From mboxrd@z Thu Jan 1 00:00:00 1970 From: Segher Boessenkool Subject: Re: [PATCH v10 2/5] powerpc/vdso: Prepare for switching VDSO to generic C implementation. Date: Wed, 5 Aug 2020 13:40:54 -0500 Message-ID: <20200805184054.GQ6753@gate.crashing.org> References: <348528c33cd4007f3fee7fe643ef160843d09a6c.1596611196.git.christophe.leroy@csgroup.eu> <20200805140307.GO6753@gate.crashing.org> <3db2a590-b842-83db-ed2b-f3ee62595f18@csgroup.eu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <3db2a590-b842-83db-ed2b-f3ee62595f18@csgroup.eu> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: linuxppc-dev-bounces+glppe-linuxppc-embedded-2=m.gmane-mx.org@lists.ozlabs.org Sender: "Linuxppc-dev" To: Christophe Leroy Cc: nathanl@linux.ibm.com, linux-arch@vger.kernel.org, vincenzo.frascino@arm.com, arnd@arndb.de, linux-kernel@vger.kernel.org, Paul Mackerras , luto@kernel.org, tglx@linutronix.de, linuxppc-dev@lists.ozlabs.org List-Id: linux-arch.vger.kernel.org Hi! On Wed, Aug 05, 2020 at 04:40:16PM +0000, Christophe Leroy wrote: > >It cannot optimise it because it does not know shift < 32. The code > >below is incorrect for shift equal to 32, fwiw. > > Is there a way to tell it ? Sure, for example the &31 should work (but it doesn't, with the GCC version you used -- which version is that?) > >What does the compiler do for just > > > >static __always_inline u64 vdso_shift_ns(u64 ns, unsigned long shift) > > return ns >> (shift & 31); > >} > > > > Worse: I cannot make heads or tails of all that branch spaghetti, sorry. > 73c: 55 8c 06 fe clrlwi r12,r12,27 > 740: 7f c8 f0 14 addc r30,r8,r30 > 744: 7c c6 4a 14 add r6,r6,r9 > 748: 7c c6 e1 14 adde r6,r6,r28 > 74c: 34 6c ff e0 addic. r3,r12,-32 > 750: 41 80 00 70 blt 7c0 <__c_kernel_clock_gettime+0x114> This branch is always true. Hrm. Segher From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gate.crashing.org ([63.228.1.57]:57203 "EHLO gate.crashing.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728019AbgHETB4 (ORCPT ); Wed, 5 Aug 2020 15:01:56 -0400 Date: Wed, 5 Aug 2020 13:40:54 -0500 From: Segher Boessenkool Subject: Re: [PATCH v10 2/5] powerpc/vdso: Prepare for switching VDSO to generic C implementation. Message-ID: <20200805184054.GQ6753@gate.crashing.org> References: <348528c33cd4007f3fee7fe643ef160843d09a6c.1596611196.git.christophe.leroy@csgroup.eu> <20200805140307.GO6753@gate.crashing.org> <3db2a590-b842-83db-ed2b-f3ee62595f18@csgroup.eu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3db2a590-b842-83db-ed2b-f3ee62595f18@csgroup.eu> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Christophe Leroy Cc: Benjamin Herrenschmidt , Paul Mackerras , Michael Ellerman , nathanl@linux.ibm.com, anton@ozlabs.org, linux-arch@vger.kernel.org, arnd@arndb.de, linux-kernel@vger.kernel.org, luto@kernel.org, tglx@linutronix.de, vincenzo.frascino@arm.com, linuxppc-dev@lists.ozlabs.org Message-ID: <20200805184054.xOnIXdPE6ot-_QFyA13htJ6b4oUiNYXUAi4YTAjG07w@z> Hi! On Wed, Aug 05, 2020 at 04:40:16PM +0000, Christophe Leroy wrote: > >It cannot optimise it because it does not know shift < 32. The code > >below is incorrect for shift equal to 32, fwiw. > > Is there a way to tell it ? Sure, for example the &31 should work (but it doesn't, with the GCC version you used -- which version is that?) > >What does the compiler do for just > > > >static __always_inline u64 vdso_shift_ns(u64 ns, unsigned long shift) > > return ns >> (shift & 31); > >} > > > > Worse: I cannot make heads or tails of all that branch spaghetti, sorry. > 73c: 55 8c 06 fe clrlwi r12,r12,27 > 740: 7f c8 f0 14 addc r30,r8,r30 > 744: 7c c6 4a 14 add r6,r6,r9 > 748: 7c c6 e1 14 adde r6,r6,r28 > 74c: 34 6c ff e0 addic. r3,r12,-32 > 750: 41 80 00 70 blt 7c0 <__c_kernel_clock_gettime+0x114> This branch is always true. Hrm. Segher