On (04/02/18 11:21), Maninder Singh wrote:
[..]
> >> static const char * const backends[] = {
> >> "lzo",
> >> #if IS_ENABLED(CONFIG_CRYPTO_LZ4)
> >> "lz4",
> >> +#if (PAGE_SIZE < (32 * KB))
> >> + "lz4_dyn",
> >> +#endif
> >
> >This is not the list of supported algorithms. It's the list of
> >recommended algorithms. You can configure zram to use any of
> >available and known to Crypto API algorithms. Including lz4_dyn
> >on PAGE_SIZE > 32K systems.
> >
> Yes, we want to integrate new compression(lz4_dyn) for ZRAM
> only if PAGE_SIZE is less than 32KB to get maximum benefit.
> so we added lz4_dyn to available list of ZRAM compression alhorithms.
Which is not what I was talking about.
You shrink a 2 bytes offset down to a 1 byte offset, thus you enforce that
'page should be less than 32KB', which I'm sure will be confusing. And you
rely on lz4_dyn users to do the right thing - namely, to use that 'nice'
`#if (PAGE_SIZE < (32 * KB))'. Apart from that, lz4_dyn supports only data
in up to page_size chunks. Suppose my system has page_size of less than 32K,
so I legitimately can enable lz4_dyn, but suppose that I will use it
somewhere where I don't work with page_size-d chunks. Will I able to just
do tfm->compress(src, sz) on random buffers? The whole thing looks to be
quite fragile.
-ss
