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:   Thu, 24 Jan 2019 07:36:19 +0800
From:   Leo Yan <leo.yan@...aro.org>
To:     Mathieu Poirier <mathieu.poirier@...aro.org>
Cc:     Arnaldo Carvalho de Melo <acme@...nel.org>,
        Suzuki K Poulose <suzuki.poulose@....com>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Jiri Olsa <jolsa@...hat.com>,
        Namhyung Kim <namhyung@...nel.org>,
        Mike Leach <mike.leach@...aro.org>,
        Robert Walker <robert.walker@....com>,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        Coresight ML <coresight@...ts.linaro.org>
Subject: Re: [PATCH v6 8/8] perf cs-etm: Set sample flags for exception
 return packet

Hi Mathieu,

On Wed, Jan 23, 2019 at 02:51:14PM -0700, Mathieu Poirier wrote:
> On Sat, Jan 19, 2019 at 09:43:47AM +0800, Leo Yan wrote:
> > When return from exception, we need to distinguish if it's system call
> > return or for other type exceptions for setting sample flags.  Due to
> > the exception return packet doesn't contain exception number, so we
> > cannot decide sample flags based on exception number.
> > 
> > On the other hand, the exception return packet is followed by an
> > instruction range packet; this range packet deliveries the start address
> > after exception handling, we can check if it is a SVC instruction just
> > before the start address.  If there has one SVC instruction is found
> > ahead the return address, this means it's an exception return for system
> > call; otherwise it is an normal return for other exceptions.
> > 
> > This patch is to set sample flags for exception return packet, firstly
> > it simply set sample flags as PERF_IP_FLAG_INTERRUPT for all exception
> > returns since at this point it doesn't know what's exactly the exception
> > type.  We will defer to decide if it's an exception return for system
> > call when the next instruction range packet comes, it checks if there
> > has one SVC instruction prior to the start address and if so we will
> > change sample flags to PERF_IP_FLAG_SYSCALLRET for system call
> > return.
> > 
> > Signed-off-by: Leo Yan <leo.yan@...aro.org>
> > ---
> >  tools/perf/util/cs-etm.c | 44 ++++++++++++++++++++++++++++++++++++++++
> >  1 file changed, 44 insertions(+)
> > 
> > diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c
> > index 052805de6513..7547a7178f46 100644
> > --- a/tools/perf/util/cs-etm.c
> > +++ b/tools/perf/util/cs-etm.c
> > @@ -1372,6 +1372,20 @@ static int cs_etm__set_sample_flags(struct cs_etm_queue *etmq)
> >  		if (prev_packet->sample_type == CS_ETM_DISCONTINUITY)
> >  			prev_packet->flags |= PERF_IP_FLAG_BRANCH |
> >  					      PERF_IP_FLAG_TRACE_BEGIN;
> > +
> > +		/*
> > +		 * If the previous packet is an exception return packet
> > +		 * and the return address just follows SVC instuction,
> > +		 * it needs to calibrate the previous packet sample flags
> > +		 * as PERF_IP_FLAG_SYSCALLRET.
> > +		 */
> > +		if (prev_packet->flags == (PERF_IP_FLAG_BRANCH |
> > +					   PERF_IP_FLAG_RETURN |
> > +					   PERF_IP_FLAG_INTERRUPT) &&
> 
> Would it make more sense to just look for prev-packet->sample_type ==
> CS_ETM_EXCEPTION_RET ?

We cannot check 'prev-packet->sample_type == CS_ETM_EXCEPTION_RET',
since CS_ETM_EXCEPTION_RET is associated to its previous instruction
range packet but not a standalone instruction range packet, we don't
swap for exception and exception return packets, 'prev_packet' is
pointed to the previous instruction range packet at here.

This is why we need to use 'prev_packet->flags' but not
'prev_packet->sample_type' as checking condition.

Thanks,
Leo Yan

> > +		    cs_etm__is_svc_instr(etmq, packet, packet->start_addr))
> > +			prev_packet->flags = PERF_IP_FLAG_BRANCH |
> > +					     PERF_IP_FLAG_RETURN |
> > +					     PERF_IP_FLAG_SYSCALLRET;
> >  		break;
> >  	case CS_ETM_DISCONTINUITY:
> >  		/*
> > @@ -1422,6 +1436,36 @@ static int cs_etm__set_sample_flags(struct cs_etm_queue *etmq)
> >  			prev_packet->flags = packet->flags;
> >  		break;
> >  	case CS_ETM_EXCEPTION_RET:
> > +		/*
> > +		 * When the exception return packet is inserted, since
> > +		 * exception return packet is not used standalone for
> > +		 * generating samples and it's affiliation to the previous
> > +		 * instruction range packet; so set previous range packet
> > +		 * flags to tell perf it is an exception return branch.
> > +		 *
> > +		 * The exception return can be for either system call or
> > +		 * other exception types; unfortunately the packet doesn't
> > +		 * contain exception type related info so we cannot decide
> > +		 * the exception type purely based on exception return packet.
> > +		 * If we record the exception number from exception packet and
> > +		 * reuse it for excpetion return packet, this is not reliable
> > +		 * due the trace can be discontinuity or the interrupt can
> > +		 * be nested, thus the recorded exception number cannot be
> > +		 * used for exception return packet for these two cases.
> > +		 *
> > +		 * For exception return packet, we only need to distinguish the
> > +		 * packet is for system call or for other types.  Thus the
> > +		 * decision can be deferred when receive the next packet which
> > +		 * contains the return address, based on the return address we
> > +		 * can read out the previous instruction and check if it's a
> > +		 * system call instruction and then calibrate the sample flag
> > +		 * as needed.
> > +		 */
> > +		if (prev_packet->sample_type == CS_ETM_RANGE)
> > +			prev_packet->flags = PERF_IP_FLAG_BRANCH |
> > +					     PERF_IP_FLAG_RETURN |
> > +					     PERF_IP_FLAG_INTERRUPT;
> > +		break;
> >  	case CS_ETM_EMPTY:
> >  	default:
> >  		break;
> > -- 
> > 2.17.1
> > 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ