Commit 4f3e8d26 authored by Roger Quadros's avatar Roger Quadros Committed by Felipe Balbi

usb: musb: use DMA mode 1 whenever possible

Do not rely on any hints from gadget drivers and use DMA mode 1
whenever we expect data of at least the endpoint's packet size and
have not yet received a short packet.

The last packet if short is always transferred using DMA mode 0.

This patch fixes USB throughput issues in mass storage mode for
host to device transfers.
Signed-off-by: default avatarRoger Quadros <rogerq@ti.com>
Signed-off-by: default avatarFelipe Balbi <balbi@ti.com>
parent 8e8a5516
...@@ -707,12 +707,11 @@ static void rxstate(struct musb *musb, struct musb_request *req) ...@@ -707,12 +707,11 @@ static void rxstate(struct musb *musb, struct musb_request *req)
fifo_count = musb_readw(epio, MUSB_RXCOUNT); fifo_count = musb_readw(epio, MUSB_RXCOUNT);
/* /*
* Enable Mode 1 on RX transfers only when short_not_ok flag * use mode 1 only if we expect data of at least ep packet_sz
* is set. Currently short_not_ok flag is set only from * and have not yet received a short packet
* file_storage and f_mass_storage drivers
*/ */
if ((request->length - request->actual >= musb_ep->packet_sz) &&
if (request->short_not_ok && fifo_count == musb_ep->packet_sz) (fifo_count >= musb_ep->packet_sz))
use_mode_1 = 1; use_mode_1 = 1;
else else
use_mode_1 = 0; use_mode_1 = 0;
...@@ -727,27 +726,6 @@ static void rxstate(struct musb *musb, struct musb_request *req) ...@@ -727,27 +726,6 @@ static void rxstate(struct musb *musb, struct musb_request *req)
c = musb->dma_controller; c = musb->dma_controller;
channel = musb_ep->dma; channel = musb_ep->dma;
/* We use DMA Req mode 0 in rx_csr, and DMA controller operates in
* mode 0 only. So we do not get endpoint interrupts due to DMA
* completion. We only get interrupts from DMA controller.
*
* We could operate in DMA mode 1 if we knew the size of the tranfer
* in advance. For mass storage class, request->length = what the host
* sends, so that'd work. But for pretty much everything else,
* request->length is routinely more than what the host sends. For
* most these gadgets, end of is signified either by a short packet,
* or filling the last byte of the buffer. (Sending extra data in
* that last pckate should trigger an overflow fault.) But in mode 1,
* we don't get DMA completion interrupt for short packets.
*
* Theoretically, we could enable DMAReq irq (MUSB_RXCSR_DMAMODE = 1),
* to get endpoint interrupt on every DMA req, but that didn't seem
* to work reliably.
*
* REVISIT an updated g_file_storage can set req->short_not_ok, which
* then becomes usable as a runtime "use mode 1" hint...
*/
/* Experimental: Mode1 works with mass storage use cases */ /* Experimental: Mode1 works with mass storage use cases */
if (use_mode_1) { if (use_mode_1) {
csr |= MUSB_RXCSR_AUTOCLEAR; csr |= MUSB_RXCSR_AUTOCLEAR;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment