From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751385AbdGQI2m (ORCPT ); Mon, 17 Jul 2017 04:28:42 -0400 Received: from mga05.intel.com ([192.55.52.43]:42403 "EHLO mga05.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751257AbdGQI2l (ORCPT ); Mon, 17 Jul 2017 04:28:41 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.40,374,1496127600"; d="scan'208";a="1196226022" Subject: Re: [PATCH v9 2/7] perf/x86/intel: Record branch type To: Jiri Olsa Cc: acme@kernel.org, jolsa@kernel.org, peterz@infradead.org, mingo@redhat.com, alexander.shishkin@linux.intel.com, mpe@ellerman.id.au, Linux-kernel@vger.kernel.org, ak@linux.intel.com, kan.liang@intel.com, yao.jin@intel.com References: <1500289603-4352-1-git-send-email-yao.jin@linux.intel.com> <1500289603-4352-3-git-send-email-yao.jin@linux.intel.com> <20170717081112.GC22046@krava> From: "Jin, Yao" Message-ID: Date: Mon, 17 Jul 2017 16:28:38 +0800 User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101 Thunderbird/52.2.1 MIME-Version: 1.0 In-Reply-To: <20170717081112.GC22046@krava> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 7/17/2017 4:11 PM, Jiri Olsa wrote: > On Mon, Jul 17, 2017 at 07:06:38PM +0800, Jin Yao wrote: > > SNIP > >> +#define X86_BR_TYPE_MAP_MAX 16 >> + >> +static int >> +common_branch_type(int type) >> +{ >> + int i; >> + const int branch_map[X86_BR_TYPE_MAP_MAX] = { >> + PERF_BR_CALL, /* X86_BR_CALL */ >> + PERF_BR_RET, /* X86_BR_RET */ >> + PERF_BR_SYSCALL, /* X86_BR_SYSCALL */ >> + PERF_BR_SYSRET, /* X86_BR_SYSRET */ >> + PERF_BR_UNKNOWN, /* X86_BR_INT */ >> + PERF_BR_UNKNOWN, /* X86_BR_IRET */ >> + PERF_BR_COND, /* X86_BR_JCC */ >> + PERF_BR_UNCOND, /* X86_BR_JMP */ >> + PERF_BR_UNKNOWN, /* X86_BR_IRQ */ >> + PERF_BR_IND_CALL, /* X86_BR_IND_CALL */ >> + PERF_BR_UNKNOWN, /* X86_BR_ABORT */ >> + PERF_BR_UNKNOWN, /* X86_BR_IN_TX */ >> + PERF_BR_UNKNOWN, /* X86_BR_NO_TX */ >> + PERF_BR_CALL, /* X86_BR_ZERO_CALL */ >> + PERF_BR_UNKNOWN, /* X86_BR_CALL_STACK */ >> + PERF_BR_IND, /* X86_BR_IND_JMP */ >> + }; > should the branch_map array be static? having it on stack makes > the compiler to create it every time we call the function > > jirka OK, agree to let branch_map array be static. I will add this in v10. I'm also waiting for Peter's review comments for this patch update. Thanks Jin Yao > >> + >> + type >>= 2; /* skip X86_BR_USER and X86_BR_KERNEL */ >> + >> + if (type) { >> + i = __ffs(type); >> + if (i < X86_BR_TYPE_MAP_MAX) >> + return branch_map[i]; >> + } >> + >> + return PERF_BR_UNKNOWN; >> +} >> + >> /* >> * implement actual branch filter based on user demand. >> * Hardware may not exactly satisfy that request, thus >> @@ -942,7 +987,8 @@ intel_pmu_lbr_filter(struct cpu_hw_events *cpuc) >> bool compress = false; >> >> /* if sampling all branches, then nothing to filter */ >> - if ((br_sel & X86_BR_ALL) == X86_BR_ALL) >> + if (((br_sel & X86_BR_ALL) == X86_BR_ALL) && >> + ((br_sel & X86_BR_TYPE_SAVE) != X86_BR_TYPE_SAVE)) >> return; >> >> for (i = 0; i < cpuc->lbr_stack.nr; i++) { >> @@ -963,6 +1009,9 @@ intel_pmu_lbr_filter(struct cpu_hw_events *cpuc) >> cpuc->lbr_entries[i].from = 0; >> compress = true; >> } >> + >> + if ((br_sel & X86_BR_TYPE_SAVE) == X86_BR_TYPE_SAVE) >> + cpuc->lbr_entries[i].type = common_branch_type(type); >> } >> >> if (!compress) >> -- >> 2.7.4 >>