if of_find_device_by_node() succeed, dwc3_octeon_device_init() doesn't have
a corresponding put_device(). Thus add put_device() to fix the exception
handling for this function implementation.

Fixes: 93e502b3 ("MIPS: OCTEON: Platform support for OCTEON III USB controller")
Signed-off-by: NYu Kuai <yukuai3@huawei.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

paravirt machine was introduced for Cavium's partial virtualization
technology, however, it's host side support and QEMU support never
landed in upstream.

As Cavium was acquired by Marvel and they have no intention to maintain
their MIPS product line, also paravirt is unlikely to be utilized by
community users, it's time to retire it if nobody steps in to maintain
it.
Signed-off-by: NJiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Add Thomas and myself as maintainers of the MIPS CPU and GIC IRQchip, MIPS
GIC timer and MIPS CPS CPUidle drivers.
Signed-off-by: NSerge Semin <Sergey.Semin@baikalelectronics.ru>
Acked-by: NMarc Zyngier <maz@kernel.org>
Acked-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

CDMM may be available not only on MIPS R2 architectures, but also on
newer MIPS R5 chips. For instance our P5600 chip has one. Let's mark
the CDMM bus being supported for that MIPS arch too.
Signed-off-by: NSerge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Since having and mapping the CDMM block is platform specific, then
instead of just returning a zero-address, lets make the default CDMM
base address search method (mips_cdmm_phys_base()) to do something
useful. For instance to find the address in a dedicated dtb-node in
order to support of-based platforms by default.
Signed-off-by: NSerge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

It's a Common Device Memory Map controller embedded into the MIPS IP
cores, which dts node is supposed to have compatible and reg properties.
Signed-off-by: NSerge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: NRob Herring <robh@kernel.org>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Modern device tree bindings are supposed to be created as YAML-files
in accordance with DT schema. This commit replaces MIPS GIC legacy bare
text binding with YAML file. As before the binding file states that the
corresponding dts node is supposed to be compatible with MIPS Global
Interrupt Controller indicated by the "mti,gic" compatible string and
to provide a mandatory interrupt-controller and '#interrupt-cells'
properties. There might be optional registers memory range,
"mti,reserved-cpu-vectors" and "mti,reserved-ipi-vectors" properties
specified.

MIPS GIC also includes a free-running global timer, per-CPU count/compare
timers, and a watchdog. Since currently the GIC Timer is only supported the
DT schema expects an IRQ and clock-phandler charged timer sub-node with
"mti,mips-gic-timer" compatible string.
Signed-off-by: NSerge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: NRob Herring <robh@kernel.org>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

It's a Cluster Power Controller embedded into the MIPS IP cores.
Currently the corresponding dts node is supposed to have compatible
and reg properties.
Signed-off-by: NSerge Semin <Sergey.Semin@baikalelectronics.ru>
Reviewed-by: NRob Herring <robh@kernel.org>
Cc: Alexey Malahov <Alexey.Malahov@baikalelectronics.ru>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

In the MIPS architecture, we should clear the security-relevant
flag READ_IMPLIES_EXEC in the function SET_PERSONALITY2() of the
file arch/mips/include/asm/elf.h.

Otherwise, with this flag set, PROT_READ implies PROT_EXEC for
mmap to make memory executable that is not safe, because this
condition allows an attacker to simply jump to and execute bytes
that are considered to be just data [1].

In mm/mmap.c:
unsigned long do_mmap(struct file *file, unsigned long addr,
			unsigned long len, unsigned long prot,
			unsigned long flags, vm_flags_t vm_flags,
			unsigned long pgoff, unsigned long *populate,
			struct list_head *uf)
{
	[...]
	if ((prot & PROT_READ) && (current->personality & READ_IMPLIES_EXEC))
		if (!(file && path_noexec(&file->f_path)))
			prot |= PROT_EXEC;
	[...]
}

By the way, x86 and ARM64 have done the similar thing.

After commit 250c2277 ("x86_64: move kernel"), in the file
arch/x86/kernel/process_64.c:
void set_personality_64bit(void)
{
	[...]
	current->personality &= ~READ_IMPLIES_EXEC;
}

After commit 48f99c8e ("arm64: Preventing READ_IMPLIES_EXEC
propagation"), in the file arch/arm64/include/asm/elf.h:
#define SET_PERSONALITY(ex)						\
({									\
	clear_thread_flag(TIF_32BIT);					\
	current->personality &= ~READ_IMPLIES_EXEC;			\
})

[1] https://insights.sei.cmu.edu/cert/2014/02/feeling-insecure-blame-your-parent.htmlReported-by: NJuxin Gao <gaojuxin@loongson.cn>
Co-developed-by: NJuxin Gao <gaojuxin@loongson.cn>
Signed-off-by: NJuxin Gao <gaojuxin@loongson.cn>
Signed-off-by: NTiezhu Yang <yangtiezhu@loongson.cn>
Reviewed-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Close "fd" before the return of map_vdso() and close "out_file"
in main().
Signed-off-by: NPeng Fan <fanpeng@loongson.cn>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Refresh CU1000-Neo's defconfig to support LED.
Tested-by: N周正 (Zhou Zheng) <sernia.zhou@foxmail.com>
Signed-off-by: N周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

1.The CU1000-Neo board actually uses X1000E instead of X1000,
  so the wrongly written "ingenic,x1000" in compatible should
  be changed to "ingenic,x1000e".
2.Adjust the order of nodes according to the corresponding
  address value.
3.Drop unnecessary node in "wlan_pwrseq".
4.Add the leds node to "cu1000-neo.dts".
Tested-by: N周正 (Zhou Zheng) <sernia.zhou@foxmail.com>
Signed-off-by: N周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Add a device tree and a defconfig for the Ingenic X1830 based
YSH & ATIL CU Neo board.
Tested-by: N周正 (Zhou Zheng) <sernia.zhou@foxmail.com>
Signed-off-by: N周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

1.Add bindings for Ingenic X1830 based board, prepare for later dts.
2.The CU1000-Neo board actually uses X1000E instead of X1000, so
  the wrongly written "ingenic,x1000" in bindings should be changed
  to "ingenic,x1000e", the corresponding dts file modification will
  be made in a patch later in this series.
Tested-by: N周正 (Zhou Zheng) <sernia.zhou@foxmail.com>
Signed-off-by: N周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Support the Ingenic X1830 SoC using the code under arch/mips/jz4740.
This is left unselectable in Kconfig until a X1830 based board is
added in a later commit.
Tested-by: N周正 (Zhou Zheng) <sernia.zhou@foxmail.com>
Signed-off-by: N周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

sparse report build warning as follows:
arch/mips/vdso/vdso-n32-image.c:13:35:
incorrect type in assignment (different address spaces) @@
expected void *[usertype] vdso @@     got void [noderef] <asn:1> * @@
Reported-by: Nkernel test robot <lkp@intel.com>
Signed-off-by: NSunguoyun <sunguoyun@loongson.cn>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

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 neither `\bgnu\.org/license`, nor `\bmozilla\.org/MPL\b`:
            If both the HTTP and HTTPS versions
            return 200 OK and serve the same content:
              Replace HTTP with HTTPS.
Signed-off-by: NAlexander A. Klimov <grandmaster@al2klimov.de>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Load correct devicetree according to PRID and PCH type.
Signed-off-by: NHuacai Chen <chenhc@lemote.com>
Signed-off-by: NJiaxun Yang <jiaxun.yang@flygoat.com>
Tested-by: NTiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Add DeviceTree files for Classic Loongson64 Quad Core + LS7A boards and
Generic Loongson64 Quad Core + LS7A boards.
Signed-off-by: NHuacai Chen <chenhc@lemote.com>
Signed-off-by: NJiaxun Yang <jiaxun.yang@flygoat.com>
Tested-by: NTiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

From previous commits, the machine names with "loongson3-" prefix have
renamed to "loongson64c-" prefix in documents, but the .dts files have
not been updated as well. So fix it.
Signed-off-by: NHuacai Chen <chenhc@lemote.com>
Tested-by: NTiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Replace hwmon_device_register() with hwmon_device_register_with_info()
to fix the following boot warning :

[    9.029924] Loongson Hwmon Enter...
[    9.106850] (NULL device *): hwmon_device_register() is deprecated. Please convert the driver to use hwmon_device_register_with_info().
Signed-off-by: NZhi Li <lizhi01@loongson.cn>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Once the temperature of any CPUs is too high, it can power off immediately,
no need to check the rest of CPUs, and it is better to print a log before
power off, this is useful when analysis the abnormal issues.
Signed-off-by: NTiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: NZhi Li <lizhi01@loongson.cn>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Fix the following checkpatch warnings and errors:

ERROR: do not initialise statics to 0
+static int csr_temp_enable = 0;

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
+static SENSOR_DEVICE_ATTR(name, S_IRUGO, get_hwmon_name, NULL, 0);

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
+static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, get_cpu_temp, NULL, 1);

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
+static SENSOR_DEVICE_ATTR(temp1_label, S_IRUGO, cpu_temp_label, NULL, 1);

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
+static SENSOR_DEVICE_ATTR(temp2_input, S_IRUGO, get_cpu_temp, NULL, 2);

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
+static SENSOR_DEVICE_ATTR(temp2_label, S_IRUGO, cpu_temp_label, NULL, 2);

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
+static SENSOR_DEVICE_ATTR(temp3_input, S_IRUGO, get_cpu_temp, NULL, 3);

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
+static SENSOR_DEVICE_ATTR(temp3_label, S_IRUGO, cpu_temp_label, NULL, 3);

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
+static SENSOR_DEVICE_ATTR(temp4_input, S_IRUGO, get_cpu_temp, NULL, 4);

WARNING: Symbolic permissions 'S_IRUGO' are not preferred. Consider using octal permissions '0444'.
+static SENSOR_DEVICE_ATTR(temp4_label, S_IRUGO, cpu_temp_label, NULL, 4);

WARNING: Missing a blank line after declarations
+	int id = (to_sensor_dev_attr(attr))->index - 1;
+	return sprintf(buf, "CPU %d Temperature\n", id);

WARNING: Missing a blank line after declarations
+	int value = loongson3_cpu_temp(id);
+	return sprintf(buf, "%d\n", value);

ERROR: spaces required around that '=' (ctx:VxV)
+	for (i=0; i<nr_packages; i++)
 	      ^

ERROR: spaces required around that '<' (ctx:VxV)
+	for (i=0; i<nr_packages; i++)
 	           ^

ERROR: spaces required around that '=' (ctx:VxV)
+	for (i=0; i<nr_packages; i++)
 	      ^

ERROR: spaces required around that '<' (ctx:VxV)
+	for (i=0; i<nr_packages; i++)
 	           ^

ERROR: spaces required around that '=' (ctx:VxV)
+	for (i=0; i<nr_packages; i++) {
 	      ^

ERROR: spaces required around that '<' (ctx:VxV)
+	for (i=0; i<nr_packages; i++) {
 	           ^

WARNING: line over 80 characters
+		csr_temp_enable = csr_readl(LOONGSON_CSR_FEATURES) & LOONGSON_CSRF_TEMP;
Signed-off-by: NTiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: NZhi Li <lizhi01@loongson.cn>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Add a basic default config for the RS-90 RetroMini board.
Signed-off-by: NPaul Cercueil <paul@crapouillou.net>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

The RS-90, better known as RetroMini, is a small and pocketable handheld
gaming console from YLMChina. It has little more than a JZ4725B SoC, a
NAND, a screen, some buttons and a speaker.
Signed-off-by: NPaul Cercueil <paul@crapouillou.net>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Add preliminary support for boards based on the JZ4725B SoC from
Ingenic.

The JZ4725B SoC is supposed to be older than the JZ4740 SoC, but its
internals are much closer to what can be found on the JZ4750 and newer
SoCs.

It is low-power SoC with a MIPS32r1 SoC running at ~360 MHz, and no FPU.
Signed-off-by: NPaul Cercueil <paul@crapouillou.net>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Use an enum instead of macros to represent the various versions of the
Ingenic SoCs, and add some of the SoC versions that were previously
missing.
Signed-off-by: NPaul Cercueil <paul@crapouillou.net>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Add compatible strings for the PWM and watchdog IPs on the Ingenic
JZ4725B SoC.
Signed-off-by: NPaul Cercueil <paul@crapouillou.net>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Add an entry to ingenic/devices.yaml for the JZ4725B-based
YLM "RetroMini" RS-90.
Signed-off-by: NPaul Cercueil <paul@crapouillou.net>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Shenzhen Yangliming Electronic Technology Co., Ltd., abbreviated YLM or
YLMChina, and known as Anbernic in the rest of the world, is a Chinese
manufacturer of handheld game consoles, some of which are known to be
running Linux.
Signed-off-by: NPaul Cercueil <paul@crapouillou.net>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

sparse report build warning as follows:

drivers/platform/mips/rs780e-acpi.c:72:6: warning:
 symbol 'acpi_registers_setup' was not declared. Should it be static?

And function acpi_registers_setup() is not used outside of this file,
so marks it static.
Reported-by: NHulk Robot <hulkci@huawei.com>
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

The size of ioports in the current RS780E dts file is not enough, which
sometimes causes device initialize fail. So we increase the size of ISA/
LPC ioports to 0x4000, and increase the size of PCI ioports to 0x8000.
Signed-off-by: NHuacai Chen <chenhc@lemote.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Document loongson64c-4core-ls7a and loongson64g-4core-ls7a, two boards
with LS7A PCH.
Signed-off-by: NHuacai Chen <chenhc@lemote.com>
Signed-off-by: NJiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Loongson-3A R1/R2/R3 and Loongson-3B R1/R2 use the same package naming
in dts, and Loongson-3A R4 will be different. In cpu.h the classic 64bit
Loongson processors are called Loongson64C (C for classic, pre Loongson-
3A R4), and the new 64bit Loongson processors are called Loongson64G (G
for generic, Loongson-3A R4+). To keep consistency and make extensible,
we rename the classic "loongson3" prefix to "loongson64c", and the new
prefix for Loongson-3A R4+ will be "loongson64g".
Signed-off-by: NHuacai Chen <chenhc@lemote.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Adjust IRQ layout in order to use IRQ resources more efficiently, which
is done by adjusting NR_IRQS and MIPS_CPU_IRQ_BASE.

Before this patch:
0~15:    ISA/LPC IRQs;
16~55:   Dynamic IRQs;
56~63:   MIPS CPU IRQs;
64~127:  PCH IRQs;
128~255: Dynamic IRQs.

After this patch:
0~15:    ISA/LPC IRQs;
16~23:   MIPS CPU IRQs;
24~87:   PCH IRQs;
88~280:  Dynamic IRQs.
Signed-off-by: NHuacai Chen <chenhc@lemote.com>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

Other vendor-defined registers use the vendor name as a prefix, not an
infix, so unify the naming style of CP0.Config6 bits.
Suggested-by: NMaciej W. Rozycki" <macro@linux-mips.org>
Signed-off-by: NHuacai Chen <chenhc@lemote.com>
Reviewed-by: NMaciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

While applying commit 9909bc43 ("dt-bindings: MIPS: Document Ingenic
 SoCs binding.") I've messed up by "fixing" indentation in a C style,
which is wrong for yaml files. Replace tabs back to spaces.

Fixes: 9909bc43 ("dt-bindings: MIPS: Document Ingenic SoCs binding.")
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

csum_fold() in MIPS' asm/checksum.h is another source of sparse flooding
when building different networking source code.
The thing is that only half of __wsum <--> u32 casts inside the function
is forced, which is insufficient.
Add all necessary forced typecasting to stop floods and simplify actual
bug hunting.
Signed-off-by: NAlexander Lobakin <alobakin@pm.me>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

MIPS MMIO macros for byteswapping from/to hardware endianness are a bit
tricky because they use cpu_to_le{16,32,64}() in both directions.
This generates a lot of questions from sparse as __le{16,32,64} types
are 'restricted' and direct cast is forbidden in order to prevent messing
up the byteorder.
As MMIO ops are used in almost every single driver, this leads to console
flooding and complicates bug hunting.

We could fix it in a more proper way, i.e. separate from device /
to device byteswap macros and expand __BUILD_MEMORY_*(), but this seems
redundant and will produce code duplication.
Instead, just expand the existing *ioswab*() macros with forced
typecasting to stop floods.
Signed-off-by: NAlexander Lobakin <alobakin@pm.me>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>

*readq() family operates with u64 arguments, so they need 64-bit
byteswaps.
Correct macros for Generic MIPS and IP-32 to match other machines'
implementations.
Signed-off-by: NAlexander Lobakin <alobakin@pm.me>
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>