From mboxrd@z Thu Jan 1 00:00:00 1970 From: Changbin Du Subject: Re: linux-next: Tree for Oct 31 (vboxguest) Date: Fri, 2 Nov 2018 23:59:33 +0000 Message-ID: <20181102235731.smqhskmxv5mu43qq@mail.google.com> References: <20181031145907.1ee2e866@canb.auug.org.au> <661e44c7-396a-6d58-efa7-ed292f2677c6@infradead.org> <20181101142946.u5luuctqav6fec6k@mail.google.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Arnd Bergmann Cc: Masahiro Yamada , Changbin Du , Randy Dunlap , Stephen Rothwell , Linux-Next Mailing List , Linux Kernel Mailing List , Steven Rostedt , Hans de Goede , Greg Kroah-Hartman List-Id: linux-next.vger.kernel.org On Fri, Nov 02, 2018 at 04:54:25PM +0100, Arnd Bergmann wrote: > On 11/2/18, Masahiro Yamada wrote: > > On Thu, Nov 1, 2018 at 11:32 PM Changbin Du wrote: > >> On Thu, Nov 01, 2018 at 12:32:48PM +0900, Masahiro Yamada wrote: > > > > How about clang? > > > > For clang, -Og might be equivalent to -O1 at this moment, but I am not > > sure. > > > > In my understanding, Clang does not inline functions marked with 'static > > inline' > > for -Og (or -O1) optimization level. > > > > Theoretically, 'inline' keyword is a just hint for the compiler, after all. > > I think this means that we cannot build the kernel in that configuration, > at least with CONFIG_OPTIMIZE_INLINING=y. Without that option, > every 'inline' becomes 'always_inline'. > > Arnd I have verified the new configuration with GCC. CONFIG_OPTIMIZE_INLINING is a whole different thing and I dont think it is so aggressive. Because many of kernel functions marked as 'inline' must be inlined. Otherwise the kernel cannot be compiled at all with CONFIG_OPTIMIZE_INLINING=y. If you wanna to try, just remove the 'inline' keyword from some kernel functions to see what happens. -- Thanks, Changbin Du