Re: [PATCH 3/7] fsdax: Copy data before write

From: Ruan Shiyang <ruansy.fnst@cn.fujitsu.com>
To: Christoph Hellwig <hch@lst.de>
Cc: <linux-kernel@vger.kernel.org>, <linux-xfs@vger.kernel.org>,
	<linux-nvdimm@lists.01.org>, <linux-fsdevel@vger.kernel.org>,
	<darrick.wong@oracle.com>, <dan.j.williams@intel.com>,
	<willy@infradead.org>, <jack@suse.cz>, <viro@zeniv.linux.org.uk>,
	<linux-btrfs@vger.kernel.org>, <ocfs2-devel@oss.oracle.com>,
	<david@fromorbit.com>, <rgoldwyn@suse.de>
Subject: Re: [PATCH 3/7] fsdax: Copy data before write
Date: Tue, 9 Feb 2021 09:53:10 +0800	[thread overview]
Message-ID: <3f2826a8-df98-e7b0-6ab8-0f410488bc55@cn.fujitsu.com> (raw)
In-Reply-To: <20210208151419.GC12872@lst.de>

On 2021/2/8 下午11:14, Christoph Hellwig wrote:
>>   	switch (iomap.type) {
>>   	case IOMAP_MAPPED:
>> +cow:
>>   		if (iomap.flags & IOMAP_F_NEW) {
>>   			count_vm_event(PGMAJFAULT);
>>   			count_memcg_event_mm(vma->vm_mm, PGMAJFAULT);
>>   			major = VM_FAULT_MAJOR;
>>   		}
>>   		error = dax_iomap_direct_access(&iomap, pos, PAGE_SIZE,
>> -						NULL, &pfn);
>> +						&kaddr, &pfn);
> 
> Any chance you could look into factoring out this code into a helper
> to avoid the goto magic, which is a little too convoluted?
> 
>>   	switch (iomap.type) {
>>   	case IOMAP_MAPPED:
>> +cow:
>>   		error = dax_iomap_direct_access(&iomap, pos, PMD_SIZE,
>> -						NULL, &pfn);
>> +						&kaddr, &pfn);
>>   		if (error < 0)
>>   			goto finish_iomap;
>>   
>>   		entry = dax_insert_entry(&xas, mapping, vmf, entry, pfn,
>>   						DAX_PMD, write && !sync);
>>   
>> +		if (srcmap.type != IOMAP_HOLE) {
> 
> Same here.

Thanks for suggestion.  I'll try it.

--
Thanks,
Ruan Shiyang.
> 
> 