From: Tan Xiaojun <tanxiaojun@huawei.com>
To: unlisted-recipients:; (no To-header on input)
Cc: <linux-kernel@vger.kernel.org>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>
Subject: [bug report & help] arm64: ltp testcase "migrate_pages01" failed
Date: Mon, 16 Oct 2017 19:42:35 +0800 [thread overview]
Message-ID: <59E49B2B.8020802@huawei.com> (raw)
Hi all,
I test ltp in Hisilicon D05 board and get a failed result about the testcase "migrate_pages01".
In fact, The sub testcase "test_invalid_nodes" failed. The testcase is to find a invalid numa node and migrate memory pages to this node via syscall of "migrate_pages".
The expected result of this case is returning "-1", but it actually return "0".
--------------------------------------------------------
# ./migrate_pages01
migrate_pages01 0 TINFO : test_empty_mask
migrate_pages01 1 TPASS : expected ret success: returned value = 0
migrate_pages01 0 TINFO : test_invalid_pid -1
migrate_pages01 2 TPASS : expected ret success: returned value = -1
migrate_pages01 3 TPASS : expected failure: TEST_ERRNO=ESRCH(3): No such process
migrate_pages01 0 TINFO : test_invalid_pid unused pid
migrate_pages01 4 TPASS : expected ret success: returned value = -1
migrate_pages01 5 TPASS : expected failure: TEST_ERRNO=ESRCH(3): No such process
migrate_pages01 0 TINFO : test_invalid_masksize
migrate_pages01 6 TPASS : expected ret success: returned value = -1
migrate_pages01 7 TPASS : expected failure: TEST_ERRNO=EINVAL(22): Invalid argument
migrate_pages01 0 TINFO : test_invalid_mem -1
migrate_pages01 8 TPASS : expected ret success: returned value = -1
migrate_pages01 9 TPASS : expected failure: TEST_ERRNO=EFAULT(14): Bad address
migrate_pages01 0 TINFO : test_invalid_mem invalid prot
migrate_pages01 10 TPASS : expected ret success: returned value = -1
migrate_pages01 11 TPASS : expected failure: TEST_ERRNO=EFAULT(14): Bad address
migrate_pages01 0 TINFO : test_invalid_mem unmmaped
migrate_pages01 12 TPASS : expected ret success: returned value = -1
migrate_pages01 13 TPASS : expected failure: TEST_ERRNO=EFAULT(14): Bad address
migrate_pages01 0 TINFO : test_invalid_nodes
migrate_pages01 14 TFAIL : migrate_pages_common.c:45: unexpected failure - returned value = 0, expected: -1
migrate_pages01 15 TFAIL : migrate_pages_common.c:55: call succeeded unexpectedly
migrate_pages01 0 TINFO : test_invalid_perm
migrate_pages01 16 TPASS : expected ret success: returned value = -1
migrate_pages01 17 TPASS : expected failure: TEST_ERRNO=EPERM(1): Operation not permitted
--------------------------------------------------------
I debug and find a interesting thing, this case does not always fail.
1) If one or several numa nodes have no memory, this case will run successfully like below:
--------------------
available: 4 nodes (0-3)
node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
node 0 size: 65309 MB
node 0 free: 61650 MB
node 1 cpus: 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
node 1 size: 65404 MB
node 1 free: 61377 MB
node 2 cpus: 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
node 2 size: 65401 MB
node 2 free: 62316 MB
node 3 cpus: 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
node 3 size: 0 MB
node 3 free: 0 MB
node distances:
node 0 1 2 3
0: 10 15 20 20
1: 15 10 20 20
2: 20 20 10 15
3: 20 20 15 10
---------------------
This testcase will find node number 3 and migrate pages to node 3. And syscall of "migrate_pages" return -1, test succeeded.
2) In most cases, all nodes have memory, and the testcase will get non-existent node like node number 4. The syscall of "migrate_pages" should also return -1, but return 0 actually.
So the testcase failed.
I think it is a problem in arm64. But I am not familiar with numa, so I ask for help from you.
Thanks.
Xiaojun.
next reply other threads:[~2017-10-16 11:43 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-16 11:42 Tan Xiaojun [this message]
2017-10-17 2:58 ` [bug report & help] arm64: ltp testcase "migrate_pages01" failed Tan Xiaojun
2017-10-17 9:23 ` Will Deacon
2017-10-17 11:33 ` Tan Xiaojun
2017-10-17 13:19 ` Yisheng Xie
2017-10-18 1:45 ` Yisheng Xie
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=59E49B2B.8020802@huawei.com \
--to=tanxiaojun@huawei.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.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