AF_XDP features are tested by the test_xsk.sh script but not by the
test_progs framework. The tests used by the script are defined in
xksxceiver.c which can't be integrated in the test_progs framework as is.
Extract these test definitions from xskxceiver{.c/.h} to put them in new
test_xsk{.c/.h} files.
Keep the main() function and its unshared dependencies in xksxceiver to
avoid impacting the test_xsk.sh script which is often used to test real
hardware.
Move ksft_test_result_*() calls to xskxceiver.c to keep the kselftest's
report valid
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Bastien Curutchet (eBPF Foundation) <bastien.curutchet@bootlin.com>
Link: https://lore.kernel.org/r/20251031-xsk-v7-1-39fe486593a3@bootlin.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
BPF stream kfuncs need to be non-sleeping as they can be called from
programs running in any context, this requires a way to allocate memory
from any context. Currently, this is done by a custom per-CPU NMI-safe
bump allocation mechanism, backed by alloc_pages_nolock() and
free_pages_nolock() primitives.
As kmalloc_nolock() and kfree_nolock() primitives are available now, the
custom allocator can be removed in favor of these.
Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Acked-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20251023161448.4263-1-puranjay@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Kumar Kartikeya Dwivedi says:
====================
Misc rqspinlock updates
A couple of changes for rqspinlock, the first disables propagation of AA
and ABBA deadlocks to waiters succeeding the deadlocking waiter. A more
verbose rationale is available in the commit log. The second commit
expands the stress test to introduce a ABBCCA mode that will reliably
exercise the timeout fallback.
====================
Link: https://lore.kernel.org/r/20251029181828.231529-1-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Introduce a new mode for the rqspinlock stress test that exercises a
deadlock that won't be detected by the AA and ABBA checks, such that we
always reliably trigger the timeout fallback. We need 4 CPUs for this
particular case, as CPU 0 is untouched, and three participant CPUs for
triggering the ABBCCA case.
Refactor the lock acquisition paths in the module to better reflect the
three modes and choose the right lock depending on the context.
Also drop ABBA case from running by default as part of test progs, since
the stress test can consume a significant amount of time.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Reviewed-by: Amery Hung <ameryhung@gmail.com>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20251029181828.231529-3-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Disable propagation and unwinding of the waiter queue in case the head
waiter detects a deadlock condition, but keep it enabled in case of the
timeout fallback.
Currently, when the head waiter experiences an AA deadlock, it will
signal all its successors in the queue to exit with an error. This is
not ideal for cases where the same lock is held in contexts which can
cause errors in an unrestricted fashion (e.g., BPF programs, or kernel
paths invoked through BPF programs), and core kernel logic which is
written in a correct fashion and does not expect deadlocks.
The same reasoning can be extended to ABBA situations. Depending on the
actual runtime schedule, one or both of the head waiters involved in an
ABBA situation can detect and exit directly without terminating their
waiter queue. If the ABBA situation manifests again, the waiters will
keep exiting until progress can be made, or a timeout is triggered in
case of more complicated locking dependencies.
We still preserve the queue destruction in case of timeouts, as either
the locking dependencies are too complex to be captured by AA and ABBA
heuristics, or the owner is perpetually stuck. As such, it would be
unwise to continue to apply the timeout for each new head waiter without
terminating the queue, since we may end up waiting for more than 250 ms
in aggregate with all participants in the locking transaction.
The patch itself is fairly simple; we can simply signal our successor to
become the next head waiter, and leave the queue without attempting to
acquire the lock.
With this change, the behavior for waiters in case of deadlocks
experienced by a predecessor changes. It is guaranteed that call sites
will no longer receive errors if the predecessors encounter deadlocks
and the successors do not participate in one. This should lower the
failure rate for waiters that are not doing improper locking opreations,
just because they were unlucky to queue behind a misbehaving waiter.
However, timeouts are still a possibility, hence they must be accounted
for, so users cannot rely upon errors not occuring at all.
Suggested-by: Amery Hung <ameryhung@gmail.com>
Suggested-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20251029181828.231529-2-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
file_reader/on_open_expect_fault intermittently fails when test_progs
runs tests in parallel, because it expects a page fault on first read.
Another file_reader test running concurrently may have already pulled
the same pages into the page cache, eliminating the fault and causing a
spurious failure.
Make file_reader/on_open_expect_fault read from a file region that does
not overlap with other file_reader tests, so the initial access still
faults even under parallel execution.
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Acked-by: Ihor Solodrai <ihor.solodrai@linux.dev>
Link: https://lore.kernel.org/r/20251029195907.858217-1-mykyta.yatsenko5@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Alexis Lothoré says:
====================
Hello,
this is the v3 of test_tc_tunnel conversion into test_progs framework.
This new revision:
- fixes a few issues spotted by the bot reviewer
- removes any test ensuring connection failure (and so depending on a
timout) to keep the execution time reasonable
test_tc_tunnel.sh tests a variety of tunnels based on BPF: packets are
encapsulated by a BPF program on the client egress. We then check that
those packets can be decapsulated on server ingress side, either thanks
to kernel-based or BPF-based decapsulation. Those tests are run thanks
to two veths in two dedicated namespaces.
- patches 1 and 2 are preparatory patches
- patch 3 introduce tc_tunnel test into test_progs
- patch 4 gets rid of the test_tc_tunnel.sh script
The new test has been executed both in some x86 local qemu machine, as
well as in CI:
# ./test_progs -a tc_tunnel
#454/1 tc_tunnel/ipip_none:OK
#454/2 tc_tunnel/ipip6_none:OK
#454/3 tc_tunnel/ip6tnl_none:OK
#454/4 tc_tunnel/sit_none:OK
#454/5 tc_tunnel/vxlan_eth:OK
#454/6 tc_tunnel/ip6vxlan_eth:OK
#454/7 tc_tunnel/gre_none:OK
#454/8 tc_tunnel/gre_eth:OK
#454/9 tc_tunnel/gre_mpls:OK
#454/10 tc_tunnel/ip6gre_none:OK
#454/11 tc_tunnel/ip6gre_eth:OK
#454/12 tc_tunnel/ip6gre_mpls:OK
#454/13 tc_tunnel/udp_none:OK
#454/14 tc_tunnel/udp_eth:OK
#454/15 tc_tunnel/udp_mpls:OK
#454/16 tc_tunnel/ip6udp_none:OK
#454/17 tc_tunnel/ip6udp_eth:OK
#454/18 tc_tunnel/ip6udp_mpls:OK
#454 tc_tunnel:OK
Summary: 1/18 PASSED, 0 SKIPPED, 0 FAILED
====================
Link: https://patch.msgid.link/20251027-tc_tunnel-v3-0-505c12019f9d@bootlin.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
The test_tc_tunnel.sh script checks that a large variety of tunneling
mechanisms handled by the kernel can be handled as well by eBPF
programs. While this test shares similarities with test_tunnel.c (which
is already integrated in test_progs), those are testing slightly
different things:
- test_tunnel.c creates a tunnel interface, and then get and set tunnel
keys in packet metadata, from BPF programs.
- test_tc_tunnels.sh manually parses/crafts packets content
Bring the tests covered by test_tc_tunnel.sh into the test_progs
framework, by creating a dedicated test_tc_tunnel.sh. This new test
defines a "generic" runner which, for each test configuration:
- will configure the relevant veth pair, each of those isolated in a
dedicated namespace
- will check that traffic will fail if there is only an encapsulating
program attached to one veth egress
- will check that traffic succeed if we enable some decapsulation module
on kernel side
- will check that traffic still succeeds if we replace the kernel
decapsulation with some eBPF ingress decapsulation.
Example of the new test execution:
# ./test_progs -a tc_tunnel
#447/1 tc_tunnel/ipip_none:OK
#447/2 tc_tunnel/ipip6_none:OK
#447/3 tc_tunnel/ip6tnl_none:OK
#447/4 tc_tunnel/sit_none:OK
#447/5 tc_tunnel/vxlan_eth:OK
#447/6 tc_tunnel/ip6vxlan_eth:OK
#447/7 tc_tunnel/gre_none:OK
#447/8 tc_tunnel/gre_eth:OK
#447/9 tc_tunnel/gre_mpls:OK
#447/10 tc_tunnel/ip6gre_none:OK
#447/11 tc_tunnel/ip6gre_eth:OK
#447/12 tc_tunnel/ip6gre_mpls:OK
#447/13 tc_tunnel/udp_none:OK
#447/14 tc_tunnel/udp_eth:OK
#447/15 tc_tunnel/udp_mpls:OK
#447/16 tc_tunnel/ip6udp_none:OK
#447/17 tc_tunnel/ip6udp_eth:OK
#447/18 tc_tunnel/ip6udp_mpls:OK
#447 tc_tunnel:OK
Summary: 1/18 PASSED, 0 SKIPPED, 0 FAILED
Signed-off-by: Alexis Lothoré (eBPF Foundation) <alexis.lothore@bootlin.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20251027-tc_tunnel-v3-3-505c12019f9d@bootlin.com
When trying to run bpf-based encapsulation in a s390x environment, some
parts of test_tc_tunnel.bpf.o do not encapsulate correctly the traffic,
leading to tests failures. Adding some logs shows for example that
packets about to be sent on an interface with the ip6vxlan_eth program
attached do not have the expected value 5 in the ip header ihl field,
and so are ignored by the program.
This phenomenon appears when trying to cross-compile the selftests,
rather than compiling it from a virtualized host: the selftests build
system may then wrongly pick some host headers. If <asm/byteorder.h>
ends up being picked on the host (and if the host has a endianness
different from the target one), it will then expose wrong endianness
defines (e.g __LITTLE_ENDIAN_BITFIELD instead of __BIT_ENDIAN_BITFIELD),
and it will for example mess up the iphdr structure layout used in the
ebpf program.
To prevent this, directly use the vmlinux.h header generated by the
selftests build system rather than including directly specific kernel
headers. As a consequence, add some missing definitions that are not
exposed by vmlinux.h, and adapt the bitfield manipulations to allow
building and using the program on both types of platforms.
Signed-off-by: Alexis Lothoré (eBPF Foundation) <alexis.lothore@bootlin.com>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20251027-tc_tunnel-v3-2-505c12019f9d@bootlin.com
In the elf_sec_data() function, the input parameter 'scn' will be
evaluated. If it is NULL, then it will directly return NULL. Therefore,
the return value of the elf_sec_data() function already takes into
account the case where the input parameter scn is NULL. Therefore,
subsequently, the code only needs to check whether the return value of
the elf_sec_data() function is NULL.
Signed-off-by: Jianyun Gao <jianyungao89@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/bpf/20251024080802.642189-1-jianyungao89@gmail.com
Extract the duplicated maximum allowed depth computation for stack
traces stored in BPF stacks from bpf_get_stackid() and __bpf_get_stack()
into a dedicated stack_map_calculate_max_depth() helper function.
This unifies the logic for:
- The max depth computation
- Enforcing the sysctl_perf_event_max_stack limit
No functional changes for existing code paths.
Signed-off-by: Arnaud Lecomte <contact@arnaud-lcm.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Acked-by: Song Liu <song@kernel.org>
Link: https://lore.kernel.org/bpf/20251025192858.31424-1-contact@arnaud-lcm.com
Mykyta Yatsenko says:
====================
bpf: Introduce file dynptr
From: Mykyta Yatsenko <yatsenko@meta.com>
This series adds a new dynptr kind, file dynptr, which enables BPF
programs to perform safe reads from files in a structured way.
Initial motivations include:
* Parsing the executable’s ELF to locate thread-local variable symbols
* Capturing stack traces when frame pointers are disabled
By leveraging the existing dynptr abstraction, we reuse the verifier’s
lifetime/size checks and keep the API consistent with existing dynptr
read helpers.
Technical details:
1. Reuses the existing freader library to read files a folio at a time.
2. bpf_dynptr_slice() and bpf_dynptr_read() always copy data from folios
into a program-provided buffer; zero-copy access is intentionally not
supported to keep it simple.
3. Reads may sleep if the requested folios are not in the page cache.
4. Few verifier changes required:
* Support dynptr destruction in kfuncs
* Add kfunc address substitution based on whether the program runs in
a sleepable or non-sleepable context.
Testing:
The final patch adds a selftest that validates BPF program reads the
same data as userspace, page faults are enabled in sleepable context and
disabled in non-sleepable.
Changelog:
---
v4 -> v5
v4: https://lore.kernel.org/all/20251021200334.220542-1-mykyta.yatsenko5@gmail.com/
* Inlined and removed kfunc_call_imm(), run overflow check for call_imm
only if !bpf_jit_supports_far_kfunc_call().
v3 -> v4
v3: https://lore.kernel.org/bpf/20251020222538.932915-1-mykyta.yatsenko5@gmail.com/
* Remove ringbuf usage from selftests
* bpf_dynptr_set_null(ptr) when discarding file dynptr
* call kfunc_call_imm() in specialize_kfunc() only, removed
call from add_kfunc_call()
v2 -> v3
v2: https://lore.kernel.org/bpf/20251015161155.120148-1-mykyta.yatsenko5@gmail.com/
* Add negative tests
* Rewrote tests to use LSM for bpf_get_task_exe_file()
* Move call_imm overflow check into kfunc_call_imm()
v1 -> v2
v1: https://lore.kernel.org/bpf/20251003160416.585080-1-mykyta.yatsenko5@gmail.com/
* Remove ELF parsing selftest
* Expanded u32 -> u64 refactoring, changes in include/uapi/linux/bpf.h
* Removed freader.{c,h}, instead move freader definitions into
buildid.h.
* Small refactoring of the multiple folios reading algorithm
* Directly return error after unmark_stack_slots_dynptr().
* Make kfuncs receive trusted arguments.
* Remove enum bpf_is_sleepable, use bool instead
* Remove unnecessary sorting from specialize_kfunc()
* Remove bool kfunc_in_sleepable_ctx; field from the struct
bpf_insn_aux_data, rely on non_sleepable field introduced by Kumar
* Refactor selftests, do madvise(...MADV_PAGEOUT) for all pages read by
the test
* Introduce the test for non-sleepable case, verify it fails with -EFAULT
====================
Link: https://lore.kernel.org/r/20251026203853.135105-1-mykyta.yatsenko5@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
File dynptr reads may sleep when the requested folios are not in
the page cache. To avoid sleeping in non-sleepable contexts while still
supporting valid sleepable use, given that dynptrs are non-sleepable by
default, enable sleeping only when bpf_dynptr_from_file() is invoked
from a sleepable context.
This change:
* Introduces a sleepable constructor: bpf_dynptr_from_file_sleepable()
* Override non-sleepable constructor with sleepable if it's always
called in sleepable context
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251026203853.135105-10-mykyta.yatsenko5@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Move kfunc specialization (function address substitution) to later stage
of verification to support a new use case, where we need to take into
consideration whether kfunc is called in sleepable context.
Minor refactoring in add_kfunc_call(), making sure that if function
fails, kfunc desc is not added to tab->descs (previously it could be
added or not, depending on what failed).
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251026203853.135105-9-mykyta.yatsenko5@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add support for file dynptr.
Introduce struct bpf_dynptr_file_impl to hold internal state for file
dynptrs, with 64-bit size and offset support.
Introduce lifecycle management kfuncs:
- bpf_dynptr_from_file() for initialization
- bpf_dynptr_file_discard() for destruction
Extend existing helpers to support file dynptrs in:
- bpf_dynptr_read()
- bpf_dynptr_slice()
Write helpers (bpf_dynptr_write() and bpf_dynptr_data()) are not
modified, as file dynptr is read-only.
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251026203853.135105-8-mykyta.yatsenko5@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add the necessary verifier plumbing for the new file-backed dynptr type.
Introduce two kfuncs for its lifecycle management:
* bpf_dynptr_from_file() for initialization
* bpf_dynptr_file_discard() for destruction
Currently there is no mechanism for kfunc to release dynptr, this patch
add one:
* Dynptr release function sets meta->release_regno
* Call unmark_stack_slots_dynptr() if meta->release_regno is set and
dynptr ref_obj_id is set as well.
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251026203853.135105-7-mykyta.yatsenko5@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
freader_fetch currently reads from at most two folios. When a read spans
into a third folio, the overflow bytes are copied adjacent to the second
folio’s data instead of being handled as a separate folio.
This patch modifies fetch algorithm to support reading from many folios.
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Reviewed-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20251026203853.135105-5-mykyta.yatsenko5@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Dynptr currently caps size and offset at 24 bits, which isn’t sufficient
for file-backed use cases; even 32 bits can be limiting. Refactor dynptr
helpers/kfuncs to use 64-bit size and offset, ensuring consistency
across the APIs.
This change does not affect internals of xdp, skb or other dynptrs,
which continue to behave as before. Also it does not break binary
compatibility.
The widening enables large-file access support via dynptr, implemented
in the next patches.
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251026203853.135105-3-mykyta.yatsenko5@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The bpf_insn_successors() function is used to return successors
to a BPF instruction. So far, an instruction could have 0, 1 or 2
successors. Prepare the verifier code to introduction of instructions
with more than 2 successors (namely, indirect jumps).
To do this, introduce a new struct, struct bpf_iarray, containing
an array of bpf instruction indexes and make bpf_insn_successors
to return a pointer of that type. The storage for all instructions
is allocated in the env->succ, which holds an array of size 2,
to be used for all instructions.
Signed-off-by: Anton Protopopov <a.s.protopopov@gmail.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251019202145.3944697-10-a.s.protopopov@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The kernel/bpf/array.c file defines the array_map_get_next_key()
function which finds the next key for array maps. It actually doesn't
use any map fields besides the generic max_entries field. Generalize
it, and export as bpf_array_get_next_key() such that it can be
re-used by other array-like maps.
Signed-off-by: Anton Protopopov <a.s.protopopov@gmail.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251019202145.3944697-4-a.s.protopopov@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Introduce a new subprog_start field in bpf_prog_aux. This field may
be used by JIT compilers wanting to know the real absolute xlated
offset of the function being jitted. The func_info[func_id] may have
served this purpose, but func_info may be NULL, so JIT compilers
can't rely on it.
Signed-off-by: Anton Protopopov <a.s.protopopov@gmail.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251019202145.3944697-3-a.s.protopopov@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
propagate_to_outer_instance() calls get_outer_instance() and uses the
returned pointer to reset and commit stack write marks. Under normal
conditions, update_instance() guarantees that an outer instance exists,
so get_outer_instance() cannot return an ERR_PTR.
However, explicitly checking for IS_ERR(outer_instance) makes this code
more robust and self-documenting. It reduces cognitive load when reading
the control flow and silences potential false-positive reports from
static analysis or automated tooling.
No functional change intended.
Signed-off-by: Shardul Bankar <shardulsb08@gmail.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20251021080849.860072-1-shardulsb08@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Yinhao et al. reported that their fuzzer tool was able to trigger a
skb_warn_bad_offload() from netif_skb_features() -> gso_features_check().
When a BPF program - triggered via BPF test infra - pushes the packet
to the loopback device via bpf_clone_redirect() then mentioned offload
warning can be seen. GSO-related features are then rightfully disabled.
We get into this situation due to convert___skb_to_skb() setting
gso_segs and gso_size but not gso_type. Technically, it makes sense
that this warning triggers since the GSO properties are malformed due
to the gso_type. Potentially, the gso_type could be marked non-trustworthy
through setting it at least to SKB_GSO_DODGY without any other specific
assumptions, but that also feels wrong given we should not go further
into the GSO engine in the first place.
The checks were added in 121d57af30 ("gso: validate gso_type in GSO
handlers") because there were malicious (syzbot) senders that combine
a protocol with a non-matching gso_type. If we would want to drop such
packets, gso_features_check() currently only returns feature flags via
netif_skb_features(), so one location for potentially dropping such skbs
could be validate_xmit_unreadable_skb(), but then otoh it would be
an additional check in the fast-path for a very corner case. Given
bpf_clone_redirect() is the only place where BPF test infra could emit
such packets, lets reject them right there.
Fixes: 850a88cc40 ("bpf: Expose __sk_buff wire_len/gso_segs to BPF_PROG_TEST_RUN")
Fixes: cf62089b0e ("bpf: Add gso_size to __sk_buff")
Reported-by: Yinhao Hu <dddddd@hust.edu.cn>
Reported-by: Kaiyan Mei <M202472210@hust.edu.cn>
Reported-by: Dongliang Mu <dzm91@hust.edu.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/20251020075441.127980-1-daniel@iogearbox.net
The __list_del fuction doesn't set the previous node's next pointer to
the next node of the node to be deleted. It just updates the local variable
and not the actual pointer in the previous node.
The test was passing up till now because the bpf code is doing bpf_free()
after list_del and therfore reading head->first from the userspace will
read all zeroes. But after arena_list_del() is finished, head->first should
point to NULL;
Signed-off-by: Puranjay Mohan <puranjay@kernel.org>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20251017141727.51355-1-puranjay@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
do_hbm_test.sh:
The comment incorrectly used "upcomming" instead of "upcoming".
hbm.c
The comment incorrectly used "Managment" instead of "Management".
The comment incorrectly used "Currrently" instead of "Currently".
tcp_cong_kern.c
The comment incorrectly used "deteremined" instead of "determined".
tracex1.bpf.c
The comment incorrectly used "loobpack" instead of "loopback".
Signed-off-by: Chu Guangqing <chuguangqing@inspur.com>
Link: https://lore.kernel.org/r/20251015015024.2212-2-chuguangqing@inspur.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The vma->vm_mm might be NULL and it can be accessed outside of RCU. Thus,
we can mark it as trusted_or_null. With this change, BPF helpers can safely
access vma->vm_mm to retrieve the associated mm_struct from the VMA.
Then we can make policy decision from the VMA.
The "trusted" annotation enables direct access to vma->vm_mm within kfuncs
marked with KF_TRUSTED_ARGS or KF_RCU, such as bpf_task_get_cgroup1() and
bpf_task_under_cgroup(). Conversely, "null" enforcement requires all
callsites using vma->vm_mm to perform NULL checks.
The lsm selftest must be modified because it directly accesses vma->vm_mm
without a NULL pointer check; otherwise it will break due to this
change.
For the VMA based THP policy, the use case is as follows,
@mm = @vma->vm_mm; // vm_area_struct::vm_mm is trusted or null
if (!@mm)
return;
bpf_rcu_read_lock(); // rcu lock must be held to dereference the owner
@owner = @mm->owner; // mm_struct::owner is rcu trusted or null
if (!@owner)
goto out;
@cgroup1 = bpf_task_get_cgroup1(@owner, MEMCG_HIERARCHY_ID);
/* make the decision based on the @cgroup1 attribute */
bpf_cgroup_release(@cgroup1); // release the associated cgroup
out:
bpf_rcu_read_unlock();
PSI memory information can be obtained from the associated cgroup to inform
policy decisions. Since upstream PSI support is currently limited to cgroup
v2, the following example demonstrates cgroup v2 implementation:
@owner = @mm->owner;
if (@owner) {
// @ancestor_cgid is user-configured
@ancestor = bpf_cgroup_from_id(@ancestor_cgid);
if (bpf_task_under_cgroup(@owner, @ancestor)) {
@psi_group = @ancestor->psi;
/* Extract PSI metrics from @psi_group and
* implement policy logic based on the values
*/
}
}
The vma::vm_file can also be marked with __safe_trusted_or_null.
No additional selftests are required since vma->vm_file and vma->vm_mm are
already validated in the existing selftest suite.
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Link: https://lore.kernel.org/r/20251016063929.13830-3-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
When CONFIG_MEMCG is enabled, we can access mm->owner under RCU. The
owner can be NULL. With this change, BPF helpers can safely access
mm->owner to retrieve the associated task from the mm. We can then make
policy decision based on the task attribute.
The typical use case is as follows,
bpf_rcu_read_lock(); // rcu lock must be held for rcu trusted field
@owner = @mm->owner; // mm_struct::owner is rcu trusted or null
if (!@owner)
goto out;
/* Do something based on the task attribute */
out:
bpf_rcu_read_unlock();
Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Link: https://lore.kernel.org/r/20251016063929.13830-2-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
There are some set but not used build errors when compiling bpf selftests
with the latest upstream mainline GCC, at the beginning add the attribute
__maybe_unused for the variables, but it is better to just add the option
-Wno-unused-but-set-variable to CFLAGS in Makefile to disable the errors
instead of hacking the tests.
tools/testing/selftests/bpf/map_tests/lpm_trie_map_basic_ops.c:229:36:
error: variable ‘n_matches_after_delete’ set but not used [-Werror=unused-but-set-variable=]
tools/testing/selftests/bpf/map_tests/lpm_trie_map_basic_ops.c:229:25:
error: variable ‘n_matches’ set but not used [-Werror=unused-but-set-variable=]
tools/testing/selftests/bpf/prog_tests/bpf_cookie.c:426:22:
error: variable ‘j’ set but not used [-Werror=unused-but-set-variable=]
tools/testing/selftests/bpf/prog_tests/find_vma.c:52:22:
error: variable ‘j’ set but not used [-Werror=unused-but-set-variable=]
tools/testing/selftests/bpf/prog_tests/perf_branches.c:67:22:
error: variable ‘j’ set but not used [-Werror=unused-but-set-variable=]
tools/testing/selftests/bpf/prog_tests/perf_link.c:15:22:
error: variable ‘j’ set but not used [-Werror=unused-but-set-variable=]
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Link: https://lore.kernel.org/r/20251018082815.20622-1-yangtiezhu@loongson.cn
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Pull rustfmt fixes from Miguel Ojeda:
"Rust 'rustfmt' cleanup
'rustfmt', by default, formats imports in a way that is prone to
conflicts while merging and rebasing, since in some cases it condenses
several items into the same line.
Document in our guidelines that we will handle this for the moment
with the trailing empty comment workaround and make the tree
'rustfmt'-clean again"
* tag 'rust-rustfmt' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux:
rust: bitmap: fix formatting
rust: cpufreq: fix formatting
rust: alloc: employ a trailing comment to keep vertical layout
docs: rust: add section on imports formatting
Pull tpm fix from Jarkko Sakkinen:
"Correct the state transitions for ARM FF-A to match the spec and how
tpm_crb behaves on other platforms"
* tag 'tpmdd-next-v6.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd:
tpm_crb: Add idle support for the Arm FF-A start method
Pull pci fixes from Bjorn Helgaas:
- Search for MSI Capability with correct ID to fix an MSI regression on
platforms with Cadence IP (Hans Zhang)
- Revert early bridge resource set up to fix resource assignment
failures that broke at least alpha boot and Snapdragon ath12k WiFi
(Ilpo Järvinen)
- Implement VMD .irq_startup()/.irq_shutdown() to fix IRQ issues that
caused boot crashes and broken devices below VMD (Inochi Amaoto)
- Select CONFIG_SCREEN_INFO on X86 to fix black screen on boot when
SCREEN_INFO not selected (Mario Limonciello)
* tag 'pci-v6.18-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
PCI/VGA: Select SCREEN_INFO on X86
PCI: vmd: Override irq_startup()/irq_shutdown() in vmd_init_dev_msi_info()
PCI: Revert early bridge resource set up
PCI: cadence: Search for MSI Capability with correct ID
Pull Compute Express Link fixes from Dave Jiang:
"A small collection of CXL fixes. In addition to some misc fixes for
the CXL subsystem, a number of fixes for CXL extended linear cache
support are included to make it functional again.
- Avoid missing port component registers setup due to dport
enumeration failure
- Add check for no entries in cxl_feature_info to address accessing
invalid pointer.
- Use %pa printk format to emit resource_size_t in
validate_region_offset()
CXL extended linear cache support fixes:
- Fix setup of memory resource in cxl_acpi_set_cache_size()
- Set range param for region_res_match_cxl_range() as const
(addresses a compile warning for match_region_by_range() fix)
- Fix match_region_by_range() to use region_res_match_cxl_range()
- Subtract to find an hpa_alias0 in cxl_poison events to correct the
alias math calculation"
* tag 'cxl-fixes-6.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl:
cxl/trace: Subtract to find an hpa_alias0 in cxl_poison events
cxl/region: Use %pa printk format to emit resource_size_t
cxl: Fix match_region_by_range() to use region_res_match_cxl_range()
cxl: Set range param for region_res_match_cxl_range() as const
cxl/acpi: Fix setup of memory resource in cxl_acpi_set_cache_size()
cxl/features: Add check for no entries in cxl_feature_info
cxl/port: Avoid missing port component registers setup
