Merge branch 'master' of git://git.denx.de/u-boot-usb

* 'master' of git://git.denx.de/u-boot-usb:
  usb_storage: fix ehci driver max transfer size
  smsc95xx: align buffers to cache line size
  ehci-hcd: change debug() to printf() in case of errors
  usb: check return value of submit_{control, bulk}_msg
  usb: pass cache-aligned buffer to usb_get_descriptor()
  ehci-hcd: fix external buffer cache handling
  ehci-hcd.c, musb_core, usb.h: Add USB_DMA_MINALIGN define for cache alignment
  ehci-hcd: program asynclistaddr before every transfer
  common.h: Introduce DEFINE_CACHE_ALIGN_BUFFER
  ehci-omap: Do not call dcache_off from omap_ehci_hcd_init

Signed-off-by: Wolfgang Denk <wd@denx.de>

4 files changed, 61 insertions, 51 deletions

diff --git a/drivers/usb/eth/smsc95xx.c b/drivers/usb/eth/smsc95xx.c
index c7aebea4e..c62a8c112 100644
--- a/drivers/usb/eth/smsc95xx.c
+++ b/drivers/usb/eth/smsc95xx.c
@@ -153,13 +153,15 @@ static int curr_eth_dev; /* index for name of next device detected */
 static int smsc95xx_write_reg(struct ueth_data *dev, u32 index, u32 data)
 {
 	int len;
+	ALLOC_CACHE_ALIGN_BUFFER(u32, tmpbuf, 1);
 
 	cpu_to_le32s(&data);
+	tmpbuf[0] = data;
 
 	len = usb_control_msg(dev->pusb_dev, usb_sndctrlpipe(dev->pusb_dev, 0),
 		USB_VENDOR_REQUEST_WRITE_REGISTER,
 		USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
-		00, index, &data, sizeof(data), USB_CTRL_SET_TIMEOUT);
+		00, index, tmpbuf, sizeof(data), USB_CTRL_SET_TIMEOUT);
 	if (len != sizeof(data)) {
 		debug("smsc95xx_write_reg failed: index=%d, data=%d, len=%d",
 		      index, data, len);
@@ -171,11 +173,13 @@ static int smsc95xx_write_reg(struct ueth_data *dev, u32 index, u32 data)
 static int smsc95xx_read_reg(struct ueth_data *dev, u32 index, u32 *data)
 {
 	int len;
+	ALLOC_CACHE_ALIGN_BUFFER(u32, tmpbuf, 1);
 
 	len = usb_control_msg(dev->pusb_dev, usb_rcvctrlpipe(dev->pusb_dev, 0),
 		USB_VENDOR_REQUEST_READ_REGISTER,
 		USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
-		00, index, data, sizeof(data), USB_CTRL_GET_TIMEOUT);
+		00, index, tmpbuf, sizeof(data), USB_CTRL_GET_TIMEOUT);
+	*data = tmpbuf[0];
 	if (len != sizeof(data)) {
 		debug("smsc95xx_read_reg failed: index=%d, len=%d",
 		      index, len);
@@ -664,7 +668,8 @@ static int smsc95xx_send(struct eth_device *eth, void* packet, int length)
 	int actual_len;
 	u32 tx_cmd_a;
 	u32 tx_cmd_b;
-	unsigned char msg[PKTSIZE + sizeof(tx_cmd_a) + sizeof(tx_cmd_b)];
+	ALLOC_CACHE_ALIGN_BUFFER(unsigned char, msg,
+				 PKTSIZE + sizeof(tx_cmd_a) + sizeof(tx_cmd_b));
 
 	debug("** %s(), len %d, buf %#x\n", __func__, length, (int)msg);
 	if (length > PKTSIZE)
@@ -695,7 +700,7 @@ static int smsc95xx_send(struct eth_device *eth, void* packet, int length)
 static int smsc95xx_recv(struct eth_device *eth)
 {
 	struct ueth_data *dev = (struct ueth_data *)eth->priv;
-	static unsigned char  recv_buf[AX_RX_URB_SIZE];
+	DEFINE_CACHE_ALIGN_BUFFER(unsigned char, recv_buf, AX_RX_URB_SIZE);
 	unsigned char *buf_ptr;
 	int err;
 	int actual_len;
diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c
index 04300be11..2a82a2912 100644
--- a/drivers/usb/host/ehci-hcd.c
+++ b/drivers/usb/host/ehci-hcd.c
@@ -34,7 +34,10 @@ struct ehci_hccr *hccr;	/* R/O registers, not need for volatile */
 volatile struct ehci_hcor *hcor;
 
 static uint16_t portreset;
-static struct QH qh_list __attribute__((aligned(32)));
+DEFINE_ALIGN_BUFFER(struct QH, qh_list, 1, USB_DMA_MINALIGN);
+
+#define ALIGN_END_ADDR(type, ptr, size)			\
+	((uint32_t)(ptr) + roundup((size) * sizeof(type), USB_DMA_MINALIGN))
 
 static struct descriptor {
 	struct usb_hub_descriptor hub;
@@ -172,18 +175,15 @@ static int ehci_td_buffer(struct qTD *td, void *buf, size_t sz)
 {
 	uint32_t delta, next;
 	uint32_t addr = (uint32_t)buf;
-	size_t rsz = roundup(sz, 32);
 	int idx;
 
-	if (sz != rsz)
-		debug("EHCI-HCD: Misaligned buffer size (%08x)\n", sz);
-
-	if (addr & 31)
+	if (addr != ALIGN(addr, ARCH_DMA_MINALIGN))
 		debug("EHCI-HCD: Misaligned buffer address (%p)\n", buf);
 
+	flush_dcache_range(addr, ALIGN(addr + sz, ARCH_DMA_MINALIGN));
+
 	idx = 0;
 	while (idx < 5) {
-		flush_dcache_range(addr, addr + rsz);
 		td->qt_buffer[idx] = cpu_to_hc32(addr);
 		td->qt_buffer_hi[idx] = 0;
 		next = (addr + 4096) & ~4095;
@@ -196,7 +196,7 @@ static int ehci_td_buffer(struct qTD *td, void *buf, size_t sz)
 	}
 
 	if (idx == 5) {
-		debug("out of buffer pointers (%u bytes left)\n", sz);
+		printf("out of buffer pointers (%u bytes left)\n", sz);
 		return -1;
 	}
 
@@ -207,8 +207,8 @@ static int
 ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 		   int length, struct devrequest *req)
 {
-	static struct QH qh __attribute__((aligned(32)));
-	static struct qTD qtd[3] __attribute__((aligned (32)));
+	ALLOC_ALIGN_BUFFER(struct QH, qh, 1, USB_DMA_MINALIGN);
+	ALLOC_ALIGN_BUFFER(struct qTD, qtd, 3, USB_DMA_MINALIGN);
 	int qtd_counter = 0;
 
 	volatile struct qTD *vtd;
@@ -229,8 +229,8 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 		      le16_to_cpu(req->value), le16_to_cpu(req->value),
 		      le16_to_cpu(req->index));
 
-	memset(&qh, 0, sizeof(struct QH));
-	memset(qtd, 0, sizeof(qtd));
+	memset(qh, 0, sizeof(struct QH));
+	memset(qtd, 0, 3 * sizeof(*qtd));
 
 	toggle = usb_gettoggle(dev, usb_pipeendpoint(pipe), usb_pipeout(pipe));
 
@@ -244,7 +244,7 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 	 *   qh_overlay.qt_next ...... 13-10 H
 	 * - qh_overlay.qt_altnext
 	 */
-	qh.qh_link = cpu_to_hc32((uint32_t)&qh_list | QH_LINK_TYPE_QH);
+	qh->qh_link = cpu_to_hc32((uint32_t)qh_list | QH_LINK_TYPE_QH);
 	c = (usb_pipespeed(pipe) != USB_SPEED_HIGH &&
 	     usb_pipeendpoint(pipe) == 0) ? 1 : 0;
 	endpt = (8 << 28) |
@@ -255,14 +255,14 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 	    (usb_pipespeed(pipe) << 12) |
 	    (usb_pipeendpoint(pipe) << 8) |
 	    (0 << 7) | (usb_pipedevice(pipe) << 0);
-	qh.qh_endpt1 = cpu_to_hc32(endpt);
+	qh->qh_endpt1 = cpu_to_hc32(endpt);
 	endpt = (1 << 30) |
 	    (dev->portnr << 23) |
 	    (dev->parent->devnum << 16) | (0 << 8) | (0 << 0);
-	qh.qh_endpt2 = cpu_to_hc32(endpt);
-	qh.qh_overlay.qt_next = cpu_to_hc32(QT_NEXT_TERMINATE);
+	qh->qh_endpt2 = cpu_to_hc32(endpt);
+	qh->qh_overlay.qt_next = cpu_to_hc32(QT_NEXT_TERMINATE);
 
-	tdp = &qh.qh_overlay.qt_next;
+	tdp = &qh->qh_overlay.qt_next;
 
 	if (req != NULL) {
 		/*
@@ -281,7 +281,7 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 		    (0 << 15) | (0 << 12) | (3 << 10) | (2 << 8) | (0x80 << 0);
 		qtd[qtd_counter].qt_token = cpu_to_hc32(token);
 		if (ehci_td_buffer(&qtd[qtd_counter], req, sizeof(*req)) != 0) {
-			debug("unable construct SETUP td\n");
+			printf("unable construct SETUP td\n");
 			goto fail;
 		}
 		/* Update previous qTD! */
@@ -310,7 +310,7 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 		    ((usb_pipein(pipe) ? 1 : 0) << 8) | (0x80 << 0);
 		qtd[qtd_counter].qt_token = cpu_to_hc32(token);
 		if (ehci_td_buffer(&qtd[qtd_counter], buffer, length) != 0) {
-			debug("unable construct DATA td\n");
+			printf("unable construct DATA td\n");
 			goto fail;
 		}
 		/* Update previous qTD! */
@@ -340,13 +340,16 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 		tdp = &qtd[qtd_counter++].qt_next;
 	}
 
-	qh_list.qh_link = cpu_to_hc32((uint32_t)&qh | QH_LINK_TYPE_QH);
+	qh_list->qh_link = cpu_to_hc32((uint32_t)qh | QH_LINK_TYPE_QH);
 
 	/* Flush dcache */
-	flush_dcache_range((uint32_t)&qh_list,
-		(uint32_t)&qh_list + sizeof(struct QH));
-	flush_dcache_range((uint32_t)&qh, (uint32_t)&qh + sizeof(struct QH));
-	flush_dcache_range((uint32_t)qtd, (uint32_t)qtd + sizeof(qtd));
+	flush_dcache_range((uint32_t)qh_list,
+		ALIGN_END_ADDR(struct QH, qh_list, 1));
+	flush_dcache_range((uint32_t)qh, ALIGN_END_ADDR(struct QH, qh, 1));
+	flush_dcache_range((uint32_t)qtd, ALIGN_END_ADDR(struct qTD, qtd, 3));
+
+	/* Set async. queue head pointer. */
+	ehci_writel(&hcor->or_asynclistaddr, (uint32_t)qh_list);
 
 	usbsts = ehci_readl(&hcor->or_usbsts);
 	ehci_writel(&hcor->or_usbsts, (usbsts & 0x3f));
@@ -369,12 +372,12 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 	timeout = USB_TIMEOUT_MS(pipe);
 	do {
 		/* Invalidate dcache */
-		invalidate_dcache_range((uint32_t)&qh_list,
-			(uint32_t)&qh_list + sizeof(struct QH));
-		invalidate_dcache_range((uint32_t)&qh,
-			(uint32_t)&qh + sizeof(struct QH));
+		invalidate_dcache_range((uint32_t)qh_list,
+			ALIGN_END_ADDR(struct QH, qh_list, 1));
+		invalidate_dcache_range((uint32_t)qh,
+			ALIGN_END_ADDR(struct QH, qh, 1));
 		invalidate_dcache_range((uint32_t)qtd,
-			(uint32_t)qtd + sizeof(qtd));
+			ALIGN_END_ADDR(struct qTD, qtd, 3));
 
 		token = hc32_to_cpu(vtd->qt_token);
 		if (!(token & 0x80))
@@ -382,9 +385,17 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 		WATCHDOG_RESET();
 	} while (get_timer(ts) < timeout);
 
-	/* Invalidate the memory area occupied by buffer */
-	invalidate_dcache_range(((uint32_t)buffer & ~31),
-		((uint32_t)buffer & ~31) + roundup(length, 32));
+	/*
+	 * Invalidate the memory area occupied by buffer
+	 * Don't try to fix the buffer alignment, if it isn't properly
+	 * aligned it's upper layer's fault so let invalidate_dcache_range()
+	 * vow about it. But we have to fix the length as it's actual
+	 * transfer length and can be unaligned. This is potentially
+	 * dangerous operation, it's responsibility of the calling
+	 * code to make sure enough space is reserved.
+	 */
+	invalidate_dcache_range((uint32_t)buffer,
+		ALIGN((uint32_t)buffer + length, ARCH_DMA_MINALIGN));
 
 	/* Check that the TD processing happened */
 	if (token & 0x80) {
@@ -403,9 +414,7 @@ ehci_submit_async(struct usb_device *dev, unsigned long pipe, void *buffer,
 		goto fail;
 	}
 
-	qh_list.qh_link = cpu_to_hc32((uint32_t)&qh_list | QH_LINK_TYPE_QH);
-
-	token = hc32_to_cpu(qh.qh_overlay.qt_token);
+	token = hc32_to_cpu(qh->qh_overlay.qt_token);
 	if (!(token & 0x80)) {
 		debug("TOKEN=%#x\n", token);
 		switch (token & 0xfc) {
@@ -733,16 +742,13 @@ int usb_lowlevel_init(void)
 #endif
 
 	/* Set head of reclaim list */
-	memset(&qh_list, 0, sizeof(qh_list));
-	qh_list.qh_link = cpu_to_hc32((uint32_t)&qh_list | QH_LINK_TYPE_QH);
-	qh_list.qh_endpt1 = cpu_to_hc32((1 << 15) | (USB_SPEED_HIGH << 12));
-	qh_list.qh_curtd = cpu_to_hc32(QT_NEXT_TERMINATE);
-	qh_list.qh_overlay.qt_next = cpu_to_hc32(QT_NEXT_TERMINATE);
-	qh_list.qh_overlay.qt_altnext = cpu_to_hc32(QT_NEXT_TERMINATE);
-	qh_list.qh_overlay.qt_token = cpu_to_hc32(0x40);
-
-	/* Set async. queue head pointer. */
-	ehci_writel(&hcor->or_asynclistaddr, (uint32_t)&qh_list);
+	memset(qh_list, 0, sizeof(*qh_list));
+	qh_list->qh_link = cpu_to_hc32((uint32_t)qh_list | QH_LINK_TYPE_QH);
+	qh_list->qh_endpt1 = cpu_to_hc32((1 << 15) | (USB_SPEED_HIGH << 12));
+	qh_list->qh_curtd = cpu_to_hc32(QT_NEXT_TERMINATE);
+	qh_list->qh_overlay.qt_next = cpu_to_hc32(QT_NEXT_TERMINATE);
+	qh_list->qh_overlay.qt_altnext = cpu_to_hc32(QT_NEXT_TERMINATE);
+	qh_list->qh_overlay.qt_token = cpu_to_hc32(0x40);
 
 	reg = ehci_readl(&hccr->cr_hcsparams);
 	descriptor.hub.bNbrPorts = HCS_N_PORTS(reg);
diff --git a/drivers/usb/host/ehci-omap.c b/drivers/usb/host/ehci-omap.c
index 1ed77107e..292673bf0 100644
--- a/drivers/usb/host/ehci-omap.c
+++ b/drivers/usb/host/ehci-omap.c
@@ -246,7 +246,6 @@ int omap_ehci_hcd_init(struct omap_usbhs_board_data *usbhs_pdata)
 		if (is_ehci_phy_mode(usbhs_pdata->port_mode[i]))
 			omap_ehci_soft_phy_reset(i);
 
-	dcache_disable();
 	hccr = (struct ehci_hccr *)(OMAP_EHCI_BASE);
 	hcor = (struct ehci_hcor *)(OMAP_EHCI_BASE + 0x10);
 
diff --git a/drivers/usb/musb/musb_core.h b/drivers/usb/musb/musb_core.h
index a8adcce00..e91436929 100644
--- a/drivers/usb/musb/musb_core.h
+++ b/drivers/usb/musb/musb_core.h
@@ -145,7 +145,7 @@ struct musb_regs {
 		struct musb_epN_regs epN;
 	} ep[16];
 
-} __attribute__((packed, aligned(32)));
+} __attribute__((packed, aligned(USB_DMA_MINALIGN)));
 #endif
 
 /*