Skip to contentSkip to conversation

MTD: Add Initial Hyperbus support

PATCH In review Vignesh Raghavendra 4 participants 19 messages Jun 25, 2019 – Jul 3, 2022

Subsystems (3)
lore
Share & tools
lore

Vignesh Raghavendra

author maintainer
#0 lore raw

v8 cover MTD: Add Initial Hyperbus support

Change log:
Since v7:
Fix up status register polling to support banked flashes in patch 1/5. Fix up comments by Sergei on documentation and style

Since v6:
Fix up DT bindings for TI HBMC driver to move hyperbus node out of syscon

Since v5:
Fix up DT bindings comments for TI HBMC driver Move calibration sequence out of core into TI HBMC driver

Since v4:
Fix Rob's comments on dt-bindings of TI HBMC driver

Since v3:
* Drop reading QRY string twice in hyperbus_calibrate()
* Fix doc/misc comments on v3.

Since RFC v2:
* use map_word_xxx() for handling status register to support interleaved flashes as suggested by Joakim Tjernlund <[email protected]>
* Report error status/messages on erase/program failure by looking at status register bits.
* Add "cfi-flash" as fallback compatible for cypress,hyperflash
* Add support to select between HyperBus and OSPI using mmio mux

Since RFC v1:
* Re-work Hyperbus core to provide separate struct representation for controller and slave devices
* Rename all files and func names to have hyperbus_ prefix
* Provide default calibration routine for use by controller drivers
* Fix up errors with patch spliting
* Address comments by Sergei Shtylyov <[email protected]> Cypress HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory(called HyperFlash) or self refresh DRAM(called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since Hyperbus is x8 DDR bus, its equivalent to x16 parallel NOR flash wrt bits per clk. But Hyperbus operates at >166MHz frequencies.
HyperRAM provides direct random read/write access to flash memory array.
Framework is modelled along the lines of spi-nor framework. HyperBus memory controller(HBMC) drivers call hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

This is an early RFC, to know if its okay to use maps framework and existing CFI compliant flash support code to support Hyperflash
Also would like input on different types of HBMC master IPs out there and their programming sequences.
Would appreciate any testing/review.

Tested on modified TI AM654 EVM with Cypress Hyperflash S26KS512 by creating a UBIFS partition and writing and reading files to it. Stress tested by writing/reading 16MB flash repeatedly at different offsets using dd commmand.

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]
TI's HBMC controller details at[3]

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

Vignesh Raghavendra (5):
mtd: cfi_cmdset_0002: Add support for polling status register
dt-bindings: mtd: Add binding documentation for HyperFlash
mtd: Add support for HyperBus memory devices
dt-bindings: mtd: Add bindings for TI's AM654 HyperBus memory
controller
mtd: hyperbus: Add driver for TI's HyperBus memory controller

 .../bindings/mtd/cypress,hyperflash.txt       |  13 ++
 .../devicetree/bindings/mtd/ti,am654-hbmc.txt |  51 ++++++
 MAINTAINERS                                   |   8 +
 drivers/mtd/Kconfig                           |   2 +
 drivers/mtd/Makefile                          |   1 +
 drivers/mtd/chips/cfi_cmdset_0002.c           | 130 +++++++++++++--
 drivers/mtd/hyperbus/Kconfig                  |  23 +++
 drivers/mtd/hyperbus/Makefile                 |   4 +
 drivers/mtd/hyperbus/hbmc-am654.c             | 141 ++++++++++++++++
 drivers/mtd/hyperbus/hyperbus-core.c          | 154 ++++++++++++++++++
 include/linux/mtd/cfi.h                       |   7 +
 include/linux/mtd/hyperbus.h                  |  86 ++++++++++
 12 files changed, 603 insertions(+), 17 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/mtd/cypress,hyperflash.txt
 create mode 100644 Documentation/devicetree/bindings/mtd/ti,am654-hbmc.txt
 create mode 100644 drivers/mtd/hyperbus/Kconfig
 create mode 100644 drivers/mtd/hyperbus/Makefile
 create mode 100644 drivers/mtd/hyperbus/hbmc-am654.c
 create mode 100644 drivers/mtd/hyperbus/hyperbus-core.c
 create mode 100644 include/linux/mtd/hyperbus.h
Signature · 1 line
2.22.0
author maintainer
#1 lore raw

PATCH v8 1/5 mtd: cfi_cmdset_0002: Add support for polling status register

HyperFlash devices are compliant with CFI AMD/Fujitsu Extended Command Set (0x0002) for flash operations, therefore drivers/mtd/chips/cfi_cmdset_0002.c can be used as is. But these devices do not support DQ polling method of determining chip ready/good status. These flashes provide Status Register whose bits can be polled to know status of flash operation.

Cypress HyperFlash datasheet here[1], talks about CFI Amd/Fujitsu Extended Query version 1.5. Bit 0 of "Software Features supported" field of CFI Primary Vendor-Specific Extended Query table indicates presence/absence of status register and Bit 1 indicates whether or not DQ polling is supported. Using these bits, its possible to determine whether flash supports DQ polling or need to use Status Register.

Add support for polling Status Register to know device ready/status of erase/write operations when DQ polling is not supported. Print error messages on erase/program failure by looking at related Status Register bits.

[1] https://www.cypress.com/file/213346/download

---
v8:
Fix up status register polling to support banked flashes in patch 1/5.

 drivers/mtd/chips/cfi_cmdset_0002.c | 130 ++++++++++++++++++++++++----
 include/linux/mtd/cfi.h             |   7 ++
 2 files changed, 120 insertions(+), 17 deletions(-)
drivers/mtd/chips/cfi_cmdset_0002.c +113 -17
index c8fa5906bdf9..09f8aaf1e763 100644
@@ -49,6 +49,16 @@
 #define SST49LF008A		0x005a
 #define AT49BV6416		0x00d6
 
+/*
+ * Status Register bit description. Used by flash devices that don't
+ * support DQ polling (e.g. HyperFlash)
+ */
+#define CFI_SR_DRB		BIT(7)
+#define CFI_SR_ESB		BIT(5)
+#define CFI_SR_PSB		BIT(4)
+#define CFI_SR_WBASB		BIT(3)
+#define CFI_SR_SLSB		BIT(1)
+
 static int cfi_amdstd_read (struct mtd_info *, loff_t, size_t, size_t *, u_char *);
 static int cfi_amdstd_write_words(struct mtd_info *, loff_t, size_t, size_t *, const u_char *);
 static int cfi_amdstd_write_buffers(struct mtd_info *, loff_t, size_t, size_t *, const u_char *);
@@ -97,6 +107,50 @@ static struct mtd_chip_driver cfi_amdstd_chipdrv = {
 	.module		= THIS_MODULE
 };
 
+/*
+ * Use status register to poll for Erase/write completion when DQ is not
+ * supported. This is indicated by Bit[1:0] of SoftwareFeatures field in
+ * CFI Primary Vendor-Specific Extended Query table 1.5
+ */
+static int cfi_use_status_reg(struct cfi_private *cfi)
+{
+	struct cfi_pri_amdstd *extp = cfi->cmdset_priv;
+	u8 poll_mask = CFI_POLL_STATUS_REG | CFI_POLL_DQ;
+
+	return extp->MinorVersion >= '5' &&
+		(extp->SoftwareFeatures & poll_mask) == CFI_POLL_STATUS_REG;
+}
+
+static void cfi_check_err_status(struct map_info *map, struct flchip *chip,
+				 unsigned long adr)
+{
+	struct cfi_private *cfi = map->fldrv_priv;
+	map_word status;
+
+	if (!cfi_use_status_reg(cfi))
+		return;
+
+	cfi_send_gen_cmd(0x70, cfi->addr_unlock1, chip->start, map, cfi,
+			 cfi->device_type, NULL);
+	status = map_read(map, adr);
+
+	if (map_word_bitsset(map, status, CMD(0x3a))) {
+		unsigned long chipstatus = MERGESTATUS(status);
+
+		if (chipstatus & CFI_SR_ESB)
+			pr_err("%s erase operation failed, status %lx\n",
+			       map->name, chipstatus);
+		if (chipstatus & CFI_SR_PSB)
+			pr_err("%s program operation failed, status %lx\n",
+			       map->name, chipstatus);
+		if (chipstatus & CFI_SR_WBASB)
+			pr_err("%s buffer program command aborted, status %lx\n",
+			       map->name, chipstatus);
+		if (chipstatus & CFI_SR_SLSB)
+			pr_err("%s sector write protected, status %lx\n",
+			       map->name, chipstatus);
+	}
+}
 
 /* #define DEBUG_CFI_FEATURES */
 
@@ -742,10 +796,25 @@ static struct mtd_info *cfi_amdstd_setup(struct mtd_info *mtd)
  * correctly and is therefore not done	(particularly with interleaved chips
  * as each chip must be checked independently of the others).
  */
-static int __xipram chip_ready(struct map_info *map, unsigned long addr)
+static int __xipram chip_ready(struct map_info *map, struct flchip *chip,
+			       unsigned long addr)
 {
+	struct cfi_private *cfi = map->fldrv_priv;
 	map_word d, t;
 
+	if (cfi_use_status_reg(cfi)) {
+		map_word ready = CMD(CFI_SR_DRB);
+		/*
+		 * For chips that support status register, check device
+		 * ready bit
+		 */
+		cfi_send_gen_cmd(0x70, cfi->addr_unlock1, chip->start, map, cfi,
+				 cfi->device_type, NULL);
+		d = map_read(map, addr);
+
+		return map_word_andequal(map, d, ready, ready);
+	}
+
 	d = map_read(map, addr);
 	t = map_read(map, addr);
 
@@ -767,10 +836,30 @@ static int __xipram chip_ready(struct map_info *map, unsigned long addr)
  * as each chip must be checked independently of the others).
  *
  */
-static int __xipram chip_good(struct map_info *map, unsigned long addr, map_word expected)
+static int __xipram chip_good(struct map_info *map, struct flchip *chip,
+			      unsigned long addr, map_word expected)
 {
+	struct cfi_private *cfi = map->fldrv_priv;
 	map_word oldd, curd;
 
+	if (cfi_use_status_reg(cfi)) {
+		map_word ready = CMD(CFI_SR_DRB);
+		map_word err = CMD(CFI_SR_PSB | CFI_SR_ESB);
+		/*
+		 * For chips that support status register, check device
+		 * ready bit and Erase/Program status bit to know if
+		 * operation succeeded.
+		 */
+		cfi_send_gen_cmd(0x70, cfi->addr_unlock1, chip->start, map, cfi,
+				 cfi->device_type, NULL);
+		curd = map_read(map, addr);
+
+		if (map_word_andequal(map, curd, ready, ready))
+			return !map_word_bitsset(map, curd, err);
+
+		return 0;
+	}
+
 	oldd = map_read(map, addr);
 	curd = map_read(map, addr);
 
@@ -792,7 +881,7 @@ static int get_chip(struct map_info *map, struct flchip *chip, unsigned long adr
 
 	case FL_STATUS:
 		for (;;) {
-			if (chip_ready(map, adr))
+			if (chip_ready(map, chip, adr))
 				break;
 
 			if (time_after(jiffies, timeo)) {
@@ -830,7 +919,7 @@ static int get_chip(struct map_info *map, struct flchip *chip, unsigned long adr
 		chip->state = FL_ERASE_SUSPENDING;
 		chip->erase_suspended = 1;
 		for (;;) {
-			if (chip_ready(map, adr))
+			if (chip_ready(map, chip, adr))
 				break;
 
 			if (time_after(jiffies, timeo)) {
@@ -1362,7 +1451,7 @@ static int do_otp_lock(struct map_info *map, struct flchip *chip, loff_t adr,
 	/* wait for chip to become ready */
 	timeo = jiffies + msecs_to_jiffies(2);
 	for (;;) {
-		if (chip_ready(map, adr))
+		if (chip_ready(map, chip, adr))
 			break;
 
 		if (time_after(jiffies, timeo)) {
@@ -1628,22 +1717,24 @@ static int __xipram do_write_oneword(struct map_info *map, struct flchip *chip,
 			continue;
 		}
 
-		if (time_after(jiffies, timeo) && !chip_ready(map, adr)){
+		if (time_after(jiffies, timeo) &&
+		    !chip_ready(map, chip, adr)) {
 			xip_enable(map, chip, adr);
 			printk(KERN_WARNING "MTD %s(): software timeout\n", __func__);
 			xip_disable(map, chip, adr);
 			break;
 		}
 
-		if (chip_ready(map, adr))
+		if (chip_ready(map, chip, adr))
 			break;
 
 		/* Latency issues. Drop the lock, wait a while and retry */
 		UDELAY(map, chip, adr, 1);
 	}
 	/* Did we succeed? */
-	if (!chip_good(map, adr, datum)) {
+	if (!chip_good(map, chip, adr, datum)) {
 		/* reset on all failures. */
+		cfi_check_err_status(map, chip, adr);
 		map_write(map, CMD(0xF0), chip->start);
 		/* FIXME - should have reset delay before continuing */
 
@@ -1881,10 +1972,11 @@ static int __xipram do_write_buffer(struct map_info *map, struct flchip *chip,
 		 * We check "time_after" and "!chip_good" before checking "chip_good" to avoid
 		 * the failure due to scheduling.
 		 */
-		if (time_after(jiffies, timeo) && !chip_good(map, adr, datum))
+		if (time_after(jiffies, timeo) &&
+		    !chip_good(map, chip, adr, datum))
 			break;
 
-		if (chip_good(map, adr, datum)) {
+		if (chip_good(map, chip, adr, datum)) {
 			xip_enable(map, chip, adr);
 			goto op_done;
 		}
@@ -1901,6 +1993,7 @@ static int __xipram do_write_buffer(struct map_info *map, struct flchip *chip,
 	 * See e.g.
 	 * http://www.spansion.com/Support/Application%20Notes/MirrorBit_Write_Buffer_Prog_Page_Buffer_Read_AN.pdf
 	 */
+	cfi_check_err_status(map, chip, adr);
 	cfi_send_gen_cmd(0xAA, cfi->addr_unlock1, chip->start, map, cfi,
 			 cfi->device_type, NULL);
 	cfi_send_gen_cmd(0x55, cfi->addr_unlock2, chip->start, map, cfi,
@@ -2018,7 +2111,7 @@ static int cfi_amdstd_panic_wait(struct map_info *map, struct flchip *chip,
 	 * If the driver thinks the chip is idle, and no toggle bits
 	 * are changing, then the chip is actually idle for sure.
 	 */
-	if (chip->state == FL_READY && chip_ready(map, adr))
+	if (chip->state == FL_READY && chip_ready(map, chip, adr))
 		return 0;
 
 	/*
@@ -2035,7 +2128,7 @@ static int cfi_amdstd_panic_wait(struct map_info *map, struct flchip *chip,
 
 		/* wait for the chip to become ready */
 		for (i = 0; i < jiffies_to_usecs(timeo); i++) {
-			if (chip_ready(map, adr))
+			if (chip_ready(map, chip, adr))
 				return 0;
 
 			udelay(1);
@@ -2099,14 +2192,15 @@ static int do_panic_write_oneword(struct map_info *map, struct flchip *chip,
 	map_write(map, datum, adr);
 
 	for (i = 0; i < jiffies_to_usecs(uWriteTimeout); i++) {
-		if (chip_ready(map, adr))
+		if (chip_ready(map, chip, adr))
 			break;
 
 		udelay(1);
 	}
 
-	if (!chip_good(map, adr, datum)) {
+	if (!chip_good(map, chip, adr, datum)) {
 		/* reset on all failures. */
+		cfi_check_err_status(map, chip, adr);
 		map_write(map, CMD(0xF0), chip->start);
 		/* FIXME - should have reset delay before continuing */
 
@@ -2300,7 +2394,7 @@ static int __xipram do_erase_chip(struct map_info *map, struct flchip *chip)
 			chip->erase_suspended = 0;
 		}
 
-		if (chip_good(map, adr, map_word_ff(map)))
+		if (chip_good(map, chip, adr, map_word_ff(map)))
 			break;
 
 		if (time_after(jiffies, timeo)) {
@@ -2316,6 +2410,7 @@ static int __xipram do_erase_chip(struct map_info *map, struct flchip *chip)
 	/* Did we succeed? */
 	if (ret) {
 		/* reset on all failures. */
+		cfi_check_err_status(map, chip, adr);
 		map_write(map, CMD(0xF0), chip->start);
 		/* FIXME - should have reset delay before continuing */
 
@@ -2396,7 +2491,7 @@ static int __xipram do_erase_oneblock(struct map_info *map, struct flchip *chip,
 			chip->erase_suspended = 0;
 		}
 
-		if (chip_good(map, adr, map_word_ff(map)))
+		if (chip_good(map, chip, adr, map_word_ff(map)))
 			break;
 
 		if (time_after(jiffies, timeo)) {
@@ -2412,6 +2507,7 @@ static int __xipram do_erase_oneblock(struct map_info *map, struct flchip *chip,
 	/* Did we succeed? */
 	if (ret) {
 		/* reset on all failures. */
+		cfi_check_err_status(map, chip, adr);
 		map_write(map, CMD(0xF0), chip->start);
 		/* FIXME - should have reset delay before continuing */
 
@@ -2589,7 +2685,7 @@ static int __maybe_unused do_ppb_xxlock(struct map_info *map,
 	 */
 	timeo = jiffies + msecs_to_jiffies(2000);	/* 2s max (un)locking */
 	for (;;) {
-		if (chip_ready(map, adr))
+		if (chip_ready(map, chip, adr))
 			break;
 
 		if (time_after(jiffies, timeo)) {
include/linux/mtd/cfi.h +7 -0
index 208c87cf2e3e..c98a21108688 100644
@@ -219,6 +219,13 @@ struct cfi_pri_amdstd {
 	uint8_t  VppMin;
 	uint8_t  VppMax;
 	uint8_t  TopBottom;
+	/* Below field are added from version 1.5 */
+	uint8_t  ProgramSuspend;
+	uint8_t  UnlockBypass;
+	uint8_t  SecureSiliconSector;
+	uint8_t  SoftwareFeatures;
+#define CFI_POLL_STATUS_REG	BIT(0)
+#define CFI_POLL_DQ		BIT(1)
 } __packed;
 
 /* Vendor-Specific PRI for Atmel chips (command set 0x0002) */
Signature · 1 line
2.22.0
Reviewed-by on v8 1/5
#2 lore raw

Re: [PATCH v8 1/5] mtd: cfi_cmdset_0002: Add support for polling status register

Hi,

Thanks for the fix.

I have just tested the patch quickly on my local environment that uses the cfi_cmdset_0002 flash device but not HyperFlash family. So tested as not affected by the change.

Quoted 61 lines · Vignesh Raghavendra

HyperFlash devices are compliant with CFI AMD/Fujitsu Extended Command Set (0x0002) for flash operations, therefore drivers/mtd/chips/cfi_cmdset_0002.c can be used as is. But these devices do not support DQ polling method of determining chip ready/good status. These flashes provide Status Register whose bits can be polled to know status of flash operation.

Cypress HyperFlash datasheet here[1], talks about CFI Amd/Fujitsu Extended Query version 1.5. Bit 0 of "Software Features supported" field of CFI Primary Vendor-Specific Extended Query table indicates presence/absence of status register and Bit 1 indicates whether or not DQ polling is supported. Using these bits, its possible to determine whether flash supports DQ polling or need to use Status Register.

Add support for polling Status Register to know device ready/status of erase/write operations when DQ polling is not supported. Print error messages on erase/program failure by looking at related Status Register bits.

[1] https://www.cypress.com/file/213346/download

---
v8:
Fix up status register polling to support banked flashes in patch 1/5.

  drivers/mtd/chips/cfi_cmdset_0002.c | 130 ++++++++++++++++++++++++----
  include/linux/mtd/cfi.h             |   7 ++
  2 files changed, 120 insertions(+), 17 deletions(-)
drivers/mtd/chips/cfi_cmdset_0002.c +15 -0
index c8fa5906bdf9..09f8aaf1e763 100644
@@ -49,6 +49,16 @@
  #define SST49LF008A		0x005a
  #define AT49BV6416		0x00d6
 
+/*
+ * Status Register bit description. Used by flash devices that don't
+ * support DQ polling (e.g. HyperFlash)
+ */
+#define CFI_SR_DRB		BIT(7)
+#define CFI_SR_ESB		BIT(5)
+#define CFI_SR_PSB		BIT(4)
+#define CFI_SR_WBASB		BIT(3)
+#define CFI_SR_SLSB		BIT(1)
+
  static int cfi_amdstd_read (struct mtd_info *, loff_t, size_t, size_t *, u_char *);
  static int cfi_amdstd_write_words(struct mtd_info *, loff_t, size_t, size_t *, const u_char *);
  static int cfi_amdstd_write_buffers(struct mtd_info *, loff_t, size_t, size_t *, const u_char *);
@@ -97,6 +107,50 @@ static struct mtd_chip_driver cfi_amdstd_chipdrv = {
  	.module		= THIS_MODULE
  };
 
+/*
+ * Use status register to poll for Erase/write completion when DQ is not
+ * supported. This is indicated by Bit[1:0] of SoftwareFeatures field in
+ * CFI Primary Vendor-Specific Extended Query table 1.5
+ */

drivers/mtd/chips/cfi_cmdset_0002.c static struct mtd_chip_driver cfi_amdstd_chipdrv = { · line 124

Quoted 10 lines
drivers/mtd/chips/cfi_cmdset_0002.c +10 -0
+static int cfi_use_status_reg(struct cfi_private *cfi)
+{
+	struct cfi_pri_amdstd *extp = cfi->cmdset_priv;
+	u8 poll_mask = CFI_POLL_STATUS_REG | CFI_POLL_DQ;
+
+	return extp->MinorVersion >= '5' &&
+		(extp->SoftwareFeatures & poll_mask) == CFI_POLL_STATUS_REG;
+}
+
+static void cfi_check_err_status(struct map_info *map, struct flchip *chip,

Just a minor comment.
I though that it is better to be called this function in chip_good() instead of to be called separately.
But I can understand that the chip_good() is called repeatedly in do_erase_chip(), do_erase_oneblock() and do_write_buffer() until timeout. Also do_write_oneword() is also possible to be changed to call the chip_good() repeatedly instead of chip_ready(). So additional logics is needed to avoid the repeated error messages in the function above.
Anyway it is okay for the current implementation to me.

Regards,
Ikegami

Quoted 265 lines
drivers/mtd/chips/cfi_cmdset_0002.c +88 -17
+				 unsigned long adr)
+{
+	struct cfi_private *cfi = map->fldrv_priv;
+	map_word status;
+
+	if (!cfi_use_status_reg(cfi))
+		return;
+
+	cfi_send_gen_cmd(0x70, cfi->addr_unlock1, chip->start, map, cfi,
+			 cfi->device_type, NULL);
+	status = map_read(map, adr);
+
+	if (map_word_bitsset(map, status, CMD(0x3a))) {
+		unsigned long chipstatus = MERGESTATUS(status);
+
+		if (chipstatus & CFI_SR_ESB)
+			pr_err("%s erase operation failed, status %lx\n",
+			       map->name, chipstatus);
+		if (chipstatus & CFI_SR_PSB)
+			pr_err("%s program operation failed, status %lx\n",
+			       map->name, chipstatus);
+		if (chipstatus & CFI_SR_WBASB)
+			pr_err("%s buffer program command aborted, status %lx\n",
+			       map->name, chipstatus);
+		if (chipstatus & CFI_SR_SLSB)
+			pr_err("%s sector write protected, status %lx\n",
+			       map->name, chipstatus);
+	}
+}
 
  /* #define DEBUG_CFI_FEATURES */
 
@@ -742,10 +796,25 @@ static struct mtd_info *cfi_amdstd_setup(struct mtd_info *mtd)
   * correctly and is therefore not done	(particularly with interleaved chips
   * as each chip must be checked independently of the others).
   */
-static int __xipram chip_ready(struct map_info *map, unsigned long addr)
+static int __xipram chip_ready(struct map_info *map, struct flchip *chip,
+			       unsigned long addr)
  {
+	struct cfi_private *cfi = map->fldrv_priv;
  	map_word d, t;
 
+	if (cfi_use_status_reg(cfi)) {
+		map_word ready = CMD(CFI_SR_DRB);
+		/*
+		 * For chips that support status register, check device
+		 * ready bit
+		 */
+		cfi_send_gen_cmd(0x70, cfi->addr_unlock1, chip->start, map, cfi,
+				 cfi->device_type, NULL);
+		d = map_read(map, addr);
+
+		return map_word_andequal(map, d, ready, ready);
+	}
+
  	d = map_read(map, addr);
  	t = map_read(map, addr);
 
@@ -767,10 +836,30 @@ static int __xipram chip_ready(struct map_info *map, unsigned long addr)
   * as each chip must be checked independently of the others).
   *
   */
-static int __xipram chip_good(struct map_info *map, unsigned long addr, map_word expected)
+static int __xipram chip_good(struct map_info *map, struct flchip *chip,
+			      unsigned long addr, map_word expected)
  {
+	struct cfi_private *cfi = map->fldrv_priv;
  	map_word oldd, curd;
 
+	if (cfi_use_status_reg(cfi)) {
+		map_word ready = CMD(CFI_SR_DRB);
+		map_word err = CMD(CFI_SR_PSB | CFI_SR_ESB);
+		/*
+		 * For chips that support status register, check device
+		 * ready bit and Erase/Program status bit to know if
+		 * operation succeeded.
+		 */
+		cfi_send_gen_cmd(0x70, cfi->addr_unlock1, chip->start, map, cfi,
+				 cfi->device_type, NULL);
+		curd = map_read(map, addr);
+
+		if (map_word_andequal(map, curd, ready, ready))
+			return !map_word_bitsset(map, curd, err);
+
+		return 0;
+	}
+
  	oldd = map_read(map, addr);
  	curd = map_read(map, addr);
 
@@ -792,7 +881,7 @@ static int get_chip(struct map_info *map, struct flchip *chip, unsigned long adr
 
  	case FL_STATUS:
  		for (;;) {
-			if (chip_ready(map, adr))
+			if (chip_ready(map, chip, adr))
  				break;
 
  			if (time_after(jiffies, timeo)) {
@@ -830,7 +919,7 @@ static int get_chip(struct map_info *map, struct flchip *chip, unsigned long adr
  		chip->state = FL_ERASE_SUSPENDING;
  		chip->erase_suspended = 1;
  		for (;;) {
-			if (chip_ready(map, adr))
+			if (chip_ready(map, chip, adr))
  				break;
 
  			if (time_after(jiffies, timeo)) {
@@ -1362,7 +1451,7 @@ static int do_otp_lock(struct map_info *map, struct flchip *chip, loff_t adr,
  	/* wait for chip to become ready */
  	timeo = jiffies + msecs_to_jiffies(2);
  	for (;;) {
-		if (chip_ready(map, adr))
+		if (chip_ready(map, chip, adr))
  			break;
 
  		if (time_after(jiffies, timeo)) {
@@ -1628,22 +1717,24 @@ static int __xipram do_write_oneword(struct map_info *map, struct flchip *chip,
  			continue;
  		}
 
-		if (time_after(jiffies, timeo) && !chip_ready(map, adr)){
+		if (time_after(jiffies, timeo) &&
+		    !chip_ready(map, chip, adr)) {
  			xip_enable(map, chip, adr);
  			printk(KERN_WARNING "MTD %s(): software timeout\n", __func__);
  			xip_disable(map, chip, adr);
  			break;
  		}
 
-		if (chip_ready(map, adr))
+		if (chip_ready(map, chip, adr))
  			break;
 
  		/* Latency issues. Drop the lock, wait a while and retry */
  		UDELAY(map, chip, adr, 1);
  	}
  	/* Did we succeed? */
-	if (!chip_good(map, adr, datum)) {
+	if (!chip_good(map, chip, adr, datum)) {
  		/* reset on all failures. */
+		cfi_check_err_status(map, chip, adr);
  		map_write(map, CMD(0xF0), chip->start);
  		/* FIXME - should have reset delay before continuing */
 
@@ -1881,10 +1972,11 @@ static int __xipram do_write_buffer(struct map_info *map, struct flchip *chip,
  		 * We check "time_after" and "!chip_good" before checking "chip_good" to avoid
  		 * the failure due to scheduling.
  		 */
-		if (time_after(jiffies, timeo) && !chip_good(map, adr, datum))
+		if (time_after(jiffies, timeo) &&
+		    !chip_good(map, chip, adr, datum))
  			break;
 
-		if (chip_good(map, adr, datum)) {
+		if (chip_good(map, chip, adr, datum)) {
  			xip_enable(map, chip, adr);
  			goto op_done;
  		}
@@ -1901,6 +1993,7 @@ static int __xipram do_write_buffer(struct map_info *map, struct flchip *chip,
  	 * See e.g.
  	 * http://www.spansion.com/Support/Application%20Notes/MirrorBit_Write_Buffer_Prog_Page_Buffer_Read_AN.pdf
  	 */
+	cfi_check_err_status(map, chip, adr);
  	cfi_send_gen_cmd(0xAA, cfi->addr_unlock1, chip->start, map, cfi,
  			 cfi->device_type, NULL);
  	cfi_send_gen_cmd(0x55, cfi->addr_unlock2, chip->start, map, cfi,
@@ -2018,7 +2111,7 @@ static int cfi_amdstd_panic_wait(struct map_info *map, struct flchip *chip,
  	 * If the driver thinks the chip is idle, and no toggle bits
  	 * are changing, then the chip is actually idle for sure.
  	 */
-	if (chip->state == FL_READY && chip_ready(map, adr))
+	if (chip->state == FL_READY && chip_ready(map, chip, adr))
  		return 0;
 
  	/*
@@ -2035,7 +2128,7 @@ static int cfi_amdstd_panic_wait(struct map_info *map, struct flchip *chip,
 
  		/* wait for the chip to become ready */
  		for (i = 0; i < jiffies_to_usecs(timeo); i++) {
-			if (chip_ready(map, adr))
+			if (chip_ready(map, chip, adr))
  				return 0;
 
  			udelay(1);
@@ -2099,14 +2192,15 @@ static int do_panic_write_oneword(struct map_info *map, struct flchip *chip,
  	map_write(map, datum, adr);
 
  	for (i = 0; i < jiffies_to_usecs(uWriteTimeout); i++) {
-		if (chip_ready(map, adr))
+		if (chip_ready(map, chip, adr))
  			break;
 
  		udelay(1);
  	}
 
-	if (!chip_good(map, adr, datum)) {
+	if (!chip_good(map, chip, adr, datum)) {
  		/* reset on all failures. */
+		cfi_check_err_status(map, chip, adr);
  		map_write(map, CMD(0xF0), chip->start);
  		/* FIXME - should have reset delay before continuing */
 
@@ -2300,7 +2394,7 @@ static int __xipram do_erase_chip(struct map_info *map, struct flchip *chip)
  			chip->erase_suspended = 0;
  		}
 
-		if (chip_good(map, adr, map_word_ff(map)))
+		if (chip_good(map, chip, adr, map_word_ff(map)))
  			break;
 
  		if (time_after(jiffies, timeo)) {
@@ -2316,6 +2410,7 @@ static int __xipram do_erase_chip(struct map_info *map, struct flchip *chip)
  	/* Did we succeed? */
  	if (ret) {
  		/* reset on all failures. */
+		cfi_check_err_status(map, chip, adr);
  		map_write(map, CMD(0xF0), chip->start);
  		/* FIXME - should have reset delay before continuing */
 
@@ -2396,7 +2491,7 @@ static int __xipram do_erase_oneblock(struct map_info *map, struct flchip *chip,
  			chip->erase_suspended = 0;
  		}
 
-		if (chip_good(map, adr, map_word_ff(map)))
+		if (chip_good(map, chip, adr, map_word_ff(map)))
  			break;
 
  		if (time_after(jiffies, timeo)) {
@@ -2412,6 +2507,7 @@ static int __xipram do_erase_oneblock(struct map_info *map, struct flchip *chip,
  	/* Did we succeed? */
  	if (ret) {
  		/* reset on all failures. */
+		cfi_check_err_status(map, chip, adr);
  		map_write(map, CMD(0xF0), chip->start);
  		/* FIXME - should have reset delay before continuing */
 
@@ -2589,7 +2685,7 @@ static int __maybe_unused do_ppb_xxlock(struct map_info *map,
  	 */
  	timeo = jiffies + msecs_to_jiffies(2000);	/* 2s max (un)locking */
  	for (;;) {
-		if (chip_ready(map, adr))
+		if (chip_ready(map, chip, adr))
  			break;
 
  		if (time_after(jiffies, timeo)) {
include/linux/mtd/cfi.h +7 -0
index 208c87cf2e3e..c98a21108688 100644
@@ -219,6 +219,13 @@ struct cfi_pri_amdstd {
  	uint8_t  VppMin;
  	uint8_t  VppMax;
  	uint8_t  TopBottom;
+	/* Below field are added from version 1.5 */
+	uint8_t  ProgramSuspend;
+	uint8_t  UnlockBypass;
+	uint8_t  SecureSiliconSector;
+	uint8_t  SoftwareFeatures;
+#define CFI_POLL_STATUS_REG	BIT(0)
+#define CFI_POLL_DQ		BIT(1)
  } __packed;
 
  /* Vendor-Specific PRI for Atmel chips (command set 0x0002) */
author maintainer
#3 lore raw

Re: [PATCH v8 1/5] mtd: cfi_cmdset_0002: Add support for polling status register

Quoted 10 lines · Tokunori Ikegami

Hi,

Thanks for the fix.

I have just tested the patch quickly on my local environment that uses the cfi_cmdset_0002 flash device but not HyperFlash family. So tested as not affected by the change.

Thanks for testing!

Quoted 78 lines · Vignesh Raghavendra
Quoted 73 lines

HyperFlash devices are compliant with CFI AMD/Fujitsu Extended Command Set (0x0002) for flash operations, therefore drivers/mtd/chips/cfi_cmdset_0002.c can be used as is. But these devices do not support DQ polling method of determining chip ready/good status. These flashes provide Status Register whose bits can be polled to know status of flash operation.

Cypress HyperFlash datasheet here[1], talks about CFI Amd/Fujitsu Extended Query version 1.5. Bit 0 of "Software Features supported" field of CFI Primary Vendor-Specific Extended Query table indicates presence/absence of status register and Bit 1 indicates whether or not DQ polling is supported. Using these bits, its possible to determine whether flash supports DQ polling or need to use Status Register.

Add support for polling Status Register to know device ready/status of erase/write operations when DQ polling is not supported. Print error messages on erase/program failure by looking at related Status Register bits.

[1] https://www.cypress.com/file/213346/download

---
v8:
Fix up status register polling to support banked flashes in patch 1/5.

  drivers/mtd/chips/cfi_cmdset_0002.c | 130 ++++++++++++++++++++++++---- include/linux/mtd/cfi.h             |   7 ++

  2 files changed, 120 insertions(+), 17 deletions(-)
drivers/mtd/chips/cfi_cmdset_0002.c +0 -0
index c8fa5906bdf9..09f8aaf1e763 100644
@@ -49,6 +49,16 @@
   #define SST49LF008A        0x005a
   #define AT49BV6416        0x00d6

  +/*
+ * Status Register bit description. Used by flash devices that don't
+ * support DQ polling (e.g. HyperFlash)
+ */

drivers/mtd/chips/cfi_cmdset_0002.c +6 -0
+#define CFI_SR_DRB        BIT(7)
+#define CFI_SR_ESB        BIT(5)
+#define CFI_SR_PSB        BIT(4)
+#define CFI_SR_WBASB        BIT(3)
+#define CFI_SR_SLSB        BIT(1)
+

  static int cfi_amdstd_read (struct mtd_info *, loff_t, size_t, size_t *, u_char *);
  static int cfi_amdstd_write_words(struct mtd_info *, loff_t, size_t, size_t *, const u_char *);
  static int cfi_amdstd_write_buffers(struct mtd_info *, loff_t, size_t, size_t *, const u_char *);

drivers/mtd/chips/cfi_cmdset_0002.c +0 -0
@@ -97,6 +107,50 @@ static struct mtd_chip_driver cfi_amdstd_chipdrv = {

      .module        = THIS_MODULE
  };
  +/*
+ * Use status register to poll for Erase/write completion when DQ is not
+ * supported. This is indicated by Bit[1:0] of SoftwareFeatures field in
+ * CFI Primary Vendor-Specific Extended Query table 1.5
+ */

drivers/mtd/chips/cfi_cmdset_0002.c +10 -0
+static int cfi_use_status_reg(struct cfi_private *cfi)
+{
+    struct cfi_pri_amdstd *extp = cfi->cmdset_priv;
+    u8 poll_mask = CFI_POLL_STATUS_REG | CFI_POLL_DQ;
+
+    return extp->MinorVersion >= '5' &&
+        (extp->SoftwareFeatures & poll_mask) == CFI_POLL_STATUS_REG;
+}
+
+static void cfi_check_err_status(struct map_info *map, struct flchip
Quoted 1 line

*chip,

Just a minor comment.
I though that it is better to be called this function in chip_good()

Quoted 9 lines

instead of to be called separately.
But I can understand that the chip_good() is called repeatedly in do_erase_chip(), do_erase_oneblock() and do_write_buffer() until timeout. Also do_write_oneword() is also possible to be changed to call the chip_good() repeatedly instead of chip_ready().
So additional logics is needed to avoid the repeated error messages in the function above.
Anyway it is okay for the current implementation to me.

Will do that as an incremental change once base support for HyperFlash is in.

Regards
Vignesh

Quoted 262 lines

Regards,
Ikegami

Quoted 259 lines

+                 unsigned long adr)

drivers/mtd/chips/cfi_cmdset_0002.c +28 -0
+{
+    struct cfi_private *cfi = map->fldrv_priv;
+    map_word status;
+
+    if (!cfi_use_status_reg(cfi))
+        return;
+
+    cfi_send_gen_cmd(0x70, cfi->addr_unlock1, chip->start, map, cfi,
+             cfi->device_type, NULL);
+    status = map_read(map, adr);
+
+    if (map_word_bitsset(map, status, CMD(0x3a))) {
+        unsigned long chipstatus = MERGESTATUS(status);
+
+        if (chipstatus & CFI_SR_ESB)
+            pr_err("%s erase operation failed, status %lx\n",
+                   map->name, chipstatus);
+        if (chipstatus & CFI_SR_PSB)
+            pr_err("%s program operation failed, status %lx\n",
+                   map->name, chipstatus);
+        if (chipstatus & CFI_SR_WBASB)
+            pr_err("%s buffer program command aborted, status %lx\n",
+                   map->name, chipstatus);
+        if (chipstatus & CFI_SR_SLSB)
+            pr_err("%s sector write protected, status %lx\n",
+                   map->name, chipstatus);
+    }
+}

    /* #define DEBUG_CFI_FEATURES */
  @@ -742,10 +796,25 @@ static struct mtd_info *cfi_amdstd_setup(struct mtd_info *mtd)
   * correctly and is therefore not done    (particularly with interleaved chips
   * as each chip must be checked independently of the others). */

drivers/mtd/chips/cfi_cmdset_0002.c +1 -1
-static int __xipram chip_ready(struct map_info *map, unsigned long addr)
+static int __xipram chip_ready(struct map_info *map, struct flchip

*chip,
+                   unsigned long addr)
  {
+    struct cfi_private *cfi = map->fldrv_priv;
      map_word d, t;
  +    if (cfi_use_status_reg(cfi)) {
+        map_word ready = CMD(CFI_SR_DRB);
+        /*
+         * For chips that support status register, check device +         * ready bit
+         */
+        cfi_send_gen_cmd(0x70, cfi->addr_unlock1, chip->start, map, cfi, +                 cfi->device_type, NULL);
+        d = map_read(map, addr);

drivers/mtd/chips/cfi_cmdset_0002.c +4 -0
+
+        return map_word_andequal(map, d, ready, ready);
+    }
+

      d = map_read(map, addr);
      t = map_read(map, addr);
  @@ -767,10 +836,30 @@ static int __xipram chip_ready(struct map_info *map, unsigned long addr)
   * as each chip must be checked independently of the others). *
   */
-static int __xipram chip_good(struct map_info *map, unsigned long addr, map_word expected)
+static int __xipram chip_good(struct map_info *map, struct flchip *chip, +                  unsigned long addr, map_word expected)
  {
+    struct cfi_private *cfi = map->fldrv_priv;
      map_word oldd, curd;
  +    if (cfi_use_status_reg(cfi)) {
+        map_word ready = CMD(CFI_SR_DRB);
+        map_word err = CMD(CFI_SR_PSB | CFI_SR_ESB);
+        /*
+         * For chips that support status register, check device +         * ready bit and Erase/Program status bit to know if +         * operation succeeded.
+         */
+        cfi_send_gen_cmd(0x70, cfi->addr_unlock1, chip->start, map, cfi, +                 cfi->device_type, NULL);
+        curd = map_read(map, addr);

drivers/mtd/chips/cfi_cmdset_0002.c +7 -0
+
+        if (map_word_andequal(map, curd, ready, ready))
+            return !map_word_bitsset(map, curd, err);
+
+        return 0;
+    }
+

      oldd = map_read(map, addr);
      curd = map_read(map, addr);
  @@ -792,7 +881,7 @@ static int get_chip(struct map_info *map, struct flchip *chip, unsigned long adr
        case FL_STATUS:
          for (;;) {
-            if (chip_ready(map, adr))
+            if (chip_ready(map, chip, adr))
                  break;
                if (time_after(jiffies, timeo)) {

drivers/mtd/chips/cfi_cmdset_0002.c +0 -0
@@ -830,7 +919,7 @@ static int get_chip(struct map_info *map, struct

flchip *chip, unsigned long adr
          chip->state = FL_ERASE_SUSPENDING;
          chip->erase_suspended = 1;
          for (;;) {
-            if (chip_ready(map, adr))
+            if (chip_ready(map, chip, adr))
                  break;
                if (time_after(jiffies, timeo)) {

drivers/mtd/chips/cfi_cmdset_0002.c +6 -2
@@ -1362,7 +1451,7 @@ static int do_otp_lock(struct map_info *map,
 struct flchip *chip, loff_t adr,
       /* wait for chip to become ready */
       timeo = jiffies + msecs_to_jiffies(2);
       for (;;) {
-        if (chip_ready(map, adr))
+        if (chip_ready(map, chip, adr))
               break;
             if (time_after(jiffies, timeo)) {
@@ -1628,22 +1717,24 @@ static int __xipram do_write_oneword(struct
 map_info *map, struct flchip *chip,
               continue;
           }
   -        if (time_after(jiffies, timeo) && !chip_ready(map, adr)){
+        if (time_after(jiffies, timeo) &&
+            !chip_ready(map, chip, adr)) {
               xip_enable(map, chip, adr);
               printk(KERN_WARNING "MTD %s(): software timeout\n",
 __func__);
               xip_disable(map, chip, adr);
               break;
           }
   -        if (chip_ready(map, adr))
+        if (chip_ready(map, chip, adr))
               break;
             /* Latency issues. Drop the lock, wait a while and retry */
           UDELAY(map, chip, adr, 1);
       }
       /* Did we succeed? */
-    if (!chip_good(map, adr, datum)) {
+    if (!chip_good(map, chip, adr, datum)) {
           /* reset on all failures. */
+        cfi_check_err_status(map, chip, adr);
           map_write(map, CMD(0xF0), chip->start);
           /* FIXME - should have reset delay before continuing */

  @@ -1881,10 +1972,11 @@ static int __xipram do_write_buffer(struct map_info *map, struct flchip *chip,
           * We check "time_after" and "!chip_good" before checking "chip_good" to avoid
           * the failure due to scheduling.
           */
-        if (time_after(jiffies, timeo) && !chip_good(map, adr, datum)) +        if (time_after(jiffies, timeo) &&
+            !chip_good(map, chip, adr, datum))
              break;
  -        if (chip_good(map, adr, datum)) {
+        if (chip_good(map, chip, adr, datum)) { xip_enable(map, chip, adr);
              goto op_done;
          }

drivers/mtd/chips/cfi_cmdset_0002.c +0 -0
@@ -1901,6 +1993,7 @@ static int __xipram do_write_buffer(struct
 map_info *map, struct flchip *chip,
        * See e.g.
        *

http://www.spansion.com/Support/Application%20Notes/MirrorBit_Write_Buffer_Prog_Page_Buffer_Read_AN.pdf

       */
+    cfi_check_err_status(map, chip, adr); cfi_send_gen_cmd(0xAA, cfi->addr_unlock1, chip->start, map, cfi, cfi->device_type, NULL); cfi_send_gen_cmd(0x55, cfi->addr_unlock2, chip->start, map, cfi,

drivers/mtd/chips/cfi_cmdset_0002.c +5 -3
@@ -2018,7 +2111,7 @@ static int cfi_amdstd_panic_wait(struct map_info
 *map, struct flchip *chip,
        * If the driver thinks the chip is idle, and no toggle bits
        * are changing, then the chip is actually idle for sure.
        */
-    if (chip->state == FL_READY && chip_ready(map, adr))
+    if (chip->state == FL_READY && chip_ready(map, chip, adr))
           return 0;
         /*
@@ -2035,7 +2128,7 @@ static int cfi_amdstd_panic_wait(struct map_info
 *map, struct flchip *chip,
             /* wait for the chip to become ready */
           for (i = 0; i < jiffies_to_usecs(timeo); i++) {
-            if (chip_ready(map, adr))
+            if (chip_ready(map, chip, adr))
                   return 0;
                 udelay(1);
@@ -2099,14 +2192,15 @@ static int do_panic_write_oneword(struct
 map_info *map, struct flchip *chip,
       map_write(map, datum, adr);
         for (i = 0; i < jiffies_to_usecs(uWriteTimeout); i++) {
-        if (chip_ready(map, adr))
+        if (chip_ready(map, chip, adr))
               break;
             udelay(1);
       }
   -    if (!chip_good(map, adr, datum)) {
+    if (!chip_good(map, chip, adr, datum)) {
           /* reset on all failures. */
+        cfi_check_err_status(map, chip, adr);
           map_write(map, CMD(0xF0), chip->start);
           /* FIXME - should have reset delay before continuing */

  @@ -2300,7 +2394,7 @@ static int __xipram do_erase_chip(struct map_info *map, struct flchip *chip) chip->erase_suspended = 0;
          }
  -        if (chip_good(map, adr, map_word_ff(map))) +        if (chip_good(map, chip, adr, map_word_ff(map))) break;
            if (time_after(jiffies, timeo)) {

drivers/mtd/chips/cfi_cmdset_0002.c +1 -0
@@ -2316,6 +2410,7 @@ static int __xipram do_erase_chip(struct
 map_info *map, struct flchip *chip)
       /* Did we succeed? */
       if (ret) {
           /* reset on all failures. */
+        cfi_check_err_status(map, chip, adr);
           map_write(map, CMD(0xF0), chip->start);
           /* FIXME - should have reset delay before continuing */

  @@ -2396,7 +2491,7 @@ static int __xipram do_erase_oneblock(struct map_info *map, struct flchip *chip,
              chip->erase_suspended = 0;
          }
  -        if (chip_good(map, adr, map_word_ff(map)))
+        if (chip_good(map, chip, adr, map_word_ff(map))) break;
            if (time_after(jiffies, timeo)) {

drivers/mtd/chips/cfi_cmdset_0002.c +1 -0
@@ -2412,6 +2507,7 @@ static int __xipram do_erase_oneblock(struct
 map_info *map, struct flchip *chip,
       /* Did we succeed? */
       if (ret) {
           /* reset on all failures. */
+        cfi_check_err_status(map, chip, adr);
           map_write(map, CMD(0xF0), chip->start);
           /* FIXME - should have reset delay before continuing */

  @@ -2589,7 +2685,7 @@ static int __maybe_unused do_ppb_xxlock(struct map_info *map,
       */
      timeo = jiffies + msecs_to_jiffies(2000);    /* 2s max (un)locking */
      for (;;) {
-        if (chip_ready(map, adr))
+        if (chip_ready(map, chip, adr))
              break;
            if (time_after(jiffies, timeo)) {

include/linux/mtd/cfi.h +7 -0
index 208c87cf2e3e..c98a21108688 100644
@@ -219,6 +219,13 @@ struct cfi_pri_amdstd {
       uint8_t  VppMin;
       uint8_t  VppMax;
       uint8_t  TopBottom;
+    /* Below field are added from version 1.5 */
+    uint8_t  ProgramSuspend;
+    uint8_t  UnlockBypass;
+    uint8_t  SecureSiliconSector;
+    uint8_t  SoftwareFeatures;
+#define CFI_POLL_STATUS_REG    BIT(0)
+#define CFI_POLL_DQ        BIT(1)
   } __packed;
     /* Vendor-Specific PRI for Atmel chips (command set 0x0002) */
Signature · 6 lines
Regards
Vignesh

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
author maintainer
#4 lore raw

PATCH v8 2/5 dt-bindings: mtd: Add binding documentation for HyperFlash

In reply to Vignesh Raghavendra

Add DT binding documentation for HyperFlash devices.

---
v8:
No changes

 .../devicetree/bindings/mtd/cypress,hyperflash.txt  | 13 +++++++++++++
 1 file changed, 13 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/mtd/cypress,hyperflash.txt
Documentation/devicetree/bindings/mtd/cypress,hyperflash.txt new file +13 -0
new file mode 100644
index 000000000000..ad42f4db32f1
@@ -0,0 +1,13 @@
+Bindings for HyperFlash NOR flash chips compliant with Cypress HyperBus
+specification and supports Cypress CFI specification 1.5 command set.
+
+Required properties:
+- compatible : "cypress,hyperflash", "cfi-flash" for HyperFlash NOR chips
+- reg : Address of flash's memory map
+
+Example:
+
+	flash@0 {
+		compatible = "cypress,hyperflash", "cfi-flash";
+		reg = <0x0 0x4000000>;
+	};
Signature · 1 line
2.22.0
author maintainer
#5 lore raw

PATCH v8 3/5 mtd: Add support for HyperBus memory devices

In reply to Vignesh Raghavendra

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

---
v8:
Fix documentation comments on core APIs

 MAINTAINERS                          |   7 ++
 drivers/mtd/Kconfig                  |   2 +
 drivers/mtd/Makefile                 |   1 +
 drivers/mtd/hyperbus/Kconfig         |  11 ++
 drivers/mtd/hyperbus/Makefile        |   3 +
 drivers/mtd/hyperbus/hyperbus-core.c | 154 +++++++++++++++++++++++++++
 include/linux/mtd/hyperbus.h         |  86 +++++++++++++++
 7 files changed, 264 insertions(+)
 create mode 100644 drivers/mtd/hyperbus/Kconfig
 create mode 100644 drivers/mtd/hyperbus/Makefile
 create mode 100644 drivers/mtd/hyperbus/hyperbus-core.c
 create mode 100644 include/linux/mtd/hyperbus.h
MAINTAINERS +7 -0
index 10359a30ed3c..c4860e53c695 100644
@@ -7363,6 +7363,13 @@ F:	include/uapi/linux/hyperv.h
 F:	tools/hv/
 F:	Documentation/ABI/stable/sysfs-bus-vmbus
 
+HYPERBUS SUPPORT
+M:	Vignesh Raghavendra <[email protected]>
+S:	Supported
+F:	drivers/mtd/hyperbus/
+F:	include/linux/mtd/hyperbus.h
+F:	Documentation/devicetree/bindings/mtd/cypress,hyperflash.txt
+
 HYPERVISOR VIRTUAL CONSOLE DRIVER
 L:	[email protected]
 S:	Odd Fixes
drivers/mtd/Kconfig +2 -0
index fb31a7f649a3..80a6e2dcd085 100644
@@ -274,4 +274,6 @@ source "drivers/mtd/spi-nor/Kconfig"
 
 source "drivers/mtd/ubi/Kconfig"
 
+source "drivers/mtd/hyperbus/Kconfig"
+
 endif # MTD
drivers/mtd/Makefile +1 -0
index 806287e80e84..62d649a959e2 100644
@@ -34,3 +34,4 @@ obj-y		+= chips/ lpddr/ maps/ devices/ nand/ tests/
 
 obj-$(CONFIG_MTD_SPI_NOR)	+= spi-nor/
 obj-$(CONFIG_MTD_UBI)		+= ubi/
+obj-$(CONFIG_MTD_HYPERBUS)	+= hyperbus/
drivers/mtd/hyperbus/Kconfig new file +11 -0
new file mode 100644
index 000000000000..98147e28caa0
@@ -0,0 +1,11 @@
+menuconfig MTD_HYPERBUS
+	tristate "HyperBus support"
+	select MTD_CFI
+	select MTD_MAP_BANK_WIDTH_2
+	select MTD_CFI_AMDSTD
+	select MTD_COMPLEX_MAPPINGS
+	help
+	  This is the framework for the HyperBus which can be used by
+	  the HyperBus Controller driver to communicate with
+	  HyperFlash. See Cypress HyperBus specification for more
+	  details
drivers/mtd/hyperbus/Makefile new file +3 -0
new file mode 100644
index 000000000000..ca61dedd730d
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-$(CONFIG_MTD_HYPERBUS)	+= hyperbus-core.o
drivers/mtd/hyperbus/hyperbus-core.c new file +154 -0

154 lines not on this page · raw message

include/linux/mtd/hyperbus.h new file +86 -0

86 lines not on this page · raw message

Signature · 1 line
2.22.0
#6 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

Hello!

Quoted 49 lines · Vignesh Raghavendra

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

[...]

Quoted 41 lines
drivers/mtd/Makefile +1 -0
index 806287e80e84..62d649a959e2 100644
@@ -34,3 +34,4 @@ obj-y		+= chips/ lpddr/ maps/ devices/ nand/ tests/
 
 obj-$(CONFIG_MTD_SPI_NOR)	+= spi-nor/
 obj-$(CONFIG_MTD_UBI)		+= ubi/
+obj-$(CONFIG_MTD_HYPERBUS)	+= hyperbus/
drivers/mtd/hyperbus/Kconfig new file +11 -0
new file mode 100644
index 000000000000..98147e28caa0
@@ -0,0 +1,11 @@
+menuconfig MTD_HYPERBUS
+	tristate "HyperBus support"
+	select MTD_CFI
+	select MTD_MAP_BANK_WIDTH_2
+	select MTD_CFI_AMDSTD
+	select MTD_COMPLEX_MAPPINGS
+	help
+	  This is the framework for the HyperBus which can be used by
+	  the HyperBus Controller driver to communicate with
+	  HyperFlash. See Cypress HyperBus specification for more
+	  details
drivers/mtd/hyperbus/Makefile new file +3 -0
new file mode 100644
index 000000000000..ca61dedd730d
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-$(CONFIG_MTD_HYPERBUS)	+= hyperbus-core.o
drivers/mtd/hyperbus/hyperbus-core.c new file +0 -0
new file mode 100644
index 000000000000..63a9e64895bc
@@ -0,0 +1,154 @@

[...]

Quoted 2 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +2 -0
+int hyperbus_register_device(struct hyperbus_device *hbdev)
+{

[...]

Quoted 6 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +6 -0
+	hbdev->mtd = do_map_probe("cfi_probe", map);
+	if (!hbdev->mtd) {
+		dev_err(dev, "probing of hyperbus device failed\n");
+		return -ENODEV;
+	}
+

drivers/mtd/hyperbus/hyperbus-core.c line 133

Quoted 10 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +10 -0
+	hbdev->mtd->dev.parent = dev;
+	mtd_set_of_node(hbdev->mtd, np);
+
+	ret = mtd_device_register(hbdev->mtd, NULL, 0);
+	if (ret) {
+		dev_err(dev, "failed to register mtd device\n");
+		map_destroy(hbdev->mtd);
+		return ret;
+	}
+	hbdev->registered = true;

I doubt that you actually need this flag...

drivers/mtd/hyperbus/hyperbus-core.c line 143

Quoted 10 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +10 -0
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hyperbus_register_device);
+
+int hyperbus_unregister_device(struct hyperbus_device *hbdev)
+{
+	int ret = 0;
+
+	if (hbdev && hbdev->mtd && hbdev->registered) {

... as you missed clearing that 'registered' flag.

drivers/mtd/hyperbus/hyperbus-core.c line 150

Quoted 7 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +7 -0
+		ret = mtd_device_unregister(hbdev->mtd);
+		map_destroy(hbdev->mtd);
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(hyperbus_unregister_device);

[...]

MBR, Sergei

#7 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

In reply to Vignesh Raghavendra

Hello!

Quoted 43 lines · Vignesh Raghavendra

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

Quoted 6 lines

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

[...]

I have at least created my HyperBus driver and unfortunately I'm having serious
issues with the design of the support core (see below)...

[...]

Quoted 6 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +0 -0
new file mode 100644
index 000000000000..63a9e64895bc
@@ -0,0 +1,154 @@

[...]

Quoted 13 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +13 -0
+int hyperbus_register_device(struct hyperbus_device *hbdev)
+{
+	const struct hyperbus_ops *ops;
+	struct hyperbus_ctlr *ctlr;
+	struct device_node *np;
+	struct map_info *map;
+	struct resource res;
+	struct device *dev;
+	int ret;
+
+	if (!hbdev || !hbdev->np || !hbdev->ctlr || !hbdev->ctlr->dev) {
+		pr_err("hyperbus: please fill all the necessary fields!\n");
+		return -EINVAL;

drivers/mtd/hyperbus/hyperbus-core.c line 81

Quoted 10 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +10 -0
+	}
+
+	np = hbdev->np;
+	ctlr = hbdev->ctlr;
+	if (!of_device_is_compatible(np, "cypress,hyperflash"))
+		return -ENODEV;
+
+	hbdev->memtype = HYPERFLASH;
+
+	ret = of_address_to_resource(np, 0, &res);

Hm, I doubt that the HB devices are wholly mapped into memory space, that seems
like a property of the HB controller. In my case, the flash device in the DT has only single-cell "reg" prop (equal to the chip select #). Then this function returns -EINVAL and the registration fails. Also, in my case such mapping is R/O, not R/W.

drivers/mtd/hyperbus/hyperbus-core.c line 90

Quoted 9 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +9 -0
+	if (ret)
+		return ret;
+
+	dev = ctlr->dev;
+	map = &hbdev->map;
+	map->size = resource_size(&res);
+	map->virt = devm_ioremap_resource(dev, &res);
+	if (IS_ERR(map->virt))
+		return PTR_ERR(map->virt);

Again, I doubt that this should be done here, and not in the HB controller driver...

[...]

MBR, Sergei

author maintainer
#8 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

Quoted 86 lines · Sergei Shtylyov

Hello!

Quoted 43 lines · Vignesh Raghavendra

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

Quoted 6 lines

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

[...]

I have at least created my HyperBus driver and unfortunately I'm having serious
issues with the design of the support core (see below)...

[...]

Quoted 6 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +0 -0
new file mode 100644
index 000000000000..63a9e64895bc
@@ -0,0 +1,154 @@

[...]

Quoted 20 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +20 -0
+int hyperbus_register_device(struct hyperbus_device *hbdev)
+{
+	const struct hyperbus_ops *ops;
+	struct hyperbus_ctlr *ctlr;
+	struct device_node *np;
+	struct map_info *map;
+	struct resource res;
+	struct device *dev;
+	int ret;
+
+	if (!hbdev || !hbdev->np || !hbdev->ctlr || !hbdev->ctlr->dev) {
+		pr_err("hyperbus: please fill all the necessary fields!\n");
+		return -EINVAL;
+	}
+
+	np = hbdev->np;
+	ctlr = hbdev->ctlr;
+	if (!of_device_is_compatible(np, "cypress,hyperflash"))
+		return -ENODEV;
+
Quoted 9 lines

drivers/mtd/hyperbus/hyperbus-core.c line 81

Quoted 3 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +3 -0
+	hbdev->memtype = HYPERFLASH;
+
+	ret = of_address_to_resource(np, 0, &res);

Hm, I doubt that the HB devices are wholly mapped into memory space, that seems
like a property of the HB controller. In my case, the flash device in the DT has only single-cell "reg" prop (equal to the chip select #). Then this function returns -EINVAL and the registration fails. Also, in my case such mapping is R/O, not R/W.

You could declare R/O MMIO region in controla and set up a translation using ranges from slave's reg CS based reg mapping like:

+ hbmc: hyperbus@47034000 {
+ compatible = "ti,am654-hbmc";
+ reg = <0x0 0x47034000 0x0 0x100>,
+ <0x5 0x00000000 0x1 0x0000000>;
+ #address-cells = <2>;
+ #size-cells = <1>;
+ ranges = <0x0 0x0 0x5 0x00000000 0x4000000>, /* CS0 - 64MB */
+ <0x1 0x0 0x5 0x04000000 0x4000000>; /* CS1 - 64MB */
+
+ /* Slave flash node */
+ flash@0,0 {
+ compatible = "cypress,hyperflash", "cfi-flash";
+ reg = <0x0 0x0 0x4000000>;
+ };
+ };

If you use just CS# how would you handle CS to MMIO region mapping? Does both CS use the same MMIO base for reads?

Quoted 11 lines

drivers/mtd/hyperbus/hyperbus-core.c line 90

Quoted 9 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +9 -0
+	if (ret)
+		return ret;
+
+	dev = ctlr->dev;
+	map = &hbdev->map;
+	map->size = resource_size(&res);
+	map->virt = devm_ioremap_resource(dev, &res);
+	if (IS_ERR(map->virt))
+		return PTR_ERR(map->virt);

Again, I doubt that this should be done here, and not in the HB controller driver...

If multiple CS use same MMIO base, then I can make this part of code non fatal when reg entry is a single cell and introduce notion of CS like SPI

Quoted 5 lines

[...]

MBR, Sergei

Signature · 2 lines
Regards
Vignesh
#9 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

Hello!

Quoted 48 lines · Vignesh Raghavendra
Quoted 48 lines
Quoted 48 lines

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

Quoted 4 lines
Quoted 4 lines
Quoted 1 line

[...]

I have at least created my HyperBus driver and unfortunately I'm having serious

At last. :-)

Quoted 40 lines
Quoted 30 lines

issues with the design of the support core (see below)...

[...]

Quoted 6 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +0 -0
new file mode 100644
index 000000000000..63a9e64895bc
@@ -0,0 +1,154 @@

[...]

Quoted 20 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +20 -0
+int hyperbus_register_device(struct hyperbus_device *hbdev)
+{
+	const struct hyperbus_ops *ops;
+	struct hyperbus_ctlr *ctlr;
+	struct device_node *np;
+	struct map_info *map;
+	struct resource res;
+	struct device *dev;
+	int ret;
+
+	if (!hbdev || !hbdev->np || !hbdev->ctlr || !hbdev->ctlr->dev) {
+		pr_err("hyperbus: please fill all the necessary fields!\n");
+		return -EINVAL;
+	}
+
+	np = hbdev->np;
+	ctlr = hbdev->ctlr;
+	if (!of_device_is_compatible(np, "cypress,hyperflash"))
+		return -ENODEV;
+
Quoted 9 lines

drivers/mtd/hyperbus/hyperbus-core.c line 81

Quoted 3 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +3 -0
+	hbdev->memtype = HYPERFLASH;
+
+	ret = of_address_to_resource(np, 0, &res);

Hm, I doubt that the HB devices are wholly mapped into memory space, that seems
like a property of the HB controller. In my case, the flash device in the DT has only single-cell "reg" prop (equal to the chip select #). Then this function returns -EINVAL and the registration fails. Also, in my case such mapping is R/O, not R/W.

Quoted 2 lines

You could declare R/O MMIO region in controla and set up a translation using ranges from slave's reg CS based reg mapping like:

No, not all HB controllers work the same (simple) way as yours. In case of RPC-IF,
the direct read map is a 64 MiB window into a possibly larger flash chip, it has a register supplying address bits 25:31...

Quoted 16 lines
(quoted diff) +15 -0
+	hbmc: hyperbus@47034000 {
+		compatible = "ti,am654-hbmc";
+		reg = <0x0 0x47034000 0x0 0x100>,
+			<0x5 0x00000000 0x1 0x0000000>;
+		#address-cells = <2>;
+		#size-cells = <1>;
+		ranges = <0x0 0x0 0x5 0x00000000 0x4000000>, /* CS0 - 64MB */
+			 <0x1 0x0 0x5 0x04000000 0x4000000>; /* CS1 - 64MB */
+
+		/* Slave flash node */
+		flash@0,0 {
+			compatible = "cypress,hyperflash", "cfi-flash";
+			reg = <0x0 0x0 0x4000000>;
+		};
+	};
Quoted 2 lines

If you use just CS# how would you handle CS to MMIO region mapping? Does both CS use the same MMIO base for reads?

The RPC-IF HF mode only has a single CS signal.

[...]

MBR, Sergei

author maintainer
#10 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

Quoted 52 lines · Sergei Shtylyov

Hello!

Quoted 48 lines · Vignesh Raghavendra
Quoted 48 lines
Quoted 48 lines

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

Quoted 7 lines
Quoted 4 lines
Quoted 4 lines
Quoted 1 line

[...]

I have at least created my HyperBus driver and unfortunately I'm having serious

   At last. :-)

So, I guess driver works for limited memory size?

Quoted 40 lines
Quoted 40 lines
Quoted 39 lines

issues with the design of the support core (see below)...

[...]

Quoted 6 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +0 -0
new file mode 100644
index 000000000000..63a9e64895bc
@@ -0,0 +1,154 @@

[...]

Quoted 13 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +13 -0
+int hyperbus_register_device(struct hyperbus_device *hbdev)
+{
+	const struct hyperbus_ops *ops;
+	struct hyperbus_ctlr *ctlr;
+	struct device_node *np;
+	struct map_info *map;
+	struct resource res;
+	struct device *dev;
+	int ret;
+
+	if (!hbdev || !hbdev->np || !hbdev->ctlr || !hbdev->ctlr->dev) {
+		pr_err("hyperbus: please fill all the necessary fields!\n");
+		return -EINVAL;

drivers/mtd/hyperbus/hyperbus-core.c line 81

Quoted 10 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +10 -0
+	}
+
+	np = hbdev->np;
+	ctlr = hbdev->ctlr;
+	if (!of_device_is_compatible(np, "cypress,hyperflash"))
+		return -ENODEV;
+
+	hbdev->memtype = HYPERFLASH;
+
+	ret = of_address_to_resource(np, 0, &res);

Hm, I doubt that the HB devices are wholly mapped into memory space, that seems
like a property of the HB controller. In my case, the flash device in the DT has only single-cell "reg" prop (equal to the chip select #). Then this function returns -EINVAL and the registration fails. Also, in my case such mapping is R/O, not R/W.

Quoted 6 lines
Quoted 2 lines

You could declare R/O MMIO region in controla and set up a translation using ranges from slave's reg CS based reg mapping like:

No, not all HB controllers work the same (simple) way as yours. In case of RPC-IF,
the direct read map is a 64 MiB window into a possibly larger flash chip, it has a register supplying address bits 25:31...

Okay, this limitation was not made clear earlier. I thought RPC-IF also supported MMIO accesses for all reads

I will look into changes needed to support HB controllers that don't have MMIO interface next week.

Regards
Vignesh

Quoted 17 lines
Quoted 17 lines
(quoted diff) +15 -0
+	hbmc: hyperbus@47034000 {
+		compatible = "ti,am654-hbmc";
+		reg = <0x0 0x47034000 0x0 0x100>,
+			<0x5 0x00000000 0x1 0x0000000>;
+		#address-cells = <2>;
+		#size-cells = <1>;
+		ranges = <0x0 0x0 0x5 0x00000000 0x4000000>, /* CS0 - 64MB */
+			 <0x1 0x0 0x5 0x04000000 0x4000000>; /* CS1 - 64MB */
+
+		/* Slave flash node */
+		flash@0,0 {
+			compatible = "cypress,hyperflash", "cfi-flash";
+			reg = <0x0 0x0 0x4000000>;
+		};
+	};
Quoted 5 lines
Quoted 2 lines

If you use just CS# how would you handle CS to MMIO region mapping? Does both CS use the same MMIO base for reads?

The RPC-IF HF mode only has a single CS signal.

I see...

Quoted 4 lines

[...]

MBR, Sergei

Regards
Vignesh

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

#11 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

In reply to Vignesh Raghavendra

Hello!

Quoted 49 lines · Vignesh Raghavendra

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

[...]

Quoted 6 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +0 -0
new file mode 100644
index 000000000000..63a9e64895bc
@@ -0,0 +1,154 @@

[...]

Quoted 2 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +2 -0
+int hyperbus_register_device(struct hyperbus_device *hbdev)
+{

[...]

drivers/mtd/hyperbus/hyperbus-core.c line 93

Quoted 2 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +2 -0
+	map->name = dev_name(dev);
+	map->bankwidth = 2;

I think this should really be 1, judging on the comment to that field (and on
Cogent's own RPC-IF HF driver).

drivers/mtd/hyperbus/hyperbus-core.c line 94

Quoted 1 line
drivers/mtd/hyperbus/hyperbus-core.c new file +1 -0
+	map->device_node = np;

[...]

MBR, Sergei

author maintainer
#12 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

Quoted 62 lines · Sergei Shtylyov

Hello!

Quoted 49 lines · Vignesh Raghavendra

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

[...]

Quoted 6 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +0 -0
new file mode 100644
index 000000000000..63a9e64895bc
@@ -0,0 +1,154 @@

[...]

Quoted 9 lines
Quoted 2 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +2 -0
+int hyperbus_register_device(struct hyperbus_device *hbdev)
+{

[...]

drivers/mtd/hyperbus/hyperbus-core.c line 93

Quoted 2 lines
drivers/mtd/hyperbus/hyperbus-core.c new file +2 -0
+	map->name = dev_name(dev);
+	map->bankwidth = 2;

I think this should really be 1, judging on the comment to that field (and on
Cogent's own RPC-IF HF driver).

I agree this setting is a bit confusing because DDR nature. What we have with HyperFlash in DDR mode is equivalent to 16bit flash on a 8bit bus and kind of equal to 2 bus cycles (in this case clock edges), therefore bandwidth would turn out to be 2. Otherwise cfi_build_cmd() would generate wrong addresses and simple map implmention of read/writes would use wrong accessors.
Only way I see map->bankwidth = 1 working is if HF is used in SDR mode. So is Cogent's HF in SDR mode? I thought HyperFlash is DDR only but I may be wrong.

Quoted 6 lines

drivers/mtd/hyperbus/hyperbus-core.c line 94

Quoted 1 line
drivers/mtd/hyperbus/hyperbus-core.c new file +1 -0
+	map->device_node = np;

[...]

MBR, Sergei

Signature · 2 lines
Regards
Vignesh
#13 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

In reply to Vignesh Raghavendra

Hi Vignesh,

Quoted 32 lines · Vignesh Raghavendra

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Quoted 7 lines

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

Thanks for your patch, which is now commit dcc7d3446a0fa19b ("mtd:
Add support for HyperBus memory devices") in v5.3.

Quoted 7 lines

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

The last link no longer works. Do you have a replacement? Thanks!

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

author maintainer
#14 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

Hi Geert,

Quoted 49 lines · Geert Uytterhoeven

Hi Vignesh,

Quoted 32 lines · Vignesh Raghavendra

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Quoted 7 lines

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

Thanks for your patch, which is now commit dcc7d3446a0fa19b ("mtd:
Add support for HyperBus memory devices") in v5.3.

Quoted 3 lines

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

Quoted 6 lines
Quoted 4 lines

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

The last link no longer works. Do you have a replacement?

Looks like I used a link point to specific version instead of top level redirector link. Please use:

https://www.ti.com/lit/pdf/spruid7

Regards
Vignesh

#15 lore raw

Re: [PATCH v8 3/5] mtd: Add support for HyperBus memory devices

Hi Vignesh,

Quoted 53 lines · Vignesh
Quoted 52 lines · Geert Uytterhoeven
Quoted 32 lines · Vignesh Raghavendra

Cypress' HyperBus is Low Signal Count, High Performance Double Data Rate Bus interface between a host system master and one or more slave interfaces. HyperBus is used to connect microprocessor, microcontroller, or ASIC devices with random access NOR flash memory (called HyperFlash) or self refresh DRAM (called HyperRAM).

Its a 8-bit data bus (DQ[7:0]) with Read-Write Data Strobe (RWDS) signal and either Single-ended clock(3.0V parts) or Differential clock (1.8V parts). It uses ChipSelect lines to select b/w multiple slaves. At bus level, it follows a separate protocol described in HyperBus specification[1].

HyperFlash follows CFI AMD/Fujitsu Extended Command Set (0x0002) similar to that of existing parallel NORs. Since HyperBus is x8 DDR bus, its equivalent to x16 parallel NOR flash with respect to bits per clock cycle. But HyperBus operates at >166MHz frequencies. HyperRAM provides direct random read/write access to flash memory array.

But, HyperBus memory controllers seem to abstract implementation details and expose a simple MMIO interface to access connected flash.

Add support for registering HyperFlash devices with MTD framework. MTD maps framework along with CFI chip support framework are used to support communicating with flash.

Framework is modelled along the lines of spi-nor framework. HyperBus memory controller (HBMC) drivers calls hyperbus_register_device() to register a single HyperFlash device. HyperFlash core parses MMIO access information from DT, sets up the map_info struct, probes CFI flash and registers it with MTD framework.

Quoted 7 lines

Some HBMC masters need calibration/training sequence[3] to be carried out, in order for DLL inside the controller to lock, by reading a known string/pattern. This is done by repeatedly reading CFI Query Identification String. Calibration needs to be done before trying to detect flash as part of CFI flash probe.

HyperRAM is not supported at the moment.

Thanks for your patch, which is now commit dcc7d3446a0fa19b ("mtd:
Add support for HyperBus memory devices") in v5.3.

Quoted 7 lines

HyperBus specification can be found at[1]
HyperFlash datasheet can be found at[2]

[1] https://www.cypress.com/file/213356/download
[2] https://www.cypress.com/file/213346/download
[3] http://www.ti.com/lit/ug/spruid7b/spruid7b.pdf
Table 12-5741. HyperFlash Access Sequence

Quoted 6 lines
Quoted 1 line

The last link no longer works. Do you have a replacement?

Looks like I used a link point to specific version instead of top level redirector link. Please use:

https://www.ti.com/lit/pdf/spruid7

Thank you, that link works for me.

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

author maintainer
#16 lore raw

PATCH v8 4/5 dt-bindings: mtd: Add bindings for TI's AM654 HyperBus memory controller

In reply to Vignesh Raghavendra

Add binding documentation for TI's HyperBus memory controller present on AM654 SoC.

---
v8:
Fix white space issues

 .../devicetree/bindings/mtd/ti,am654-hbmc.txt | 51 +++++++++++++++++++
 MAINTAINERS                                   |  1 +
 2 files changed, 52 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/mtd/ti,am654-hbmc.txt
Documentation/devicetree/bindings/mtd/ti,am654-hbmc.txt new file +51 -0
new file mode 100644
index 000000000000..faa81c2e5da6
@@ -0,0 +1,51 @@
+Bindings for HyperBus Memory Controller (HBMC) on TI's K3 family of SoCs
+
+Required properties:
+- compatible : "ti,am654-hbmc" for AM654 SoC
+- reg : Two entries:
+	First entry pointed to the register space of HBMC controller
+	Second entry pointing to the memory map region dedicated for
+	MMIO access to attached flash devices
+- ranges : Address translation from offset within CS to allocated MMIO
+	   space in SoC
+
+Optional properties:
+- mux-controls : phandle to the multiplexer that controls selection of
+		 HBMC vs OSPI inside Flash SubSystem (FSS). Default is OSPI,
+		 if property is absent.
+		 See Documentation/devicetree/bindings/mux/reg-mux.txt
+		 for mmio-mux binding details
+
+Example:
+
+	system-controller@47000000 {
+		compatible = "syscon", "simple-mfd";
+		reg = <0x0 0x47000000 0x0 0x100>;
+		#address-cells = <2>;
+		#size-cells = <2>;
+		ranges;
+
+		hbmc_mux: multiplexer {
+			compatible = "mmio-mux";
+			#mux-control-cells = <1>;
+			mux-reg-masks = <0x4 0x2>; /* 0: reg 0x4, bit 1 */
+		};
+	};
+
+	hbmc: hyperbus@47034000 {
+		compatible = "ti,am654-hbmc";
+		reg = <0x0 0x47034000 0x0 0x100>,
+			<0x5 0x00000000 0x1 0x0000000>;
+		power-domains = <&k3_pds 55>;
+		#address-cells = <2>;
+		#size-cells = <1>;
+		ranges = <0x0 0x0 0x5 0x00000000 0x4000000>, /* CS0 - 64MB */
+			 <0x1 0x0 0x5 0x04000000 0x4000000>; /* CS1 - 64MB */
+		mux-controls = <&hbmc_mux 0>;
+
+		/* Slave flash node */
+		flash@0,0 {
+			compatible = "cypress,hyperflash", "cfi-flash";
+			reg = <0x0 0x0 0x4000000>;
+		};
+	};
MAINTAINERS +1 -0
index c4860e53c695..4631c08d94e2 100644
@@ -7369,6 +7369,7 @@ S:	Supported
 F:	drivers/mtd/hyperbus/
 F:	include/linux/mtd/hyperbus.h
 F:	Documentation/devicetree/bindings/mtd/cypress,hyperflash.txt
+F:	Documentation/devicetree/bindings/mtd/ti,am654-hbmc.txt
 
 HYPERVISOR VIRTUAL CONSOLE DRIVER
 L:	[email protected]
Signature · 1 line
2.22.0
author maintainer
#17 lore raw

PATCH v8 5/5 mtd: hyperbus: Add driver for TI's HyperBus memory controller

In reply to Vignesh Raghavendra

Add driver for HyperBus memory controller on TI's AM654 SoC. Programming IP is pretty simple and provides direct memory mapped access to connected Flash devices.

Add basic support for the IP without DMA. Second chipSelect is not supported for now.

---
v8:
No change

 drivers/mtd/hyperbus/Kconfig      |  12 +++
 drivers/mtd/hyperbus/Makefile     |   1 +
 drivers/mtd/hyperbus/hbmc-am654.c | 141 ++++++++++++++++++++++++++++++
 3 files changed, 154 insertions(+)
 create mode 100644 drivers/mtd/hyperbus/hbmc-am654.c
drivers/mtd/hyperbus/Kconfig +12 -0
index 98147e28caa0..cff6bbd226f5 100644
@@ -9,3 +9,15 @@ menuconfig MTD_HYPERBUS
 	  the HyperBus Controller driver to communicate with
 	  HyperFlash. See Cypress HyperBus specification for more
 	  details
+
+if MTD_HYPERBUS
+
+config HBMC_AM654
+	tristate "HyperBus controller driver for AM65x SoC"
+	select MULTIPLEXER
+	select MUX_MMIO
+	help
+	 This is the driver for HyperBus controller on TI's AM65x and
+	 other SoCs
+
+endif # MTD_HYPERBUS
drivers/mtd/hyperbus/Makefile +1 -0
index ca61dedd730d..8a936e066f48 100644
@@ -1,3 +1,4 @@
 # SPDX-License-Identifier: GPL-2.0
 
 obj-$(CONFIG_MTD_HYPERBUS)	+= hyperbus-core.o
+obj-$(CONFIG_HBMC_AM654)	+= hbmc-am654.o
drivers/mtd/hyperbus/hbmc-am654.c new file +141 -0

141 lines not on this page · raw message

Signature · 1 line
2.22.0
author maintainer
#18 lore raw
In reply to Vignesh Raghavendra

Hi,

On 25-Jun-19 1:27 PM, Vignesh Raghavendra wrote:
[...]

Quoted 8 lines

Vignesh Raghavendra (5):
mtd: cfi_cmdset_0002: Add support for polling status register
dt-bindings: mtd: Add binding documentation for HyperFlash
mtd: Add support for HyperBus memory devices
dt-bindings: mtd: Add bindings for TI's AM654 HyperBus memory
controller
mtd: hyperbus: Add driver for TI's HyperBus memory controller

Fixed comments on patch 3/5 locally and series applied to
https://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux.git
branch mtd/next.

Regards
Vignesh

Quoted 21 lines
 .../bindings/mtd/cypress,hyperflash.txt       |  13 ++
 .../devicetree/bindings/mtd/ti,am654-hbmc.txt |  51 ++++++
 MAINTAINERS                                   |   8 +
 drivers/mtd/Kconfig                           |   2 +
 drivers/mtd/Makefile                          |   1 +
 drivers/mtd/chips/cfi_cmdset_0002.c           | 130 +++++++++++++--
 drivers/mtd/hyperbus/Kconfig                  |  23 +++
 drivers/mtd/hyperbus/Makefile                 |   4 +
 drivers/mtd/hyperbus/hbmc-am654.c             | 141 ++++++++++++++++
 drivers/mtd/hyperbus/hyperbus-core.c          | 154 ++++++++++++++++++
 include/linux/mtd/cfi.h                       |   7 +
 include/linux/mtd/hyperbus.h                  |  86 ++++++++++
 12 files changed, 603 insertions(+), 17 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/mtd/cypress,hyperflash.txt
 create mode 100644 Documentation/devicetree/bindings/mtd/ti,am654-hbmc.txt
 create mode 100644 drivers/mtd/hyperbus/Kconfig
 create mode 100644 drivers/mtd/hyperbus/Makefile
 create mode 100644 drivers/mtd/hyperbus/hbmc-am654.c
 create mode 100644 drivers/mtd/hyperbus/hyperbus-core.c
 create mode 100644 include/linux/mtd/hyperbus.h