linux-kernel - Re: [PATCHv3 1/2] zsmalloc: introduce zs_huge

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

Message-ID: <20180226074652.GB168047@rodete-desktop-imager.corp.google.com>
Date:   Mon, 26 Feb 2018 16:46:52 +0900
From:   Minchan Kim <minchan@...nel.org>
To:     Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Mike Rapoport <rppt@...ux.vnet.ibm.com>,
        linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        Sergey Senozhatsky <sergey.senozhatsky@...il.com>
Subject: Re: [PATCHv3 1/2] zsmalloc: introduce zs_huge_object() function

On Mon, Feb 26, 2018 at 03:50:35PM +0900, Sergey Senozhatsky wrote:
> On (02/26/18 14:58), Minchan Kim wrote:
> [..]
> > > Right. The changes are pretty trivial, that's why I kept then in
> > > 2 simple patches. Besides, I didn't want to mix zsmalloc and zram
> > > changes.
> > 
> > As I said earlier, it's not thing we usually do, at least, MM.
> > Anyway, I don't want to insist on it because it depends each
> > person's point of view what's the better for review, git-bisect.
> 
> Thanks :)
> 
> > > > 	size_t huge_size = _zs_huge_object(pool);
> > > > 	..
> > > > 	..
> > > > 	if (comp_size >= huge_size)
> > > > 		memcpy(dst, src, 4K);
> > > 
> > > Yes, can do. My plan was to keep it completely internally to zsmalloc.
> > > Who knows, it might become smart enough one day to do something more
> > > than just size comparison. Any reason you used that leading underscore
> > 
> > Let's do that in future if someone want it. :)
> 
> OK.
> 
> > > in _zs_huge_object()?
> > 
> > 
> > Nope. It's just typo. Let's think better name.
> > How about using zs_huge_size()?
> 
> hm, I think `huge_size' on it's own is a bit general and cryptic.
> zs_huge_object_size() or zs_huge_class_size()?

I wanted to use more general word to hide zsmalloc internal but
I realized it's really impossible to hide them all.
If so, let's use zs_huge_class_size and then let's add big fat
comment what the API represents in there.

Thanks, Sergey!