[3/8] arch/x86: Remove sentinel elem from ctl_table arrays

Commit Message

Joel Granados via B4 Relay Sept. 6, 2023, 10:03 a.m. UTC

From: Joel Granados <[e-mail]>

This commit comes at the tail end of a greater effort to remove the
empty elements at the end of the ctl_table arrays (sentinels) which
will reduce the overall build time size of the kernel and run time
memory bloat by ~64 bytes per sentinel (further information Link :
https://lore.kernel.org/all/[e-mail]/)

Remove sentinel element from sld_sysctl and itmt_kern_table.

Signed-off-by: Joel Granados <[e-mail]>
---
 arch/x86/kernel/cpu/intel.c | 3 +--
 arch/x86/kernel/itmt.c      | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

Comments

Dave Hansen Sept. 6, 2023, 2:45 p.m. UTC | #1

On 9/6/23 03:03, Joel Granados via B4 Relay wrote:
> This commit comes at the tail end of a greater effort to remove the
> empty elements at the end of the ctl_table arrays (sentinels) which
> will reduce the overall build time size of the kernel and run time
> memory bloat by ~64 bytes per sentinel (further information Link :
> https://lore.kernel.org/all/[e-mail]/)
> 
> Remove sentinel element from sld_sysctl and itmt_kern_table.

There's a *LOT* of content to read for a reviewer to figure out what's
going on here between all the links.  I would have appreciated one more
sentence here, maybe:

	This is now safe because the sysctl registration code
	(register_sysctl()) implicitly uses ARRAY_SIZE() in addition
	to checking for a sentinel.

That needs to be more prominent _somewhere_.  Maybe here, or maybe in
the cover letter, but _somewhere_.

That said, feel free to add this to the two x86 patches:

Acked-by: Dave Hansen <[e-mail]> # for x86

Ingo Molnar Sept. 6, 2023, 9:58 p.m. UTC | #2

* Dave Hansen <[e-mail]> wrote:

> On 9/6/23 03:03, Joel Granados via B4 Relay wrote:
> > This commit comes at the tail end of a greater effort to remove the
> > empty elements at the end of the ctl_table arrays (sentinels) which
> > will reduce the overall build time size of the kernel and run time
> > memory bloat by ~64 bytes per sentinel (further information Link :
> > https://lore.kernel.org/all/[e-mail]/)
> > 
> > Remove sentinel element from sld_sysctl and itmt_kern_table.
> 
> There's a *LOT* of content to read for a reviewer to figure out what's
> going on here between all the links.  I would have appreciated one more
> sentence here, maybe:
> 
> 	This is now safe because the sysctl registration code
> 	(register_sysctl()) implicitly uses ARRAY_SIZE() in addition
> 	to checking for a sentinel.
> 
> That needs to be more prominent _somewhere_.  Maybe here, or maybe in
> the cover letter, but _somewhere_.
> 
> That said, feel free to add this to the two x86 patches:
> 
> Acked-by: Dave Hansen <[e-mail]> # for x86

Absolutely needs to be in the title as well, something like:

   arch/x86: Remove now superfluous sentinel elem from ctl_table arrays

With that propagated into the whole series:

   Reviewed-by: Ingo Molnar <[e-mail]>

Thanks,

	Ingo

Joel Granados Sept. 7, 2023, 7:38 a.m. UTC | #3

On Wed, Sep 06, 2023 at 07:45:09AM -0700, Dave Hansen wrote:
> On 9/6/23 03:03, Joel Granados via B4 Relay wrote:
> > This commit comes at the tail end of a greater effort to remove the
> > empty elements at the end of the ctl_table arrays (sentinels) which
> > will reduce the overall build time size of the kernel and run time
> > memory bloat by ~64 bytes per sentinel (further information Link :
> > https://lore.kernel.org/all/[e-mail]/)
> > 
> > Remove sentinel element from sld_sysctl and itmt_kern_table.
> 
> There's a *LOT* of content to read for a reviewer to figure out what's
> going on here between all the links.  I would have appreciated one more
> sentence here, maybe:
> 
> 	This is now safe because the sysctl registration code
> 	(register_sysctl()) implicitly uses ARRAY_SIZE() in addition
> 	to checking for a sentinel.
Thx for the feedback. This is a great sentence to add at the end of the
first paragraph instead of the link. I'll add it with a few changes as
there are more than just one register function and the use of ARRAY_SIZE
is implicit most of the time.

  This is now safe because the sysctl registration code
  (register_sysctl() and friends) use the array size in addition to
  checking for a sentinel.

I have changed my cover letter in case I send a V2 and for the other
batches that are coming after the architecture one.

> 
> That needs to be more prominent _somewhere_.  Maybe here, or maybe in
> the cover letter, but _somewhere_.
This is also a good point. I think having it in both the cover letter
and each of the commits is an added value.

> 
> That said, feel free to add this to the two x86 patches:
> 
> Acked-by: Dave Hansen <[e-mail]> # for x86

Best

Joel Granados Sept. 7, 2023, 8:24 a.m. UTC | #4

On Wed, Sep 06, 2023 at 11:58:47PM +0200, Ingo Molnar wrote:
> 
> * Dave Hansen <[e-mail]> wrote:
> 
> > On 9/6/23 03:03, Joel Granados via B4 Relay wrote:
> > > This commit comes at the tail end of a greater effort to remove the
> > > empty elements at the end of the ctl_table arrays (sentinels) which
> > > will reduce the overall build time size of the kernel and run time
> > > memory bloat by ~64 bytes per sentinel (further information Link :
> > > https://lore.kernel.org/all/[e-mail]/)
> > > 
> > > Remove sentinel element from sld_sysctl and itmt_kern_table.
> > 
> > There's a *LOT* of content to read for a reviewer to figure out what's
> > going on here between all the links.  I would have appreciated one more
> > sentence here, maybe:
> > 
> > 	This is now safe because the sysctl registration code
> > 	(register_sysctl()) implicitly uses ARRAY_SIZE() in addition
> > 	to checking for a sentinel.
> > 
> > That needs to be more prominent _somewhere_.  Maybe here, or maybe in
> > the cover letter, but _somewhere_.
> > 
> > That said, feel free to add this to the two x86 patches:
> > 
> > Acked-by: Dave Hansen <[e-mail]> # for x86
> 
> Absolutely needs to be in the title as well, something like:
> 
>    arch/x86: Remove now superfluous sentinel elem from ctl_table arrays
Done. Will wait to see if other ppl have more comments to send out V2

Thx.
> 
> With that propagated into the whole series:
> 
>    Reviewed-by: Ingo Molnar <[e-mail]>
> 
> Thanks,
> 
> 	Ingo

Patch

diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index be4045628fd3..e63391b82624 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -1015,8 +1015,7 @@  static struct ctl_table sld_sysctls[] = {
 		.proc_handler	= proc_douintvec_minmax,
 		.extra1         = SYSCTL_ZERO,
 		.extra2         = SYSCTL_ONE,
-	},
-	{}
+	}
 };
 
 static int __init sld_mitigate_sysctl_init(void)
diff --git a/arch/x86/kernel/itmt.c b/arch/x86/kernel/itmt.c
index ee4fe8cdb857..5f2ccff38297 100644
--- a/arch/x86/kernel/itmt.c
+++ b/arch/x86/kernel/itmt.c
@@ -73,8 +73,7 @@  static struct ctl_table itmt_kern_table[] = {
 		.proc_handler	= sched_itmt_update_handler,
 		.extra1		= SYSCTL_ZERO,
 		.extra2		= SYSCTL_ONE,
-	},
-	{}
+	}
 };
 
 static struct ctl_table_header *itmt_sysctl_header;