[RFC,v5,1/6] eal: add static per-lcore memory allocation facility

Message ID 20240228100928.524277-2-mattias.ronnblom@ericsson.com (mailing list archive)
State Superseded, archived
Delegated to: David Marchand
Headers
Series Lcore variables |

Checks

Context Check Description
ci/checkpatch warning coding style issues

Commit Message

Mattias Rönnblom Feb. 28, 2024, 10:09 a.m. UTC
Introduce DPDK per-lcore id variables, or lcore variables for short.

An lcore variable has one value for every current and future lcore
id-equipped thread.

The primary <rte_lcore_var.h> use case is for statically allocating
small chunks of often-used data, which is related logically, but where
there are performance benefits to reap from having updates being local
to an lcore.

Lcore variables are similar to thread-local storage (TLS, e.g., C11
_Thread_local), but decoupling the values' life time with that of the
threads.

Lcore variables are also similar in terms of functionality provided by
FreeBSD kernel's DPCPU_*() family of macros and the associated
build-time machinery. DPCPU uses linker scripts, which effectively
prevents the reuse of its, otherwise seemingly viable, approach.

The currently-prevailing way to solve the same problem as lcore
variables is to keep a module's per-lcore data as RTE_MAX_LCORE-sized
array of cache-aligned, RTE_CACHE_GUARDed structs. The benefit of
lcore variables over this approach is that data related to the same
lcore now is close (spatially, in memory), rather than data used by
the same module, which in turn avoid excessive use of padding,
polluting caches with unused data.

RFC v5:
 * In Doxygen, consistenly use @<cmd> (and not \<cmd>).
 * The RTE_LCORE_VAR_GET() and SET() convience access macros
   covered an uncommon use case, where the lcore value is of a
   primitive type, rather than a struct, and is thus eliminated
   from the API. (Morten Brørup)
 * In the wake up GET()/SET() removeal, rename RTE_LCORE_VAR_PTR()
   RTE_LCORE_VAR_VALUE().
 * The underscores are removed from __rte_lcore_var_lcore_ptr() to
   signal that this function is a part of the public API.
 * Macro arguments are documented.

RFV v4:
 * Replace large static array with libc heap-allocated memory. One
   implication of this change is there no longer exists a fixed upper
   bound for the total amount of memory used by lcore variables.
   RTE_MAX_LCORE_VAR has changed meaning, and now represent the
   maximum size of any individual lcore variable value.
 * Fix issues in example. (Morten Brørup)
 * Improve access macro type checking. (Morten Brørup)
 * Refer to the lcore variable handle as "handle" and not "name" in
   various macros.
 * Document lack of thread safety in rte_lcore_var_alloc().
 * Provide API-level assurance the lcore variable handle is
   always non-NULL, to all applications to use NULL to mean
   "not yet allocated".
 * Note zero-sized allocations are not allowed.
 * Give API-level guarantee the lcore variable values are zeroed.

RFC v3:
 * Replace use of GCC-specific alignof(<expression>) with alignof(<type>).
 * Update example to reflect FOREACH macro name change (in RFC v2).

RFC v2:
 * Use alignof to derive alignment requirements. (Morten Brørup)
 * Change name of FOREACH to make it distinct from <rte_lcore.h>'s
   *per-EAL-thread* RTE_LCORE_FOREACH(). (Morten Brørup)
 * Allow user-specified alignment, but limit max to cache line size.

Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
Acked-by: Morten Brørup <mb@smartsharesystems.com>
---
 config/rte_config.h                   |   1 +
 doc/api/doxy-api-index.md             |   1 +
 lib/eal/common/eal_common_lcore_var.c |  68 +++++
 lib/eal/common/meson.build            |   1 +
 lib/eal/include/meson.build           |   1 +
 lib/eal/include/rte_lcore_var.h       | 368 ++++++++++++++++++++++++++
 lib/eal/version.map                   |   4 +
 7 files changed, 444 insertions(+)
 create mode 100644 lib/eal/common/eal_common_lcore_var.c
 create mode 100644 lib/eal/include/rte_lcore_var.h
  

Comments

Konstantin Ananyev March 19, 2024, 12:52 p.m. UTC | #1
Hi Mattias,
> Introduce DPDK per-lcore id variables, or lcore variables for short.
> 
> An lcore variable has one value for every current and future lcore
> id-equipped thread.
> 
> The primary <rte_lcore_var.h> use case is for statically allocating
> small chunks of often-used data, which is related logically, but where
> there are performance benefits to reap from having updates being local
> to an lcore.
> 
> Lcore variables are similar to thread-local storage (TLS, e.g., C11
> _Thread_local), but decoupling the values' life time with that of the
> threads.
> 
> Lcore variables are also similar in terms of functionality provided by
> FreeBSD kernel's DPCPU_*() family of macros and the associated
> build-time machinery. DPCPU uses linker scripts, which effectively
> prevents the reuse of its, otherwise seemingly viable, approach.
> 
> The currently-prevailing way to solve the same problem as lcore
> variables is to keep a module's per-lcore data as RTE_MAX_LCORE-sized
> array of cache-aligned, RTE_CACHE_GUARDed structs. The benefit of
> lcore variables over this approach is that data related to the same
> lcore now is close (spatially, in memory), rather than data used by
> the same module, which in turn avoid excessive use of padding,
> polluting caches with unused data.

Thanks for the RFC, very interesting one.
Few comments/questions below. 

 
> RFC v5:
>  * In Doxygen, consistenly use @<cmd> (and not \<cmd>).
>  * The RTE_LCORE_VAR_GET() and SET() convience access macros
>    covered an uncommon use case, where the lcore value is of a
>    primitive type, rather than a struct, and is thus eliminated
>    from the API. (Morten Brørup)
>  * In the wake up GET()/SET() removeal, rename RTE_LCORE_VAR_PTR()
>    RTE_LCORE_VAR_VALUE().
>  * The underscores are removed from __rte_lcore_var_lcore_ptr() to
>    signal that this function is a part of the public API.
>  * Macro arguments are documented.
> 
> RFV v4:
>  * Replace large static array with libc heap-allocated memory. One
>    implication of this change is there no longer exists a fixed upper
>    bound for the total amount of memory used by lcore variables.
>    RTE_MAX_LCORE_VAR has changed meaning, and now represent the
>    maximum size of any individual lcore variable value.
>  * Fix issues in example. (Morten Brørup)
>  * Improve access macro type checking. (Morten Brørup)
>  * Refer to the lcore variable handle as "handle" and not "name" in
>    various macros.
>  * Document lack of thread safety in rte_lcore_var_alloc().
>  * Provide API-level assurance the lcore variable handle is
>    always non-NULL, to all applications to use NULL to mean
>    "not yet allocated".
>  * Note zero-sized allocations are not allowed.
>  * Give API-level guarantee the lcore variable values are zeroed.
> 
> RFC v3:
>  * Replace use of GCC-specific alignof(<expression>) with alignof(<type>).
>  * Update example to reflect FOREACH macro name change (in RFC v2).
> 
> RFC v2:
>  * Use alignof to derive alignment requirements. (Morten Brørup)
>  * Change name of FOREACH to make it distinct from <rte_lcore.h>'s
>    *per-EAL-thread* RTE_LCORE_FOREACH(). (Morten Brørup)
>  * Allow user-specified alignment, but limit max to cache line size.
> 
> Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
> Acked-by: Morten Brørup <mb@smartsharesystems.com>
> ---
>  config/rte_config.h                   |   1 +
>  doc/api/doxy-api-index.md             |   1 +
>  lib/eal/common/eal_common_lcore_var.c |  68 +++++
>  lib/eal/common/meson.build            |   1 +
>  lib/eal/include/meson.build           |   1 +
>  lib/eal/include/rte_lcore_var.h       | 368 ++++++++++++++++++++++++++
>  lib/eal/version.map                   |   4 +
>  7 files changed, 444 insertions(+)
>  create mode 100644 lib/eal/common/eal_common_lcore_var.c
>  create mode 100644 lib/eal/include/rte_lcore_var.h
> 
> diff --git a/config/rte_config.h b/config/rte_config.h
> index d743a5c3d3..0dac33d3b9 100644
> --- a/config/rte_config.h
> +++ b/config/rte_config.h
> @@ -41,6 +41,7 @@
>  /* EAL defines */
>  #define RTE_CACHE_GUARD_LINES 1
>  #define RTE_MAX_HEAPS 32
> +#define RTE_MAX_LCORE_VAR 1048576
>  #define RTE_MAX_MEMSEG_LISTS 128
>  #define RTE_MAX_MEMSEG_PER_LIST 8192
>  #define RTE_MAX_MEM_MB_PER_LIST 32768
> diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
> index 8c1eb8fafa..a3b8391570 100644
> --- a/doc/api/doxy-api-index.md
> +++ b/doc/api/doxy-api-index.md
> @@ -99,6 +99,7 @@ The public API headers are grouped by topics:
>    [interrupts](@ref rte_interrupts.h),
>    [launch](@ref rte_launch.h),
>    [lcore](@ref rte_lcore.h),
> +  [lcore-varible](@ref rte_lcore_var.h),
>    [per-lcore](@ref rte_per_lcore.h),
>    [service cores](@ref rte_service.h),
>    [keepalive](@ref rte_keepalive.h),
> diff --git a/lib/eal/common/eal_common_lcore_var.c b/lib/eal/common/eal_common_lcore_var.c
> new file mode 100644
> index 0000000000..5c353ebd46
> --- /dev/null
> +++ b/lib/eal/common/eal_common_lcore_var.c
> @@ -0,0 +1,68 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2024 Ericsson AB
> + */
> +
> +#include <inttypes.h>
> +
> +#include <rte_common.h>
> +#include <rte_debug.h>
> +#include <rte_log.h>
> +
> +#include <rte_lcore_var.h>
> +
> +#include "eal_private.h"
> +
> +#define LCORE_BUFFER_SIZE (RTE_MAX_LCORE_VAR * RTE_MAX_LCORE)
> +
> +static void *lcore_buffer;
> +static size_t offset = RTE_MAX_LCORE_VAR;
> +
> +static void *
> +lcore_var_alloc(size_t size, size_t align)
> +{
> +	void *handle;
> +	void *value;
> +
> +	offset = RTE_ALIGN_CEIL(offset, align);
> +
> +	if (offset + size > RTE_MAX_LCORE_VAR) {
> +		lcore_buffer = aligned_alloc(RTE_CACHE_LINE_SIZE,
> +					     LCORE_BUFFER_SIZE);

Hmm... do I get it right: if offset is <= then RTE_MAX_LCORE_VAR,  and  offset + size > RTE_MAX_LCORE_VAR
we simply overwrite lcore_buffer with newly allocated buffer of the same size?
I understand that you expect it just never to happen (total size of all lcore vars never exceed 1MB), but still
I think we need to handle it in a some better way then just ignoring such possibility...
Might be RTE_VERIFY() at least?   

As a more generic question - do we need to support LCORE_VAR for dlopen()s that could happen after rte_eal_init()
is called and LCORE threads were created?
Because, if no, then we probably can make this construction much more flexible:
one buffer per LCORE, allocate on demand, etc. 

> +		RTE_VERIFY(lcore_buffer != NULL);
> +
> +		offset = 0;
> +	}
> +
> +	handle = RTE_PTR_ADD(lcore_buffer, offset);
> +
> +	offset += size;
> +
> +	RTE_LCORE_VAR_FOREACH_VALUE(value, handle)
> +		memset(value, 0, size);
> +
> +	EAL_LOG(DEBUG, "Allocated %"PRIuPTR" bytes of per-lcore data with a "
> +		"%"PRIuPTR"-byte alignment", size, align);
> +
> +	return handle;
> +}
> +
> +void *
> +rte_lcore_var_alloc(size_t size, size_t align)
> +{
> +	/* Having the per-lcore buffer size aligned on cache lines
> +	 * assures as well as having the base pointer aligned on cache
> +	 * size assures that aligned offsets also translate to alipgned
> +	 * pointers across all values.
> +	 */
> +	RTE_BUILD_BUG_ON(RTE_MAX_LCORE_VAR % RTE_CACHE_LINE_SIZE != 0);
> +	RTE_ASSERT(align <= RTE_CACHE_LINE_SIZE);
> +	RTE_ASSERT(size <= RTE_MAX_LCORE_VAR);
> +
> +	/* '0' means asking for worst-case alignment requirements */
> +	if (align == 0)
> +		align = alignof(max_align_t);
> +
> +	RTE_ASSERT(rte_is_power_of_2(align));
> +
> +	return lcore_var_alloc(size, align);
> +}
  
Mattias Rönnblom March 20, 2024, 10:24 a.m. UTC | #2
On 2024-03-19 13:52, Konstantin Ananyev wrote:
> 
> Hi Mattias,
>> Introduce DPDK per-lcore id variables, or lcore variables for short.
>>
>> An lcore variable has one value for every current and future lcore
>> id-equipped thread.
>>
>> The primary <rte_lcore_var.h> use case is for statically allocating
>> small chunks of often-used data, which is related logically, but where
>> there are performance benefits to reap from having updates being local
>> to an lcore.
>>
>> Lcore variables are similar to thread-local storage (TLS, e.g., C11
>> _Thread_local), but decoupling the values' life time with that of the
>> threads.
>>
>> Lcore variables are also similar in terms of functionality provided by
>> FreeBSD kernel's DPCPU_*() family of macros and the associated
>> build-time machinery. DPCPU uses linker scripts, which effectively
>> prevents the reuse of its, otherwise seemingly viable, approach.
>>
>> The currently-prevailing way to solve the same problem as lcore
>> variables is to keep a module's per-lcore data as RTE_MAX_LCORE-sized
>> array of cache-aligned, RTE_CACHE_GUARDed structs. The benefit of
>> lcore variables over this approach is that data related to the same
>> lcore now is close (spatially, in memory), rather than data used by
>> the same module, which in turn avoid excessive use of padding,
>> polluting caches with unused data.
> 
> Thanks for the RFC, very interesting one.
> Few comments/questions below.
> 
>   
>> RFC v5:
>>   * In Doxygen, consistenly use @<cmd> (and not \<cmd>).
>>   * The RTE_LCORE_VAR_GET() and SET() convience access macros
>>     covered an uncommon use case, where the lcore value is of a
>>     primitive type, rather than a struct, and is thus eliminated
>>     from the API. (Morten Brørup)
>>   * In the wake up GET()/SET() removeal, rename RTE_LCORE_VAR_PTR()
>>     RTE_LCORE_VAR_VALUE().
>>   * The underscores are removed from __rte_lcore_var_lcore_ptr() to
>>     signal that this function is a part of the public API.
>>   * Macro arguments are documented.
>>
>> RFV v4:
>>   * Replace large static array with libc heap-allocated memory. One
>>     implication of this change is there no longer exists a fixed upper
>>     bound for the total amount of memory used by lcore variables.
>>     RTE_MAX_LCORE_VAR has changed meaning, and now represent the
>>     maximum size of any individual lcore variable value.
>>   * Fix issues in example. (Morten Brørup)
>>   * Improve access macro type checking. (Morten Brørup)
>>   * Refer to the lcore variable handle as "handle" and not "name" in
>>     various macros.
>>   * Document lack of thread safety in rte_lcore_var_alloc().
>>   * Provide API-level assurance the lcore variable handle is
>>     always non-NULL, to all applications to use NULL to mean
>>     "not yet allocated".
>>   * Note zero-sized allocations are not allowed.
>>   * Give API-level guarantee the lcore variable values are zeroed.
>>
>> RFC v3:
>>   * Replace use of GCC-specific alignof(<expression>) with alignof(<type>).
>>   * Update example to reflect FOREACH macro name change (in RFC v2).
>>
>> RFC v2:
>>   * Use alignof to derive alignment requirements. (Morten Brørup)
>>   * Change name of FOREACH to make it distinct from <rte_lcore.h>'s
>>     *per-EAL-thread* RTE_LCORE_FOREACH(). (Morten Brørup)
>>   * Allow user-specified alignment, but limit max to cache line size.
>>
>> Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
>> Acked-by: Morten Brørup <mb@smartsharesystems.com>
>> ---
>>   config/rte_config.h                   |   1 +
>>   doc/api/doxy-api-index.md             |   1 +
>>   lib/eal/common/eal_common_lcore_var.c |  68 +++++
>>   lib/eal/common/meson.build            |   1 +
>>   lib/eal/include/meson.build           |   1 +
>>   lib/eal/include/rte_lcore_var.h       | 368 ++++++++++++++++++++++++++
>>   lib/eal/version.map                   |   4 +
>>   7 files changed, 444 insertions(+)
>>   create mode 100644 lib/eal/common/eal_common_lcore_var.c
>>   create mode 100644 lib/eal/include/rte_lcore_var.h
>>
>> diff --git a/config/rte_config.h b/config/rte_config.h
>> index d743a5c3d3..0dac33d3b9 100644
>> --- a/config/rte_config.h
>> +++ b/config/rte_config.h
>> @@ -41,6 +41,7 @@
>>   /* EAL defines */
>>   #define RTE_CACHE_GUARD_LINES 1
>>   #define RTE_MAX_HEAPS 32
>> +#define RTE_MAX_LCORE_VAR 1048576
>>   #define RTE_MAX_MEMSEG_LISTS 128
>>   #define RTE_MAX_MEMSEG_PER_LIST 8192
>>   #define RTE_MAX_MEM_MB_PER_LIST 32768
>> diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
>> index 8c1eb8fafa..a3b8391570 100644
>> --- a/doc/api/doxy-api-index.md
>> +++ b/doc/api/doxy-api-index.md
>> @@ -99,6 +99,7 @@ The public API headers are grouped by topics:
>>     [interrupts](@ref rte_interrupts.h),
>>     [launch](@ref rte_launch.h),
>>     [lcore](@ref rte_lcore.h),
>> +  [lcore-varible](@ref rte_lcore_var.h),
>>     [per-lcore](@ref rte_per_lcore.h),
>>     [service cores](@ref rte_service.h),
>>     [keepalive](@ref rte_keepalive.h),
>> diff --git a/lib/eal/common/eal_common_lcore_var.c b/lib/eal/common/eal_common_lcore_var.c
>> new file mode 100644
>> index 0000000000..5c353ebd46
>> --- /dev/null
>> +++ b/lib/eal/common/eal_common_lcore_var.c
>> @@ -0,0 +1,68 @@
>> +/* SPDX-License-Identifier: BSD-3-Clause
>> + * Copyright(c) 2024 Ericsson AB
>> + */
>> +
>> +#include <inttypes.h>
>> +
>> +#include <rte_common.h>
>> +#include <rte_debug.h>
>> +#include <rte_log.h>
>> +
>> +#include <rte_lcore_var.h>
>> +
>> +#include "eal_private.h"
>> +
>> +#define LCORE_BUFFER_SIZE (RTE_MAX_LCORE_VAR * RTE_MAX_LCORE)
>> +
>> +static void *lcore_buffer;
>> +static size_t offset = RTE_MAX_LCORE_VAR;
>> +
>> +static void *
>> +lcore_var_alloc(size_t size, size_t align)
>> +{
>> +	void *handle;
>> +	void *value;
>> +
>> +	offset = RTE_ALIGN_CEIL(offset, align);
>> +
>> +	if (offset + size > RTE_MAX_LCORE_VAR) {
>> +		lcore_buffer = aligned_alloc(RTE_CACHE_LINE_SIZE,
>> +					     LCORE_BUFFER_SIZE);
> 
> Hmm... do I get it right: if offset is <= then RTE_MAX_LCORE_VAR,  and  offset + size > RTE_MAX_LCORE_VAR
> we simply overwrite lcore_buffer with newly allocated buffer of the same size?

No, it's just the pointer that is overwritten. The old buffer will 
remain in memory.

> I understand that you expect it just never to happen (total size of all lcore vars never exceed 1MB), but still
> I think we need to handle it in a some better way then just ignoring such possibility...
> Might be RTE_VERIFY() at least?
> 

In this revision of the patch set, RTE_MAX_LCORE_VAR does not represent 
an upper bound for the sum of all lcore variables' size, but rather only 
the maximum size of a single lcore variable.

Variable alignment and size constraints are RTE_ASSERT()ed at the point 
of allocation. One could argue they should be RTE_VERIFY()-ed instead, 
since there aren't any performance constraints.

> As a more generic question - do we need to support LCORE_VAR for dlopen()s that could happen after rte_eal_init()
> is called and LCORE threads were created?

Yes, allocations after rte_eal_init() (caused by dlopen() or otherwise) 
must be allowed imo, and are allowed. Otherwise applications sitting on 
top of DPDK can't use this facility.

> Because, if no, then we probably can make this construction much more flexible:
> one buffer per LCORE, allocate on demand, etc.
> 

On-demand allocations are already supported, but one can't do free(). 
That's why I've called what this module provide "static allocation", 
while it may be more appropriately described as "dynamic allocation 
without deallocation".

"True" dynamic memory allocation of per-lcore memory would be very 
useful, but is an entirely different beast in terms of complexity and 
(if to be usable in the packet processing fast path) performance 
requirements.

"True" dynamic memory allocation would also result in something less 
compact (at least if you use the usual pattern with a per-object heap 
header).

>> +		RTE_VERIFY(lcore_buffer != NULL);
>> +
>> +		offset = 0;
>> +	}
>> +
>> +	handle = RTE_PTR_ADD(lcore_buffer, offset);
>> +
>> +	offset += size;
>> +
>> +	RTE_LCORE_VAR_FOREACH_VALUE(value, handle)
>> +		memset(value, 0, size);
>> +
>> +	EAL_LOG(DEBUG, "Allocated %"PRIuPTR" bytes of per-lcore data with a "
>> +		"%"PRIuPTR"-byte alignment", size, align);
>> +
>> +	return handle;
>> +}
>> +
>> +void *
>> +rte_lcore_var_alloc(size_t size, size_t align)
>> +{
>> +	/* Having the per-lcore buffer size aligned on cache lines
>> +	 * assures as well as having the base pointer aligned on cache
>> +	 * size assures that aligned offsets also translate to alipgned
>> +	 * pointers across all values.
>> +	 */
>> +	RTE_BUILD_BUG_ON(RTE_MAX_LCORE_VAR % RTE_CACHE_LINE_SIZE != 0);
>> +	RTE_ASSERT(align <= RTE_CACHE_LINE_SIZE);
>> +	RTE_ASSERT(size <= RTE_MAX_LCORE_VAR);
>> +
>> +	/* '0' means asking for worst-case alignment requirements */
>> +	if (align == 0)
>> +		align = alignof(max_align_t);
>> +
>> +	RTE_ASSERT(rte_is_power_of_2(align));
>> +
>> +	return lcore_var_alloc(size, align);
>> +}
  
Konstantin Ananyev March 20, 2024, 2:18 p.m. UTC | #3
> >> Introduce DPDK per-lcore id variables, or lcore variables for short.
> >>
> >> An lcore variable has one value for every current and future lcore
> >> id-equipped thread.
> >>
> >> The primary <rte_lcore_var.h> use case is for statically allocating
> >> small chunks of often-used data, which is related logically, but where
> >> there are performance benefits to reap from having updates being local
> >> to an lcore.
> >>
> >> Lcore variables are similar to thread-local storage (TLS, e.g., C11
> >> _Thread_local), but decoupling the values' life time with that of the
> >> threads.
> >>
> >> Lcore variables are also similar in terms of functionality provided by
> >> FreeBSD kernel's DPCPU_*() family of macros and the associated
> >> build-time machinery. DPCPU uses linker scripts, which effectively
> >> prevents the reuse of its, otherwise seemingly viable, approach.
> >>
> >> The currently-prevailing way to solve the same problem as lcore
> >> variables is to keep a module's per-lcore data as RTE_MAX_LCORE-sized
> >> array of cache-aligned, RTE_CACHE_GUARDed structs. The benefit of
> >> lcore variables over this approach is that data related to the same
> >> lcore now is close (spatially, in memory), rather than data used by
> >> the same module, which in turn avoid excessive use of padding,
> >> polluting caches with unused data.
> >
> > Thanks for the RFC, very interesting one.
> > Few comments/questions below.
> >
> >
> >> RFC v5:
> >>   * In Doxygen, consistenly use @<cmd> (and not \<cmd>).
> >>   * The RTE_LCORE_VAR_GET() and SET() convience access macros
> >>     covered an uncommon use case, where the lcore value is of a
> >>     primitive type, rather than a struct, and is thus eliminated
> >>     from the API. (Morten Brørup)
> >>   * In the wake up GET()/SET() removeal, rename RTE_LCORE_VAR_PTR()
> >>     RTE_LCORE_VAR_VALUE().
> >>   * The underscores are removed from __rte_lcore_var_lcore_ptr() to
> >>     signal that this function is a part of the public API.
> >>   * Macro arguments are documented.
> >>
> >> RFV v4:
> >>   * Replace large static array with libc heap-allocated memory. One
> >>     implication of this change is there no longer exists a fixed upper
> >>     bound for the total amount of memory used by lcore variables.
> >>     RTE_MAX_LCORE_VAR has changed meaning, and now represent the
> >>     maximum size of any individual lcore variable value.
> >>   * Fix issues in example. (Morten Brørup)
> >>   * Improve access macro type checking. (Morten Brørup)
> >>   * Refer to the lcore variable handle as "handle" and not "name" in
> >>     various macros.
> >>   * Document lack of thread safety in rte_lcore_var_alloc().
> >>   * Provide API-level assurance the lcore variable handle is
> >>     always non-NULL, to all applications to use NULL to mean
> >>     "not yet allocated".
> >>   * Note zero-sized allocations are not allowed.
> >>   * Give API-level guarantee the lcore variable values are zeroed.
> >>
> >> RFC v3:
> >>   * Replace use of GCC-specific alignof(<expression>) with alignof(<type>).
> >>   * Update example to reflect FOREACH macro name change (in RFC v2).
> >>
> >> RFC v2:
> >>   * Use alignof to derive alignment requirements. (Morten Brørup)
> >>   * Change name of FOREACH to make it distinct from <rte_lcore.h>'s
> >>     *per-EAL-thread* RTE_LCORE_FOREACH(). (Morten Brørup)
> >>   * Allow user-specified alignment, but limit max to cache line size.
> >>
> >> Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
> >> Acked-by: Morten Brørup <mb@smartsharesystems.com>
> >> ---
> >>   config/rte_config.h                   |   1 +
> >>   doc/api/doxy-api-index.md             |   1 +
> >>   lib/eal/common/eal_common_lcore_var.c |  68 +++++
> >>   lib/eal/common/meson.build            |   1 +
> >>   lib/eal/include/meson.build           |   1 +
> >>   lib/eal/include/rte_lcore_var.h       | 368 ++++++++++++++++++++++++++
> >>   lib/eal/version.map                   |   4 +
> >>   7 files changed, 444 insertions(+)
> >>   create mode 100644 lib/eal/common/eal_common_lcore_var.c
> >>   create mode 100644 lib/eal/include/rte_lcore_var.h
> >>
> >> diff --git a/config/rte_config.h b/config/rte_config.h
> >> index d743a5c3d3..0dac33d3b9 100644
> >> --- a/config/rte_config.h
> >> +++ b/config/rte_config.h
> >> @@ -41,6 +41,7 @@
> >>   /* EAL defines */
> >>   #define RTE_CACHE_GUARD_LINES 1
> >>   #define RTE_MAX_HEAPS 32
> >> +#define RTE_MAX_LCORE_VAR 1048576
> >>   #define RTE_MAX_MEMSEG_LISTS 128
> >>   #define RTE_MAX_MEMSEG_PER_LIST 8192
> >>   #define RTE_MAX_MEM_MB_PER_LIST 32768
> >> diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
> >> index 8c1eb8fafa..a3b8391570 100644
> >> --- a/doc/api/doxy-api-index.md
> >> +++ b/doc/api/doxy-api-index.md
> >> @@ -99,6 +99,7 @@ The public API headers are grouped by topics:
> >>     [interrupts](@ref rte_interrupts.h),
> >>     [launch](@ref rte_launch.h),
> >>     [lcore](@ref rte_lcore.h),
> >> +  [lcore-varible](@ref rte_lcore_var.h),
> >>     [per-lcore](@ref rte_per_lcore.h),
> >>     [service cores](@ref rte_service.h),
> >>     [keepalive](@ref rte_keepalive.h),
> >> diff --git a/lib/eal/common/eal_common_lcore_var.c b/lib/eal/common/eal_common_lcore_var.c
> >> new file mode 100644
> >> index 0000000000..5c353ebd46
> >> --- /dev/null
> >> +++ b/lib/eal/common/eal_common_lcore_var.c
> >> @@ -0,0 +1,68 @@
> >> +/* SPDX-License-Identifier: BSD-3-Clause
> >> + * Copyright(c) 2024 Ericsson AB
> >> + */
> >> +
> >> +#include <inttypes.h>
> >> +
> >> +#include <rte_common.h>
> >> +#include <rte_debug.h>
> >> +#include <rte_log.h>
> >> +
> >> +#include <rte_lcore_var.h>
> >> +
> >> +#include "eal_private.h"
> >> +
> >> +#define LCORE_BUFFER_SIZE (RTE_MAX_LCORE_VAR * RTE_MAX_LCORE)
> >> +
> >> +static void *lcore_buffer;
> >> +static size_t offset = RTE_MAX_LCORE_VAR;
> >> +
> >> +static void *
> >> +lcore_var_alloc(size_t size, size_t align)
> >> +{
> >> +	void *handle;
> >> +	void *value;
> >> +
> >> +	offset = RTE_ALIGN_CEIL(offset, align);
> >> +
> >> +	if (offset + size > RTE_MAX_LCORE_VAR) {
> >> +		lcore_buffer = aligned_alloc(RTE_CACHE_LINE_SIZE,
> >> +					     LCORE_BUFFER_SIZE);
> >
> > Hmm... do I get it right: if offset is <= then RTE_MAX_LCORE_VAR,  and  offset + size > RTE_MAX_LCORE_VAR
> > we simply overwrite lcore_buffer with newly allocated buffer of the same size?
> 
> No, it's just the pointer that is overwritten. The old buffer will
> remain in memory.

Ah ok, I missed that you changed the handle to pointer conversion in new version too.
Now handle is not just an offset, but an actual pointer to lcore0 var, so all we have is to add
lcore_idx offset.
Makes sense, thanks for clarifying.
LGTM then.
 

> 
> > I understand that you expect it just never to happen (total size of all lcore vars never exceed 1MB), but still
> > I think we need to handle it in a some better way then just ignoring such possibility...
> > Might be RTE_VERIFY() at least?
> >
> 
> In this revision of the patch set, RTE_MAX_LCORE_VAR does not represent
> an upper bound for the sum of all lcore variables' size, but rather only
> the maximum size of a single lcore variable.
> 
> Variable alignment and size constraints are RTE_ASSERT()ed at the point
> of allocation. One could argue they should be RTE_VERIFY()-ed instead,
> since there aren't any performance constraints.
> 
> > As a more generic question - do we need to support LCORE_VAR for dlopen()s that could happen after rte_eal_init()
> > is called and LCORE threads were created?
> 
> Yes, allocations after rte_eal_init() (caused by dlopen() or otherwise)
> must be allowed imo, and are allowed. Otherwise applications sitting on
> top of DPDK can't use this facility.
> 
> > Because, if no, then we probably can make this construction much more flexible:
> > one buffer per LCORE, allocate on demand, etc.
> >
> 
> On-demand allocations are already supported, but one can't do free().
> That's why I've called what this module provide "static allocation",
> while it may be more appropriately described as "dynamic allocation
> without deallocation".
> 
> "True" dynamic memory allocation of per-lcore memory would be very
> useful, but is an entirely different beast in terms of complexity and
> (if to be usable in the packet processing fast path) performance
> requirements.
> 
> "True" dynamic memory allocation would also result in something less
> compact (at least if you use the usual pattern with a per-object heap
> header).
> 
> >> +		RTE_VERIFY(lcore_buffer != NULL);
> >> +
> >> +		offset = 0;
> >> +	}
> >> +
> >> +	handle = RTE_PTR_ADD(lcore_buffer, offset);
> >> +
> >> +	offset += size;
> >> +
> >> +	RTE_LCORE_VAR_FOREACH_VALUE(value, handle)
> >> +		memset(value, 0, size);
> >> +
> >> +	EAL_LOG(DEBUG, "Allocated %"PRIuPTR" bytes of per-lcore data with a "
> >> +		"%"PRIuPTR"-byte alignment", size, align);
> >> +
> >> +	return handle;
> >> +}
> >> +
> >> +void *
> >> +rte_lcore_var_alloc(size_t size, size_t align)
> >> +{
> >> +	/* Having the per-lcore buffer size aligned on cache lines
> >> +	 * assures as well as having the base pointer aligned on cache
> >> +	 * size assures that aligned offsets also translate to alipgned
> >> +	 * pointers across all values.
> >> +	 */
> >> +	RTE_BUILD_BUG_ON(RTE_MAX_LCORE_VAR % RTE_CACHE_LINE_SIZE != 0);
> >> +	RTE_ASSERT(align <= RTE_CACHE_LINE_SIZE);
> >> +	RTE_ASSERT(size <= RTE_MAX_LCORE_VAR);
> >> +
> >> +	/* '0' means asking for worst-case alignment requirements */
> >> +	if (align == 0)
> >> +		align = alignof(max_align_t);
> >> +
> >> +	RTE_ASSERT(rte_is_power_of_2(align));
> >> +
> >> +	return lcore_var_alloc(size, align);
> >> +}
  
Morten Brørup Sept. 2, 2024, 2:42 p.m. UTC | #4
> From: Mattias Rönnblom [mailto:mattias.ronnblom@ericsson.com]
> Sent: Monday, 6 May 2024 10.27
> 
> This RFC presents a new API <rte_lcore_var.h> for static per-lcore id
> data allocation.
> 
> Please refer to the <rte_lcore_var.h> API documentation for both a
> rationale for this new API, and a comparison to the alternatives
> available.
> 
> The adoption of this API would affect many different DPDK modules, but
> the author updated only a few, mostly to serve as examples in this
> RFC, and to iron out some, but surely not all, wrinkles in the API.
> 
> The question on how to best allocate static per-lcore memory has been
> up several times on the dev mailing list, for example in the thread on
> "random: use per lcore state" RFC by Stephen Hemminger.
> 
> Lcore variables are surely not the answer to all your per-lcore-data
> needs, since it only allows for more-or-less static allocation. In the
> author's opinion, it does however provide a reasonably simple and
> clean and seemingly very much performant solution to a real problem.

This RFC is an improvement of the design pattern of allocating a RTE_MAX_LCORE sized array of structs per library, which typically introduces a lot of padding, and thus wastes L1 data cache.

I would like to see it as a patch getting into DPDK 24.11.

> 
> One thing is unclear to the author is how this API relates to a
> potential future per-lcore dynamic allocator (e.g., a per-lcore heap).

Perfection is the enemy of progress.
Let's consider this a 1:1 upgrade of a existing design pattern, and not worry about how to broaden its scope in the future.

-Morten
  
Mattias Rönnblom Sept. 10, 2024, 6:41 a.m. UTC | #5
On 2024-09-02 16:42, Morten Brørup wrote:
>> From: Mattias Rönnblom [mailto:mattias.ronnblom@ericsson.com]
>> Sent: Monday, 6 May 2024 10.27
>>
>> This RFC presents a new API <rte_lcore_var.h> for static per-lcore id
>> data allocation.
>>
>> Please refer to the <rte_lcore_var.h> API documentation for both a
>> rationale for this new API, and a comparison to the alternatives
>> available.
>>
>> The adoption of this API would affect many different DPDK modules, but
>> the author updated only a few, mostly to serve as examples in this
>> RFC, and to iron out some, but surely not all, wrinkles in the API.
>>
>> The question on how to best allocate static per-lcore memory has been
>> up several times on the dev mailing list, for example in the thread on
>> "random: use per lcore state" RFC by Stephen Hemminger.
>>
>> Lcore variables are surely not the answer to all your per-lcore-data
>> needs, since it only allows for more-or-less static allocation. In the
>> author's opinion, it does however provide a reasonably simple and
>> clean and seemingly very much performant solution to a real problem.
> 
> This RFC is an improvement of the design pattern of allocating a RTE_MAX_LCORE sized array of structs per library, which typically introduces a lot of padding, and thus wastes L1 data cache.
> 
> I would like to see it as a patch getting into DPDK 24.11.
> 

I would be happy to develop and maintain this DPDK module.

I will submit this as a v1 PATCH.

>>
>> One thing is unclear to the author is how this API relates to a
>> potential future per-lcore dynamic allocator (e.g., a per-lcore heap).
> 
> Perfection is the enemy of progress.
> Let's consider this a 1:1 upgrade of a existing design pattern, and not worry about how to broaden its scope in the future.
> 
> -Morten
>
  
Stephen Hemminger Sept. 10, 2024, 3:41 p.m. UTC | #6
On Tue, 10 Sep 2024 08:41:19 +0200
Mattias Rönnblom <hofors@lysator.liu.se> wrote:

> On 2024-09-02 16:42, Morten Brørup wrote:

On a related note, latest GCC supports annotating the address space
of variables. Kernel uses it for RCU.

It would be good if DPDK could do this for:
	- per lcore data
	- data in huge pages
	- data protected by rcu

With these annotations various checkers and compilers can warn about
places where data is passed (with cast to override).
  

Patch

diff --git a/config/rte_config.h b/config/rte_config.h
index d743a5c3d3..0dac33d3b9 100644
--- a/config/rte_config.h
+++ b/config/rte_config.h
@@ -41,6 +41,7 @@ 
 /* EAL defines */
 #define RTE_CACHE_GUARD_LINES 1
 #define RTE_MAX_HEAPS 32
+#define RTE_MAX_LCORE_VAR 1048576
 #define RTE_MAX_MEMSEG_LISTS 128
 #define RTE_MAX_MEMSEG_PER_LIST 8192
 #define RTE_MAX_MEM_MB_PER_LIST 32768
diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
index 8c1eb8fafa..a3b8391570 100644
--- a/doc/api/doxy-api-index.md
+++ b/doc/api/doxy-api-index.md
@@ -99,6 +99,7 @@  The public API headers are grouped by topics:
   [interrupts](@ref rte_interrupts.h),
   [launch](@ref rte_launch.h),
   [lcore](@ref rte_lcore.h),
+  [lcore-varible](@ref rte_lcore_var.h),
   [per-lcore](@ref rte_per_lcore.h),
   [service cores](@ref rte_service.h),
   [keepalive](@ref rte_keepalive.h),
diff --git a/lib/eal/common/eal_common_lcore_var.c b/lib/eal/common/eal_common_lcore_var.c
new file mode 100644
index 0000000000..5c353ebd46
--- /dev/null
+++ b/lib/eal/common/eal_common_lcore_var.c
@@ -0,0 +1,68 @@ 
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2024 Ericsson AB
+ */
+
+#include <inttypes.h>
+
+#include <rte_common.h>
+#include <rte_debug.h>
+#include <rte_log.h>
+
+#include <rte_lcore_var.h>
+
+#include "eal_private.h"
+
+#define LCORE_BUFFER_SIZE (RTE_MAX_LCORE_VAR * RTE_MAX_LCORE)
+
+static void *lcore_buffer;
+static size_t offset = RTE_MAX_LCORE_VAR;
+
+static void *
+lcore_var_alloc(size_t size, size_t align)
+{
+	void *handle;
+	void *value;
+
+	offset = RTE_ALIGN_CEIL(offset, align);
+
+	if (offset + size > RTE_MAX_LCORE_VAR) {
+		lcore_buffer = aligned_alloc(RTE_CACHE_LINE_SIZE,
+					     LCORE_BUFFER_SIZE);
+		RTE_VERIFY(lcore_buffer != NULL);
+
+		offset = 0;
+	}
+
+	handle = RTE_PTR_ADD(lcore_buffer, offset);
+
+	offset += size;
+
+	RTE_LCORE_VAR_FOREACH_VALUE(value, handle)
+		memset(value, 0, size);
+
+	EAL_LOG(DEBUG, "Allocated %"PRIuPTR" bytes of per-lcore data with a "
+		"%"PRIuPTR"-byte alignment", size, align);
+
+	return handle;
+}
+
+void *
+rte_lcore_var_alloc(size_t size, size_t align)
+{
+	/* Having the per-lcore buffer size aligned on cache lines
+	 * assures as well as having the base pointer aligned on cache
+	 * size assures that aligned offsets also translate to alipgned
+	 * pointers across all values.
+	 */
+	RTE_BUILD_BUG_ON(RTE_MAX_LCORE_VAR % RTE_CACHE_LINE_SIZE != 0);
+	RTE_ASSERT(align <= RTE_CACHE_LINE_SIZE);
+	RTE_ASSERT(size <= RTE_MAX_LCORE_VAR);
+
+	/* '0' means asking for worst-case alignment requirements */
+	if (align == 0)
+		align = alignof(max_align_t);
+
+	RTE_ASSERT(rte_is_power_of_2(align));
+
+	return lcore_var_alloc(size, align);
+}
diff --git a/lib/eal/common/meson.build b/lib/eal/common/meson.build
index 22a626ba6f..d41403680b 100644
--- a/lib/eal/common/meson.build
+++ b/lib/eal/common/meson.build
@@ -18,6 +18,7 @@  sources += files(
         'eal_common_interrupts.c',
         'eal_common_launch.c',
         'eal_common_lcore.c',
+        'eal_common_lcore_var.c',
         'eal_common_mcfg.c',
         'eal_common_memalloc.c',
         'eal_common_memory.c',
diff --git a/lib/eal/include/meson.build b/lib/eal/include/meson.build
index e94b056d46..9449253e23 100644
--- a/lib/eal/include/meson.build
+++ b/lib/eal/include/meson.build
@@ -27,6 +27,7 @@  headers += files(
         'rte_keepalive.h',
         'rte_launch.h',
         'rte_lcore.h',
+        'rte_lcore_var.h',
         'rte_lock_annotations.h',
         'rte_malloc.h',
         'rte_mcslock.h',
diff --git a/lib/eal/include/rte_lcore_var.h b/lib/eal/include/rte_lcore_var.h
new file mode 100644
index 0000000000..1db479253d
--- /dev/null
+++ b/lib/eal/include/rte_lcore_var.h
@@ -0,0 +1,368 @@ 
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2024 Ericsson AB
+ */
+
+#ifndef _RTE_LCORE_VAR_H_
+#define _RTE_LCORE_VAR_H_
+
+/**
+ * @file
+ *
+ * RTE Per-lcore id variables
+ *
+ * This API provides a mechanism to create and access per-lcore id
+ * variables in a space- and cycle-efficient manner.
+ *
+ * A per-lcore id variable (or lcore variable for short) has one value
+ * for each EAL thread and registered non-EAL thread. In other words,
+ * there's one copy of its value for each and every current and future
+ * lcore id-equipped thread, with the total number of copies amounting
+ * to @c RTE_MAX_LCORE.
+ *
+ * In order to access the values of an lcore variable, a handle is
+ * used. The type of the handle is a pointer to the value's type
+ * (e.g., for @c uint32_t lcore variable, the handle is a
+ * <code>uint32_t *</code>. A handle may be passed between modules and
+ * threads just like any pointer, but its value is not the address of
+ * any particular object, but rather just an opaque identifier, stored
+ * in a typed pointer (to inform the access macro the type of values).
+ *
+ * @b Creation
+ *
+ * An lcore variable is created in two steps:
+ *  1. Define a lcore variable handle by using @ref RTE_LCORE_VAR_HANDLE.
+ *  2. Allocate lcore variable storage and initialize the handle with
+ *     a unique identifier by @ref RTE_LCORE_VAR_ALLOC or
+ *     @ref RTE_LCORE_VAR_INIT. Allocation generally occurs the time of
+ *     module initialization, but may be done at any time.
+ *
+ * An lcore variable is not tied to the owning thread's lifetime. It's
+ * available for use by any thread immediately after having been
+ * allocated, and continues to be available throughout the lifetime of
+ * the EAL.
+ *
+ * Lcore variables cannot and need not be freed.
+ *
+ * @b Access
+ *
+ * The value of any lcore variable for any lcore id may be accessed
+ * from any thread (including unregistered threads), but is should
+ * generally only *frequently* read from or written to by the owner.
+ *
+ * Values of the same lcore variable but owned by to different lcore
+ * ids *may* be frequently read or written by the owners without the
+ * risk of false sharing.
+ *
+ * An appropriate synchronization mechanism (e.g., atomics) should
+ * employed to assure there are no data races between the owning
+ * thread and any non-owner threads accessing the same lcore variable
+ * instance.
+ *
+ * The value of the lcore variable for a particular lcore id is
+ * accessed using @ref RTE_LCORE_VAR_LCORE_VALUE.
+ *
+ * A common pattern is for an EAL thread or a registered non-EAL
+ * thread to access its own lcore variable value, for which a
+ * short-hand exists in the form of @ref RTE_LCORE_VAR_VALUE.
+ *
+ * Although the handle (as defined by @ref RTE_LCORE_VAR_HANDLE) is a
+ * pointer with the same type as the value, it may not be directly
+ * dereferenced and must be treated as an opaque identifier.
+ *
+ * Lcore variable handles and value pointers may be freely passed
+ * between different threads.
+ *
+ * @b Storage
+ *
+ * An lcore variable's values may by of a primitive type like @c int,
+ * but would more typically be a @c struct. An application may choose
+ * to define an lcore variable, which it then it goes on to never
+ * allocate.
+ *
+ * The lcore variable handle introduces a per-variable (not
+ * per-value/per-lcore id) overhead of @c sizeof(void *) bytes, so
+ * there are some memory footprint gains to be made by organizing all
+ * per-lcore id data for a particular module as one lcore variable
+ * (e.g., as a struct).
+ *
+ * The size of a lcore variable's value must be less than the DPDK
+ * build-time constant @c RTE_MAX_LCORE_VAR.
+ *
+ * The lcore variable are stored in a series of lcore buffers, which
+ * are allocated from the libc heap. Heap allocation failures are
+ * treated as fatal.
+ *
+ * Lcore variables should generally *not* be @ref __rte_cache_aligned
+ * and need *not* include a @ref RTE_CACHE_GUARD field, since the use
+ * of these constructs are designed to avoid false sharing. In the
+ * case of an lcore variable instance, all nearby data structures
+ * should almost-always be written to by a single thread (the lcore
+ * variable owner). Adding padding will increase the effective memory
+ * working set size, and potentially reducing performance.
+ *
+ * @b Example
+ *
+ * Below is an example of the use of an lcore variable:
+ *
+ * @code{.c}
+ * struct foo_lcore_state {
+ *         int a;
+ *         long b;
+ * };
+ *
+ * static RTE_LCORE_VAR_HANDLE(struct foo_lcore_state, lcore_states);
+ *
+ * long foo_get_a_plus_b(void)
+ * {
+ *         struct foo_lcore_state *state = RTE_LCORE_VAR_VALUE(lcore_states);
+ *
+ *         return state->a + state->b;
+ * }
+ *
+ * RTE_INIT(rte_foo_init)
+ * {
+ *         RTE_LCORE_VAR_ALLOC(lcore_states);
+ *
+ *         struct foo_lcore_state *state;
+ *         RTE_LCORE_VAR_FOREACH_VALUE(state, lcore_states) {
+ *                 (initialize 'state')
+ *         }
+ *
+ *         (other initialization)
+ * }
+ * @endcode
+ *
+ *
+ * @b Alternatives
+ *
+ * Lcore variables are designed to replace a pattern exemplified below:
+ * @code{.c}
+ * struct foo_lcore_state {
+ *         int a;
+ *         long b;
+ *         RTE_CACHE_GUARD;
+ * } __rte_cache_aligned;
+ *
+ * static struct foo_lcore_state lcore_states[RTE_MAX_LCORE];
+ * @endcode
+ *
+ * This scheme is simple and effective, but has one drawback: the data
+ * is organized so that objects related to all lcores for a particular
+ * module is kept close in memory. At a bare minimum, this forces the
+ * use of cache-line alignment to avoid false sharing. With CPU
+ * hardware prefetching and memory loads resulting from speculative
+ * execution (functions which seemingly are getting more eager faster
+ * than they are getting more intelligent), one or more "guard" cache
+ * lines may be required to separate one lcore's data from another's.
+ *
+ * Lcore variables has the upside of working with, not against, the
+ * CPU's assumptions and for example next-line prefetchers may well
+ * work the way its designers intended (i.e., to the benefit, not
+ * detriment, of system performance).
+ *
+ * Another alternative to @ref rte_lcore_var.h is the @ref
+ * rte_per_lcore.h API, which make use of thread-local storage (TLS,
+ * e.g., GCC __thread or C11 _Thread_local). The main differences
+ * between by using the various forms of TLS (e.g., @ref
+ * RTE_DEFINE_PER_LCORE or _Thread_local) and the use of lcore
+ * variables are:
+ *
+ *   * The existence and non-existence of a thread-local variable
+ *     instance follow that of particular thread's. The data cannot be
+ *     accessed before the thread has been created, nor after it has
+ *     exited. One effect of this is thread-local variables must
+ *     initialized in a "lazy" manner (e.g., at the point of thread
+ *     creation). Lcore variables may be accessed immediately after
+ *     having been allocated (which is usually prior any thread beyond
+ *     the main thread is running).
+ *   * A thread-local variable is duplicated across all threads in the
+ *     process, including unregistered non-EAL threads (i.e.,
+ *     "regular" threads). For DPDK applications heavily relying on
+ *     multi-threading (in conjunction to DPDK's "one thread per core"
+ *     pattern), either by having many concurrent threads or
+ *     creating/destroying threads at a high rate, an excessive use of
+ *     thread-local variables may cause inefficiencies (e.g.,
+ *     increased thread creation overhead due to thread-local storage
+ *     initialization or increased total RAM footprint usage). Lcore
+ *     variables *only* exist for threads with an lcore id, and thus
+ *     not for such "regular" threads.
+ *   * If data in thread-local storage may be shared between threads
+ *     (i.e., can a pointer to a thread-local variable be passed to
+ *     and successfully dereferenced by non-owning thread) depends on
+ *     the details of the TLS implementation. With GCC __thread and
+ *     GCC _Thread_local, such data sharing is supported. In the C11
+ *     standard, the result of accessing another thread's
+ *     _Thread_local object is implementation-defined. Lcore variable
+ *     instances may be accessed reliably by any thread.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <stddef.h>
+#include <stdalign.h>
+
+#include <rte_common.h>
+#include <rte_config.h>
+#include <rte_lcore.h>
+
+/**
+ * Given the lcore variable type, produces the type of the lcore
+ * variable handle.
+ */
+#define RTE_LCORE_VAR_HANDLE_TYPE(type)		\
+	type *
+
+/**
+ * Define a lcore variable handle.
+ *
+ * This macro defines a variable which is used as a handle to access
+ * the various per-lcore id instances of a per-lcore id variable.
+ *
+ * The aim with this macro is to make clear at the point of
+ * declaration that this is an lcore handler, rather than a regular
+ * pointer.
+ *
+ * Add @b static as a prefix in case the lcore variable are only to be
+ * accessed from a particular translation unit.
+ */
+#define RTE_LCORE_VAR_HANDLE(type, name)	\
+	RTE_LCORE_VAR_HANDLE_TYPE(type) name
+
+/**
+ * Allocate space for an lcore variable, and initialize its handle.
+ */
+#define RTE_LCORE_VAR_ALLOC_SIZE_ALIGN(handle, size, align)	\
+	handle = rte_lcore_var_alloc(size, align)
+
+/**
+ * Allocate space for an lcore variable, and initialize its handle,
+ * with values aligned for any type of object.
+ */
+#define RTE_LCORE_VAR_ALLOC_SIZE(handle, size)	\
+	RTE_LCORE_VAR_ALLOC_SIZE_ALIGN(handle, size, 0)
+
+/**
+ * Allocate space for an lcore variable of the size and alignment requirements
+ * suggested by the handler pointer type, and initialize its handle.
+ */
+#define RTE_LCORE_VAR_ALLOC(handle)					\
+	RTE_LCORE_VAR_ALLOC_SIZE_ALIGN(handle, sizeof(*(handle)),	\
+				       alignof(typeof(*(handle))))
+
+/**
+ * Allocate an explicitly-sized, explicitly-aligned lcore variable by
+ * means of a @ref RTE_INIT constructor.
+ */
+#define RTE_LCORE_VAR_INIT_SIZE_ALIGN(name, size, align)		\
+	RTE_INIT(rte_lcore_var_init_ ## name)				\
+	{								\
+		RTE_LCORE_VAR_ALLOC_SIZE_ALIGN(name, size, align);	\
+	}
+
+/**
+ * Allocate an explicitly-sized lcore variable by means of a @ref
+ * RTE_INIT constructor.
+ */
+#define RTE_LCORE_VAR_INIT_SIZE(name, size)		\
+	RTE_LCORE_VAR_INIT_SIZE_ALIGN(name, size, 0)
+
+/**
+ * Allocate an lcore variable by means of a @ref RTE_INIT constructor.
+ */
+#define RTE_LCORE_VAR_INIT(name)					\
+	RTE_INIT(rte_lcore_var_init_ ## name)				\
+	{								\
+		RTE_LCORE_VAR_ALLOC(name);				\
+	}
+
+/**
+ * Get void pointer to lcore variable instance with the specified
+ * lcore id.
+ *
+ * @param lcore_id
+ *   The lcore id specifying which of the @c RTE_MAX_LCORE value
+ *   instances should be accessed. The lcore id need not be valid
+ *   (e.g., may be @ref LCORE_ID_ANY), but in such a case, the pointer
+ *   is also not valid (and thus should not be dereferenced).
+ * @param handle
+ *   The lcore variable handle.
+ */
+static inline void *
+rte_lcore_var_lcore_ptr(unsigned int lcore_id, void *handle)
+{
+	return RTE_PTR_ADD(handle, lcore_id * RTE_MAX_LCORE_VAR);
+}
+
+/**
+ * Get pointer to lcore variable instance with the specified lcore id.
+ *
+ * @param lcore_id
+ *   The lcore id specifying which of the @c RTE_MAX_LCORE value
+ *   instances should be accessed. The lcore id need not be valid
+ *   (e.g., may be @ref LCORE_ID_ANY), but in such a case, the pointer
+ *   is also not valid (and thus should not be dereferenced).
+ * @param handle
+ *   The lcore variable handle.
+ */
+#define RTE_LCORE_VAR_LCORE_VALUE(lcore_id, handle)			\
+	((typeof(handle))rte_lcore_var_lcore_ptr(lcore_id, handle))
+
+/**
+ * Get pointer to lcore variable instance of the current thread.
+ *
+ * May only be used by EAL threads and registered non-EAL threads.
+ */
+#define RTE_LCORE_VAR_VALUE(handle) \
+	RTE_LCORE_VAR_LCORE_VALUE(rte_lcore_id(), handle)
+
+/**
+ * Iterate over each lcore id's value for a lcore variable.
+ *
+ * @param value
+ *   A pointer set successivly set to point to lcore variable value
+ *   corresponding to every lcore id (up to @c RTE_MAX_LCORE).
+ * @param handle
+ *   The lcore variable handle.
+ */
+#define RTE_LCORE_VAR_FOREACH_VALUE(value, handle)			\
+	for (unsigned int lcore_id =					\
+		     (((value) = RTE_LCORE_VAR_LCORE_VALUE(0, handle)), 0); \
+	     lcore_id < RTE_MAX_LCORE;					\
+	     lcore_id++, (value) = RTE_LCORE_VAR_LCORE_VALUE(lcore_id, handle))
+
+/**
+ * Allocate space in the per-lcore id buffers for a lcore variable.
+ *
+ * The pointer returned is only an opaque identifer of the variable. To
+ * get an actual pointer to a particular instance of the variable use
+ * @ref RTE_LCORE_VAR_VALUE or @ref RTE_LCORE_VAR_LCORE_VALUE.
+ *
+ * The lcore variable values' memory is set to zero.
+ *
+ * The allocation is always successful, barring a fatal exhaustion of
+ * the per-lcore id buffer space.
+ *
+ * rte_lcore_var_alloc() is not multi-thread safe.
+ *
+ * @param size
+ *   The size (in bytes) of the variable's per-lcore id value. Must be > 0.
+ * @param align
+ *   If 0, the values will be suitably aligned for any kind of type
+ *   (i.e., alignof(max_align_t)). Otherwise, the values will be aligned
+ *   on a multiple of *align*, which must be a power of 2 and equal or
+ *   less than @c RTE_CACHE_LINE_SIZE.
+ * @return
+ *   The id of the variable, stored in a void pointer value. The value
+ *   is always non-NULL.
+ */
+__rte_experimental
+void *
+rte_lcore_var_alloc(size_t size, size_t align);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_LCORE_VAR_H_ */
diff --git a/lib/eal/version.map b/lib/eal/version.map
index 5e0cd47c82..e90b86115a 100644
--- a/lib/eal/version.map
+++ b/lib/eal/version.map
@@ -393,6 +393,10 @@  EXPERIMENTAL {
 	# added in 23.07
 	rte_memzone_max_get;
 	rte_memzone_max_set;
+
+	# added in 24.03
+	rte_lcore_var_alloc;
+	rte_lcore_var;
 };
 
 INTERNAL {