At present dm/device.h includes the linux-compatible features. This
requires including linux/compat.h which in turn includes a lot of headers.
One of these is malloc.h which we thus end up including in every file in
U-Boot. Apart from the inefficiency of this, it is problematic for sandbox
which needs to use the system malloc() in some files.
Move the compatibility features into a separate header file.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present devres.h is included in all files that include dm.h but few
make use of it. Also this pulls in linux/compat which adds several more
headers. Drop the automatic inclusion and require files to include devres
themselves. This provides a good indication of which files use devres.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Anatolij Gustschin <agust@denx.de>
At present if CONFIG_SPL_GPIO_SUPPORT is enabled then the GPIO uclass
is included in SPL/TPL without any control for boards. Some boards may
want to disable this to reduce code size where GPIOs are not needed in
SPL or TPL.
Add a new Kconfig option to permit this. Default it to 'y' so that
existing boards work correctly.
Change existing uses of CONFIG_DM_GPIO to CONFIG_IS_ENABLED(DM_GPIO) to
preserve the current behaviour. Also update the 74x164 GPIO driver since
it cannot build with SPL.
This allows us to remove the hacks in config_uncmd_spl.h and
Makefile.uncmd_spl (eventually those files should be removed).
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
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 commit
642b80d256 ("net: designware: drop compatible altr, socfpga-stmmac")
breaks designware ethernet for all ARC boards. It removes
"altr, socfpga-stmmac" compatible from "drivers/net/designware.c"
without changing compatible in the boards which use it.
Fix that by adding "snps,arc-dwmac-3.70a" compatible string to
"drivers/net/designware.c" and using it in ARC boards device tree.
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
The same compatible = "altr,socfpga-stmmac" appears in both
drivers/net/designware.c and drivers/net/dwmac_socfgpa.c,
creating ambiguity in which driver will be bound.
For Intel/Altera SoC devices, dwmac_socfpga.c is the correct driver.
So drop the compatible string from designware.c.
Signed-off-by: Ralph Siemsen <ralph.siemsen@linaro.org>
Reviewed-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Using 'phy_connect' instead of 'phy_find_by_mask' and 'phy_connect_dev'
both deduplicates code and adds support for 'fixed-link'.
Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
The designware eth driver registers an mdio bus during probe, but if no
PHY is found, this bus is never removed although probe failes and the
driver is shown as not probed in the dm tree.
This later leads to errors when e.g. the mii or mdio commands try to
use available mdio buses because the mdio bus is still registered but
all corresponding data structures are invalid because probe failed.
Fix this by unregistering the mdio bus on probe failure (just as it is
unregistered in the .remove callback, too).
Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Short frames are padded to the minimum allowed size of 60 bytes.
However, the designware driver sends old data in these padding bytes.
It is common practice to zero out these padding bytes ro prevent
leaking memory contents to other hosts.
Fix the padding code to zero out the padded bytes at the end.
Tested on socfpga gen5.
Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
The designware driver has a bug in setting the tx length into the dma
descriptor: it always or's the length into the descriptor without
zeroing out the length mask before.
This results in occasional packets being transmitted with a length
greater than they should be (trailer). Due to the nature of Ethernet
allowing such a trailer, most packets seem to be parsed fine by remote
hosts, which is probably why this hasn't been noticed.
Fix this by correctly clearing the size mask before setting the new
length.
Tested on socfpga gen5.
Signed-off-by: Simon Goldschmidt <simon.k.r.goldschmidt@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Add code to reset all reset signals as in Ethernet DT node. A reset
property is an optional feature, so only print out a warning and do not
fail if a reset property is not present.
If a reset property is discovered, then use it to deassert, thus
bringing the IP out of reset.
Signed-off-by: Ley Foon Tan <ley.foon.tan@intel.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
On the SPEAr600 SoC, which has the dwmac1000 variant of the IP block,
the DMA reset never succeeds when a MII PHY is used (no problem with a
GMII PHY). The designware_eth_init() function sets the
DMAMAC_SRST bit in the DMA_BUS_MODE register, and then
polls until this bit clears. When a MII PHY is used, with the current
driver, this bit never clears and the driver therefore doesn't work.
The reason is that the PS bit of the GMAC_CONTROL register should be
correctly configured for the DMA reset to work. When the PS bit is 0,
it tells the MAC we have a GMII PHY, when the PS bit is 1, it tells
the MAC we have a MII PHY.
Doing a DMA reset clears all registers, so the PS bit is cleared as
well. This makes the DMA reset work fine with a GMII PHY. However,
with MII PHY, the PS bit should be set.
We have identified this issue thanks to two SPEAr600 platform:
- One equipped with a GMII PHY, with which the existing driver was
working fine.
- One equipped with a MII PHY, where the current driver fails because
the DMA reset times out.
Note: Taken from https://www.spinics.net/lists/netdev/msg432578.html
Signed-off-by: Quentin Schulz <quentin.schulz@bootlin.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.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>
We have a large number of places where while we historically referenced
gd in the code we no longer do, as well as cases where the code added
that line "just in case" during development and never dropped it.
Signed-off-by: Tom Rini <trini@konsulko.com>
After commit ba1f966725 ("net: designware: add clock support")
we got NET broken on axs101 and axs103 platforms.
Some clock don't support gating so their clock drivers don't
implement .enable/.disable callbacks. In such case clk_enable
returns -ENOSYS.
Also some clock drivers implement .enable/.disable callbacks not for all
clock IDs and return -ENOSYS (or -ENOTSUPP) for others.
If we have such clock in 'clocks' list of designware ethernet controller
node we fail to probe designware ethernet.
Fix it.
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Reviewed-by: Patrice Chotard <patrice.chotard@st.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Make sure that we pad small packets to a minimum length of 60 bytes
(without FCS). This is necessary to interface with Ethernet switches
that will reject RUNT frames unless padded correctly.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
This implementation manages several clocks, disable and
free all of them in case of error during probe and in remove
callback.
Signed-off-by: Patrice Chotard <patrice.chotard@st.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Update the Designware Ethernet MAC driver to support a live device
tree.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
With the new dev_read functions available, we can convert the rockchip
architecture-specific drivers and common drivers used by these devices
over to the dev_read family of calls.
This covers the Gigabit Ethernet MAC (i.e. common designware driver and
rockchip-specific wrapper).
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Some board need a regulator for gmac phy, so add this code to handle it.
Signed-off-by: Jacob Chen <jacob2.chen@rock-chips.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
These support the flat device tree. We want to use the dev_read_..()
prefix for functions that support both flat tree and live tree. So rename
the existing functions to avoid confusion.
In the end we will have:
1. dev_read_addr...() - works on devices, supports flat/live tree
2. devfdt_get_addr...() - current functions, flat tree only
3. of_get_address() etc. - new functions, live tree only
All drivers will be written to use 1. That function will in turn call
either 2 or 3 depending on whether the flat or live tree is in use.
Note this involves changing some dead code - the imx_lpi2c.c file.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present devices use a simple integer offset to record the device tree
node associated with the device. In preparation for supporting a live
device tree, which uses a node pointer instead, refactor existing code to
access this field through an inline function.
Signed-off-by: Simon Glass <sjg@chromium.org>
In Uboot for Meson GX the compatible string in meson-gxbb.dtsi so far is:
compatible = "amlogic,meson6-dwmac", "snps,dwmac";
On Linux in the same dt file it's
compatible = "amlogic,meson-gx-dwmac", "amlogic,meson-gxbb-dwmac", "snps,dwmac";
To avoid breaking ethernet with the next DT synch from Linux to U-Boot
(planned as prerequisite for adding Meson GX MMC driver to U-Boot) add
"amlogic,meson-gx-dwmac" to the compatibility list in the designware
driver.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
This patch adds glue code required for enabling the designware
mac on stm32f7 devices.
Signed-off-by: Michael Kurz <michi.kurz@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Export all functions so that drivers can use them, or not, as the need
arises.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Romain Perier <romain.perier@collabora.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
With rockchip we need to make adjustments after the link speed is set but
before enabling received/transmit. In preparation for this, split these
two pieces into separate functions.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Romain Perier <romain.perier@collabora.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
This function can fail, so return the error if there is one.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Romain Perier <romain.perier@collabora.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
To allow other DM drivers to subclass the designware driver various
functions and structures need to be exported. Export these.
Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
Signed-off-by: Romain Perier <romain.perier@collabora.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Remove the device definition from board file, update the driver with
the new compatible property and update config with necessary options.
Signed-off-by: Beniamino Galvani <b.galvani@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Commit 90b7fc924a "net: designware: support phy reset device-tree
bindings" made DW GMAC driver dependent on DM_GPIO by unconditional
usage of purely DM_GPIO stuff like:
* dm_gpio_XXX()
* gpio_request_by_name()
But since that driver as of today might be easily used without
DM_GPIO (that's the case for Synopsys AXS10x boards) we're
shielding all DM_GPIO things by ifdefs.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Beniamino Galvani <b.galvani@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
Cc: Sonic Zhang <sonic.zhang@analog.com>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Marek Vasut <marex@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
All members of the DMA descriptor must be 32-bit, even on 64-bit
architectures: change the type to u32 to ensure this. Also, fix
other warnings.
Signed-off-by: Beniamino Galvani <b.galvani@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
[trini: Use phys_addr_t not unsigned long long to test that we're within
DMA'able memory]
Signed-off-by: Tom Rini <trini@konsulko.com>
Add support for the snps,reset-gpio, snps,reset-active-low (optional) and
snps,reset-delays-us device-tree bindings. The combination of these
three define how the PHY should be reset to ensure it's in a sane state.
Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
Reviewed-by: Simon Glass <sjg@chromium.org>
This property allows to specify fastest connection mode supported by
the MAC (as opposed to features of the phy).
There are situations when phy may handle faster modes than the
MAC (or even it's particular implementation or even due to CPU being too
slow).
This property is a standard one in Linux kernel these days and some
boards do already use it in their device tree descriptions.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Sonic Zhang <sonic.zhang@analog.com>
cc: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Current implementation only sets "port select" bit for non-1Gb mode.
That works fine if GMAC has just exited reset state but we may as well
change connection mode in runtime. Then we'll need to reprogram GMAC for
that new mode of operation and if previous mode was 10 or 100 Mb and new
one is 1 Gb we'll need to reset port mode bit.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Sonic Zhang <sonic.zhang@analog.com>
cc: Simon Glass <sjg@chromium.org>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
With format-security errors turned on, GCC picks up the use of sprintf with
a format parameter not being a string literal.
Simple uses of sprintf are also converted to use strcpy.
Signed-off-by: Ben Whitten <ben.whitten@gmail.com>
Acked-by: Wolfgang Denk <wd@denx.de>
Reviewed-by: Tom Rini <trini@konsulko.com>
Most driver model PCI functions have a dm_ prefix. At some point, when the
old code is converted to driver model and the old functions are removed, we
will drop that prefix.
For consistency, we should use the dm_ prefix for all driver model
functions. Update pci_get_bdf() accordingly.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Trailing backslashes are necessary only in macros, not in the actual
code, so remove them.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Select PHYLIB in drivers/net/Kconfig. And remove CONFIG_PHYLIB
from legacy board header files.
This fixed the warnings when both ALTERA_TSE and ETH_DESIGNWARE
are selected.
Signed-off-by: Thomas Chou <thomas@wytron.com.tw>
Reported-by: Pavel Machek <pavel@denx.de>
Acked-by: Chin Liang See <clsee@altera.com>
Acked-by: Pavel Machek <pavel@denx.de>
Tested-by: Pavel Machek <pavel@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
In designware_eth_probe(), some additional resources are allocated
(eg: mdio, phy). We should free these in the driver remove phase.
Add designware_eth_remove() to clean it up.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
The Designware ethernet controller is also seen on PCI bus, e.g.
on Intel Quark SoC. Add this support in the DM version driver.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
When building dm version of designware eth driver on a platform
with 64-bit phys_addr_t, it reports the following warnings:
drivers/net/designware.c: In function 'designware_eth_probe':
drivers/net/designware.c:599:2:
warning: format '%lx' expects argument of type 'long unsigned int',
but argument 3 has type 'phys_addr_t' [-Wformat]
drivers/net/designware.c:600:21:
warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
drivers/net/designware.c:601:21:
warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
This commit fixes the build warnings.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The driver variable name is eth_sandbox, which is probably a copy-paste
mistake. Fix it.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Add the OF compatible property to match the SoCFPGA GMAC.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Dinh Nguyen <dinguyen@opensource.altera.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Some devices can take a long time to work out whether they have a new packet
or now. For example the ASIX USB Ethernet dongle can take 5 seconds to do
this, since it waits until it gets a new packet on the wire before allowing
the USB bulk read packet to be submitted.
At present with driver mode the Ethernet receive code reads 32 packets. This
can take a very long time if we must wait for all 32 packets. The old code
(before driver model) worked by reading a single set of packets from the USB
device, then processing all the packets with in. It would be nice to use
the same behaviour with driver model.
Add a flag to the receive method which indicates that the driver should try
to find a packet if available, by consulting the hardware. When the flag is
not set, it should just return any packet data it has already received. If
there is none, it should return -EAGAIN so that the loop will terminate.
Signed-off-by: Simon Glass <sjg@chromium.org>
