Skip to contentSkip to conversation

API: Allow testing of kernel features in development

RFC PATCH In review Richard Palethorpe via ltp 6 participants 19 messages Dec 21, 2021 – Jun 16, 2022 1 message not in the archive Library code

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw

RFC PATCH 1/1 API: Allow testing of kernel features in development

Add an unstable kernel ABI flag and a runtest file for unstable tests. This means we can add tests which are likely to be broken by changes in the kernel ABI. Without disrupting LTP releases which are required to be stable.

Users who require stability can filter the tests with this flag or not schedule the unstable runtest file(s).

---

How about adding this to the fanotify22 patch?

 include/tst_test.h        | 12 ++++++++++++
 lib/tst_test.c            |  6 ++++++
 runtest/syscalls-unstable |  3 +++
 3 files changed, 21 insertions(+)
 create mode 100644 runtest/syscalls-unstable
include/tst_test.h +12 -0
index 450ddf086..ff31e972e 100644
@@ -174,6 +174,18 @@ struct tst_test {
 	int skip_in_lockdown:1;
 	int skip_in_compat:1;
 
+	/*
+	 * Test is for a feature that has not been added to the stable
+	 * kernel ABI. That is, it's for a feature only available in
+	 * linux-next, an RC or some other development branch.
+	 *
+	 * This string should be set to some text describing the
+	 * kernel branch and version the test was developed
+	 * against. e.g. "5.16 RC2", "linux-next-20211220",
+	 * "net-next".
+	 */
+	const char *const unstable_abi_version;
+
 	/*
 	 * The skip_filesystem is a NULL terminated list of filesystems the
 	 * test does not support. It can also be used to disable whole class of
lib/tst_test.c +6 -0
index 9b51bb5be..babb250d0 100644
@@ -1013,6 +1013,12 @@ static void do_setup(int argc, char *argv[])
 	if (!tst_test)
 		tst_brk(TBROK, "No tests to run");
 
+	if (tst_test->unstable_abi_version) {
+		tst_res(TINFO,
+			"This test was developed against pre-release kernel: %s",
+			tst_test->unstable_abi_version);
+	}
+
 	if (tst_test->tconf_msg)
 		tst_brk(TCONF, "%s", tst_test->tconf_msg);
 
runtest/syscalls-unstable new file +3 -0
new file mode 100644
index 000000000..a87284afe
@@ -0,0 +1,3 @@
+# Tests for kernel features which are not finalized
+
+fanotify22 fanotify22
Signature · 5 lines
2.34.0


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#2 lore raw

Hi Richard,

Quoted 29 lines · Richard Palethorpe via ltp

Add an unstable kernel ABI flag and a runtest file for unstable tests. This means we can add tests which are likely to be broken by changes in the kernel ABI. Without disrupting LTP releases which are required to be stable.

Users who require stability can filter the tests with this flag or not schedule the unstable runtest file(s).

---

How about adding this to the fanotify22 patch?

 include/tst_test.h        | 12 ++++++++++++
 lib/tst_test.c            |  6 ++++++
 runtest/syscalls-unstable |  3 +++
 3 files changed, 21 insertions(+)
 create mode 100644 runtest/syscalls-unstable
include/tst_test.h +1 -0
index 450ddf086..ff31e972e 100644
@@ -174,6 +174,18 @@ struct tst_test {
        int skip_in_lockdown:1;
        int skip_in_compat:1;
 
+       /*

include/tst_test.h struct tst_test { · line 187

Quoted 10 lines
include/tst_test.h +10 -0
+        * Test is for a feature that has not been added to the stable
+        * kernel ABI. That is, it's for a feature only available in
+        * linux-next, an RC or some other development branch.
+        *
+        * This string should be set to some text describing the
+        * kernel branch and version the test was developed
+        * against. e.g. "5.16 RC2", "linux-next-20211220",
+        * "net-next".
+        */
+       const char *const unstable_abi_version;

I'm not sure if this will bring convenience or trouble to LTP.

If a new syscall is in development (or merged in linux-next) but drop/change in mainline-kernel. We have to adjust the test case accordingly.

And after the feature is go into the mainline kernel finally, should we move that from 'syscalls-unstable' to 'syscalls' and remove the tst_test->unstable_abi_version field at the same time?

Btw, why not we just keep the unstable-syscall test case stay in the review phase until it gets merged in mailline-kernel? I guess that wouldn't block anything or make anyone unhappy.

Quoted 37 lines
include/tst_test.h +1 -0
+
        /*
         * The skip_filesystem is a NULL terminated list of filesystems the
         * test does not support. It can also be used to disable whole class of
lib/tst_test.c +6 -0
index 9b51bb5be..babb250d0 100644
@@ -1013,6 +1013,12 @@ static void do_setup(int argc, char *argv[])
        if (!tst_test)
                tst_brk(TBROK, "No tests to run");
 
+       if (tst_test->unstable_abi_version) {
+               tst_res(TINFO,
+                       "This test was developed against pre-release kernel: %s",
+                       tst_test->unstable_abi_version);
+       }
+
        if (tst_test->tconf_msg)
                tst_brk(TCONF, "%s", tst_test->tconf_msg);
 
runtest/syscalls-unstable new file +3 -1
new file mode 100644
index 000000000..a87284afe
@@ -0,0 +1,3 @@
+# Tests for kernel features which are not finalized
+
+fanotify22 fanotify22
--

2.34.0

--
Mailing list info: https://lists.linux.it/listinfo/ltp

Signature · 6 lines
Regards,
Li Wang


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#3 lore raw

Hello Li,

Quoted 46 lines · Li Wang

Hi Richard,

Quoted 29 lines · Richard Palethorpe via ltp

Add an unstable kernel ABI flag and a runtest file for unstable tests. This means we can add tests which are likely to be broken by changes in the kernel ABI. Without disrupting LTP releases which are required to be stable.

Users who require stability can filter the tests with this flag or not schedule the unstable runtest file(s).

---

How about adding this to the fanotify22 patch?

 include/tst_test.h        | 12 ++++++++++++
 lib/tst_test.c            |  6 ++++++
 runtest/syscalls-unstable |  3 +++
 3 files changed, 21 insertions(+)
 create mode 100644 runtest/syscalls-unstable
include/tst_test.h +1 -0
index 450ddf086..ff31e972e 100644
@@ -174,6 +174,18 @@ struct tst_test {
        int skip_in_lockdown:1;
        int skip_in_compat:1;
 
+       /*

include/tst_test.h struct tst_test { · line 187

Quoted 10 lines
include/tst_test.h +10 -0
+        * Test is for a feature that has not been added to the stable
+        * kernel ABI. That is, it's for a feature only available in
+        * linux-next, an RC or some other development branch.
+        *
+        * This string should be set to some text describing the
+        * kernel branch and version the test was developed
+        * against. e.g. "5.16 RC2", "linux-next-20211220",
+        * "net-next".
+        */
+       const char *const unstable_abi_version;

I'm not sure if this will bring convenience or trouble to LTP.

Quoted 3 lines

If a new syscall is in development (or merged in linux-next) but drop/change in mainline-kernel. We have to adjust the test case accordingly.

If someone is motivated enough to contribute tests for an unreleased feature. There is a high chance they will make the changes. In the worst case we just delete the test.

Quoted 4 lines

And after the feature is go into the mainline kernel finally, should we move that from 'syscalls-unstable' to 'syscalls' and remove the tst_test->unstable_abi_version field at the same time?

Yes.

Quoted 4 lines

Btw, why not we just keep the unstable-syscall test case stay in the review phase until it gets merged in mailline-kernel? I guess that wouldn't block anything or make anyone unhappy.

It's a lot of work to merge patches into your own branch for testing RCs and linux-next. It makes collaboration a lot harder. It's easy for people just to enable the unstable tests. Then new features will get tested on a much wider range of setups.

It increases the chances of finding bugs before release.

Signature · 5 lines
Thank you,
Richard.

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
Acked-by on v1 1/1
#4 lore raw
In reply to Richard Palethorpe via ltp

i all,

[ Cc automated-testing and people who might be interested ]

Quoted 7 lines

Add an unstable kernel ABI flag and a runtest file for unstable tests. This means we can add tests which are likely to be broken by changes in the kernel ABI. Without disrupting LTP releases which are required to be stable.

Users who require stability can filter the tests with this flag or not schedule the unstable runtest file(s).

I'm ok for this from a long term perspective, because agree Richie it can help people to run tests on kernel from next tree or mainline rc kernel).

It's not much work to maintain this.

It'd also help people writing tests for fanotify and IMA not having wait several weeks.

Yes, we could add it to fanotify22 [1], but not to ima_conditionals.sh [2], which is shell. But adding support to shell is trivial.

....

Quoted 1 line

+++ b/runtest/syscalls-unstable

How about having name syscalls-next? Although there can be tests which are from some kernel maintainer tree (it does not have to be limited to next tree), unstable can mean "tests which aren't fixed yet and thus buggy".

runtest/syscalls-unstable line 3

Quoted 4 lines
runtest/syscalls-unstable new file +3 -0
@@ -0,0 +1,3 @@
+# Tests for kernel features which are not finalized
+
+fanotify22 fanotify22

Kind regards,
Petr

[1] https://patchwork.ozlabs.org/project/ltp/list/?series=272782
[2] https://patchwork.ozlabs.org/project/ltp/list/?series=265664

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
#5 lore raw
Quoted 24 lines · Petr Vorel

i all,

[ Cc automated-testing and people who might be interested ]

Quoted 7 lines

Add an unstable kernel ABI flag and a runtest file for unstable tests. This means we can add tests which are likely to be broken by changes in the kernel ABI. Without disrupting LTP releases which are required to be stable.

Users who require stability can filter the tests with this flag or not schedule the unstable runtest file(s).

I'm ok for this from a long term perspective, because agree Richie it can help people to run tests on kernel from next tree or mainline rc kernel).

It's not much work to maintain this.

It'd also help people writing tests for fanotify and IMA not having wait several weeks.

Yes, we could add it to fanotify22 [1], but not to ima_conditionals.sh [2], which is shell. But adding support to shell is trivial.

Quoted 7 lines

....

Quoted 1 line

+++ b/runtest/syscalls-unstable

How about having name syscalls-next? Although there can be tests which are from some kernel maintainer tree (it does not have to be limited to next tree), unstable can mean "tests which aren't fixed yet and thus buggy".

staging?

IMO separate runtest would be enough, any notes why and how test was developed could be in comments in code, where people can find it (less metadata to maintain),
and those comments could stay there after feature is accepted to mainline, we just
move test between runtest files.

Quoted 14 lines

runtest/syscalls-unstable line 3

Quoted 4 lines
runtest/syscalls-unstable new file +3 -0
@@ -0,0 +1,3 @@
+# Tests for kernel features which are not finalized
+
+fanotify22 fanotify22

Kind regards,
Petr

[1] https://patchwork.ozlabs.org/project/ltp/list/?series=272782
[2] https://patchwork.ozlabs.org/project/ltp/list/?series=265664

--
Mailing list info: https://lists.linux.it/listinfo/ltp

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
#6 lore raw

Hi Jan, all,

Quoted 7 lines
Quoted 5 lines

....

Quoted 1 line

+++ b/runtest/syscalls-unstable

How about having name syscalls-next? Although there can be tests which are from some kernel maintainer tree (it does not have to be limited to next tree), unstable can mean "tests which aren't fixed yet and thus buggy".

staging?

IMHO better than unstable, just staging is in kernel for not yet ready drivers which need to cleanup, thus it could be confusing similar way as unstable.

But this should be ok, we document the purpose.

Quoted 6 lines

IMO separate runtest would be enough, any notes why and how test was developed could be in comments in code, where people can find it (less metadata to maintain),
and those comments could stay there after feature is accepted to mainline, we just
move test between runtest files.

+1, actually the simplest solution.

Kind regards,
Petr

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
#7 lore raw
In reply to Jan Stancek

Hello,

Quoted 27 lines · Jan Stancek
Quoted 26 lines · Petr Vorel

i all,

[ Cc automated-testing and people who might be interested ]

Quoted 7 lines

Add an unstable kernel ABI flag and a runtest file for unstable tests. This means we can add tests which are likely to be broken by changes in the kernel ABI. Without disrupting LTP releases which are required to be stable.

Users who require stability can filter the tests with this flag or not schedule the unstable runtest file(s).

I'm ok for this from a long term perspective, because agree Richie it can help people to run tests on kernel from next tree or mainline rc kernel).

It's not much work to maintain this.

It'd also help people writing tests for fanotify and IMA not having wait several weeks.

Yes, we could add it to fanotify22 [1], but not to ima_conditionals.sh [2], which is shell. But adding support to shell is trivial.

Quoted 7 lines
Quoted 5 lines

....

Quoted 1 line

+++ b/runtest/syscalls-unstable

How about having name syscalls-next? Although there can be tests which are from some kernel maintainer tree (it does not have to be limited to next tree), unstable can mean "tests which aren't fixed yet and thus buggy".

staging?

Staging and unstable could equally mean the test itself is not fininshed IMO. I didn't suggest next for exactly the reason mentioned, but it might be the better choice.

Quoted 7 lines

IMO separate runtest would be enough, any notes why and how test was developed could be in comments in code, where people can find it (less metadata to maintain),
and those comments could stay there after feature is accepted to mainline, we just
move test between runtest files.

Then the test has a useless or misleading comment saying it was developed against a feature still in development. It's trivial to remove such comments or meta-data. I expect test authors will do it themselves and if they don't we can rethink accepting such tests.

Also the patch uses the meta-data to print a hint. That way we do not need to look at the source code, runtest file and LTP version before deciding on the severity of a problem. Doing extra work upstream saves a lot of work downstream.

Finally note that the plan is to schedule tests without runtest files for parallel execution. That requires meta-data.

Quoted 15 lines
Quoted 15 lines

runtest/syscalls-unstable line 3

Quoted 5 lines
runtest/syscalls-unstable new file +3 -0
@@ -0,0 +1,3 @@
+# Tests for kernel features which are not finalized
+
+fanotify22 fanotify22

Kind regards,
Petr

[1] https://patchwork.ozlabs.org/project/ltp/list/?series=272782
[2] https://patchwork.ozlabs.org/project/ltp/list/?series=265664

--
Mailing list info: https://lists.linux.it/listinfo/ltp

Signature · 5 lines
Thank you,
Richard.

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#8 lore raw
In reply to Richard Palethorpe via ltp

Hi!

Quoted 28 lines

Add an unstable kernel ABI flag and a runtest file for unstable tests. This means we can add tests which are likely to be broken by changes in the kernel ABI. Without disrupting LTP releases which are required to be stable.

Users who require stability can filter the tests with this flag or not schedule the unstable runtest file(s).

---

How about adding this to the fanotify22 patch?

 include/tst_test.h        | 12 ++++++++++++
 lib/tst_test.c            |  6 ++++++
 runtest/syscalls-unstable |  3 +++
 3 files changed, 21 insertions(+)
 create mode 100644 runtest/syscalls-unstable
include/tst_test.h +1 -0
index 450ddf086..ff31e972e 100644
@@ -174,6 +174,18 @@ struct tst_test {
 	int skip_in_lockdown:1;
 	int skip_in_compat:1;
 
+	/*

include/tst_test.h struct tst_test { · line 187

Quoted 10 lines
include/tst_test.h +10 -0
+	 * Test is for a feature that has not been added to the stable
+	 * kernel ABI. That is, it's for a feature only available in
+	 * linux-next, an RC or some other development branch.
+	 *
+	 * This string should be set to some text describing the
+	 * kernel branch and version the test was developed
+	 * against. e.g. "5.16 RC2", "linux-next-20211220",
+	 * "net-next".
+	 */
+	const char *const unstable_abi_version;

I would rather call this 'remove_after_release' or 'remove_after_kernel_release' but that is very minor.

Quoted 26 lines
include/tst_test.h +0 -0
 	/*
 	 * The skip_filesystem is a NULL terminated list of filesystems the
 	 * test does not support. It can also be used to disable whole class of
lib/tst_test.c +6 -0
index 9b51bb5be..babb250d0 100644
@@ -1013,6 +1013,12 @@ static void do_setup(int argc, char *argv[])
 	if (!tst_test)
 		tst_brk(TBROK, "No tests to run");
 
+	if (tst_test->unstable_abi_version) {
+		tst_res(TINFO,
+			"This test was developed against pre-release kernel: %s",
+			tst_test->unstable_abi_version);
+	}
+
 	if (tst_test->tconf_msg)
 		tst_brk(TCONF, "%s", tst_test->tconf_msg);
 
runtest/syscalls-unstable new file +0 -0
new file mode 100644
index 000000000..a87284afe

runtest/syscalls-unstable line 3

Quoted 3 lines
runtest/syscalls-unstable new file +3 -0
@@ -0,0 +1,3 @@
+# Tests for kernel features which are not finalized
+
+fanotify22 fanotify22

I would be even tempted to add this to the default syscalls runtest file but render the test resultless, e.g. replace the tst_res() and tst_brk() handlers in tst_res.c so that it will print only INFO messages.

Signature · 5 lines
Cyril Hrubis
[email protected]

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#9 lore raw

Hi!

Quoted 3 lines

I would be even tempted to add this to the default syscalls runtest file but render the test resultless, e.g. replace the tst_res() and tst_brk() handlers in tst_res.c so that it will print only INFO messages.

And that would also allow us to have env variable switch that would enable all tests for unrelased features, something as LTP_ENABLE_UNRELASED or so. Also with that we can really just TCONF if that variable is not set which would be much simpler.

Signature · 5 lines
Cyril Hrubis
[email protected]

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#10 lore raw

Hello,

Quoted 9 lines · Cyril Hrubis

Hi!

Quoted 3 lines

I would be even tempted to add this to the default syscalls runtest file but render the test resultless, e.g. replace the tst_res() and tst_brk() handlers in tst_res.c so that it will print only INFO messages.

And that would also allow us to have env variable switch that would enable all tests for unrelased features, something as LTP_ENABLE_UNRELASED or so. Also with that we can really just TCONF if that variable is not set which would be much simpler.

I worry this will cause extra work for people reviewing TCONFs. Obviously the meta-data can be used to avoid any issues. However few people have integrated that into their tooling and workflow yet.

Signature · 5 lines
Thank you,
Richard.

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#11 lore raw

Hi all,

Quoted 9 lines

Hello,

Quoted 4 lines · Cyril Hrubis

Hi!

Quoted 3 lines

I would be even tempted to add this to the default syscalls runtest file but render the test resultless, e.g. replace the tst_res() and tst_brk() handlers in tst_res.c so that it will print only INFO messages.

Quoted 8 lines
Quoted 4 lines

And that would also allow us to have env variable switch that would enable all tests for unrelased features, something as LTP_ENABLE_UNRELASED or so. Also with that we can really just TCONF if that variable is not set which would be much simpler.

I worry this will cause extra work for people reviewing TCONFs. Obviously the meta-data can be used to avoid any issues. However few people have integrated that into their tooling and workflow yet.

Can we solve this to have the policy? I don't have preference between having special foo-unstable (or foo-staging) runtest files or TCONF unless LTP_ENABLE_UNRELASED

(If we chose the latter, we could add this info into -h, thus people could run ./foo -h | grep ... to catch this before running; there could be even --json to print metadata in json, which would include this info).

Kind regards,
Petr

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
#12 lore raw

PATCH 1/1 Create policy for testing unstable kernel features

It's not clear if anything more than this is required for now.

---

 doc/test-writing-guidelines.txt | 10 ++++++++++
 runtest/staging                 |  1 +
 2 files changed, 11 insertions(+)
 create mode 100644 runtest/staging
doc/test-writing-guidelines.txt +10 -0
index 1fa751ed5..8ca014d45 100644
@@ -334,3 +334,13 @@ It's easier to maintain a '.gitignore' file per directory with tests, rather
 than having single file in the project root directory. This way, we don't have
 to update all the gitignore files when moving directories, and they get deleted
 automatically when a directory with tests is removed.
+
+7 Testing pre-release kernel features
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Tests for features not yet in a mainline kernel release are accepted. However
+they must only be added to the +staging+ runtest file. Once a feature is part
+of the stable kernel ABI the associated test must be moved out of staging.
+
+This is primarily to help test kernel RCs by avoiding the need to download
+separate LTP patchsets.
runtest/staging new file +1 -0
new file mode 100644
index 000000000..ef1cdea15
@@ -0,0 +1 @@
+# Tests for features that are not yet in the stable kernel ABI
Signature · 5 lines
2.34.1


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
Acked-by on v1 1/1
#13 lore raw

Re: [LTP] [PATCH] Create policy for testing unstable kernel features

Hi Richie,

Thanks!
I'd consider name next, but does not really matter that much.

+1 for single file, because not only syscalls have tests for RC (at least IMA tests also do).

Kind regards,
Petr

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
Acked-by on v1 1/1
#14 lore raw

Re: [LTP] [PATCH] Create policy for testing unstable kernel features

In reply to Richard Palethorpe via ltp
Quoted 4 lines · Richard Palethorpe via ltp

It's not clear if anything more than this is required for now.

Quoted 39 lines

---

 doc/test-writing-guidelines.txt | 10 ++++++++++
 runtest/staging                 |  1 +
 2 files changed, 11 insertions(+)
 create mode 100644 runtest/staging
doc/test-writing-guidelines.txt +10 -0
index 1fa751ed5..8ca014d45 100644
@@ -334,3 +334,13 @@ It's easier to maintain a '.gitignore' file per directory with tests, rather
 than having single file in the project root directory. This way, we don't have
 to update all the gitignore files when moving directories, and they get deleted
 automatically when a directory with tests is removed.
+
+7 Testing pre-release kernel features
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Tests for features not yet in a mainline kernel release are accepted. However
+they must only be added to the +staging+ runtest file. Once a feature is part
+of the stable kernel ABI the associated test must be moved out of staging.
+
+This is primarily to help test kernel RCs by avoiding the need to download
+separate LTP patchsets.
runtest/staging new file +1 -1
new file mode 100644
index 000000000..ef1cdea15
@@ -0,0 +1 @@
+# Tests for features that are not yet in the stable kernel ABI
--

2.34.1

--
Mailing list info: https://lists.linux.it/listinfo/ltp

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
Reviewed-by on v1 1/1
#15 lore raw

Re: [LTP] [PATCH] Create policy for testing unstable kernel features

In reply to Richard Palethorpe via ltp
Quoted 4 lines · Richard Palethorpe

It's not clear if anything more than this is required for now.

Quoted 42 lines

---

 doc/test-writing-guidelines.txt | 10 ++++++++++
 runtest/staging                 |  1 +
 2 files changed, 11 insertions(+)
 create mode 100644 runtest/staging
doc/test-writing-guidelines.txt +0 -0
index 1fa751ed5..8ca014d45 100644
@@ -334,3 +334,13 @@ It's easier to maintain a '.gitignore' file per

directory with tests, rather
than having single file in the project root directory. This way, we don't
have
to update all the gitignore files when moving directories, and they get
deleted
automatically when a directory with tests is removed.

doc/test-writing-guidelines.txt +5 -0
+
+7 Testing pre-release kernel features
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Tests for features not yet in a mainline kernel release are accepted.

However
+they must only be added to the +staging+ runtest file. Once a feature is part

doc/test-writing-guidelines.txt +4 -0
+of the stable kernel ABI the associated test must be moved out of staging.
+
+This is primarily to help test kernel RCs by avoiding the need to download
+separate LTP patchsets.
runtest/staging new file +1 -1
new file mode 100644
index 000000000..ef1cdea15
@@ -0,0 +1 @@
+# Tests for features that are not yet in the stable kernel ABI
--

2.34.1

Signature · 2 lines
Regards,
Li Wang
Signature · 2 lines
-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#16 lore raw

Re: [LTP] [PATCH] Create policy for testing unstable kernel features

Hi all,

...

Quoted 4 lines
Quoted 4 lines

+7 Testing pre-release kernel features
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

(quoted diff) +2 -0
+
+Tests for features not yet in a mainline kernel release are accepted.
Quoted 8 lines

runtest/staging

Quoted 8 lines

However
+they must only be added to the +staging+ runtest file. Once a feature is part

doc/test-writing-guidelines.txt +4 -0
+of the stable kernel ABI the associated test must be moved out of staging.
+
+This is primarily to help test kernel RCs by avoiding the need to download
+separate LTP patchsets.
runtest/staging +0 -0

...

FYI keyctl09 [1] patch is a candidate for merging under this policy.

Cyril wasn't sure [2] that we haven't agreed yet about how this should be handled. Not sure what exactly is he missing.

I guess the description is quite open to allow testing both tests in next tree or even in maintainer tree (before endup in next tree). From my point it should be obvious that not *anything* out of mainline can be added, just tests which sooner or later endup in mainline (and if not, they should be deleted).

Also procedure after release is pretty clear: just move everything out of staging to expected runtest files (mostly syscalls).

=> I'd merge this.

Kind regards,
Petr

[1] https://patchwork.ozlabs.org/project/ltp/patch/[email protected]/
[2] https://lore.kernel.org/ltp/YiC4Pj1sH8UIHY7k@yuki/

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
#17 lore raw

Re: [LTP] [PATCH] Create policy for testing unstable kernel features

Hi all,

as there is another set of fanotify tests for rc kernel, I'd really like to have some policy. Therefore I suggest to stick policy suggested by Richie in this patchset [1], i.e. adding tests into runtest/staging until mainline kernel with this functionality is released (feature is part of the stable kernel ABI). i.e. each kernel release content of this file must be revised and (ideally) moved to particular runtest file where it should belong or fixed (in rare situations when feature changed) or removed (feature was reverted).

I volunteer to maintain runtest/staging.

This feature got already ack from Mike Frysinger, Li Wang, Jan Stancek (and me). I'll wait little longer for ack from others (Yang Xu, Cyril Hrubis).

Anybody against it?

"remove_after_release" solution [2] suggested by Cyril and Richie, which would keep kernel version could be quite easily implemented. But IMHO it's not worth with current number of tests which need it (< 5, often 1 or 0).

Obviously I suggest to drop my original suggestion [2] (it didn't include runtest/staging).

Kind regards,
Petr

[1] https://lore.kernel.org/ltp/[email protected]/
[2] https://lore.kernel.org/ltp/YdW5WEXgrotentzM@yuki/
[3] https://lore.kernel.org/ltp/[email protected]/

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
#18 lore raw

Re: [LTP] [PATCH] Create policy for testing unstable kernel features

Hi!

Quoted 9 lines

as there is another set of fanotify tests for rc kernel, I'd really like to have some policy. Therefore I suggest to stick policy suggested by Richie in this patchset [1], i.e. adding tests into runtest/staging until mainline kernel with this functionality is released (feature is part of the stable kernel ABI). i.e. each kernel release content of this file must be revised and (ideally) moved to particular runtest file where it should belong or fixed (in rare situations when feature changed) or removed (feature was reverted).

I volunteer to maintain runtest/staging.

As long as you commit to maintain the staging you have my ack as well.

Signature · 5 lines
Cyril Hrubis
[email protected]

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#19 lore raw

Re: [LTP] [PATCH] Create policy for testing unstable kernel features

Hi all,

Quoted 9 lines

Hi!

Quoted 7 lines

as there is another set of fanotify tests for rc kernel, I'd really like to have some policy. Therefore I suggest to stick policy suggested by Richie in this patchset [1], i.e. adding tests into runtest/staging until mainline kernel with this functionality is released (feature is part of the stable kernel ABI). i.e. each kernel release content of this file must be revised and (ideally) moved to particular runtest file where it should belong or fixed (in rare situations when feature changed) or removed (feature was reverted).

Quoted 3 lines
Quoted 1 line

I volunteer to maintain runtest/staging.

As long as you commit to maintain the staging you have my ack as well.

Thx! I merged this patch then and I'll merge Amir's fanotify{10,23} tests for 5.19-rc1 kernel.

Kind regards,
Petr

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp