Skip to contentSkip to conversation

syscalls/preadv203: Add basic RWF_NOWAIT test

PATCH New Cyril Hrubis 5 participants 8 messages Jan 28, 2019 – Feb 21, 2019

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 syscalls/preadv203: Add basic RWF_NOWAIT test

From: Cyril Hrubis <[email protected]>

We are attempting to trigger the EAGAIN path for the RWF_NOWAIT flag.

In order to do so the test runs three threads:

* nowait_reader: reads from a random offset from a random file with RWF_NOWAIT flag and expects to get EAGAIN and short read sooner or later

* writer_thread: rewrites random file in order to keep the underlying device bussy so that pages evicted from cache cannot be faulted immediatelly

* cache_dropper: attempts to evict pages from a cache in order for reader to hit evicted page sooner or later

CC Jiri Kosina <[email protected]>
CC Linux-MM <[email protected]>
CC kernel list <[email protected]>
CC Linux API <[email protected]>

---

I was wondering if we can do a better job at flushing the caches. Is there an interface for flusing caches just for the device we are using for the test?

Also the RWF_NOWAIT should probably be benchmarked as well but that is completely out of scope for LTP.

 runtest/syscalls                              |   2 +
 testcases/kernel/syscalls/preadv2/.gitignore  |   2 +
 testcases/kernel/syscalls/preadv2/Makefile    |   4 +
 testcases/kernel/syscalls/preadv2/preadv203.c | 266 ++++++++++++++++++
 4 files changed, 274 insertions(+)
 create mode 100644 testcases/kernel/syscalls/preadv2/preadv203.c
runtest/syscalls +2 -0
index 34b47f36b..a69c431f1 100644
@@ -853,6 +853,8 @@ preadv201 preadv201
 preadv201_64 preadv201_64
 preadv202 preadv202
 preadv202_64 preadv202_64
+preadv203 preadv203
+preadv203_64 preadv203_64
 
 profil01 profil01
 
testcases/kernel/syscalls/preadv2/.gitignore +2 -0
index 759d9ef5b..98b81abea 100644
@@ -2,3 +2,5 @@
 /preadv201_64
 /preadv202
 /preadv202_64
+/preadv203
+/preadv203_64
testcases/kernel/syscalls/preadv2/Makefile +4 -0
index fc1fbf3c7..fbedd0287 100644
@@ -11,4 +11,8 @@ include $(abs_srcdir)/../utils/newer_64.mk
 
 %_64: CPPFLAGS += -D_FILE_OFFSET_BITS=64
 
+preadv203: CFLAGS += -pthread
+preadv203_64: CFLAGS += -pthread
+preadv203_64: LDFLAGS += -pthread
+
 include $(top_srcdir)/include/mk/generic_leaf_target.mk
testcases/kernel/syscalls/preadv2/preadv203.c new file +266 -0

266 lines not on this page · raw message

Signature · 1 line
2.19.2
#1 lore raw

PATCH 1/1 syscalls/preadv203: Add basic RWF_NOWAIT test

Hi Cyril,

Quoted 1 line

From: Cyril Hrubis <[email protected]>

Quoted 10 lines

We are attempting to trigger the EAGAIN path for the RWF_NOWAIT flag.

In order to do so the test runs three threads:

* nowait_reader: reads from a random offset from a random file with RWF_NOWAIT flag and expects to get EAGAIN and short read sooner or later

* writer_thread: rewrites random file in order to keep the underlying device bussy so that pages evicted from cache cannot be faulted

typo => busy

Quoted 1 line

immediatelly

typo => immediately

...

testcases/kernel/syscalls/preadv2/preadv203.c line 9

Quoted 11 lines

+++ b/testcases/kernel/syscalls/preadv2/preadv203.c

testcases/kernel/syscalls/preadv2/preadv203.c new file +9 -0
@@ -0,0 +1,266 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) 2019 Cyril Hrubis <[email protected]>
+ */
+
+/*
+ * This is a basic functional test for RWF_NOWAIT flag, we are attempting to
+ * force preadv2() either to return a short read or EAGAIN with three
+ * concurelntly running threads:

typo => concurrently

testcases/kernel/syscalls/preadv2/preadv203.c line 16

Quoted 7 lines
testcases/kernel/syscalls/preadv2/preadv203.c new file +7 -0
+ *
+ *  nowait_reader: reads from a random offset from a random file with
+ *                 RWF_NOWAIT flag and expects to get EAGAIN and short
+ *                 read sooner or later
+ *
+ *  writer_thread: rewrites random file in order to keep the underlying device
+ *                 bussy so that pages evicted from cache cannot be faulted

typo => busy

testcases/kernel/syscalls/preadv2/preadv203.c line 17

Quoted 1 line
testcases/kernel/syscalls/preadv2/preadv203.c new file +1 -0
+ *                 immediatelly

typo => immediately

Quoted 7 lines
testcases/kernel/syscalls/preadv2/preadv203.c new file +7 -0
+ *
+ *  cache_dropper: attempts to evict pages from a cache in order for reader to
+ *                 hit evicted page sooner or later
+ */
+
+/*
+ * If test fails with EOPNOTSUPP you have likely hit a glibc bug:

testcases/kernel/syscalls/preadv2/preadv203.c line 34

Quoted 10 lines
testcases/kernel/syscalls/preadv2/preadv203.c new file +10 -0
+ *
+ * https://sourceware.org/bugzilla/show_bug.cgi?id=23579
+ *
+ * Which can be worked around by calling preadv2() directly by syscall() such as:
+ *
+ * static ssize_t sys_preadv2(int fd, const struct iovec *iov, int iovcnt,
+ *                            off_t offset, int flags)
+ * {
+ *	return syscall(SYS_preadv2, fd, iov, iovcnt, offset, offset>>32, flags);
+ * }

I wonder if we want either warn user or run it both via (g)libc wrapper and directly the syscall.

BTW testing on kernel 4.20.0, glibc 2.28 with sys_preadv2() and still get TBROK EOPNOTSUPP. I tried to test on ext[2-4], btrfs. I wonder what I'm missing.

Kind regards,
Petr

#2 lore raw

PATCH 1/1 syscalls/preadv203: Add basic RWF_NOWAIT test

Hi!

Quoted 2 lines

I wonder if we want either warn user or run it both via (g)libc wrapper and directly the syscall.

I guess that we may drop a hint in the error message.

Quoted 2 lines

BTW testing on kernel 4.20.0, glibc 2.28 with sys_preadv2() and still get TBROK EOPNOTSUPP. I tried to test on ext[2-4], btrfs. I wonder what I'm missing.

That's strange, are you sure that the test was calling the wrapper instead of the glibc call?

This should really work unless the support in kernel has been disabled in distribution kernel for some reason. I was testing it on vanilla kernel where it worked as expected.

Signature · 2 lines
Cyril Hrubis
[email protected]
#3 lore raw

PATCH 1/1 syscalls/preadv203: Add basic RWF_NOWAIT test

Hi Cyril,

Quoted 10 lines
Quoted 2 lines

I wonder if we want either warn user or run it both via (g)libc wrapper and directly the syscall.

I guess that we may drop a hint in the error message.

Quoted 2 lines

BTW testing on kernel 4.20.0, glibc 2.28 with sys_preadv2() and still get TBROK EOPNOTSUPP. I tried to test on ext[2-4], btrfs. I wonder what I'm missing.

That's strange, are you sure that the test was calling the wrapper instead of the glibc call?

See diff below, but I guess it's what you meant :).

Quoted 3 lines

This should really work unless the support in kernel has been disabled in distribution kernel for some reason. I was testing it on vanilla kernel where it worked as expected.

What config option is needed?

It does not work on both my machines (glibc 2.27 and 2.28) and it's working on various VMs with these versions.

Kind regards,
Petr

+++ testcases/kernel/syscalls/preadv2/preadv203.c
@@ -95,6 +95,12 @@ static int verify_short_read(struct iovec *iov, size_t iov_cnt,
return 0;
}

+ static ssize_t sys_preadv2(int fd, const struct iovec *iov, int iovcnt,
+ off_t offset, int flags)
+ {
+ return syscall(SYS_preadv2, fd, iov, iovcnt, offset, offset>>32, flags);
+ }
+
static void *nowait_reader(void *unused LTP_ATTRIBUTE_UNUSED)
{
char buf1[CHUNK_SZ/2];
@@ -115,7 +121,7 @@ static void *nowait_reader(void *unused LTP_ATTRIBUTE_UNUSED)
off_t off = random() % ((CHUNKS - 2) * CHUNK_SZ); int fd = fds[random() % FILES];

- TEST(preadv2(fd, rd_iovec, 2, off, RWF_NOWAIT));
+ TEST(sys_preadv2(fd, rd_iovec, 2, off, RWF_NOWAIT));

if (TST_RET < 0) {
if (TST_ERR != EAGAIN)

#4 lore raw

PATCH 1/1 syscalls/preadv203: Add basic RWF_NOWAIT test

In reply to Cyril Hrubis

Hi Cyril,

Quoted 1 line

+static void *nowait_reader(void *unused LTP_ATTRIBUTE_UNUSED)

...

testcases/kernel/syscalls/preadv2/preadv203.c line 118

Quoted 1 line
testcases/kernel/syscalls/preadv2/preadv203.c new file +1 -0
+		TEST(preadv2(fd, rd_iovec, 2, off, RWF_NOWAIT));

RWF_NOWAIT needs to be declared in lapi for old distros.

Kind regards,
Petr

#5 lore raw

PATCH 1/1 syscalls/preadv203: Add basic RWF_NOWAIT test

Hi!

Quoted 4 lines
Quoted 1 line

+static void *nowait_reader(void *unused LTP_ATTRIBUTE_UNUSED)

...

testcases/kernel/syscalls/preadv2/preadv203.c line 118

Quoted 1 line
testcases/kernel/syscalls/preadv2/preadv203.c new file +1 -0
+		TEST(preadv2(fd, rd_iovec, 2, off, RWF_NOWAIT));

RWF_NOWAIT needs to be declared in lapi for old distros.

Right and we should as well return TCONF on older kernels, I haven't tested that properly yet but I guess that we should check for EOPNOTSUPP and EINVAL. I will do that later on, I just didn't want to delay the initial test patch because it seem that kernel developers wanted to have it ASAP.

Signature · 2 lines
Cyril Hrubis
[email protected]
#6 lore raw

PATCH 1/1 syscalls/preadv203: Add basic RWF_NOWAIT test

Quoted 4 lines · Cyril Hrubis

Hi!

Quoted 2 lines
Quoted 1 line

+static void *nowait_reader(void *unused LTP_ATTRIBUTE_UNUSED)

...

Quoted 8 lines
Quoted 2 lines
Quoted 1 line

+ TEST(preadv2(fd, rd_iovec, 2, off, RWF_NOWAIT));

RWF_NOWAIT needs to be declared in lapi for old distros.

Right and we should as well return TCONF on older kernels, I haven't tested that properly yet but I guess that we should check for EOPNOTSUPP and EINVAL. I will do that later on, I just didn't want to delay the initial test patch because it seem that kernel developers wanted to have it ASAP.

LTP syscalls test preadv201 and preadv201_64 failed on 4.4 branch kernel on all devices.
PASS 4.9, 4.14, 4.20 kernel versions.

OTOH,
preadv202 and preadv202_64 PASS on 4.4, 4.9, 4.14 and 4.20 kernel versions.

Test output log,
tst_test.c:1085: INFO: Timeout per run is 0h 15m 00s preadv201.c:91: PASS: preadv2() read 64 bytes with content 'a' expectedly preadv201.c:91: PASS: preadv2() read 64 bytes with content 'a' expectedly preadv201.c:91: PASS: preadv2() read 32 bytes with content 'b' expectedly preadv201.c:64: FAIL: preadv2() failed: EINVAL preadv201.c:64: FAIL: preadv2() failed: EINVAL preadv201.c:64: FAIL: preadv2() failed: EINVAL
Summary:

passed   3
failed   3
skipped  0

warnings 0
tst_test.c:1085: INFO: Timeout per run is 0h 15m 00s preadv201.c:91: PASS: preadv2() read 64 bytes with content 'a' expectedly preadv201.c:91: PASS: preadv2() read 64 bytes with content 'a' expectedly preadv201.c:91: PASS: preadv2() read 32 bytes with content 'b' expectedly preadv201.c:64: FAIL: preadv2() failed: EINVAL preadv201.c:64: FAIL: preadv2() failed: EINVAL preadv201.c:64: FAIL: preadv2() failed: EINVAL
Summary:

passed   3
failed   3
skipped  0

warnings 0

Full test log from 4.4 branch kernel x86_64,
https://lkft.validation.linaro.org/scheduler/job/614534#L7922

https://qa-reports.linaro.org/lkft/linux-stable-rc-4.4-oe/tests/ltp-syscalls-tests/preadv201

- Naresh

#7 lore raw

PATCH 1/1 syscalls/preadv203: Add basic RWF_NOWAIT test

In reply to Cyril Hrubis
Quoted 23 lines · Cyril Hrubis

From: Cyril Hrubis <[email protected]>

We are attempting to trigger the EAGAIN path for the RWF_NOWAIT flag.

In order to do so the test runs three threads:

* nowait_reader: reads from a random offset from a random file with RWF_NOWAIT flag and expects to get EAGAIN and short read sooner or later

* writer_thread: rewrites random file in order to keep the underlying device bussy so that pages evicted from cache cannot be faulted immediatelly

* cache_dropper: attempts to evict pages from a cache in order for reader to hit evicted page sooner or later

CC Jiri Kosina <[email protected]>
CC Linux-MM <[email protected]>
CC kernel list <[email protected]>
CC Linux API <[email protected]>
Quoted 6 lines

---

I was wondering if we can do a better job at flushing the caches. Is there an interface for flusing caches just for the device we are using for the test?

Will the BLKFLSBUF ioctl be suitable for this purpose?

Regards,
Vijay