From: Adrian Hunter <adrian.hunter@intel.com>
To: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Ingo Molnar <mingo@redhat.com>,
Arnaldo Carvalho de Melo <acme@kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Jiri Olsa <jolsa@kernel.org>, Miguel Ojeda <ojeda@kernel.org>,
Liam Howlett <liam.howlett@oracle.com>,
Namhyung Kim <namhyung@kernel.org>,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org
Subject: Re: [PATCH v1] perf intel-pt: pkt-decoder: Fix alignment issues
Date: Thu, 5 Oct 2023 22:04:37 +0300 [thread overview]
Message-ID: <93afcf44-5f8b-49c1-abc5-5304cba5f991@intel.com> (raw)
In-Reply-To: <CAP-5=fX-VOeCkOJY5xeW67x0+A0tGAHM4VYBBz46L-g2eRCR+w@mail.gmail.com>
On 5/10/23 18:48, Ian Rogers wrote:
> On Mon, Oct 2, 2023 at 3:19 PM Ian Rogers <irogers@google.com> wrote:
>>
>> The byte aligned buffer is cast to large types and dereferenced
>> causing misaligned pointer warnings from undefined behavior sanitizer.
>> Fix the alignment issues with memcpy which may require the
>> introduction of temporaries.
>>
>> Signed-off-by: Ian Rogers <irogers@google.com>
>> ---
>
> This is a relatively small change that fixes building with
> -fsanitize=alignment -fsanitize-undefined-trap-on-error. Adrian, as
> this is Intel-PT could you take a look?
Thanks! This has been down my list of things to do for ages,
but using get_unaligned_le16() etc seems nicer. I sent a patch
set for that.
>
> Thanks,
> Ian
>
>> .../intel-pt-decoder/intel-pt-pkt-decoder.c | 21 ++++++++++++-------
>> 1 file changed, 14 insertions(+), 7 deletions(-)
>>
>> diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.c
>> index af9710622a1f..28659874d84e 100644
>> --- a/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.c
>> +++ b/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.c
>> @@ -83,7 +83,7 @@ static int intel_pt_get_long_tnt(const unsigned char *buf, size_t len,
>> if (len < 8)
>> return INTEL_PT_NEED_MORE_BYTES;
>>
>> - payload = le64_to_cpu(*(uint64_t *)buf);
>> + memcpy_le64(&payload, buf, sizeof(payload));
>>
>> for (count = 47; count; count--) {
>> if (payload & BIT63)
>> @@ -220,6 +220,8 @@ static int intel_pt_get_3byte(const unsigned char *buf, size_t len,
>> static int intel_pt_get_ptwrite(const unsigned char *buf, size_t len,
>> struct intel_pt_pkt *packet)
>> {
>> + uint32_t tmp;
>> +
>> packet->count = (buf[1] >> 5) & 0x3;
>> packet->type = buf[1] & BIT(7) ? INTEL_PT_PTWRITE_IP :
>> INTEL_PT_PTWRITE;
>> @@ -228,12 +230,13 @@ static int intel_pt_get_ptwrite(const unsigned char *buf, size_t len,
>> case 0:
>> if (len < 6)
>> return INTEL_PT_NEED_MORE_BYTES;
>> - packet->payload = le32_to_cpu(*(uint32_t *)(buf + 2));
>> + memcpy(&tmp, buf + 2, sizeof(tmp));
>> + packet->payload = le32_to_cpu(tmp);
>> return 6;
>> case 1:
>> if (len < 10)
>> return INTEL_PT_NEED_MORE_BYTES;
>> - packet->payload = le64_to_cpu(*(uint64_t *)(buf + 2));
>> + memcpy_le64(&packet->payload, buf + 2, sizeof(packet->payload));
>> return 10;
>> default:
>> return INTEL_PT_BAD_PACKET;
>> @@ -258,7 +261,7 @@ static int intel_pt_get_mwait(const unsigned char *buf, size_t len,
>> if (len < 10)
>> return INTEL_PT_NEED_MORE_BYTES;
>> packet->type = INTEL_PT_MWAIT;
>> - packet->payload = le64_to_cpu(*(uint64_t *)(buf + 2));
>> + memcpy_le64(&packet->payload, buf + 2, sizeof(packet->payload));
>> return 10;
>> }
>>
>> @@ -454,6 +457,8 @@ static int intel_pt_get_ip(enum intel_pt_pkt_type type, unsigned int byte,
>> struct intel_pt_pkt *packet)
>> {
>> int ip_len;
>> + uint16_t tmp16;
>> + uint32_t tmp32;
>>
>> packet->count = byte >> 5;
>>
>> @@ -465,13 +470,15 @@ static int intel_pt_get_ip(enum intel_pt_pkt_type type, unsigned int byte,
>> if (len < 3)
>> return INTEL_PT_NEED_MORE_BYTES;
>> ip_len = 2;
>> - packet->payload = le16_to_cpu(*(uint16_t *)(buf + 1));
>> + memcpy(&tmp16, buf + 1, sizeof(tmp16));
>> + packet->payload = le16_to_cpu(tmp16);
>> break;
>> case 2:
>> if (len < 5)
>> return INTEL_PT_NEED_MORE_BYTES;
>> ip_len = 4;
>> - packet->payload = le32_to_cpu(*(uint32_t *)(buf + 1));
>> + memcpy(&tmp32, buf + 1, sizeof(tmp32));
>> + packet->payload = le32_to_cpu(tmp32);
>> break;
>> case 3:
>> case 4:
>> @@ -484,7 +491,7 @@ static int intel_pt_get_ip(enum intel_pt_pkt_type type, unsigned int byte,
>> if (len < 9)
>> return INTEL_PT_NEED_MORE_BYTES;
>> ip_len = 8;
>> - packet->payload = le64_to_cpu(*(uint64_t *)(buf + 1));
>> + memcpy_le64(&packet->payload, buf + 1, sizeof(packet->payload));
>> break;
>> default:
>> return INTEL_PT_BAD_PACKET;
>> --
>> 2.42.0.582.g8ccd20d70d-goog
>>
next prev parent reply other threads:[~2023-10-05 19:06 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-02 22:19 Ian Rogers
2023-10-05 15:48 ` Ian Rogers
2023-10-05 19:04 ` Adrian Hunter [this message]
2023-10-05 21:24 ` Ian Rogers
2023-10-09 5:29 ` Namhyung Kim
2023-10-09 15:31 ` Ian Rogers
2023-10-11 5:56 ` Namhyung Kim
2023-10-11 6:50 ` Ian Rogers
2023-10-12 12:27 ` Adrian Hunter
2023-10-13 4:23 ` Namhyung Kim
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=93afcf44-5f8b-49c1-abc5-5304cba5f991@intel.com \
--to=adrian.hunter@intel.com \
--cc=acme@kernel.org \
--cc=alexander.shishkin@linux.intel.com \
--cc=irogers@google.com \
--cc=jolsa@kernel.org \
--cc=liam.howlett@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=ojeda@kernel.org \
--cc=peterz@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®