linux-kernel - Re: [PATCH v3 6/9] perf, x86: handle multiple records in PEBS buffer

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 28 Jul 2014 11:20:20 +0800
From:	"Yan, Zheng" <zheng.z.yan@...el.com>
To:	Peter Zijlstra <peterz@...radead.org>
CC:	Andi Kleen <andi@...stfloor.org>, linux-kernel@...r.kernel.org,
	mingo@...nel.org, acme@...radead.org, eranian@...gle.com
Subject: Re: [PATCH v3 6/9] perf, x86: handle multiple records in PEBS buffer

On 07/26/2014 12:40 AM, Andi Kleen wrote:
>> Suppose two pebs events, one has exclude_kernel set. It overflows,
>> before entering the kernel, the other event generates PEBS records from
>> inside the kernel with both events marked in the overflow field.
>>
>> And only once we leave the kernel can the exclude_kernel event tick
>> again and trigger the assist, finalyl clearing the bit.
>>
>> If you were to report the records to both events, one would get a lot of
>> kernel info he was not entitled to.
> 
> Ok that case can be filtered in software. Shouldn't be too difficult.
> Perhaps just using ip
> 
> 	if (event->attr.exclude_kernel && pebs->ip >= __PAGE_OFFSET)
> 	    skip;
> 	if (event->attr.exclude_user && pebs->ip < __PAGE_OFFSET)
> 	    skip;
> 
> This would also help with the existing skid.
> 
> Any other concerns?
> 
> -Andi
> 

how about following patch
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index 33b4c0e..ea76507 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -1016,6 +1016,16 @@ static void setup_pebs_sample_data(struct perf_event *event,
 		data->br_stack = &cpuc->lbr_stack;
 }
 
+static inline bool intel_pmu_pebs_filter(struct perf_event *event,
+					 struct pebs_record_nhm *record)
+{
+	if (event->attr.exclude_user && !kernel_ip(record->ip))
+		return true;
+	if (event->attr.exclude_kernel && kernel_ip(record->ip))
+		return true;
+	return false;
+}
+
 static void __intel_pmu_pebs_event(struct perf_event *event,
 				   struct pt_regs *iregs,
 				   void *at, void *top, int count)
@@ -1052,6 +1062,8 @@ static void __intel_pmu_pebs_event(struct perf_event *event,
 		struct pebs_record_nhm *p = at;
 		if (!(p->status & (1 << event->hw.idx)))
 			continue;
+		if (intel_pmu_pebs_filter(event, p))
+			continue;
 
 		setup_pebs_sample_data(event, iregs, at, &data, &regs);
 		perf_output_sample(&handle, &header, &data, event);
@@ -1139,6 +1151,8 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs)
 			WARN_ON_ONCE(!event);
 			if (!event->attr.precise_ip)
 				continue;
+			if (intel_pmu_pebs_filter(event, p))
+				continue;
 			counts[bit]++;
 		}
 	}
@@ -1149,7 +1163,8 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs)
 		event = cpuc->events[bit];
 		for (at = base; at < top; at += x86_pmu.pebs_record_size) {
 			struct pebs_record_nhm *p = at;
-			if (p->status & (1 << bit))
+			if ((p->status & (1 << bit)) &&
+			    !intel_pmu_pebs_filter(event, p))
 				break;
 		}
 
---

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/