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: [Linaro-mm-sig] [PATCH v3] staging: android: ion: Zero CMA
 allocated memory
To:     Chen Feng <puck.chen@hisilicon.com>,
        Liam Mark <lmark@codeaurora.org>,
        Sumit Semwal <sumit.semwal@linaro.org>
Cc:     devel@driverdev.osuosl.org, Greg KH <gregkh@linuxfoundation.org>,
        linaro-mm-sig@lists.linaro.org, linux-kernel@vger.kernel.org,
        Dan Carpenter <dan.carpenter@oracle.com>,
        "Xiaqing (A)" <saberlily.xia@hisilicon.com>,
        Zhuangluan Su <suzhuangluan@hisilicon.com>
References: <alpine.DEB.2.02.1801260942330.13788@lmark-linux.qualcomm.com>
 <5A6BDE11.9070403@hisilicon.com>
From:   Laura Abbott <labbott@redhat.com>
Message-ID: <b0380b1b-2583-89fb-72c9-b9a606ee6a8a@redhat.com>
Date:   Fri, 26 Jan 2018 22:52:28 -0800
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.5.2
MIME-Version: 1.0
In-Reply-To: <5A6BDE11.9070403@hisilicon.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 01/26/2018 06:04 PM, Chen Feng wrote:
> 
> 
> On 2018/1/27 1:48, Liam Mark wrote:
>> Since commit 204f672255c2 ("staging: android: ion: Use CMA APIs directly")
>> the CMA API is now used directly and therefore the allocated memory is no
>> longer automatically zeroed.
>>
>> Explicitly zero CMA allocated memory to ensure that no data is exposed to
>> userspace.
>>
>> Fixes: 204f672255c2 ("staging: android: ion: Use CMA APIs directly")
>> Signed-off-by: Liam Mark <lmark@codeaurora.org>
>> ---
>> Changes in v2:
>>    - Clean up the commit message.
>>    - Add 'Fixes:'
>>
>> Changes in v3:
>>   - Add support for highmem pages
>>
>>   drivers/staging/android/ion/ion_cma_heap.c | 17 +++++++++++++++++
>>   1 file changed, 17 insertions(+)
>>
>> diff --git a/drivers/staging/android/ion/ion_cma_heap.c b/drivers/staging/android/ion/ion_cma_heap.c
>> index 86196ffd2faf..fa3e4b7e0c9f 100644
>> --- a/drivers/staging/android/ion/ion_cma_heap.c
>> +++ b/drivers/staging/android/ion/ion_cma_heap.c
>> @@ -21,6 +21,7 @@
>>   #include <linux/err.h>
>>   #include <linux/cma.h>
>>   #include <linux/scatterlist.h>
>> +#include <linux/highmem.h>
>>   
>>   #include "ion.h"
>>   
>> @@ -51,6 +52,22 @@ static int ion_cma_allocate(struct ion_heap *heap, struct ion_buffer *buffer,
>>   	if (!pages)
>>   		return -ENOMEM;
>>   
>> +	if (PageHighMem(pages)) {
>> +		unsigned long nr_clear_pages = nr_pages;
>> +		struct page *page = pages;
>> +
>> +		while (nr_clear_pages > 0) {
>> +			void *vaddr = kmap_atomic(page);
>> +
>> +			memset(vaddr, 0, PAGE_SIZE);
>> +			kunmap_atomic(vaddr);
> 
> Here. This way may cause performance latency at mapping-memset-umap page one bye one.
> 
> Take a look at ion_heap_pages_zero.
> 
> Not very critical, arm64 always have linear mapping.
> 

This is under a PageHighMem check so arm64 isn't affected. It's also
the same algorithm arm32 dma-mapping.c uses so I'd like to see some
data about the performance improvement before we go changing things
too much.

Thanks,
Laura

> 
>> +			page++;
>> +			nr_clear_pages--;
>> +		}
>> +	} else {
>> +		memset(page_address(pages), 0, size);
>> +	}
>> +
>>   	table = kmalloc(sizeof(*table), GFP_KERNEL);
>>   	if (!table)
>>   		goto err;
>>
>