An XHCI controller that sits behind an IOMMU needs to map and unmap
its memory buffers to do DMA. Implement this by inroducing new
xhci_dma_map() and xhci_dma_unmap() helper functions. The
xhci_dma_map() function replaces the existing xhci_virt_to_bus()
function in the sense that it returns the bus address in the case
of simple address translation in the absence of an IOMMU. The
xhci_bus_to_virt() function is eliminated by storing the CPU
address of the allocated scratchpad memory in struct xhci_ctrl.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Marek Vasut <marex@denx.de>
There are devices which cause a USB stall when trying to read strings.
Specifically Arduino Mega R3 stalls when trying to read the product
string.
The stall currently remains unhandled, and subsequent retries submit new
transfers on a stopped endpoint which ultimately cause a crash in
abort_td():
WARN halted endpoint, queueing URB anyway.
XHCI control transfer timed out, aborting...
Unexpected XHCI event TRB, skipping... (3affe040 00000000 13000000 02008401)
BUG at drivers/usb/host/xhci-ring.c:505/abort_td()!
BUG!
resetting ...
Linux seems to be able to recover from the stall by issuing a
TRB_RESET_EP command.
Introduce reset_ep() which issues a TRB_RESET_EP followed by setting the
transfer ring dequeue pointer via TRB_SET_DEQ. This allows to properly
recover from a USB stall error and continue communicating with the USB
device.
Signed-off-by: Stefan Agner <stefan@agner.ch>
This change makes debugging a bit easier as the output is better
readable with the added space. The explicit le16_to_cpu() is not
needed in the output. Also this patch moves the strings into one line
to make the patch checkpatch clean.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Aaron Williams <awilliams@marvell.com>
Cc: Chandrakala Chavva <cchavva@marvell.com>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Marek Vasut <marex@denx.de>
So far we've been content with passing physical addresses when
configuring memory addresses into XHCI controllers, but not all
platforms have buses with transparent mappings. Specifically the
Raspberry Pi 4 might introduce an offset to memory accesses incoming
from its PCIe port.
Introduce xhci_virt_to_bus() and xhci_bus_to_virt() to cater with these
limitations, and make sure we don't break non DM users.
Signed-off-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Stefan Roese <sr@denx.de>
Tested-by: Peter Robinson <pbrobinson@gmail.com>
[mb: fix compilation for 32 bit]
Signed-off-by: Matthias Brugger <mbrugger@suse.com>
fix from nicolas
Testing with v2021.01 on MIPS Octeon has shown, that the latest patch
for the "short packet event trb handling" did introduce a bug on
platforms with virtual address != physical address. This patch fixes
this issue by using the correct address types in the compare (both
physical in this case).
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Aaron Williams <awilliams@marvell.com>
Cc: Chandrakala Chavva <cchavva@marvell.com>
Cc: Ran Wang <ran.wang_1@nxp.com>
Cc: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
Cc: Marek Vasut <marex@denx.de>
Cc: Bin Meng <bmeng.cn@gmail.com>
For bulk IN transfer, the codes will set ISP flag to request event TRB
being generated by xHC for the case of short packet. So when encountering
buffer-cross-64K-boundary (which we will divide payload and enqueuqe
more than 1 transfer TRB), and the first TRB ends up with a short packet
condition it will trigger an short packet code transfer event per that
flag and cause more than 1 event TRB generated for this transfer.
However, current codes will only handle the first transfer event TRB
then mark current transfer completed, causing next transfer
failure due to event TRB mis-match.
Such issue has been observed on some Layerscape platforms (LS1028A,
LS1088A, etc) with USB ethernet device.
This patch adds a loop to make sure the event TRB for last transfer TRB
has been handled in time.
Signed-off-by: Ran Wang <ran.wang_1@nxp.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Octeon uses mapped addresses for virtual and physical memory. It's not
that easy to calculate the resulting addresses here. So let's remove
this BUG_ON() completely, as it's not really helpful.
Please also note, that BUG_ON() is not recommended any more in the Linux
kernel.
Signed-off-by: Stefan Roese <sr@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Marek Vasut <marex@denx.de>
Use TRB_TX_TYPE() instead of (TRB_DATA_OUT/IN << TRB_TX_TYPE_SHIFT)
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
For normal TRB fields:
use TRB_LEN(x) instead of ((x) & TRB_LEN_MASK);
and use TRB_INTR_TARGET(x) instead of
(((x) & TRB_INTR_TARGET_MASK) << TRB_INTR_TARGET_SHIFT)
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
There some vendor quirks for MTK xHCI 0.96 host controller:
1. It defines some extra SW scheduling parameters for HW
to minimize the scheduling effort for synchronous and
interrupt endpoints. The parameters are put into reserved
DWs of slot context and endpoint context.
2. Its TDS in Normal TRB defines a number of packets that
remains to be transferred for a TD after processing all
Max packets in all previous TRBs.
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Tested-by: Frank Wunderlich <frank-w@public-files.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
xhci versions 1.0 and later report the untransferred data remaining in a
TD a bit differently than older hosts.
We used to have separate functions for these, and needed to check host
version before calling the right function.
Now Mediatek host has an additional quirk on how it uses the TD Size
field for remaining data. To prevent yet another function for calculating
remainder we instead want to make one quirk friendly unified function.
Porting from the Linux:
c840d6ce772d("xhci: create one unified function to calculate TRB TD remainder.")
124c39371114("xhci: use boolean to indicate last trb in td remainder calculation")
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Add a member to save xHCI version, it's used some times.
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Some platforms, like MIPS Octeon, use mapped addresses (virtual address
!= physical address). On these platforms we need to make sure, that the
local virtual addresses are converted to physical (DMA) addresses for
the xHCI controller. This patch adds the missing virt_to_phys() calls,
so that the correct addresses are used.
Signed-off-by: Stefan Roese <sr@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Marek Vasut <marex@denx.de>
These functions are CPU-related and do not use driver model. Move them to
cpu_func.h
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
The xHCI 1.1 version also need set Transfer Type field
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
The xhci.h header file is currently located under drivers/usb/xhci
Move it to the include/usb folder to make it available to drivers that
are not under drivers/usb/xhci
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
When U-Boot started using SPDX tags we were among the early adopters and
there weren't a lot of other examples to borrow from. So we picked the
area of the file that usually had a full license text and replaced it
with an appropriate SPDX-License-Identifier: entry. Since then, the
Linux Kernel has adopted SPDX tags and they place it as the very first
line in a file (except where shebangs are used, then it's second line)
and with slightly different comment styles than us.
In part due to community overlap, in part due to better tag visibility
and in part for other minor reasons, switch over to that style.
This commit changes all instances where we have a single declared
license in the tag as both the before and after are identical in tag
contents. There's also a few places where I found we did not have a tag
and have introduced one.
Signed-off-by: Tom Rini <trini@konsulko.com>
Bool initializations should use true and false.
This issue was detected with the help of Coccinelle.
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Marek Vasut <marex@denx.de>
In xhci_queue_command(), when the command is not 'reset endpoint',
'stop endpoint' or 'set TR dequeue pointer', endpoint ID should not
be encoded in the TRB.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Stefan Roese <sr@denx.de>
Tested-by: Stefan Roese <sr@denx.de>
Now, include/linux/errno.h is a wrapper of <asm-generic/errno.h>.
Replace all include directives for <asm-generic/errno.h> with
<linux/errno.h>.
<asm-generic/...> is supposed to be included from <asm/...> when
arch-headers fall back into generic implementation. Generally, they
should not be directly included from .c files.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
[trini: Add drivers/usb/host/xhci-rockchip.c]
Signed-off-by: Tom Rini <trini@konsulko.com>
Rather than getting this directly from struct usb_device, call a function
to obtain it. This will make it possible for driver model to provide it
another way.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Marek Vasut <marex@denx.de>
This commit allows xHCI to use both 64 and 32 bit memory
physical addresses depending on architecture it's being built for.
Also it makes use of readq()/writeq() on 64-bit systems
Signed-off-by: Sergey Temerkhanov <s.temerkhanov@gmail.com>
Signed-off-by: Radha Mohan Chintakuntla <rchintakuntla@cavium.com>
U-Boot has never cared about the type when we get max/min of two
values, but Linux Kernel does. This commit gets min, max, min3, max3
macros synced with the kernel introducing type checks.
Many of references of those macros must be fixed to suppress warnings.
We have two options:
- Use min, max, min3, max3 only when the arguments have the same type
(or add casts to the arguments)
- Use min_t/max_t instead with the appropriate type for the first
argument
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: Pavel Machek <pavel@denx.de>
Acked-by: Lukasz Majewski <l.majewski@samsung.com>
Tested-by: Lukasz Majewski <l.majewski@samsung.com>
[trini: Fixup arch/blackfin/lib/string.c]
Signed-off-by: Tom Rini <trini@ti.com>
This adds stack layer for eXtensible Host Controller Interface
which facilitates use of USB 3.0 in host mode.
Adapting xHCI host controller driver in linux-kernel
by Sarah Sharp to needs in u-boot.
Initial porting from Linux kernel version 3.4, with following
top commit history of drivers/usb/host/xhci* :
cf84055 xHCI: Cleanup isoc transfer ring when TD length mismatch found
This adds the basic xHCI host controller driver with bare minimum
features:
- Control/Bulk transfer support has been added with required
infrastructure for necessary xHC data structures.
- Stream protocol hasn't been supported yet.
- No support for quirky devices has been added.
Signed-off-by: Vikas C Sajjan <vikas.sajjan@samsung.com>
Signed-off-by: Julius Werner <jwerner@chromium.org>
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Minkyu Kang <mk7.kang@samsung.com>
Cc: Dan Murphy <dmurphy@ti.com>
Cc: Marek Vasut <marex@denx.de>
