linux-kernel - Re: Re: [PATCH] ARM64: Setup DMA32 zone size by bootargs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20200816125033.4565-1-phil.chang@mediatek.com>
Date:   Sun, 16 Aug 2020 20:50:34 +0800
From:   Phil Chang <phil.chang@...iatek.com>
To:     <robin.murphy@....com>
CC:     <alix.wu@...iatek.com>, <linux-arm-kernel@...ts.infradead.org>,
        <linux-doc@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <linux-mediatek@...ts.infradead.org>, <phil.chang@...iatek.com>,
        <yj.chiang@...iatek.com>
Subject: Re: Re: [PATCH] ARM64: Setup DMA32 zone size by bootargs

>> this patch allowing the arm64 DMA zone be configurable.
>> 
>> Signed-off-by: Alix Wu <alix.wu@...iatek.com>
>> Signed-off-by: YJ Chiang <yj.chiang@...iatek.com>
>> Signed-off-by: Phil Chang <phil.chang@...iatek.com>
>> ---
>> Hi
>> 
>> For some devices, the main memory split into 2 part due to the memory architecture,
>> the efficient and less inefficient part.
>> One of the use case is fine-tune the dma32 size to contain all the
>> efficient part of memory block on this kind of architecture,
>> but in general, the DMA32 zone size is hardcode to 4G.
>
> The sole point of ZONE_DMA32 is to contain all the 32-bit addressable 
> memory which is not covered by ZONE_DMA. As such the only possible thing 
> that could be configured is the size of ZONE_DMA (if present), and this 
> patch makes no sense.
>
> If you want to describe NUMA characteristics, use NUMA properties. 
> Giving users freedom to break fundamental assumptions of the DMA layer 
> by arbitrarily changing the meaning of "32-bit" is not an appropriate 
> solution.
>
> Robin.

Hi Robin

Thannks for your reply, how about add a condiction to
limit the customize dma32 zone size under 4G ?

>>   .../admin-guide/kernel-parameters.txt         |  3 ++
>>   arch/arm64/include/asm/memory.h               |  2 ++
>>   arch/arm64/mm/init.c                          | 33 +++++++++++++++++--
>>   3 files changed, 35 insertions(+), 3 deletions(-)
>> 
>> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
>> index fb95fad81c79..441ad3cb8ee8 100644
>> --- a/Documentation/admin-guide/kernel-parameters.txt
>> +++ b/Documentation/admin-guide/kernel-parameters.txt
>> @@ -956,6 +956,9 @@
>>   			The filter can be disabled or changed to another
>>   			driver later using sysfs.
>>   
>> +	dma32_zone=nn	[KMG] [KNL,BOOT]
>> +			Forces the DMA32 zone size of <nn> in mb, arm64 only.
>> +
>>   	driver_async_probe=  [KNL]
>>   			List of driver names to be probed asynchronously.
>>   			Format: <driver_name1>,<driver_name2>...
>> diff --git a/arch/arm64/include/asm/memory.h b/arch/arm64/include/asm/memory.h
>> index a1871bb32bb1..377f2252618a 100644
>> --- a/arch/arm64/include/asm/memory.h
>> +++ b/arch/arm64/include/asm/memory.h
>> @@ -174,6 +174,8 @@ extern u64			kimage_vaddr;
>>   /* the offset between the kernel virtual and physical mappings */
>>   extern u64			kimage_voffset;
>>   
>> +extern phys_addr_t		arm_dma_zone_size;
>> +
>>   static inline unsigned long kaslr_offset(void)
>>   {
>>   	return kimage_vaddr - KIMAGE_VADDR;
>> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
>> index 1e93cfc7c47a..642ab323392c 100644
>> --- a/arch/arm64/mm/init.c
>> +++ b/arch/arm64/mm/init.c
>> @@ -60,6 +60,9 @@ EXPORT_SYMBOL(physvirt_offset);
>>   struct page *vmemmap __ro_after_init;
>>   EXPORT_SYMBOL(vmemmap);
>>   
>> +phys_addr_t arm_dma_zone_size __ro_after_init;
>> +EXPORT_SYMBOL(arm_dma_zone_size);
>> +
>>   /*
>>    * We create both ZONE_DMA and ZONE_DMA32. ZONE_DMA covers the first 1G of
>>    * memory as some devices, namely the Raspberry Pi 4, have peripherals with
>> @@ -242,6 +245,25 @@ static int __init early_mem(char *p)
>>   }
>>   early_param("mem", early_mem);
>>   
>> +/*
>> + * Setup the dma32 zone size
>> + */
>> +static int __init setup_dma32_zone(char *p)
>> +{
>> +	if (!p)
>> +		return -EINVAL;
>> +
>> +	if (kstrtoull(p, 0, &arm_dma_zone_size))
>> +		return -EINVAL;
>> +
>> +	arm_dma_zone_size *= SZ_1M;
>> +	pr_notice("Setup dma32 zone size to %llu Mb\n", arm_dma_zone_size);
>> +
>> +	return 0;
>> +}
>> +
>> +early_param("dma32_zone", setup_dma32_zone);
>> +
>>   static int __init early_init_dt_scan_usablemem(unsigned long node,
>>   		const char *uname, int depth, void *data)
>>   {
>> @@ -392,10 +414,15 @@ void __init arm64_memblock_init(void)
>>   		arm64_dma_phys_limit = max_zone_phys(ARM64_ZONE_DMA_BITS);
>>   	}
>>   
>> -	if (IS_ENABLED(CONFIG_ZONE_DMA32))
>> -		arm64_dma32_phys_limit = max_zone_phys(32);
>> -	else
>> +	if (IS_ENABLED(CONFIG_ZONE_DMA32)) {
>> +		if (arm_dma_zone_size)
>> +			arm64_dma32_phys_limit = arm_dma_zone_size +
>> +						memblock_start_of_DRAM();
>> +		else
>> +			arm64_dma32_phys_limit = max_zone_phys(32);
>> +	} else {
>>   		arm64_dma32_phys_limit = PHYS_MASK + 1;
>> +	}
>>   
>>   	reserve_crashkernel();
>> 
>>