2019-03-07 17:58:38 +03:00
|
|
|
#!/usr/bin/env bash
|
2015-05-08 19:06:47 +03:00
|
|
|
#
|
|
|
|
|
# Test for configuring cache modes of arbitrary nodes (requires O_DIRECT)
|
|
|
|
|
#
|
|
|
|
|
# Copyright (C) 2015 Red Hat, Inc.
|
|
|
|
|
#
|
|
|
|
|
# This program is free software; you can redistribute it and/or modify
|
|
|
|
|
# it under the terms of the GNU General Public License as published by
|
|
|
|
|
# the Free Software Foundation; either version 2 of the License, or
|
|
|
|
|
# (at your option) any later version.
|
|
|
|
|
#
|
|
|
|
|
# This program is distributed in the hope that it will be useful,
|
|
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
|
# GNU General Public License for more details.
|
|
|
|
|
#
|
|
|
|
|
# You should have received a copy of the GNU General Public License
|
|
|
|
|
# along with this program. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
|
#
|
|
|
|
|
|
|
|
|
|
# creator
|
|
|
|
|
owner=kwolf@redhat.com
|
|
|
|
|
|
|
|
|
|
seq=`basename $0`
|
|
|
|
|
echo "QA output created by $seq"
|
|
|
|
|
|
|
|
|
|
status=1 # failure is the default!
|
|
|
|
|
|
|
|
|
|
_cleanup()
|
|
|
|
|
{
|
|
|
|
|
_cleanup_test_img
|
2019-11-07 19:37:01 +03:00
|
|
|
_rm_test_img "$TEST_IMG.snap"
|
2015-05-08 19:06:47 +03:00
|
|
|
}
|
|
|
|
|
trap "_cleanup; exit \$status" 0 1 2 3 15
|
|
|
|
|
|
|
|
|
|
# get standard environment, filters and checks
|
|
|
|
|
. ./common.rc
|
|
|
|
|
. ./common.filter
|
|
|
|
|
|
|
|
|
|
_supported_fmt qcow2
|
|
|
|
|
_supported_proto file
|
|
|
|
|
|
|
|
|
|
# We test all cache modes anyway, but O_DIRECT needs to be supported
|
|
|
|
|
_default_cache_mode none
|
|
|
|
|
_supported_cache_modes none directsync
|
|
|
|
|
|
2018-11-17 00:50:02 +03:00
|
|
|
do_run_qemu()
|
2015-05-08 19:06:47 +03:00
|
|
|
{
|
|
|
|
|
echo Testing: "$@"
|
|
|
|
|
(
|
|
|
|
|
if ! test -t 0; then
|
|
|
|
|
while read cmd; do
|
|
|
|
|
echo $cmd
|
|
|
|
|
done
|
|
|
|
|
fi
|
|
|
|
|
echo quit
|
|
|
|
|
) | $QEMU -nographic -monitor stdio -nodefaults "$@"
|
|
|
|
|
echo
|
|
|
|
|
}
|
|
|
|
|
|
2018-11-17 00:50:02 +03:00
|
|
|
run_qemu()
|
2015-05-08 19:06:47 +03:00
|
|
|
{
|
2017-04-13 20:19:51 +03:00
|
|
|
do_run_qemu "$@" 2>&1 | _filter_testdir | _filter_qemu | _filter_hmp
|
2015-05-08 19:06:47 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
size=128M
|
|
|
|
|
|
|
|
|
|
TEST_IMG="$TEST_IMG.base" _make_test_img $size
|
|
|
|
|
TEST_IMG="$TEST_IMG.snap" _make_test_img $size
|
iotests: Specify explicit backing format where sensible
There are many existing qcow2 images that specify a backing file but
no format. This has been the source of CVEs in the past, but has
become more prominent of a problem now that libvirt has switched to
-blockdev. With older -drive, at least the probing was always done by
qemu (so the only risk of a changed format between successive boots of
a guest was if qemu was upgraded and probed differently). But with
newer -blockdev, libvirt must specify a format; if libvirt guesses raw
where the image was formatted, this results in data corruption visible
to the guest; conversely, if libvirt guesses qcow2 where qemu was
using raw, this can result in potential security holes, so modern
libvirt instead refuses to use images without explicit backing format.
The change in libvirt to reject images without explicit backing format
has pointed out that a number of tools have been far too reliant on
probing in the past. It's time to set a better example in our own
iotests of properly setting this parameter.
iotest calls to create, rebase, and convert are all impacted to some
degree. It's a bit annoying that we are inconsistent on command line
- while all of those accept -o backing_file=...,backing_fmt=..., the
shortcuts are different: create and rebase have -b and -F, while
convert has -B but no -F. (amend has no shortcuts, but the previous
patch just deprecated the use of amend to change backing chains).
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <20200706203954.341758-9-eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2020-07-06 23:39:52 +03:00
|
|
|
_make_test_img -b "$TEST_IMG.base" $size -F $IMGFMT
|
2015-05-08 19:06:47 +03:00
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo === Simple test for all cache modes ===
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
run_qemu -drive file="$TEST_IMG",cache=none
|
|
|
|
|
run_qemu -drive file="$TEST_IMG",cache=directsync
|
|
|
|
|
run_qemu -drive file="$TEST_IMG",cache=writeback
|
|
|
|
|
run_qemu -drive file="$TEST_IMG",cache=writethrough
|
|
|
|
|
run_qemu -drive file="$TEST_IMG",cache=unsafe
|
|
|
|
|
run_qemu -drive file="$TEST_IMG",cache=invalid_value
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo === Check inheritance of cache modes ===
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
files="if=none,file=$TEST_IMG,backing.file.filename=$TEST_IMG.base"
|
|
|
|
|
ids="node-name=image,backing.node-name=backing,backing.file.node-name=backing-file,file.node-name=file"
|
|
|
|
|
|
2018-11-17 00:50:02 +03:00
|
|
|
check_cache_all()
|
2015-05-08 19:06:47 +03:00
|
|
|
{
|
|
|
|
|
# cache.direct is supposed to be inherited by both bs->file and
|
|
|
|
|
# bs->backing
|
|
|
|
|
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "cache.direct=on on none0\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",cache.direct=on | grep -e "Cache" -e "[Cc]annot|[Cc]ould not|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.direct=on on file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",file.cache.direct=on | grep -e "Cache" -e "[Cc]annot|[Cc]ould not|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.direct=on on backing\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",backing.cache.direct=on | grep -e "Cache" -e "[Cc]annot|[Cc]ould not|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.direct=on on backing-file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",backing.file.cache.direct=on | grep -e "Cache" -e "[Cc]annot|[Cc]ould not|[Cc]an't"
|
2015-05-08 19:06:47 +03:00
|
|
|
|
|
|
|
|
# cache.writeback is supposed to be inherited by bs->backing; bs->file
|
|
|
|
|
# always gets cache.writeback=on
|
|
|
|
|
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\n\ncache.writeback=off on none0\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",cache.writeback=off | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.writeback=off on file\n"
|
2016-03-18 19:46:45 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",file.cache.writeback=off | grep -e "doesn't" -e "does not"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.writeback=off on backing\n"
|
2016-03-18 19:46:45 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",backing.cache.writeback=off | grep -e "doesn't" -e "does not"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.writeback=off on backing-file\n"
|
2016-03-18 19:46:45 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",backing.file.cache.writeback=off | grep -e "doesn't" -e "does not"
|
2015-05-08 19:06:47 +03:00
|
|
|
|
|
|
|
|
# cache.no-flush is supposed to be inherited by both bs->file and bs->backing
|
|
|
|
|
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\n\ncache.no-flush=on on none0\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",cache.no-flush=on | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.no-flush=on on file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",file.cache.no-flush=on | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.no-flush=on on backing\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",backing.cache.no-flush=on | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.no-flush=on on backing-file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids",backing.file.cache.no-flush=on | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
2015-05-08 19:06:47 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Configure cache modes on the command line ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# First check the inherited cache mode after opening the image.
|
|
|
|
|
|
2016-03-03 13:37:48 +03:00
|
|
|
hmp_cmds="info block none0
|
|
|
|
|
info block image
|
2015-05-08 19:06:47 +03:00
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file"
|
|
|
|
|
|
|
|
|
|
check_cache_all
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Cache modes after reopen (live snapshot) ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# Then trigger a reopen and check that the cache modes are still the same.
|
|
|
|
|
|
|
|
|
|
hmp_cmds="snapshot_blkdev -n none0 $TEST_IMG.snap $IMGFMT
|
|
|
|
|
info block
|
|
|
|
|
info block image
|
|
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file"
|
|
|
|
|
|
|
|
|
|
check_cache_all
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Change cache modes with reopen (qemu-io command, flags) ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# This one actually changes the cache mode with the reopen. For this test, the
|
|
|
|
|
# new cache mode is specified in the flags, not as an option.
|
|
|
|
|
|
|
|
|
|
hmp_cmds='qemu-io none0 "reopen -c none"
|
2016-03-03 13:37:48 +03:00
|
|
|
info block none0
|
2015-05-08 19:06:47 +03:00
|
|
|
info block image
|
|
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file'
|
|
|
|
|
|
|
|
|
|
check_cache_all
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Change cache modes with reopen (qemu-io command, options) ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# This one actually changes the cache mode with the reopen. For this test, the
|
|
|
|
|
# new cache mode is specified as an option, not in the flags.
|
|
|
|
|
|
|
|
|
|
hmp_cmds='qemu-io none0 "reopen -o cache.direct=on"
|
2016-03-03 13:37:48 +03:00
|
|
|
info block none0
|
2015-05-08 19:06:47 +03:00
|
|
|
info block image
|
|
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file'
|
|
|
|
|
|
|
|
|
|
check_cache_all
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Change cache modes after snapshot ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# This checks that the original image doesn't inherit from the snapshot
|
|
|
|
|
|
|
|
|
|
hmp_cmds="snapshot_blkdev -n none0 $TEST_IMG.snap $IMGFMT
|
|
|
|
|
qemu-io none0 \"reopen -c none\"
|
|
|
|
|
info block none0
|
|
|
|
|
info block image
|
|
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file"
|
|
|
|
|
|
|
|
|
|
check_cache_all
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Change cache mode in parent, child has explicit option in JSON ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# This checks that children with options explicitly set by the json:
|
|
|
|
|
# pseudo-protocol don't inherit these options from their parents.
|
|
|
|
|
#
|
|
|
|
|
# Yes, blkdebug::json:... is criminal, but I can't see another way to have a
|
|
|
|
|
# BDS initialised with the json: pseudo-protocol, but still have it inherit
|
|
|
|
|
# options from its parent node.
|
|
|
|
|
|
2016-03-18 17:36:31 +03:00
|
|
|
hmp_cmds="qemu-io none0 \"reopen -o cache.direct=on,cache.no-flush=on\"
|
2016-03-03 13:37:48 +03:00
|
|
|
info block none0
|
2015-05-08 19:06:47 +03:00
|
|
|
info block image
|
|
|
|
|
info block blkdebug
|
|
|
|
|
info block file"
|
|
|
|
|
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive if=none,file="blkdebug::json:{\"filename\":\"$TEST_IMG\",,\"cache\":{\"direct\":false}}",node-name=image,file.node-name=blkdebug,file.image.node-name=file | grep "Cache"
|
2015-05-08 19:06:47 +03:00
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "=== Check that referenced BDSes don't inherit ==="
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
drv_bkfile="if=none,driver=file,filename=$TEST_IMG.base,node-name=backing-file"
|
|
|
|
|
drv_bk="if=none,file=json:{'driver':'$IMGFMT',,'file':'backing-file',,'node-name':'backing'}"
|
|
|
|
|
drv_file="if=none,driver=file,filename=$TEST_IMG,node-name=file"
|
|
|
|
|
drv_img="if=none,id=blk,file=json:{'driver':'$IMGFMT',,'file':'file',,'backing':'backing',,'node-name':'image'}"
|
|
|
|
|
|
2018-11-17 00:50:02 +03:00
|
|
|
check_cache_all_separate()
|
2015-05-08 19:06:47 +03:00
|
|
|
{
|
|
|
|
|
# Check cache.direct
|
|
|
|
|
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "cache.direct=on on blk\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile" -drive "$drv_bk" -drive "$drv_file" -drive "$drv_img",cache.direct=on | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.direct=on on file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile" -drive "$drv_bk" -drive "$drv_file",cache.direct=on -drive "$drv_img" | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.direct=on on backing\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile" -drive "$drv_bk",cache.direct=on -drive "$drv_file" -drive "$drv_img" | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.direct=on on backing-file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile",cache.direct=on -drive "$drv_bk" -drive "$drv_file" -drive "$drv_img" | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
2015-05-08 19:06:47 +03:00
|
|
|
|
|
|
|
|
# Check cache.writeback
|
|
|
|
|
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\n\ncache.writeback=off on blk\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile" -drive "$drv_bk" -drive "$drv_file" -drive "$drv_img",cache.writeback=off | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.writeback=off on file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile" -drive "$drv_bk" -drive "$drv_file",cache.writeback=off -drive "$drv_img" | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.writeback=off on backing\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile" -drive "$drv_bk",cache.writeback=off -drive "$drv_file" -drive "$drv_img" | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.writeback=off on backing-file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile",cache.writeback=off -drive "$drv_bk" -drive "$drv_file" -drive "$drv_img" | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
2015-05-08 19:06:47 +03:00
|
|
|
|
|
|
|
|
# Check cache.no-flush
|
|
|
|
|
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\n\ncache.no-flush=on on blk\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile" -drive "$drv_bk" -drive "$drv_file" -drive "$drv_img",cache.no-flush=on | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.no-flush=on on file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile" -drive "$drv_bk" -drive "$drv_file",cache.no-flush=on -drive "$drv_img" | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.no-flush=on on backing\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile" -drive "$drv_bk",cache.no-flush=on -drive "$drv_file" -drive "$drv_img" | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
tests: Avoid non-portable 'echo -ARG'
POSIX says that backslashes in the arguments to 'echo', as well as
any use of 'echo -n' and 'echo -e', are non-portable; it recommends
people should favor 'printf' instead. This is definitely true where
we do not control which shell is running (such as in makefile snippets
or in documentation examples). But even for scripts where we
require bash (and therefore, where echo does what we want by default),
it is still possible to use 'shopt -s xpg_echo' to change bash's
behavior of echo. And setting a good example never hurts when we are
not sure if a snippet will be copied from a bash-only script to a
general shell script (although I don't change the use of non-portable
\e for ESC when we know the running shell is bash).
Replace 'echo -n "..."' with 'printf %s "..."', and 'echo -e "..."'
with 'printf %b "...\n"', with the optimization that the %s/%b
argument can be omitted if the string being printed is a strict
literal with no '%', '$', or '`' (we could technically also make
this optimization when there are $ or `` substitutions but where
we can prove their results will not be problematic, but proving
that such substitutions are safe makes the patch less trivial
compared to just being consistent).
In the qemu-iotests check script, fix unusual shell quoting
that would result in word-splitting if 'date' outputs a space.
In test 051, take an opportunity to shorten the line.
In test 068, get rid of a pointless second invocation of bash.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-id: 20170703180950.9895-1-eblake@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2017-07-03 21:09:50 +03:00
|
|
|
printf "\ncache.no-flush=on on backing-file\n"
|
2016-03-14 17:46:03 +03:00
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$drv_bkfile",cache.no-flush=on -drive "$drv_bk" -drive "$drv_file" -drive "$drv_img" | grep -e "Cache" -e "[Cc]annot\|[Cc]ould not\|[Cc]an't"
|
2015-05-08 19:06:47 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Configure cache modes on the command line ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# First check the inherited cache mode after opening the image.
|
|
|
|
|
|
|
|
|
|
hmp_cmds="info block image
|
|
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file"
|
|
|
|
|
|
|
|
|
|
check_cache_all_separate
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Cache modes after reopen (live snapshot) ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# Then trigger a reopen and check that the cache modes are still the same.
|
|
|
|
|
|
|
|
|
|
hmp_cmds="snapshot_blkdev -n blk $TEST_IMG.snap $IMGFMT
|
|
|
|
|
info block blk
|
|
|
|
|
info block image
|
|
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file"
|
|
|
|
|
|
|
|
|
|
check_cache_all_separate
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Change cache modes with reopen (qemu-io command, flags) ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# This one actually changes the cache mode with the reopen. For this test, the
|
|
|
|
|
# new cache mode is specified as flags, not as option.
|
|
|
|
|
|
|
|
|
|
hmp_cmds='qemu-io blk "reopen -c none"
|
|
|
|
|
info block image
|
|
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file'
|
|
|
|
|
|
|
|
|
|
check_cache_all_separate
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "=== Reopening children instead of the root ==="
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
files="if=none,file=$TEST_IMG,backing.file.filename=$TEST_IMG.base"
|
|
|
|
|
ids="node-name=image,backing.node-name=backing,backing.file.node-name=backing-file,file.node-name=file"
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Basic reopen ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
hmp_cmds='qemu-io none0 "reopen -o backing.cache.direct=on"
|
2016-03-03 13:37:48 +03:00
|
|
|
info block none0
|
2015-05-08 19:06:47 +03:00
|
|
|
info block image
|
|
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file'
|
|
|
|
|
|
|
|
|
|
check_cache_all
|
|
|
|
|
|
|
|
|
|
echo
|
|
|
|
|
echo "--- Change cache mode after reopening child ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# This checks that children with options explicitly set with reopen don't
|
|
|
|
|
# inherit these options from their parents any more
|
|
|
|
|
|
|
|
|
|
# TODO Implement node-name support for 'qemu-io' HMP command for -c
|
|
|
|
|
# Can use only -o to access child node options for now
|
|
|
|
|
|
2016-03-04 16:28:01 +03:00
|
|
|
hmp_cmds="qemu-io none0 \"reopen -o file.cache.direct=off,file.cache.no-flush=off\"
|
|
|
|
|
qemu-io none0 \"reopen -o backing.file.cache.direct=off,backing.file.cache.no-flush=on\"
|
2015-05-08 19:06:47 +03:00
|
|
|
qemu-io none0 \"reopen -c none\"
|
|
|
|
|
info block image
|
|
|
|
|
info block file
|
|
|
|
|
info block backing
|
|
|
|
|
info block backing-file"
|
|
|
|
|
|
|
|
|
|
echo "$hmp_cmds" | run_qemu -drive "$files","$ids" | grep "Cache"
|
|
|
|
|
|
2021-11-16 13:14:31 +03:00
|
|
|
echo
|
|
|
|
|
echo "--- Alignment after changing O_DIRECT ---"
|
|
|
|
|
echo
|
|
|
|
|
|
|
|
|
|
# Directly test the protocol level: Can unaligned requests succeed even if
|
|
|
|
|
# O_DIRECT was only enabled through a reopen and vice versa?
|
|
|
|
|
|
|
|
|
|
# Ensure image size is a multiple of the sector size (required for O_DIRECT)
|
|
|
|
|
$QEMU_IMG create -f file "$TEST_IMG" 1M | _filter_img_create
|
|
|
|
|
|
|
|
|
|
# And write some data (not strictly necessary, but it feels better to actually
|
|
|
|
|
# have something to be read)
|
|
|
|
|
$QEMU_IO -f file -c 'write 0 4096' "$TEST_IMG" | _filter_qemu_io
|
|
|
|
|
|
|
|
|
|
$QEMU_IO --cache=writeback -f file $TEST_IMG <<EOF | _filter_qemu_io
|
|
|
|
|
read 42 42
|
|
|
|
|
reopen -o cache.direct=on
|
|
|
|
|
read 42 42
|
|
|
|
|
reopen -o cache.direct=off
|
|
|
|
|
read 42 42
|
|
|
|
|
EOF
|
|
|
|
|
$QEMU_IO --cache=none -f file $TEST_IMG <<EOF | _filter_qemu_io
|
|
|
|
|
read 42 42
|
|
|
|
|
reopen -o cache.direct=off
|
|
|
|
|
read 42 42
|
|
|
|
|
reopen -o cache.direct=on
|
|
|
|
|
read 42 42
|
|
|
|
|
EOF
|
|
|
|
|
|
2015-05-08 19:06:47 +03:00
|
|
|
# success, all done
|
|
|
|
|
echo "*** done"
|
|
|
|
|
rm -f $seq.full
|
|
|
|
|
status=0
|