All of lore.kernel.org
 help / color / mirror / Atom feed
From: Petr Vorel <pvorel@suse.cz>
To: ltp@lists.linux.it
Subject: [LTP] [PATCH v3] sysctl/sysctl02: Add new regression test for overflow file-max
Date: Thu, 20 Jun 2019 14:01:22 +0200	[thread overview]
Message-ID: <20190620120122.GB31382@dell5510> (raw)
In-Reply-To: <1560156706-13617-1-git-send-email-xuyang2018.jy@cn.fujitsu.com>

Hi,

LGTM, below are just small formatting issues.
I'm going to merge it with these changes below.

> On upstream kernel, before commit[1], the max value in proc_get_long based on
> the number of chars(21). It rejects values such as 184467440737095516160 (21 chars)
> but accepts values such as 18446744073709551616 (20 chars). But we should reject all
> because both are overflows. After this commit,the permitted max value is 2^64-1.

> Before commit[2], when writing echo 18446744073709551616 > /proc/sys/fs/file-max
> /proc/sys/fs/file-max will overflow and be set to 0.  It may crash the system. This
> commit sets the max and min value for file-max. After this commit,the permitted max
> value is 2^63-1.

> Unfortunately, commit[2] introduced the minimum value points at the global 'zero'
> variable which is an int. This results in a KASAN splat when accessed as a long by
> proc_doulongvec_minmax on 64-bit architectures. This bug has been fixed by commit[3].

> I will set 2^64 ,2^64-1,2^63 and 0 to file-max in case and test it.

> [1]https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=7f2923c4f73f21cfd714d12a2d48de8c21f11cfe
> [2]https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=32a5ad9c22852e6bd9e74bdec5934ef9d1480bc5
> [3]https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=9002b21465fa4d829edfc94a5a441005cffaa972

> Signed-off-by: Yang Xu <xuyang2018.jy@cn.fujitsu.com>
> ---
>  runtest/commands                      |  1 +
>  testcases/commands/sysctl/sysctl02.sh | 96 +++++++++++++++++++++++++++
>  2 files changed, 97 insertions(+)
>  create mode 100755 testcases/commands/sysctl/sysctl02.sh

> diff --git a/runtest/commands b/runtest/commands
> index ac15e8b23..1870c4209 100644
> --- a/runtest/commands
> +++ b/runtest/commands
> @@ -40,3 +40,4 @@ keyctl01_sh keyctl01.sh
>  gdb01_sh gdb01.sh
>  unshare01_sh unshare01.sh
>  sysctl01_sh sysctl01.sh
> +sysctl02_sh sysctl02.sh
> diff --git a/testcases/commands/sysctl/sysctl02.sh b/testcases/commands/sysctl/sysctl02.sh
> new file mode 100755
> index 000000000..22dd0a429
> --- /dev/null
> +++ b/testcases/commands/sysctl/sysctl02.sh
> @@ -0,0 +1,96 @@
> +#!/bin/sh
> +
> +# SPDX-License-Identifier: GPL-2.0-or-later
> +# Copyright (c) 2019 FUJITSU LIMITED. All rights reserved.
> +# Author: Yang Xu<xuyang2018.jy@cn.fujitsu.com>
> +#
> +# Description:
> +# This is a regression test for handling overflow for file-max.
Instead these 2 lines I'd put just this (nit):
# Regression test for handling overflow for file-max.

> +#
> +# when writing 2^64 to /proc/sys/fs/file-max. It will overflow
> +# and be set to 0. It crash system quickly.
> +#
> +# The kernel bug has been fixed in kernel:
> +# '7f2923c4f' (sysctl: handle overflow in proc_get_long)
> +# the permitted max value is  2^64-1.
> +# '32a5ad9c2' (sysctl: handle overflow for file-max)
> +# the permitted max value is 2^63-1
> +#
> +# After merged this patchset, if we exceed the max value, it will
> +# keep old value. Unfortunately, it introudced a new bug when set it
> +# to 0 and it will lead to system crash.
> +# This bug has been fixed by commit 9002b2146
> +# (kernel/sysctl.c: fix out-of-bounds access when setting file-max)
> +
> +TST_TESTFUNC=do_test
> +TST_SETUP=setup
> +TST_CLEANUP=cleanup
> +TST_CNT=4
> +TST_NEEDS_ROOT=1
> +TST_NEEDS_CMDS="sysctl"
> +dir="/proc/sys/fs/"
> +syms_file="/proc/kallsyms"
> +name="file-max"
> +orig_value=200000
> +
> +. tst_test.sh
> +
> +setup()
> +{
> +	[ ! -f "$dir""$name" ] && tst_brk TCONF \
> +		"$name was not supported"
[ ! -f "$dir$name" ] && tst_brk TCONF "$name was not supported"

> +	orig_value=$(cat "$dir""$name")
> +}
> +
> +do_test()
> +{
> +	case $1 in
> +	1)sysctl_test_overflow 18446744073709551616;;
> +	2)sysctl_test_overflow 18446744073709551615;;
> +	3)sysctl_test_overflow 9223372036854775808;;
> +	4)sysctl_test_zero;;
> +	esac
> +}
> +
> +sysctl_test_overflow()
> +{
> +	local old_value=$(cat "$dir""$name")
> +
> +	sysctl -w "fs.file-max"=$1 >/dev/null 2>&1
> +
> +	local test_value=$(cat "$dir""$name")
again, why 2 stings instead of one?
local test_value=$(cat "$dir$name")

> +
> +	echo ${test_value} |grep -q ${old_value}
> +	if [ $? -eq 0 ]; then
> +		tst_res TPASS "file-max overflow, reject it and keep old value."
Please don't use dot

> +	else
> +		tst_res TFAIL "file-max overflow and set it to ${test_value}."
Here as well.

> +	fi
> +	cleanup
> +}
> +
> +sysctl_test_zero()
> +{
> +	sysctl -w "fs.file-max"=0 >/dev/null 2>&1
Please use -q option instead of redirecting.
> +	[ ! -f "$syms_file" ] && tst_brk TCONF \
> +		"$syms_file was not supported"

> +	cat $syms_file  |grep kasan_report >/dev/null 2>&1
> +	if [ $? -eq 0 ]; then
Also grep has -q, no need to use pipe:

if grep -q kasan_report $syms_file; then

> +		dmesg | grep "KASAN: global-out-of-bounds in __do_proc_doulongvec_minmax" >/dev/null
-q here as well.

> +		if [ $? -eq 0 ]; then
> +			tst_res TFAIL "file-max is set 0 and trigger a KASAN error"
> +		else
> +			tst_res TPASS \
> +				"file-max is set 0 and doesn't trigger a KASAN error"
Probably can be on single line.

> +		fi
> +	else
> +		tst_res TCONF "kernel doesn't support KASAN"
> +	fi
> +}
> +
> +cleanup()
> +{
> +	sysctl -w "fs.""$name"=${orig_value} >/dev/null 2>&1
It's safe not quote string ($name is safe, but you can of course)
sysctl -q -w fs.$name=$orig_value

> +}
> +
> +tst_run

  reply	other threads:[~2019-06-20 12:01 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-06 10:31 [LTP] [PATCH] sysctl/sysctl02: Add new regression test for overflow file-max Yang Xu
2019-06-06 10:41 ` Yang Xu
2019-06-06 10:45 ` [LTP] [PATCH v2] " Yang Xu
2019-06-06 11:41   ` Cyril Hrubis
2019-06-10  8:51     ` [LTP] [PATCH v3] " Yang Xu
2019-06-20 12:01       ` Petr Vorel [this message]
2019-06-21  1:26         ` Yang Xu
2019-07-03 14:42           ` Petr Vorel
2019-07-04  1:40             ` Yang Xu
2019-06-20 12:52       ` Petr Vorel
2019-06-10  9:06     ` [LTP] [PATCH v2] " Yang Xu

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=20190620120122.GB31382@dell5510 \
    --to=pvorel@suse.cz \
    --cc=ltp@lists.linux.it \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.