qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Darren Kenny <darren.kenny@oracle.com>
To: Alexander Bulekov <alxndr@bu.edu>, qemu-devel@nongnu.org
Cc: Thomas Huth <thuth@redhat.com>,
	f4bug@amsat.org, Alexander Bulekov <alxndr@bu.edu>,
	bsd@redhat.com, stefanha@redhat.com,
	Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [PATCH v2 15/15] scripts/oss-fuzz: Add crash trace minimization script
Date: Thu, 03 Sep 2020 10:28:29 +0100	[thread overview]
Message-ID: <m2y2lrqkea.fsf@oracle.com> (raw)
In-Reply-To: <20200819061110.1320568-16-alxndr@bu.edu>

On Wednesday, 2020-08-19 at 02:11:10 -04, Alexander Bulekov wrote:
> Once we find a crash, we can convert it into a QTest trace. Usually this
> trace will contain many operations that are unneeded to reproduce the
> crash. This script tries to minimize the crashing trace, by removing
> operations and trimming QTest bufwrite(write addr len data...) commands.
>
> Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
> ---
>  scripts/oss-fuzz/minimize_qtest_trace.py | 118 +++++++++++++++++++++++
>  1 file changed, 118 insertions(+)
>  create mode 100755 scripts/oss-fuzz/minimize_qtest_trace.py
>
> diff --git a/scripts/oss-fuzz/minimize_qtest_trace.py b/scripts/oss-fuzz/minimize_qtest_trace.py
> new file mode 100755
> index 0000000000..2f1f4f368e
> --- /dev/null
> +++ b/scripts/oss-fuzz/minimize_qtest_trace.py
> @@ -0,0 +1,118 @@
> +#!/usr/bin/env python3
> +# -*- coding: utf-8 -*-
> +
> +"""
> +This takes a crashing qtest trace and tries to remove superflous operations
> +"""
> +
> +import sys
> +import os
> +import subprocess
> +import time
> +
> +QEMU_ARGS = None
> +QEMU_PATH = None
> +TIMEOUT = 5
> +CRASH_TOKEN = None
> +
> +
> +def usage():
> +    sys.exit("""\
> +Usage: QEMU_PATH="/path/to/qemu" QEMU_ARGS="args" {} input_trace output_trace
> +By default, will try to use the second-to-last line in the output to identify
> +whether the crash occred. Optionally, manually set a string that idenitifes the
> +crash by setting CRASH_TOKEN=
> +""".format((sys.argv[0])))
> +
> +
> +def check_if_trace_crashes(trace, path):
> +    global CRASH_TOKEN
> +    with open(path, "w") as tracefile:
> +        tracefile.write("".join(trace))
> +    rc = subprocess.Popen("timeout -s 9 {}s {} {} 2>&1 < {}".format(TIMEOUT,
> +                          QEMU_PATH, QEMU_ARGS, path),
> +                          shell=True, stdin=subprocess.PIPE,
> +                          stdout=subprocess.PIPE)

NIT: Similar comment to before, it is nicer to name the placeholders if
     there are more than 1 and you can.

> +    stdo = rc.communicate()[0]
> +    output = stdo.decode('unicode_escape')
> +    if rc.returncode == 137:    # Timed Out
> +        return False
> +    if len(output.splitlines()) < 2:
> +        return False
> +
> +    if CRASH_TOKEN is None:
> +        CRASH_TOKEN = output.splitlines()[-2]
> +
> +    return CRASH_TOKEN in output
> +
> +
> +def minimize_trace(inpath, outpath):
> +    global TIMEOUT
> +    with open(inpath) as f:
> +        trace = f.readlines()
> +    start = time.time()
> +    if not check_if_trace_crashes(trace, outpath):
> +        sys.exit("The input qtest trace didn't cause a crash...")
> +    end = time.time()
> +    print("Crashed in {} seconds".format(end-start))
> +    TIMEOUT = (end-start)*5
> +    print("Setting the timeout for {} seconds".format(TIMEOUT))
> +    print("Identifying Crashes by this string: {}".format(CRASH_TOKEN))
> +
> +    i = 0
> +    newtrace = trace[:]
> +    while i < len(newtrace):
> +        prior = newtrace[i]
> +        print("Trying to remove {}".format(newtrace[i]))
> +        # Try to remove the line completely
> +        newtrace[i] = ""
> +        if check_if_trace_crashes(newtrace, outpath):
> +            i += 1
> +            continue
> +        newtrace[i] = prior
> +        # Try to split up writes into multiple commands, each of which can be
> +        # removed.
> +        if newtrace[i].startswith("write "):

NIT: Would be good to document the assumptions here, just in case things
     change in future.

> +            addr = int(newtrace[i].split()[1], 16)
> +            length = int(newtrace[i].split()[2], 16)
> +            data = newtrace[i].split()[3][2:]
> +            if length > 1:
> +                leftlength = int(length/2)
> +                rightlength = length - leftlength
> +                newtrace.insert(i+1, "")
> +                while leftlength > 0:
> +                    newtrace[i] = "write {} {} 0x{}\n".format(
> +                            hex(addr),
> +                            hex(leftlength),
> +                            data[:leftlength*2])
> +                    newtrace[i+1] = "write {} {} 0x{}\n".format(
> +                            hex(addr+leftlength),
> +                            hex(rightlength),
> +                            data[leftlength*2:])

NIT: Similar comment w.r.t. naming the placeholders.

> +                    if check_if_trace_crashes(newtrace, outpath):
> +                        break
> +                    else:
> +                        leftlength -= 1
> +                        rightlength += 1
> +                if check_if_trace_crashes(newtrace, outpath):
> +                    i -= 1
> +                else:
> +                    newtrace[i] = prior
> +                    del newtrace[i+1]
> +        i += 1
> +    check_if_trace_crashes(newtrace, outpath)
> +
> +
> +if __name__ == '__main__':
> +    if len(sys.argv) < 3:
> +        usage()
> +
> +    QEMU_PATH = os.getenv("QEMU_PATH")
> +    QEMU_ARGS = os.getenv("QEMU_ARGS")
> +    if QEMU_PATH is None or QEMU_ARGS is None:
> +        usage()
> +    if "accel" not in QEMU_ARGS:
> +        QEMU_ARGS += " -accel qtest"
> +    CRASH_TOKEN = os.getenv("CRASH_TOKEN")
> +    QEMU_ARGS += " -qtest stdio -monitor none -serial none "
> +    minimize_trace(sys.argv[1], sys.argv[2])

Since all only nits:

Reviewed-by: Darren Kenny <darren.kenny@oracle.com>

Thanks,

Darren.


  reply	other threads:[~2020-09-03  9:29 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-19  6:10 [PATCH v2 00/15] Add a General Virtual Device Fuzzer Alexander Bulekov
2020-08-19  6:10 ` [PATCH v2 01/15] fuzz: Change the way we write qtest log to stderr Alexander Bulekov
2020-08-19  6:10 ` [PATCH v2 02/15] fuzz: Add general virtual-device fuzzer Alexander Bulekov
2020-09-02 10:03   ` Darren Kenny
2020-09-07 15:39     ` Alexander Bulekov
2020-09-07 15:55       ` Darren Kenny
2020-08-19  6:10 ` [PATCH v2 03/15] fuzz: Add PCI features to the general fuzzer Alexander Bulekov
2020-09-02 11:01   ` Darren Kenny
2020-08-19  6:10 ` [PATCH v2 04/15] fuzz: Add DMA support to the generic-fuzzer Alexander Bulekov
2020-09-03  8:43   ` Darren Kenny
2020-09-07 15:45     ` Alexander Bulekov
2020-08-19  6:11 ` [PATCH v2 05/15] fuzz: Declare DMA Read callback function Alexander Bulekov
2020-09-03  8:44   ` Darren Kenny
2020-08-19  6:11 ` [PATCH v2 06/15] fuzz: Add fuzzer callbacks to DMA-read functions Alexander Bulekov
2020-09-03  8:46   ` Darren Kenny
2020-08-19  6:11 ` [PATCH v2 07/15] fuzz: Add support for custom crossover functions Alexander Bulekov
2020-09-03  8:50   ` Darren Kenny
2020-08-19  6:11 ` [PATCH v2 08/15] fuzz: add a DISABLE_PCI op to general-fuzzer Alexander Bulekov
2020-09-03  8:49   ` Darren Kenny
2020-08-19  6:11 ` [PATCH v2 09/15] fuzz: add a crossover function to generic-fuzzer Alexander Bulekov
2020-09-03  9:04   ` Darren Kenny
2020-08-19  6:11 ` [PATCH v2 10/15] scripts/oss-fuzz: Add wrapper program for generic fuzzer Alexander Bulekov
2020-09-03  9:07   ` Darren Kenny
2020-09-03  9:10   ` Darren Kenny
2020-08-19  6:11 ` [PATCH v2 11/15] scripts/oss-fuzz: Add general-fuzzer build script Alexander Bulekov
2020-09-03  9:15   ` Darren Kenny
2020-08-19  6:11 ` [PATCH v2 12/15] scripts/oss-fuzz: Add general-fuzzer configs for oss-fuzz Alexander Bulekov
2020-09-03  9:16   ` Darren Kenny
2020-08-19  6:11 ` [PATCH v2 13/15] scripts/oss-fuzz: build the general-fuzzer configs Alexander Bulekov
2020-09-03  9:17   ` Darren Kenny
2020-09-07 15:49     ` Alexander Bulekov
2020-08-19  6:11 ` [PATCH v2 14/15] scripts/oss-fuzz: Add script to reorder a general-fuzzer trace Alexander Bulekov
2020-09-03  9:20   ` Darren Kenny
2020-08-19  6:11 ` [PATCH v2 15/15] scripts/oss-fuzz: Add crash trace minimization script Alexander Bulekov
2020-09-03  9:28   ` Darren Kenny [this message]
2020-08-19  6:32 ` [PATCH v2 00/15] Add a General Virtual Device Fuzzer no-reply
2020-08-19 16:23   ` Alexander Bulekov

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=m2y2lrqkea.fsf@oracle.com \
    --to=darren.kenny@oracle.com \
    --cc=alxndr@bu.edu \
    --cc=bsd@redhat.com \
    --cc=f4bug@amsat.org \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    --cc=thuth@redhat.com \
    /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;
as well as URLs for NNTP newsgroup(s).