From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pg0-f72.google.com (mail-pg0-f72.google.com [74.125.83.72]) by kanga.kvack.org (Postfix) with ESMTP id 247D26B0069 for ; Wed, 29 Nov 2017 10:21:27 -0500 (EST) Received: by mail-pg0-f72.google.com with SMTP id y2so2319188pgv.8 for ; Wed, 29 Nov 2017 07:21:27 -0800 (PST) Received: from mga05.intel.com (mga05.intel.com. [192.55.52.43]) by mx.google.com with ESMTPS id b184si1380404pgc.787.2017.11.29.07.21.25 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Wed, 29 Nov 2017 07:21:25 -0800 (PST) Subject: Re: [PATCH] x86/mm/kaiser: Flush the correct ASID in __native_flush_tlb_single() References: <20171128095531.F32E1BC7@viggo.jf.intel.com> <20171129143526.GP3326@worktop> From: Dave Hansen Message-ID: <27729551-ecd6-e4e9-d214-4ab03d8008da@linux.intel.com> Date: Wed, 29 Nov 2017 07:21:23 -0800 MIME-Version: 1.0 In-Reply-To: <20171129143526.GP3326@worktop> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Peter Zijlstra Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, tglx@linutronix.de, richard.fellner@student.tugraz.at, moritz.lipp@iaik.tugraz.at, daniel.gruss@iaik.tugraz.at, michael.schwarz@iaik.tugraz.at, luto@kernel.org, torvalds@linux-foundation.org, keescook@google.com, hughd@google.com, bp@alien8.de, x86@kernel.org On 11/29/2017 06:35 AM, Peter Zijlstra wrote: >> @@ -451,6 +474,9 @@ static inline void __native_flush_tlb_si >> if (kern_asid(loaded_mm_asid) != user_asid(loaded_mm_asid)) >> invpcid_flush_one(user_asid(loaded_mm_asid), addr); >> invpcid_flush_one(kern_asid(loaded_mm_asid), addr); >> + >> + /* Check that we are flushing the active ASID: */ >> + VM_WARN_ON_ONCE(kern_asid(loaded_mm_asid) != cr3_asid()); >> } > > Can't we do this differently (after my recent patches)? It appears to me > we can unconditionally do INVLPG to shoot down the kernel mapping, and > then, depending on INVPCID support we can either use that to shoot down > a single page or simply invalidate the entire user mapping. Yes, that works. Also, as I think about it, INVLPG is a safer (bug-resistant) instruction to use too. INVPCID _can_ get the current (kernel) ASID wrong, as we saw. But INVLPG always uses the current one and can't be wrong about flushing the *current* ASID. I think Andy measured it to be faster than INVPCID too. So, maybe we should just remove INVPCID's use entirely. > arch/x86/include/asm/tlbflush.h | 23 +++++++---------------- > 1 file changed, 7 insertions(+), 16 deletions(-) > > diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h > index 481d5094559e..9587722162ee 100644 > --- a/arch/x86/include/asm/tlbflush.h > +++ b/arch/x86/include/asm/tlbflush.h > @@ -438,29 +438,20 @@ static inline void __native_flush_tlb_single(unsigned long addr) > { > u32 loaded_mm_asid = this_cpu_read(cpu_tlbstate.loaded_mm_asid); > > + asm volatile("invlpg (%0)" ::"r" (addr) : "memory"); > + > + if (!kaiser_enabled) > + return; > + > /* > * Some platforms #GP if we call invpcid(type=1/2) before > * CR4.PCIDE=1. Just call invpcid in the case we are called > * early. > */ > - if (!this_cpu_has(X86_FEATURE_INVPCID_SINGLE)) { > + if (!this_cpu_has(X86_FEATURE_INVPCID_SINGLE)) > flush_user_asid(loaded_mm_asid); > - asm volatile("invlpg (%0)" ::"r" (addr) : "memory"); > - return; > - } > - /* Flush the address out of both PCIDs. */ > - /* > - * An optimization here might be to determine addresses > - * that are only kernel-mapped and only flush the kernel > - * ASID. But, userspace flushes are probably much more > - * important performance-wise. > - * > - * Make sure to do only a single invpcid when KAISER is > - * disabled and we have only a single ASID. > - */ > - if (kern_asid(loaded_mm_asid) != user_asid(loaded_mm_asid)) > + else > invpcid_flush_one(user_asid(loaded_mm_asid), addr); > - invpcid_flush_one(kern_asid(loaded_mm_asid), addr); > } > > static inline void __flush_tlb_all(void) > -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org