From: Ahmad Fatoum <a.fatoum@pengutronix.de>
To: barebox@lists.infradead.org
Cc: lst@pengutronix.de, Ahmad Fatoum <a.fatoum@pengutronix.de>
Subject: [PATCH RFC 02/12] usb: xhci: pass physical device to DMA API
Date: Tue, 21 Feb 2023 09:05:14 +0100 [thread overview]
Message-ID: <20230221080524.607241-3-a.fatoum@pengutronix.de> (raw)
In-Reply-To: <20230221080524.607241-1-a.fatoum@pengutronix.de>
The xHCI device is just a child of the DWC3 with no DT node assigned.
As such, it lacks all DMA settings that may be set in the DT.
Fix this by using the hardware device instead in case the xHCI lacks
its own DT node.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
---
drivers/usb/host/xhci-ring.c | 8 ++++----
drivers/usb/host/xhci.c | 6 +++++-
2 files changed, 9 insertions(+), 5 deletions(-)
diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c
index 60764222af3d..0abe8a67a392 100644
--- a/drivers/usb/host/xhci-ring.c
+++ b/drivers/usb/host/xhci-ring.c
@@ -594,7 +594,7 @@ int xhci_bulk_tx(struct usb_device *udev, unsigned long pipe,
memcpy(bounce, buffer, length);
}
- map = addr = dma_map_single(ctrl->dev, bounce, length, direction);
+ map = addr = dma_map_single(ctrl->host.hw_dev, bounce, length, direction);
dev_dbg(&udev->dev, "pipe=0x%lx, buffer=%p, length=%d\n",
pipe, buffer, length);
@@ -740,7 +740,7 @@ int xhci_bulk_tx(struct usb_device *udev, unsigned long pipe,
record_transfer_result(udev, event, length);
xhci_acknowledge_event(ctrl);
- dma_unmap_single(ctrl->dev, map, length, direction);
+ dma_unmap_single(ctrl->host.hw_dev, map, length, direction);
if (usb_pipein(pipe))
memcpy(buffer, bounce, length);
@@ -895,7 +895,7 @@ int xhci_ctrl_tx(struct usb_device *udev, unsigned long pipe,
if (length > 0) {
if (req->requesttype & USB_DIR_IN)
field |= TRB_DIR_IN;
- map = buf_64 = dma_map_single(ctrl->dev, buffer, length, direction);
+ map = buf_64 = dma_map_single(ctrl->host.hw_dev, buffer, length, direction);
trb_fields[0] = lower_32_bits(buf_64);
trb_fields[1] = upper_32_bits(buf_64);
@@ -947,7 +947,7 @@ int xhci_ctrl_tx(struct usb_device *udev, unsigned long pipe,
/* Invalidate buffer to make it available to usb-core */
if (length > 0)
- dma_unmap_single(ctrl->dev, map, length, direction);
+ dma_unmap_single(ctrl->host.hw_dev, map, length, direction);
if (GET_COMP_CODE(le32_to_cpu(event->trans_event.transfer_len))
== COMP_SHORT_TX) {
diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c
index 55f51ca95192..930bb4b6c608 100644
--- a/drivers/usb/host/xhci.c
+++ b/drivers/usb/host/xhci.c
@@ -1359,7 +1359,11 @@ int xhci_register(struct xhci_ctrl *ctrl)
*/
host->no_desc_before_addr = true;
- host->hw_dev = dev;
+ /*
+ * If xHCI doesn't have our own DT node, it'll be a child of a
+ * physical USB host controller device that should be used for DMA
+ */
+ host->hw_dev = dev_of_node(dev) ? dev : dev->parent;
host->submit_int_msg = xhci_submit_int_msg;
host->submit_control_msg = xhci_submit_control_msg;
host->submit_bulk_msg = xhci_submit_bulk_msg;
--
2.30.2
next prev parent reply other threads:[~2023-02-21 8:07 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-21 8:05 [PATCH RFC 00/12] ARM64: layerscape: make LS1046 DMA coherent Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 01/12] usb: dwc3: populate parent of xHCI dev Ahmad Fatoum
2023-02-21 8:05 ` Ahmad Fatoum [this message]
2023-02-21 8:05 ` [PATCH RFC 03/12] net: rtl8169: pass physical device for DMA API Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 04/12] firmware: zynqmp-fpga: pass physical device to " Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 05/12] net: designware: eqos: " Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 06/12] x86: select OF_DMA_DEFAULT_COHERENT Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 07/12] dma: define CONFIG_OF_DMA_COHERENCY Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 08/12] RISC-V: StarFive: J7100: set /soc/dma-noncoherent Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 09/12] of: populate new device_d::dma_coherent attribute Ahmad Fatoum
2023-02-27 10:41 ` Sascha Hauer
2024-01-11 7:05 ` Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 10/12] dma: fix dma_sync when not all device DMA is equally coherent Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 11/12] dma: provide of_dma_coherent_fixup helper Ahmad Fatoum
2023-02-21 8:05 ` [PATCH RFC 12/12] ARM64: layerscape: configure all DMA masters to be cache-coherent Ahmad Fatoum
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=20230221080524.607241-3-a.fatoum@pengutronix.de \
--to=a.fatoum@pengutronix.de \
--cc=barebox@lists.infradead.org \
--cc=lst@pengutronix.de \
/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