s3c-hsotg fixes

PATCH New Ben Dooks 3 participants 20 messages Jun 11, 2010 – Jun 14, 2010

Subsystems (4)

Arm (32-bit) USB Arm sub-architectures Arm Samsung Exynos

lore

Share & tools

lore

Ben Dooks

author

2010-06-11 08:20 UTC #0 lore raw

Some further fixes for the s3c-hsotg block, as well as a repost of the missed USB phy clock setting patch which either got missed or passed over from last time.

As a note, this series has a patch adding support for dedicated FIFO mode, this is in my view a fix, as when the hsotg block is compiled with dedicated fifos then each USB IN non-periodic pipe needs to be alloacted a unique TX FIFO. We have no actual data on how well the block performs when all IN TX NP FIFOs are all set to the same one, but it really should be fixed.

I am not sure if I have been subscribed to linux-usb, please CC: me just in case.

Signature · 1 line

Ben

Ben Dooks

author

2010-06-11 08:20 UTC #1 lore raw

PATCH 1/11 USB: s3c_hsotg: Add support for external USB clock

From: Maurus Cuelenaere <[email protected]>

The PLL that drives the USB clock supports 3 input clocks: 12, 24 and 48Mhz. This patch adds support to the USB driver for setting the correct register bit according to the given clock.

This depends on the following patch:
[PATCH] ARM: S3C64XX: Add USB external clock definition

Signed-off-by Maurus Cuelenaere <[email protected]>

Signed-off-by Ben Dooks <[email protected]>

---
 drivers/usb/gadget/s3c-hsotg.c |   19 +++++++++++++++++++
 1 files changed, 19 insertions(+), 0 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +19 -0

index 26193ec..6a303ce 100644
@@ -23,6 +23,7 @@
 #include <linux/delay.h>
 #include <linux/io.h>
 #include <linux/slab.h>
+#include <linux/clk.h>
 
 #include <linux/usb/ch9.h>
 #include <linux/usb/gadget.h>

@@ -2753,6 +2754,7 @@ static void __devinit s3c_hsotg_initep(struct s3c_hsotg *hsotg,
  */
 static void s3c_hsotg_otgreset(struct s3c_hsotg *hsotg)
 {
+	struct clk *xusbxti;
 	u32 osc;
 
 	writel(0, S3C_PHYPWR);

@@ -2760,6 +2762,23 @@ static void s3c_hsotg_otgreset(struct s3c_hsotg *hsotg)
 
 	osc = hsotg->plat->is_osc ? S3C_PHYCLK_EXT_OSC : 0;
 
+	xusbxti = clk_get(hsotg->dev, "xusbxti");
+	if (xusbxti && !IS_ERR(xusbxti)) {
+		switch (clk_get_rate(xusbxti)) {
+		case 12000000:
+		    osc |= S3C_PHYCLK_CLKSEL_12M;
+		    break;
+		case 24000000:
+		    osc |= S3C_PHYCLK_CLKSEL_24M;
+		    break;
+		default:
+		case 48000000:
+		    /* default reference clock */
+		    break;
+		}
+		clk_put(xusbxti);
+	}
+
 	writel(osc | 0x10, S3C_PHYCLK);
 
 	/* issue a full set of resets to the otg and core */

Signature · 1 line

1.7.0.4

Marek Szyprowski

2010-06-11 09:35 UTC #2 lore raw

PATCH 1/11 USB: s3c_hsotg: Add support for external USB clock

Hello,

Quoted 46 lines · Ben Dooks

From: Maurus Cuelenaere <[email protected]>
The PLL that drives the USB clock supports 3 input clocks: 12, 24 and 48Mhz. This patch adds support to the USB driver for setting the correct register bit
according to the given clock.
This depends on the following patch:
[PATCH] ARM: S3C64XX: Add USB external clock definition
Signed-off-by Maurus Cuelenaere <[email protected]>
Signed-off-by Ben Dooks <[email protected]>
---
 drivers/usb/gadget/s3c-hsotg.c |   19 +++++++++++++++++++
 1 files changed, 19 insertions(+), 0 deletions(-)
drivers/usb/gadget/s3c- renamed from drivers/usb/gadget/s3c-hsotg.c +0 -0
hsotg.c
index 26193ec..6a303ce 100644
drivers/usb/gadget/s3c-hsotg.c +2 -0
@@ -23,6 +23,7 @@
 #include <linux/delay.h>
 #include <linux/io.h>
 #include <linux/slab.h>
+#include <linux/clk.h>
 
 #include <linux/usb/ch9.h>
 #include <linux/usb/gadget.h>
@@ -2753,6 +2754,7 @@ static void __devinit s3c_hsotg_initep(struct
 s3c_hsotg *hsotg,
  */
 static void s3c_hsotg_otgreset(struct s3c_hsotg *hsotg)
 {
+	struct clk *xusbxti;
 	u32 osc;
 
drivers/usb/gadget/s3c-hsotg.c +3 -0
 	writel(0, S3C_PHYPWR);
@@ -2760,6 +2762,23 @@ static void s3c_hsotg_otgreset(struct s3c_hsotg
 *hsotg)
 
 	osc = hsotg->plat->is_osc ? S3C_PHYCLK_EXT_OSC : 0;
 
+	xusbxti = clk_get(hsotg->dev, "xusbxti");
+	if (xusbxti && !IS_ERR(xusbxti)) {
+		switch (clk_get_rate(xusbxti)) {

drivers/usb/gadget/s3c-hsotg.c static void s3c_hsotg_otgreset(struct s3c_hsotg · line 2778

Quoted 10 lines

drivers/usb/gadget/s3c-hsotg.c +10 -0

+		case 12000000:
+		    osc |= S3C_PHYCLK_CLKSEL_12M;
+		    break;
+		case 24000000:
+		    osc |= S3C_PHYCLK_CLKSEL_24M;
+		    break;
+		default:
+		case 48000000:
+		    /* default reference clock */
+		    break;

One quick style issue: there is some tabs-vs-spaces mess in the above code.

Best regards
--
Marek Szyprowski
Samsung Poland R&D Center

Maurus Cuelenaere

2010-06-11 09:46 UTC #3 lore raw

PATCH 1/11 USB: s3c_hsotg: Add support for external USB clock

The PLL that drives the USB clock supports 3 input clocks: 12, 24 and 48Mhz. This patch adds support to the USB driver for setting the correct register bit according to the given clock.

This depends on the following patch:
[PATCH] ARM: S3C64XX: Add USB external clock definition

Signed-off-by Maurus Cuelenaere <[email protected]>

---

v2: - fixes space issues
- use MHZ define

---
 drivers/usb/gadget/s3c-hsotg.c |   19 +++++++++++++++++++
 1 files changed, 19 insertions(+), 0 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +19 -0

index 1f73b48..40ee63a 100644
@@ -23,6 +23,7 @@
 #include <linux/delay.h>
 #include <linux/io.h>
 #include <linux/slab.h>
+#include <linux/clk.h>
 
 #include <linux/usb/ch9.h>
 #include <linux/usb/gadget.h>

@@ -2699,6 +2700,7 @@ static void __devinit s3c_hsotg_initep(struct s3c_hsotg *hsotg,
  */
 static void s3c_hsotg_otgreset(struct s3c_hsotg *hsotg)
 {
+	struct clk *xusbxti;
 	u32 osc;
 
 	writel(0, S3C_PHYPWR);

@@ -2706,6 +2708,23 @@ static void s3c_hsotg_otgreset(struct s3c_hsotg *hsotg)
 
 	osc = hsotg->plat->is_osc ? S3C_PHYCLK_EXT_OSC : 0;
 
+	xusbxti = clk_get(hsotg->dev, "xusbxti");
+	if (xusbxti && !IS_ERR(xusbxti)) {
+		switch (clk_get_rate(xusbxti)) {
+		case 12*MHZ:
+			osc |= S3C_PHYCLK_CLKSEL_12M;
+			break;
+		case 24*MHZ:
+			osc |= S3C_PHYCLK_CLKSEL_24M;
+			break;
+		default:
+		case 48*MHZ:
+			/* default reference clock */
+			break;
+		}
+		clk_put(xusbxti);
+	}
+
 	writel(osc | 0x10, S3C_PHYCLK);
 
 	/* issue a full set of resets to the otg and core */

Signature · 1 line

1.7.1

Ben Dooks

author

2010-06-11 08:20 UTC #4 lore raw

PATCH 2/11 USB: s3c-hsotg: Increase TX fifo limit

In reply to Ben Dooks

Up the FIFO size for the TX to 1024 entries, as this now seems to work with all the cores. This fixes a problem when using large packets on a core with MPS set to 512 can hang due to insufficient space for the writes.

The hang arises due to getting the non-periodic FIFO empty IRQ but not being able to satisfy any requests since there is never enough space to write 512 bytes into the buffer. This means we end up with a stream of interrupt requests.

It is easier to up the TX FIFO to fill the space we left for it than to try and fix the positions in the code where we should have limited the max-packet size to < TXFIFOSIZE, since the TXFIFOSIZE depends on how the TX FIFOs have been setup.

Signed-off-by Ben Dooks <[email protected]>

---
 drivers/usb/gadget/s3c-hsotg.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +2 -2

index 6a303ce..ba00e6e 100644
@@ -311,11 +311,11 @@ static void s3c_hsotg_init_fifo(struct s3c_hsotg *hsotg)
 		hsotg->regs + S3C_GNPTXFSIZ);
 	*/
 
-	/* set FIFO sizes to 2048/0x1C0 */
+	/* set FIFO sizes to 2048/1024 */
 
 	writel(2048, hsotg->regs + S3C_GRXFSIZ);
 	writel(S3C_GNPTXFSIZ_NPTxFStAddr(2048) |
-	       S3C_GNPTXFSIZ_NPTxFDep(0x1C0),
+	       S3C_GNPTXFSIZ_NPTxFDep(1024),
 	       hsotg->regs + S3C_GNPTXFSIZ);
 
 	/* arange all the rest of the TX FIFOs, as some versions of this

Signature · 1 line

1.7.0.4

Ben Dooks

author

2010-06-11 08:20 UTC #5 lore raw

PATCH 3/11 USB: s3c-hsotg: The NPTX/PTX FIFO sizes are in words, not bytes

In reply to Ben Dooks

Fix a problem where we have been underestimating the space available in the IN PTX/NPTX FIFOs by assuming that they where simply word aligned instead of in number-of-words. This means all length calculations need to be multiplied-by-4.

Note, we do not change the information about fifo size or start addresses available to userspace as we assume the user can multiply by four easily and is already knows these values are in words.

Signed-off-by Ben Dooks <[email protected]>

---
 drivers/usb/gadget/s3c-hsotg.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +2 -1

index ba00e6e..a018b6a 100644
@@ -506,6 +506,7 @@ static int s3c_hsotg_write_fifo(struct s3c_hsotg *hsotg,
 		}
 
 		can_write = S3C_GNPTXSTS_NPTxFSpcAvail_GET(gnptxsts);
+		can_write *= 4;	/* fifo size is in 32bit quantities. */
 	}
 
 	dev_dbg(hsotg->dev, "%s: GNPTXSTS=%08x, can=%d, to=%d, mps %d\n",

@@ -2733,7 +2734,7 @@ static void __devinit s3c_hsotg_initep(struct s3c_hsotg *hsotg,
 	 */
 
 	ptxfifo = readl(hsotg->regs + S3C_DPTXFSIZn(epnum));
-	hs_ep->fifo_size = S3C_DPTXFSIZn_DPTxFSize_GET(ptxfifo);
+	hs_ep->fifo_size = S3C_DPTXFSIZn_DPTxFSize_GET(ptxfifo) * 4;
 
 	/* if we're using dma, we need to set the next-endpoint pointer
 	 * to be something valid.

Signature · 1 line

1.7.0.4

Ben Dooks

author

2010-06-11 08:20 UTC #6 lore raw

PATCH 4/11 USB: s3c-hsotg: Avoid overwriting contents of perodic in 'fifo'

In reply to Ben Dooks

In shared fifo mode (used on older SoCs) the periodic in fifo beahves much more like a packet buffer, discarding old data when writing new data. Avoid this by ensuring that we do not load new transactions in when there is data sitting already in the FIFO.

Note, this may not be an observed bug, we are fixing the case that this may happen.

Signed-off-by Ben Dooks <[email protected]>

---
 drivers/usb/gadget/s3c-hsotg.c |   12 +++++++++++-
 1 files changed, 11 insertions(+), 1 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +11 -1

index a018b6a..df3e8d1 100644
@@ -92,7 +92,9 @@ struct s3c_hsotg_req;
  * For periodic IN endpoints, we have fifo_size and fifo_load to try
  * and keep track of the amount of data in the periodic FIFO for each
  * of these as we don't have a status register that tells us how much
- * is in each of them.
+ * is in each of them. (note, this may actually be useless information
+ * as in shared-fifo mode periodic in acts like a single-frame packet
+ * buffer than a fifo)
  */
 struct s3c_hsotg_ep {
 	struct usb_ep		ep;

@@ -475,6 +477,14 @@ static int s3c_hsotg_write_fifo(struct s3c_hsotg *hsotg,
 
 		size_left = S3C_DxEPTSIZ_XferSize_GET(epsize);
 
+		/* if shared fifo, we cannot write anything until the
+		 * previous data has been completely sent.
+		 */
+		if (hs_ep->fifo_load != 0) {
+			s3c_hsotg_en_gsint(hsotg, S3C_GINTSTS_PTxFEmp);
+			return -ENOSPC;
+		}
+
 		dev_dbg(hsotg->dev, "%s: left=%d, load=%d, fifo=%d, size %d\n",
 			__func__, size_left,
 			hs_ep->size_loaded, hs_ep->fifo_load, hs_ep->fifo_size);

Signature · 1 line

1.7.0.4

Ben Dooks

author

2010-06-11 08:20 UTC #7 lore raw

PATCH 5/11 USB: s3c-hsotg: Re-initialise all FIFOs on USB bus reset

In reply to Ben Dooks

The USB documentation suggest that the FIFOs should be reset when a bus reset event happens. Use the s3c_hsotg_init_fifo() to ensure that the FIFO layout is correct and that the FIFOs are flushed before acknowledging the reset.

Signed-off-by Ben Dooks <[email protected]>

---
 drivers/usb/gadget/s3c-hsotg.c |   13 ++++---------
 1 files changed, 4 insertions(+), 9 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +4 -9

index df3e8d1..5b61bcf 100644
@@ -2083,17 +2083,12 @@ irq_retry:
 		kill_all_requests(hsotg, &hsotg->eps[0], -ECONNRESET, true);
 
 		/* it seems after a reset we can end up with a situation
-		 * where the TXFIFO still has data in it... try flushing
-		 * it to remove anything that may still be in it.
+		 * where the TXFIFO still has data in it... the docs
+		 * suggest resetting all the fifos, so use the init_fifo
+		 * code to relayout and flush the fifos.
 		 */
 
-		if (1) {
-			writel(S3C_GRSTCTL_TxFNum(0) | S3C_GRSTCTL_TxFFlsh,
-			       hsotg->regs + S3C_GRSTCTL);
-
-			dev_info(hsotg->dev, "GNPTXSTS=%08x\n",
-				 readl(hsotg->regs + S3C_GNPTXSTS));
-		}
+		s3c_hsotg_init_fifo(hsotg);
 
 		s3c_hsotg_enqueue_setup(hsotg);

Signature · 1 line

1.7.0.4

Ben Dooks

author

2010-06-11 08:20 UTC #8 lore raw

PATCH 6/11 USB: s3c-hsotg: Add initial detection and setup for dedicated FIFO mode

In reply to Ben Dooks

Add support for the dedicated FIFO mode on newer SoCs such as the S5PV210 partly to improve support and to fix the bug where any non-EP0 IN endpoint requires its own FIFO allocation.

To fix this, we ensure that any non-zero IN endpoint is given a TXFIFO using the same allocation method as the periodic case (all our current hardware has enough FIFOs and FIFO memory for a 1:1 mapping) and ensure that the necessary transmission done interrupt is enabled.

The default settings from reset for the core point all EPs at FIFO0, used for the control endpoint. However, the controller documentation states that all IN endpoints _must_ have a unique FIFO to avoid any contention during transmission.

Note, this leaves us with a large IN FIFO for EP0 (which re-uses the old NPTXFIFO) for an endpoint which cannot shift more than a pair of packets at a time... this is a waste, but it looks like we cannot re-allocate space to the individual IN FIFOs as they are already maxed out (to be confirmed).

Signed-off-by Ben Dooks <[email protected]?

---
 .../arm/plat-samsung/include/plat/regs-usb-hsotg.h |    2 +
 drivers/usb/gadget/s3c-hsotg.c                     |   40 +++++++++++++++++++-
 2 files changed, 40 insertions(+), 2 deletions(-)

arch/arm/plat-samsung/include/plat/regs-usb-hsotg.h +2 -0

index 8d18d9d..dc90f5e 100644
@@ -226,6 +226,7 @@
 
 #define S3C_DIEPMSK				S3C_HSOTG_REG(0x810)
 
+#define S3C_DIEPMSK_TxFIFOEmpty			(1 << 7)
 #define S3C_DIEPMSK_INEPNakEffMsk		(1 << 6)
 #define S3C_DIEPMSK_INTknEPMisMsk		(1 << 5)
 #define S3C_DIEPMSK_INTknTXFEmpMsk		(1 << 4)

@@ -371,6 +372,7 @@
 
 #define S3C_DIEPDMA(_a)				S3C_HSOTG_REG(0x914 + ((_a) * 0x20))
 #define S3C_DOEPDMA(_a)				S3C_HSOTG_REG(0xB14 + ((_a) * 0x20))
+#define S3C_DTXFSTS(_a)				S3C_HSOTG_REG(0x918 + ((_a) * 0x20))
 
 #define S3C_EPFIFO(_a)				S3C_HSOTG_REG(0x1000 + ((_a) * 0x1000))

drivers/usb/gadget/s3c-hsotg.c +38 -2

index 5b61bcf..819e75a 100644
@@ -12,6 +12,8 @@
  * published by the Free Software Foundation.
 */
 
+#define DEBUG
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/spinlock.h>

@@ -131,6 +133,7 @@ struct s3c_hsotg_ep {
  * @regs: The memory area mapped for accessing registers.
  * @regs_res: The resource that was allocated when claiming register space.
  * @irq: The IRQ number we are using
+ * @dedicated_fifos: Set if the hardware has dedicated IN-EP fifos.
  * @debug_root: root directrory for debugfs.
  * @debug_file: main status file for debugfs.
  * @debug_fifo: FIFO status file for debugfs.

@@ -149,6 +152,8 @@ struct s3c_hsotg {
 	struct resource		*regs_res;
 	int			irq;
 
+	unsigned int		dedicated_fifos:1;
+
 	struct dentry		*debug_root;
 	struct dentry		*debug_file;
 	struct dentry		*debug_fifo;

@@ -467,7 +472,7 @@ static int s3c_hsotg_write_fifo(struct s3c_hsotg *hsotg,
 	if (to_write == 0)
 		return 0;
 
-	if (periodic) {
+	if (periodic && !hsotg->dedicated_fifos) {
 		u32 epsize = readl(hsotg->regs + S3C_DIEPTSIZ(hs_ep->index));
 		int size_left;
 		int size_done;

@@ -505,6 +510,11 @@ static int s3c_hsotg_write_fifo(struct s3c_hsotg *hsotg,
 			s3c_hsotg_en_gsint(hsotg, S3C_GINTSTS_PTxFEmp);
 			return -ENOSPC;
 		}
+	} else if (hsotg->dedicated_fifos && hs_ep->index != 0) {
+		can_write = readl(hsotg->regs + S3C_DTXFSTS(hs_ep->index));
+
+		can_write &= 0xffff;
+		can_write *= 4;
 	} else {
 		if (S3C_GNPTXSTS_NPTxQSpcAvail_GET(gnptxsts) == 0) {
 			dev_dbg(hsotg->dev,

@@ -1830,6 +1840,15 @@ static void s3c_hsotg_epint(struct s3c_hsotg *hsotg, unsigned int idx,
 				 __func__, idx);
 			clear |= S3C_DIEPMSK_INTknEPMisMsk;
 		}
+
+		/* FIFO has space or is empty (see GAHBCFG) */
+		if (hsotg->dedicated_fifos &&
+		    ints & S3C_DIEPMSK_TxFIFOEmpty) {
+			dev_dbg(hsotg->dev, "%s: ep%d: TxFIFOEmpty\n",
+				__func__, idx);
+			s3c_hsotg_trytx(hsotg, hs_ep);
+			clear |= S3C_DIEPMSK_TxFIFOEmpty;
+		}
 	}
 
 	writel(clear, hsotg->regs + epint_reg);

@@ -2281,6 +2300,12 @@ static int s3c_hsotg_ep_enable(struct usb_ep *ep,
 		break;
 	}
 
+	/* if the hardware has dedicated fifos, we must give each IN EP
+	 * a unique tx-fifo even if it is non-periodic.
+	 */
+	if (dir_in && hsotg->dedicated_fifos)
+		epctrl |= S3C_DxEPCTL_TxFNum(index);
+
 	/* for non control endpoints, set PID to D0 */
 	if (index)
 		epctrl |= S3C_DxEPCTL_SetD0PID;

@@ -2570,7 +2595,8 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	writel(S3C_DIEPMSK_TimeOUTMsk | S3C_DIEPMSK_AHBErrMsk |
 	       S3C_DIEPMSK_INTknEPMisMsk |
-	       S3C_DIEPMSK_EPDisbldMsk | S3C_DIEPMSK_XferComplMsk,
+	       S3C_DIEPMSK_EPDisbldMsk | S3C_DIEPMSK_XferComplMsk |
+	       ((hsotg->dedicated_fifos) ? S3C_DIEPMSK_TxFIFOEmpty : 0),
 	       hsotg->regs + S3C_DIEPMSK);
 
 	/* don't need XferCompl, we get that from RXFIFO in slave mode. In

@@ -2797,6 +2823,8 @@ static void s3c_hsotg_otgreset(struct s3c_hsotg *hsotg)
 
 static void s3c_hsotg_init(struct s3c_hsotg *hsotg)
 {
+	u32 cfg4;
+
 	/* unmask subset of endpoint interrupts */
 
 	writel(S3C_DIEPMSK_TimeOUTMsk | S3C_DIEPMSK_AHBErrMsk |

@@ -2832,6 +2860,14 @@ static void s3c_hsotg_init(struct s3c_hsotg *hsotg)
 
 	writel(using_dma(hsotg) ? S3C_GAHBCFG_DMAEn : 0x0,
 	       hsotg->regs + S3C_GAHBCFG);
+
+	/* check hardware configuration */
+
+	cfg4 = readl(hsotg->regs + 0x50);
+	hsotg->dedicated_fifos = (cfg4 >> 25) & 1;
+
+	dev_info(hsotg->dev, "%s fifos\n",
+		 hsotg->dedicated_fifos ? "dedicated" : "shared");
 }
 
 static void s3c_hsotg_dump(struct s3c_hsotg *hsotg)

Signature · 1 line

1.7.0.4

Maurus Cuelenaere

2010-06-11 08:54 UTC #9 lore raw

PATCH 6/11 USB: s3c-hsotg: Add initial detection and setup for dedicated FIFO mode

Op 11-06-10 10:20, Ben Dooks schreef:

Quoted 14 lines

Add support for the dedicated FIFO mode on newer SoCs such as the S5PV210 partly to improve support and to fix the bug where any non-EP0 IN endpoint requires its own FIFO allocation.
To fix this, we ensure that any non-zero IN endpoint is given a TXFIFO using the same allocation method as the periodic case (all our current hardware has enough FIFOs and FIFO memory for a 1:1 mapping) and ensure that the necessary transmission done interrupt is enabled.
The default settings from reset for the core point all EPs at FIFO0, used for the control endpoint. However, the controller documentation states that all IN endpoints _must_ have a unique FIFO to avoid any contention during transmission.

Quoted 7 lines

Note, this leaves us with a large IN FIFO for EP0 (which re-uses the old NPTXFIFO) for an endpoint which cannot shift more than a pair of packets at a time... this is a waste, but it looks like we cannot re-allocate space to the individual IN FIFOs as they are already maxed out (to be confirmed).
Signed-off-by Ben Dooks <[email protected]?

Shouldn't this be s/\?/>/ ?

Quoted 30 lines

---
 .../arm/plat-samsung/include/plat/regs-usb-hsotg.h |    2 +
 drivers/usb/gadget/s3c-hsotg.c                     |   40 +++++++++++++++++++-
 2 files changed, 40 insertions(+), 2 deletions(-)

arch/arm/plat-samsung/include/plat/regs-usb-hsotg.h +2 -0

index 8d18d9d..dc90f5e 100644
@@ -226,6 +226,7 @@
 
 #define S3C_DIEPMSK				S3C_HSOTG_REG(0x810)
 
+#define S3C_DIEPMSK_TxFIFOEmpty			(1 << 7)
 #define S3C_DIEPMSK_INEPNakEffMsk		(1 << 6)
 #define S3C_DIEPMSK_INTknEPMisMsk		(1 << 5)
 #define S3C_DIEPMSK_INTknTXFEmpMsk		(1 << 4)

@@ -371,6 +372,7 @@
 
 #define S3C_DIEPDMA(_a)				S3C_HSOTG_REG(0x914 + ((_a) * 0x20))
 #define S3C_DOEPDMA(_a)				S3C_HSOTG_REG(0xB14 + ((_a) * 0x20))
+#define S3C_DTXFSTS(_a)				S3C_HSOTG_REG(0x918 + ((_a) * 0x20))
 
 #define S3C_EPFIFO(_a)				S3C_HSOTG_REG(0x1000 + ((_a) * 0x1000))

drivers/usb/gadget/s3c-hsotg.c +0 -0

index 5b61bcf..819e75a 100644

drivers/usb/gadget/s3c-hsotg.c line 15

Quoted 4 lines

drivers/usb/gadget/s3c-hsotg.c +1 -0

@@ -12,6 +12,8 @@
  * published by the Free Software Foundation.
 */
 
+#define DEBUG

I don't think this is an intended change?

Quoted 105 lines

drivers/usb/gadget/s3c-hsotg.c +37 -2

+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/spinlock.h>

@@ -131,6 +133,7 @@ struct s3c_hsotg_ep {
  * @regs: The memory area mapped for accessing registers.
  * @regs_res: The resource that was allocated when claiming register space.
  * @irq: The IRQ number we are using
+ * @dedicated_fifos: Set if the hardware has dedicated IN-EP fifos.
  * @debug_root: root directrory for debugfs.
  * @debug_file: main status file for debugfs.
  * @debug_fifo: FIFO status file for debugfs.

@@ -149,6 +152,8 @@ struct s3c_hsotg {
 	struct resource		*regs_res;
 	int			irq;
 
+	unsigned int		dedicated_fifos:1;
+
 	struct dentry		*debug_root;
 	struct dentry		*debug_file;
 	struct dentry		*debug_fifo;

@@ -467,7 +472,7 @@ static int s3c_hsotg_write_fifo(struct s3c_hsotg *hsotg,
 	if (to_write == 0)
 		return 0;
 
-	if (periodic) {
+	if (periodic && !hsotg->dedicated_fifos) {
 		u32 epsize = readl(hsotg->regs + S3C_DIEPTSIZ(hs_ep->index));
 		int size_left;
 		int size_done;

@@ -505,6 +510,11 @@ static int s3c_hsotg_write_fifo(struct s3c_hsotg *hsotg,
 			s3c_hsotg_en_gsint(hsotg, S3C_GINTSTS_PTxFEmp);
 			return -ENOSPC;
 		}
+	} else if (hsotg->dedicated_fifos && hs_ep->index != 0) {
+		can_write = readl(hsotg->regs + S3C_DTXFSTS(hs_ep->index));
+
+		can_write &= 0xffff;
+		can_write *= 4;
 	} else {
 		if (S3C_GNPTXSTS_NPTxQSpcAvail_GET(gnptxsts) == 0) {
 			dev_dbg(hsotg->dev,

@@ -1830,6 +1840,15 @@ static void s3c_hsotg_epint(struct s3c_hsotg *hsotg, unsigned int idx,
 				 __func__, idx);
 			clear |= S3C_DIEPMSK_INTknEPMisMsk;
 		}
+
+		/* FIFO has space or is empty (see GAHBCFG) */
+		if (hsotg->dedicated_fifos &&
+		    ints & S3C_DIEPMSK_TxFIFOEmpty) {
+			dev_dbg(hsotg->dev, "%s: ep%d: TxFIFOEmpty\n",
+				__func__, idx);
+			s3c_hsotg_trytx(hsotg, hs_ep);
+			clear |= S3C_DIEPMSK_TxFIFOEmpty;
+		}
 	}
 
 	writel(clear, hsotg->regs + epint_reg);

@@ -2281,6 +2300,12 @@ static int s3c_hsotg_ep_enable(struct usb_ep *ep,
 		break;
 	}
 
+	/* if the hardware has dedicated fifos, we must give each IN EP
+	 * a unique tx-fifo even if it is non-periodic.
+	 */
+	if (dir_in && hsotg->dedicated_fifos)
+		epctrl |= S3C_DxEPCTL_TxFNum(index);
+
 	/* for non control endpoints, set PID to D0 */
 	if (index)
 		epctrl |= S3C_DxEPCTL_SetD0PID;

@@ -2570,7 +2595,8 @@ int usb_gadget_register_driver(struct usb_gadget_driver *driver)
 
 	writel(S3C_DIEPMSK_TimeOUTMsk | S3C_DIEPMSK_AHBErrMsk |
 	       S3C_DIEPMSK_INTknEPMisMsk |
-	       S3C_DIEPMSK_EPDisbldMsk | S3C_DIEPMSK_XferComplMsk,
+	       S3C_DIEPMSK_EPDisbldMsk | S3C_DIEPMSK_XferComplMsk |
+	       ((hsotg->dedicated_fifos) ? S3C_DIEPMSK_TxFIFOEmpty : 0),
 	       hsotg->regs + S3C_DIEPMSK);
 
 	/* don't need XferCompl, we get that from RXFIFO in slave mode. In

@@ -2797,6 +2823,8 @@ static void s3c_hsotg_otgreset(struct s3c_hsotg *hsotg)
 
 static void s3c_hsotg_init(struct s3c_hsotg *hsotg)
 {
+	u32 cfg4;
+
 	/* unmask subset of endpoint interrupts */
 
 	writel(S3C_DIEPMSK_TimeOUTMsk | S3C_DIEPMSK_AHBErrMsk |

@@ -2832,6 +2860,14 @@ static void s3c_hsotg_init(struct s3c_hsotg *hsotg)
 
 	writel(using_dma(hsotg) ? S3C_GAHBCFG_DMAEn : 0x0,
 	       hsotg->regs + S3C_GAHBCFG);
+
+	/* check hardware configuration */
+
+	cfg4 = readl(hsotg->regs + 0x50);
+	hsotg->dedicated_fifos = (cfg4 >> 25) & 1;
+
+	dev_info(hsotg->dev, "%s fifos\n",
+		 hsotg->dedicated_fifos ? "dedicated" : "shared");
 }
 
 static void s3c_hsotg_dump(struct s3c_hsotg *hsotg)

Ben Dooks

author

2010-06-11 08:20 UTC #10 lore raw

PATCH 7/11 USB: s3c-hsotg: Only load packet per fifo write

In reply to Ben Dooks

Limit the IN FIFO write to a single packet per attempt at writing, as per the specifications and ensure that we don't return fifo-full so that we can continue writing packets if we have the space.

Signed-off-by Ben Dooks <[email protected]>

---
 drivers/usb/gadget/s3c-hsotg.c |   11 +++++++++++
 1 files changed, 11 insertions(+), 0 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +11 -0

index 819e75a..7c35bfe 100644
@@ -539,6 +539,17 @@ static int s3c_hsotg_write_fifo(struct s3c_hsotg *hsotg,
 	if (can_write > 512)
 		can_write = 512;
 
+	/* limit the write to one max-packet size worth of data, but allow
+	 * the transfer to return that it did not run out of fifo space
+	 * doing it. */
+	if (to_write > hs_ep->ep.maxpacket) {
+		to_write = hs_ep->ep.maxpacket;
+
+		s3c_hsotg_en_gsint(hsotg,
+				   periodic ? S3C_GINTSTS_PTxFEmp :
+				   S3C_GINTSTS_NPTxFEmp);
+	}
+
 	/* see if we can write data */
 
 	if (to_write > can_write) {

Signature · 1 line

1.7.0.4

Ben Dooks

author

2010-06-11 08:20 UTC #11 lore raw

PATCH 8/11 USB: s3c-hsotg: Check for new request before enqueing new setup

In reply to Ben Dooks

Before trying a new setup transaction after getting an EP0 in complete interrupt, check that the driver did not try and send more EP0 IN data before enqueing a new setup transaction.

This fixes a bug where we cannot send all of the IN data in one go so split the transfer, but then fail to send all the data as we start waiting for a new OUT transaction

Signed-off-by Ben Dooks <[email protected]>

---
 drivers/usb/gadget/s3c-hsotg.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +1 -1

index 7c35bfe..2555f27 100644
@@ -1791,7 +1791,7 @@ static void s3c_hsotg_epint(struct s3c_hsotg *hsotg, unsigned int idx,
 		if (dir_in) {
 			s3c_hsotg_complete_in(hsotg, hs_ep);
 
-			if (idx == 0)
+			if (idx == 0 && !hs_ep->req)
 				s3c_hsotg_enqueue_setup(hsotg);
 		} else if (using_dma(hsotg)) {
 			/* We're using DMA, we need to fire an OutDone here

Signature · 1 line

1.7.0.4

Ben Dooks

author

2010-06-11 08:20 UTC #12 lore raw

PATCH 9/11 USB: s3c-hsotg: Fix max EP0 IN request length

In reply to Ben Dooks

The maximum length for any EP0 IN request on EP0 is 127 bytes, not 128 as the driver currently has it.

Signed-off-by Ben Dooks <[email protected]>

---
 drivers/usb/gadget/s3c-hsotg.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +1 -2

index 2555f27..17b0779 100644
@@ -613,8 +613,7 @@ static unsigned get_ep_limit(struct s3c_hsotg_ep *hs_ep)
 		maxpkt = S3C_DxEPTSIZ_PktCnt_LIMIT + 1;
 	} else {
 		if (hs_ep->dir_in) {
-			/* maxsize = S3C_DIEPTSIZ0_XferSize_LIMIT + 1; */
-			maxsize = 64+64+1;
+			maxsize = 64+64;
 			maxpkt = S3C_DIEPTSIZ0_PktCnt_LIMIT + 1;
 		} else {
 			maxsize = 0x3f;

Signature · 1 line

1.7.0.4

Ben Dooks

author

2010-06-11 08:20 UTC #13 lore raw

PATCH 10/11 USB: s3c-hsotg: Fix the OUT EP0 limit

In reply to Ben Dooks

The EP0 out limit is the same as the IN limit, so make them the same.

Signed-off-by Ben Dooks <[email protected]>

tmp-fix

---
 drivers/usb/gadget/s3c-hsotg.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +1 -2

index 17b0779..b34a05d 100644
@@ -612,11 +612,10 @@ static unsigned get_ep_limit(struct s3c_hsotg_ep *hs_ep)
 		maxsize = S3C_DxEPTSIZ_XferSize_LIMIT + 1;
 		maxpkt = S3C_DxEPTSIZ_PktCnt_LIMIT + 1;
 	} else {
+		maxsize = 64+64;
 		if (hs_ep->dir_in) {
-			maxsize = 64+64;
 			maxpkt = S3C_DIEPTSIZ0_PktCnt_LIMIT + 1;
 		} else {
-			maxsize = 0x3f;
 			maxpkt = 2;
 		}
 	}

Signature · 1 line

1.7.0.4

Maurus Cuelenaere

2010-06-11 09:01 UTC #14 lore raw

PATCH 10/11 USB: s3c-hsotg: Fix the OUT EP0 limit

Op 11-06-10 10:20, Ben Dooks schreef:

Quoted 5 lines

The EP0 out limit is the same as the IN limit, so make them the same.
Signed-off-by Ben Dooks <[email protected]>
tmp-fix

Shouldn't this line be removed?

Quoted 21 lines

---
 drivers/usb/gadget/s3c-hsotg.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +1 -2

index 17b0779..b34a05d 100644
@@ -612,11 +612,10 @@ static unsigned get_ep_limit(struct s3c_hsotg_ep *hs_ep)
 		maxsize = S3C_DxEPTSIZ_XferSize_LIMIT + 1;
 		maxpkt = S3C_DxEPTSIZ_PktCnt_LIMIT + 1;
 	} else {
+		maxsize = 64+64;
 		if (hs_ep->dir_in) {
-			maxsize = 64+64;
 			maxpkt = S3C_DIEPTSIZ0_PktCnt_LIMIT + 1;
 		} else {
-			maxsize = 0x3f;
 			maxpkt = 2;
 		}
 	}

Marek Szyprowski

2010-06-11 09:07 UTC #15 lore raw

PATCH 10/11 USB: s3c-hsotg: Fix the OUT EP0 limit

In reply to Ben Dooks

Hello,

Quoted 5 lines · Ben Dooks

The EP0 out limit is the same as the IN limit, so make them the same.
Signed-off-by Ben Dooks <[email protected]>
tmp-fix

^ looks like this temporary fix managed to get into the final patch series without proper cleanup.

Best regards
--
Marek Szyprowski
Samsung Poland R&D Center

Ben Dooks

author

2010-06-11 08:20 UTC #16 lore raw

PATCH 11/11 USB: s3c-hsotg: Fix OUT packet request retry

In reply to Ben Dooks

If there is more data in the request than we could fit into a single hardware request, then check when the OutDone event is received if we have more data, and if so, schedule the new data instead of trying to complete the request (and in the case of EP0, sending a 0 packet in the middle of a transfer).

Also, move the debug message about the current transfer state before the warning about a bad transfer.

Signed-off-by Ben Dooks <[email protected]>

---
 drivers/usb/gadget/s3c-hsotg.c |   18 +++++++++++-------
 1 files changed, 11 insertions(+), 7 deletions(-)

drivers/usb/gadget/s3c-hsotg.c +11 -7

index b34a05d..3da45f5 100644
@@ -1384,6 +1384,9 @@ static void s3c_hsotg_rx_data(struct s3c_hsotg *hsotg, int ep_idx, int size)
 	read_ptr = hs_req->req.actual;
 	max_req = hs_req->req.length - read_ptr;
 
+	dev_dbg(hsotg->dev, "%s: read %d/%d, done %d/%d\n",
+		__func__, to_read, max_req, read_ptr, hs_req->req.length);
+
 	if (to_read > max_req) {
 		/* more data appeared than we where willing
 		 * to deal with in this request.

@@ -1393,9 +1396,6 @@ static void s3c_hsotg_rx_data(struct s3c_hsotg *hsotg, int ep_idx, int size)
 		WARN_ON_ONCE(1);
 	}
 
-	dev_dbg(hsotg->dev, "%s: read %d/%d, done %d/%d\n",
-		__func__, to_read, max_req, read_ptr, hs_req->req.length);
-
 	hs_ep->total_data += to_read;
 	hs_req->req.actual += to_read;
 	to_read = DIV_ROUND_UP(to_read, 4);

@@ -1464,9 +1464,11 @@ static void s3c_hsotg_send_zlp(struct s3c_hsotg *hsotg,
 static void s3c_hsotg_handle_outdone(struct s3c_hsotg *hsotg,
 				     int epnum, bool was_setup)
 {
+	u32 epsize = readl(hsotg->regs + S3C_DOEPTSIZ(epnum));
 	struct s3c_hsotg_ep *hs_ep = &hsotg->eps[epnum];
 	struct s3c_hsotg_req *hs_req = hs_ep->req;
 	struct usb_request *req = &hs_req->req;
+	unsigned size_left = S3C_DxEPTSIZ_XferSize_GET(epsize);
 	int result = 0;
 
 	if (!hs_req) {

@@ -1475,9 +1477,7 @@ static void s3c_hsotg_handle_outdone(struct s3c_hsotg *hsotg,
 	}
 
 	if (using_dma(hsotg)) {
-		u32 epsize = readl(hsotg->regs + S3C_DOEPTSIZ(epnum));
 		unsigned size_done;
-		unsigned size_left;
 
 		/* Calculate the size of the transfer by checking how much
 		 * is left in the endpoint size register and then working it

@@ -1487,14 +1487,18 @@ static void s3c_hsotg_handle_outdone(struct s3c_hsotg *hsotg,
 		 * so may overshoot/undershoot the transfer.
 		 */
 
-		size_left = S3C_DxEPTSIZ_XferSize_GET(epsize);
-
 		size_done = hs_ep->size_loaded - size_left;
 		size_done += hs_ep->last_load;
 
 		req->actual = size_done;
 	}
 
+	/* if there is more request to do, schedule new transfer */
+	if (req->actual < req->length && size_left == 0) {
+		s3c_hsotg_start_req(hsotg, hs_ep, hs_req, true);
+		return;
+	}
+
 	if (req->actual < req->length && req->short_not_ok) {
 		dev_dbg(hsotg->dev, "%s: got %d/%d (short not ok) => error\n",
 			__func__, req->actual, req->length);

Signature · 1 line

1.7.0.4

Marek Szyprowski

2010-06-11 10:25 UTC #17 lore raw

In reply to Ben Dooks

Hello,

Quoted 10 lines · Ben Dooks

Some further fixes for the s3c-hsotg block, as well as a repost of the missed USB phy clock setting patch which either got missed or passed over from last time.
As a note, this series has a patch adding support for dedicated FIFO mode, this is in my view a fix, as when the hsotg block is compiled with dedicated fifos then each USB IN non-periodic pipe needs to be alloacted a unique TX FIFO. We have no actual data on how well the block performs when all IN TX NP FIFOs are all set to the same one, but it really should be fixed.

This patch series improves the driver a lot. Now it is possible to get CDC Ethernet gadget working on S5PV210 SoC (Samsung Aquila machine). However there are still some issues left to be resolved. I've noticed the following things:

1. CDC Ethernet gadget: there are problems after unplugging and plugging the usb cable again while data is being transferred. To trigger this it is enough to run 'ping host' command on the device and replug the usb cable. After that no data packets are sent do the usb host (checked with hardware usb analyser), although the ep0 control transfers are performed correctly (device has been enumerated correctly).

2. RNDIS Ethernet gadget: after some stress tests I get an error, which causes RNDIS session to hang (what is probably a direct result of the otg block reinit/reset). Here is an example log:

[   69.155000] s3c-hsotg s3c-hsotg: s3c_hsotg_handle_rx: unknown status 00000000
[   69.155000] s3c-hsotg s3c-hsotg: DCFG=0x00040020, DCTL=0x00000000,
DIEPMSK=000000af
[   69.155000] s3c-hsotg s3c-hsotg: GAHBCFG=0x00000001, 0x44=0x00000000
[   69.155000] s3c-hsotg s3c-hsotg: GRXFSIZ=0x00000800, GNPTXFSIZ=0x04000800
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[1] FSize=768, StAddr=0x00000f00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[2] FSize=768, StAddr=0x00001200
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[3] FSize=768, StAddr=0x00001500
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[4] FSize=768, StAddr=0x00001800
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[5] FSize=768, StAddr=0x00001b00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[6] FSize=768, StAddr=0x00001e00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[7] FSize=768, StAddr=0x00002100
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[8] FSize=768, StAddr=0x00002400
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[9] FSize=768, StAddr=0x00002700
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[10] FSize=768, StAddr=0x00002a00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[11] FSize=768, StAddr=0x00002d00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[12] FSize=768, StAddr=0x00003000
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[13] FSize=768, StAddr=0x00003300
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[14] FSize=768, StAddr=0x00003600
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[15] FSize=768, StAddr=0x00003900
[   69.155000] s3c-hsotg s3c-hsotg: ep0-in: EPCTL=0x00028000, SIZ=0x00000000,
DMA=0x085290ed
[   69.155000] s3c-hsotg s3c-hsotg: ep0-out: EPCTL=0x80028000, SIZ=0x60000000,
DMA=0xca625c65
[   69.155000] s3c-hsotg s3c-hsotg: ep1-in: EPCTL=0x00488200, SIZ=0x20000000,
DMA=0x5f9184b4
[   69.155000] s3c-hsotg s3c-hsotg: ep1-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x318b408d
[   69.155000] s3c-hsotg s3c-hsotg: ep2-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x50d8cf8b
[   69.155000] s3c-hsotg s3c-hsotg: ep2-out: EPCTL=0x800b8200, SIZ=0x00180800,
DMA=0x5792d85e
[   69.155000] s3c-hsotg s3c-hsotg: ep3-in: EPCTL=0x00cc8008, SIZ=0x00000000,
DMA=0x5dee1680
[   69.155000] s3c-hsotg s3c-hsotg: ep3-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0xa96c912b
[   69.155000] s3c-hsotg s3c-hsotg: ep4-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x5e84ae7a
[   69.155000] s3c-hsotg s3c-hsotg: ep4-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0xef2e9c78
[   69.155000] s3c-hsotg s3c-hsotg: ep5-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x9d8f4d72
[   69.155000] s3c-hsotg s3c-hsotg: ep5-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x93e82a21
[   69.155000] s3c-hsotg s3c-hsotg: ep6-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x202acadf
[   69.155000] s3c-hsotg s3c-hsotg: ep6-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x5520c3c9
[   69.155000] s3c-hsotg s3c-hsotg: ep7-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x782200c2
[   69.155000] s3c-hsotg s3c-hsotg: ep7-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0xeb0b8035
[   69.155000] s3c-hsotg s3c-hsotg: ep8-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x002de535
[   69.155000] s3c-hsotg s3c-hsotg: ep8-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x8c0b77f7
[   69.155000] s3c-hsotg s3c-hsotg: ep9-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xa59e32a0
[   69.155000] s3c-hsotg s3c-hsotg: ep9-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x07aa36a2
[   69.155000] s3c-hsotg s3c-hsotg: ep10-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xc6732fd5
[   69.155000] s3c-hsotg s3c-hsotg: ep10-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xfbf3f43d
[   69.155000] s3c-hsotg s3c-hsotg: ep11-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xb1d750a4
[   69.155000] s3c-hsotg s3c-hsotg: ep11-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x30024c06
[   69.155000] s3c-hsotg s3c-hsotg: ep12-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x154162af
[   69.155000] s3c-hsotg s3c-hsotg: ep12-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x60f9ef07
[   69.155000] s3c-hsotg s3c-hsotg: ep13-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xbea16bb2
[   69.155000] s3c-hsotg s3c-hsotg: ep13-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xd1723f0c
[   69.155000] s3c-hsotg s3c-hsotg: ep14-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x40d1795b
[   69.155000] s3c-hsotg s3c-hsotg: ep14-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xc0b155d6
[   69.155000] s3c-hsotg s3c-hsotg: DVBUSDIS=0x00000b8f, DVBUSPULSE=000002c6

I will try to investigate these issues further.

Best regards
--
Marek Szyprowski
Samsung Poland R&D Center

Ben Dooks

author

2010-06-12 03:34 UTC #18 lore raw

Quoted 20 lines · Marek Szyprowski

Hello,
Quoted 10 lines · Ben Dooks
Some further fixes for the s3c-hsotg block, as well as a repost of the missed USB phy clock setting patch which either got missed or passed over from last time.
As a note, this series has a patch adding support for dedicated FIFO mode, this is in my view a fix, as when the hsotg block is compiled with dedicated fifos then each USB IN non-periodic pipe needs to be alloacted a unique TX FIFO. We have no actual data on how well the block performs when all IN TX NP FIFOs are all set to the same one, but it really should be fixed.
This patch series improves the driver a lot. Now it is possible to get CDC Ethernet gadget working on S5PV210 SoC (Samsung Aquila machine). However there are still some issues left to be resolved. I've noticed the following things:

Quoted 6 lines

1. CDC Ethernet gadget: there are problems after unplugging and plugging the usb cable again while data is being transferred. To trigger this it is enough to run 'ping host' command on the device and replug the usb cable. After that no data packets are sent do the usb host (checked with hardware usb analyser), although the ep0 control transfers are performed correctly (device has been enumerated correctly).

Hmm, that is intereting. Maybe we broke non-ep0 fifos somehow

Quoted 3 lines

2. RNDIS Ethernet gadget: after some stress tests I get an error, which causes RNDIS session to hang (what is probably a direct result of the otg block reinit/reset). Here is an example log:

any particular kind of stress tests, or are you just telling it nasty things to upset it? will try and look into this too.

Quoted 90 lines

[   69.155000] s3c-hsotg s3c-hsotg: s3c_hsotg_handle_rx: unknown status 00000000
[   69.155000] s3c-hsotg s3c-hsotg: DCFG=0x00040020, DCTL=0x00000000,
DIEPMSK=000000af
[   69.155000] s3c-hsotg s3c-hsotg: GAHBCFG=0x00000001, 0x44=0x00000000
[   69.155000] s3c-hsotg s3c-hsotg: GRXFSIZ=0x00000800, GNPTXFSIZ=0x04000800
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[1] FSize=768, StAddr=0x00000f00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[2] FSize=768, StAddr=0x00001200
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[3] FSize=768, StAddr=0x00001500
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[4] FSize=768, StAddr=0x00001800
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[5] FSize=768, StAddr=0x00001b00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[6] FSize=768, StAddr=0x00001e00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[7] FSize=768, StAddr=0x00002100
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[8] FSize=768, StAddr=0x00002400
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[9] FSize=768, StAddr=0x00002700
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[10] FSize=768, StAddr=0x00002a00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[11] FSize=768, StAddr=0x00002d00
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[12] FSize=768, StAddr=0x00003000
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[13] FSize=768, StAddr=0x00003300
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[14] FSize=768, StAddr=0x00003600
[   69.155000] s3c-hsotg s3c-hsotg: DPTx[15] FSize=768, StAddr=0x00003900
[   69.155000] s3c-hsotg s3c-hsotg: ep0-in: EPCTL=0x00028000, SIZ=0x00000000,
DMA=0x085290ed
[   69.155000] s3c-hsotg s3c-hsotg: ep0-out: EPCTL=0x80028000, SIZ=0x60000000,
DMA=0xca625c65
[   69.155000] s3c-hsotg s3c-hsotg: ep1-in: EPCTL=0x00488200, SIZ=0x20000000,
DMA=0x5f9184b4
[   69.155000] s3c-hsotg s3c-hsotg: ep1-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x318b408d
[   69.155000] s3c-hsotg s3c-hsotg: ep2-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x50d8cf8b
[   69.155000] s3c-hsotg s3c-hsotg: ep2-out: EPCTL=0x800b8200, SIZ=0x00180800,
DMA=0x5792d85e
[   69.155000] s3c-hsotg s3c-hsotg: ep3-in: EPCTL=0x00cc8008, SIZ=0x00000000,
DMA=0x5dee1680
[   69.155000] s3c-hsotg s3c-hsotg: ep3-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0xa96c912b
[   69.155000] s3c-hsotg s3c-hsotg: ep4-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x5e84ae7a
[   69.155000] s3c-hsotg s3c-hsotg: ep4-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0xef2e9c78
[   69.155000] s3c-hsotg s3c-hsotg: ep5-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x9d8f4d72
[   69.155000] s3c-hsotg s3c-hsotg: ep5-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x93e82a21
[   69.155000] s3c-hsotg s3c-hsotg: ep6-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x202acadf
[   69.155000] s3c-hsotg s3c-hsotg: ep6-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x5520c3c9
[   69.155000] s3c-hsotg s3c-hsotg: ep7-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x782200c2
[   69.155000] s3c-hsotg s3c-hsotg: ep7-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0xeb0b8035
[   69.155000] s3c-hsotg s3c-hsotg: ep8-in: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x002de535
[   69.155000] s3c-hsotg s3c-hsotg: ep8-out: EPCTL=0x00000200, SIZ=0x00000000,
DMA=0x8c0b77f7
[   69.155000] s3c-hsotg s3c-hsotg: ep9-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xa59e32a0
[   69.155000] s3c-hsotg s3c-hsotg: ep9-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x07aa36a2
[   69.155000] s3c-hsotg s3c-hsotg: ep10-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xc6732fd5
[   69.155000] s3c-hsotg s3c-hsotg: ep10-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xfbf3f43d
[   69.155000] s3c-hsotg s3c-hsotg: ep11-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xb1d750a4
[   69.155000] s3c-hsotg s3c-hsotg: ep11-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x30024c06
[   69.155000] s3c-hsotg s3c-hsotg: ep12-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x154162af
[   69.155000] s3c-hsotg s3c-hsotg: ep12-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x60f9ef07
[   69.155000] s3c-hsotg s3c-hsotg: ep13-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xbea16bb2
[   69.155000] s3c-hsotg s3c-hsotg: ep13-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xd1723f0c
[   69.155000] s3c-hsotg s3c-hsotg: ep14-in: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0x40d1795b
[   69.155000] s3c-hsotg s3c-hsotg: ep14-out: EPCTL=0x00000000, SIZ=0x00000000,
DMA=0xc0b155d6
[   69.155000] s3c-hsotg s3c-hsotg: DVBUSDIS=0x00000b8f, DVBUSPULSE=000002c6

I will try to investigate these issues further.

Best regards
--
Marek Szyprowski
Samsung Poland R&D Center

Signature · 5 lines

-- 
Ben

Q:      What's a light-year?
A:      One-third less calories than a regular year.

Marek Szyprowski

2010-06-14 05:40 UTC #19 lore raw

Hello,

Quoted 28 lines · Ben Dooks

Quoted 20 lines · Marek Szyprowski
Hello,
Quoted 10 lines · Ben Dooks
Some further fixes for the s3c-hsotg block, as well as a repost of the missed USB phy clock setting patch which either got missed or passed over from last time.
As a note, this series has a patch adding support for dedicated FIFO mode, this is in my view a fix, as when the hsotg block is compiled with dedicated fifos then each USB IN non-periodic pipe needs to be alloacted a unique TX FIFO. We have no actual data on how well the block performs when all IN TX NP FIFOs are all set to the same one, but it really should be fixed.
This patch series improves the driver a lot. Now it is possible to get CDC Ethernet gadget working on S5PV210 SoC (Samsung Aquila machine). However there are still some issues left to be resolved. I've noticed the following things:
Quoted 6 lines
1. CDC Ethernet gadget: there are problems after unplugging and plugging the usb cable again while data is being transferred. To trigger this it is enough to run 'ping host' command on the device and replug the usb cable. After that no data packets are sent do the usb host (checked with hardware usb analyser), although the ep0 control transfers are performed correctly (device has been enumerated correctly).

Quoted 8 lines

Hmm, that is intereting. Maybe we broke non-ep0 fifos somehow
Quoted 3 lines
2. RNDIS Ethernet gadget: after some stress tests I get an error, which causes RNDIS session to hang (what is probably a direct result of the otg block reinit/reset). Here is an example log:
any particular kind of stress tests, or are you just telling it nasty things to upset it? will try and look into this too.

I've tested it with Windows XP SP3 host (Pentium4 2.8GHz, VIA USB 2.0 EHCI controller) and the following script:

--->8--[killer.bat]---
:foo
ping 192.168.129.3
goto foo
--->8-----------------

192.168.129.3 is an IP address of the client system (Linux on S5PV210).

Four instances of this script was running simultaneously and a Task Manager with Network Status tab was opened. Each time when Windows runs ping command the RNDIS transaction is performed. So this test basically checked how the driver would act on a flood of RNDIS transactions. Each NDIS transactions consists of the following transfers: RNDIS Command (ep0-control in), RNDIS notification (ep3-interrupt in) and RNDIS Response (ep0-control out). All 3 parts of RNDIS transaction must be performed correctly, otherwise Windows host locks (it is quite easy to notice this, Network Status graph stops scrolling). The test took about 5 minutes to trigger the issue.

The flood of RNDIS transactions happens also during normal usage, some firewalls or anti-virus scanning tools also can cause it, so it shouldn't be considered something really unusual.

Best regards
--
Marek Szyprowski
Samsung Poland R&D Center