[PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Siddhesh Poyarekar]

The mallopt parameters manual does not mention the environment
variables that can be used to set these parameters at program startup.
Mention those environment variables for completeness.

	* manual/memory.texi: Add environment variable alternatives to
	setting mallopt parameters.
---
 manual/memory.texi | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/manual/memory.texi b/manual/memory.texi
index 5383105..222f126 100644
--- a/manual/memory.texi
+++ b/manual/memory.texi
@@ -1113,12 +1113,18 @@ choices for @var{param}, as defined in @file{malloc.h}, are:
 @item M_MMAP_MAX
 The maximum number of chunks to allocate with @code{mmap}.  Setting this
 to zero disables all use of @code{mmap}.
+
+This parameter can also be set for the process at startup by setting the
+environment variable @code{MALLOC_MMAP_MAX_} to the desired value.
 @item M_MMAP_THRESHOLD
 All chunks larger than this value are allocated outside the normal
 heap, using the @code{mmap} system call.  This way it is guaranteed
 that the memory for these chunks can be returned to the system on
 @code{free}.  Note that requests smaller than this threshold might still
 be allocated via @code{mmap}.
+
+This parameter can also be set for the process at startup by setting the
+environment variable @code{MALLOC_MMAP_THRESHOLD_} to the desired value.
 @comment TODO: @item M_MXFAST
 @item M_PERTURB
 If non-zero, memory blocks are filled with values depending on some
@@ -1128,16 +1134,25 @@ use of uninitialized or freed heap memory.  Note that this option does not
 guarantee that the freed block will have any specific values.  It only
 guarantees that the content the block had before it was freed will be
 overwritten.
+
+This parameter can also be set for the process at startup by setting the
+environment variable @code{MALLOC_MMAP_PERTURB_} to the desired value.
 @item M_TOP_PAD
 This parameter determines the amount of extra memory to obtain from the
 system when a call to @code{sbrk} is required.  It also specifies the
 number of bytes to retain when shrinking the heap by calling @code{sbrk}
 with a negative argument.  This provides the necessary hysteresis in
 heap size such that excessive amounts of system calls can be avoided.
+
+This parameter can also be set for the process at startup by setting the
+environment variable @code{MALLOC_TOP_PAD_} to the desired value.
 @item M_TRIM_THRESHOLD
 This is the minimum size (in bytes) of the top-most, releasable chunk
 that will cause @code{sbrk} to be called with a negative argument in
 order to return memory to the system.
+
+This parameter can also be set for the process at startup by setting the
+environment variable @code{MALLOC_TRIM_THRESHOLD_} to the desired value.
 @end table
 
 @end deftypefun
-- 
2.7.4

[PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Siddhesh Poyarekar]

The M_ARENA_* mallopt parameters are in wide use in production to
control the number of arenas that a long lived process creates and
hence there is no point in stating that this interface is non-public.
Document this interface and remove the obsolete comment.

	* manual/memory.texi (M_ARENA_TEST): Add documentation.
	(M_ARENA_MAX): Likewise.
	* malloc/malloc.c: Remove obsolete comment.
---
 malloc/malloc.c    |  1 -
 manual/memory.texi | 21 +++++++++++++++++++++
 2 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/malloc/malloc.c b/malloc/malloc.c
index 09e004b..0011a6d 100644
--- a/malloc/malloc.c
+++ b/malloc/malloc.c
@@ -1718,7 +1718,6 @@ static struct malloc_par mp_ =
 };
 
 
-/*  Non public mallopt parameters.  */
 #define M_ARENA_TEST -7
 #define M_ARENA_MAX  -8
 
diff --git a/manual/memory.texi b/manual/memory.texi
index 222f126..b98dcf2 100644
--- a/manual/memory.texi
+++ b/manual/memory.texi
@@ -1153,6 +1153,27 @@ order to return memory to the system.
 
 This parameter can also be set for the process at startup by setting the
 environment variable @code{MALLOC_TRIM_THRESHOLD_} to the desired value.
+
+@item M_ARENA_TEST
+An arena is a memory pool that is allocated to act as a heap alongside the
+system heap that the kernel creates for the process.  This is to provide
+multiple heap structures for multiple threads of a process to reduce contention
+between them.  The limit to the number of such arenas per process is determined
+by the number of cores of the system and whether it is a 32-bit or a 64-bit
+processor.
+
+The value is ignored if @code{M_ARENA_MAX} is set either via @code{mallopt} or
+via the @code{MALLOC_ARENA_MAX} environment variable.
+
+This parameter can also be set for the process at startup by setting the
+environment variable @code{MALLOC_ARENA_TEST} to the desired value.
+
+@item M_ARENA_MAX
+This parameter sets the number of arenas to use regardless of the number of
+cores in the system.
+
+This parameter can also be set for the process at startup by setting the
+environment variable @code{MALLOC_ARENA_MAX} to the desired value.
 @end table
 
 @end deftypefun
-- 
2.7.4

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[DJ Delorie]

These variables all end with "_", most likely the original intention is
that they are not official and may be removed or changed at any time.

By making them official, we lock in that ABI.  Is that your intention?

(likewise your other patch).

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Siddhesh Poyarekar]

On Monday 10 October 2016 11:03 PM, DJ Delorie wrote:
> These variables all end with "_", most likely the original intention is
> that they are not official and may be removed or changed at any time.
> 
> By making them official, we lock in that ABI.  Is that your intention?
> 
> (likewise your other patch).

I don't see the point in not doing so because they're too widely in use
for us to take them back anyway.

Siddhesh

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Michael Kerrisk]

On Mon, Oct 10, 2016 at 7:33 PM, DJ Delorie <dj@redhat.com> wrote:
>
> These variables all end with "_", most likely the original intention is
> that they are not official and may be removed or changed at any time.
>
> By making them official, we lock in that ABI.  Is that your intention?

I think the notion that because one does not document something, it's
not an official part of the ABI is at best highly dubious. Especially
when so many "official" parts of the ABI are also not documented.

The only way that documentation might be able to help in such
situations is where pieces are clearly and loudly documented right
from the beginning, in the official documentation, as "not supported,
may disappear at any moment in the future, use at your own risk", but
even then people are likely to ignore the documentation or be unaware
of it.

And in any case, these environment vars have long been documented in
various "unofficial" places including (since 2012)
http://man7.org/linux/man-pages/man3/mallopt.3.html

Cheers,

Michael

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[DJ Delorie]

Michael Kerrisk <mtk.manpages@gmail.com> writes:
> I think the notion that because one does not document something, it's
> not an official part of the ABI is at best highly dubious.

I wasn't here for the origin of those variables, I'm just thinking that
we use a leading underscore elsewhere to mean "not official" so the
trailing underscore here might have similar intent.

FTR I have no problem with documenting unsupported behavior, I just want
to make sure we understand how this patch affects the officialness of
this ABI.

> The only way that documentation might be able to help in such
> situations is where pieces are clearly and loudly documented right
> from the beginning, in the official documentation, as "not supported,
> may disappear at any moment in the future, use at your own risk", but
> even then people are likely to ignore the documentation or be unaware
> of it.

That would be fine too.  My concern is: if we intended for those
variables to be unofficial before, do we want to (1) make them official
now, or (2) be careful not to *accidentally* make them official?

And if we decide they're official (and/or always have been), should we
add in variants without underscores to be the official ones?

So I guess the next step is to have someone in authority (or consensus?)
decide if those variables are "official" or not, and if they should
become so if not.

Based on that we may want to tweak the documentation patch, or the code.

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Michael Kerrisk (man-pages)]

On 10/11/2016 08:18 PM, DJ Delorie wrote:
> 
> Michael Kerrisk <mtk.manpages@gmail.com> writes:
>> I think the notion that because one does not document something, it's
>> not an official part of the ABI is at best highly dubious.
> 
> I wasn't here for the origin of those variables, I'm just thinking that
> we use a leading underscore elsewhere to mean "not official" so the
> trailing underscore here might have similar intent.
> 
> FTR I have no problem with documenting unsupported behavior, I just want
> to make sure we understand how this patch affects the officialness of
> this ABI.

My thought is: not very much. The ABI is out there and is well known.
There's not really anyway to argue to users that it is not "official",
no matter what the original developers might once have thought about
these interfaces.

Cheers,

Michael

-- 
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Linux/UNIX System Programming Training: http://man7.org/training/

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Kalle Olavi Niemitalo]

Siddhesh Poyarekar <siddhesh@sourceware.org> writes:

> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_MMAP_MAX_} to the desired value.

startup.texi has e.g. "@cindex @code{TZ} environment variable".
Could add similar index entries for MALLOC_MMAP_MAX_ and others.

Inconsistently, @vindex is used for the PWD, LANGUAGE, and
COREFILE environment variables.

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Siddhesh Poyarekar]

On Tuesday 11 October 2016 11:48 PM, DJ Delorie wrote:
> I wasn't here for the origin of those variables, I'm just thinking that
> we use a leading underscore elsewhere to mean "not official" so the
> trailing underscore here might have similar intent.

I agree, that may have been the intent.  However what happened
afterwards (i.e. their widespread use and the lack of communication to
the contrary from the then maintainers) have invalidated that intent
IMO.  What's more, other important sources (the man pages, Red Hat
product documentation) document these environment variables and that
gives them as much a sense of credibility as anything else.

> And if we decide they're official (and/or always have been), should we
> add in variants without underscores to be the official ones?

I'll be adding in tunables, which would become the 'official' versions
of these envvars.  Tunables come with the disclaimer that they may
disappear/reappear in releases, but I don't think we will do that for
the malloc tunables in practice.

> So I guess the next step is to have someone in authority (or consensus?)
> decide if those variables are "official" or not, and if they should
> become so if not.

It is usually consensus.

Siddhesh

[PING][PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Siddhesh Poyarekar]

Ping!

On Monday 10 October 2016 10:56 PM, Siddhesh Poyarekar wrote:
> The mallopt parameters manual does not mention the environment
> variables that can be used to set these parameters at program startup.
> Mention those environment variables for completeness.
> 
> 	* manual/memory.texi: Add environment variable alternatives to
> 	setting mallopt parameters.
> ---
>  manual/memory.texi | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/manual/memory.texi b/manual/memory.texi
> index 5383105..222f126 100644
> --- a/manual/memory.texi
> +++ b/manual/memory.texi
> @@ -1113,12 +1113,18 @@ choices for @var{param}, as defined in @file{malloc.h}, are:
>  @item M_MMAP_MAX
>  The maximum number of chunks to allocate with @code{mmap}.  Setting this
>  to zero disables all use of @code{mmap}.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_MMAP_MAX_} to the desired value.
>  @item M_MMAP_THRESHOLD
>  All chunks larger than this value are allocated outside the normal
>  heap, using the @code{mmap} system call.  This way it is guaranteed
>  that the memory for these chunks can be returned to the system on
>  @code{free}.  Note that requests smaller than this threshold might still
>  be allocated via @code{mmap}.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_MMAP_THRESHOLD_} to the desired value.
>  @comment TODO: @item M_MXFAST
>  @item M_PERTURB
>  If non-zero, memory blocks are filled with values depending on some
> @@ -1128,16 +1134,25 @@ use of uninitialized or freed heap memory.  Note that this option does not
>  guarantee that the freed block will have any specific values.  It only
>  guarantees that the content the block had before it was freed will be
>  overwritten.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_MMAP_PERTURB_} to the desired value.
>  @item M_TOP_PAD
>  This parameter determines the amount of extra memory to obtain from the
>  system when a call to @code{sbrk} is required.  It also specifies the
>  number of bytes to retain when shrinking the heap by calling @code{sbrk}
>  with a negative argument.  This provides the necessary hysteresis in
>  heap size such that excessive amounts of system calls can be avoided.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_TOP_PAD_} to the desired value.
>  @item M_TRIM_THRESHOLD
>  This is the minimum size (in bytes) of the top-most, releasable chunk
>  that will cause @code{sbrk} to be called with a negative argument in
>  order to return memory to the system.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_TRIM_THRESHOLD_} to the desired value.
>  @end table
>  
>  @end deftypefun
> 

[PING][PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Siddhesh Poyarekar]

Ping!

On Monday 10 October 2016 10:56 PM, Siddhesh Poyarekar wrote:
> The M_ARENA_* mallopt parameters are in wide use in production to
> control the number of arenas that a long lived process creates and
> hence there is no point in stating that this interface is non-public.
> Document this interface and remove the obsolete comment.
> 
> 	* manual/memory.texi (M_ARENA_TEST): Add documentation.
> 	(M_ARENA_MAX): Likewise.
> 	* malloc/malloc.c: Remove obsolete comment.
> ---
>  malloc/malloc.c    |  1 -
>  manual/memory.texi | 21 +++++++++++++++++++++
>  2 files changed, 21 insertions(+), 1 deletion(-)
> 
> diff --git a/malloc/malloc.c b/malloc/malloc.c
> index 09e004b..0011a6d 100644
> --- a/malloc/malloc.c
> +++ b/malloc/malloc.c
> @@ -1718,7 +1718,6 @@ static struct malloc_par mp_ =
>  };
>  
>  
> -/*  Non public mallopt parameters.  */
>  #define M_ARENA_TEST -7
>  #define M_ARENA_MAX  -8
>  
> diff --git a/manual/memory.texi b/manual/memory.texi
> index 222f126..b98dcf2 100644
> --- a/manual/memory.texi
> +++ b/manual/memory.texi
> @@ -1153,6 +1153,27 @@ order to return memory to the system.
>  
>  This parameter can also be set for the process at startup by setting the
>  environment variable @code{MALLOC_TRIM_THRESHOLD_} to the desired value.
> +
> +@item M_ARENA_TEST
> +An arena is a memory pool that is allocated to act as a heap alongside the
> +system heap that the kernel creates for the process.  This is to provide
> +multiple heap structures for multiple threads of a process to reduce contention
> +between them.  The limit to the number of such arenas per process is determined
> +by the number of cores of the system and whether it is a 32-bit or a 64-bit
> +processor.
> +
> +The value is ignored if @code{M_ARENA_MAX} is set either via @code{mallopt} or
> +via the @code{MALLOC_ARENA_MAX} environment variable.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_ARENA_TEST} to the desired value.
> +
> +@item M_ARENA_MAX
> +This parameter sets the number of arenas to use regardless of the number of
> +cores in the system.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_ARENA_MAX} to the desired value.
>  @end table
>  
>  @end deftypefun
> 

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Carlos O'Donell]

On 10/10/2016 01:26 PM, Siddhesh Poyarekar wrote:
> The M_ARENA_* mallopt parameters are in wide use in production to
> control the number of arenas that a long lived process creates and
> hence there is no point in stating that this interface is non-public.
> Document this interface and remove the obsolete comment.
> 
> 	* manual/memory.texi (M_ARENA_TEST): Add documentation.
> 	(M_ARENA_MAX): Likewise.
> 	* malloc/malloc.c: Remove obsolete comment.
> ---
>  malloc/malloc.c    |  1 -
>  manual/memory.texi | 21 +++++++++++++++++++++
>  2 files changed, 21 insertions(+), 1 deletion(-)
> 
> diff --git a/malloc/malloc.c b/malloc/malloc.c
> index 09e004b..0011a6d 100644
> --- a/malloc/malloc.c
> +++ b/malloc/malloc.c
> @@ -1718,7 +1718,6 @@ static struct malloc_par mp_ =
>  };
>  
>  
> -/*  Non public mallopt parameters.  */
>  #define M_ARENA_TEST -7
>  #define M_ARENA_MAX  -8
>  
> diff --git a/manual/memory.texi b/manual/memory.texi
> index 222f126..b98dcf2 100644
> --- a/manual/memory.texi
> +++ b/manual/memory.texi
> @@ -1153,6 +1153,27 @@ order to return memory to the system.
>  
>  This parameter can also be set for the process at startup by setting the
>  environment variable @code{MALLOC_TRIM_THRESHOLD_} to the desired value.
> +
> +@item M_ARENA_TEST

This description of an arena doesn't belong here, it belongs higher up in the
manual, and in fact I would suggest the following:

- Move 3.2.2.6 "Efficiency considerations for malloc" up to just under
  "Dynamic Memory allocation" and put it under a new subsection "The GNU allocator".

- Under "The GNU allocator" give a brief 1 paragraph description of how
  the allocator works e.g. https://sourceware.org/glibc/wiki/MallocInternals
  and talk about arenas.

> +An arena is a memory pool that is allocated to act as a heap alongside the
> +system heap that the kernel creates for the process.  This is to provide
> +multiple heap structures for multiple threads of a process to reduce contention
> +between them.  The limit to the number of such arenas per process is determined
> +by the number of cores of the system and whether it is a 32-bit or a 64-bit
> +processor.

You don't say what M_ARENA_TEST is for?

> +
> +The value is ignored if @code{M_ARENA_MAX} is set either via @code{mallopt} or
> +via the @code{MALLOC_ARENA_MAX} environment variable.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_ARENA_TEST} to the desired value.

What is the default?

> +
> +@item M_ARENA_MAX
> +This parameter sets the number of arenas to use regardless of the number of
> +cores in the system.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_ARENA_MAX} to the desired value.

What is the default?

>  @end table
>  
>  @end deftypefun
> 


-- 
Cheers,
Carlos.

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Carlos O'Donell]

On 10/10/2016 01:26 PM, Siddhesh Poyarekar wrote:
> The mallopt parameters manual does not mention the environment
> variables that can be used to set these parameters at program startup.
> Mention those environment variables for completeness.

Please don't get upset with me :-)

I'm asking you to document the default for these env var tunables.

I think it's valid for users to want to know what value they would have
had if they _didn't_ set the environment variable.

Throwing our hands up in the air and saying "it's implementation defined"
isn't very useful IMO.

> 	* manual/memory.texi: Add environment variable alternatives to
> 	setting mallopt parameters.
> ---
>  manual/memory.texi | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/manual/memory.texi b/manual/memory.texi
> index 5383105..222f126 100644
> --- a/manual/memory.texi
> +++ b/manual/memory.texi
> @@ -1113,12 +1113,18 @@ choices for @var{param}, as defined in @file{malloc.h}, are:
>  @item M_MMAP_MAX
>  The maximum number of chunks to allocate with @code{mmap}.  Setting this
>  to zero disables all use of @code{mmap}.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_MMAP_MAX_} to the desired value.

What's the default?

>  @item M_MMAP_THRESHOLD
>  All chunks larger than this value are allocated outside the normal
>  heap, using the @code{mmap} system call.  This way it is guaranteed
>  that the memory for these chunks can be returned to the system on
>  @code{free}.  Note that requests smaller than this threshold might still
>  be allocated via @code{mmap}.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_MMAP_THRESHOLD_} to the desired value.

What's the default?

>  @comment TODO: @item M_MXFAST
>  @item M_PERTURB
>  If non-zero, memory blocks are filled with values depending on some
> @@ -1128,16 +1134,25 @@ use of uninitialized or freed heap memory.  Note that this option does not
>  guarantee that the freed block will have any specific values.  It only
>  guarantees that the content the block had before it was freed will be
>  overwritten.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_MMAP_PERTURB_} to the desired value.

What is the default?

>  @item M_TOP_PAD
>  This parameter determines the amount of extra memory to obtain from the
>  system when a call to @code{sbrk} is required.  It also specifies the
>  number of bytes to retain when shrinking the heap by calling @code{sbrk}
>  with a negative argument.  This provides the necessary hysteresis in
>  heap size such that excessive amounts of system calls can be avoided.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_TOP_PAD_} to the desired value.

What's the default?

>  @item M_TRIM_THRESHOLD
>  This is the minimum size (in bytes) of the top-most, releasable chunk
>  that will cause @code{sbrk} to be called with a negative argument in
>  order to return memory to the system.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_TRIM_THRESHOLD_} to the desired value.

What's the default? Note: There is a limit even with dynamic trim threshold on.

This disables the dynamic trim threshold, and that's a very important thing to
mention.

>  @end table
>  
>  @end deftypefun
> 


-- 
Cheers,
Carlos.

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Siddhesh Poyarekar]

On Monday 17 October 2016 09:43 PM, Carlos O'Donell wrote:
> On 10/10/2016 01:26 PM, Siddhesh Poyarekar wrote:
>> The mallopt parameters manual does not mention the environment
>> variables that can be used to set these parameters at program startup.
>> Mention those environment variables for completeness.
> 
> Please don't get upset with me :-)

Grrr...

> I'm asking you to document the default for these env var tunables.
> 
> I think it's valid for users to want to know what value they would have
> had if they _didn't_ set the environment variable.
> 
> Throwing our hands up in the air and saying "it's implementation defined"
> isn't very useful IMO.

No I agree that is a good point.  I'll work on it based on your comments
on the MALLOC_ARENA_* doc patch, i.e. first describe the GNU allocator
and then use that information to explain the envvars/mallopt params and
also add in defaults.

Siddhesh

Re: [PATCH 1/2] Add note on MALLOC_MMAP_* environment variables

[Carlos O'Donell]

On 10/11/2016 02:20 AM, Michael Kerrisk wrote:
> On Mon, Oct 10, 2016 at 7:33 PM, DJ Delorie <dj@redhat.com> wrote:
>>
>> These variables all end with "_", most likely the original intention is
>> that they are not official and may be removed or changed at any time.
>>
>> By making them official, we lock in that ABI.  Is that your intention?
> 
> I think the notion that because one does not document something, it's
> not an official part of the ABI is at best highly dubious. Especially
> when so many "official" parts of the ABI are also not documented.
> 
> The only way that documentation might be able to help in such
> situations is where pieces are clearly and loudly documented right
> from the beginning, in the official documentation, as "not supported,
> may disappear at any moment in the future, use at your own risk", but
> even then people are likely to ignore the documentation or be unaware
> of it.
> 
> And in any case, these environment vars have long been documented in
> various "unofficial" places including (since 2012)
> http://man7.org/linux/man-pages/man3/mallopt.3.html

The things that DJ is worried about are the arena ones
e.g. M_ARENA_MAX, and M_ARENA_TEST and their env vars.

In 2013 I brought up the discussion about ABI implications:
https://sourceware.org/ml/libc-alpha/2013-03/msg00376.html

Only Siddhesh and I commented, and we both agreed it was a forgone
conclusion that these were stable parts of the ABI/API.

In 2015 I documented them in the linux man pages project:
https://www.sourceware.org/ml/libc-alpha/2015-08/msg00991.html
~~~
Consensus among Siddhesh and myself was that they should be public,
and in fact they were already in the public header. Therefore there
may already be applications uses these constants and expecting them
to work. At best we could limit mallopt's acceptance of the options,
but that seems like a bad solution that could lead to unexpected
behaviour for user applications. A quick google search shows that
there are packages relying on these constants to tune the glibc
malloc implementation.
~~~

They are part of the ABI and public, and should be documented
in the manual as Siddhesh's patches do.

We should however be very very circumspect about adding more
of these mallopt tunables without first seeing that such tunables
are stable and long-term useful as generic tunables (via the tunables
interface). So go review Siddhesh's patches on tunables :-)

-- 
Cheers,
Carlos.

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Michael Kerrisk]

On Mon, Oct 17, 2016 at 6:08 PM, Carlos O'Donell <carlos@redhat.com> wrote:
> On 10/10/2016 01:26 PM, Siddhesh Poyarekar wrote:
>> The M_ARENA_* mallopt parameters are in wide use in production to
>> control the number of arenas that a long lived process creates and
>> hence there is no point in stating that this interface is non-public.
>> Document this interface and remove the obsolete comment.
>>
>>       * manual/memory.texi (M_ARENA_TEST): Add documentation.
>>       (M_ARENA_MAX): Likewise.
>>       * malloc/malloc.c: Remove obsolete comment.
>> ---
>>  malloc/malloc.c    |  1 -
>>  manual/memory.texi | 21 +++++++++++++++++++++
>>  2 files changed, 21 insertions(+), 1 deletion(-)
>>
>> diff --git a/malloc/malloc.c b/malloc/malloc.c
>> index 09e004b..0011a6d 100644
>> --- a/malloc/malloc.c
>> +++ b/malloc/malloc.c
>> @@ -1718,7 +1718,6 @@ static struct malloc_par mp_ =
>>  };
>>
>>
>> -/*  Non public mallopt parameters.  */
>>  #define M_ARENA_TEST -7
>>  #define M_ARENA_MAX  -8
>>
>> diff --git a/manual/memory.texi b/manual/memory.texi
>> index 222f126..b98dcf2 100644
>> --- a/manual/memory.texi
>> +++ b/manual/memory.texi
>> @@ -1153,6 +1153,27 @@ order to return memory to the system.
>>
>>  This parameter can also be set for the process at startup by setting the
>>  environment variable @code{MALLOC_TRIM_THRESHOLD_} to the desired value.
>> +
>> +@item M_ARENA_TEST
>
> This description of an arena doesn't belong here, it belongs higher up in the
> manual, and in fact I would suggest the following:
>
> - Move 3.2.2.6 "Efficiency considerations for malloc" up to just under
>   "Dynamic Memory allocation" and put it under a new subsection "The GNU allocator".
>
> - Under "The GNU allocator" give a brief 1 paragraph description of how
>   the allocator works e.g. https://sourceware.org/glibc/wiki/MallocInternals
>   and talk about arenas.
>
>> +An arena is a memory pool that is allocated to act as a heap alongside the
>> +system heap that the kernel creates for the process.  This is to provide
>> +multiple heap structures for multiple threads of a process to reduce contention
>> +between them.  The limit to the number of such arenas per process is determined
>> +by the number of cores of the system and whether it is a 32-bit or a 64-bit
>> +processor.
>
> You don't say what M_ARENA_TEST is for?
>
>> +
>> +The value is ignored if @code{M_ARENA_MAX} is set either via @code{mallopt} or
>> +via the @code{MALLOC_ARENA_MAX} environment variable.
>> +
>> +This parameter can also be set for the process at startup by setting the
>> +environment variable @code{MALLOC_ARENA_TEST} to the desired value.
>
> What is the default?

So my reading from malloc/malloc.c:

#define NARENAS_FROM_NCORES(n) ((n) * (sizeof (long) == 4 ? 2 : 8))
  .arena_test = NARENAS_FROM_NCORES (1)

So, the default value for this parameter is 2 on systems where
sizeof(long) is 4; otherwise the default value is 8.

>> +
>> +@item M_ARENA_MAX
>> +This parameter sets the number of arenas to use regardless of the number of
>> +cores in the system.
>> +
>> +This parameter can also be set for the process at startup by setting the
>> +environment variable @code{MALLOC_ARENA_MAX} to the desired value.
>
> What is the default?

So, IIUC, the default value of this parameter is 0, meaning that there
is no limit on the number of arenas that can be created. Do you
confirm, Siddhesh?

Siddhesh, in case you want to use any of my wordings (assuming they
are correct), I place them in the public domain. (The text will also
land in the mallopt(3) manual page.)

Cheers,

Michael

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Siddhesh Poyarekar]

On Tuesday 18 October 2016 12:45 PM, Michael Kerrisk wrote:
> So my reading from malloc/malloc.c:
> 
> #define NARENAS_FROM_NCORES(n) ((n) * (sizeof (long) == 4 ? 2 : 8))
>   .arena_test = NARENAS_FROM_NCORES (1)
> 
> So, the default value for this parameter is 2 on systems where
> sizeof(long) is 4; otherwise the default value is 8.

That is correct.

> So, IIUC, the default value of this parameter is 0, meaning that there
> is no limit on the number of arenas that can be created. Do you
> confirm, Siddhesh?

No, the default is a function of the number of cores using the
NARENAS_FROM_NCORES macro.  That is, 2 * (number of cores) where
sizeof(long) == 4 and 8 * (number of cores) otherwise.  If the number of
cores is not available (i.e. we can't read this info for some reason)
then we default to 4 and 16 respectively, i.e. assume 2 cores.

Note that this default only comes into force once the number of arenas
cross arena_test.

> Siddhesh, in case you want to use any of my wordings (assuming they
> are correct), I place them in the public domain. (The text will also
> land in the mallopt(3) manual page.)

Thanks.

Siddhesh

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Michael Kerrisk (man-pages)]

Hi Siddhesh,

On 10/18/2016 12:07 PM, Siddhesh Poyarekar wrote:
> On Tuesday 18 October 2016 12:45 PM, Michael Kerrisk wrote:
>> So my reading from malloc/malloc.c:
>>
>> #define NARENAS_FROM_NCORES(n) ((n) * (sizeof (long) == 4 ? 2 : 8))
>>   .arena_test = NARENAS_FROM_NCORES (1)
>>
>> So, the default value for this parameter is 2 on systems where
>> sizeof(long) is 4; otherwise the default value is 8.
> 
> That is correct.

Thanks for the confirmation.

[You've over trimmed the mail for this next piece. To be clear, below,
we are talking about M_ARENA_MAX; or at least I was.]

>> So, IIUC, the default value of this parameter is 0, meaning that there
>> is no limit on the number of arenas that can be created. Do you
>> confirm, Siddhesh?
> 
> No, the default is a function of the number of cores using the
> NARENAS_FROM_NCORES macro.  That is, 2 * (number of cores) where
> sizeof(long) == 4 and 8 * (number of cores) otherwise.  If the number of
> cores is not available (i.e. we can't read this info for some reason)
> then we default to 4 and 16 respectively, i.e. assume 2 cores.
> 
> Note that this default only comes into force once the number of arenas
> cross arena_test.

I don't think you're correct here. 'arena_max' is a field in a static
structure that is not otherwise initialized, AFAICT. So, it has the
value zero. (Some dirty hacking with a program that uses
malloc_get_state() and inspects the internal data structure seems
to confirm this.)

And then in malloc/arena.c we have

          if (mp_.arena_max != 0)
            narenas_limit = mp_.arena_max;
          else if (narenas > mp_.arena_test)
            {
              int n = __get_nprocs ();
              
              if (n >= 1)
                narenas_limit = NARENAS_FROM_NCORES (n);
              else
                /* We have no information about the system.  Assume two
                   cores.  */ 
                narenas_limit = NARENAS_FROM_NCORES (2);

So, I believe my original statement about M_ARENA_MAX is correct.
Have I missed something?

Cheers,

Michael

>> Siddhesh, in case you want to use any of my wordings (assuming they
>> are correct), I place them in the public domain. (The text will also
>> land in the mallopt(3) manual page.)
> 
> Thanks.
> 
> Siddhesh
> 


-- 
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Linux/UNIX System Programming Training: http://man7.org/training/

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Siddhesh Poyarekar]

On Tuesday 18 October 2016 07:20 PM, Michael Kerrisk (man-pages) wrote:
> Hi Siddhesh,
> 
> On 10/18/2016 12:07 PM, Siddhesh Poyarekar wrote:
>> On Tuesday 18 October 2016 12:45 PM, Michael Kerrisk wrote:
>>> So my reading from malloc/malloc.c:
>>>
>>> #define NARENAS_FROM_NCORES(n) ((n) * (sizeof (long) == 4 ? 2 : 8))
>>>   .arena_test = NARENAS_FROM_NCORES (1)
>>>
>>> So, the default value for this parameter is 2 on systems where
>>> sizeof(long) is 4; otherwise the default value is 8.
>>
>> That is correct.
> 
> Thanks for the confirmation.
> 
> [You've over trimmed the mail for this next piece. To be clear, below,
> we are talking about M_ARENA_MAX; or at least I was.]
> 
>>> So, IIUC, the default value of this parameter is 0, meaning that there
>>> is no limit on the number of arenas that can be created. Do you
>>> confirm, Siddhesh?
>>
>> No, the default is a function of the number of cores using the
>> NARENAS_FROM_NCORES macro.  That is, 2 * (number of cores) where
>> sizeof(long) == 4 and 8 * (number of cores) otherwise.  If the number of
>> cores is not available (i.e. we can't read this info for some reason)
>> then we default to 4 and 16 respectively, i.e. assume 2 cores.
>>
>> Note that this default only comes into force once the number of arenas
>> cross arena_test.
> 
> I don't think you're correct here. 'arena_max' is a field in a static
> structure that is not otherwise initialized, AFAICT. So, it has the
> value zero. (Some dirty hacking with a program that uses
> malloc_get_state() and inspects the internal data structure seems
> to confirm this.)
> 
> And then in malloc/arena.c we have
> 
>           if (mp_.arena_max != 0)
>             narenas_limit = mp_.arena_max;
>           else if (narenas > mp_.arena_test)
>             {
>               int n = __get_nprocs ();
>               
>               if (n >= 1)
>                 narenas_limit = NARENAS_FROM_NCORES (n);
>               else
>                 /* We have no information about the system.  Assume two
>                    cores.  */ 
>                 narenas_limit = NARENAS_FROM_NCORES (2);
> 
> So, I believe my original statement about M_ARENA_MAX is correct.
> Have I missed something?

You're right in that the variable arena_max is initialized to 0.
However you also concluded that there is no limit to the number of
arenas that can be created when arena_max is 0, which is incorrect.  As
the code snippet you pasted above shows that if arena_max is 0, once we
cross arena_test arenas, the narenas_limit static variable is set to a
default upper limit based on the number of cores.  That acts as the
upper limit to the number of arenas that can be created when arena_max is 0.

Siddhesh

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Michael Kerrisk (man-pages)]

Hi Siddhesh,

On 10/18/2016 04:30 PM, Siddhesh Poyarekar wrote:
> On Tuesday 18 October 2016 07:20 PM, Michael Kerrisk (man-pages) wrote:
>> Hi Siddhesh,
>>
>> On 10/18/2016 12:07 PM, Siddhesh Poyarekar wrote:
>>> On Tuesday 18 October 2016 12:45 PM, Michael Kerrisk wrote:
>>>> So my reading from malloc/malloc.c:
>>>>
>>>> #define NARENAS_FROM_NCORES(n) ((n) * (sizeof (long) == 4 ? 2 : 8))
>>>>   .arena_test = NARENAS_FROM_NCORES (1)
>>>>
>>>> So, the default value for this parameter is 2 on systems where
>>>> sizeof(long) is 4; otherwise the default value is 8.
>>>
>>> That is correct.
>>
>> Thanks for the confirmation.
>>
>> [You've over trimmed the mail for this next piece. To be clear, below,
>> we are talking about M_ARENA_MAX; or at least I was.]
>>
>>>> So, IIUC, the default value of this parameter is 0, meaning that there
>>>> is no limit on the number of arenas that can be created. Do you
>>>> confirm, Siddhesh?
>>>
>>> No, the default is a function of the number of cores using the
>>> NARENAS_FROM_NCORES macro.  That is, 2 * (number of cores) where
>>> sizeof(long) == 4 and 8 * (number of cores) otherwise.  If the number of
>>> cores is not available (i.e. we can't read this info for some reason)
>>> then we default to 4 and 16 respectively, i.e. assume 2 cores.
>>>
>>> Note that this default only comes into force once the number of arenas
>>> cross arena_test.
>>
>> I don't think you're correct here. 'arena_max' is a field in a static
>> structure that is not otherwise initialized, AFAICT. So, it has the
>> value zero. (Some dirty hacking with a program that uses
>> malloc_get_state() and inspects the internal data structure seems
>> to confirm this.)
>>
>> And then in malloc/arena.c we have
>>
>>           if (mp_.arena_max != 0)
>>             narenas_limit = mp_.arena_max;
>>           else if (narenas > mp_.arena_test)
>>             {
>>               int n = __get_nprocs ();
>>               
>>               if (n >= 1)
>>                 narenas_limit = NARENAS_FROM_NCORES (n);
>>               else
>>                 /* We have no information about the system.  Assume two
>>                    cores.  */ 
>>                 narenas_limit = NARENAS_FROM_NCORES (2);
>>
>> So, I believe my original statement about M_ARENA_MAX is correct.
>> Have I missed something?
> 
> You're right in that the variable arena_max is initialized to 0.

Okay.

> However you also concluded that there is no limit to the number of
> arenas that can be created when arena_max is 0, which is incorrect.  As
> the code snippet you pasted above shows that if arena_max is 0, once we
> cross arena_test arenas, the narenas_limit static variable is set to a
> default upper limit based on the number of cores.  That acts as the
> upper limit to the number of arenas that can be created when arena_max is 0.

D'oh!  Yes, of course. Thanks for that. Not sure how I managed to 
misread that code :-}. So a better formulation would be something like:

    The default value of this parameter is 0, meaning that the limit on 
    the number of arenas is determined according to the setting of
    M_ARENA_TEST.

Seem okay?

Cheers,

Michael

-- 
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Linux/UNIX System Programming Training: http://man7.org/training/

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Siddhesh Poyarekar]

On Tuesday 18 October 2016 09:33 PM, Michael Kerrisk (man-pages) wrote:
>>> I don't think you're correct here. 'arena_max' is a field in a static
>>> structure that is not otherwise initialized, AFAICT. So, it has the
>>> value zero. (Some dirty hacking with a program that uses
>>> malloc_get_state() and inspects the internal data structure seems
>>> to confirm this.)
>>>
>>> And then in malloc/arena.c we have
>>>
>>>           if (mp_.arena_max != 0)
>>>             narenas_limit = mp_.arena_max;
>>>           else if (narenas > mp_.arena_test)
>>>             {
>>>               int n = __get_nprocs ();
>>>               
>>>               if (n >= 1)
>>>                 narenas_limit = NARENAS_FROM_NCORES (n);
>>>               else
>>>                 /* We have no information about the system.  Assume two
>>>                    cores.  */ 
>>>                 narenas_limit = NARENAS_FROM_NCORES (2);
>>>
>>> So, I believe my original statement about M_ARENA_MAX is correct.
>>> Have I missed something?
>>
>> You're right in that the variable arena_max is initialized to 0.
> 
> Okay.
> 
>> However you also concluded that there is no limit to the number of
>> arenas that can be created when arena_max is 0, which is incorrect.  As
>> the code snippet you pasted above shows that if arena_max is 0, once we
>> cross arena_test arenas, the narenas_limit static variable is set to a
>> default upper limit based on the number of cores.  That acts as the
>> upper limit to the number of arenas that can be created when arena_max is 0.
> 
> D'oh!  Yes, of course. Thanks for that. Not sure how I managed to 
> misread that code :-}. So a better formulation would be something like:
> 
>     The default value of this parameter is 0, meaning that the limit on 
>     the number of arenas is determined according to the setting of
>     M_ARENA_TEST.

No, it is not the value of M_ARENA_TEST :)

M_ARENA_TEST is set by default to NARENAS_FROM_CORES(1).  The limit on
the number of arenas is decided after M_ARENA_TEST number of arenas have
already created, but it is actually set to NARENAS_FROM_CORES(n), where
n is the number of cores.  If we can't find the number of cores for some
reason, then we set the limit to NARENAS_FROM_CORES(2).

So something like this is more accurate:

    The default value of this parameter is 0, meaning that the limit on
    the number of arenas is determined by the number of CPU cores
    online and the size of the '''long''' datatype.  For 32-bit systems
    the limit of on the number of arenas is 2 * '''number of CPU cores
    online''' while 64-bit systems, the limit on the number of arenas
    is 8 * '''number of CPU cores online'''.  If information on CPU
    cores is not available, it is assumed that there are 2 CPU cores
    online.

I have implicitly stated here that 32-bit long == 32-bit pointers since
I don't think there are Unix systems that have 32-bit long and 64-bit
pointers and also because the choice of multipliers for cores (2 and 8)
correlates better to the size of the address space than to size of long.

Siddhesh

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Andreas Schwab]

On Okt 18 2016, Siddhesh Poyarekar <siddhesh@sourceware.org> wrote:

> So something like this is more accurate:
>
>     The default value of this parameter is 0, meaning that the limit on
>     the number of arenas is determined by the number of CPU cores
>     online and the size of the '''long''' datatype.  For 32-bit systems
>     the limit of on the number of arenas is 2 * '''number of CPU cores
                                              twice the
>     online''' while 64-bit systems, the limit on the number of arenas
>     is 8 * '''number of CPU cores online'''.  If information on CPU
         eight times the
>     cores is not available, it is assumed that there are 2 CPU cores
>     online.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Michael Kerrisk (man-pages)]

Hi Siddhesh,

On 10/18/2016 06:46 PM, Siddhesh Poyarekar wrote:
> On Tuesday 18 October 2016 09:33 PM, Michael Kerrisk (man-pages) wrote:
>>>> I don't think you're correct here. 'arena_max' is a field in a static
>>>> structure that is not otherwise initialized, AFAICT. So, it has the
>>>> value zero. (Some dirty hacking with a program that uses
>>>> malloc_get_state() and inspects the internal data structure seems
>>>> to confirm this.)
>>>>
>>>> And then in malloc/arena.c we have
>>>>
>>>>           if (mp_.arena_max != 0)
>>>>             narenas_limit = mp_.arena_max;
>>>>           else if (narenas > mp_.arena_test)
>>>>             {
>>>>               int n = __get_nprocs ();
>>>>               
>>>>               if (n >= 1)
>>>>                 narenas_limit = NARENAS_FROM_NCORES (n);
>>>>               else
>>>>                 /* We have no information about the system.  Assume two
>>>>                    cores.  */ 
>>>>                 narenas_limit = NARENAS_FROM_NCORES (2);
>>>>
>>>> So, I believe my original statement about M_ARENA_MAX is correct.
>>>> Have I missed something?
>>>
>>> You're right in that the variable arena_max is initialized to 0.
>>
>> Okay.
>>
>>> However you also concluded that there is no limit to the number of
>>> arenas that can be created when arena_max is 0, which is incorrect.  As
>>> the code snippet you pasted above shows that if arena_max is 0, once we
>>> cross arena_test arenas, the narenas_limit static variable is set to a
>>> default upper limit based on the number of cores.  That acts as the
>>> upper limit to the number of arenas that can be created when arena_max is 0.
>>
>> D'oh!  Yes, of course. Thanks for that. Not sure how I managed to 
>> misread that code :-}. So a better formulation would be something like:
>>
>>     The default value of this parameter is 0, meaning that the limit on 
>>     the number of arenas is determined according to the setting of
>>     M_ARENA_TEST.
> 
> No, it is not the value of M_ARENA_TEST :)

I agree. But it's also not what I said ;-). I said "determined according
to the value of..."

Currently, the M_ARENA_TEST text in mallopt(3) says:

       M_ARENA_TEST
              This  parameter specifies a value, in number of arenas cre‐
              ated, at which point the system configuration will be exam‐
              ined  to  determine  a  hard limit on the number of created
              arenas.  (See M_ARENA_MAX for the definition of an arena.)

              The computation of the arena hard limit is  implementation-
              defined and is usually calculated as a multiple of the num‐
              ber of available CPUs.  Once the hard  limit  is  computed,
              the result is final and constrains the total number of are‐
              nas.

              The default value for the M_ARENA_TEST parameter  is  2  on
              systems  where  sizeof(long)  is  4;  otherwise the default
              value is 8.

              This parameter has been  available  since  glibc  2.10  via
              --enable-experimental-malloc,   and  since  glibc  2.15  by
              default.

              The value of M_ARENA_TEST is not used when M_ARENA_MAX  has
              a nonzero value.

> M_ARENA_TEST is set by default to NARENAS_FROM_CORES(1).  The limit on
> the number of arenas is decided after M_ARENA_TEST number of arenas have
> already created, but it is actually set to NARENAS_FROM_CORES(n), where
> n is the number of cores.  If we can't find the number of cores for some
> reason, then we set the limit to NARENAS_FROM_CORES(2).
> 
> So something like this is more accurate:
> 
>     The default value of this parameter is 0, meaning that the limit on
>     the number of arenas is determined by the number of CPU cores
>     online and the size of the '''long''' datatype.  For 32-bit systems
>     the limit of on the number of arenas is 2 * '''number of CPU cores
>     online''' while 64-bit systems, the limit on the number of arenas
>     is 8 * '''number of CPU cores online'''.  If information on CPU
>     cores is not available, it is assumed that there are 2 CPU cores
>     online.

Your details following are a little more precise. But I believe Carlos,
when he wrote the text about how the limit was calculated, deliberately
chose to be more vague. I'm okay with including this level of detail 
in the man page, but, if I recall correctly, Carlos wanted to avoid it.

Cheers,

Michael

> I have implicitly stated here that 32-bit long == 32-bit pointers since
> I don't think there are Unix systems that have 32-bit long and 64-bit
> pointers and also because the choice of multipliers for cores (2 and 8)
> correlates better to the size of the address space than to size of long.
> 
> Siddhesh
> 


-- 
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Linux/UNIX System Programming Training: http://man7.org/training/

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Siddhesh Poyarekar]

On Wednesday 19 October 2016 12:23 PM, Michael Kerrisk (man-pages) wrote:
> I agree. But it's also not what I said ;-). I said "determined according
> to the value of..."

Ahh, I wouldn't have caught that if you hadn't pointed it out.

> Your details following are a little more precise. But I believe Carlos,
> when he wrote the text about how the limit was calculated, deliberately
> chose to be more vague. I'm okay with including this level of detail 
> in the man page, but, if I recall correctly, Carlos wanted to avoid it.

I'll let Carlos take the call on that.  He asked for detail in the glibc
manual, so I'll propose that level and detail for the manual and see
what he thinks.

Siddhesh

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Siddhesh Poyarekar]

On Wednesday 26 October 2016 03:25 AM, DJ Delorie wrote:
> Siddhesh Poyarekar <siddhesh@sourceware.org> writes:
>> -/*  Non public mallopt parameters.  */
>>  #define M_ARENA_TEST -7
>>  #define M_ARENA_MAX  -8
> 
> Hmmm... if these are now public, do they need to be moved elsewhere?
> Like, malloc.h ?

They're already there in malloc.h.  I'll push a follow-up patch if they
can be removed.

>> +The presence of multiple @code{arenas} allows multiple threads to allocate
>> +memory simultaneously in their own separate arenas, thus improving performance.
> 
> Not quite true - there isn't one arena per thread, there are N arenas
> per M threads.  Probably better to say "... simultaneously in separate
> arenas ..."

OK.

>> +environment variable @code{MALLOC_ARENA_TEST} to the desired value.
>> +@item M_ARENA_MAX
> 
> Missing blank line.
> 
> Otherwise, the content looks OK to me from a technical point of view.

Thanks, I'll fix up formatting comments from Rical and push this.

Siddhesh

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[DJ Delorie]

Siddhesh Poyarekar <siddhesh@sourceware.org> writes:
> -/*  Non public mallopt parameters.  */
>  #define M_ARENA_TEST -7
>  #define M_ARENA_MAX  -8

Hmmm... if these are now public, do they need to be moved elsewhere?
Like, malloc.h ?

> +The presence of multiple @code{arenas} allows multiple threads to allocate
> +memory simultaneously in their own separate arenas, thus improving performance.

Not quite true - there isn't one arena per thread, there are N arenas
per M threads.  Probably better to say "... simultaneously in separate
arenas ..."

> +environment variable @code{MALLOC_ARENA_TEST} to the desired value.
> +@item M_ARENA_MAX

Missing blank line.

Otherwise, the content looks OK to me from a technical point of view.

Re: [PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Rical Jasan]

Hi Siddhesh,

Most of this is formatting or cleanup/improvement of the pre-existing
chunks that were moved, but the comment on M_ARENA_MAX regards content.

On 10/24/2016 07:07 AM, Siddhesh Poyarekar wrote:
> The M_ARENA_* mallopt parameters are in wide use in production to
> control the number of arenas that a long lived process creates and
> hence there is no point in stating that this interface is non-public.
> Document this interface and remove the obsolete comment.
> 
> 	* manual/memory.texi (M_ARENA_TEST): Add documentation.
> 	(M_ARENA_MAX): Likewise.
> 	* malloc/malloc.c: Remove obsolete comment.
> ---
>  malloc/malloc.c    |   1 -
>  manual/memory.texi | 122 +++++++++++++++++++++++++++--------------------------
>  2 files changed, 62 insertions(+), 61 deletions(-)
> 
> diff --git a/malloc/malloc.c b/malloc/malloc.c
> index ef04360..a849901 100644
> --- a/malloc/malloc.c
> +++ b/malloc/malloc.c
> @@ -1718,7 +1718,6 @@ static struct malloc_par mp_ =
>  };
>  
>  
> -/*  Non public mallopt parameters.  */
>  #define M_ARENA_TEST -7
>  #define M_ARENA_MAX  -8
>  
> diff --git a/manual/memory.texi b/manual/memory.texi
> index 6f33455..198a933 100644
> --- a/manual/memory.texi
> +++ b/manual/memory.texi
> @@ -162,6 +162,8 @@ special to @theglibc{} and GNU Compiler.
>  
>  @menu
>  * Memory Allocation and C::     How to get different kinds of allocation in C.
> +* The GNU allocator::		An overview of the GNU @code{malloc}

I think "Allocator" should be capitalized.  The manual is inconsistent
on @subsection capitalization, but the majority of it is (capitalized).
Also, all the other info menu entries visible in the context here are
fully capitalized.

> +				implementation.
>  * Unconstrained Allocation::    The @code{malloc} facility allows fully general
>  		 		 dynamic allocation.
>  * Allocation Debugging::        Finding memory leaks and not freed memory.
> @@ -258,6 +260,43 @@ address of the space.  Then you can use the operators @samp{*} and
>  @}
>  @end smallexample
>  
> +@node The GNU allocator
> +@subsection The GNU allocator

Allocator

> +@cindex gnu allocator
> +
> +The @code{malloc} implementation in @theglibc{} is derived from ptmalloc
> +(pthreads malloc), which in turn is derived from dlmalloc (Doug Lea malloc).
> +This malloc may allocate memory in two different ways depending on their size
> +and certain parameters that may be controlled by users. The most common way is
> +to allocate portions of memory (called chunks) from a large contiguous area of
> +memory and manage these areas to optimize their use and reduce wastage in the
> +form of unusable chunks. Traditionally the system heap was set up to be the one
> +large memory area but @theglibc{} @code{malloc} implementation maintains

This should be "the @glibcadj{} @code{malloc}".

> +multiple such areas to optimize their use in multi-threaded applications.  Each
> +such area is internally referred to as an @code{arena}.

@dfn{arena}

Unless this is a function name (or literal string one would be using in
code, for example), this is simply a term we use to describe the
concept.  On first use, where we define the term, it should have @dfn{},
and otherwise it doesn't need to be stylized at all, such as...

> +
> +As opposed to other versions, the @code{malloc} in @theglibc{} does not round
> +up chunk sizes to powers of two, neither for large nor for small sizes.
> +Neighboring chunks can be coalesced on a @code{free} no matter what their size
> +is.  This makes the implementation suitable for all kinds of allocation
> +patterns without generally incurring high memory waste through fragmentation.
> +The presence of multiple @code{arenas} allows multiple threads to allocate

here (no @code{}).

> +memory simultaneously in their own separate arenas, thus improving performance.
> +
> +The other way of memory allocation is for very large blocks, i.e. much larger
> +than a page. These requests are allocated with @code{mmap} (anonymous or via
> +@code{/dev/zero}). This has the great advantage that these chunks are returned

Should be @file{}.

A reference to mmap would be good.  Something like, "... (anonymous or
via @file{/dev/zero}; @pxref{Memory-mapped I/O})."

> +to the system immediately when they are freed.  Therefore, it cannot happen
> +that a large chunk becomes ``locked'' in between smaller ones and even after
> +calling @code{free} wastes memory.  The size threshold for @code{mmap} to be
> +used is dynamic and gets adjusted according to allocation patterns of the
> +program.  This can also be statically adjusted with @code{mallopt}.  The use of

A reference to mallopt would be good here.

> +@code{mmap} can also be disabled completely.

Should briefly say how and/or give a reference.  I believe mallopt
applies to both, so maybe something like, "@code{mallopt} can be used to
statically adjust the threshold using @code{M_MMAP_THRESHOLD}, and the
use of @code{mmap} can be disabled completely with @code{M_MMAP_MAX};
@pxref{Malloc Tunable Parameters}."

> +
> +A more detailed technical description of the GNU allocator is maintained in
> +@theglibc{} wiki. See

the @glibcadj{}

> +@uref{https://sourceware.org/glibc/wiki/MallocInternals}.
> +
>  @node Unconstrained Allocation
>  @subsection Unconstrained Allocation
>  @cindex unconstrained memory allocation
> @@ -278,8 +317,6 @@ any time (or never).
>  				 bigger or smaller.
>  * Allocating Cleared Space::    Use @code{calloc} to allocate a
>  				 block and clear it.
> -* Efficiency and Malloc::       Efficiency considerations in use of
> -				 these functions.
>  * Aligned Memory Blocks::       Allocating specially aligned memory.
>  * Malloc Tunable Parameters::   Use @code{mallopt} to adjust allocation
>                                   parameters.
> @@ -867,59 +904,6 @@ But in general, it is not guaranteed that @code{calloc} calls
>  @code{malloc}/@code{realloc}/@code{free} outside the C library, it
>  should always define @code{calloc}, too.
>  
> -@node Efficiency and Malloc
> -@subsubsection Efficiency Considerations for @code{malloc}
> -@cindex efficiency and @code{malloc}
> -
> -
> -
> -
> -@ignore
> -
> -@c No longer true, see below instead.
> -To make the best use of @code{malloc}, it helps to know that the GNU
> -version of @code{malloc} always dispenses small amounts of memory in
> -blocks whose sizes are powers of two.  It keeps separate pools for each
> -power of two.  This holds for sizes up to a page size.  Therefore, if
> -you are free to choose the size of a small block in order to make
> -@code{malloc} more efficient, make it a power of two.
> -@c !!! xref getpagesize
> -
> -Once a page is split up for a particular block size, it can't be reused
> -for another size unless all the blocks in it are freed.  In many
> -programs, this is unlikely to happen.  Thus, you can sometimes make a
> -program use memory more efficiently by using blocks of the same size for
> -many different purposes.
> -
> -When you ask for memory blocks of a page or larger, @code{malloc} uses a
> -different strategy; it rounds the size up to a multiple of a page, and
> -it can coalesce and split blocks as needed.
> -
> -The reason for the two strategies is that it is important to allocate
> -and free small blocks as fast as possible, but speed is less important
> -for a large block since the program normally spends a fair amount of
> -time using it.  Also, large blocks are normally fewer in number.
> -Therefore, for large blocks, it makes sense to use a method which takes
> -more time to minimize the wasted space.
> -
> -@end ignore
> -
> -As opposed to other versions, the @code{malloc} in @theglibc{}
> -does not round up block sizes to powers of two, neither for large nor
> -for small sizes.  Neighboring chunks can be coalesced on a @code{free}
> -no matter what their size is.  This makes the implementation suitable
> -for all kinds of allocation patterns without generally incurring high
> -memory waste through fragmentation.
> -
> -Very large blocks (much larger than a page) are allocated with
> -@code{mmap} (anonymous or via @code{/dev/zero}) by this implementation.
> -This has the great advantage that these chunks are returned to the
> -system immediately when they are freed.  Therefore, it cannot happen
> -that a large chunk becomes ``locked'' in between smaller ones and even
> -after calling @code{free} wastes memory.  The size threshold for
> -@code{mmap} to be used can be adjusted with @code{mallopt}.  The use of
> -@code{mmap} can also be disabled completely.
> -
>  @node Aligned Memory Blocks
>  @subsubsection Allocating Aligned Memory Blocks
>  
> @@ -1105,10 +1089,6 @@ parameter to be set, and @var{value} the new value to be set.  Possible
>  choices for @var{param}, as defined in @file{malloc.h}, are:
>  
>  @table @code
> -@comment TODO: @item M_ARENA_MAX
> -@comment       - Document ARENA_MAX env var.
> -@comment TODO: @item M_ARENA_TEST
> -@comment       - Document ARENA_TEST env var.
>  @comment TODO: @item M_CHECK_ACTION
>  @item M_MMAP_MAX
>  The maximum number of chunks to allocate with @code{mmap}.  Setting this
> @@ -1169,6 +1149,28 @@ value is set statically to the provided input.
>  
>  This parameter can also be set for the process at startup by setting the
>  environment variable @code{MALLOC_TRIM_THRESHOLD_} to the desired value.
> +
> +@item M_ARENA_TEST
> +This parameter specifies the number of arenas that can be created before the
> +test on the limit to the number of arenas is conducted. The value is ignored if
> +@code{M_ARENA_MAX} is set.
> +
> +The default value of this parameter is 2 on 32-bit systems and 8 on 64-bit
> +systems.
> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_ARENA_TEST} to the desired value.
> +@item M_ARENA_MAX
> +This parameter sets the number of arenas to use regardless of the number of
> +cores in the system.
> +
> +The default value of this tunable is @code{0}, meaning that the limit on the
> +number of arenas is determined by the number of CPU cores online. For 32-bit
> +systems the limit is twice the number of cores online and on 64-bit systems, it
> +is eight times the number of cores online.

Even though I had followed the thread, I immediately jumped to the same
confusion with M_ARENA_TEST that was resolved in [1].  Explicitly
stating here that 2 and 8 are not derived from M_ARENA_TEST defaults
might save a lot of general confusion down the road.

> +
> +This parameter can also be set for the process at startup by setting the
> +environment variable @code{MALLOC_ARENA_MAX} to the desired value.
>  @end table
>  
>  @end deftypefun
> @@ -1511,7 +1513,7 @@ This is the total size of memory allocated with @code{sbrk} by
>  This is the number of chunks not in use.  (The memory allocator
>  internally gets chunks of memory from the operating system, and then
>  carves them up to satisfy individual @code{malloc} requests; see
> -@ref{Efficiency and Malloc}.)
> +@ref{The GNU allocator}.)

Allocator

Also, making this an @pxref{} now would be nice, since this renders
wrong in info as-is.

>  
>  @item int smblks
>  This field is unused.
> 

Rical

[1] https://sourceware.org/ml/libc-alpha/2016-10/msg00310.html

[PATCH 2/2] Document the M_ARENA_* mallopt parameters

[Siddhesh Poyarekar]

The M_ARENA_* mallopt parameters are in wide use in production to
control the number of arenas that a long lived process creates and
hence there is no point in stating that this interface is non-public.
Document this interface and remove the obsolete comment.

	* manual/memory.texi (M_ARENA_TEST): Add documentation.
	(M_ARENA_MAX): Likewise.
	* malloc/malloc.c: Remove obsolete comment.
---
 malloc/malloc.c    |   1 -
 manual/memory.texi | 122 +++++++++++++++++++++++++++--------------------------
 2 files changed, 62 insertions(+), 61 deletions(-)

diff --git a/malloc/malloc.c b/malloc/malloc.c
index ef04360..a849901 100644
--- a/malloc/malloc.c
+++ b/malloc/malloc.c
@@ -1718,7 +1718,6 @@ static struct malloc_par mp_ =
 };
 
 
-/*  Non public mallopt parameters.  */
 #define M_ARENA_TEST -7
 #define M_ARENA_MAX  -8
 
diff --git a/manual/memory.texi b/manual/memory.texi
index 6f33455..198a933 100644
--- a/manual/memory.texi
+++ b/manual/memory.texi
@@ -162,6 +162,8 @@ special to @theglibc{} and GNU Compiler.
 
 @menu
 * Memory Allocation and C::     How to get different kinds of allocation in C.
+* The GNU allocator::		An overview of the GNU @code{malloc}
+				implementation.
 * Unconstrained Allocation::    The @code{malloc} facility allows fully general
 		 		 dynamic allocation.
 * Allocation Debugging::        Finding memory leaks and not freed memory.
@@ -258,6 +260,43 @@ address of the space.  Then you can use the operators @samp{*} and
 @}
 @end smallexample
 
+@node The GNU allocator
+@subsection The GNU allocator
+@cindex gnu allocator
+
+The @code{malloc} implementation in @theglibc{} is derived from ptmalloc
+(pthreads malloc), which in turn is derived from dlmalloc (Doug Lea malloc).
+This malloc may allocate memory in two different ways depending on their size
+and certain parameters that may be controlled by users. The most common way is
+to allocate portions of memory (called chunks) from a large contiguous area of
+memory and manage these areas to optimize their use and reduce wastage in the
+form of unusable chunks. Traditionally the system heap was set up to be the one
+large memory area but @theglibc{} @code{malloc} implementation maintains
+multiple such areas to optimize their use in multi-threaded applications.  Each
+such area is internally referred to as an @code{arena}.
+
+As opposed to other versions, the @code{malloc} in @theglibc{} does not round
+up chunk sizes to powers of two, neither for large nor for small sizes.
+Neighboring chunks can be coalesced on a @code{free} no matter what their size
+is.  This makes the implementation suitable for all kinds of allocation
+patterns without generally incurring high memory waste through fragmentation.
+The presence of multiple @code{arenas} allows multiple threads to allocate
+memory simultaneously in their own separate arenas, thus improving performance.
+
+The other way of memory allocation is for very large blocks, i.e. much larger
+than a page. These requests are allocated with @code{mmap} (anonymous or via
+@code{/dev/zero}). This has the great advantage that these chunks are returned
+to the system immediately when they are freed.  Therefore, it cannot happen
+that a large chunk becomes ``locked'' in between smaller ones and even after
+calling @code{free} wastes memory.  The size threshold for @code{mmap} to be
+used is dynamic and gets adjusted according to allocation patterns of the
+program.  This can also be statically adjusted with @code{mallopt}.  The use of
+@code{mmap} can also be disabled completely.
+
+A more detailed technical description of the GNU allocator is maintained in
+@theglibc{} wiki. See
+@uref{https://sourceware.org/glibc/wiki/MallocInternals}.
+
 @node Unconstrained Allocation
 @subsection Unconstrained Allocation
 @cindex unconstrained memory allocation
@@ -278,8 +317,6 @@ any time (or never).
 				 bigger or smaller.
 * Allocating Cleared Space::    Use @code{calloc} to allocate a
 				 block and clear it.
-* Efficiency and Malloc::       Efficiency considerations in use of
-				 these functions.
 * Aligned Memory Blocks::       Allocating specially aligned memory.
 * Malloc Tunable Parameters::   Use @code{mallopt} to adjust allocation
                                  parameters.
@@ -867,59 +904,6 @@ But in general, it is not guaranteed that @code{calloc} calls
 @code{malloc}/@code{realloc}/@code{free} outside the C library, it
 should always define @code{calloc}, too.
 
-@node Efficiency and Malloc
-@subsubsection Efficiency Considerations for @code{malloc}
-@cindex efficiency and @code{malloc}
-
-
-
-
-@ignore
-
-@c No longer true, see below instead.
-To make the best use of @code{malloc}, it helps to know that the GNU
-version of @code{malloc} always dispenses small amounts of memory in
-blocks whose sizes are powers of two.  It keeps separate pools for each
-power of two.  This holds for sizes up to a page size.  Therefore, if
-you are free to choose the size of a small block in order to make
-@code{malloc} more efficient, make it a power of two.
-@c !!! xref getpagesize
-
-Once a page is split up for a particular block size, it can't be reused
-for another size unless all the blocks in it are freed.  In many
-programs, this is unlikely to happen.  Thus, you can sometimes make a
-program use memory more efficiently by using blocks of the same size for
-many different purposes.
-
-When you ask for memory blocks of a page or larger, @code{malloc} uses a
-different strategy; it rounds the size up to a multiple of a page, and
-it can coalesce and split blocks as needed.
-
-The reason for the two strategies is that it is important to allocate
-and free small blocks as fast as possible, but speed is less important
-for a large block since the program normally spends a fair amount of
-time using it.  Also, large blocks are normally fewer in number.
-Therefore, for large blocks, it makes sense to use a method which takes
-more time to minimize the wasted space.
-
-@end ignore
-
-As opposed to other versions, the @code{malloc} in @theglibc{}
-does not round up block sizes to powers of two, neither for large nor
-for small sizes.  Neighboring chunks can be coalesced on a @code{free}
-no matter what their size is.  This makes the implementation suitable
-for all kinds of allocation patterns without generally incurring high
-memory waste through fragmentation.
-
-Very large blocks (much larger than a page) are allocated with
-@code{mmap} (anonymous or via @code{/dev/zero}) by this implementation.
-This has the great advantage that these chunks are returned to the
-system immediately when they are freed.  Therefore, it cannot happen
-that a large chunk becomes ``locked'' in between smaller ones and even
-after calling @code{free} wastes memory.  The size threshold for
-@code{mmap} to be used can be adjusted with @code{mallopt}.  The use of
-@code{mmap} can also be disabled completely.
-
 @node Aligned Memory Blocks
 @subsubsection Allocating Aligned Memory Blocks
 
@@ -1105,10 +1089,6 @@ parameter to be set, and @var{value} the new value to be set.  Possible
 choices for @var{param}, as defined in @file{malloc.h}, are:
 
 @table @code
-@comment TODO: @item M_ARENA_MAX
-@comment       - Document ARENA_MAX env var.
-@comment TODO: @item M_ARENA_TEST
-@comment       - Document ARENA_TEST env var.
 @comment TODO: @item M_CHECK_ACTION
 @item M_MMAP_MAX
 The maximum number of chunks to allocate with @code{mmap}.  Setting this
@@ -1169,6 +1149,28 @@ value is set statically to the provided input.
 
 This parameter can also be set for the process at startup by setting the
 environment variable @code{MALLOC_TRIM_THRESHOLD_} to the desired value.
+
+@item M_ARENA_TEST
+This parameter specifies the number of arenas that can be created before the
+test on the limit to the number of arenas is conducted. The value is ignored if
+@code{M_ARENA_MAX} is set.
+
+The default value of this parameter is 2 on 32-bit systems and 8 on 64-bit
+systems.
+
+This parameter can also be set for the process at startup by setting the
+environment variable @code{MALLOC_ARENA_TEST} to the desired value.
+@item M_ARENA_MAX
+This parameter sets the number of arenas to use regardless of the number of
+cores in the system.
+
+The default value of this tunable is @code{0}, meaning that the limit on the
+number of arenas is determined by the number of CPU cores online. For 32-bit
+systems the limit is twice the number of cores online and on 64-bit systems, it
+is eight times the number of cores online.
+
+This parameter can also be set for the process at startup by setting the
+environment variable @code{MALLOC_ARENA_MAX} to the desired value.
 @end table
 
 @end deftypefun
@@ -1511,7 +1513,7 @@ This is the total size of memory allocated with @code{sbrk} by
 This is the number of chunks not in use.  (The memory allocator
 internally gets chunks of memory from the operating system, and then
 carves them up to satisfy individual @code{malloc} requests; see
-@ref{Efficiency and Malloc}.)
+@ref{The GNU allocator}.)
 
 @item int smblks
 This field is unused.
-- 
2.7.4