| d8cd45ca8cb268a3fcf4d0679d5fdd24f499ece6 |
|
03-Oct-2017 |
Cyril Hrubis <chrubis@suse.cz> |
Make use of SAFE_OPEN()
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 7bf0cbd87d8981d0a1d034ca4e6a24219e5a77c1 |
|
03-Oct-2017 |
Cyril Hrubis <chrubis@suse.cz> |
Make use of SAFE_MUNMAP()
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| df007542635aed8686b5ebc111b0807a67349b41 |
|
03-Oct-2017 |
Cyril Hrubis <chrubis@suse.cz> |
Make use of SAFE_CLOSE()
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 817d8095fbfe5ebe22b29f58b7483f3d8d5e3fcc |
|
24-Jul-2017 |
Richard Palethorpe <rpalethorpe@suse.com> |
Rename linux_syscall_numbers.h to lapi/syscalls.h
Move testcases/kernel/include/linux_syscall_numbers.h to lapi/syscalls.h so
that it is easily accessible from the LTP library as well as the tests.
Signed-off-by: Richard Palethorpe <rpalethorpe@suse.com>
Acked-by: Cyril Hrubis <chrubis@suse.cz>
Acked-by: Jan Stancek <jstancek@redhat.com>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| d62f6f8486394a6d35ca5cc1f8069c1e26485188 |
|
26-May-2017 |
Cyril Hrubis <chrubis@suse.cz> |
readahead02: Fix warnings.
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| eac3375d4b90c9784c20e893ab732def44ed0c31 |
|
15-May-2017 |
Jan Stancek <jstancek@redhat.com> |
readahead02: include errno in failure message
Signed-off-by: Jan Stancek <jstancek@redhat.com>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| fb5ce4796adb606cc32b673b5187aba36fd7cbe8 |
|
10-Oct-2016 |
Jan Stancek <jstancek@redhat.com> |
readahead02: estimate max readahead size
Max readahead size is kernel implementation detail, which can and
already has changed in past (and probably will again [1]).
Futher, current (4.8) implementation defines it as block device's
read_ahead_kb, which means its value varies based on storage/fs
setup.
This patch estimates max readahead size based on cache increase
from first readahead call and then advances offset for subsequent
calls by that amount. It also makes sure it's > 0 to guarantee,
that it eventually reaches end of file.
This also fixes test on Btrfs, where it wasn't working previously
since there is no /sys/dev/block/$major:$minor/ for the anonymous
block devices that are used internaly for subvolumes.
[1] https://lkml.org/lkml/2016/7/25/308
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Tested-by: Cyril Hrubis <chrubis@suse.cz>
Acked-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 29a24639f7b5c69cd9a397201e364a5e863d57cc |
|
14-Sep-2016 |
Jan Stancek <jstancek@redhat.com> |
readahead02: fix path for read_ahead_kb on raw partitions
Path to read_ahead_kb is not consistent for all block devices,
for example for LVM it's:
/sys/dev/block/$maj:$min/queue/read_ahead_kb
but for raw partitions "queue" is only in "parent" device:
/sys/dev/block/$maj:$min/../queue/read_ahead_kb
This patch checks both types and reads first one it can find.
Tested on LVM and raw partitions with 4.7.3.
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Acked-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 38396b3e4c72a04679f6a209861f032f711f5cc5 |
|
24-Mar-2016 |
Jan Stancek <jstancek@redhat.com> |
readahead: use glibc wrapper function
We have had reports of this testcase not functioning
properly on 32bit powerpc systems, because syscall()
is not passing 64bit parameters correctly.
This patch is using glibc wrapper, which should be available
for a long time now (since glibc 2.3).
Reported-and-tested-by: Julio Cruz Barroso <julio.cruz@smartmatic.com>
Reported-and-tested-by: Kshitij Malik <Kshitij.Malik@mitel.com>
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Acked-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| f6b7a503db7dd5c9c957e64a90604791f8f6c19c |
|
15-Mar-2016 |
Roman Gushchin <klamm@yandex-team.ru> |
syscalls/readahead02: use bdi max readahead limit
After commit 600e19afc5f8a6 ("mm: use only per-device readahead limit")
readahead size is not limited anymore by 2 megabytes. Instead,
the per-bdi limit is used for kernels 4.4 and older.
Signed-off-by: Roman Gushchin <klamm@yandex-team.ru>
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Acked-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| ae4e530df2df456749a0afdce8b22c27249a0af0 |
|
15-Mar-2016 |
Jan Stancek <jstancek@redhat.com> |
readahead02: don't underflow cached/cached_ra
Julio Cruz reported a failure, where readahead failed to utilize
any cache, moreover cache usage went down, which resulted in
cached_ra underflow.
Reported-by: Julio Cruz <jcsistemas2001@gmail.com>
Signed-off-by: Jan Stancek <jstancek@redhat.com>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| d1e794d62b1bf619df8390535e4c2a58899b1145 |
|
30-Jul-2015 |
Cyril Hrubis <chrubis@suse.cz> |
lib: Get rid of unused tst_require_root() callback
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| d6d11d08678aac1ed2c370ea8e42e5f45aea07be |
|
09-Mar-2015 |
Cyril Hrubis <chrubis@suse.cz> |
Introduce tst_parse_opts()
The pattern that was used in all testcases is:
const char *msg;
msg = parse_opts(...);
if (msg)
tst_brkm(TBROK, NULL, "OPTION PARSING ERROR - %s", msg);
This change simplifies the steps to just calling:
tst_parse_opts(...);
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| aabb8340f63ed31afe995fd97795e542dc68b93c |
|
04-Feb-2015 |
Cyril Hrubis <chrubis@suse.cz> |
Include usctest.h in test.h
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 605fa3362fd7cef0baa2131be32cf44661783d3e |
|
04-Feb-2015 |
Cyril Hrubis <chrubis@suse.cz> |
Get rid of TEST_CLEANUP
Special thanks to Coccinelle for making this easy job.
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 0b9589f3f9c0345b29cfcf7da5a1253c708303eb |
|
27-May-2014 |
Cyril Hrubis <chrubis@suse.cz> |
testcases: Constify char *msg variable.
The parse_opts() now returns const char * instead of char * this commit
fixes all warnings caused by the change.
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 09bea28b888f0fd22edda956243a1dac00617af8 |
|
22-Apr-2014 |
Jan Stancek <jstancek@redhat.com> |
readahead02: read in 2M chunks
Max readahead size has been limited since commit:
commit 6d2be915e589b58cb11418cbe1f22ff90732b6ac
Author: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
Date: Thu Apr 3 14:48:23 2014 -0700
mm/readahead.c: fix readahead failure for memoryless NUMA nodes
and limit readahead pages
Update testcase to read ahead in 2M chunks, this change is
compatible with older kernels.
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Acked-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 02ee88cf252e530f2fabd1bb2b72945232771176 |
|
22-Apr-2014 |
Jan Stancek <jstancek@redhat.com> |
readahead02: fix signed/unsigned comparison warnings
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Acked-by: Cyril Hrubis <chrubis@suse.cz>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| d59a659cd639ca2780b00049d102acd2a783d585 |
|
07-Mar-2013 |
Caspar Zhang <caspar@casparzhang.com> |
s/Tst_count/tst_count/ to avoid CamelCase
when using latest checkpatch.pl, CamelCase variables are not encouraged,
this patch does a batch update to replace all var `Tst_count` with
`tst_count`.
Signed-off-by: Caspar Zhang <caspar@casparzhang.com>
Acked-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 359980f68b19c77c698b121b57a071dfe6e3ca31 |
|
15-Feb-2013 |
Jan Stancek <jstancek@redhat.com> |
rename syscall() macro to ltp_syscall()
This patch renames syscall() macro defined in linux_syscall_numbers.h
to reduce confusion with syscall(2). It also updates all places
where it was previously in use.
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| 354ebb48db8e66a853a58379a4808d5dcd1ceac3 |
|
07-Dec-2012 |
Wanlong Gao <gaowanlong@cn.fujitsu.com> |
cleanup code indent
Cleanup the code indent using:
find . -name *.c -exec Lindent {} \;
It's really a big change, but can fix almost all of
the indent problem in C code, although we can't
ensure all of the changes are right, but the error
changes are really few.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|
| dd4927f8f9a05148343fefe4b08561783621ac98 |
|
15-Oct-2012 |
Jan Stancek <jstancek@redhat.com> |
new syscall test: readahead02
Use readahead() syscall and check if subsequent
read saves some I/O and if cache size increased.
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Reviewed-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
/external/ltp/testcases/kernel/syscalls/readahead/readahead02.c
|