linux-next: manual merge of the sysctl tree with the mm-nonmm-unstable tree

PATCH Mark Brown 4 participants 9 messages Jul 28 – Aug 7 Linux-next tree

lore

Share & tools

lore

Mark Brown

author maintainer

2026-07-28 17:39 UTC #0 lore raw

Hi all,

Today's linux-next merge of the sysctl tree got a conflict in:

kernel/time/jiffies.c

between commit:

842912dea6d51 ("sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL")

from the mm-nonmm-unstable tree and commit:

a3c70a5a9f944 ("sysctl: Generate do_proc_doulongvec_minmax with do_proc_dotypevec macro")

from the sysctl tree.

I fixed it up (see below) and can carry the fix as necessary. This is now fixed as far as linux-next is concerned, but any non trivial conflicts should be mentioned to your upstream maintainer when your tree is submitted for merging. You may also want to consider cooperating with the maintainer of the conflicting tree to minimise any particularly complex conflicts.

kernel/time/jiffies.c +0 -0

diff --combined kernel/time/jiffies.c
index 84c6bd8e742e9,af529d558fa92..0000000000000
@@@ -98,7 -98,7 +98,7 @@@ void __init register_refined_jiffies(lo
  	__clocksource_register(&refined_jiffies);
  }
  
 -#ifdef CONFIG_PROC_SYSCTL
 +#ifdef CONFIG_SYSCTL
  static ulong mult_hz(const ulong val)
  {
  	return val * HZ;
@@@ -185,7 -185,25 +185,7 @@@ static int do_proc_int_conv_ms_jiffies_
  			     sysctl_u2k_int_conv_ms, sysctl_k2u_int_conv_ms);
  }
  
 -static int sysctl_u2k_ulong_conv_ms(const ulong *u_ptr, ulong *k_ptr)
 -{
 -	return proc_ulong_u2k_conv_uop(u_ptr, k_ptr, sysctl_msecs_to_jiffies);
 -}
 -
 -static int sysctl_k2u_ulong_conv_ms(ulong *u_ptr, const ulong *k_ptr)
 -{
 -	return proc_ulong_k2u_conv_kop(u_ptr, k_ptr, sysctl_jiffies_to_msecs);
 -}
 -
 -static int do_proc_ulong_conv_ms_jiffies(bool *negp, ulong *u_ptr, ulong *k_ptr,
 -					 int dir, const struct ctl_table *tbl)
 -{
 -	return proc_ulong_conv(u_ptr, k_ptr, dir, tbl, false,
 -			       sysctl_u2k_ulong_conv_ms, sysctl_k2u_ulong_conv_ms);
 -}
 -
 -
 -#else // CONFIG_PROC_SYSCTL
 +#else // CONFIG_SYSCTL
  static int do_proc_int_conv_jiffies(bool *negp, ulong *u_ptr, int *k_ptr,
  				    int dir, const struct ctl_table *tbl)
  {
@@@ -211,6 -229,12 +211,12 @@@ static int do_proc_int_conv_ms_jiffies_
  {
  	return -ENOSYS;
  }
+ 
+ static int do_proc_ulong_conv_ms_jiffies(bool *negp, ulong *u_ptr, ulong *k_ptr,
+ 					 int dir, const struct ctl_table *tbl)
+ {
+ 	return -ENOSYS;
+ }
  #endif
  
  /**
@@@ -309,8 -333,8 +315,8 @@@ int proc_dointvec_ms_jiffies_minmax(con
  int proc_doulongvec_ms_jiffies_minmax(const struct ctl_table *table, int dir,
  				      void *buffer, size_t *lenp, loff_t *ppos)
  {
- 	return proc_doulongvec_minmax_conv(table, dir, buffer, lenp, ppos,
- 					   HZ, 1000l);
+ 	return proc_doulongvec_conv(table, dir, buffer, lenp, ppos,
+ 				    do_proc_ulong_conv_ms_jiffies);
  }
  EXPORT_SYMBOL(proc_doulongvec_ms_jiffies_minmax);

Andrew Morton

2026-07-28 20:47 UTC #1 lore raw

Quoted 6 lines · On Tue, 28 Jul 2026 18:39:51 +0100 Mark Brown <[email protected]> wrote:

Today's linux-next merge of the sysctl tree got a conflict in:
kernel/time/jiffies.c
between commit:

Quoted 7 lines

842912dea6d51 ("sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL")
from the mm-nonmm-unstable tree and commit:
a3c70a5a9f944 ("sysctl: Generate do_proc_doulongvec_minmax with do_proc_dotypevec macro")
from the sysctl tree.

Luis, can you please grab both

	sysctl: move the "cad_pid" entry from pid_table[] to kern_reboot_table[]
	https://lore.kernel.org/[email protected]

and

sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
https://lore.kernel.org/[email protected]

I'll email them at you...

Joel Granados

2026-08-04 07:36 UTC #2 lore raw

Quoted 10 lines · Andrew Morton

Quoted 8 lines · On Tue, 28 Jul 2026 18:39:51 +0100 Mark Brown <[email protected]> wrote:
Today's linux-next merge of the sysctl tree got a conflict in:
kernel/time/jiffies.c
between commit:
842912dea6d51 ("sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL")

Quoted 7 lines

Quoted 5 lines
from the mm-nonmm-unstable tree and commit:
a3c70a5a9f944 ("sysctl: Generate do_proc_doulongvec_minmax with do_proc_dotypevec macro")
from the sysctl tree.
Luis, can you please grab both

I can put these into sysctl next if needed. Do you still need both of them to be there?

Quoted 3 lines

	sysctl: move the "cad_pid" entry from pid_table[] to kern_reboot_table[]
	https://lore.kernel.org/[email protected]

I do **not** see this one in next-20260803. I'll just add this to the sysctl-next.

Quoted 5 lines

and
sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
https://lore.kernel.org/[email protected]

I see that this one is already in next-20260803. probably went in through mm tree.

23a24c8a95c5 sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL [Oleg Nesterov]

Let me know if you prefer some other solution.

Best

Signature · 1 line

Joel Granados

Joel Granados

2026-08-04 07:54 UTC #3 lore raw

Quoted 18 lines · Joel Granados

Quoted 10 lines · Andrew Morton
Quoted 8 lines · On Tue, 28 Jul 2026 18:39:51 +0100 Mark Brown <[email protected]> wrote:
Today's linux-next merge of the sysctl tree got a conflict in:
kernel/time/jiffies.c
between commit:
842912dea6d51 ("sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL")
Quoted 7 lines
Quoted 5 lines
from the mm-nonmm-unstable tree and commit:
a3c70a5a9f944 ("sysctl: Generate do_proc_doulongvec_minmax with do_proc_dotypevec macro")
from the sysctl tree.
Luis, can you please grab both

Quoted 8 lines

I can put these into sysctl next if needed. Do you still need both of them to be there?
Quoted 3 lines
	sysctl: move the "cad_pid" entry from pid_table[] to kern_reboot_table[]
	https://lore.kernel.org/[email protected]
I do **not** see this one in next-20260803. I'll just add this to the sysctl-next.

This one might cause another conflict in linux next. To solve it just replace CONFIG_PROC_SYSCTL with CONFIG_SYSCTL.

Quoted 10 lines

Quoted 6 lines
and
sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
https://lore.kernel.org/[email protected]
I see that this one is already in next-20260803. probably went in through mm tree.
23a24c8a95c5 sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL [Oleg Nesterov]

To avoid further conflicts in next I can also take this one into sysctl-next and handle every thing in my PR to Linus.

Let me know what you decide.

Best

Signature · 1 line

Joel Granados

Andrew Morton

2026-08-04 20:14 UTC #4 lore raw

Quoted 8 lines · On Tue, 4 Aug 2026 09:54:04 +0200 Joel Granados <[email protected]>…

Quoted 8 lines
Quoted 5 lines
and
sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
https://lore.kernel.org/[email protected]
I see that this one is already in next-20260803. probably went in through mm tree.

Quoted 6 lines

Quoted 1 line
23a24c8a95c5 sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL [Oleg Nesterov]
To avoid further conflicts in next I can also take this one into sysctl-next and handle every thing in my PR to Linus.
Let me know what you decide.

How about I remove

	sysctl: move the "cad_pid" entry from pid_table[] to kern_reboot_table[]
	https://lore.kernel.org/[email protected]

and

sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
https://lore.kernel.org/[email protected]

from mm.git then forget all about them ;)

Joel Granados

2026-08-05 08:06 UTC #5 lore raw

Quoted 22 lines · Andrew Morton

Quoted 8 lines · On Tue, 4 Aug 2026 09:54:04 +0200 Joel Granados <[email protected]>…
Quoted 8 lines
Quoted 5 lines
and
sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
https://lore.kernel.org/[email protected]
I see that this one is already in next-20260803. probably went in through mm tree.
Quoted 6 lines
Quoted 1 line
23a24c8a95c5 sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL [Oleg Nesterov]
To avoid further conflicts in next I can also take this one into sysctl-next and handle every thing in my PR to Linus.
Let me know what you decide.
How about I remove
	sysctl: move the "cad_pid" entry from pid_table[] to kern_reboot_table[]
	https://lore.kernel.org/[email protected]

Quoted 6 lines

and
sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
https://lore.kernel.org/[email protected]
from mm.git then forget all about them ;)

Works for me. I had already picked up the first, I'll add the second to sysctl-next and they will go upstream through the sysctl tree.

Best

Joel

Oleg Nesterov

2026-08-06 18:04 UTC #6 lore raw

Joel, et al,

First of all, sorry for the hassle caused by my patches ;)

Quoted 10 lines · Joel Granados

Quoted 8 lines · Andrew Morton

How about I remove

	sysctl: move the "cad_pid" entry from pid_table[] to kern_reboot_table[]
	https://lore.kernel.org/[email protected]

and

Quoted 7 lines

Quoted 4 lines
sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
https://lore.kernel.org/[email protected]
from mm.git then forget all about them ;)
Works for me. I had already picked up the first, I'll add the second to sysctl-next and they will go upstream through the sysctl tree.

From [PATCH -mm v2 0/1] sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL

If this patch is accepted, I'll send a couple of followup cleanups:

kernel/sysctl.c:
collapse the now-identical CONFIG_SYSCTL nesting

drivers/parport/procfs.c, fs/dcache.c, and fs/file_table.c:
drop the CONFIG_PROC_FS checks, now it is clear that CONFIG_SYSCTL depends on it

Now that I see both patches in
https://git.kernel.org/pub/scm/linux/kernel/git/joel.granados/linux.git/log/?h=jag/sysctl
let me ask..

Can I send these (minor) cleanups based on sysctl-next to you right now (before I forget it ;) ?

Or it would be better to do this later, after they go upstream?

Oleg.

Joel Granados

2026-08-07 09:21 UTC #7 lore raw

Hey Oleg

Quoted 3 lines · Oleg Nesterov

Joel, et al,
First of all, sorry for the hassle caused by my patches ;)

No worries. I think removing one of those configs is the right thing to do.

Quoted 27 lines · Joel Granados

Quoted 10 lines · Andrew Morton
Quoted 8 lines
How about I remove
	sysctl: move the "cad_pid" entry from pid_table[] to kern_reboot_table[]
	https://lore.kernel.org/[email protected]
and
Quoted 7 lines
Quoted 4 lines
sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
https://lore.kernel.org/[email protected]
from mm.git then forget all about them ;)
Works for me. I had already picked up the first, I'll add the second to sysctl-next and they will go upstream through the sysctl tree.
From [PATCH -mm v2 0/1] sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL
If this patch is accepted, I'll send a couple of followup cleanups:
kernel/sysctl.c:
collapse the now-identical CONFIG_SYSCTL nesting

Quoted 7 lines

drivers/parport/procfs.c, fs/dcache.c, and fs/file_table.c:
drop the CONFIG_PROC_FS checks, now it is clear that CONFIG_SYSCTL depends on it
Now that I see both patches in
https://git.kernel.org/pub/scm/linux/kernel/git/joel.granados/linux.git/log/?h=jag/sysctl
let me ask..

Its already in sysctl-next https://git.kernel.org/pub/scm/linux/kernel/git/sysctl/sysctl.git/log/?h=sysctl-next
which means it is in linux-next (have not checked).

Quoted 3 lines

Can I send these (minor) cleanups based on sysctl-next to you right now (before I forget it ;) ?

Send them now based on sysctl-next. My guess is that it is too late to include them in this release. But they will get queued up for the next one.

Best

Joel

Joel Granados

2026-08-04 07:44 UTC #8 lore raw

In reply to Mark Brown

Quoted 12 lines · Mark Brown

Hi all,
Today's linux-next merge of the sysctl tree got a conflict in:
kernel/time/jiffies.c
between commit:
842912dea6d51 ("sysctl: remove CONFIG_PROC_SYSCTL, it just mirrors CONFIG_SYSCTL")
from the mm-nonmm-unstable tree and commit:

Quoted 8 lines

a3c70a5a9f944 ("sysctl: Generate do_proc_doulongvec_minmax with do_proc_dotypevec macro")
from the sysctl tree.
I fixed it up (see below) and can carry the fix as necessary. This is now fixed as far as linux-next is concerned, but any non trivial conflicts should be mentioned to your upstream maintainer when your tree is submitted for merging. You may also want to consider cooperating

I just checked next-20260803 and I see that all the sysctl changes are maintained. So from my side all is well.

Best

Signature · 1 line

Joel Granados