linux.git/Documentation/bpf, branch v6.12.80 Clone of https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git bpf: Adjust free target to avoid global starvation of LRU map 2025-07-17T16:37:22+00:00 Willem de Bruijn willemb@google.com 2025-06-18T21:57:40+00:00 9ef5d4748dfeddc3548590cf125cff34f84cffe6 [ Upstream commit d4adf1c9ee7722545450608bcb095fb31512f0c6 ] BPF_MAP_TYPE_LRU_HASH can recycle most recent elements well before the map is full, due to percpu reservations and force shrink before neighbor stealing. Once a CPU is unable to borrow from the global map, it will once steal one elem from a neighbor and after that each time flush this one element to the global list and immediately recycle it. Batch value LOCAL_FREE_TARGET (128) will exhaust a 10K element map with 79 CPUs. CPU 79 will observe this behavior even while its neighbors hold 78 * 127 + 1 * 15 == 9921 free elements (99%). CPUs need not be active concurrently. The issue can appear with affinity migration, e.g., irqbalance. Each CPU can reserve and then hold onto its 128 elements indefinitely. Avoid global list exhaustion by limiting aggregate percpu caches to half of map size, by adjusting LOCAL_FREE_TARGET based on cpu count. This change has no effect on sufficiently large tables. Similar to LOCAL_NR_SCANS and lru->nr_scans, introduce a map variable lru->free_target. The extra field fits in a hole in struct bpf_lru. The cacheline is already warm where read in the hot path. The field is only accessed with the lru lock held. Tested-by: Anton Protopopov <a.s.protopopov@gmail.com> Signed-off-by: Willem de Bruijn <willemb@google.com> Acked-by: Stanislav Fomichev <sdf@fomichev.me> Link: https://lore.kernel.org/r/20250618215803.3587312-1-willemdebruijn.kernel@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit d4adf1c9ee7722545450608bcb095fb31512f0c6 ]

BPF_MAP_TYPE_LRU_HASH can recycle most recent elements well before the
map is full, due to percpu reservations and force shrink before
neighbor stealing. Once a CPU is unable to borrow from the global map,
it will once steal one elem from a neighbor and after that each time
flush this one element to the global list and immediately recycle it.

Batch value LOCAL_FREE_TARGET (128) will exhaust a 10K element map
with 79 CPUs. CPU 79 will observe this behavior even while its
neighbors hold 78 * 127 + 1 * 15 == 9921 free elements (99%).

CPUs need not be active concurrently. The issue can appear with
affinity migration, e.g., irqbalance. Each CPU can reserve and then
hold onto its 128 elements indefinitely.

Avoid global list exhaustion by limiting aggregate percpu caches to
half of map size, by adjusting LOCAL_FREE_TARGET based on cpu count.
This change has no effect on sufficiently large tables.

Similar to LOCAL_NR_SCANS and lru->nr_scans, introduce a map variable
lru->free_target. The extra field fits in a hole in struct bpf_lru.
The cacheline is already warm where read in the hot path. The field is
only accessed with the lru lock held.

Tested-by: Anton Protopopov <a.s.protopopov@gmail.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Link: https://lore.kernel.org/r/20250618215803.3587312-1-willemdebruijn.kernel@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
bpf: Add namespace to BPF internal symbols 2025-05-02T05:59:04+00:00 Alexei Starovoitov ast@kernel.org 2025-04-25T01:45:42+00:00 955f9ede52b8617e454fcfd4528f90bd23a30753 [ Upstream commit f88886de0927a2adf4c1b4c5c1f1d31d2023ef74 ] Add namespace to BPF internal symbols used by light skeleton to prevent abuse and document with the code their allowed usage. Fixes: b1d18a7574d0 ("bpf: Extend sys_bpf commands for bpf_syscall programs.") Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com> Link: https://lore.kernel.org/bpf/20250425014542.62385-1-alexei.starovoitov@gmail.com Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit f88886de0927a2adf4c1b4c5c1f1d31d2023ef74 ]

Add namespace to BPF internal symbols used by light skeleton
to prevent abuse and document with the code their allowed usage.

Fixes: b1d18a7574d0 ("bpf: Extend sys_bpf commands for bpf_syscall programs.")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/bpf/20250425014542.62385-1-alexei.starovoitov@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
docs/bpf: Add missing BPF program types to docs 2024-09-12T17:56:41+00:00 Donald Hunter donald.hunter@gmail.com 2024-09-12T09:59:44+00:00 6182e0b80f9b5c8da19ab3a60db2e4cc10b259e8 Update the table of program types in the libbpf documentation with the recently added program types. Signed-off-by: Donald Hunter <donald.hunter@gmail.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20240912095944.6386-1-donald.hunter@gmail.com 
Update the table of program types in the libbpf documentation with the
recently added program types.

Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240912095944.6386-1-donald.hunter@gmail.com
docs/bpf: Add constant values for linkages 2024-09-11T20:47:04+00:00 Will Hawkins hawkinsw@obs.cr 2024-09-11T05:50:32+00:00 c229c17a76e9c52950ea345323d5a19993610f57 Make the values of the symbolic constants that define the valid linkages for functions and variables explicit. Signed-off-by: Will Hawkins <hawkinsw@obs.cr> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Acked-by: Yonghong Song <yonghong.song@linux.dev> Link: https://lore.kernel.org/bpf/20240911055033.2084881-1-hawkinsw@obs.cr 
Make the values of the symbolic constants that define the valid linkages
for functions and variables explicit.

Signed-off-by: Will Hawkins <hawkinsw@obs.cr>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/bpf/20240911055033.2084881-1-hawkinsw@obs.cr
docs/bpf: Fix a typo in verifier.rst 2024-08-29T19:19:30+00:00 Yiming Xiang kxiang@umich.edu 2024-08-29T03:17:12+00:00 89dd9bb25597621ca0500aa598140ff0858091e2 In verifier.rst, there is a typo in section 'Register parentage chains'. Caller saved registers are r0-r5, callee saved registers are r6-r9. Here by context it means callee saved registers rather than caller saved registers. This may confuse users. Signed-off-by: Yiming Xiang <kxiang@umich.edu> Link: https://lore.kernel.org/r/20240829031712.198489-1-kxiang@umich.edu Signed-off-by: Alexei Starovoitov <ast@kernel.org> 
In verifier.rst, there is a typo in section 'Register parentage chains'.
Caller saved registers are r0-r5, callee saved registers are r6-r9.

Here by context it means callee saved registers rather than caller saved
registers. This may confuse users.

Signed-off-by: Yiming Xiang <kxiang@umich.edu>
Link: https://lore.kernel.org/r/20240829031712.198489-1-kxiang@umich.edu
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
bpf, docs: Address comments from IETF Area Directors 2024-06-23T16:10:26+00:00 Dave Thaler dthaler1968@googlemail.com 2024-06-23T15:04:53+00:00 04efaebd72d1d3d9991841051fafc6b195f3676d This patch does the following to address IETF feedback: * Remove mention of "program type" and reference future docs (and mention platform-specific docs exist) for helper functions and BTF. Addresses Roman Danyliw's comments based on GENART review from Ines Robles [0]. * Add reference for endianness as requested by John Scudder [1]. * Added bit numbers to top of 32-bit wide format diagrams as requested by Paul Wouters [2]. * Added more text about why BPF doesn't stand for anything, based on text from ebpf.io [3], as requested by Eric Vyncke and Gunter Van de Velde [4]. * Replaced "htobe16" (and similar) and the direction-specific description with just "be16" (and similar) and a direction-agnostic description, to match the direction-agnostic description in the Byteswap Instructions section. Based on feedback from Eric Vyncke [5]. [0] https://mailarchive.ietf.org/arch/msg/bpf/DvDgDWOiwk05OyNlWlAmELZFPlM/ [1] https://mailarchive.ietf.org/arch/msg/bpf/eKNXpU4jCLjsbZDSw8LjI29M3tM/ [2] https://mailarchive.ietf.org/arch/msg/bpf/hGk8HkYxeZTpdu9qW_MvbGKj7WU/ [3] https://ebpf.io/what-is-ebpf/#what-do-ebpf-and-bpf-stand-for [4] https://mailarchive.ietf.org/arch/msg/bpf/i93lzdN3ewnzzS_JMbinCIYxAIU/ [5] https://mailarchive.ietf.org/arch/msg/bpf/KBWXbMeDcSrq4vsKR_KkBbV6hI4/ Acked-by: David Vernet <void@manifault.com> Signed-off-by: Dave Thaler <dthaler1968@googlemail.com> Link: https://lore.kernel.org/r/20240623150453.10613-1-dthaler1968@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> 
This patch does the following to address IETF feedback:

* Remove mention of "program type" and reference future
  docs (and mention platform-specific docs exist) for
  helper functions and BTF. Addresses Roman Danyliw's
  comments based on GENART review from Ines Robles [0].

* Add reference for endianness as requested by John
  Scudder [1].

* Added bit numbers to top of 32-bit wide format diagrams
  as requested by Paul Wouters [2].

* Added more text about why BPF doesn't stand for anything, based
  on text from ebpf.io [3], as requested by Eric Vyncke and
  Gunter Van de Velde [4].

* Replaced "htobe16" (and similar) and the direction-specific
  description with just "be16" (and similar) and a direction-agnostic
  description, to match the direction-agnostic description in
  the Byteswap Instructions section. Based on feedback from Eric
  Vyncke [5].

[0] https://mailarchive.ietf.org/arch/msg/bpf/DvDgDWOiwk05OyNlWlAmELZFPlM/

[1] https://mailarchive.ietf.org/arch/msg/bpf/eKNXpU4jCLjsbZDSw8LjI29M3tM/

[2] https://mailarchive.ietf.org/arch/msg/bpf/hGk8HkYxeZTpdu9qW_MvbGKj7WU/

[3] https://ebpf.io/what-is-ebpf/#what-do-ebpf-and-bpf-stand-for

[4] https://mailarchive.ietf.org/arch/msg/bpf/i93lzdN3ewnzzS_JMbinCIYxAIU/

[5] https://mailarchive.ietf.org/arch/msg/bpf/KBWXbMeDcSrq4vsKR_KkBbV6hI4/

Acked-by: David Vernet <void@manifault.com>
Signed-off-by: Dave Thaler <dthaler1968@googlemail.com>
Link: https://lore.kernel.org/r/20240623150453.10613-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
libbpf: Configure log verbosity with env variable 2024-05-28T23:25:06+00:00 Mykyta Yatsenko yatsenko@meta.com 2024-05-24T13:18:40+00:00 eb4e7726279a344c82e3c23be396bcfd0a4d5669 Configure logging verbosity by setting LIBBPF_LOG_LEVEL environment variable, which is applied only to default logger. Once user set their custom logging callback, it is up to them to handle filtering. Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20240524131840.114289-1-yatsenko@meta.com 
Configure logging verbosity by setting LIBBPF_LOG_LEVEL environment
variable, which is applied only to default logger. Once user set their
custom logging callback, it is up to them to handle filtering.

Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240524131840.114289-1-yatsenko@meta.com
bpf, docs: Fix instruction.rst indentation 2024-05-26T16:58:42+00:00 Dave Thaler dthaler1968@googlemail.com 2024-05-26T06:18:15+00:00 e245ef8a0b06fa5a18ca3bdcf82430166419283e The table captions patch corrected indented most tables to work with the table directive for adding a caption but missed two of them. Signed-off-by: Dave Thaler <dthaler1968@gmail.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Link: https://lore.kernel.org/r/20240526061815.22497-1-dthaler1968@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> 
The table captions patch corrected indented most tables to work with
the table directive for adding a caption but missed two of them.

Signed-off-by: Dave Thaler <dthaler1968@gmail.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20240526061815.22497-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
bpf, docs: Clarify call local offset 2024-05-25T17:41:57+00:00 Dave Thaler dthaler1968@googlemail.com 2024-05-25T15:33:32+00:00 f980f13e4eb299abba6692365315196e1ba6fd2c In the Jump instructions section it explains that the offset is "relative to the instruction following the jump instruction". But the program-local section confusingly said "referenced by offset from the call instruction, similar to JA". This patch updates that sentence with consistent wording, saying it's relative to the instruction following the call instruction. Signed-off-by: Dave Thaler <dthaler1968@gmail.com> Acked-by: Yonghong Song <yonghong.song@linux.dev> Link: https://lore.kernel.org/r/20240525153332.21355-1-dthaler1968@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> 
In the Jump instructions section it explains that the offset is
"relative to the instruction following the jump instruction".
But the program-local section confusingly said "referenced by
offset from the call instruction, similar to JA".

This patch updates that sentence with consistent wording, saying
it's relative to the instruction following the call instruction.

Signed-off-by: Dave Thaler <dthaler1968@gmail.com>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20240525153332.21355-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
bpf, docs: Add table captions 2024-05-25T17:39:31+00:00 Dave Thaler dthaler1968@googlemail.com 2024-05-24T16:46:18+00:00 6a6d8b6f00ade597e0030669fae3fdf57cfba33b As suggested by Ines Robles in his IETF GENART review at https://datatracker.ietf.org/doc/review-ietf-bpf-isa-02-genart-lc-robles-2024-05-16/ Signed-off-by: Dave Thaler <dthaler1968@gmail.com> Link: https://lore.kernel.org/r/20240524164618.18894-1-dthaler1968@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org> 
As suggested by Ines Robles in his IETF GENART review at
https://datatracker.ietf.org/doc/review-ietf-bpf-isa-02-genart-lc-robles-2024-05-16/

Signed-off-by: Dave Thaler <dthaler1968@gmail.com>
Link: https://lore.kernel.org/r/20240524164618.18894-1-dthaler1968@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>