Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH v14 06/16] of/fdt: add helper functions for handling
 properties
To:     AKASHI Takahiro <takahiro.akashi@linaro.org>,
        catalin.marinas@arm.com, will.deacon@arm.com, dhowells@redhat.com,
        vgoyal@redhat.com, herbert@gondor.apana.org.au,
        davem@davemloft.net, dyoung@redhat.com, bhe@redhat.com,
        arnd@arndb.de, schwidefsky@de.ibm.com, heiko.carstens@de.ibm.com,
        prudo@linux.ibm.com, ard.biesheuvel@linaro.org,
        james.morse@arm.com, bhsharma@redhat.com,
        kexec@lists.infradead.org, linux-arm-kernel@lists.infradead.org,
        linux-kernel@vger.kernel.org, Rob Herring <robh+dt@kernel.org>
References: <20180907080040.4967-1-takahiro.akashi@linaro.org>
 <a535515a-92f1-fc33-bdb9-6f5e0392cb11@gmail.com>
 <20180910023819.GA30795@linaro.org>
From:   Frank Rowand <frowand.list@gmail.com>
Message-ID: <59d0bc6f-ab71-eb21-d438-da3cd09aaa79@gmail.com>
Date:   Thu, 13 Sep 2018 18:42:35 -0700
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.9.1
MIME-Version: 1.0
In-Reply-To: <20180910023819.GA30795@linaro.org>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 09/09/18 19:38, AKASHI Takahiro wrote:
> Frank,
> 
> Thank you for the comments. I will address all of them, but
> have one question:
> 
> On Fri, Sep 07, 2018 at 12:53:58PM -0700, Frank Rowand wrote:
>> On 09/07/18 01:00, AKASHI Takahiro wrote:
>>> These functions will be used later to handle kexec-specific properties
>>> in arm64's kexec_file implementation.
>>>
>>> Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
>>> Cc: Rob Herring <robh+dt@kernel.org>
>>> Cc: Frank Rowand <frowand.list@gmail.com>
>>> ---
>>>  drivers/of/fdt.c       | 62 ++++++++++++++++++++++++++++++++++++++++--
>>>  include/linux/of_fdt.h | 10 +++++--
>>>  2 files changed, 68 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c
>>> index 800ad252cf9c..dc960cea1355 100644
>>> --- a/drivers/of/fdt.c
>>> +++ b/drivers/of/fdt.c
>>> @@ -25,6 +25,7 @@
>>>  #include <linux/debugfs.h>
>>>  #include <linux/serial_core.h>
>>>  #include <linux/sysfs.h>
>>> +#include <linux/types.h>
>>>  
>>>  #include <asm/setup.h>  /* for COMMAND_LINE_SIZE */
>>>  #include <asm/page.h>
>>> @@ -537,8 +538,8 @@ void *of_fdt_unflatten_tree(const unsigned long *blob,
>>>  EXPORT_SYMBOL_GPL(of_fdt_unflatten_tree);
>>>  
>>>  /* Everything below here references initial_boot_params directly. */
>>> -int __initdata dt_root_addr_cells;
>>> -int __initdata dt_root_size_cells;
>>> +int dt_root_addr_cells;
>>> +int dt_root_size_cells;
>>>  
>>>  void *initial_boot_params;
>>>  
>>> @@ -1323,3 +1324,60 @@ late_initcall(of_fdt_raw_init);
>>>  #endif
>>>  
>>>  #endif /* CONFIG_OF_EARLY_FLATTREE */
>>> +
>>
>> Please add comment:
>>
>> /* helper functions for arm64 kexec */
>>
>>
>>> +bool of_fdt_cells_size_fitted(u64 base, u64 size)
>>
>> Please rename as of_fdt_range_valid()
>>
>>
>>> +{
>>> +	/* if *_cells >= 2, cells can hold 64-bit values anyway */
>>> +	if ((dt_root_addr_cells == 1) && (base > U32_MAX))
>>> +		return false;
>>> +
>>> +	if ((dt_root_size_cells == 1) && (size > U32_MAX))
>>> +		return false;
>>
>> Should also check that base + size does not wrap around.
> 
> What is the upper limit here?
> For instance, #address_cells = <1> and #size_cells = <1>,
> and can 'base + size' be over U32_MAX?
> Assuming 'not' is quite reasonable, but it seems to me
> that devicetree spec doesn't exclude it, as least I couldn't
> find any notes about such a case.
> (In my understands, #address_cells only restricts a size in 'reg' property.)

(See my other reply in this thread a few minutes ago -- the context of the
ranges is essentially what is valid for the "reg" property of the "/memory"
nodes.)

The "Devicetree Specification" does not specify whether a memory range can
wrap around.  For example, can a region of size 0x2000 that wraps around
the highest 32 bit address be specified as:

   / memory@0xfffff000 {
        reg = <0xfffff000 0x2000>;
     }

or must it be specified as (option 1):

   / memory@0xfffff000 {
        reg = <0xfffff000 0x1000 0x0 0x1000>
     }

or (option 2):

   / memory@0xfffff000 {
        reg = <0xfffff000 0x1000>;
     }
     memory@0 {
        reg = <0x0 0x1000>;
     }


I suggest you start a thread on the devicetree specification list asking
that the spec be updated to state whether wrap around is allowed.


> 
> Thanks,
> -Takahiro Akashi
> 
>>
>>> +
>>> +	return true;
>>> +}
>>> +
>>> +size_t of_fdt_reg_cells_size(void)
>>
>> Please rename as of_fdt_root_range_size()
>>
>>
>>> +{
>>> +	return (dt_root_addr_cells + dt_root_size_cells) * sizeof(u32);
>>> +}
>>> +
>>> +#define FDT_ALIGN(x, a)	(((x) + (a) - 1) & ~((a) - 1))
>>> +#define FDT_TAGALIGN(x)	(FDT_ALIGN((x), FDT_TAGSIZE))
>>> +
>>> +int fdt_prop_len(const char *prop_name, int len)
>>
>> Please rename as fdt_len_added_prop()
>>
>>
>>> +{
>>> +	return (strlen(prop_name) + 1) +
>>> +		sizeof(struct fdt_property) +
>>> +		FDT_TAGALIGN(len);
>>> +}
>>> +
>>
>> Please add comment, something like:
>>
>> /* cells must be 1 or 2 */
>>
>>
>>> +static void fill_property(void *buf, u64 val64, int cells)
>>
>> Please rename as cpu64_to_fdt_cells()
>>
>> Thanks,
>>
>> Frank
>>
>>> +{
>>> +	__be32 val32;
>>> +
>>> +	while (cells) {
>>> +		val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX);
>>> +		memcpy(buf, &val32, sizeof(val32));
>>> +		buf += sizeof(val32);
>>> +	}
>>> +}
>>> +
>>> +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name,
>>> +						u64 addr, u64 size)
>>> +{
>>> +	char buf[sizeof(__be32) * 2 * 2];
>>> +		/* assume dt_root_[addr|size]_cells <= 2 */
>>> +	void *prop;
>>> +	size_t buf_size;
>>> +
>>> +	buf_size = of_fdt_reg_cells_size();
>>> +	prop = buf;
>>> +
>>> +	fill_property(prop, addr, dt_root_addr_cells);
>>> +	prop += dt_root_addr_cells * sizeof(u32);
>>> +
>>> +	fill_property(prop, size, dt_root_size_cells);
>>> +
>>> +	return fdt_setprop(fdt, nodeoffset, name, buf, buf_size);
>>> +}
>>> diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h
>>> index b9cd9ebdf9b9..9615d6142578 100644
>>> --- a/include/linux/of_fdt.h
>>> +++ b/include/linux/of_fdt.h
>>> @@ -37,8 +37,8 @@ extern void *of_fdt_unflatten_tree(const unsigned long *blob,
>>>  				   struct device_node **mynodes);
>>>  
>>>  /* TBD: Temporary export of fdt globals - remove when code fully merged */
>>> -extern int __initdata dt_root_addr_cells;
>>> -extern int __initdata dt_root_size_cells;
>>> +extern int dt_root_addr_cells;
>>> +extern int dt_root_size_cells;
>>>  extern void *initial_boot_params;
>>>  
>>>  extern char __dtb_start[];
>>> @@ -108,5 +108,11 @@ static inline void unflatten_device_tree(void) {}
>>>  static inline void unflatten_and_copy_device_tree(void) {}
>>>  #endif /* CONFIG_OF_EARLY_FLATTREE */
>>>  
>>> +bool of_fdt_cells_size_fitted(u64 base, u64 size);
>>> +size_t of_fdt_reg_cells_size(void);
>>> +int fdt_prop_len(const char *prop_name, int len);
>>> +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name,
>>> +						u64 addr, u64 size);
>>> +
>>>  #endif /* __ASSEMBLY__ */
>>>  #endif /* _LINUX_OF_FDT_H */
>>>
>>
>