After the discussion here:
https://lore.kernel.org/netdev/20210603143453.if7hgifupx5k433b@pali/
which resulted in this patch:
https://patchwork.kernel.org/project/netdevbpf/patch/20210704134325.24842-1-pali@kernel.org/
and many other discussions before it, notably:
https://patchwork.kernel.org/project/linux-arm-kernel/patch/1512016235-15909-1-git-send-email-Bhaskar.Upadhaya@nxp.com/
it became apparent that nobody really knows what "SGMII 2500" is.
Certainly, Freescale/NXP hardware engineers name this protocol
"SGMII 2500" in the reference manuals, but the PCS devices do not
support any "SGMII" specific features when operating at the speed of
2500 Mbps, no in-band autoneg and no speed change via symbol replication
. So that leaves a fixed speed of 2500 Mbps using a coding of 8b/10b
with a SERDES lane frequency of 3.125 GHz. In fact, "SGMII 2500 without
in-band autoneg and at a fixed speed" is indistinguishable from
"2500base-x without in-band autoneg", which is precisely what these NXP
devices support.
So it just appears that "SGMII 2500" is an unclear name with no clear
definition that stuck.
As such, in the Linux kernel, the drivers which use this SERDES protocol
use the 2500base-x phy-mode.
This patch converts U-Boot to use 2500base-x too, or at least, as much
as it can.
Note that I would have really liked to delete PHY_INTERFACE_MODE_SGMII_2500
completely, but the mvpp2 driver seems to even distinguish between SGMII
2500 and 2500base-X. Namely, it enables in-band autoneg for one but not
the other, and forces flow control for one but not the other. This goes
back to the idea that maybe 2500base-X is a fiber protocol and SGMII-2500
is an MII protocol (connects a MAC to a PHY such as Aquantia), but the
two are practically indistinguishable through everything except use case.
NXP devices can support both use cases through an identical configuration,
for example RX flow control can be unconditionally enabled in order to
support rate adaptation performed by an Aquantia PHY. At least I can
find no indication in online documents published by Cisco which would
point towards "SGMII-2500" being an actual standard with an actual
definition, so I cannot say "yes, NXP devices support it".
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
At present ofnode is present in the device even if it is never used. With
of-platdata this field is not used, so can be removed. In preparation for
this, change the access to go through inline functions.
Signed-off-by: Simon Glass <sjg@chromium.org>
Most drivers use these access methods but a few do not. Update them.
In some cases the access is not permitted, so mark those with a FIXME tag
for the maintainer to check.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Pratyush Yadav <p.yadav@ti.com>
We use 'priv' for private data but often use 'platdata' for platform data.
We can't really use 'pdata' since that is ambiguous (it could mean private
or platform data).
Rename some of the latter variables to end with 'plat' for consistency.
Signed-off-by: Simon Glass <sjg@chromium.org>
This construct is quite long-winded. In earlier days it made some sense
since auto-allocation was a strange concept. But with driver model now
used pretty universally, we can shorten this to 'auto'. This reduces
verbosity and makes it easier to read.
Coincidentally it also ensures that every declaration is on one line,
thus making dtoc's job easier.
Signed-off-by: Simon Glass <sjg@chromium.org>
The Linux coding style guide (Documentation/process/coding-style.rst)
clearly says:
It's a **mistake** to use typedef for structures and pointers.
Besides, using typedef for structures is annoying when you try to make
headers self-contained.
Let's say you have the following function declaration in a header:
void foo(bd_t *bd);
This is not self-contained since bd_t is not defined.
To tell the compiler what 'bd_t' is, you need to include <asm/u-boot.h>
#include <asm/u-boot.h>
void foo(bd_t *bd);
Then, the include direcective pulls in more bloat needlessly.
If you use 'struct bd_info' instead, it is enough to put a forward
declaration as follows:
struct bd_info;
void foo(struct bd_info *bd);
Right, typedef'ing bd_t is a mistake.
I used coccinelle to generate this commit.
The semantic patch that makes this change is as follows:
<smpl>
@@
typedef bd_t;
@@
-bd_t
+struct bd_info
</smpl>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
- Add DM_ETH support for lx2160aqds, ls2080aqds, ls1088aqds
- QSI related fixes on ls1012a, ls2080a, ls1046a, ls1088a, ls1043a based
platforms
- Bug-fixes/updtaes related to ls1046afrwy, fsl-mc, msi-map property
Bus callback functions for read/write/reset need to be set only for
DM_ETH, moving endif a bit lower.
Signed-off-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Reviewed-by: Priyanka Jain <priyanka.jain@nxp.com>
Probe the FMan MACs based on the device tree while
retaining the legacy code/functionality.
One notable change introduced here is that, for DM_ETH,
the name of the interfaces is corrected to the fmX-macY
format, that avoids the referral to the MAC block names
which were incorrect for FMan v3 devices (i.e. DTSEC,
TGEC) and had weird formatting (i.e. FM1@DTSEC6, FM1@TGEC1).
The legacy code is left unchanged in this respect.
Signed-off-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Reviewed-by: Priyanka Jain <priyanka.jain@nxp.com>
Move the receive buffer free code in a separate function.
Signed-off-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Reviewed-by: Priyanka Jain <priyanka.jain@nxp.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>
The settings for 2.5G SGMII are wrong, which the 2.5G case is missed in
set_if_mode(), and the serdes PCS configuration are wrong, this patch uses
the correct settings took from Linux.
Signed-off-by: Shaohui Xie <Shaohui.Xie@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
QSGMII PCS needed to be programmed same as SGMII PCS, and there are
four ports in QSGMII PCS, port 0, 1, 2, 3, all the four ports shared
port 0's MDIO controller, so when programming port 0, we continue to
program other three ports.
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: Mingkai Hu <Mingkai.Hu@freescale.com>
Signed-off-by: Gong Qianyu <Qianyu.Gong@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
The fsl_dtsec.h & fsl_tgec.h & fsl_fman.h can be shared on both ARM
and PPC, move it out of ppc to include/, and change the path in
drivers accordingly.
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: Gong Qianyu <Qianyu.Gong@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
codes related to phylib operations should be wrapped by CONFIG_PHYLIB.
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: Gong Qianyu <Qianyu.Gong@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
In convention, the '0' is a normal return value indicating there isn't
an error. While some functions of FMan IM driver treat '0' as an error
return value.
Signed-off-by: Hou Zhiqiang <B48286@freescale.com>
Signed-off-by: Gong Qianyu <Qianyu.Gong@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
The FMan IM driver is developed for 32-bit platfroms and isn't
friendly to 64-bit platforms, so do the minimal refactor:
1. Refine the MURAM management and access.
2. Correct the initialization and operations for QDs and BDs.
Signed-off-by: Hou Zhiqiang <B48286@freescale.com>
Signed-off-by: Gong Qianyu <Qianyu.Gong@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
The Frame Manager(FMan) is a big-endian peripheral, so the
registers, internal MURAM and BDs, which are allocated in main
memory and used to communication between core and FMan, should
be accessed in big-endian. The big-endian platforms can access
them directly as the code implemented so far, while for the
little-endian platforms it need to swap the byte-order.
Signed-off-by: Hou Zhiqiang <B48286@freescale.com>
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: Mingkai Hu <Mingkai.Hu@freescale.com>
Signed-off-by: Gong Qianyu <Qianyu.Gong@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
The rxbd is not correctly handled in case of a frame physical error
(FPE) or frame size error (FSE). The rxbd must be cleared and
advanced in case of an error to avoid receive stall.
Signed-off-by: Daniel Inderbitzin <daniel.inderbitzin@gmail.com>
The memac for PHY management on little endian SoCs is similar on big
endian SoCs, so we modify the driver by using I/O accessor function to
handle the endianness, so the driver can be reused on little endian
SoCs, we introduce CONFIG_SYS_MEMAC_LITTLE_ENDIAN for little endian
SoCs, if the CONFIG_SYS_MEMAC_LITTLE_ENDIAN is defined, the I/O access
is little endian, if not, the I/O access is big endian. Move fsl_memac.h
out of powerpc include.
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: York Sun <yorksun@freescale.com>
Update the naming convention used in the network stack functions and
variables that Ethernet drivers use to interact with it.
This cleans up the temporary hacks that were added to this interface
along with the DM support.
This patch has a few remaining checkpatch.pl failures that would be out
of the scope of this patch to fix (drivers that are in gross violation
of checkpatch.pl).
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Acked-by: Simon Glass <sjg@chromium.org>
U-boot assumes that all FMAN ports have a PHY. Some SoCs (like T1040)
have fixed links. This means that the ports are connected MAC to MAc
and there is no Ethernet PHY attatched. This patch initializes a
FMAN MAC even if it doesn't have a PHY attached.
Signed-off-by: Codrin Ciubotariu <codrin.ciubotariu@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
fm_standard_init() initializes each 10G port by FM_TGEC_INFO_INITIALIZER.
but it needs different implementation of FM_TGEC_INFO_INITIALIZER on different SoCs.
on SoCs earlier(e.g. T4240, T2080), the notation between 10GEC and MAC as below:
10GEC1->MAC9, 10GEC2->MAC10, 10GEC3->MAC1, 10GEC4->MAC2
on SoCs later(e.g. T1024, etc), the notation between 10GEC and MAC as below:
10GEC1->MAC1, 10GEC2->MAC2
so we introduce CONFIG_FSL_FM_10GEC_REGULAR_NOTATION to fit the new SoCs on
which 10GEC enumeration is consistent with MAC enumeration.
Signed-off-by: Shengzhou Liu <Shengzhou.Liu@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
As auto-negotiation is not supported for 2.5G SGMII, we need
to add a new type PHY_INTERFACE_MODE_SGMII_2500 to differentiate
SGMII-1G and SGMII-2.5G with different setting for auto-negotiation.
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: Shengzhou Liu <Shengzhou.Liu@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
There are more than two 10GEC in single FMAN in some SoCs(e.g. T2080).
This patch adds support for 10GEC3 and 10GEC4.
Signed-off-by: Shengzhou Liu <Shengzhou.Liu@freescale.com>
Also some fix for QSGMII.
1. fix QSGMII configure of Serdes2.
2. fix PHY address of QSGMII MAC9 & MAC10 for each FMAN.
3. fix dtb for QSGMII interface.
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Acked-by: York Sun <yorksun@freescale.com>
1. fix 10G mac offset by plus 8;
2. add second 10G port info for FM1 & FM2 when init ethernet info;
3. fix 10G lanes name to match lane protocol table;
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
Signed-off-by: Andy Fleming <afleming@freescale.com>
phy.c:46:5: warning: symbol 'genphy_config_advert' was not declared. Should it be static?
phy.c:121:5: warning: symbol 'genphy_setup_forced' was not declared. Should it be static?
phy.c:468:5: warning: symbol 'phy_probe' was not declared. Should it be static?
phy.c:491:19: warning: symbol 'get_phy_driver' was not declared. Should it be static?
phy.c:508:19: warning: symbol 'phy_device_create' was not declared. Should it be static?
phy.c:552:5: warning: symbol 'get_phy_id' was not declared. Should it be static?
phy.c:584:19: warning: symbol 'get_phy_device' was not declared. Should it be sta
vitesse.c:126:5: warning: symbol 'vsc8601_config' was not declared. Should it be static?
vsc7385.c:33:5: warning: symbol 'vsc7385_upload_firmware' was not declared. Should it be static?
tgec_phy.c:33:5: warning: symbol 'tgec_mdio_write' was not declared. Should it be static?
tgec_phy.c:75:5: warning: symbol 'tgec_mdio_read' was not declared. Should it be static?
tgec_phy.c:117:5: warning: symbol 'tgec_mdio_reset' was not declared. Should it be static?
eth.c:48:6: warning: symbol 'dtsec_configure_serdes' was not declared. Should it be static?
p4080.c:26:5: warning: symbol 'port_to_devdisr' was not declared. Should it be static?
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
TBI PHY address (TBIPA) register is set in general frame manager
phy init funciton dtsec_init_phy() in drivers/net/fm/eth.c, and
it is supposed to set TBIPA on FM1@DTSEC1 in case of FM1@DTSEC1
isn't used directly, which provides MDIO for other ports. So
following code is wrong in case of FM2, which has a different
mac base.
struct dtsec *regs = (struct dtsec *)fm_eth->mac->base;
/* Assign a Physical address to the TBI */
out_be32(®s->tbipa, CONFIG_SYS_TBIPA_VALUE);
Signed-off-by: Shaohui Xie <Shaohui.Xie@freescale.com>
Signed-off-by: Andy Fleming <afleming@freescale.com>
The multirate ethernet media access controller (mEMAC) interfaces to
10Gbps and below Ethernet/IEEE 802.3 networks via either RGMII/RMII
interfaces or XAUI/XFI/SGMII/QSGMII using the high-speed SerDes interface.
Signed-off-by: Sandeep Singh <Sandeep@freescale.com>
Signed-off-by: Poonam Aggrwal <poonam.aggrwal@freescale.com>
Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
Signed-off-by: Andy Fleming <afleming@freescale.com>
Now that phy_startup() can return an actual error code, check for that error
code and abort network initialization if the PHY fails.
Signed-off-by: Timur Tabi <timur@freescale.com>
Acked-by: Nobuhiro Iwamamatsu <nobuhiro.iwamatsu.yj@renesas.com> (sh_eth part)
Acked-by: Stephan Linz <linz@li-pro.net> (Xilinx part, xilinx_axi_emac and xilinx_ll_temac)
Reviewed-by: Marek Vasut <marex@denx.de> (FEC part)
Fix this:
eth.c: In function 'fm_eth_initialize':
eth.c:651:12: warning: assignment from incompatible pointer type
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Function dtsec_configure_serdes() needs to know where the TBI PHY registers
are in order to configure SGMII for proper SerDes operation.
During SGMII initialzation, fm_eth_init_mac() passing NULL for 'phyregs'
when it called init_dtsec(), because it was believed that phyregs was not
used. In fact, it is used by dtsec_configure_serdes() to configure the TBI
PHY registers.
We also need to define the PHY registers in struct fm_mdio.
Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
The Frame Manager (FMan) on QorIQ SoCs with DPAA (datapath acceleration
architecture) is the ethernet contoller block. Normally it is utilized
via Queue Manager (Qman) and Buffer Manager (Bman). However for boot
usage the FMan supports a mode similar to QE or CPM ethernet collers
called Independent mode.
Additionally the FMan block supports multiple 1g and 10g interfaces as a
single entity in the system rather than each controller being managed
uniquely. This means we have to initialize all of Fman regardless of
the number of interfaces we utilize.
Different SoCs support different combinations of the number of FMan as
well as the number of 1g & 10g interfaces support per Fman.
We add support for the following SoCs:
* P1023 - 1 Fman, 2x1g
* P4080 - 2 Fman, each Fman has 4x1g and 1x10g
* P204x/P3041/P5020 - 1 Fman, 5x1g, 1x10g
Signed-off-by: Dave Liu <daveliu@freescale.com>
Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
Signed-off-by: Dai Haruki <dai.haruki@freescale.com>
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@freescale.com>
Signed-off-by: Lei Xu <B33228@freescale.com>
Signed-off-by: Mingkai Hu <Mingkai.hu@freescale.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Shaohui Xie <b21989@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
