From: Richard Palethorpe <rpalethorpe@suse.de>
To: Andrea Cervesato <andrea.cervesato@suse.com>
Cc: ltp@lists.linux.it
Subject: Re: [LTP] [PATCH v1 2/3] Add process_madvise02 test
Date: Tue, 18 Oct 2022 13:24:18 +0100 [thread overview]
Message-ID: <87a65tjp22.fsf@suse.de> (raw)
In-Reply-To: <20221006110642.12410-4-andrea.cervesato@suse.com>
Hello,
Looks like largely the same problems as with the previous test. Let's
get that one merged then lessons learned from that can be applied to
this.
Andrea Cervesato via ltp <ltp@lists.linux.it> writes:
> Test for checking MADV_PAGEOUT functionality over anonymous memory in
> process_madvise syscall.
>
> Signed-off-by: Andrea Cervesato <andrea.cervesato@suse.com>
> ---
> testcases/kernel/syscalls/cma/.gitignore | 1 +
> .../kernel/syscalls/cma/process_madvise02.c | 128 ++++++++++++++++++
> 2 files changed, 129 insertions(+)
> create mode 100644 testcases/kernel/syscalls/cma/process_madvise02.c
>
> diff --git a/testcases/kernel/syscalls/cma/.gitignore b/testcases/kernel/syscalls/cma/.gitignore
> index 846704294..47ae3e445 100644
> --- a/testcases/kernel/syscalls/cma/.gitignore
> +++ b/testcases/kernel/syscalls/cma/.gitignore
> @@ -3,3 +3,4 @@
> /process_vm_readv03
> /process_vm_writev02
> /process_madvise01
> +/process_madvise02
> diff --git a/testcases/kernel/syscalls/cma/process_madvise02.c b/testcases/kernel/syscalls/cma/process_madvise02.c
> new file mode 100644
> index 000000000..9f42dd48a
> --- /dev/null
> +++ b/testcases/kernel/syscalls/cma/process_madvise02.c
> @@ -0,0 +1,128 @@
> +// SPDX-License-Identifier: GPL-2.0-or-later
> +/*
> + * Copyright (C) 2022 SUSE LLC Andrea Cervesato <andrea.cervesato@suse.com>
> + */
> +
> +/*\
> + * [Description]
> + *
> + * Spawn child inside cgroup and set max memory. Allocate anonymous memory
> + * pages inside child and reclaim it with MADV_PAGEOUT. Then check if memory
> + * pages have been swapped out.
> + *
> + * The advice might be ignored for some pages in the range when it is
> + * not applicable, so test passes if swap memory increases after
> + * reclaiming memory with MADV_PAGEOUT.
> + */
> +
> +#define _GNU_SOURCE
> +
> +#include <sys/mman.h>
> +#include "tst_test.h"
> +#include "lapi/mmap.h"
> +#include "lapi/syscalls.h"
> +#include "cma.h"
> +
> +#define MEM_CHILD (10 * 1024 * 1024)
> +
> +static void **data_ptr;
> +
> +static void child_alloc(void)
> +{
> + char *ptr;
> + int freed = 1;
> + struct addr_mapping map_before;
> + struct addr_mapping map_after;
> +
> + tst_res(TINFO, "Allocate memory");
> +
> + *data_ptr = SAFE_MMAP(NULL, MEM_CHILD,
> + PROT_READ | PROT_WRITE,
> + MAP_SHARED | MAP_ANONYMOUS, -1, 0);
> + memset(*data_ptr, 'a', MEM_CHILD);
> +
> + read_address_mapping((unsigned long)*data_ptr, &map_before);
> +
> + TST_CHECKPOINT_WAKE_AND_WAIT(0);
> +
> + for (ptr = *data_ptr; *ptr != '\0'; ptr++) {
> + if (*ptr == 'a') {
> + freed = 0;
> + break;
> + }
> + }
> +
> + if (freed) {
> + tst_res(TFAIL, "Memory has been freed");
> + return;
> + }
> +
> + read_address_mapping((unsigned long)*data_ptr, &map_after);
> +
> + SAFE_MUNMAP(*data_ptr, MEM_CHILD);
> + *data_ptr = NULL;
> +
> + if (map_before.swap < map_after.swap)
> + tst_res(TPASS, "Memory has been swapped out");
> + else
> + tst_res(TFAIL, "Swap memory has decreased");
> +}
> +
> +static void setup(void)
> +{
> + data_ptr = SAFE_MMAP(NULL, sizeof(void *),
> + PROT_READ | PROT_WRITE,
> + MAP_SHARED | MAP_ANONYMOUS, -1, 0);
> +}
> +
> +static void cleanup(void)
> +{
> + if (*data_ptr)
> + SAFE_MUNMAP(*data_ptr, MEM_CHILD);
> +
> + if (data_ptr)
> + SAFE_MUNMAP(data_ptr, sizeof(void *));
> +}
> +
> +static void run(void)
> +{
> + int ret;
> + int pidfd;
> + pid_t pid_alloc;
> + struct iovec vec;
> +
> + pid_alloc = SAFE_FORK();
> + if (!pid_alloc) {
> + child_alloc();
> + return;
> + }
> +
> + TST_CHECKPOINT_WAIT(0);
> +
> + tst_res(TINFO, "Apply MADV_PAGEOUT advise rule");
> +
> + pidfd = SAFE_PIDFD_OPEN(pid_alloc, 0);
> +
> + vec.iov_base = *data_ptr;
> + vec.iov_len = MEM_CHILD;
> +
> + ret = tst_syscall(__NR_process_madvise, pidfd, &vec, 1UL,
> + MADV_PAGEOUT, 0UL);
> +
> + if (ret == -1)
> + tst_brk(TBROK | TERRNO, "process_madvise failed");
> +
> + if (ret != MEM_CHILD)
> + tst_brk(TBROK, "process_madvise reclaimed only %d bytes", ret);
> +
> + TST_CHECKPOINT_WAKE(0);
> +}
> +
> +static struct tst_test test = {
> + .setup = setup,
> + .cleanup = cleanup,
> + .test_all = run,
> + .forks_child = 1,
> + .min_kver = "5.10",
> + .needs_checkpoints = 1,
> +};
> --
> 2.35.3
--
Thank you,
Richard.
--
Mailing list info: https://lists.linux.it/listinfo/ltp
next prev parent reply other threads:[~2022-10-18 12:26 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-06 11:06 [LTP] [PATCH v1 0/3] Add process_madvise support Andrea Cervesato via ltp
2022-10-06 11:06 ` [LTP] [PATCH v1 1/3] Add process_madvise01 test Andrea Cervesato via ltp
2022-10-18 11:51 ` Richard Palethorpe
2022-10-06 11:06 ` [LTP] [PATCH v1] Rewrite process_vm01 test using new LTP API Andrea Cervesato via ltp
2022-10-06 11:10 ` Andrea Cervesato via ltp
2022-10-06 11:06 ` [LTP] [PATCH v1 2/3] Add process_madvise02 test Andrea Cervesato via ltp
2022-10-18 12:24 ` Richard Palethorpe [this message]
2022-10-26 7:50 ` Andrea Cervesato via ltp
2022-10-06 11:06 ` [LTP] [PATCH v1 3/3] Add process_madvise03 test Andrea Cervesato via ltp
2022-10-18 12:29 ` Richard Palethorpe
2022-10-18 12:49 ` Richard Palethorpe
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=87a65tjp22.fsf@suse.de \
--to=rpalethorpe@suse.de \
--cc=andrea.cervesato@suse.com \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox