crypto: ccp - Let a v5 CCP provide the same function as v3

Enable equivalent function on a v5 CCP. Add support for a
version 5 CCP which enables AES/XTS/SHA services. Also,
more work on the data structures to virtualize
functionality.
Signed-off-by: Gary R Hook <gary.hook@amd.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

drivers/crypto/ccp/Makefile

@@ -2,6 +2,7 @@ obj-$(CONFIG_CRYPTO_DEV_CCP_DD) += ccp.o
 ccp-objs := ccp-dev.o \
 	    ccp-ops.o \
 	    ccp-dev-v3.o \
+	    ccp-dev-v5.o \
 	    ccp-platform.o \
 	    ccp-dmaengine.o
 ccp-$(CONFIG_PCI) += ccp-pci.o

drivers/crypto/ccp/ccp-crypto-sha.c

@@ -4,6 +4,7 @@
  * Copyright (C) 2013,2016 Advanced Micro Devices, Inc.
  *
  * Author: Tom Lendacky <thomas.lendacky@amd.com>
+ * Author: Gary R Hook <gary.hook@amd.com>
  *
  * 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
@@ -134,7 +135,22 @@ static int ccp_do_sha_update(struct ahash_request *req, unsigned int nbytes,
 	rctx->cmd.engine = CCP_ENGINE_SHA;
 	rctx->cmd.u.sha.type = rctx->type;
 	rctx->cmd.u.sha.ctx = &rctx->ctx_sg;
-	rctx->cmd.u.sha.ctx_len = sizeof(rctx->ctx);
+
+	switch (rctx->type) {
+	case CCP_SHA_TYPE_1:
+		rctx->cmd.u.sha.ctx_len = SHA1_DIGEST_SIZE;
+		break;
+	case CCP_SHA_TYPE_224:
+		rctx->cmd.u.sha.ctx_len = SHA224_DIGEST_SIZE;
+		break;
+	case CCP_SHA_TYPE_256:
+		rctx->cmd.u.sha.ctx_len = SHA256_DIGEST_SIZE;
+		break;
+	default:
+		/* Should never get here */
+		break;
+	}
+
 	rctx->cmd.u.sha.src = sg;
 	rctx->cmd.u.sha.src_len = rctx->hash_cnt;
 	rctx->cmd.u.sha.opad = ctx->u.sha.key_len ?

drivers/crypto/ccp/ccp-dev-v3.c

@@ -405,6 +405,7 @@ static int ccp_init(struct ccp_device *ccp)
 	init_waitqueue_head(&ccp->sb_queue);
 	init_waitqueue_head(&ccp->suspend_queue);
 
+	dev_dbg(dev, "Starting threads...\n");
 	/* Create a kthread for each queue */
 	for (i = 0; i < ccp->cmd_q_count; i++) {
 		struct task_struct *kthread;
@@ -424,6 +425,13 @@ static int ccp_init(struct ccp_device *ccp)
 		wake_up_process(kthread);
 	}
 
+	dev_dbg(dev, "Enabling interrupts...\n");
+	/* Enable interrupts */
+	iowrite32(qim, ccp->io_regs + IRQ_MASK_REG);
+
+	dev_dbg(dev, "Registering device...\n");
+	ccp_add_device(ccp);
+
 	/* Register the RNG */
 	ccp->hwrng.name = ccp->rngname;
 	ccp->hwrng.read = ccp_trng_read;
@@ -438,11 +446,6 @@ static int ccp_init(struct ccp_device *ccp)
 	if (ret)
 		goto e_hwrng;
 
-	ccp_add_device(ccp);
-
-	/* Enable interrupts */
-	iowrite32(qim, ccp->io_regs + IRQ_MASK_REG);
-
 	return 0;
 
 e_hwrng:
@@ -468,7 +471,13 @@ static void ccp_destroy(struct ccp_device *ccp)
 	struct ccp_cmd *cmd;
 	unsigned int qim, i;
 
-	/* Remove this device from the list of available units first */
+	/* Unregister the DMA engine */
+	ccp_dmaengine_unregister(ccp);
+
+	/* Unregister the RNG */
+	hwrng_unregister(&ccp->hwrng);
+
+	/* Remove this device from the list of available units */
 	ccp_del_device(ccp);
 
 	/* Build queue interrupt mask (two interrupt masks per queue) */
@@ -488,12 +497,6 @@ static void ccp_destroy(struct ccp_device *ccp)
 	}
 	iowrite32(qim, ccp->io_regs + IRQ_STATUS_REG);
 
-	/* Unregister the DMA engine */
-	ccp_dmaengine_unregister(ccp);
-
-	/* Unregister the RNG */
-	hwrng_unregister(&ccp->hwrng);
-
 	/* Stop the queue kthreads */
 	for (i = 0; i < ccp->cmd_q_count; i++)
 		if (ccp->cmd_q[i].kthread)
@@ -570,6 +573,7 @@ static const struct ccp_actions ccp3_actions = {
 
 struct ccp_vdata ccpv3 = {
 	.version = CCP_VERSION(3, 0),
+	.setup = NULL,
 	.perform = &ccp3_actions,
 	.bar = 2,
 	.offset = 0x20000,

drivers/crypto/ccp/ccp-dev.h

@@ -61,7 +61,62 @@
 #define CMD_Q_ERROR(__qs)		((__qs) & 0x0000003f)
 #define CMD_Q_DEPTH(__qs)		(((__qs) >> 12) & 0x0000000f)
 
-/****** REQ0 Related Values ******/
+/* ------------------------ CCP Version 5 Specifics ------------------------ */
+#define CMD5_QUEUE_MASK_OFFSET		0x00
+#define CMD5_REQID_CONFIG_OFFSET	0x08
+#define LSB_PUBLIC_MASK_LO_OFFSET	0x18
+#define LSB_PUBLIC_MASK_HI_OFFSET	0x1C
+#define LSB_PRIVATE_MASK_LO_OFFSET	0x20
+#define LSB_PRIVATE_MASK_HI_OFFSET	0x24
+
+#define CMD5_Q_CONTROL_BASE		0x0000
+#define CMD5_Q_TAIL_LO_BASE		0x0004
+#define CMD5_Q_HEAD_LO_BASE		0x0008
+#define CMD5_Q_INT_ENABLE_BASE		0x000C
+#define CMD5_Q_INTERRUPT_STATUS_BASE	0x0010
+
+#define CMD5_Q_STATUS_BASE		0x0100
+#define CMD5_Q_INT_STATUS_BASE		0x0104
+#define CMD5_Q_DMA_STATUS_BASE		0x0108
+#define CMD5_Q_DMA_READ_STATUS_BASE	0x010C
+#define CMD5_Q_DMA_WRITE_STATUS_BASE	0x0110
+#define CMD5_Q_ABORT_BASE		0x0114
+#define CMD5_Q_AX_CACHE_BASE		0x0118
+
+/* Address offset between two virtual queue registers */
+#define CMD5_Q_STATUS_INCR		0x1000
+
+/* Bit masks */
+#define CMD5_Q_RUN			0x1
+#define CMD5_Q_HALT			0x2
+#define CMD5_Q_MEM_LOCATION		0x4
+#define CMD5_Q_SIZE			0x1F
+#define CMD5_Q_SHIFT			3
+#define COMMANDS_PER_QUEUE		16
+#define QUEUE_SIZE_VAL			((ffs(COMMANDS_PER_QUEUE) - 2) & \
+					  CMD5_Q_SIZE)
+#define Q_PTR_MASK			(2 << (QUEUE_SIZE_VAL + 5) - 1)
+#define Q_DESC_SIZE			sizeof(struct ccp5_desc)
+#define Q_SIZE(n)			(COMMANDS_PER_QUEUE*(n))
+
+#define INT_COMPLETION			0x1
+#define INT_ERROR			0x2
+#define INT_QUEUE_STOPPED		0x4
+#define ALL_INTERRUPTS			(INT_COMPLETION| \
+					 INT_ERROR| \
+					 INT_QUEUE_STOPPED)
+
+#define LSB_REGION_WIDTH		5
+#define MAX_LSB_CNT			8
+
+#define LSB_SIZE			16
+#define LSB_ITEM_SIZE			32
+#define PLSB_MAP_SIZE			(LSB_SIZE)
+#define SLSB_MAP_SIZE			(MAX_LSB_CNT * LSB_SIZE)
+
+#define LSB_ENTRY_NUMBER(LSB_ADDR)	(LSB_ADDR / LSB_ITEM_SIZE)
+
+/* ------------------------ CCP Version 3 Specifics ------------------------ */
 #define REQ0_WAIT_FOR_WRITE		0x00000004
 #define REQ0_INT_ON_COMPLETE		0x00000002
 #define REQ0_STOP_ON_COMPLETE		0x00000001
@@ -115,6 +170,8 @@
 
 #define CCP_JOBID_MASK			0x0000003f
 
+/* ------------------------ General CCP Defines ------------------------ */
+
 #define CCP_DMAPOOL_MAX_SIZE		64
 #define CCP_DMAPOOL_ALIGN		BIT(5)
 
@@ -149,6 +206,7 @@
 struct ccp_op;
 struct ccp_device;
 struct ccp_cmd;
+struct ccp_fns;
 
 struct ccp_dma_cmd {
 	struct list_head entry;
@@ -192,10 +250,30 @@ struct ccp_cmd_queue {
 	/* Queue dma pool */
 	struct dma_pool *dma_pool;
 
+	/* Queue base address (not neccessarily aligned)*/
+	struct ccp5_desc *qbase;
+
+	/* Aligned queue start address (per requirement) */
+	struct mutex q_mutex ____cacheline_aligned;
+	unsigned int qidx;
+
+	/* Version 5 has different requirements for queue memory */
+	unsigned int qsize;
+	dma_addr_t qbase_dma;
+	dma_addr_t qdma_tail;
+
 	/* Per-queue reserved storage block(s) */
 	u32 sb_key;
 	u32 sb_ctx;
 
+	/* Bitmap of LSBs that can be accessed by this queue */
+	DECLARE_BITMAP(lsbmask, MAX_LSB_CNT);
+	/* Private LSB that is assigned to this queue, or -1 if none.
+	 * Bitmap for my private LSB, unused otherwise
+	 */
+	unsigned int lsb;
+	DECLARE_BITMAP(lsbmap, PLSB_MAP_SIZE);
+
 	/* Queue processing thread */
 	struct task_struct *kthread;
 	unsigned int active;
@@ -209,8 +287,17 @@ struct ccp_cmd_queue {
 	u32 int_err;
 
 	/* Register addresses for queue */
+	void __iomem *reg_control;
+	void __iomem *reg_tail_lo;
+	void __iomem *reg_head_lo;
+	void __iomem *reg_int_enable;
+	void __iomem *reg_interrupt_status;
 	void __iomem *reg_status;
 	void __iomem *reg_int_status;
+	void __iomem *reg_dma_status;
+	void __iomem *reg_dma_read_status;
+	void __iomem *reg_dma_write_status;
+	u32 qcontrol; /* Cached control register */
 
 	/* Status values from job */
 	u32 int_status;
@@ -306,6 +393,9 @@ struct ccp_device {
 	unsigned int sb_count;
 	u32 sb_start;
 
+	/* Bitmap of shared LSBs, if any */
+	DECLARE_BITMAP(lsbmap, SLSB_MAP_SIZE);
+
 	/* Suspend support */
 	unsigned int suspending;
 	wait_queue_head_t suspend_queue;
@@ -320,6 +410,7 @@ enum ccp_memtype {
 	CCP_MEMTYPE_LOCAL,
 	CCP_MEMTYPE__LAST,
 };
+#define	CCP_MEMTYPE_LSB	CCP_MEMTYPE_KSB
 
 struct ccp_dma_info {
 	dma_addr_t address;
@@ -407,6 +498,7 @@ struct ccp_op {
 
 	struct ccp_mem src;
 	struct ccp_mem dst;
+	struct ccp_mem exp;
 
 	union {
 		struct ccp_aes_op aes;
@@ -416,6 +508,7 @@ struct ccp_op {
 		struct ccp_passthru_op passthru;
 		struct ccp_ecc_op ecc;
 	} u;
+	struct ccp_mem key;
 };
 
 static inline u32 ccp_addr_lo(struct ccp_dma_info *info)
@@ -428,6 +521,70 @@ static inline u32 ccp_addr_hi(struct ccp_dma_info *info)
 	return upper_32_bits(info->address + info->offset) & 0x0000ffff;
 }
 
+/**
+ * descriptor for version 5 CPP commands
+ * 8 32-bit words:
+ * word 0: function; engine; control bits
+ * word 1: length of source data
+ * word 2: low 32 bits of source pointer
+ * word 3: upper 16 bits of source pointer; source memory type
+ * word 4: low 32 bits of destination pointer
+ * word 5: upper 16 bits of destination pointer; destination memory type
+ * word 6: low 32 bits of key pointer
+ * word 7: upper 16 bits of key pointer; key memory type
+ */
+struct dword0 {
+	__le32 soc:1;
+	__le32 ioc:1;
+	__le32 rsvd1:1;
+	__le32 init:1;
+	__le32 eom:1;		/* AES/SHA only */
+	__le32 function:15;
+	__le32 engine:4;
+	__le32 prot:1;
+	__le32 rsvd2:7;
+};
+
+struct dword3 {
+	__le32 src_hi:16;
+	__le32 src_mem:2;
+	__le32 lsb_cxt_id:8;
+	__le32 rsvd1:5;
+	__le32 fixed:1;
+};
+
+union dword4 {
+	__le32 dst_lo;		/* NON-SHA	*/
+	__le32 sha_len_lo;	/* SHA		*/
+};
+
+union dword5 {
+	struct {
+		__le32 dst_hi:16;
+		__le32 dst_mem:2;
+		__le32 rsvd1:13;
+		__le32 fixed:1;
+	} fields;
+	__le32 sha_len_hi;
+};
+
+struct dword7 {
+	__le32 key_hi:16;
+	__le32 key_mem:2;
+	__le32 rsvd1:14;
+};
+
+struct ccp5_desc {
+	struct dword0 dw0;
+	__le32 length;
+	__le32 src_lo;
+	struct dword3 dw3;
+	union dword4 dw4;
+	union dword5 dw5;
+	__le32 key_lo;
+	struct dword7 dw7;
+};
+
 int ccp_pci_init(void);
 void ccp_pci_exit(void);
 
@@ -466,13 +623,14 @@ struct ccp_actions {
 
 /* Structure to hold CCP version-specific values */
 struct ccp_vdata {
-	unsigned int version;
-	int (*init)(struct ccp_device *);
+	const unsigned int version;
+	void (*setup)(struct ccp_device *);
 	const struct ccp_actions *perform;
 	const unsigned int bar;
 	const unsigned int offset;
 };
 
 extern	struct ccp_vdata ccpv3;
+extern	struct ccp_vdata ccpv5;
 
 #endif

drivers/crypto/ccp/ccp-pci.c

@@ -141,10 +141,11 @@ static void ccp_free_irqs(struct ccp_device *ccp)
 			free_irq(ccp_pci->msix[ccp_pci->msix_count].vector,
 				 dev);
 		pci_disable_msix(pdev);
-	} else {
+	} else if (ccp->irq) {
 		free_irq(ccp->irq, dev);
 		pci_disable_msi(pdev);
 	}
+	ccp->irq = 0;
 }
 
 static int ccp_find_mmio_area(struct ccp_device *ccp)
@@ -229,6 +230,8 @@ static int ccp_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	dev_set_drvdata(dev, ccp);
 
+	if (ccp->vdata->setup)
+		ccp->vdata->setup(ccp);
 	ret = ccp->vdata->perform->init(ccp);
 	if (ret)
 		goto e_iomap;
@@ -321,6 +324,7 @@ static int ccp_pci_resume(struct pci_dev *pdev)
 
 static const struct pci_device_id ccp_pci_table[] = {
 	{ PCI_VDEVICE(AMD, 0x1537), (kernel_ulong_t)&ccpv3 },
+	{ PCI_VDEVICE(AMD, 0x1456), (kernel_ulong_t)&ccpv5 },
 	/* Last entry must be zero */
 	{ 0, }
 };

include/linux/ccp.h

@@ -238,9 +238,6 @@ struct ccp_xts_aes_engine {
 };
 
 /***** SHA engine *****/
-#define CCP_SHA_BLOCKSIZE               SHA256_BLOCK_SIZE
-#define CCP_SHA_CTXSIZE                 SHA256_DIGEST_SIZE
-
 /**
  * ccp_sha_type - type of SHA operation
  *