Re: [PATCH v13 3/9] block: add emulation for copy

[Kent Overstreet]

On Tue, Aug 01, 2023 at 06:37:02PM +0530, Nitesh Shetty wrote:
> On 23/07/20 09:50AM, Christoph Hellwig wrote:
> > > +static void *blkdev_copy_alloc_buf(sector_t req_size, sector_t *alloc_size,
> > > +		gfp_t gfp_mask)
> > > +{
> > > +	int min_size = PAGE_SIZE;
> > > +	void *buf;
> > > +
> > > +	while (req_size >= min_size) {
> > > +		buf = kvmalloc(req_size, gfp_mask);
> > > +		if (buf) {
> > > +			*alloc_size = req_size;
> > > +			return buf;
> > > +		}
> > > +		/* retry half the requested size */
> > > +		req_size >>= 1;
> > > +	}
> > > +
> > > +	return NULL;
> > 
> > Is there any good reason for using vmalloc instead of a bunch
> > of distcontiguous pages?
> > 
> 
> kvmalloc seemed convenient for the purpose. We will need to call alloc_page
> in a loop to guarantee discontigous pages. Do you prefer that over kvmalloc?

No, kvmalloc should be the preferred approach here now: with large
folios, we're now getting better about doing more large memory
allocations and avoiding fragmentation, so in practice this won't be a
vmalloc allocation except in exceptional circumstances, and performance
will be better and the code will be simpler doing a single large
allocation.