public inbox for linux-man@vger.kernel.org
 help / color / mirror / Atom feed
* Issue in man page PR_SET_SECUREBITS.2const
@ 2025-08-24 14:48 Helge Kreutzmann
  2025-08-24 19:21 ` Alejandro Colomar
  0 siblings, 1 reply; 9+ messages in thread
From: Helge Kreutzmann @ 2025-08-24 14:48 UTC (permalink / raw)
  To: alx; +Cc: mario.blaettermann, linux-man

Without further ado, the following was found:

Issue:    malloc → B<malloc>(3) ?

"Setting this flag provides a method for disabling transparent huge pages for "
"jobs where the code cannot be modified, and using a malloc hook with "
"B<madvise>(2)  is not an option (i.e., statically allocated data).  The "
"setting of the \"THP disable\" flag is inherited by a child created via "
"B<fork>(2)  and is preserved across B<execve>(2)."

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Issue in man page PR_SET_SECUREBITS.2const
@ 2025-08-24 14:48 Helge Kreutzmann
  2025-08-24 19:25 ` Alejandro Colomar
  0 siblings, 1 reply; 9+ messages in thread
From: Helge Kreutzmann @ 2025-08-24 14:48 UTC (permalink / raw)
  To: alx; +Cc: mario.blaettermann, linux-man

Without further ado, the following was found:

Issue:    I<op> → I<flags> ?

"I<op> is B<PR_SET_SECUREBITS>, and the caller does not have the "
"B<CAP_SETPCAP> capability, or tried to unset a \"locked\" flag, or tried to "
"set a flag whose corresponding locked flag was set (see B<capabilities>(7))."

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: Issue in man page PR_SET_SECUREBITS.2const
  2025-08-24 14:48 Issue in man page PR_SET_SECUREBITS.2const Helge Kreutzmann
@ 2025-08-24 19:21 ` Alejandro Colomar
  2025-08-25 16:11   ` Helge Kreutzmann
  2025-08-25 17:52   ` Carlos O'Donell
  0 siblings, 2 replies; 9+ messages in thread
From: Alejandro Colomar @ 2025-08-24 19:21 UTC (permalink / raw)
  To: Helge Kreutzmann; +Cc: mario.blaettermann, linux-man

[-- Attachment #1: Type: text/plain, Size: 1389 bytes --]

Hi Helge,

> Subject: Re: Issue in man page PR_SET_SECUREBITS.2const

You probably meant PR_SET_THP_DISABLE.2const.  The text you quoted is
there:

	$ grep -rn THP.disable
	man/man2const/PR_SET_THP_DISABLE.2const:9:set the state of the "THP disable" flag for the calling thread
	man/man2const/PR_SET_THP_DISABLE.2const:21:Set the state of the "THP disable" flag for the calling thread.
	man/man2const/PR_SET_THP_DISABLE.2const:32:The setting of the "THP disable" flag is inherited by a child created via
	man/man2const/PR_GET_THP_DISABLE.2const:9:get the state of the "THP disable" flag for the calling thread
	man/man2const/PR_GET_THP_DISABLE.2const:22:the "THP disable" flag for the calling thread:

On Sun, Aug 24, 2025 at 02:48:48PM +0000, Helge Kreutzmann wrote:
> Without further ado, the following was found:
> 
> Issue:    malloc → B<malloc>(3) ?

Maybe, but we should probably add a reference to malloc_hook(3) instead?


Cheers,
Alex

> 
> "Setting this flag provides a method for disabling transparent huge pages for "
> "jobs where the code cannot be modified, and using a malloc hook with "
> "B<madvise>(2)  is not an option (i.e., statically allocated data).  The "
> "setting of the \"THP disable\" flag is inherited by a child created via "
> "B<fork>(2)  and is preserved across B<execve>(2)."

-- 
<https://www.alejandro-colomar.es/>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: Issue in man page PR_SET_SECUREBITS.2const
  2025-08-24 14:48 Helge Kreutzmann
@ 2025-08-24 19:25 ` Alejandro Colomar
  0 siblings, 0 replies; 9+ messages in thread
From: Alejandro Colomar @ 2025-08-24 19:25 UTC (permalink / raw)
  To: Helge Kreutzmann; +Cc: mario.blaettermann, linux-man

[-- Attachment #1: Type: text/plain, Size: 1494 bytes --]

Hi Helge,

On Sun, Aug 24, 2025 at 02:48:48PM +0000, Helge Kreutzmann wrote:
> Without further ado, the following was found:
> 
> Issue:    I<op> → I<flags> ?
> 
> "I<op> is B<PR_SET_SECUREBITS>, and the caller does not have the "
> "B<CAP_SETPCAP> capability, or tried to unset a \"locked\" flag, or tried to "
> "set a flag whose corresponding locked flag was set (see B<capabilities>(7))."

Hmmm, this is cruft that I forgot to update after splitting the page.
I've fixed it with this:

	commit c71d7a2e31c0689d942bc3b030284144bd331f04 (HEAD -> contrib)
	Author: Alejandro Colomar <alx@kernel.org>
	Date:   Sun Aug 24 21:23:39 2025 +0200

	    man/man2const/PR_SET_SECUREBITS.2const: wfix
	    
	    Fixes: 2dcad3cde34e (2024-05-31; "PR_SET_SECUREBITS.2const: Tweak after split")
	    Reported-by: Helge Kreutzmann <debian@helgefjell.de>
	    Signed-off-by: Alejandro Colomar <alx@kernel.org>

	diff --git a/man/man2const/PR_SET_SECUREBITS.2const b/man/man2const/PR_SET_SECUREBITS.2const
	index ae7268d7d..c80ce0f2e 100644
	--- a/man/man2const/PR_SET_SECUREBITS.2const
	+++ b/man/man2const/PR_SET_SECUREBITS.2const
	@@ -35,10 +35,7 @@ .SH ERRORS
	 is not a valid value.
	 .TP
	 .B EPERM
	-.I op
	-is
	-.BR PR_SET_SECUREBITS ,
	-and the caller does not have the
	+The caller does not have the
	 .B CAP_SETPCAP
	 capability,
	 or tried to unset a "locked" flag,

Thanks for the report!


Cheers,
Alex

-- 
<https://www.alejandro-colomar.es/>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: Issue in man page PR_SET_SECUREBITS.2const
  2025-08-24 19:21 ` Alejandro Colomar
@ 2025-08-25 16:11   ` Helge Kreutzmann
  2025-08-25 17:52   ` Carlos O'Donell
  1 sibling, 0 replies; 9+ messages in thread
From: Helge Kreutzmann @ 2025-08-25 16:11 UTC (permalink / raw)
  To: Alejandro Colomar; +Cc: mario.blaettermann, linux-man

[-- Attachment #1: Type: text/plain, Size: 2084 bytes --]

Hello Alejandro,
Am Sun, Aug 24, 2025 at 09:21:39PM +0200 schrieb Alejandro Colomar:
> Hi Helge,
> 
> > Subject: Re: Issue in man page PR_SET_SECUREBITS.2const
> 
> You probably meant PR_SET_THP_DISABLE.2const.  The text you quoted is
> there:
> 
> 	$ grep -rn THP.disable
> 	man/man2const/PR_SET_THP_DISABLE.2const:9:set the state of the "THP disable" flag for the calling thread
> 	man/man2const/PR_SET_THP_DISABLE.2const:21:Set the state of the "THP disable" flag for the calling thread.
> 	man/man2const/PR_SET_THP_DISABLE.2const:32:The setting of the "THP disable" flag is inherited by a child created via
> 	man/man2const/PR_GET_THP_DISABLE.2const:9:get the state of the "THP disable" flag for the calling thread
> 	man/man2const/PR_GET_THP_DISABLE.2const:22:the "THP disable" flag for the calling thread:

Yes, sorry.

> On Sun, Aug 24, 2025 at 02:48:48PM +0000, Helge Kreutzmann wrote:
> > Without further ado, the following was found:
> > 
> > Issue:    malloc → B<malloc>(3) ?
> 
> Maybe, but we should probably add a reference to malloc_hook(3) instead?

This I cannot judge on, I just see the reference to "malloc" in the
text and hence the question, if this should be turned into a
reference. If malloc_hook(3) is better, than of coures it should be
used.

> Cheers,
> Alex
> 
> > 
> > "Setting this flag provides a method for disabling transparent huge pages for "
> > "jobs where the code cannot be modified, and using a malloc hook with "
> > "B<madvise>(2)  is not an option (i.e., statically allocated data).  The "
> > "setting of the \"THP disable\" flag is inherited by a child created via "
> > "B<fork>(2)  and is preserved across B<execve>(2)."
> 
> -- 
> <https://www.alejandro-colomar.es/>

Greetings

          Helge

-- 
      Dr. Helge Kreutzmann                     debian@helgefjell.de
           Dipl.-Phys.                   http://www.helgefjell.de/debian.php
        64bit GNU powered                     gpg signed mail preferred
           Help keep free software "libre": http://www.ffii.de/

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: Issue in man page PR_SET_SECUREBITS.2const
  2025-08-24 19:21 ` Alejandro Colomar
  2025-08-25 16:11   ` Helge Kreutzmann
@ 2025-08-25 17:52   ` Carlos O'Donell
  2025-08-25 19:41     ` Alejandro Colomar
  1 sibling, 1 reply; 9+ messages in thread
From: Carlos O'Donell @ 2025-08-25 17:52 UTC (permalink / raw)
  To: Alejandro Colomar, Helge Kreutzmann; +Cc: mario.blaettermann, linux-man

On 8/24/25 3:21 PM, Alejandro Colomar wrote:
> Hi Helge,
> 
>> Subject: Re: Issue in man page PR_SET_SECUREBITS.2const
> 
> You probably meant PR_SET_THP_DISABLE.2const.  The text you quoted is
> there:
> 
> 	$ grep -rn THP.disable
> 	man/man2const/PR_SET_THP_DISABLE.2const:9:set the state of the "THP disable" flag for the calling thread
> 	man/man2const/PR_SET_THP_DISABLE.2const:21:Set the state of the "THP disable" flag for the calling thread.
> 	man/man2const/PR_SET_THP_DISABLE.2const:32:The setting of the "THP disable" flag is inherited by a child created via
> 	man/man2const/PR_GET_THP_DISABLE.2const:9:get the state of the "THP disable" flag for the calling thread
> 	man/man2const/PR_GET_THP_DISABLE.2const:22:the "THP disable" flag for the calling thread:
> 
> On Sun, Aug 24, 2025 at 02:48:48PM +0000, Helge Kreutzmann wrote:
>> Without further ado, the following was found:
>>
>> Issue:    malloc → B<malloc>(3) ?
> 
> Maybe, but we should probably add a reference to malloc_hook(3) instead?

In a modern system you care about malloc not malloc_hook.

The hooks have been removed and glibc has directly integrated THP usage in malloc instead.

https://sourceware.org/glibc/manual/2.42/html_node/Memory-Allocation-Tunables.html

Tunable: glibc.malloc.hugetlb

-- 
Cheers,
Carlos.


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: Issue in man page PR_SET_SECUREBITS.2const
  2025-08-25 17:52   ` Carlos O'Donell
@ 2025-08-25 19:41     ` Alejandro Colomar
  2025-08-25 20:00       ` Carlos O'Donell
  0 siblings, 1 reply; 9+ messages in thread
From: Alejandro Colomar @ 2025-08-25 19:41 UTC (permalink / raw)
  To: Carlos O'Donell; +Cc: Helge Kreutzmann, mario.blaettermann, linux-man

[-- Attachment #1: Type: text/plain, Size: 1673 bytes --]

Hi Carlos,

On Mon, Aug 25, 2025 at 01:52:03PM -0400, Carlos O'Donell wrote:
> On 8/24/25 3:21 PM, Alejandro Colomar wrote:
> > Hi Helge,
> > 
> > > Subject: Re: Issue in man page PR_SET_SECUREBITS.2const
> > 
> > You probably meant PR_SET_THP_DISABLE.2const.  The text you quoted is
> > there:
> > 
> > 	$ grep -rn THP.disable
> > 	man/man2const/PR_SET_THP_DISABLE.2const:9:set the state of the "THP disable" flag for the calling thread
> > 	man/man2const/PR_SET_THP_DISABLE.2const:21:Set the state of the "THP disable" flag for the calling thread.
> > 	man/man2const/PR_SET_THP_DISABLE.2const:32:The setting of the "THP disable" flag is inherited by a child created via
> > 	man/man2const/PR_GET_THP_DISABLE.2const:9:get the state of the "THP disable" flag for the calling thread
> > 	man/man2const/PR_GET_THP_DISABLE.2const:22:the "THP disable" flag for the calling thread:
> > 
> > On Sun, Aug 24, 2025 at 02:48:48PM +0000, Helge Kreutzmann wrote:
> > > Without further ado, the following was found:
> > > 
> > > Issue:    malloc → B<malloc>(3) ?
> > 
> > Maybe, but we should probably add a reference to malloc_hook(3) instead?
> 
> In a modern system you care about malloc not malloc_hook.
> 
> The hooks have been removed and glibc has directly integrated THP usage in malloc instead.
> 
> https://sourceware.org/glibc/manual/2.42/html_node/Memory-Allocation-Tunables.html

Thanks!  Do we need a more thorough rewrite of this page is needed?
Or just replace malloc by malloc(3)?


Have a lovely night!
Alex

> 
> Tunable: glibc.malloc.hugetlb
> 
> -- 
> Cheers,
> Carlos.
> 

-- 
<https://www.alejandro-colomar.es/>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: Issue in man page PR_SET_SECUREBITS.2const
  2025-08-25 19:41     ` Alejandro Colomar
@ 2025-08-25 20:00       ` Carlos O'Donell
  2025-08-31  9:19         ` Alejandro Colomar
  0 siblings, 1 reply; 9+ messages in thread
From: Carlos O'Donell @ 2025-08-25 20:00 UTC (permalink / raw)
  To: Alejandro Colomar; +Cc: Helge Kreutzmann, mario.blaettermann, linux-man

On 8/25/25 3:41 PM, Alejandro Colomar wrote:
> Hi Carlos,
> 
> On Mon, Aug 25, 2025 at 01:52:03PM -0400, Carlos O'Donell wrote:
>> On 8/24/25 3:21 PM, Alejandro Colomar wrote:
>>> Hi Helge,
>>>
>>>> Subject: Re: Issue in man page PR_SET_SECUREBITS.2const
>>>
>>> You probably meant PR_SET_THP_DISABLE.2const.  The text you quoted is
>>> there:
>>>
>>> 	$ grep -rn THP.disable
>>> 	man/man2const/PR_SET_THP_DISABLE.2const:9:set the state of the "THP disable" flag for the calling thread
>>> 	man/man2const/PR_SET_THP_DISABLE.2const:21:Set the state of the "THP disable" flag for the calling thread.
>>> 	man/man2const/PR_SET_THP_DISABLE.2const:32:The setting of the "THP disable" flag is inherited by a child created via
>>> 	man/man2const/PR_GET_THP_DISABLE.2const:9:get the state of the "THP disable" flag for the calling thread
>>> 	man/man2const/PR_GET_THP_DISABLE.2const:22:the "THP disable" flag for the calling thread:
>>>
>>> On Sun, Aug 24, 2025 at 02:48:48PM +0000, Helge Kreutzmann wrote:
>>>> Without further ado, the following was found:
>>>>
>>>> Issue:    malloc → B<malloc>(3) ?
>>>
>>> Maybe, but we should probably add a reference to malloc_hook(3) instead?
>>
>> In a modern system you care about malloc not malloc_hook.
>>
>> The hooks have been removed and glibc has directly integrated THP usage in malloc instead.
>>
>> https://sourceware.org/glibc/manual/2.42/html_node/Memory-Allocation-Tunables.html
> 
> Thanks!  Do we need a more thorough rewrite of this page is needed?
> Or just replace malloc by malloc(3)?

Yes, the existing text is only correct for < 2.34 glibc system.

A > 2.34 system has THP support with the tunable to support THP (for legacy libhugetlbfs users).

And a glibc == 2.34 system lacks this feature and doesn't support such hook + interposer deployments.

The quickest fix is to reference malloc(3) for now.

-- 
Cheers,
Carlos.


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: Issue in man page PR_SET_SECUREBITS.2const
  2025-08-25 20:00       ` Carlos O'Donell
@ 2025-08-31  9:19         ` Alejandro Colomar
  0 siblings, 0 replies; 9+ messages in thread
From: Alejandro Colomar @ 2025-08-31  9:19 UTC (permalink / raw)
  To: Carlos O'Donell; +Cc: Helge Kreutzmann, mario.blaettermann, linux-man

[-- Attachment #1: Type: text/plain, Size: 2426 bytes --]

Hi Carlos, Helge,

On Mon, Aug 25, 2025 at 04:00:58PM -0400, Carlos O'Donell wrote:
> On 8/25/25 3:41 PM, Alejandro Colomar wrote:
> > Hi Carlos,
> > 
> > On Mon, Aug 25, 2025 at 01:52:03PM -0400, Carlos O'Donell wrote:
> > > On 8/24/25 3:21 PM, Alejandro Colomar wrote:
> > > > Hi Helge,
> > > > 
> > > > > Subject: Re: Issue in man page PR_SET_SECUREBITS.2const
> > > > 
> > > > You probably meant PR_SET_THP_DISABLE.2const.  The text you quoted is
> > > > there:
> > > > 
> > > > 	$ grep -rn THP.disable
> > > > 	man/man2const/PR_SET_THP_DISABLE.2const:9:set the state of the "THP disable" flag for the calling thread
> > > > 	man/man2const/PR_SET_THP_DISABLE.2const:21:Set the state of the "THP disable" flag for the calling thread.
> > > > 	man/man2const/PR_SET_THP_DISABLE.2const:32:The setting of the "THP disable" flag is inherited by a child created via
> > > > 	man/man2const/PR_GET_THP_DISABLE.2const:9:get the state of the "THP disable" flag for the calling thread
> > > > 	man/man2const/PR_GET_THP_DISABLE.2const:22:the "THP disable" flag for the calling thread:
> > > > 
> > > > On Sun, Aug 24, 2025 at 02:48:48PM +0000, Helge Kreutzmann wrote:
> > > > > Without further ado, the following was found:
> > > > > 
> > > > > Issue:    malloc → B<malloc>(3) ?
> > > > 
> > > > Maybe, but we should probably add a reference to malloc_hook(3) instead?
> > > 
> > > In a modern system you care about malloc not malloc_hook.
> > > 
> > > The hooks have been removed and glibc has directly integrated THP usage in malloc instead.
> > > 
> > > https://sourceware.org/glibc/manual/2.42/html_node/Memory-Allocation-Tunables.html
> > 
> > Thanks!  Do we need a more thorough rewrite of this page is needed?
> > Or just replace malloc by malloc(3)?
> 
> Yes, the existing text is only correct for < 2.34 glibc system.
> 
> A > 2.34 system has THP support with the tunable to support THP (for legacy libhugetlbfs users).
> 
> And a glibc == 2.34 system lacks this feature and doesn't support such hook + interposer deployments.
> 
> The quickest fix is to reference malloc(3) for now.

Thanks!  I've applied that fix.
<https://www.alejandro-colomar.es/src/alx/linux/man-pages/man-pages.git/commit/?h=contrib&id=4de21772d70d210ff9085c65ed380b0463d47a49>
(use port 80.)


Have a lovely day!
Alex

-- 
<https://www.alejandro-colomar.es>
Use port 80 (that is, <...:80/>).

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2025-08-31  9:19 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-08-24 14:48 Issue in man page PR_SET_SECUREBITS.2const Helge Kreutzmann
2025-08-24 19:21 ` Alejandro Colomar
2025-08-25 16:11   ` Helge Kreutzmann
2025-08-25 17:52   ` Carlos O'Donell
2025-08-25 19:41     ` Alejandro Colomar
2025-08-25 20:00       ` Carlos O'Donell
2025-08-31  9:19         ` Alejandro Colomar
  -- strict thread matches above, loose matches on Subject: below --
2025-08-24 14:48 Helge Kreutzmann
2025-08-24 19:25 ` Alejandro Colomar

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox