding required. Put all padding in one
         * buffer.
         */
        data_pad_len = spu->spu_gcm_ccm_pad_len(ctx->cipher.mode, chunksize);
        db_size = spu_real_db_size(0, 0, local_nbuf, new_data_len,
                                   0, 0, hash_parms.pad_len);
        if (spu->spu_tx_status_len())
                stat_pad_len = spu->spu_wordalign_padlen(db_size);
        if (stat_pad_len)
                rx_frag_num++;
        pad_len = hash_parms.pad_len + data_pad_len + stat_pad_len;
        if (pad_len) {
                tx_frag_num++;
                spu->spu_request_pad(rctx->msg_buf.spu_req_pad, data_pad_len,
                                     hash_parms.pad_len, ctx->auth.alg,
                                     ctx->auth.mode, rctx->total_sent,
                                     stat_pad_len);
        }

        spu->spu_dump_msg_hdr(rctx->msg_buf.bcm_spu_req_hdr + BCM_HDR_LEN,
                              spu_hdr_len);
        packet_dump("    prebuf: ", rctx->hash_carry, local_nbuf);
        flow_log("Data:\n");
        dump_sg(rctx->src_sg, rctx->src_skip, new_data_len);
        packet_dump("   pad: ", rctx->msg_buf.spu_req_pad, pad_len);

        /*
         * Build mailbox message containing SPU request msg and rx buffers
         * to catch response message
         */
        memset(mssg, 0, sizeof(*mssg));
        mssg->type = BRCM_MESSAGE_SPU;
        mssg->ctx = rctx;       /* Will be returned in response */

        /* Create rx scatterlist to catch result */
        err = spu_ahash_rx_sg_create(mssg, rctx, rx_frag_num, digestsize,
                                     stat_pad_len);
        if (err)
                return err;

        /* Create tx scatterlist containing SPU request message */
        tx_frag_num += rctx->src_nents;
        if (spu->spu_tx_status_len())
                tx_frag_num++;
        err = spu_ahash_tx_sg_create(mssg, rctx, tx_frag_num, spu_hdr_len,
                                     local_nbuf, new_data_len, pad_len);
        if (err)
                return err;

        err = mailbox_send_message(mssg, req->base.flags, rctx->chan_idx);
        if (unlikely(err < 0))
                return err;

        return -EINPROGRESS;
}

/**
 * spu_hmac_outer_hash() - Request synchonous software compute of the outer hash
 * for an HMAC request.
 * @req:  The HMAC request from the crypto API
 * @ctx:  The session context
 *
 * Return: 0 if synchronous hash operation successful
 *         -EINVAL if the hash algo is unrecognized
 *         any other value indicates an error
 */
static int spu_hmac_outer_hash(struct ahash_request *req,
                               struct iproc_ctx_s *ctx)
{
        struct crypto_ahash *ahash = crypto_ahash_reqtfm(req);
        unsigned int blocksize =
                crypto_tfm_alg_blocksize(crypto_ahash_tfm(ahash));
        int rc;

        switch (ctx->auth.alg) {
        case HASH_ALG_MD5:
                rc = do_shash("md5", req->result, ctx->opad, blocksize,
                              req->result, ctx->digestsize, NULL, 0);
                break;
        case HASH_ALG_SHA1:
                rc = do_shash("sha1", req->result, ctx->opad, blocksize,
                              req->result, ctx->digestsize, NULL, 0);
                break;
        case HASH_ALG_SHA224:
                rc = do_shash("sha224", req->result, ctx->opad, blocksize,
                              req->result, ctx->digestsize, NULL, 0);
                break;
        case HASH_ALG_SHA256:
                rc = do_shash("sha256", req->result, ctx->opad, blocksize,
                              req->result, ctx->digestsize, NULL, 0);
                break;
        case HASH_ALG_SHA384:
                rc = do_shash("sha384", req->result, ctx->opad, blocksize,
                              req->result, ctx->digestsize, NULL, 0);
                break;
        case HASH_ALG_SHA512:
                rc = do_shash("sha512", req->result, ctx->opad, blocksize,
                              req->result, ctx->digestsize, NULL, 0);
                break;
        default:
                pr_err("%s() Error : unknown hmac type\n", __func__);
                rc = -EINVAL;
        }
        return rc;
}

/**
 * ahash_req_done() - Process a hash result from the SPU hardware.
 * @rctx: Crypto request context
 *
 * Return: 0 if successful
 *         < 0 if an error
 */
static int ahash_req_done(struct iproc_reqctx_s *rctx)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct crypto_async_request *areq = rctx->parent;
        struct ahash_request *req = ahash_request_cast(areq);
        struct iproc_ctx_s *ctx = rctx->ctx;
        int err;

        memcpy(req->result, rctx->msg_buf.digest, ctx->digestsize);

        if (spu->spu_type == SPU_TYPE_SPUM) {
                /* byte swap the output from the UPDT function to network byte
                 * order
                 */
                if (ctx->auth.alg == HASH_ALG_MD5) {
                        __swab32s((u32 *)req->result);
                        __swab32s(((u32 *)req->result) + 1);
                        __swab32s(((u32 *)req->result) + 2);
                        __swab32s(((u32 *)req->result) + 3);
                        __swab32s(((u32 *)req->result) + 4);
                }
        }

        flow_dump("  digest ", req->result, ctx->digestsize);

        /* if this an HMAC then do the outer hash */
        if (rctx->is_sw_hmac) {
                err = spu_hmac_outer_hash(req, ctx);
                if (err < 0)
                        return err;
                flow_dump("  hmac: ", req->result, ctx->digestsize);
        }

        if (rctx->is_sw_hmac || ctx->auth.mode == HASH_MODE_HMAC) {
                atomic_inc(&iproc_priv.op_counts[SPU_OP_HMAC]);
                atomic_inc(&iproc_priv.hmac_cnt[ctx->auth.alg]);
        } else {
                atomic_inc(&iproc_priv.op_counts[SPU_OP_HASH]);
                atomic_inc(&iproc_priv.hash_cnt[ctx->auth.alg]);
        }

        return 0;
}

/**
 * handle_ahash_resp() - Process a SPU response message for a hash request.
 * Checks if the entire crypto API request has been processed, and if so,
 * invokes post processing on the result.
 * @rctx: Crypto request context
 */
static void handle_ahash_resp(struct iproc_reqctx_s *rctx)
{
        struct iproc_ctx_s *ctx = rctx->ctx;
#ifdef DEBUG
        struct crypto_async_request *areq = rctx->parent;
        struct ahash_request *req = ahash_request_cast(areq);
        struct crypto_ahash *ahash = crypto_ahash_reqtfm(req);
        unsigned int blocksize =
                crypto_tfm_alg_blocksize(crypto_ahash_tfm(ahash));
#endif
        /*
         * Save hash to use as input to next op if incremental. Might be copying
         * too much, but that's easier than figuring out actual digest size here
         */
        memcpy(rctx->incr_hash, rctx->msg_buf.digest, MAX_DIGEST_SIZE);

        flow_log("%s() blocksize:%u digestsize:%u\n",
                 __func__, blocksize, ctx->digestsize);

        atomic64_add(ctx->digestsize, &iproc_priv.bytes_in);

        if (rctx->is_final && (rctx->total_sent == rctx->total_todo))
                ahash_req_done(rctx);
}

/**
 * spu_aead_rx_sg_create() - Build up the scatterlist of buffers used to receive
 * a SPU response message for an AEAD request. Includes buffers to catch SPU
 * message headers and the response data.
 * @mssg:       mailbox message containing the receive sg
 * @rctx:       crypto request context
 * @rx_frag_num: number of scatterlist elements required to hold the
 *              SPU response message
 * @assoc_len:  Length of associated data included in the crypto request
 * @ret_iv_len: Length of IV returned in response
 * @resp_len:   Number of bytes of response data expected to be written to
 *              dst buffer from crypto API
 * @digestsize: Length of hash digest, in bytes
 * @stat_pad_len: Number of bytes required to pad the STAT field to
 *              a 4-byte boundary
 *
 * The scatterlist that gets allocated here is freed in spu_chunk_cleanup()
 * when the request completes, whether the request is handled successfully or
 * there is an error.
 *
 * Returns:
 *   0 if successful
 *   < 0 if an error
 */
static int spu_aead_rx_sg_create(struct brcm_message *mssg,
                                 struct aead_request *req,
                                 struct iproc_reqctx_s *rctx,
                                 u8 rx_frag_num,
                                 unsigned int assoc_len,
                                 u32 ret_iv_len, unsigned int resp_len,
                                 unsigned int digestsize, u32 stat_pad_len)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct scatterlist *sg; /* used to build sgs in mbox message */
        struct iproc_ctx_s *ctx = rctx->ctx;
        u32 datalen;            /* Number of bytes of response data expected */
        u32 assoc_buf_len;
        u8 data_padlen = 0;

        if (ctx->is_rfc4543) {
                /* RFC4543: only pad after data, not after AAD */
                data_padlen = spu->spu_gcm_ccm_pad_len(ctx->cipher.mode,
                                                          assoc_len + resp_len);
                assoc_buf_len = assoc_len;
        } else {
                data_padlen = spu->spu_gcm_ccm_pad_len(ctx->cipher.mode,
                                                          resp_len);
                assoc_buf_len = spu->spu_assoc_resp_len(ctx->cipher.mode,
                                                assoc_len, ret_iv_len,
                                                rctx->is_encrypt);
        }

        if (ctx->cipher.mode == CIPHER_MODE_CCM)
                /* ICV (after data) must be in the next 32-bit word for CCM */
                data_padlen += spu->spu_wordalign_padlen(assoc_buf_len +
                                                         resp_len +
                                                         data_padlen);

        if (data_padlen)
                /* have to catch gcm pad in separate buffer */
                rx_frag_num++;

        mssg->spu.dst = kcalloc(rx_frag_num, sizeof(struct scatterlist),
                                rctx->gfp);
        if (!mssg->spu.dst)
                return -ENOMEM;

        sg = mssg->spu.dst;
        sg_init_table(sg, rx_frag_num);

        /* Space for SPU message header */
        sg_set_buf(sg++, rctx->msg_buf.spu_resp_hdr, ctx->spu_resp_hdr_len);

        if (assoc_buf_len) {
                /*
                 * Don't write directly to req->dst, because SPU may pad the
                 * assoc data in the response
                 */
                memset(rctx->msg_buf.a.resp_aad, 0, assoc_buf_len);
                sg_set_buf(sg++, rctx->msg_buf.a.resp_aad, assoc_buf_len);
        }

        if (resp_len) {
                /*
                 * Copy in each dst sg entry from request, up to chunksize.
                 * dst sg catches just the data. digest caught in separate buf.
                 */
                datalen = spu_msg_sg_add(&sg, &rctx->dst_sg, &rctx->dst_skip,
                                         rctx->dst_nents, resp_len);
                if (datalen < (resp_len)) {
                        pr_err("%s(): failed to copy dst sg to mbox msg. expected len %u, datalen %u",
                               __func__, resp_len, datalen);
                        return -EFAULT;
                }
        }

        /* If GCM/CCM data is padded, catch padding in separate buffer */
        if (data_padlen) {
                memset(rctx->msg_buf.a.gcmpad, 0, data_padlen);
                sg_set_buf(sg++, rctx->msg_buf.a.gcmpad, data_padlen);
        }

        /* Always catch ICV in separate buffer */
        sg_set_buf(sg++, rctx->msg_buf.digest, digestsize);

        flow_log("stat_pad_len %u\n", stat_pad_len);
        if (stat_pad_len) {
                memset(rctx->msg_buf.rx_stat_pad, 0, stat_pad_len);
                sg_set_buf(sg++, rctx->msg_buf.rx_stat_pad, stat_pad_len);
        }

        memset(rctx->msg_buf.rx_stat, 0, SPU_RX_STATUS_LEN);
        sg_set_buf(sg, rctx->msg_buf.rx_stat, spu->spu_rx_status_len());

        return 0;
}

/**
 * spu_aead_tx_sg_create() - Build up the scatterlist of buffers used to send a
 * SPU request message for an AEAD request. Includes SPU message headers and the
 * request data.
 * @mssg:       mailbox message containing the transmit sg
 * @rctx:       crypto request context
 * @tx_frag_num: number of scatterlist elements required to construct the
 *              SPU request message
 * @spu_hdr_len: length of SPU message header in bytes
 * @assoc:      crypto API associated data scatterlist
 * @assoc_len:  length of associated data
 * @assoc_nents: number of scatterlist entries containing assoc data
 * @aead_iv_len: length of AEAD IV, if included
 * @chunksize:  Number of bytes of request data
 * @aad_pad_len: Number of bytes of padding at end of AAD. For GCM/CCM.
 * @pad_len:    Number of pad bytes
 * @incl_icv:   If true, write separate ICV buffer after data and
 *              any padding
 *
 * The scatterlist that gets allocated here is freed in spu_chunk_cleanup()
 * when the request completes, whether the request is handled successfully or
 * there is an error.
 *
 * Return:
 *   0 if successful
 *   < 0 if an error
 */
static int spu_aead_tx_sg_create(struct brcm_message *mssg,
                                 struct iproc_reqctx_s *rctx,
                                 u8 tx_frag_num,
                                 u32 spu_hdr_len,
                                 struct scatterlist *assoc,
                                 unsigned int assoc_len,
                                 int assoc_nents,
                                 unsigned int aead_iv_len,
                                 unsigned int chunksize,
                                 u32 aad_pad_len, u32 pad_len, bool incl_icv)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct scatterlist *sg; /* used to build sgs in mbox message */
        struct scatterlist *assoc_sg = assoc;
        struct iproc_ctx_s *ctx = rctx->ctx;
        u32 datalen;            /* Number of bytes of data to write */
        u32 written;            /* Number of bytes of data written */
        u32 assoc_offset = 0;
        u32 stat_len;

        mssg->spu.src = kcalloc(tx_frag_num, sizeof(struct scatterlist),
                                rctx->gfp);
        if (!mssg->spu.src)
                return -ENOMEM;

        sg = mssg->spu.src;
        sg_init_table(sg, tx_frag_num);

        sg_set_buf(sg++, rctx->msg_buf.bcm_spu_req_hdr,
                   BCM_HDR_LEN + spu_hdr_len);

        if (assoc_len) {
                /* Copy in each associated data sg entry from request */
                written = spu_msg_sg_add(&sg, &assoc_sg, &assoc_offset,
                                         assoc_nents, assoc_len);
                if (written < assoc_len) {
                        pr_err("%s(): failed to copy assoc sg to mbox msg",
                               __func__);
                        return -EFAULT;
                }
        }

        if (aead_iv_len)
                sg_set_buf(sg++, rctx->msg_buf.iv_ctr, aead_iv_len);

        if (aad_pad_len) {
                memset(rctx->msg_buf.a.req_aad_pad, 0, aad_pad_len);
                sg_set_buf(sg++, rctx->msg_buf.a.req_aad_pad, aad_pad_len);
        }

        datalen = chunksize;
        if ((chunksize > ctx->digestsize) && incl_icv)
                datalen -= ctx->digestsize;
        if (datalen) {
                /* For aead, a single msg should consume the entire src sg */
                written = spu_msg_sg_add(&sg, &rctx->src_sg, &rctx->src_skip,
                                         rctx->src_nents, datalen);
                if (written < datalen) {
                        pr_err("%s(): failed to copy src sg to mbox msg",
                               __func__);
                        return -EFAULT;
                }
        }

        if (pad_len) {
                memset(rctx->msg_buf.spu_req_pad, 0, pad_len);
                sg_set_buf(sg++, rctx->msg_buf.spu_req_pad, pad_len);
        }

        if (incl_icv)
                sg_set_buf(sg++, rctx->msg_buf.digest, ctx->digestsize);

        stat_len = spu->spu_tx_status_len();
        if (stat_len) {
                memset(rctx->msg_buf.tx_stat, 0, stat_len);
                sg_set_buf(sg, rctx->msg_buf.tx_stat, stat_len);
        }
        return 0;
}

/**
 * handle_aead_req() - Submit a SPU request message for the next chunk of the
 * current AEAD request.
 * @rctx:  Crypto request context
 *
 * Unlike other operation types, we assume the length of the request fits in
 * a single SPU request message. aead_enqueue() makes sure this is true.
 * Comments for other op types regarding threads applies here as well.
 *
 * Unlike incremental hash ops, where the spu returns the entire hash for
 * truncated algs like sha-224, the SPU returns just the truncated hash in
 * response to aead requests. So digestsize is always ctx->digestsize here.
 *
 * Return: -EINPROGRESS: crypto request has been accepted and result will be
 *                       returned asynchronously
 *         Any other value indicates an error
 */
static int handle_aead_req(struct iproc_reqctx_s *rctx)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct crypto_async_request *areq = rctx->parent;
        struct aead_request *req = container_of(areq,
                                                struct aead_request, base);
        struct iproc_ctx_s *ctx = rctx->ctx;
        int err;
        unsigned int chunksize;
        unsigned int resp_len;
        u32 spu_hdr_len;
        u32 db_size;
        u32 stat_pad_len;
        u32 pad_len;
        struct brcm_message *mssg;      /* mailbox message */
        struct spu_request_opts req_opts;
        struct spu_cipher_parms cipher_parms;
        struct spu_hash_parms hash_parms;
        struct spu_aead_parms aead_parms;
        int assoc_nents = 0;
        bool incl_icv = false;
        unsigned int digestsize = ctx->digestsize;

        /* number of entries in src and dst sg. Always includes SPU msg header.
         */
        u8 rx_frag_num = 2;     /* and STATUS */
        u8 tx_frag_num = 1;

        /* doing the whole thing at once */
        chunksize = rctx->total_todo;

        flow_log("%s: chunksize %u\n", __func__, chunksize);

        memset(&req_opts, 0, sizeof(req_opts));
        memset(&hash_parms, 0, sizeof(hash_parms));
        memset(&aead_parms, 0, sizeof(aead_parms));

        req_opts.is_inbound = !(rctx->is_encrypt);
        req_opts.auth_first = ctx->auth_first;
        req_opts.is_aead = true;
        req_opts.is_esp = ctx->is_esp;

        cipher_parms.alg = ctx->cipher.alg;
        cipher_parms.mode = ctx->cipher.mode;
        cipher_parms.type = ctx->cipher_type;
        cipher_parms.key_buf = ctx->enckey;
        cipher_parms.key_len = ctx->enckeylen;
        cipher_parms.iv_buf = rctx->msg_buf.iv_ctr;
        cipher_parms.iv_len = rctx->iv_ctr_len;

        hash_parms.alg = ctx->auth.alg;
        hash_parms.mode = ctx->auth.mode;
        hash_parms.type = HASH_TYPE_NONE;
        hash_parms.key_buf = (u8 *)ctx->authkey;
        hash_parms.key_len = ctx->authkeylen;
        hash_parms.digestsize = digestsize;

        if ((ctx->auth.alg == HASH_ALG_SHA224) &&
            (ctx->authkeylen < SHA224_DIGEST_SIZE))
                hash_parms.key_len = SHA224_DIGEST_SIZE;

        aead_parms.assoc_size = req->assoclen;
        if (ctx->is_esp && !ctx->is_rfc4543) {
                /*
                 * 8-byte IV is included assoc data in request. SPU2
                 * expects AAD to include just SPI and seqno. So
                 * subtract off the IV len.
                 */
                aead_parms.assoc_size -= GCM_RFC4106_IV_SIZE;

                if (rctx->is_encrypt) {
                        aead_parms.return_iv = true;
                        aead_parms.ret_iv_len = GCM_RFC4106_IV_SIZE;
                        aead_parms.ret_iv_off = GCM_ESP_SALT_SIZE;
                }
        } else {
                aead_parms.ret_iv_len = 0;
        }

        /*
         * Count number of sg entries from the crypto API request that are to
         * be included in this mailbox message. For dst sg, don't count space
         * for digest. Digest gets caught in a separate buffer and copied back
         * to dst sg when processing response.
         */
        rctx->src_nents = spu_sg_count(rctx->src_sg, rctx->src_skip, chunksize);
        rctx->dst_nents = spu_sg_count(rctx->dst_sg, rctx->dst_skip, chunksize);
        if (aead_parms.assoc_size)
                assoc_nents = spu_sg_count(rctx->assoc, 0,
                                           aead_parms.assoc_size);

        mssg = &rctx->mb_mssg;

        rctx->total_sent = chunksize;
        rctx->src_sent = chunksize;
        if (spu->spu_assoc_resp_len(ctx->cipher.mode,
                                    aead_parms.assoc_size,
                                    aead_parms.ret_iv_len,
                                    rctx->is_encrypt))
                rx_frag_num++;

        aead_parms.iv_len = spu->spu_aead_ivlen(ctx->cipher.mode,
                                                rctx->iv_ctr_len);

        if (ctx->auth.alg == HASH_ALG_AES)
                hash_parms.type = (enum hash_type)ctx->cipher_type;

        /* General case AAD padding (CCM and RFC4543 special cases below) */
        aead_parms.aad_pad_len = spu->spu_gcm_ccm_pad_len(ctx->cipher.mode,
                                                 aead_parms.assoc_size);

        /* General case data padding (CCM decrypt special case below) */
        aead_parms.data_pad_len = spu->spu_gcm_ccm_pad_len(ctx->cipher.mode,
                                                           chunksize);

        if (ctx->cipher.mode == CIPHER_MODE_CCM) {
                /*
                 * for CCM, AAD len + 2 (rather than AAD len) needs to be
                 * 128-bit aligned
                 */
                aead_parms.aad_pad_len = spu->spu_gcm_ccm_pad_len(
                                         ctx->cipher.mode,
                                         aead_parms.assoc_size + 2);

                /*
                 * And when decrypting CCM, need to pad without including
                 * size of ICV which is tacked on to end of chunk
                 */
                if (!rctx->is_encrypt)
                        aead_parms.data_pad_len =
                                spu->spu_gcm_ccm_pad_len(ctx->cipher.mode,
                                                        chunksize - digestsize);

                /* CCM also requires software to rewrite portions of IV: */
                spu->spu_ccm_update_iv(digestsize, &cipher_parms, req->assoclen,
                                       chunksize, rctx->is_encrypt,
                                       ctx->is_esp);
        }

        if (ctx->is_rfc4543) {
                /*
                 * RFC4543: data is included in AAD, so don't pad after AAD
                 * and pad data based on both AAD + data size
                 */
                aead_parms.aad_pad_len = 0;
                if (!rctx->is_encrypt)
                        aead_parms.data_pad_len = spu->spu_gcm_ccm_pad_len(
                                        ctx->cipher.mode,
                                        aead_parms.assoc_size + chunksize -
                                        digestsize);
                else
                        aead_parms.data_pad_len = spu->spu_gcm_ccm_pad_len(
                                        ctx->cipher.mode,
                                        aead_parms.assoc_size + chunksize);

                req_opts.is_rfc4543 = true;
        }

        if (spu_req_incl_icv(ctx->cipher.mode, rctx->is_encrypt)) {
                incl_icv = true;
                tx_frag_num++;
                /* Copy ICV from end of src scatterlist to digest buf */
                sg_copy_part_to_buf(req->src, rctx->msg_buf.digest, digestsize,
                                    req->assoclen + rctx->total_sent -
                                    digestsize);
        }

        atomic64_add(chunksize, &iproc_priv.bytes_out);

        flow_log("%s()-sent chunksize:%u\n", __func__, chunksize);

        /* Prepend SPU header with type 3 BCM header */
        memcpy(rctx->msg_buf.bcm_spu_req_hdr, BCMHEADER, BCM_HDR_LEN);

        spu_hdr_len = spu->spu_create_request(rctx->msg_buf.bcm_spu_req_hdr +
                                              BCM_HDR_LEN, &req_opts,
                                              &cipher_parms, &hash_parms,
                                              &aead_parms, chunksize);

        /* Determine total length of padding. Put all padding in one buffer. */
        db_size = spu_real_db_size(aead_parms.assoc_size, aead_parms.iv_len, 0,
                                   chunksize, aead_parms.aad_pad_len,
                                   aead_parms.data_pad_len, 0);

        stat_pad_len = spu->spu_wordalign_padlen(db_size);

        if (stat_pad_len)
                rx_frag_num++;
        pad_len = aead_parms.data_pad_len + stat_pad_len;
        if (pad_len) {
                tx_frag_num++;
                spu->spu_request_pad(rctx->msg_buf.spu_req_pad,
                                     aead_parms.data_pad_len, 0,
                                     ctx->auth.alg, ctx->auth.mode,
                                     rctx->total_sent, stat_pad_len);
        }

        spu->spu_dump_msg_hdr(rctx->msg_buf.bcm_spu_req_hdr + BCM_HDR_LEN,
                              spu_hdr_len);
        dump_sg(rctx->assoc, 0, aead_parms.assoc_size);
        packet_dump("    aead iv: ", rctx->msg_buf.iv_ctr, aead_parms.iv_len);
        packet_log("BD:\n");
        dump_sg(rctx->src_sg, rctx->src_skip, chunksize);
        packet_dump("   pad: ", rctx->msg_buf.spu_req_pad, pad_len);

        /*
         * Build mailbox message containing SPU request msg and rx buffers
         * to catch response message
         */
        memset(mssg, 0, sizeof(*mssg));
        mssg->type = BRCM_MESSAGE_SPU;
        mssg->ctx = rctx;       /* Will be returned in response */

        /* Create rx scatterlist to catch result */
        rx_frag_num += rctx->dst_nents;
        resp_len = chunksize;

        /*
         * Always catch ICV in separate buffer. Have to for GCM/CCM because of
         * padding. Have to for SHA-224 and other truncated SHAs because SPU
         * sends entire digest back.
         */
        rx_frag_num++;

        if (((ctx->cipher.mode == CIPHER_MODE_GCM) ||
             (ctx->cipher.mode == CIPHER_MODE_CCM)) && !rctx->is_encrypt) {
                /*
                 * Input is ciphertxt plus ICV, but ICV not incl
                 * in output.
                 */
                resp_len -= ctx->digestsize;
                if (resp_len == 0)
                        /* no rx frags to catch output data */
                        rx_frag_num -= rctx->dst_nents;
        }

        err = spu_aead_rx_sg_create(mssg, req, rctx, rx_frag_num,
                                    aead_parms.assoc_size,
                                    aead_parms.ret_iv_len, resp_len, digestsize,
                                    stat_pad_len);
        if (err)
                return err;

        /* Create tx scatterlist containing SPU request message */
        tx_frag_num += rctx->src_nents;
        tx_frag_num += assoc_nents;
        if (aead_parms.aad_pad_len)
                tx_frag_num++;
        if (aead_parms.iv_len)
                tx_frag_num++;
        if (spu->spu_tx_status_len())
                tx_frag_num++;
        err = spu_aead_tx_sg_create(mssg, rctx, tx_frag_num, spu_hdr_len,
                                    rctx->assoc, aead_parms.assoc_size,
                                    assoc_nents, aead_parms.iv_len, chunksize,
                                    aead_parms.aad_pad_len, pad_len, incl_icv);
        if (err)
                return err;

        err = mailbox_send_message(mssg, req->base.flags, rctx->chan_idx);
        if (unlikely(err < 0))
                return err;

        return -EINPROGRESS;
}

/**
 * handle_aead_resp() - Process a SPU response message for an AEAD request.
 * @rctx:  Crypto request context
 */
static void handle_aead_resp(struct iproc_reqctx_s *rctx)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct crypto_async_request *areq = rctx->parent;
        struct aead_request *req = container_of(areq,
                                                struct aead_request, base);
        struct iproc_ctx_s *ctx = rctx->ctx;
        u32 payload_len;
        unsigned int icv_offset;
        u32 result_len;

        /* See how much data was returned */
        payload_len = spu->spu_payload_length(rctx->msg_buf.spu_resp_hdr);
        flow_log("payload_len %u\n", payload_len);

        /* only count payload */
        atomic64_add(payload_len, &iproc_priv.bytes_in);

        if (req->assoclen)
                packet_dump("  assoc_data ", rctx->msg_buf.a.resp_aad,
                            req->assoclen);

        /*
         * Copy the ICV back to the destination
         * buffer. In decrypt case, SPU gives us back the digest, but crypto
         * API doesn't expect ICV in dst buffer.
         */
        result_len = req->cryptlen;
        if (rctx->is_encrypt) {
                icv_offset = req->assoclen + rctx->total_sent;
                packet_dump("  ICV: ", rctx->msg_buf.digest, ctx->digestsize);
                flow_log("copying ICV to dst sg at offset %u\n", icv_offset);
                sg_copy_part_from_buf(req->dst, rctx->msg_buf.digest,
                                      ctx->digestsize, icv_offset);
                result_len += ctx->digestsize;
        }

        packet_log("response data:  ");
        dump_sg(req->dst, req->assoclen, result_len);

        atomic_inc(&iproc_priv.op_counts[SPU_OP_AEAD]);
        if (ctx->cipher.alg == CIPHER_ALG_AES) {
                if (ctx->cipher.mode == CIPHER_MODE_CCM)
                        atomic_inc(&iproc_priv.aead_cnt[AES_CCM]);
                else if (ctx->cipher.mode == CIPHER_MODE_GCM)
                        atomic_inc(&iproc_priv.aead_cnt[AES_GCM]);
                else
                        atomic_inc(&iproc_priv.aead_cnt[AUTHENC]);
        } else {
                atomic_inc(&iproc_priv.aead_cnt[AUTHENC]);
        }
}

/**
 * spu_chunk_cleanup() - Do cleanup after processing one chunk of a request
 * @rctx:  request context
 *
 * Mailbox scatterlists are allocated for each chunk. So free them after
 * processing each chunk.
 */
static void spu_chunk_cleanup(struct iproc_reqctx_s *rctx)
{
        /* mailbox message used to tx request */
        struct brcm_message *mssg = &rctx->mb_mssg;

        kfree(mssg->spu.src);
        kfree(mssg->spu.dst);
        memset(mssg, 0, sizeof(struct brcm_message));
}

/**
 * finish_req() - Used to invoke the complete callback from the requester when
 * a request has been handled asynchronously.
 * @rctx:  Request context
 * @err:   Indicates whether the request was successful or not
 *
 * Ensures that cleanup has been done for request
 */
static void finish_req(struct iproc_reqctx_s *rctx, int err)
{
        struct crypto_async_request *areq = rctx->parent;

        flow_log("%s() err:%d\n\n", __func__, err);

        /* No harm done if already called */
        spu_chunk_cleanup(rctx);

        if (areq)
                areq->complete(areq, err);
}

/**
 * spu_rx_callback() - Callback from mailbox framework with a SPU response.
 * @cl:         mailbox client structure for SPU driver
 * @msg:        mailbox message containing SPU response
 */
static void spu_rx_callback(struct mbox_client *cl, void *msg)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct brcm_message *mssg = msg;
        struct iproc_reqctx_s *rctx;
        struct iproc_ctx_s *ctx;
        struct crypto_async_request *areq;
        int err = 0;

        rctx = mssg->ctx;
        if (unlikely(!rctx)) {
                /* This is fatal */
                pr_err("%s(): no request context", __func__);
                err = -EFAULT;
                goto cb_finish;
        }
        areq = rctx->parent;
        ctx = rctx->ctx;

        /* process the SPU status */
        err = spu->spu_status_process(rctx->msg_buf.rx_stat);
        if (err != 0) {
                if (err == SPU_INVALID_ICV)
                        atomic_inc(&iproc_priv.bad_icv);
                err = -EBADMSG;
                goto cb_finish;
        }

        /* Process the SPU response message */
        switch (rctx->ctx->alg->type) {
        case CRYPTO_ALG_TYPE_ABLKCIPHER:
                handle_ablkcipher_resp(rctx);
                break;
        case CRYPTO_ALG_TYPE_AHASH:
                handle_ahash_resp(rctx);
                break;
        case CRYPTO_ALG_TYPE_AEAD:
                handle_aead_resp(rctx);
                break;
        default:
                err = -EINVAL;
                goto cb_finish;
        }

        /*
         * If this response does not complete the request, then send the next
         * request chunk.
         */
        if (rctx->total_sent < rctx->total_todo) {
                /* Deallocate anything specific to previous chunk */
                spu_chunk_cleanup(rctx);

                switch (rctx->ctx->alg->type) {
                case CRYPTO_ALG_TYPE_ABLKCIPHER:
                        err = handle_ablkcipher_req(rctx);
                        break;
                case CRYPTO_ALG_TYPE_AHASH:
                        err = handle_ahash_req(rctx);
                        if (err == -EAGAIN)
                                /*
                                 * we saved data in hash carry, but tell crypto
                                 * API we successfully completed request.
                                 */
                                err = 0;
                        break;
                case CRYPTO_ALG_TYPE_AEAD:
                        err = handle_aead_req(rctx);
                        break;
                default:
                        err = -EINVAL;
                }

                if (err == -EINPROGRESS)
                        /* Successfully submitted request for next chunk */
                        return;
        }

cb_finish:
        finish_req(rctx, err);
}

/* ==================== Kernel Cryptographic API ==================== */

/**
 * ablkcipher_enqueue() - Handle ablkcipher encrypt or decrypt request.
 * @req:        Crypto API request
 * @encrypt:    true if encrypting; false if decrypting
 *
 * Return: -EINPROGRESS if request accepted and result will be returned
 *                      asynchronously
 *         < 0 if an error
 */
static int ablkcipher_enqueue(struct ablkcipher_request *req, bool encrypt)
{
        struct iproc_reqctx_s *rctx = ablkcipher_request_ctx(req);
        struct iproc_ctx_s *ctx =
            crypto_ablkcipher_ctx(crypto_ablkcipher_reqtfm(req));
        int err;

        flow_log("%s() enc:%u\n", __func__, encrypt);

        rctx->gfp = (req->base.flags & (CRYPTO_TFM_REQ_MAY_BACKLOG |
                       CRYPTO_TFM_REQ_MAY_SLEEP)) ? GFP_KERNEL : GFP_ATOMIC;
        rctx->parent = &req->base;
        rctx->is_encrypt = encrypt;
        rctx->bd_suppress = false;
        rctx->total_todo = req->nbytes;
        rctx->src_sent = 0;
        rctx->total_sent = 0;
        rctx->total_received = 0;
        rctx->ctx = ctx;

        /* Initialize current position in src and dst scatterlists */
        rctx->src_sg = req->src;
        rctx->src_nents = 0;
        rctx->src_skip = 0;
        rctx->dst_sg = req->dst;
        rctx->dst_nents = 0;
        rctx->dst_skip = 0;

        if (ctx->cipher.mode == CIPHER_MODE_CBC ||
            ctx->cipher.mode == CIPHER_MODE_CTR ||
            ctx->cipher.mode == CIPHER_MODE_OFB ||
            ctx->cipher.mode == CIPHER_MODE_XTS ||
            ctx->cipher.mode == CIPHER_MODE_GCM ||
            ctx->cipher.mode == CIPHER_MODE_CCM) {
                rctx->iv_ctr_len =
                    crypto_ablkcipher_ivsize(crypto_ablkcipher_reqtfm(req));
                memcpy(rctx->msg_buf.iv_ctr, req->info, rctx->iv_ctr_len);
        } else {
                rctx->iv_ctr_len = 0;
        }

        /* Choose a SPU to process this request */
        rctx->chan_idx = select_channel();
        err = handle_ablkcipher_req(rctx);
        if (err != -EINPROGRESS)
                /* synchronous result */
                spu_chunk_cleanup(rctx);

        return err;
}

static int des_setkey(struct crypto_ablkcipher *cipher, const u8 *key,
                      unsigned int keylen)
{
        struct iproc_ctx_s *ctx = crypto_ablkcipher_ctx(cipher);
        u32 tmp[DES_EXPKEY_WORDS];

        if (keylen == DES_KEY_SIZE) {
                if (des_ekey(tmp, key) == 0) {
                        if (crypto_ablkcipher_get_flags(cipher) &
                            CRYPTO_TFM_REQ_WEAK_KEY) {
                                u32 flags = CRYPTO_TFM_RES_WEAK_KEY;

                                crypto_ablkcipher_set_flags(cipher, flags);
                                return -EINVAL;
                        }
                }

                ctx->cipher_type = CIPHER_TYPE_DES;
        } else {
                crypto_ablkcipher_set_flags(cipher, CRYPTO_TFM_RES_BAD_KEY_LEN);
                return -EINVAL;
        }
        return 0;
}

static int threedes_setkey(struct crypto_ablkcipher *cipher, const u8 *key,
                           unsigned int keylen)
{
        struct iproc_ctx_s *ctx = crypto_ablkcipher_ctx(cipher);

        if (keylen == (DES_KEY_SIZE * 3)) {
                u32 flags;
                int ret;

                flags = crypto_ablkcipher_get_flags(cipher);
                ret = __des3_verify_key(&flags, key);
                if (unlikely(ret)) {
                        crypto_ablkcipher_set_flags(cipher, flags);
                        return ret;
                }

                ctx->cipher_type = CIPHER_TYPE_3DES;
        } else {
                crypto_ablkcipher_set_flags(cipher, CRYPTO_TFM_RES_BAD_KEY_LEN);
                return -EINVAL;
        }
        return 0;
}

static int aes_setkey(struct crypto_ablkcipher *cipher, const u8 *key,
                      unsigned int keylen)
{
        struct iproc_ctx_s *ctx = crypto_ablkcipher_ctx(cipher);

        if (ctx->cipher.mode == CIPHER_MODE_XTS)
                /* XTS includes two keys of equal length */
                keylen = keylen / 2;

        switch (keylen) {
        case AES_KEYSIZE_128:
                ctx->cipher_type = CIPHER_TYPE_AES128;
                break;
        case AES_KEYSIZE_192:
                ctx->cipher_type = CIPHER_TYPE_AES192;
                break;
        case AES_KEYSIZE_256:
                ctx->cipher_type = CIPHER_TYPE_AES256;
                break;
        default:
                crypto_ablkcipher_set_flags(cipher, CRYPTO_TFM_RES_BAD_KEY_LEN);
                return -EINVAL;
        }
        WARN_ON((ctx->max_payload != SPU_MAX_PAYLOAD_INF) &&
                ((ctx->max_payload % AES_BLOCK_SIZE) != 0));
        return 0;
}

static int rc4_setkey(struct crypto_ablkcipher *cipher, const u8 *key,
                      unsigned int keylen)
{
        struct iproc_ctx_s *ctx = crypto_ablkcipher_ctx(cipher);
        int i;

        ctx->enckeylen = ARC4_MAX_KEY_SIZE + ARC4_STATE_SIZE;

        ctx->enckey[0] = 0x00;  /* 0x00 */
        ctx->enckey[1] = 0x00;  /* i    */
        ctx->enckey[2] = 0x00;  /* 0x00 */
        ctx->enckey[3] = 0x00;  /* j    */
        for (i = 0; i < ARC4_MAX_KEY_SIZE; i++)
                ctx->enckey[i + ARC4_STATE_SIZE] = key[i % keylen];

        ctx->cipher_type = CIPHER_TYPE_INIT;

        return 0;
}

static int ablkcipher_setkey(struct crypto_ablkcipher *cipher, const u8 *key,
                             unsigned int keylen)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct iproc_ctx_s *ctx = crypto_ablkcipher_ctx(cipher);
        struct spu_cipher_parms cipher_parms;
        u32 alloc_len = 0;
        int err;

        flow_log("ablkcipher_setkey() keylen: %d\n", keylen);
        flow_dump("  key: ", key, keylen);

        switch (ctx->cipher.alg) {
        case CIPHER_ALG_DES:
                err = des_setkey(cipher, key, keylen);
                break;
        case CIPHER_ALG_3DES:
                err = threedes_setkey(cipher, key, keylen);
                break;
        case CIPHER_ALG_AES:
                err = aes_setkey(cipher, key, keylen);
                break;
        case CIPHER_ALG_RC4:
                err = rc4_setkey(cipher, key, keylen);
                break;
        default:
                pr_err("%s() Error: unknown cipher alg\n", __func__);
                err = -EINVAL;
        }
        if (err)
                return err;

        /* RC4 already populated ctx->enkey */
        if (ctx->cipher.alg != CIPHER_ALG_RC4) {
                memcpy(ctx->enckey, key, keylen);
                ctx->enckeylen = keylen;
        }
        /* SPU needs XTS keys in the reverse order the crypto API presents */
        if ((ctx->cipher.alg == CIPHER_ALG_AES) &&
            (ctx->cipher.mode == CIPHER_MODE_XTS)) {
                unsigned int xts_keylen = keylen / 2;

                memcpy(ctx->enckey, key + xts_keylen, xts_keylen);
                memcpy(ctx->enckey + xts_keylen, key, xts_keylen);
        }

        if (spu->spu_type == SPU_TYPE_SPUM)
                alloc_len = BCM_HDR_LEN + SPU_HEADER_ALLOC_LEN;
        else if (spu->spu_type == SPU_TYPE_SPU2)
                alloc_len = BCM_HDR_LEN + SPU2_HEADER_ALLOC_LEN;
        memset(ctx->bcm_spu_req_hdr, 0, alloc_len);
        cipher_parms.iv_buf = NULL;
        cipher_parms.iv_len = crypto_ablkcipher_ivsize(cipher);
        flow_log("%s: iv_len %u\n", __func__, cipher_parms.iv_len);

        cipher_parms.alg = ctx->cipher.alg;
        cipher_parms.mode = ctx->cipher.mode;
        cipher_parms.type = ctx->cipher_type;
        cipher_parms.key_buf = ctx->enckey;
        cipher_parms.key_len = ctx->enckeylen;

        /* Prepend SPU request message with BCM header */
        memcpy(ctx->bcm_spu_req_hdr, BCMHEADER, BCM_HDR_LEN);
        ctx->spu_req_hdr_len =
            spu->spu_cipher_req_init(ctx->bcm_spu_req_hdr + BCM_HDR_LEN,
                                     &cipher_parms);

        ctx->spu_resp_hdr_len = spu->spu_response_hdr_len(ctx->authkeylen,
                                                          ctx->enckeylen,
                                                          false);

        atomic_inc(&iproc_priv.setkey_cnt[SPU_OP_CIPHER]);

        return 0;
}

static int ablkcipher_encrypt(struct ablkcipher_request *req)
{
        flow_log("ablkcipher_encrypt() nbytes:%u\n", req->nbytes);

        return ablkcipher_enqueue(req, true);
}

static int ablkcipher_decrypt(struct ablkcipher_request *req)
{
        flow_log("ablkcipher_decrypt() nbytes:%u\n", req->nbytes);
        return ablkcipher_enqueue(req, false);
}

static int ahash_enqueue(struct ahash_request *req)
{
        struct iproc_reqctx_s *rctx = ahash_request_ctx(req);
        struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(tfm);
        int err = 0;
        const char *alg_name;

        flow_log("ahash_enqueue() nbytes:%u\n", req->nbytes);

        rctx->gfp = (req->base.flags & (CRYPTO_TFM_REQ_MAY_BACKLOG |
                       CRYPTO_TFM_REQ_MAY_SLEEP)) ? GFP_KERNEL : GFP_ATOMIC;
        rctx->parent = &req->base;
        rctx->ctx = ctx;
        rctx->bd_suppress = true;
        memset(&rctx->mb_mssg, 0, sizeof(struct brcm_message));

        /* Initialize position in src scatterlist */
        rctx->src_sg = req->src;
        rctx->src_skip = 0;
        rctx->src_nents = 0;
        rctx->dst_sg = NULL;
        rctx->dst_skip = 0;
        rctx->dst_nents = 0;

        /* SPU2 hardware does not compute hash of zero length data */
        if ((rctx->is_final == 1) && (rctx->total_todo == 0) &&
            (iproc_priv.spu.spu_type == SPU_TYPE_SPU2)) {
                alg_name = crypto_tfm_alg_name(crypto_ahash_tfm(tfm));
                flow_log("Doing %sfinal %s zero-len hash request in software\n",
                         rctx->is_final ? "" : "non-", alg_name);
                err = do_shash((unsigned char *)alg_name, req->result,
                               NULL, 0, NULL, 0, ctx->authkey,
                               ctx->authkeylen);
                if (err < 0)
                        flow_log("Hash request failed with error %d\n", err);
                return err;
        }
        /* Choose a SPU to process this request */
        rctx->chan_idx = select_channel();

        err = handle_ahash_req(rctx);
        if (err != -EINPROGRESS)
                /* synchronous result */
                spu_chunk_cleanup(rctx);

        if (err == -EAGAIN)
                /*
                 * we saved data in hash carry, but tell crypto API
                 * we successfully completed request.
                 */
                err = 0;

        return err;
}

static int __ahash_init(struct ahash_request *req)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct iproc_reqctx_s *rctx = ahash_request_ctx(req);
        struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(tfm);

        flow_log("%s()\n", __func__);

        /* Initialize the context */
        rctx->hash_carry_len = 0;
        rctx->is_final = 0;

        rctx->total_todo = 0;
        rctx->src_sent = 0;
        rctx->total_sent = 0;
        rctx->total_received = 0;

        ctx->digestsize = crypto_ahash_digestsize(tfm);
        /* If we add a hash whose digest is larger, catch it here. */
        WARN_ON(ctx->digestsize > MAX_DIGEST_SIZE);

        rctx->is_sw_hmac = false;

        ctx->spu_resp_hdr_len = spu->spu_response_hdr_len(ctx->authkeylen, 0,
                                                          true);

        return 0;
}

/**
 * spu_no_incr_hash() - Determine whether incremental hashing is supported.
 * @ctx:  Crypto session context
 *
 * SPU-2 does not support incremental hashing (we'll have to revisit and
 * condition based on chip revision or device tree entry if future versions do
 * support incremental hash)
 *
 * SPU-M also doesn't support incremental hashing of AES-XCBC
 *
 * Return: true if incremental hashing is not supported
 *         false otherwise
 */
bool spu_no_incr_hash(struct iproc_ctx_s *ctx)
{
        struct spu_hw *spu = &iproc_priv.spu;

        if (spu->spu_type == SPU_TYPE_SPU2)
                return true;

        if ((ctx->auth.alg == HASH_ALG_AES) &&
            (ctx->auth.mode == HASH_MODE_XCBC))
                return true;

        /* Otherwise, incremental hashing is supported */
        return false;
}

static int ahash_init(struct ahash_request *req)
{
        struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(tfm);
        const char *alg_name;
        struct crypto_shash *hash;
        int ret;
        gfp_t gfp;

        if (spu_no_incr_hash(ctx)) {
                /*
                 * If we get an incremental hashing request and it's not
                 * supported by the hardware, we need to handle it in software
                 * by calling synchronous hash functions.
                 */
                alg_name = crypto_tfm_alg_name(crypto_ahash_tfm(tfm));
                hash = crypto_alloc_shash(alg_name, 0, 0);
                if (IS_ERR(hash)) {
                        ret = PTR_ERR(hash);
                        goto err;
                }

                gfp = (req->base.flags & (CRYPTO_TFM_REQ_MAY_BACKLOG |
                       CRYPTO_TFM_REQ_MAY_SLEEP)) ? GFP_KERNEL : GFP_ATOMIC;
                ctx->shash = kmalloc(sizeof(*ctx->shash) +
                                     crypto_shash_descsize(hash), gfp);
                if (!ctx->shash) {
                        ret = -ENOMEM;
                        goto err_hash;
                }
                ctx->shash->tfm = hash;
                ctx->shash->flags = 0;

                /* Set the key using data we already have from setkey */
                if (ctx->authkeylen > 0) {
                        ret = crypto_shash_setkey(hash, ctx->authkey,
                                                  ctx->authkeylen);
                        if (ret)
                                goto err_shash;
                }

                /* Initialize hash w/ this key and other params */
                ret = crypto_shash_init(ctx->shash);
                if (ret)
                        goto err_shash;
        } else {
                /* Otherwise call the internal function which uses SPU hw */
                ret = __ahash_init(req);
        }

        return ret;

err_shash:
        kfree(ctx->shash);
err_hash:
        crypto_free_shash(hash);
err:
        return ret;
}

static int __ahash_update(struct ahash_request *req)
{
        struct iproc_reqctx_s *rctx = ahash_request_ctx(req);

        flow_log("ahash_update() nbytes:%u\n", req->nbytes);

        if (!req->nbytes)
                return 0;
        rctx->total_todo += req->nbytes;
        rctx->src_sent = 0;

        return ahash_enqueue(req);
}

static int ahash_update(struct ahash_request *req)
{
        struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(tfm);
        u8 *tmpbuf;
        int ret;
        int nents;
        gfp_t gfp;

        if (spu_no_incr_hash(ctx)) {
                /*
                 * If we get an incremental hashing request and it's not
                 * supported by the hardware, we need to handle it in software
                 * by calling synchronous hash functions.
                 */
                if (req->src)
                        nents = sg_nents(req->src);
                else
                        return -EINVAL;

                /* Copy data from req scatterlist to tmp buffer */
                gfp = (req->base.flags & (CRYPTO_TFM_REQ_MAY_BACKLOG |
                       CRYPTO_TFM_REQ_MAY_SLEEP)) ? GFP_KERNEL : GFP_ATOMIC;
                tmpbuf = kmalloc(req->nbytes, gfp);
                if (!tmpbuf)
                        return -ENOMEM;

                if (sg_copy_to_buffer(req->src, nents, tmpbuf, req->nbytes) !=
                                req->nbytes) {
                        kfree(tmpbuf);
                        return -EINVAL;
                }

                /* Call synchronous update */
                ret = crypto_shash_update(ctx->shash, tmpbuf, req->nbytes);
                kfree(tmpbuf);
        } else {
                /* Otherwise call the internal function which uses SPU hw */
                ret = __ahash_update(req);
        }

        return ret;
}

static int __ahash_final(struct ahash_request *req)
{
        struct iproc_reqctx_s *rctx = ahash_request_ctx(req);

        flow_log("ahash_final() nbytes:%u\n", req->nbytes);

        rctx->is_final = 1;

        return ahash_enqueue(req);
}

static int ahash_final(struct ahash_request *req)
{
        struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(tfm);
        int ret;

        if (spu_no_incr_hash(ctx)) {
                /*
                 * If we get an incremental hashing request and it's not
                 * supported by the hardware, we need to handle it in software
                 * by calling synchronous hash functions.
                 */
                ret = crypto_shash_final(ctx->shash, req->result);

                /* Done with hash, can deallocate it now */
                crypto_free_shash(ctx->shash->tfm);
                kfree(ctx->shash);

        } else {
                /* Otherwise call the internal function which uses SPU hw */
                ret = __ahash_final(req);
        }

        return ret;
}

static int __ahash_finup(struct ahash_request *req)
{
        struct iproc_reqctx_s *rctx = ahash_request_ctx(req);

        flow_log("ahash_finup() nbytes:%u\n", req->nbytes);

        rctx->total_todo += req->nbytes;
        rctx->src_sent = 0;
        rctx->is_final = 1;

        return ahash_enqueue(req);
}

static int ahash_finup(struct ahash_request *req)
{
        struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(tfm);
        u8 *tmpbuf;
        int ret;
        int nents;
        gfp_t gfp;

        if (spu_no_incr_hash(ctx)) {
                /*
                 * If we get an incremental hashing request and it's not
                 * supported by the hardware, we need to handle it in software
                 * by calling synchronous hash functions.
                 */
                if (req->src) {
                        nents = sg_nents(req->src);
                } else {
                        ret = -EINVAL;
                        goto ahash_finup_exit;
                }

                /* Copy data from req scatterlist to tmp buffer */
                gfp = (req->base.flags & (CRYPTO_TFM_REQ_MAY_BACKLOG |
                       CRYPTO_TFM_REQ_MAY_SLEEP)) ? GFP_KERNEL : GFP_ATOMIC;
                tmpbuf = kmalloc(req->nbytes, gfp);
                if (!tmpbuf) {
                        ret = -ENOMEM;
                        goto ahash_finup_exit;
                }

                if (sg_copy_to_buffer(req->src, nents, tmpbuf, req->nbytes) !=
                                req->nbytes) {
                        ret = -EINVAL;
                        goto ahash_finup_free;
                }

                /* Call synchronous update */
                ret = crypto_shash_finup(ctx->shash, tmpbuf, req->nbytes,
                                         req->result);
        } else {
                /* Otherwise call the internal function which uses SPU hw */
                return __ahash_finup(req);
        }
ahash_finup_free:
        kfree(tmpbuf);

ahash_finup_exit:
        /* Done with hash, can deallocate it now */
        crypto_free_shash(ctx->shash->tfm);
        kfree(ctx->shash);
        return ret;
}

static int ahash_digest(struct ahash_request *req)
{
        int err = 0;

        flow_log("ahash_digest() nbytes:%u\n", req->nbytes);

        /* whole thing at once */
        err = __ahash_init(req);
        if (!err)
                err = __ahash_finup(req);

        return err;
}

static int ahash_setkey(struct crypto_ahash *ahash, const u8 *key,
                        unsigned int keylen)
{
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(ahash);

        flow_log("%s() ahash:%p key:%p keylen:%u\n",
                 __func__, ahash, key, keylen);
        flow_dump("  key: ", key, keylen);

        if (ctx->auth.alg == HASH_ALG_AES) {
                switch (keylen) {
                case AES_KEYSIZE_128:
                        ctx->cipher_type = CIPHER_TYPE_AES128;
                        break;
                case AES_KEYSIZE_192:
                        ctx->cipher_type = CIPHER_TYPE_AES192;
                        break;
                case AES_KEYSIZE_256:
                        ctx->cipher_type = CIPHER_TYPE_AES256;
                        break;
                default:
                        pr_err("%s() Error: Invalid key length\n", __func__);
                        return -EINVAL;
                }
        } else {
                pr_err("%s() Error: unknown hash alg\n", __func__);
                return -EINVAL;
        }
        memcpy(ctx->authkey, key, keylen);
        ctx->authkeylen = keylen;

        return 0;
}

static int ahash_export(struct ahash_request *req, void *out)
{
        const struct iproc_reqctx_s *rctx = ahash_request_ctx(req);
        struct spu_hash_export_s *spu_exp = (struct spu_hash_export_s *)out;

        spu_exp->total_todo = rctx->total_todo;
        spu_exp->total_sent = rctx->total_sent;
        spu_exp->is_sw_hmac = rctx->is_sw_hmac;
        memcpy(spu_exp->hash_carry, rctx->hash_carry, sizeof(rctx->hash_carry));
        spu_exp->hash_carry_len = rctx->hash_carry_len;
        memcpy(spu_exp->incr_hash, rctx->incr_hash, sizeof(rctx->incr_hash));

        return 0;
}

static int ahash_import(struct ahash_request *req, const void *in)
{
        struct iproc_reqctx_s *rctx = ahash_request_ctx(req);
        struct spu_hash_export_s *spu_exp = (struct spu_hash_export_s *)in;

        rctx->total_todo = spu_exp->total_todo;
        rctx->total_sent = spu_exp->total_sent;
        rctx->is_sw_hmac = spu_exp->is_sw_hmac;
        memcpy(rctx->hash_carry, spu_exp->hash_carry, sizeof(rctx->hash_carry));
        rctx->hash_carry_len = spu_exp->hash_carry_len;
        memcpy(rctx->incr_hash, spu_exp->incr_hash, sizeof(rctx->incr_hash));

        return 0;
}

static int ahash_hmac_setkey(struct crypto_ahash *ahash, const u8 *key,
                             unsigned int keylen)
{
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(ahash);
        unsigned int blocksize =
                crypto_tfm_alg_blocksize(crypto_ahash_tfm(ahash));
        unsigned int digestsize = crypto_ahash_digestsize(ahash);
        unsigned int index;
        int rc;

        flow_log("%s() ahash:%p key:%p keylen:%u blksz:%u digestsz:%u\n",
                 __func__, ahash, key, keylen, blocksize, digestsize);
        flow_dump("  key: ", key, keylen);

        if (keylen > blocksize) {
                switch (ctx->auth.alg) {
                case HASH_ALG_MD5:
                        rc = do_shash("md5", ctx->authkey, key, keylen, NULL,
                                      0, NULL, 0);
                        break;
                case HASH_ALG_SHA1:
                        rc = do_shash("sha1", ctx->authkey, key, keylen, NULL,
                                      0, NULL, 0);
                        break;
                case HASH_ALG_SHA224:
                        rc = do_shash("sha224", ctx->authkey, key, keylen, NULL,
                                      0, NULL, 0);
                        break;
                case HASH_ALG_SHA256:
                        rc = do_shash("sha256", ctx->authkey, key, keylen, NULL,
                                      0, NULL, 0);
                        break;
                case HASH_ALG_SHA384:
                        rc = do_shash("sha384", ctx->authkey, key, keylen, NULL,
                                      0, NULL, 0);
                        break;
                case HASH_ALG_SHA512:
                        rc = do_shash("sha512", ctx->authkey, key, keylen, NULL,
                                      0, NULL, 0);
                        break;
                case HASH_ALG_SHA3_224:
                        rc = do_shash("sha3-224", ctx->authkey, key, keylen,
                                      NULL, 0, NULL, 0);
                        break;
                case HASH_ALG_SHA3_256:
                        rc = do_shash("sha3-256", ctx->authkey, key, keylen,
                                      NULL, 0, NULL, 0);
                        break;
                case HASH_ALG_SHA3_384:
                        rc = do_shash("sha3-384", ctx->authkey, key, keylen,
                                      NULL, 0, NULL, 0);
                        break;
                case HASH_ALG_SHA3_512:
                        rc = do_shash("sha3-512", ctx->authkey, key, keylen,
                                      NULL, 0, NULL, 0);
                        break;
                default:
                        pr_err("%s() Error: unknown hash alg\n", __func__);
                        return -EINVAL;
                }
                if (rc < 0) {
                        pr_err("%s() Error %d computing shash for %s\n",
                               __func__, rc, hash_alg_name[ctx->auth.alg]);
                        return rc;
                }
                ctx->authkeylen = digestsize;

                flow_log("  keylen > digestsize... hashed\n");
                flow_dump("  newkey: ", ctx->authkey, ctx->authkeylen);
        } else {
                memcpy(ctx->authkey, key, keylen);
                ctx->authkeylen = keylen;
        }

        /*
         * Full HMAC operation in SPUM is not verified,
         * So keeping the generation of IPAD, OPAD and
         * outer hashing in software.
         */
        if (iproc_priv.spu.spu_type == SPU_TYPE_SPUM) {
                memcpy(ctx->ipad, ctx->authkey, ctx->authkeylen);
                memset(ctx->ipad + ctx->authkeylen, 0,
                       blocksize - ctx->authkeylen);
                ctx->authkeylen = 0;
                memcpy(ctx->opad, ctx->ipad, blocksize);

                for (index = 0; index < blocksize; index++) {
                        ctx->ipad[index] ^= HMAC_IPAD_VALUE;
                        ctx->opad[index] ^= HMAC_OPAD_VALUE;
                }

                flow_dump("  ipad: ", ctx->ipad, blocksize);
                flow_dump("  opad: ", ctx->opad, blocksize);
        }
        ctx->digestsize = digestsize;
        atomic_inc(&iproc_priv.setkey_cnt[SPU_OP_HMAC]);

        return 0;
}

static int ahash_hmac_init(struct ahash_request *req)
{
        struct iproc_reqctx_s *rctx = ahash_request_ctx(req);
        struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(tfm);
        unsigned int blocksize =
                        crypto_tfm_alg_blocksize(crypto_ahash_tfm(tfm));

        flow_log("ahash_hmac_init()\n");

        /* init the context as a hash */
        ahash_init(req);

        if (!spu_no_incr_hash(ctx)) {
                /* SPU-M can do incr hashing but needs sw for outer HMAC */
                rctx->is_sw_hmac = true;
                ctx->auth.mode = HASH_MODE_HASH;
                /* start with a prepended ipad */
                memcpy(rctx->hash_carry, ctx->ipad, blocksize);
                rctx->hash_carry_len = blocksize;
                rctx->total_todo += blocksize;
        }

        return 0;
}

static int ahash_hmac_update(struct ahash_request *req)
{
        flow_log("ahash_hmac_update() nbytes:%u\n", req->nbytes);

        if (!req->nbytes)
                return 0;

        return ahash_update(req);
}

static int ahash_hmac_final(struct ahash_request *req)
{
        flow_log("ahash_hmac_final() nbytes:%u\n", req->nbytes);

        return ahash_final(req);
}

static int ahash_hmac_finup(struct ahash_request *req)
{
        flow_log("ahash_hmac_finupl() nbytes:%u\n", req->nbytes);

        return ahash_finup(req);
}

static int ahash_hmac_digest(struct ahash_request *req)
{
        struct iproc_reqctx_s *rctx = ahash_request_ctx(req);
        struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_ahash_ctx(tfm);
        unsigned int blocksize =
                        crypto_tfm_alg_blocksize(crypto_ahash_tfm(tfm));

        flow_log("ahash_hmac_digest() nbytes:%u\n", req->nbytes);

        /* Perform initialization and then call finup */
        __ahash_init(req);

        if (iproc_priv.spu.spu_type == SPU_TYPE_SPU2) {
                /*
                 * SPU2 supports full HMAC implementation in the
                 * hardware, need not to generate IPAD, OPAD and
                 * outer hash in software.
                 * Only for hash key len > hash block size, SPU2
                 * expects to perform hashing on the key, shorten
                 * it to digest size and feed it as hash key.
                 */
                rctx->is_sw_hmac = false;
                ctx->auth.mode = HASH_MODE_HMAC;
        } else {
                rctx->is_sw_hmac = true;
                ctx->auth.mode = HASH_MODE_HASH;
                /* start with a prepended ipad */
                memcpy(rctx->hash_carry, ctx->ipad, blocksize);
                rctx->hash_carry_len = blocksize;
                rctx->total_todo += blocksize;
        }

        return __ahash_finup(req);
}

/* aead helpers */

static int aead_need_fallback(struct aead_request *req)
{
        struct iproc_reqctx_s *rctx = aead_request_ctx(req);
        struct spu_hw *spu = &iproc_priv.spu;
        struct crypto_aead *aead = crypto_aead_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_aead_ctx(aead);
        u32 payload_len;

        /*
         * SPU hardware cannot handle the AES-GCM/CCM case where plaintext
         * and AAD are both 0 bytes long. So use fallback in this case.
         */
        if (((ctx->cipher.mode == CIPHER_MODE_GCM) ||
             (ctx->cipher.mode == CIPHER_MODE_CCM)) &&
            (req->assoclen == 0)) {
                if ((rctx->is_encrypt && (req->cryptlen == 0)) ||
                    (!rctx->is_encrypt && (req->cryptlen == ctx->digestsize))) {
                        flow_log("AES GCM/CCM needs fallback for 0 len req\n");
                        return 1;
                }
        }

        /* SPU-M hardware only supports CCM digest size of 8, 12, or 16 bytes */
        if ((ctx->cipher.mode == CIPHER_MODE_CCM) &&
            (spu->spu_type == SPU_TYPE_SPUM) &&
            (ctx->digestsize != 8) && (ctx->digestsize != 12) &&
            (ctx->digestsize != 16)) {
                flow_log("%s() AES CCM needs fallback for digest size %d\n",
                         __func__, ctx->digestsize);
                return 1;
        }

        /*
         * SPU-M on NSP has an issue where AES-CCM hash is not correct
         * when AAD size is 0
         */
        if ((ctx->cipher.mode == CIPHER_MODE_CCM) &&
            (spu->spu_subtype == SPU_SUBTYPE_SPUM_NSP) &&
            (req->assoclen == 0)) {
                flow_log("%s() AES_CCM needs fallback for 0 len AAD on NSP\n",
                         __func__);
                return 1;
        }

        payload_len = req->cryptlen;
        if (spu->spu_type == SPU_TYPE_SPUM)
                payload_len += req->assoclen;

        flow_log("%s() payload len: %u\n", __func__, payload_len);

        if (ctx->max_payload == SPU_MAX_PAYLOAD_INF)
                return 0;
        else
                return payload_len > ctx->max_payload;
}

static void aead_complete(struct crypto_async_request *areq, int err)
{
        struct aead_request *req =
            container_of(areq, struct aead_request, base);
        struct iproc_reqctx_s *rctx = aead_request_ctx(req);
        struct crypto_aead *aead = crypto_aead_reqtfm(req);

        flow_log("%s() err:%d\n", __func__, err);

        areq->tfm = crypto_aead_tfm(aead);

        areq->complete = rctx->old_complete;
        areq->data = rctx->old_data;

        areq->complete(areq, err);
}

static int aead_do_fallback(struct aead_request *req, bool is_encrypt)
{
        struct crypto_aead *aead = crypto_aead_reqtfm(req);
        struct crypto_tfm *tfm = crypto_aead_tfm(aead);
        struct iproc_reqctx_s *rctx = aead_request_ctx(req);
        struct iproc_ctx_s *ctx = crypto_tfm_ctx(tfm);
        int err;
        u32 req_flags;

        flow_log("%s() enc:%u\n", __func__, is_encrypt);

        if (ctx->fallback_cipher) {
                /* Store the cipher tfm and then use the fallback tfm */
                rctx->old_tfm = tfm;
                aead_request_set_tfm(req, ctx->fallback_cipher);
                /*
                 * Save the callback and chain ourselves in, so we can restore
                 * the tfm
                 */
                rctx->old_complete = req->base.complete;
                rctx->old_data = req->base.data;
                req_flags = aead_request_flags(req);
                aead_request_set_callback(req, req_flags, aead_complete, req);
                err = is_encrypt ? crypto_aead_encrypt(req) :
                    crypto_aead_decrypt(req);

                if (err == 0) {
                        /*
                         * fallback was synchronous (did not return
                         * -EINPROGRESS). So restore request state here.
                         */
                        aead_request_set_callback(req, req_flags,
                                                  rctx->old_complete, req);
                        req->base.data = rctx->old_data;
                        aead_request_set_tfm(req, aead);
                        flow_log("%s() fallback completed successfully\n\n",
                                 __func__);
                }
        } else {
                err = -EINVAL;
        }

        return err;
}

static int aead_enqueue(struct aead_request *req, bool is_encrypt)
{
        struct iproc_reqctx_s *rctx = aead_request_ctx(req);
        struct crypto_aead *aead = crypto_aead_reqtfm(req);
        struct iproc_ctx_s *ctx = crypto_aead_ctx(aead);
        int err;

        flow_log("%s() enc:%u\n", __func__, is_encrypt);

        if (req->assoclen > MAX_ASSOC_SIZE) {
                pr_err
                    ("%s() Error: associated data too long. (%u > %u bytes)\n",
                     __func__, req->assoclen, MAX_ASSOC_SIZE);
                return -EINVAL;
        }

        rctx->gfp = (req->base.flags & (CRYPTO_TFM_REQ_MAY_BACKLOG |
                       CRYPTO_TFM_REQ_MAY_SLEEP)) ? GFP_KERNEL : GFP_ATOMIC;
        rctx->parent = &req->base;
        rctx->is_encrypt = is_encrypt;
        rctx->bd_suppress = false;
        rctx->total_todo = req->cryptlen;
        rctx->src_sent = 0;
        rctx->total_sent = 0;
        rctx->total_received = 0;
        rctx->is_sw_hmac = false;
        rctx->ctx = ctx;
        memset(&rctx->mb_mssg, 0, sizeof(struct brcm_message));

        /* assoc data is at start of src sg */
        rctx->assoc = req->src;

        /*
         * Init current position in src scatterlist to be after assoc data.
         * src_skip set to buffer offset where data begins. (Assoc data could
         * end in the middle of a buffer.)
         */
        if (spu_sg_at_offset(req->src, req->assoclen, &rctx->src_sg,
                             &rctx->src_skip) < 0) {
                pr_err("%s() Error: Unable to find start of src data\n",
                       __func__);
                return -EINVAL;
        }

        rctx->src_nents = 0;
        rctx->dst_nents = 0;
        if (req->dst == req->src) {
                rctx->dst_sg = rctx->src_sg;
                rctx->dst_skip = rctx->src_skip;
        } else {
                /*
                 * Expect req->dst to have room for assoc data followed by
                 * output data and ICV, if encrypt. So initialize dst_sg
                 * to point beyond assoc len offset.
                 */
                if (spu_sg_at_offset(req->dst, req->assoclen, &rctx->dst_sg,
                                     &rctx->dst_skip) < 0) {
                        pr_err("%s() Error: Unable to find start of dst data\n",
                               __func__);
                        return -EINVAL;
                }
        }

        if (ctx->cipher.mode == CIPHER_MODE_CBC ||
            ctx->cipher.mode == CIPHER_MODE_CTR ||
            ctx->cipher.mode == CIPHER_MODE_OFB ||
            ctx->cipher.mode == CIPHER_MODE_XTS ||
            ctx->cipher.mode == CIPHER_MODE_GCM) {
                rctx->iv_ctr_len =
                        ctx->salt_len +
                        crypto_aead_ivsize(crypto_aead_reqtfm(req));
        } else if (ctx->cipher.mode == CIPHER_MODE_CCM) {
                rctx->iv_ctr_len = CCM_AES_IV_SIZE;
        } else {
                rctx->iv_ctr_len = 0;
        }

        rctx->hash_carry_len = 0;

        flow_log("  src sg: %p\n", req->src);
        flow_log("  rctx->src_sg: %p, src_skip %u\n",
                 rctx->src_sg, rctx->src_skip);
        flow_log("  assoc:  %p, assoclen %u\n", rctx->assoc, req->assoclen);
        flow_log("  dst sg: %p\n", req->dst);
        flow_log("  rctx->dst_sg: %p, dst_skip %u\n",
                 rctx->dst_sg, rctx->dst_skip);
        flow_log("  iv_ctr_len:%u\n", rctx->iv_ctr_len);
        flow_dump("  iv: ", req->iv, rctx->iv_ctr_len);
        flow_log("  authkeylen:%u\n", ctx->authkeylen);
        flow_log("  is_esp: %s\n", ctx->is_esp ? "yes" : "no");

        if (ctx->max_payload == SPU_MAX_PAYLOAD_INF)
                flow_log("  max_payload infinite");
        else
                flow_log("  max_payload: %u\n", ctx->max_payload);

        if (unlikely(aead_need_fallback(req)))
                return aead_do_fallback(req, is_encrypt);

        /*
         * Do memory allocations for request after fallback check, because if we
         * do fallback, we won't call finish_req() to dealloc.
         */
        if (rctx->iv_ctr_len) {
                if (ctx->salt_len)
                        memcpy(rctx->msg_buf.iv_ctr + ctx->salt_offset,
                               ctx->salt, ctx->salt_len);
                memcpy(rctx->msg_buf.iv_ctr + ctx->salt_offset + ctx->salt_len,
                       req->iv,
                       rctx->iv_ctr_len - ctx->salt_len - ctx->salt_offset);
        }

        rctx->chan_idx = select_channel();
        err = handle_aead_req(rctx);
        if (err != -EINPROGRESS)
                /* synchronous result */
                spu_chunk_cleanup(rctx);

        return err;
}

static int aead_authenc_setkey(struct crypto_aead *cipher,
                               const u8 *key, unsigned int keylen)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct iproc_ctx_s *ctx = crypto_aead_ctx(cipher);
        struct crypto_tfm *tfm = crypto_aead_tfm(cipher);
        struct rtattr *rta = (void *)key;
        struct crypto_authenc_key_param *param;
        const u8 *origkey = key;
        const unsigned int origkeylen = keylen;

        int ret = 0;

        flow_log("%s() aead:%p key:%p keylen:%u\n", __func__, cipher, key,
                 keylen);
        flow_dump("  key: ", key, keylen);

        if (!RTA_OK(rta, keylen))
                goto badkey;
        if (rta->rta_type != CRYPTO_AUTHENC_KEYA_PARAM)
                goto badkey;
        if (RTA_PAYLOAD(rta) < sizeof(*param))
                goto badkey;

        param = RTA_DATA(rta);
        ctx->enckeylen = be32_to_cpu(param->enckeylen);

        key += RTA_ALIGN(rta->rta_len);
        keylen -= RTA_ALIGN(rta->rta_len);

        if (keylen < ctx->enckeylen)
                goto badkey;
        if (ctx->enckeylen > MAX_KEY_SIZE)
                goto badkey;

        ctx->authkeylen = keylen - ctx->enckeylen;

        if (ctx->authkeylen > MAX_KEY_SIZE)
                goto badkey;

        memcpy(ctx->enckey, key + ctx->authkeylen, ctx->enckeylen);
        /* May end up padding auth key. So make sure it's zeroed. */
        memset(ctx->authkey, 0, sizeof(ctx->authkey));
        memcpy(ctx->authkey, key, ctx->authkeylen);

        switch (ctx->alg->cipher_info.alg) {
        case CIPHER_ALG_DES:
                if (ctx->enckeylen == DES_KEY_SIZE) {
                        u32 tmp[DES_EXPKEY_WORDS];
                        u32 flags = CRYPTO_TFM_RES_WEAK_KEY;

                        if (des_ekey(tmp, key) == 0) {
                                if (crypto_aead_get_flags(cipher) &
                                    CRYPTO_TFM_REQ_WEAK_KEY) {
                                        crypto_aead_set_flags(cipher, flags);
                                        return -EINVAL;
                                }
                        }

                        ctx->cipher_type = CIPHER_TYPE_DES;
                } else {
                        goto badkey;
                }
                break;
        case CIPHER_ALG_3DES:
                if (ctx->enckeylen == (DES_KEY_SIZE * 3)) {
                        u32 flags;

                        flags = crypto_aead_get_flags(cipher);
                        ret = __des3_verify_key(&flags, ctx->enckey);
                        if (unlikely(ret)) {
                                crypto_aead_set_flags(cipher, flags);
                                return ret;
                        }

                        ctx->cipher_type = CIPHER_TYPE_3DES;
                } else {
                        crypto_aead_set_flags(cipher,
                                              CRYPTO_TFM_RES_BAD_KEY_LEN);
                        return -EINVAL;
                }
                break;
        case CIPHER_ALG_AES:
                switch (ctx->enckeylen) {
                case AES_KEYSIZE_128:
                        ctx->cipher_type = CIPHER_TYPE_AES128;
                        break;
                case AES_KEYSIZE_192:
                        ctx->cipher_type = CIPHER_TYPE_AES192;
                        break;
                case AES_KEYSIZE_256:
                        ctx->cipher_type = CIPHER_TYPE_AES256;
                        break;
                default:
                        goto badkey;
                }
                break;
        case CIPHER_ALG_RC4:
                ctx->cipher_type = CIPHER_TYPE_INIT;
                break;
        default:
                pr_err("%s() Error: Unknown cipher alg\n", __func__);
                return -EINVAL;
        }

        flow_log("  enckeylen:%u authkeylen:%u\n", ctx->enckeylen,
                 ctx->authkeylen);
        flow_dump("  enc: ", ctx->enckey, ctx->enckeylen);
        flow_dump("  auth: ", ctx->authkey, ctx->authkeylen);

        /* setkey the fallback just in case we needto use it */
        if (ctx->fallback_cipher) {
                flow_log("  running fallback setkey()\n");

                ctx->fallback_cipher->base.crt_flags &= ~CRYPTO_TFM_REQ_MASK;
                ctx->fallback_cipher->base.crt_flags |=
                    tfm->crt_flags & CRYPTO_TFM_REQ_MASK;
                ret =
                    crypto_aead_setkey(ctx->fallback_cipher, origkey,
                                       origkeylen);
                if (ret) {
                        flow_log("  fallback setkey() returned:%d\n", ret);
                        tfm->crt_flags &= ~CRYPTO_TFM_RES_MASK;
                        tfm->crt_flags |=
                            (ctx->fallback_cipher->base.crt_flags &
                             CRYPTO_TFM_RES_MASK);
                }
        }

        ctx->spu_resp_hdr_len = spu->spu_response_hdr_len(ctx->authkeylen,
                                                          ctx->enckeylen,
                                                          false);

        atomic_inc(&iproc_priv.setkey_cnt[SPU_OP_AEAD]);

        return ret;

badkey:
        ctx->enckeylen = 0;
        ctx->authkeylen = 0;
        ctx->digestsize = 0;

        crypto_aead_set_flags(cipher, CRYPTO_TFM_RES_BAD_KEY_LEN);
        return -EINVAL;
}

static int aead_gcm_ccm_setkey(struct crypto_aead *cipher,
                               const u8 *key, unsigned int keylen)
{
        struct spu_hw *spu = &iproc_priv.spu;
        struct iproc_ctx_s *ctx = crypto_aead_ctx(cipher);
        struct crypto_tfm *tfm = crypto_aead_tfm(cipher);

        int ret = 0;

        flow_log("%s() keylen:%u\n", __func__, keylen);
        flow_dump("  key: ", key, keylen);