On Tue, Apr 05, 2022 at 12:27:50PM +0200, Riccardo Schirone wrote: > Adding Kirill in CC as well. > > On Tue, Apr 05, 2022 at 11:18:57AM +0200, Borislav Petkov wrote: > > On Tue, Apr 05, 2022 at 09:55:09AM +0200, Riccardo Schirone wrote: > > > The reason why I thought this change made sense is just because I think the > > > two operations should be mostly the same and if one can do one operation, he > > > should be able to access also the other as they are defined in the same header > > > file. > > > > Does this subthread answer your question: > > > > https://lore.kernel.org/r/87bkzqw1vr.ffs@tglx > > > > ? > > > > Not really, as the only answer I saw is: > > > > I guess we can get away not exporting pgprot_encrypted(), but this > > > asymmetry bothers me :) > > > > Well, no. We export only stuff which is needed. Exporting just because > > is a NONO. > > I understand that only the minimum set of really used functions should be > exported, but cc_mkenc() and cc_mkdec() do mostly the same thing so it feels > strange. > > Kirill, is there any reason to not exporting cc_mkenc() apart from the fact that > it is not currently used anywhere else in the in-kernel code? Like, external > modules are not supposed to use cc_mkenc()/pgprot_encrypted()? I see > pgprot_encrypted() used only in fs/proc/vmcore.c outside of arch/x86. The answer you quote is the resason I removed the EXPORT_SYMBOL(). -- Kirill A. Shutemov
