crypto: caam/qi2 - add DPAA2-CAAM driver

Add CAAM driver that works using the DPSECI backend, i.e. manages
DPSECI DPAA2 objects sitting on the Management Complex (MC) fsl-mc bus.

Data transfers (crypto requests) are sent/received to/from CAAM crypto
engine via Queue Interface (v2), this being similar to existing caam/qi.
OTOH, configuration/setup (obtaining virtual queue IDs, authorization
etc.) is done by sending commands to the MC f/w.

Note that the CAAM accelerator included in DPAA2 platforms still has
Job Rings. However, the driver being added does not handle access
via this backend. Kconfig & Makefile are updated such that DPAA2-CAAM
(a.k.a. "caam/qi2") driver does not depend on caam/jr or caam/qi
backends - which rely on platform bus support (ctrl.c).

Support for the following aead and authenc algorithms is also added
in this patch:
-aead:
gcm(aes)
rfc4106(gcm(aes))
rfc4543(gcm(aes))
-authenc:
authenc(hmac({md5,sha*}),cbc({aes,des,des3_ede}))
echainiv(authenc(hmac({md5,sha*}),cbc({aes,des,des3_ede})))
authenc(hmac({md5,sha*}),rfc3686(ctr(aes))
seqiv(authenc(hmac({md5,sha*}),rfc3686(ctr(aes)))

Signed-off-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

drivers/crypto/Makefile

@@ -10,7 +10,7 @@ obj-$(CONFIG_CRYPTO_DEV_CHELSIO) += chelsio/
 obj-$(CONFIG_CRYPTO_DEV_CPT) += cavium/cpt/
 obj-$(CONFIG_CRYPTO_DEV_NITROX) += cavium/nitrox/
 obj-$(CONFIG_CRYPTO_DEV_EXYNOS_RNG) += exynos-rng.o
-obj-$(CONFIG_CRYPTO_DEV_FSL_CAAM) += caam/
+obj-$(CONFIG_CRYPTO_DEV_FSL_CAAM_COMMON) += caam/
 obj-$(CONFIG_CRYPTO_DEV_GEODE) += geode-aes.o
 obj-$(CONFIG_CRYPTO_DEV_HIFN_795X) += hifn_795x.o
 obj-$(CONFIG_CRYPTO_DEV_IMGTEC_HASH) += img-hash.o

drivers/crypto/caam/Kconfig

@@ -1,7 +1,11 @@
+config CRYPTO_DEV_FSL_CAAM_COMMON
+	tristate
+
 config CRYPTO_DEV_FSL_CAAM
-	tristate "Freescale CAAM-Multicore driver backend"
+	tristate "Freescale CAAM-Multicore platform driver backend"
 	depends on FSL_SOC || ARCH_MXC || ARCH_LAYERSCAPE
 	select SOC_BUS
+	select CRYPTO_DEV_FSL_CAAM_COMMON
 	help
 	  Enables the driver module for Freescale's Cryptographic Accelerator
 	  and Assurance Module (CAAM), also known as the SEC version 4 (SEC4).
@@ -12,9 +16,16 @@ config CRYPTO_DEV_FSL_CAAM
 	  To compile this driver as a module, choose M here: the module
 	  will be called caam.
 
+if CRYPTO_DEV_FSL_CAAM
+
+config CRYPTO_DEV_FSL_CAAM_DEBUG
+	bool "Enable debug output in CAAM driver"
+	help
+	  Selecting this will enable printing of various debug
+	  information in the CAAM driver.
+
 config CRYPTO_DEV_FSL_CAAM_JR
 	tristate "Freescale CAAM Job Ring driver backend"
-	depends on CRYPTO_DEV_FSL_CAAM
 	default y
 	help
 	  Enables the driver module for Job Rings which are part of
@@ -25,9 +36,10 @@ config CRYPTO_DEV_FSL_CAAM_JR
 	  To compile this driver as a module, choose M here: the module
 	  will be called caam_jr.
 
+if CRYPTO_DEV_FSL_CAAM_JR
+
 config CRYPTO_DEV_FSL_CAAM_RINGSIZE
 	int "Job Ring size"
-	depends on CRYPTO_DEV_FSL_CAAM_JR
 	range 2 9
 	default "9"
 	help
@@ -45,7 +57,6 @@ config CRYPTO_DEV_FSL_CAAM_RINGSIZE
 
 config CRYPTO_DEV_FSL_CAAM_INTC
 	bool "Job Ring interrupt coalescing"
-	depends on CRYPTO_DEV_FSL_CAAM_JR
 	help
 	  Enable the Job Ring's interrupt coalescing feature.
 
@@ -75,7 +86,6 @@ config CRYPTO_DEV_FSL_CAAM_INTC_TIME_THLD
 
 config CRYPTO_DEV_FSL_CAAM_CRYPTO_API
 	tristate "Register algorithm implementations with the Crypto API"
-	depends on CRYPTO_DEV_FSL_CAAM_JR
 	default y
 	select CRYPTO_AEAD
 	select CRYPTO_AUTHENC
@@ -90,7 +100,7 @@ config CRYPTO_DEV_FSL_CAAM_CRYPTO_API
 
 config CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI
 	tristate "Queue Interface as Crypto API backend"
-	depends on CRYPTO_DEV_FSL_CAAM_JR && FSL_DPAA && NET
+	depends on FSL_DPAA && NET
 	default y
 	select CRYPTO_AUTHENC
 	select CRYPTO_BLKCIPHER
@@ -107,7 +117,6 @@ config CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI
 
 config CRYPTO_DEV_FSL_CAAM_AHASH_API
 	tristate "Register hash algorithm implementations with Crypto API"
-	depends on CRYPTO_DEV_FSL_CAAM_JR
 	default y
 	select CRYPTO_HASH
 	help
@@ -119,7 +128,6 @@ config CRYPTO_DEV_FSL_CAAM_AHASH_API
 
 config CRYPTO_DEV_FSL_CAAM_PKC_API
         tristate "Register public key cryptography implementations with Crypto API"
-        depends on CRYPTO_DEV_FSL_CAAM_JR
         default y
         select CRYPTO_RSA
         help
@@ -131,7 +139,6 @@ config CRYPTO_DEV_FSL_CAAM_PKC_API
 
 config CRYPTO_DEV_FSL_CAAM_RNG_API
 	tristate "Register caam device for hwrng API"
-	depends on CRYPTO_DEV_FSL_CAAM_JR
 	default y
 	select CRYPTO_RNG
 	select HW_RANDOM
@@ -142,13 +149,25 @@ config CRYPTO_DEV_FSL_CAAM_RNG_API
 	  To compile this as a module, choose M here: the module
 	  will be called caamrng.
 
-config CRYPTO_DEV_FSL_CAAM_DEBUG
+endif # CRYPTO_DEV_FSL_CAAM_JR
-	bool "Enable debug output in CAAM driver"
+
-	depends on CRYPTO_DEV_FSL_CAAM
+endif # CRYPTO_DEV_FSL_CAAM
+
+config CRYPTO_DEV_FSL_DPAA2_CAAM
+	tristate "QorIQ DPAA2 CAAM (DPSECI) driver"
+	depends on FSL_MC_DPIO
+	select CRYPTO_DEV_FSL_CAAM_COMMON
+	select CRYPTO_AUTHENC
+	select CRYPTO_AEAD
 	help
-	  Selecting this will enable printing of various debug
+	  CAAM driver for QorIQ Data Path Acceleration Architecture 2.
-	  information in the CAAM driver.
+	  It handles DPSECI DPAA2 objects that sit on the Management Complex
+	  (MC) fsl-mc bus.
+
+	  To compile this as a module, choose M here: the module
+	  will be called dpaa2_caam.
 
 config CRYPTO_DEV_FSL_CAAM_CRYPTO_API_DESC
 	def_tristate (CRYPTO_DEV_FSL_CAAM_CRYPTO_API || \
-		      CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI)
+		      CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI || \
+		      CRYPTO_DEV_FSL_DPAA2_CAAM)

drivers/crypto/caam/Makefile

@@ -6,6 +6,9 @@ ifeq ($(CONFIG_CRYPTO_DEV_FSL_CAAM_DEBUG), y)
 	ccflags-y := -DDEBUG
 endif
 
+ccflags-y += -DVERSION=\"\"
+
+obj-$(CONFIG_CRYPTO_DEV_FSL_CAAM_COMMON) += error.o
 obj-$(CONFIG_CRYPTO_DEV_FSL_CAAM) += caam.o
 obj-$(CONFIG_CRYPTO_DEV_FSL_CAAM_JR) += caam_jr.o
 obj-$(CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API) += caamalg.o
@@ -16,9 +19,13 @@ obj-$(CONFIG_CRYPTO_DEV_FSL_CAAM_RNG_API) += caamrng.o
 obj-$(CONFIG_CRYPTO_DEV_FSL_CAAM_PKC_API) += caam_pkc.o
 
 caam-objs := ctrl.o
-caam_jr-objs := jr.o key_gen.o error.o
+caam_jr-objs := jr.o key_gen.o
 caam_pkc-y := caampkc.o pkc_desc.o
 ifneq ($(CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI),)
 	ccflags-y += -DCONFIG_CAAM_QI
 	caam-objs += qi.o
 endif
+
+obj-$(CONFIG_CRYPTO_DEV_FSL_DPAA2_CAAM) += dpaa2_caam.o
+
+dpaa2_caam-y    := caamalg_qi2.o dpseci.o

drivers/crypto/caam/caamalg_qi2.h

@@ -0,0 +1,189 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
+/*
+ * Copyright 2015-2016 Freescale Semiconductor Inc.
+ * Copyright 2017-2018 NXP
+ */
+
+#ifndef _CAAMALG_QI2_H_
+#define _CAAMALG_QI2_H_
+
+#include <soc/fsl/dpaa2-io.h>
+#include <soc/fsl/dpaa2-fd.h>
+#include <linux/threads.h>
+#include "dpseci.h"
+#include "desc_constr.h"
+
+#define DPAA2_CAAM_STORE_SIZE	16
+/* NAPI weight *must* be a multiple of the store size. */
+#define DPAA2_CAAM_NAPI_WEIGHT	64
+
+/* The congestion entrance threshold was chosen so that on LS2088
+ * we support the maximum throughput for the available memory
+ */
+#define DPAA2_SEC_CONG_ENTRY_THRESH	(128 * 1024 * 1024)
+#define DPAA2_SEC_CONG_EXIT_THRESH	(DPAA2_SEC_CONG_ENTRY_THRESH * 9 / 10)
+
+/**
+ * dpaa2_caam_priv - driver private data
+ * @dpseci_id: DPSECI object unique ID
+ * @major_ver: DPSECI major version
+ * @minor_ver: DPSECI minor version
+ * @dpseci_attr: DPSECI attributes
+ * @sec_attr: SEC engine attributes
+ * @rx_queue_attr: array of Rx queue attributes
+ * @tx_queue_attr: array of Tx queue attributes
+ * @cscn_mem: pointer to memory region containing the congestion SCN
+ *	it's size is larger than to accommodate alignment
+ * @cscn_mem_aligned: pointer to congestion SCN; it is computed as
+ *	PTR_ALIGN(cscn_mem, DPAA2_CSCN_ALIGN)
+ * @cscn_dma: dma address used by the QMAN to write CSCN messages
+ * @dev: device associated with the DPSECI object
+ * @mc_io: pointer to MC portal's I/O object
+ * @domain: IOMMU domain
+ * @ppriv: per CPU pointers to privata data
+ */
+struct dpaa2_caam_priv {
+	int dpsec_id;
+
+	u16 major_ver;
+	u16 minor_ver;
+
+	struct dpseci_attr dpseci_attr;
+	struct dpseci_sec_attr sec_attr;
+	struct dpseci_rx_queue_attr rx_queue_attr[DPSECI_MAX_QUEUE_NUM];
+	struct dpseci_tx_queue_attr tx_queue_attr[DPSECI_MAX_QUEUE_NUM];
+	int num_pairs;
+
+	/* congestion */
+	void *cscn_mem;
+	void *cscn_mem_aligned;
+	dma_addr_t cscn_dma;
+
+	struct device *dev;
+	struct fsl_mc_io *mc_io;
+	struct iommu_domain *domain;
+
+	struct dpaa2_caam_priv_per_cpu __percpu *ppriv;
+};
+
+/**
+ * dpaa2_caam_priv_per_cpu - per CPU private data
+ * @napi: napi structure
+ * @net_dev: netdev used by napi
+ * @req_fqid: (virtual) request (Tx / enqueue) FQID
+ * @rsp_fqid: (virtual) response (Rx / dequeue) FQID
+ * @prio: internal queue number - index for dpaa2_caam_priv.*_queue_attr
+ * @nctx: notification context of response FQ
+ * @store: where dequeued frames are stored
+ * @priv: backpointer to dpaa2_caam_priv
+ */
+struct dpaa2_caam_priv_per_cpu {
+	struct napi_struct napi;
+	struct net_device net_dev;
+	int req_fqid;
+	int rsp_fqid;
+	int prio;
+	struct dpaa2_io_notification_ctx nctx;
+	struct dpaa2_io_store *store;
+	struct dpaa2_caam_priv *priv;
+};
+
+/*
+ * The CAAM QI hardware constructs a job descriptor which points
+ * to shared descriptor (as pointed by context_a of FQ to CAAM).
+ * When the job descriptor is executed by deco, the whole job
+ * descriptor together with shared descriptor gets loaded in
+ * deco buffer which is 64 words long (each 32-bit).
+ *
+ * The job descriptor constructed by QI hardware has layout:
+ *
+ *	HEADER		(1 word)
+ *	Shdesc ptr	(1 or 2 words)
+ *	SEQ_OUT_PTR	(1 word)
+ *	Out ptr		(1 or 2 words)
+ *	Out length	(1 word)
+ *	SEQ_IN_PTR	(1 word)
+ *	In ptr		(1 or 2 words)
+ *	In length	(1 word)
+ *
+ * The shdesc ptr is used to fetch shared descriptor contents
+ * into deco buffer.
+ *
+ * Apart from shdesc contents, the total number of words that
+ * get loaded in deco buffer are '8' or '11'. The remaining words
+ * in deco buffer can be used for storing shared descriptor.
+ */
+#define MAX_SDLEN	((CAAM_DESC_BYTES_MAX - DESC_JOB_IO_LEN) / CAAM_CMD_SZ)
+
+/* Length of a single buffer in the QI driver memory cache */
+#define CAAM_QI_MEMCACHE_SIZE	512
+
+/*
+ * aead_edesc - s/w-extended aead descriptor
+ * @src_nents: number of segments in input scatterlist
+ * @dst_nents: number of segments in output scatterlist
+ * @iv_dma: dma address of iv for checking continuity and link table
+ * @qm_sg_bytes: length of dma mapped h/w link table
+ * @qm_sg_dma: bus physical mapped address of h/w link table
+ * @assoclen: associated data length, in CAAM endianness
+ * @assoclen_dma: bus physical mapped address of req->assoclen
+ * @sgt: the h/w link table, followed by IV
+ */
+struct aead_edesc {
+	int src_nents;
+	int dst_nents;
+	dma_addr_t iv_dma;
+	int qm_sg_bytes;
+	dma_addr_t qm_sg_dma;
+	unsigned int assoclen;
+	dma_addr_t assoclen_dma;
+	struct dpaa2_sg_entry sgt[0];
+};
+
+/**
+ * caam_flc - Flow Context (FLC)
+ * @flc: Flow Context options
+ * @sh_desc: Shared Descriptor
+ */
+struct caam_flc {
+	u32 flc[16];
+	u32 sh_desc[MAX_SDLEN];
+} ____cacheline_aligned;
+
+enum optype {
+	ENCRYPT = 0,
+	DECRYPT,
+	NUM_OP
+};
+
+/**
+ * caam_request - the request structure the driver application should fill while
+ *                submitting a job to driver.
+ * @fd_flt: Frame list table defining input and output
+ *          fd_flt[0] - FLE pointing to output buffer
+ *          fd_flt[1] - FLE pointing to input buffer
+ * @fd_flt_dma: DMA address for the frame list table
+ * @flc: Flow Context
+ * @flc_dma: I/O virtual address of Flow Context
+ * @cbk: Callback function to invoke when job is completed
+ * @ctx: arbit context attached with request by the application
+ * @edesc: extended descriptor; points to aead_edesc
+ */
+struct caam_request {
+	struct dpaa2_fl_entry fd_flt[2];
+	dma_addr_t fd_flt_dma;
+	struct caam_flc *flc;
+	dma_addr_t flc_dma;
+	void (*cbk)(void *ctx, u32 err);
+	void *ctx;
+	void *edesc;
+};
+
+/**
+ * dpaa2_caam_enqueue() - enqueue a crypto request
+ * @dev: device associated with the DPSECI object
+ * @req: pointer to caam_request
+ */
+int dpaa2_caam_enqueue(struct device *dev, struct caam_request *req);
+
+#endif	/* _CAAMALG_QI2_H_ */

drivers/crypto/caam/compat.h

@@ -17,6 +17,7 @@
 #include <linux/of_platform.h>
 #include <linux/dma-mapping.h>
 #include <linux/io.h>
+#include <linux/iommu.h>
 #include <linux/spinlock.h>
 #include <linux/rtnetlink.h>
 #include <linux/in.h>

drivers/crypto/caam/error.c

@@ -358,3 +358,7 @@ void caam_strstatus(struct device *jrdev, u32 status, bool qi_v2)
 		dev_err(jrdev, "%d: unknown error source\n", ssrc);
 }
 EXPORT_SYMBOL(caam_strstatus);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("FSL CAAM error reporting");
+MODULE_AUTHOR("Freescale Semiconductor");