Once all the other drivers got torn down in preparation for the OS
to start, tear down the clock controller last. The clock controller
must be torn down last as some of the clock which get turned off
might have still been needed during the teardown stage of the other
drivers.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Signed-off-by: Tam Nguyen <tam.nguyen.xa@renesas.com>
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Update R-Car Gen4 support in Gen3 clock driver. This patch renames the
V3U clock parts to Gen4 and extends them by new PLL2, PLL3, PLL4, PLL6
as well as SDSRC clock which use undocumented bits so far, and RPCSRC
clock which uses its own more capable divider table. The Gen4 module
standby and reset tables are also updated.
This patch makes use of union to alias Gen3 and more extensive Gen4
PLL tables, as the driver cannot ever be instantiated on hardware
that would identify itself as both Gen3 and Gen4.
The V3U clock driver is updated to match Gen4 clock driver behavior,
it is augmented with a more extensive PLL table and a valid MODEMR
register offset.
This supersedes "clk: renesas: Introduce R-Car Gen4 CPG driver"
from Hai Pham as the R-Car Gen3 and Gen4 clock core drivers are
extremely similar. That implementation was in turn based on Linux
commit 470e3f0d0b15 ("clk: renesas: rcar-gen4: Introduce R-Car Gen4 CPG driver")
by Yoshihiro Shimoda .
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
The reset identifier must be processed via MOD_CLK_PACK() before it is
used to look up register and bit within reset_regs or reset_clear_regs
arrays, otherwise completely bogus register and bit is picked from the
arrays, one which may even be out of range.
Fixes: 326e05c5e2 ("clk: renesas: Add and enable CPG reset driver")
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Do not cache the single CPG MODE register bit 12, instead cache the
entire register value, and only pick the matching bit from the cached
value when core clock of type MDSEL or PE are used. Both MDSEL and PE
clock type currently define .offset field as 12 on Gen3, which means
this code will use bit 12 on Gen3 again, however there are additional
clock on Gen4 which use different bits, and having this flexibility
in place now will be useful when adding Gen4.
No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
R-Car V3U has a CPG different enough to not be a generic Gen3 CPG but
similar enough to reuse code. Introduce a new CPG library, factor out
the SD clock and RPC clock handling and hook them to the generic Gen3
CPG driver so we have an equal state.
Based on Linux commit [1] and [2] by Wolfram Sang
[1] 8bb67d87346a ("clk: renesas: rcar-gen3: Factor out CPG library")
[2] 6f21d145b90f ("clk: renesas: cpg-lib: Move RPC clock registration to
the library")
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Marek: - Add rcar_clk_* prefix to all functions
- Rebase on changes to
clk: renesas: Introduce and use rcar_clk_get_rate64_div_table function
- Use u32_encode_bits/GENMASK bitfield ops
On R-Car V3M (AKA R8A77970), the SD0CKCR is laid out differently than on
the other R-Car gen3 SoCs. Hence, new clock types are introduced
respectively.
Based on Linux commit 381081ffc294 ("clk: renesas: r8a77970: Add SD0H/SD0
clocks for SDHI") by Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Marek: - Fix missing ~ in GENMASK(a, b), use clrsetbits_le32 instead
- Do not modify r8a77970-cpg-mssr.c much, drop enum r8a77970_clk_types
which is now part of common clock types in rcar-gen3-cpg.h instead
The old SD handling code was huge and could not handle all the details
which showed up on R-Car Gen3 SoCs meanwhile. It is time to switch to
another design. Have SDnH a separate clock, use the existing divider
clocks and move the errata handling from the clock driver to the SDHI
driver where it belongs.
Based on Linux series by Wolfram Sang, commit bb6d3fa98a41 ("clk:
renesas: rcar-gen3: Switch to new SD clock handling") and commit
e5f7e81ee430a ("mmc: renesas_sdhi: Parse DT for SDnH")
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Marek: - Add rcar_clk_* prefix to all functions
- Fix missing ~ in GENMASK(a, b), use clrsetbits_le32 instead
- Use DIV_ROUND_CLOSEST, else if parent clock = 199999992 and
rate = 200000000, the divider would be 0 and table lookup
would fail.
- Turn rcar_clk_get_table_val into signed integer, so it can
return 0 as a valid value and negative values as errors.
- Make the code operate on correct clock and add comment
which explains the reasoning behind it.
- Rebase on changes to
clk: renesas: Introduce and use rcar_clk_get_rate64_div_table function
The RPCSRC clock divider on R-Car D3 is very similar to the one on R-Car
E3, but uses a different pre-divider for the PLL0 parent. Add a new
macro to describe it, reusing the existing clock type for R-Car E3.
As both E3/D3 RPCSRC clock divider are different from the rest of R-Car
Gen3, keep the original implementation from Linux.
Based on Linux commit 40745482eec8 ("clk: renesas: r8a774c0: Add RPC
clocks") by Lad Prabhakar and 9d18f81b3535 ("clk: renesas: r8a77995: Add
RPC clocks") by Geert Uytterhoeven.
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org> # Add D3 tweaks
Introduce new helper function to handle clock type that uses
clk_div_table struct. Based vaguely on Linux code. Make use
of clk_div_table in RPC clocks handling.
The E3/D3 RPCSRC need to be handled differently and will be addressed in
subsequence patch.
Based on Linux commit db4a0073cc82 ("clk: renesas: rcar-gen3: Add RPC
clocks") by Sergei Shtylyov.
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Marek: - Squash patches to avoid adding unused code:
clk: renesas: Make use of clk_div_table in RPC clocks handling
clk: renesas: Introduce rcar_clk_get_rate64_div_table function
- Move the new code to the beginning of clk-rcar-gen3 to avoid
tables mixed with code
- Use rcar_ prefix for get_table_div function
- Get rid of custom macros, use GENMASK. Use custom field_get
implementation as the generic FIELD_GET does not support
constant mask and u32_get_bits requires higher optimization level
- Pass in the register bit mask instead of width/shift combination
- Turn rcar_clk_get_rate64_div_table into s64, as it can return -EINVAL
Replace custom local structure with matching one from clk-provider.h .
No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Drop 'core' parameter from gen3_clk_get_rate64_pll_mul_reg() function
as it is only used in debug print. No functional change except for the
debug print, which is disabled by default.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Since commit f7b4e4c094 ("clk: renesas: Synchronize R-Car Gen3 tables
with Linux 5.12"), the custom macros for RPC clocks were dropped.
Use pre-defined offset for RPC clocks, same as what Linux does, instead
of retrieving it from the macros
Reviewed-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Add trivial reset driver extension to the CPG clock driver. The change
turns current CPG UCLASS_CLK driver instance into an UCLASS_NOP proxy
driver, which in turn binds both generic rcar3_clk UCLASS_CLK clock
driver as well as generic rcar_rst UCLASS_RESET reset driver to the
CPG DT node. This way, any other drivers which use the 'reset' DT
property can now obtain valid reset handle backed by a reset driver.
The clock tables have been updated to represent the CPG driver and only
implement the generic CPG proxy driver bind call, which binds the clock
and reset drivers.
The DM_RESET is now enabled for all R-Car Gen3 platforms.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Synchronize R-Car R8A779A0 V3U clock tables with Linux 6.1.7,
commit 21e996306a6afaae88295858de0ffb8955173a15 .
Rename CLK_TYPE_R8A779A0_ to CLK_TYPE_GEN4_ to match the new
clock tables. Add CLK_TYPE_GEN4_SD, CLK_TYPE_GEN4_RPC and
CLK_TYPE_GEN4_RPCD2 macros and handling into Gen3 CPG core.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Currently, SDnH is handled together with SDn. This caused lots of
problems, so we want SDnH as a separate clock. Introduce a dummy SDnH
type here which creates a fixed-factor clock with factor 1. That allows
us to convert the per-SoC CPG drivers while keeping the old behaviour
for now. A later patch then will add the proper functionality.
Based on Linux series by Wolfram Sang:
commit a31cf51bf6b4b ("clk: renesas: rcar-gen3: Add dummy SDnH clock"),
commit 1abd04480866c ("clk: renesas: rcar-gen3: Add SDnH clock"),
commit 63494b6f98f26 ("clk: renesas: r8a779a0: Add SDnH clock to V3U")
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org> # Switch to gen3_clk_get_rate64
Somewhere along the way, someone misspelt "invalid" and it got copied
everywhere. Fix it.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Alexander Dahl <ada@thorsis.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add clock tables for R8A779A0 V3U SoC from Linux 5.12,
commit 9f4ad9e425a1 ("Linux 5.12")
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
--
Marek: - Add .reset_modemr_offset
- Sync tables from Linux 5.12
- Rebase on latest u-boot
On R8A779A0 V3U SoC, PLL1 and PLL5 use a divider value
from cpg_pll_configs table while PLL{20,21,30,31,4} use
different control offset. Introduce new types to handle
this and handle those types in the Gen3 clock code.
Based on "clk: renesas: Add support for R8A779A0 V3U PLLn"
by Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Most of the PLLx, MAIN, FIXED clock handlers are calling very similar
code, which determines parent rate and then applies multiplication and
division. The only difference is whether multiplication is fixed factor
or coming from CRx register. Deduplicate the code into a single function.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Base on Linux v5.10-rc2, commit 8b652aa8a1fb by Yoshihiro Shimoda
To support other register layouts in the future, add register pointers
of {control,status,reset,reset_clear}_regs into struct cpg_mssr_info
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
CPG IP in some specific Renesas SoCs (i.e. new R8A779A0 V3U SoC)
requires a different setting procedure. Make struct cpg_mssr_info
accessible to handle the clock setting in that case.
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
The MODEMR register offset changed on R8A779A0, make the MODEMR offset
configurable. Fill the offset in on all clock drivers. No functional
change.
Based off "clk: renesas: Make CPG Reset MODEMR offset accessible from
struct cpg_mssr_info" by Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
This supports RPCD2 clock handling. While at it, add the check point
for RPC-IF clock RPCD2 Frequency Division Ratio, since it must be odd
number
Signed-off-by: Hai Pham <hai.pham.ud@renesas.com>
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Move this out of the common header and include it only where needed. In
a number of cases this requires adding "struct udevice;" to avoid adding
another large header or in other cases replacing / adding missing header
files that had been pulled in, very indirectly. Finally, we have a few
cases where we did not need to include <asm/global_data.h> at all, so
remove that include.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Use the _ptr suffixed variant instead of casting. Also, convert it to
dev_read_addr_ptr(), which is safe to CONFIG_OF_LIVE.
One curious part is an error check like follows in
drivers/watchdog/omap_wdt.c:
priv->regs = (struct wd_timer *)devfdt_get_addr(dev);
if (!priv->regs)
return -EINVAL;
devfdt_get_addr() returns FDT_ADDR_T_NONE (i.e. -1) on error.
So, this code does not catch any error in DT parsing.
dev_read_addr_ptr() returns NULL on error, so this error check
will work.
I generated this commit by the following command:
$ find . -name .git -prune -o -name '*.[ch]' -type f -print | \
xargs sed -i -e 's/([^*)]*\*)devfdt_get_addr(/dev_read_addr_ptr(/'
I manually fixed drivers/usb/host/ehci-mx6.c
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Use the _ptr suffixed variant instead of casting. Also, convert it to
dev_read_addr_ptr(), which is safe to CONFIG_OF_LIVE.
One curious part is an error check like follows in
drivers/watchdog/omap_wdt.c:
priv->regs = (struct wd_timer *)devfdt_get_addr(dev);
if (!priv->regs)
return -EINVAL;
devfdt_get_addr() returns FDT_ADDR_T_NONE (i.e. -1) on error.
So, this code does not catch any error in DT parsing.
dev_read_addr_ptr() returns NULL on error, so this error check
will work.
I generated this commit by the following command:
$ find . -name .git -prune -o -name '*.[ch]' -type f -print | \
xargs sed -i -e 's/([^*)]*\*)devfdt_get_addr(/dev_read_addr_ptr(/'
I manually fixed drivers/usb/host/ehci-mx6.c
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
The SDHI clock must be configured differently for HS200/HS400/SDR104
modes. Add support for reconfiguring the SDHI clock settings into the
clock driver.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
The PE clock have two parents, add support for picking the correct
one and deriving the clock from it.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
The PLL rate could be in the GHz range, which could overflow a 32bit
data type. Since the hardware is 64bit anyway, pass the clock rates
as 64bit number internally to avoid this.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
The mul and div arguments were reported in reverse order in the debug
message, swap them to fix this.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
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 SD driver calls clk_set_rate() before clk_enable(), yet clk_set_rate()
implementation in the clock driver does not set the SD-IF divider. Fix it.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Pull code which is common for RCar Gen2 and RCar Gen3 into
separate source file. No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
The Gen2 requires setting RMSTPCR before booting, while on Gen3 this
is thus far always zero. Split the tables so the RMSTPCR can be set
too.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Not all SoCs have the same PLL configuration options,
so make those PLL configuraion tables per-SoC.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Not all drivers use the same IDs, so make those IDs per-driver.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Split the massive driver into smaller per-SoC drivers and pull the
common code into a separate file. This would allow configuring out
unnecessary clock drivers once the Kconfig changes are in and also
allow adding more clock tables easily.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Fix a copy-paste typo in the clock table entry, s/7795/7796/.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Add xHCI entry into the clock tables, so that the xHCI USB driver
can enable the clock for the xHCI block via clock framework.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
The MSTP registers were poked in boards/renesas/rcar-common/common.c
in arch_preboot_os hook thus far to shut down the clock before Linux
takes over. With DM, this is no longer needed and we can do the same
in the clock driver .remove callback. This patch adds such a .remove
callback for R8A7795 and R8A7796.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com>
Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
