syscon_node_to_regmap() will make the created regmap get and enable the
first clock it can parse from the device tree. This clock is not needed to
access the registers and should not be enabled at that time.
Use device_node_to_regmap to resolve this as it looks up the regmap in
the same list but doesn't care about the clocks. This issue is detected
by lockdep when booting the sama5d3 with a device tree containing the
new clk bindings.
This fix already happened in 6956eb33ab ("clk: at91: fix possible
deadlock") for the drivers that had been migrated to the new clk binding
back then. This does the same for the new drivers as well.
Fixes: 01e2113de9 ("clk: at91: add sam9x60 pmc driver")
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.kernel.org/r/20200703073236.23923-1-a.fatoum@pengutronix.de
Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
For test and debug purposes, it's simple enough to enable or disable
clocks from shell. Add a new debugfs file 'clk_prepare_enable' that
calls clk_prepare_enable() when writing "1" and clk_disable_unprepare()
when writing "0".
This can have security implications, so only support it when the code
has been modified to #define CLOCK_ALLOW_WRITE_DEBUGFS.
Signed-off-by: Mike Tipton <mdtipton@codeaurora.org>
Link: https://lore.kernel.org/r/20200630003024.6282-1-mdtipton@codeaurora.org
[sboyd@kernel.org: Reword commit text and remove comment update]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Fixed ops were already identical, this adds support for non-fixed ops by
sharing between trion and lucid.
This also changes the names for trion ops to be consistent with the rest.
Note LUCID_PCAL_DONE is renamed to TRION_PCAL_DONE because it is wrong for
lucid, LUCID_PCAL_DONE should be BIT(27). Next patch will address this.
Signed-off-by: Jonathan Marek <jonathan@marek.ca>
Tested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/20200709135251.643-4-jonathan@marek.ca
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Reusing the generic struct vc5_hw_data for all blocks is handy. However it
implies we allocate space the div_int and div_frc fields even for the
output drivers where they are unused, and the clk_output_cfg0 and
clk_output_cfg0_mask fields for all components even though they are used
only for the output drivers.
Use a dedicated struct for the output drivers so that each block uses
exactly the fields it needs, not more.
Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Link: https://lore.kernel.org/r/20200723072603.1795-1-luca@lucaceresoli.net
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
When getting the names of the child nodes, kasprintf is used to
allocate memory which is used to create the string for the node
name. Unfortunately, there is no memory check to determine
if this allocation fails, it may cause an error when trying
to get child node name.
This patch will check if the memory allocation fails, and returns
and -ENOMEM error instead of blindly moving on.
Fixes: 260249f929 ("clk: vc5: Enable addition output configurations of the Versaclock")
Suggested-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Link: https://lore.kernel.org/r/20200716122620.4538-1-aford173@gmail.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
There are a several places where printing an error message of
init.name occurs after init.name has been kfree'd. Also the failure
message is duplicated each time in the code. Fix this by adding
a registration error failure path for these cases, moving the
duplicated error messages to one common point and kfree'ing init.name
only after it has been used.
Changes also shrink the object code size by 171 bytes (x86-64, gcc 9.3):
Before:
text data bss dec hex filename
21057 3960 64 25081 61f9 drivers/clk/clk-versaclock5.o
After:
text data bss dec hex filename
20886 3960 64 24910 614e drivers/clk/clk-versaclock5.o
Addresses-Coverity: ("Use after free")
Fixes: f491276a51 ("clk: vc5: Allow Versaclock driver to support multiple instances")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Link: https://lore.kernel.org/r/20200625132736.88832-1-colin.king@canonical.com
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
[sboyd@kernel.org: Drop stray newline]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
* clk-qcom:
clk: qcom: gcc: Make disp gpll0 branch aon for sc7180/sdm845
ipq806x: gcc: add support for child probe
clk: qcom: msm8996: Make symbol 'cpu_msm8996_clks' static
clk: qcom: ipq8074: Add correct index for PCIe clocks
The display gpll0 branch clock inside GCC needs to always be enabled.
Otherwise the AHB clk (disp_cc_mdss_ahb_clk_src) for the display clk
controller (dispcc) will stop clocking while sourcing from gpll0 when
this branch inside GCC is turned off during unused clk disabling. We can
never turn this branch off because the AHB clk for the display subsystem
is needed to read/write any registers inside the display subsystem
including clk related ones. This makes this branch a really easy way to
turn off AHB access to the display subsystem and cause all sorts of
mayhem. Let's just make the clk ops keep the clk enabled forever and
ignore any attempts to disable this clk so that dispcc accesses keep
working.
Signed-off-by: Taniya Das <tdas@codeaurora.org>
Reported-by: Evan Green <evgreen@chromium.org>
Link: https://lore.kernel.org/r/1594796050-14511-1-git-send-email-tdas@codeaurora.org
Fixes: 17269568f7 ("clk: qcom: Add Global Clock controller (GCC) driver for SC7180")
Fixes: 06391eddb6 ("clk: qcom: Add Global Clock controller (GCC) driver for SDM845")
[sboyd@kernel.org: Fill out commit text more]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Pull clk fixes from Stephen Boyd:
"A couple build fixes for issues exposed this merge window and a fix
for the eMMC clk on AST2600 SoCs that fixes the rate that is
calculated by the clk framework"
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: Specify IOMEM dependency for HSDK pll driver
clk: AST2600: Add mux for EMMC clock
clk: mvebu: ARMADA_AP_CPU_CLK needs to select ARMADA_AP_CP_HELPER
Currently we are not initializing the scmi clock with discrete rates
correctly. We fetch the min_rate and max_rate value only for clocks with
ranges and ignore the ones with discrete rates. This will lead to wrong
initialization of rate range when clock supports discrete rate.
Fix this by using the first and the last rate in the sorted list of the
discrete clock rates while registering the clock.
Link: https://lore.kernel.org/r/20200709081705.46084-2-sudeep.holla@arm.com
Fixes: 6d6a1d82ea ("clk: add support for clocks provided by SCMI")
Reviewed-by: Stephen Boyd <sboyd@kernel.org>
Reported-and-tested-by: Dien Pham <dien.pham.ry@renesas.com>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
The HSDK pll driver uses the devm_ioremap_resource function, but does
not specify a dependency on IOMEM in Kconfig. This causes a build
failure on architectures without IOMEM, for example, UML (notably with
make allyesconfig).
Fix this by making CONFIG_CLK_HSDK depend on CONFIG_IOMEM.
Signed-off-by: David Gow <davidgow@google.com>
Link: https://lore.kernel.org/r/20200630043214.1080961-1-davidgow@google.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
When building arm32 allmodconfig:
ld.lld: error: undefined symbol: ap_cp_unique_name
>>> referenced by ap-cpu-clk.c
>>> clk/mvebu/ap-cpu-clk.o:(ap_cpu_clock_probe) in archive drivers/built-in.a
ap_cp_unique_name is only compiled into the kernel image when
CONFIG_ARMADA_AP_CP_HELPER is selected (as it is not user selectable).
However, CONFIG_ARMADA_AP_CPU_CLK does not select it.
This has been a problem since the driver was added to the kernel but it
was not built before commit c318ea261749 ("cpufreq: ap806: fix cpufreq
driver needs ap cpu clk") so it was never noticed.
Fixes: f756e362d9 ("clk: mvebu: add CPU clock driver for Armada 7K/8K")
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Link: https://lore.kernel.org/r/20200701201128.2448427-1-natechancellor@gmail.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
* clk-renesas:
clk: renesas: rzg2: Mark RWDT clocks as critical
clk: renesas: rcar-gen3: Mark RWDT clocks as critical
clk: renesas: cpg-mssr: Mark clocks as critical only if on at boot
dt-bindings: clock: renesas: cpg: Convert to json-schema
Rationale:
Reduces attack surface on kernel devs opening the links for MITM
as HTTPS traffic is much harder to manipulate.
Deterministic algorithm:
For each file:
If not .svg:
For each line:
If doesn't contain `\bxmlns\b`:
For each link, `\bhttp://[^# \t\r\n]*(?:\w|/)`:
If both the HTTP and HTTPS versions
return 200 OK and serve the same content:
Replace HTTP with HTTPS.
Signed-off-by: Alexander A. Klimov <grandmaster@al2klimov.de>
Link: https://lore.kernel.org/r/20200703175114.15027-1-grandmaster@al2klimov.de
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
