From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:54263) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dUgvf-000484-U9 for qemu-devel@nongnu.org; Mon, 10 Jul 2017 18:13:20 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dUgvb-0006Ax-Tc for qemu-devel@nongnu.org; Mon, 10 Jul 2017 18:13:19 -0400 Received: from out3-smtp.messagingengine.com ([66.111.4.27]:60237) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dUgvb-0006Ai-L1 for qemu-devel@nongnu.org; Mon, 10 Jul 2017 18:13:15 -0400 Date: Mon, 10 Jul 2017 18:13:14 -0400 From: "Emilio G. Cota" Message-ID: <20170710221314.GA1051@flamenco> References: <1499586614-20507-1-git-send-email-cota@braap.org> <1499586614-20507-22-git-send-email-cota@braap.org> <20170710211446.GA25777@flamenco> <790239404.15218867.1499722387288.JavaMail.zimbra@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <790239404.15218867.1499722387288.JavaMail.zimbra@redhat.com> Subject: Re: [Qemu-devel] [PATCH 21/22] tcg: enable per-thread TCG for softmmu List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini Cc: qemu-devel@nongnu.org, Richard Henderson On Mon, Jul 10, 2017 at 17:33:07 -0400, Paolo Bonzini wrote: > > > I agree that it would be nice to have the same mechanism for all. > > > > The main hurdle I see is how to allow for concurrent code generation while > > minimizing flushes of the single, fixed-size[*] code_gen_buffer. > > In user-mode this is tricky because there is no way to bound the number > > of threads that might be spawned by the guest code (I don't think reading > > /proc/sys/kernel/threads-max is a viable solution here). > > > > Switching to a "__thread *tcg_ctx_ptr" model will help minimize > > user-mode/softmmu differences though. The only remaining difference would be > > that user-mode would need tb_lock() around tb_gen_code, whereas softmmu > > wouldn't, but everything else would be the same. > > Hmm, tb_gen_code is already protected by mmap_lock in linux-user, so you wouldn't > get any parallelism. On the other hand, you could just say that the fixed-size > code_gen_buffer is protected by mmap_lock, which doesn't exist for softmmu. Yes. tb_lock/mmap_lock, or like they're called in some asserts, memory_lock. A way to get some parallelism in user-mode given the constraints would be to share regions among TCG threads. Threads would still need to take a per-region lock, but it wouldn't be a global lock so that would scale better. I'm not sure we really need that much parallelism for code generation in user-mode, though. So I wouldn't focus on this until seeing benchmarks that have a clear bottleneck due to "memory_lock". E.