Linux 5.17 API updates #12989

nabijaczleweli · 2022-01-19T20:36:14Z

Motivation and Context

The buildd was failing.

Description

See individual commits.

How Has This Been Tested?

Pre-5.16 local builds, 5.17 on CI.

Types of changes

Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Performance enhancement (non-breaking change which improves efficiency)
Code cleanup (non-breaking change which makes code smaller or more readable)
Breaking change (fix or feature that would cause existing functionality to change)
Library ABI change (libzfs, libzfs_core, libnvpair, libuutil and libzfsbootenv)
Documentation (a change to man pages or other documentation)

Checklist:

My code follows the OpenZFS code style requirements.
I have updated the documentation accordingly.
I have read the contributing document.
I have added tests to cover my changes.
I have run the ZFS Test Suite with this change applied. – CI take my hand
All commit messages are properly formatted and contain Signed-off-by.

nabijaczleweli · 2022-01-19T21:04:09Z

The buildd now also notes this:

fs/zfs/zfs/../os/linux/zfs/zvol_os.c: In function ‘zvol_os_create_minor’:
fs/zfs/zfs/../os/linux/zfs/zvol_os.c:1113:3: warning: ignoring return value of ‘add_disk’, declared with attribute warn_unused_result [-Wunused-result]
   add_disk(zv->zv_zso->zvo_disk);
   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

And, indeed, there are no provisions for device_add_disk() failing, but I personally don't know what they'd, well, be, so.

gamanakis · 2022-01-19T21:06:52Z

Have you seen this #12975 ?
Edit: I think you address different issues here. The add_disk() issue is handled in the above PR.

nabijaczleweli · 2022-01-19T21:09:45Z

I haven't, but neither of these is touched by that PR (add_disk() is, though).

Both of these commits landed in this merge, and the PR is from the 14th:

commit 35ce8ae9ae2e471f92759f9d6880eab42cc1c3b6
Merge: 6661224e66f0 a403df29789b
Author: Linus Torvalds <[email protected]>
Date:   Mon Jan 17 05:49:30 2022 +0200

    Merge branch 'signal-for-v5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace

rincebrain · 2022-01-20T06:24:00Z

I would suspect you mean s/5.16/5.17/, since 5.16 builds without these for me, and the merge commit you cite is labeled "for 5.17"?

nabijaczleweli · 2022-01-20T12:02:46Z

Hm, the describes say v5.16-rc1-2-g5768d8906bc2 and v5.16-rc1-15-gcead18552660.

rincebrain · 2022-01-20T12:21:36Z

Sure, but 5.16 came out on 1/9, and those were pulled on 1/17.

…

On Thu, Jan 20, 2022 at 7:02 AM наб ***@***.***> wrote: Hm, the describes say v5.16-rc1-2-g5768d8906bc2 and v5.16-rc1-15-gcead18552660. — Reply to this email directly, view it on GitHub <#12989 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABUI7KTCWOCS4COPXRXJITUW72XHANCNFSM5MK6DAEA> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you commented.Message ID: ***@***.***>

nabijaczleweli · 2022-01-20T12:21:37Z

And the merge is v5.16-10543-g35ce8ae9ae2e, so fair cop. Relabeled.

behlendorf

Looks good. Nice and straight forward, thanks.

tonyhutter · 2022-01-22T01:16:12Z

config/kernel-kthread.m4

+	dnl # 5.17 API,
+	dnl # cead18552660702a4a46f58e65188fe5f36e9dfe ("exit: Rename complete_and_exit to kthread_complete_and_exit")
+	dnl #
+	dnl # Also moves the definition from i/l/kernel.h to i/l/kthread.h


- dnl # Also moves the definition from i/l/kernel.h to i/l/kthread.h + dnl # Also moves the definition from include/linux/kernel.h to include/linux/kthread.h

tonyhutter · 2022-01-22T01:27:20Z

We typically use "Linux x.y compat:" commit labeling for tracking kernel changes. Would you mind changing your commit messages to read:

Linux 5.17 compat: dequeue_signal() added a 3rd argument

dequeue_signal() now takes a 3rd argument for pid_type.

Upstream commit cead18552660702a4a46f58e65188fe5f36e9dfe
("exit: Rename complete_and_exit to kthread_complete_and_exit")

Signed-off-by: Ahelenia Ziemiańska <[email protected]>

Linux 5.17 compat: detect complete_and_exit() rename

The 5.17 kernel renames complete_and_exit() to
kthread_complete_and_exit().

Upstream commit cead18552660702a4a46f58e65188fe5f36e9dfe
("exit: Rename complete_and_exit to kthread_complete_and_exit")

Signed-off-by: Ahelenia Ziemiańska <[email protected]>

nabijaczleweli · 2022-01-22T01:40:35Z

Rebased, both applied, messages reworded

behlendorf · 2022-01-24T21:08:46Z

@nabijaczleweli this needs a rebase to resolve the minor conflict after merging the 5.16 changes. Then it should be good to go.

nabijaczleweli · 2022-01-24T21:13:18Z

Rebased

config/kernel-kthread.m4

nabijaczleweli · 2022-01-24T22:17:40Z

Done to both

behlendorf · 2022-01-25T00:21:32Z

config/kernel-kthread.m4

+		#include <linux/sched/signal.h>
+	], [
+		struct task_struct *task = NULL;
+		signet_t *mask = NULL;


Suggested change

signet_t *mask = NULL;

sigset_t *mask = NULL;

I made a typo here. I should have mentioned that I hadn't actually compiled this to verify it works. After fixing the typo it correctly detects the kernel interface for me. I also checked the kthread_complete_and_exit() test.

What I'd suggest is adding the other 5.17 fix to this PR when you update it. This should at least confirm via the CI that it compiles for the latest kernel.

nabijaczleweli · 2022-01-25T00:26:56Z

Both done

nabijaczleweli · 2022-01-25T00:47:19Z

Aaand the built-in buildd passes this day.

behlendorf · 2022-01-25T01:15:22Z

Skimming the kernel build-in build log I see we have a couple more warnings which I haven't seen before with the normal builds.

fs/zfs/zfs/txg.o: warning: objtool: txg_thread_exit() falls through to next function txg_do_callbacks()

Seems like the __noreturn added to __thread_exit in this PR may have caused this, and

ld: warning: orphan section `.eh_frame' from `fs/zfs/icp/asm-x86_64/modes/aesni-gcm-x86_64.o' being placed in section `.eh_frame'.
ld: warning: orphan section `.eh_frame' from `fs/zfs/icp/asm-x86_64/modes/ghash-x86_64.o' being placed in section `.eh_frame'.
ld: warning: orphan section `.eh_frame' from `fs/zfs/icp/asm-x86_64/sha2/sha256_impl.o' being placed in section `.eh_frame'.
ld: warning: orphan section `.eh_frame' from `fs/zfs/icp/asm-x86_64/sha2/sha512_impl.o' being placed in section `.eh_frame'.

which are likely a consequence of the recent icp cleanup.

ckane · 2022-01-25T02:45:16Z

I have this code running on my laptop now, seems to be working

AttilaFueloep · 2022-01-25T06:53:06Z

@behlendorf I'm to blame for the ld warnings and I've seen them way before @nabijaczleweli changes. They are due to the cfi instruction added to the assembler files and me failing to figure out how to add linker scripts to the build (or locate existing ones). I'd say they are harmless since the .eh_frame is the right place for them.

Linux 5.17 sees a rename from complete_and_exit() to kthread complete_and_exit() Upstream commit cead18552660702a4a46f58e65188fe5f36e9dfe ("exit: Rename complete_and_exit to kthread_complete_and_exit") Ref: http://build.zfsonlinux.org/builders/Kernel.org%20Built-in%20x86_64%20%28BUILD%29/builds/43905/steps/shell_1/logs/make Signed-off-by: Ahelenia Ziemiańska <[email protected]>

Linux 5.17's dequeue_signal() takes an additional enum pid_type * output argument Upstream commit 5768d8906bc23d512b1a736c1e198aa833a6daa4 ("signal: Requeue signals in the appropriate queue") Ref: http://build.zfsonlinux.org/builders/Kernel.org%20Built-in%20x86_64%20%28BUILD%29/builds/43905/steps/shell_1/logs/make Signed-off-by: Ahelenia Ziemiańska <[email protected]>

Upstream commit 359745d78351c6f5442435f81549f0207ece28aa ("proc: remove PDE_DATA() completely") Link: https://lore.kernel.org/all/[email protected]/T/#u Closes: openzfs#13004 Signed-off-by: Ahelenia Ziemiańska <[email protected]>

nabijaczleweli · 2022-01-25T12:38:19Z

Yeah, dunno why, especially seeing as both KTHREAD_COMPLETE_AND_EXITs are declared __noreturn. Removed the attributes.

behlendorf · 2022-01-25T20:27:29Z

@AttilaFueloep got it, thanks. I just hadn't noticed them until now.

Linux 5.17's dequeue_signal() takes an additional enum pid_type * output argument Upstream commit 5768d8906bc23d512b1a736c1e198aa833a6daa4 ("signal: Requeue signals in the appropriate queue") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #12989

Upstream commit 359745d78351c6f5442435f81549f0207ece28aa ("proc: remove PDE_DATA() completely") Link: https://lore.kernel.org/all/[email protected]/T/#u Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13004 Closes #12989

Linux 5.17 sees a rename from complete_and_exit() to kthread complete_and_exit() Upstream commit cead18552660702a4a46f58e65188fe5f36e9dfe ("exit: Rename complete_and_exit to kthread_complete_and_exit") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #12989

Linux 5.17's dequeue_signal() takes an additional enum pid_type * output argument Upstream commit 5768d8906bc23d512b1a736c1e198aa833a6daa4 ("signal: Requeue signals in the appropriate queue") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #12989

Upstream commit 359745d78351c6f5442435f81549f0207ece28aa ("proc: remove PDE_DATA() completely") Link: https://lore.kernel.org/all/[email protected]/T/#u Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13004 Closes #12989

Linux 5.17 sees a rename from complete_and_exit() to kthread complete_and_exit() Upstream commit cead18552660702a4a46f58e65188fe5f36e9dfe ("exit: Rename complete_and_exit to kthread_complete_and_exit") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes openzfs#12989

Linux 5.17's dequeue_signal() takes an additional enum pid_type * output argument Upstream commit 5768d8906bc23d512b1a736c1e198aa833a6daa4 ("signal: Requeue signals in the appropriate queue") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes openzfs#12989

Upstream commit 359745d78351c6f5442435f81549f0207ece28aa ("proc: remove PDE_DATA() completely") Link: https://lore.kernel.org/all/[email protected]/T/#u Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes openzfs#13004 Closes openzfs#12989

Linux 5.17 sees a rename from complete_and_exit() to kthread complete_and_exit() Upstream commit cead18552660702a4a46f58e65188fe5f36e9dfe ("exit: Rename complete_and_exit to kthread_complete_and_exit") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes openzfs#12989

Linux 5.17's dequeue_signal() takes an additional enum pid_type * output argument Upstream commit 5768d8906bc23d512b1a736c1e198aa833a6daa4 ("signal: Requeue signals in the appropriate queue") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes openzfs#12989

Upstream commit 359745d78351c6f5442435f81549f0207ece28aa ("proc: remove PDE_DATA() completely") Link: https://lore.kernel.org/all/[email protected]/T/#u Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes openzfs#13004 Closes openzfs#12989

nabijaczleweli force-pushed the new-linux branch 3 times, most recently from 1aea2e4 to 9059895 Compare January 19, 2022 20:47

nabijaczleweli force-pushed the new-linux branch from 9059895 to bd6fbc5 Compare January 19, 2022 21:06

nabijaczleweli force-pushed the new-linux branch from bd6fbc5 to d10cdb9 Compare January 20, 2022 12:21

nabijaczleweli changed the title ~~Linux 5.16 API updates~~ Linux 5.17 API updates Jan 20, 2022

behlendorf approved these changes Jan 21, 2022

View reviewed changes

behlendorf added the Status: Accepted Ready to integrate (reviewed, tested) label Jan 21, 2022

behlendorf requested a review from tonyhutter January 21, 2022 18:21

tonyhutter reviewed Jan 22, 2022

View reviewed changes

nabijaczleweli force-pushed the new-linux branch from d10cdb9 to 7044b8f Compare January 22, 2022 01:38

satmandu mentioned this pull request Jan 24, 2022

[5.17] PDE_DATA() replaced by pde_data() #13004

Closed

nabijaczleweli force-pushed the new-linux branch from 7044b8f to 243c265 Compare January 24, 2022 21:13

behlendorf reviewed Jan 24, 2022

View reviewed changes

config/kernel-kthread.m4 Outdated Show resolved Hide resolved

nabijaczleweli force-pushed the new-linux branch from 243c265 to d00e2ad Compare January 24, 2022 22:17

behlendorf requested changes Jan 25, 2022

View reviewed changes

nabijaczleweli force-pushed the new-linux branch from d00e2ad to 92bd508 Compare January 25, 2022 00:25

nabijaczleweli mentioned this pull request Jan 25, 2022

Linux 5.17 compat: PDE_DATA() renamed to pde_data() #13007

Closed

13 tasks

behlendorf approved these changes Jan 25, 2022

View reviewed changes

tonyhutter approved these changes Jan 25, 2022

View reviewed changes

nabijaczleweli force-pushed the new-linux branch from aef6ae7 to da94f00 Compare January 25, 2022 12:17

nabijaczleweli added 3 commits January 25, 2022 13:36

Linux 5.17 compat: PDE_DATA() renamed to pde_data()

c76d51e

Upstream commit 359745d78351c6f5442435f81549f0207ece28aa ("proc: remove PDE_DATA() completely") Link: https://lore.kernel.org/all/[email protected]/T/#u Closes: openzfs#13004 Signed-off-by: Ahelenia Ziemiańska <[email protected]>

nabijaczleweli force-pushed the new-linux branch from da94f00 to c76d51e Compare January 25, 2022 12:37

behlendorf closed this in a985657 Jan 25, 2022

fabianishere mentioned this pull request Feb 14, 2022

Support for kernel 5.17? fabianishere/pve-edge-kernel#241

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Linux 5.17 API updates #12989

Linux 5.17 API updates #12989

nabijaczleweli commented Jan 19, 2022 •

edited

Loading

nabijaczleweli commented Jan 19, 2022 •

edited

Loading

gamanakis commented Jan 19, 2022 •

edited

Loading

nabijaczleweli commented Jan 19, 2022 •

edited

Loading

rincebrain commented Jan 20, 2022

nabijaczleweli commented Jan 20, 2022

rincebrain commented Jan 20, 2022 via email

nabijaczleweli commented Jan 20, 2022 •

edited

Loading

behlendorf left a comment

tonyhutter Jan 22, 2022

tonyhutter commented Jan 22, 2022 •

edited

Loading

nabijaczleweli commented Jan 22, 2022

behlendorf commented Jan 24, 2022

nabijaczleweli commented Jan 24, 2022

nabijaczleweli commented Jan 24, 2022

behlendorf Jan 25, 2022

nabijaczleweli commented Jan 25, 2022

nabijaczleweli commented Jan 25, 2022

behlendorf commented Jan 25, 2022

ckane commented Jan 25, 2022

AttilaFueloep commented Jan 25, 2022

nabijaczleweli commented Jan 25, 2022

behlendorf commented Jan 25, 2022

Linux 5.17 API updates #12989

Linux 5.17 API updates #12989

Conversation

nabijaczleweli commented Jan 19, 2022 • edited Loading

Motivation and Context

Description

How Has This Been Tested?

Types of changes

Checklist:

nabijaczleweli commented Jan 19, 2022 • edited Loading

gamanakis commented Jan 19, 2022 • edited Loading

nabijaczleweli commented Jan 19, 2022 • edited Loading

rincebrain commented Jan 20, 2022

nabijaczleweli commented Jan 20, 2022

rincebrain commented Jan 20, 2022 via email

nabijaczleweli commented Jan 20, 2022 • edited Loading

behlendorf left a comment

Choose a reason for hiding this comment

tonyhutter Jan 22, 2022

Choose a reason for hiding this comment

tonyhutter commented Jan 22, 2022 • edited Loading

nabijaczleweli commented Jan 22, 2022

behlendorf commented Jan 24, 2022

nabijaczleweli commented Jan 24, 2022

nabijaczleweli commented Jan 24, 2022

behlendorf Jan 25, 2022

Choose a reason for hiding this comment

nabijaczleweli commented Jan 25, 2022

nabijaczleweli commented Jan 25, 2022

behlendorf commented Jan 25, 2022

ckane commented Jan 25, 2022

AttilaFueloep commented Jan 25, 2022

nabijaczleweli commented Jan 25, 2022

behlendorf commented Jan 25, 2022

nabijaczleweli commented Jan 19, 2022 •

edited

Loading

nabijaczleweli commented Jan 19, 2022 •

edited

Loading

gamanakis commented Jan 19, 2022 •

edited

Loading

nabijaczleweli commented Jan 19, 2022 •

edited

Loading

nabijaczleweli commented Jan 20, 2022 •

edited

Loading

tonyhutter commented Jan 22, 2022 •

edited

Loading