linux-omap.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Felipe Balbi <balbi@ti.com>
To: Vikram Pandita <vikram.pandita@ti.com>
Cc: balbi@ti.com, linux-usb@vger.kernel.org, gadiyar@ti.com,
	linux-omap@vger.kernel.org, Moiz Sonasath <m-sonasath@ti.com>
Subject: Re: [PATCH v4] usb: musb: Enable DMA mode1 RX for USB-Mass-Storage
Date: Fri, 29 Jul 2011 07:45:10 +0300	[thread overview]
Message-ID: <20110729044508.GY9069@legolas.emea.dhcp.ti.com> (raw)
In-Reply-To: <1311138718-5072-1-git-send-email-vikram.pandita@ti.com>

[-- Attachment #1: Type: text/plain, Size: 4800 bytes --]

Hi,

On Tue, Jul 19, 2011 at 10:11:58PM -0700, Vikram Pandita wrote:
> From: Anand Gadiyar <gadiyar@ti.com>
> 
> This patch enables the DMA mode1 RX support.
> This feature is enabled based on the short_not_ok flag passed from
> gadget drivers.
> 
> This will result in a thruput performance gain of around
> 40% for USB mass-storage/mtp use cases.
> 
> Signed-off-by: Anand Gadiyar <gadiyar@ti.com>
> Signed-off-by: Moiz Sonasath <m-sonasath@ti.com>
> Signed-off-by: Vikram Pandita <vikram.pandita@ti.com>
> Tested-by: Vikram Pandita <vikram.pandita@ti.com>

applied this one, but I changed commit log and code comment a little
bit. Here's updated commit:

commit e9c281b174f188adb7950ea8f6a55ca07be69914
Author: Anand Gadiyar <gadiyar@ti.com>
Date:   Tue Jul 19 22:11:58 2011 -0700

    usb: musb: Enable DMA mode1 RX for transfers without short packets
    
    This patch enables DMA mode1 for the RX patch when we know
    there won't be any short packets. We check that by looking
    into the short_no_ok flag, if it's true we enable mode1, otherwise
    we use mode0 to transfer the data.
    
    This will result in a throughput performance gain of around
    40% for USB mass-storage/mtp use cases.
    
    [ balbi@ti.com : updated commit log and code comments slightly ]
    
    Signed-off-by: Anand Gadiyar <gadiyar@ti.com>
    Signed-off-by: Moiz Sonasath <m-sonasath@ti.com>
    Signed-off-by: Vikram Pandita <vikram.pandita@ti.com>
    Tested-by: Vikram Pandita <vikram.pandita@ti.com>
    Signed-off-by: Felipe Balbi <balbi@ti.com>

diff --git a/drivers/usb/musb/musb_gadget.c b/drivers/usb/musb/musb_gadget.c
index b67a062..d314f58 100644
--- a/drivers/usb/musb/musb_gadget.c
+++ b/drivers/usb/musb/musb_gadget.c
@@ -634,6 +634,7 @@ static void rxstate(struct musb *musb, struct musb_request *req)
 	u16			len;
 	u16			csr = musb_readw(epio, MUSB_RXCSR);
 	struct musb_hw_ep	*hw_ep = &musb->endpoints[epnum];
+	u8			use_mode_1;
 
 	if (hw_ep->is_shared_fifo)
 		musb_ep = &hw_ep->ep_in;
@@ -683,6 +684,18 @@ static void rxstate(struct musb *musb, struct musb_request *req)
 
 	if (csr & MUSB_RXCSR_RXPKTRDY) {
 		len = musb_readw(epio, MUSB_RXCOUNT);
+
+		/*
+		 * Enable Mode 1 on RX transfers only when short_not_ok flag
+		 * is set. Currently short_not_ok flag is set only from
+		 * file_storage and f_mass_storage drivers
+		 */
+
+		if (request->short_not_ok && len == musb_ep->packet_sz)
+			use_mode_1 = 1;
+		else
+			use_mode_1 = 0;
+
 		if (request->actual < request->length) {
 #ifdef CONFIG_USB_INVENTRA_DMA
 			if (is_buffer_mapped(req)) {
@@ -714,37 +727,41 @@ static void rxstate(struct musb *musb, struct musb_request *req)
 	 * then becomes usable as a runtime "use mode 1" hint...
 	 */
 
-				csr |= MUSB_RXCSR_DMAENAB;
-#ifdef USE_MODE1
-				csr |= MUSB_RXCSR_AUTOCLEAR;
-				/* csr |= MUSB_RXCSR_DMAMODE; */
-
-				/* this special sequence (enabling and then
-				 * disabling MUSB_RXCSR_DMAMODE) is required
-				 * to get DMAReq to activate
-				 */
-				musb_writew(epio, MUSB_RXCSR,
-					csr | MUSB_RXCSR_DMAMODE);
-#else
-				if (!musb_ep->hb_mult &&
-					musb_ep->hw_ep->rx_double_buffered)
+				/* Experimental: Mode1 works with mass storage use cases */
+				if (use_mode_1) {
 					csr |= MUSB_RXCSR_AUTOCLEAR;
-#endif
-				musb_writew(epio, MUSB_RXCSR, csr);
+					musb_writew(epio, MUSB_RXCSR, csr);
+					csr |= MUSB_RXCSR_DMAENAB;
+					musb_writew(epio, MUSB_RXCSR, csr);
+
+					/*
+					 * this special sequence (enabling and then
+					 * disabling MUSB_RXCSR_DMAMODE) is required
+					 * to get DMAReq to activate
+					 */
+					musb_writew(epio, MUSB_RXCSR,
+						csr | MUSB_RXCSR_DMAMODE);
+					musb_writew(epio, MUSB_RXCSR, csr);
+
+				} else {
+					if (!musb_ep->hb_mult &&
+						musb_ep->hw_ep->rx_double_buffered)
+						csr |= MUSB_RXCSR_AUTOCLEAR;
+					csr |= MUSB_RXCSR_DMAENAB;
+					musb_writew(epio, MUSB_RXCSR, csr);
+				}
 
 				if (request->actual < request->length) {
 					int transfer_size = 0;
-#ifdef USE_MODE1
-					transfer_size = min(request->length - request->actual,
-							channel->max_len);
-#else
-					transfer_size = min(request->length - request->actual,
-							(unsigned)len);
-#endif
-					if (transfer_size <= musb_ep->packet_sz)
-						musb_ep->dma->desired_mode = 0;
-					else
+					if (use_mode_1) {
+						transfer_size = min(request->length - request->actual,
+								channel->max_len);
 						musb_ep->dma->desired_mode = 1;
+					} else {
+						transfer_size = min(request->length - request->actual,
+								(unsigned)len);
+						musb_ep->dma->desired_mode = 0;
+					}
 
 					use_dma = c->channel_program(
 							channel,

-- 
balbi

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 490 bytes --]

  parent reply	other threads:[~2011-07-29  4:45 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-07-20  5:11 [PATCH v4] usb: musb: Enable DMA mode1 RX for USB-Mass-Storage Vikram Pandita
2011-07-20  5:34 ` Jassi Brar
2011-07-20  5:45   ` Pandita, Vikram
2011-07-20  5:53     ` Jassi Brar
2011-07-20 16:48       ` Pandita, Vikram
2011-07-26 15:06         ` Felipe Balbi
     [not found]           ` <20110726150655.GK32582-UiBtZHVXSwEVvW8u9ZQWYwjfymiNCTlR@public.gmane.org>
2011-07-27  3:50             ` Jassi Brar
2011-07-27 11:30               ` Felipe Balbi
2011-07-29  4:45 ` Felipe Balbi [this message]
2011-07-29  4:50   ` Pandita, Vikram

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110729044508.GY9069@legolas.emea.dhcp.ti.com \
    --to=balbi@ti.com \
    --cc=gadiyar@ti.com \
    --cc=linux-omap@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=m-sonasath@ti.com \
    --cc=vikram.pandita@ti.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).