Return-Path: Received: from zen.linaro.local ([81.128.185.34]) by smtp.gmail.com with ESMTPSA id i64sm15472514wmd.33.2017.06.14.05.35.00 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Wed, 14 Jun 2017 05:35:00 -0700 (PDT) Received: from zen (localhost [127.0.0.1]) by zen.linaro.local (Postfix) with ESMTPS id B231C3E009E; Wed, 14 Jun 2017 13:35:35 +0100 (BST) References: <20170609170100.3599-1-alex.bennee@linaro.org> <20170609170100.3599-4-alex.bennee@linaro.org> <87vao4b4z5.fsf@linaro.org> <9776b437-90b4-f2c2-4a0c-c1c6585379bf@twiddle.net> <20170611050730.GA12317@flamenco> <20170613225352.GA26288@flamenco> <5036e9ad-257a-b098-5093-050fca825627@redhat.com> <87r2ymiyht.fsf@linaro.org> <9c7ea148-2ca7-ec56-48ed-eabac0fe7896@redhat.com> <87poe6ix5u.fsf@linaro.org> User-agent: mu4e 0.9.19; emacs 25.2.50.3 From: Alex =?utf-8?Q?Benn=C3=A9e?= To: Paolo Bonzini Cc: Richard Henderson , "Emilio G. Cota" , peter.maydell@linaro.org, edgar.iglesias@xilinx.com, qemu-devel@nongnu.org, Peter Crosthwaite , "open list\:ARM" Subject: Re: [PATCH] target/aarch64: exit to main loop after handling MSR In-reply-to: Date: Wed, 14 Jun 2017 13:35:35 +0100 Message-ID: <87o9tqiw6g.fsf@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit X-TUID: UyuKisNjt/G+ Paolo Bonzini writes: > On 14/06/2017 14:14, Alex Bennée wrote: >>> Then Emilio's patch, if a bit of a heavy hammer, is correct. After >>> aa64_daif_write needs you need an exit_tb so that arm_cpu_exec_interrupt >>> is executed again. >> >> This is a case of cpu->interrupt_request being pending but not having >> set cpu->icount_decr yet to signal the exit. > > Rather than "yet", "anymore". So far it has always been an invariant > that anything that re-enabled an interrupt had to do exit_tb. > >> Wouldn't another approach >> (that didn't involve futzing with each front-end) to be to check >> cpu->interrupt_request and force the exit in lookup_tb_ptr? > > That would cause an unnecessary slowdown in code that runs with > interrupts disabled but does a lot of indirect jumps... ppc's SLOF > firmware probably qualifies. Really? I'd have to measure the change it makes. Is there a benchmark stanza for measuring the PPC slof firmware time? I have 3 patches now which all fix the same thing so we can pick and choose which we should apply. Patches incoming... -- Alex Bennée