Daniel Borkmann says:
====================
bpf-next-for-netdev
The following pull-request contains BPF updates for your *net-next* tree.
We've added 22 non-merge commits during the last 3 day(s) which contain
a total of 23 files changed, 652 insertions(+), 431 deletions(-).
The main changes are:
1) Add verifier support for annotating user's global BPF subprogram arguments
with few commonly requested annotations for a better developer experience,
from Andrii Nakryiko.
These tags are:
- Ability to annotate a special PTR_TO_CTX argument
- Ability to annotate a generic PTR_TO_MEM as non-NULL
2) Support BPF verifier tracking of BPF_JNE which helps cases when the compiler
transforms (unsigned) "a > 0" into "if a == 0 goto xxx" and the like, from
Menglong Dong.
3) Fix a warning in bpf_mem_cache's check_obj_size() as reported by LKP, from Hou Tao.
4) Re-support uid/gid options when mounting bpffs which had to be reverted with
the prior token series revert to avoid conflicts, from Daniel Borkmann.
5) Fix a libbpf NULL pointer dereference in bpf_object__collect_prog_relos() found
from fuzzing the library with malformed ELF files, from Mingyi Zhang.
6) Skip DWARF sections in libbpf's linker sanity check given compiler options to
generate compressed debug sections can trigger a rejection due to misalignment,
from Alyssa Ross.
7) Fix an unnecessary use of the comma operator in BPF verifier, from Simon Horman.
8) Fix format specifier for unsigned long values in cpustat sample, from Colin Ian King.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
For a clean, conflict-free revert of the token-related patches in commit
d17aff807f ("Revert BPF token-related functionality"), the bpf fs commit
750e785796 ("bpf: Support uid and gid when mounting bpffs") was undone
temporarily as well.
This patch manually re-adds the functionality from the original one back
in 750e785796, no other functional changes intended.
Testing:
# mount -t bpf -o uid=65534,gid=65534 bpffs ./foo
# ls -la . | grep foo
drwxrwxrwt 2 nobody nogroup 0 Dec 20 13:16 foo
# mount -t bpf
bpffs on /root/foo type bpf (rw,relatime,uid=65534,gid=65534)
Also, passing invalid arguments for uid/gid are properly rejected as expected.
Fixes: d17aff807f ("Revert BPF token-related functionality")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Cc: Jie Jiang <jiejiang@chromium.org>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: linux-fsdevel@vger.kernel.org
Link: https://lore.kernel.org/bpf/20231220133805.20953-1-daniel@iogearbox.net
Only the posix timer system calls use this (when the posix timer support
is disabled, which does not actually happen in any normal case), because
they had debug code to print out a warning about missing system calls.
Get rid of that special case, and just use the standard COND_SYSCALL
interface that creates weak system call stubs that return -ENOSYS for
when the system call does not exist.
This fixes a kCFI issue with the SYS_NI() hackery:
CFI failure at int80_emulation+0x67/0xb0 (target: sys_ni_posix_timers+0x0/0x70; expected type: 0xb02b34d9)
WARNING: CPU: 0 PID: 48 at int80_emulation+0x67/0xb0
Reported-by: kernel test robot <oliver.sang@intel.com>
Reviewed-by: Sami Tolvanen <samitolvanen@google.com>
Tested-by: Sami Tolvanen <samitolvanen@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Borislav Petkov <bp@alien8.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
At present, bpf memory allocator uses check_obj_size() to ensure that
ksize() of allocated pointer is equal with the unit_size of used
bpf_mem_cache. Its purpose is to prevent bpf_mem_free() from selecting
a bpf_mem_cache which has different unit_size compared with the
bpf_mem_cache used for allocation. But as reported by lkp, the return
value of ksize() or kmalloc_size_roundup() may change due to slab merge
and it will lead to the warning report in check_obj_size().
The reported warning happened as follows:
(1) in bpf_mem_cache_adjust_size(), kmalloc_size_roundup(96) returns the
object_size of kmalloc-96 instead of kmalloc-cg-96. The object_size of
kmalloc-96 is 96, so size_index for 96 is not adjusted accordingly.
(2) the object_size of kmalloc-cg-96 is adjust from 96 to 128 due to
slab merge in __kmem_cache_alias(). For SLAB, SLAB_HWCACHE_ALIGN is
enabled by default for kmalloc slab, so align is 64 and size is 128 for
kmalloc-cg-96. SLUB has a similar merge logic, but its object_size will
not be changed, because its align is 8 under x86-64.
(3) when unit_alloc() does kmalloc_node(96, __GFP_ACCOUNT, node),
ksize() returns 128 instead of 96 for the returned pointer.
(4) the warning in check_obj_size() is triggered.
Considering the slab merge can happen in anytime (e.g, a slab created in
a new module), the following case is also possible: during the
initialization of bpf_global_ma, there is no slab merge and ksize() for
a 96-bytes object returns 96. But after that a new slab created by a
kernel module is merged to kmalloc-cg-96 and the object_size of
kmalloc-cg-96 is adjust from 96 to 128 (which is possible for x86-64 +
CONFIG_SLAB, because its alignment requirement is 64 for 96-bytes slab).
So soon or later, when bpf_global_ma frees a 96-byte-sized pointer
which is allocated from bpf_mem_cache with unit_size=96, bpf_mem_free()
will free the pointer through a bpf_mem_cache in which unit_size is 128,
because the return value of ksize() changes. The warning for the
mismatch will be triggered again.
A feasible fix is introducing similar APIs compared with ksize() and
kmalloc_size_roundup() to return the actually-allocated size instead of
size which may change due to slab merge, but it will introduce
unnecessary dependency on the implementation details of mm subsystem.
As for now the pointer of bpf_mem_cache is saved in the 8-bytes area
(or 4-bytes under 32-bit host) above the returned pointer, using
unit_size in the saved bpf_mem_cache to select the target cache instead
of inferring the size from the pointer itself. Beside no extra
dependency on mm subsystem, the performance for bpf_mem_free_rcu() is
also improved as shown below.
Before applying the patch, the performances of bpf_mem_alloc() and
bpf_mem_free_rcu() on 8-CPUs VM with one producer are as follows:
kmalloc : alloc 11.69 ± 0.28M/s free 29.58 ± 0.93M/s
percpu : alloc 14.11 ± 0.52M/s free 14.29 ± 0.99M/s
After apply the patch, the performance for bpf_mem_free_rcu() increases
9% and 146% for kmalloc memory and per-cpu memory respectively:
kmalloc: alloc 11.01 ± 0.03M/s free 32.42 ± 0.48M/s
percpu: alloc 12.84 ± 0.12M/s free 35.24 ± 0.23M/s
After the fixes, there is no need to adjust size_index to fix the
mismatch between allocation and free, so remove it as well. Also return
NULL instead of ZERO_SIZE_PTR for zero-sized alloc in bpf_mem_alloc(),
because there is no bpf_mem_cache pointer saved above ZERO_SIZE_PTR.
Fixes: 9077fc228f ("bpf: Use kmalloc_size_roundup() to adjust size_index")
Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/bpf/202310302113.9f8fe705-oliver.sang@intel.com
Signed-off-by: Hou Tao <houtao1@huawei.com>
Link: https://lore.kernel.org/r/20231216131052.27621-2-houtao@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add ability to pass a pointer to dynptr into global functions.
This allows to have global subprogs that accept and work with generic
dynptrs that are created by caller. Dynptr argument is detected based on
the name of a struct type, if it's "bpf_dynptr", it's assumed to be
a proper dynptr pointer. Both actual struct and forward struct
declaration types are supported.
This is conceptually exactly the same semantics as
bpf_user_ringbuf_drain()'s use of dynptr to pass a variable-sized
pointer to ringbuf record. So we heavily rely on CONST_PTR_TO_DYNPTR
bits of already existing logic in the verifier.
During global subprog validation, we mark such CONST_PTR_TO_DYNPTR as
having LOCAL type, as that's the most unassuming type of dynptr and it
doesn't have any special helpers that can try to free or acquire extra
references (unlike skb, xdp, or ringbuf dynptr). So that seems like a safe
"choice" to make from correctness standpoint. It's still possible to
pass any type of dynptr to such subprog, though, because generic dynptr
helpers, like getting data/slice pointers, read/write memory copying
routines, dynptr adjustment and getter routines all work correctly with
any type of dynptr.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20231215011334.2307144-8-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add support for annotating global BPF subprog arguments to provide more
information about expected semantics of the argument. Currently,
verifier relies purely on argument's BTF type information, and supports
three general use cases: scalar, pointer-to-context, and
pointer-to-fixed-size-memory.
Scalar and pointer-to-fixed-mem work well in practice and are quite
natural to use. But pointer-to-context is a bit problematic, as typical
BPF users don't realize that they need to use a special type name to
signal to verifier that argument is not just some pointer, but actually
a PTR_TO_CTX. Further, even if users do know which type to use, it is
limiting in situations where the same BPF program logic is used across
few different program types. Common case is kprobes, tracepoints, and
perf_event programs having a helper to send some data over BPF perf
buffer. bpf_perf_event_output() requires `ctx` argument, and so it's
quite cumbersome to share such global subprog across few BPF programs of
different types, necessitating extra static subprog that is context
type-agnostic.
Long story short, there is a need to go beyond types and allow users to
add hints to global subprog arguments to define expectations.
This patch adds such support for two initial special tags:
- pointer to context;
- non-null qualifier for generic pointer arguments.
All of the above came up in practice already and seem generally useful
additions. Non-null qualifier is an often requested feature, which
currently has to be worked around by having unnecessary NULL checks
inside subprogs even if we know that arguments are never NULL. Pointer
to context was discussed earlier.
As for implementation, we utilize btf_decl_tag attribute and set up an
"arg:xxx" convention to specify argument hint. As such:
- btf_decl_tag("arg:ctx") is a PTR_TO_CTX hint;
- btf_decl_tag("arg:nonnull") marks pointer argument as not allowed to
be NULL, making NULL check inside global subprog unnecessary.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20231215011334.2307144-7-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Remove duplicated BTF parsing logic when it comes to subprog call check.
Instead, use (potentially cached) results of btf_prepare_func_args() to
abstract away expectations of each subprog argument in generic terms
(e.g., "this is pointer to context", or "this is a pointer to memory of
size X"), and then use those simple high-level argument type
expectations to validate actual register states to check if they match
expectations.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20231215011334.2307144-6-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Subprog call logic in btf_check_subprog_call() currently has both a lot
of BTF parsing logic (which is, presumably, what justified putting it
into btf.c), but also a bunch of register state checks, some of each
utilize deep verifier logic helpers, necessarily exported from
verifier.c: check_ptr_off_reg(), check_func_arg_reg_off(),
and check_mem_reg().
Going forward, btf_check_subprog_call() will have a minimum of
BTF-related logic, but will get more internal verifier logic related to
register state manipulation. So move it into verifier.c to minimize
amount of verifier-specific logic exposed to btf.c.
We do this move before refactoring btf_check_func_arg_match() to
preserve as much history post-refactoring as possible.
No functional changes.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20231215011334.2307144-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Generalize btf_prepare_func_args() to support both global and static
subprogs. We are going to utilize this property in the next patch,
reusing btf_prepare_func_args() for subprog call logic instead of
reparsing BTF information in a completely separate implementation.
btf_prepare_func_args() now detects whether subprog is global or static
makes slight logic adjustments for static func cases, like not failing
fatally (-EFAULT) for conditions that are allowable for static subprogs.
Somewhat subtle (but major!) difference is the handling of pointer arguments.
Both global and static functions need to handle special context
arguments (which are pointers to predefined type names), but static
subprogs give up on any other pointers, falling back to marking subprog
as "unreliable", disabling the use of BTF type information altogether.
For global functions, though, we are assuming that such pointers to
unrecognized types are just pointers to fixed-sized memory region (or
error out if size cannot be established, like for `void *` pointers).
This patch accommodates these small differences and sets up a stage for
refactoring in the next patch, eliminating a separate BTF-based parsing
logic in btf_check_func_arg_match().
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20231215011334.2307144-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Instead of btf_check_subprog_arg_match(), use btf_prepare_func_args()
logic to validate "trustworthiness" of main BPF program's BTF information,
if it is present.
We ignored results of original BTF check anyway, often times producing
confusing and ominously-sounding "reg type unsupported for arg#0
function" message, which has no apparent effect on program correctness
and verification process.
All the -EFAULT returning sanity checks are already performed in
check_btf_info_early(), so there is zero reason to have this duplication
of logic between btf_check_subprog_call() and btf_check_subprog_arg_match().
Dropping btf_check_subprog_arg_match() simplifies
btf_check_func_arg_match() further removing `bool processing_call` flag.
One subtle bit that was done by btf_check_subprog_arg_match() was
potentially marking main program's BTF as unreliable. We do this
explicitly now with a dedicated simple check, preserving the original
behavior, but now based on well factored btf_prepare_func_args() logic.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20231215011334.2307144-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
btf_prepare_func_args() is used to understand expectations and
restrictions on global subprog arguments. But current implementation is
hard to extend, as it intermixes BTF-based func prototype parsing and
interpretation logic with setting up register state at subprog entry.
Worse still, those registers are not completely set up inside
btf_prepare_func_args(), requiring some more logic later in
do_check_common(). Like calling mark_reg_unknown() and similar
initialization operations.
This intermixing of BTF interpretation and register state setup is
problematic. First, it causes duplication of BTF parsing logic for global
subprog verification (to set up initial state of global subprog) and
global subprog call sites analysis (when we need to check that whatever
is being passed into global subprog matches expectations), performed in
btf_check_subprog_call().
Given we want to extend global func argument with tags later, this
duplication is problematic. So refactor btf_prepare_func_args() to do
only BTF-based func proto and args parsing, returning high-level
argument "expectations" only, with no regard to specifics of register
state. I.e., if it's a context argument, instead of setting register
state to PTR_TO_CTX, we return ARG_PTR_TO_CTX enum for that argument as
"an argument specification" for further processing inside
do_check_common(). Similarly for SCALAR arguments, PTR_TO_MEM, etc.
This allows to reuse btf_prepare_func_args() in following patches at
global subprog call site analysis time. It also keeps register setup
code consistently in one place, do_check_common().
Besides all this, we cache this argument specs information inside
env->subprog_info, eliminating the need to redo these potentially
expensive BTF traversals, especially if BPF program's BTF is big and/or
there are lots of global subprog calls.
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20231215011334.2307144-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
We can derive some new information for BPF_JNE in regs_refine_cond_op().
Take following code for example:
/* The type of "a" is u32 */
if (a > 0 && a < 100) {
/* the range of the register for a is [0, 99], not [1, 99],
* and will cause the following error:
*
* invalid zero-sized read
*
* as a can be 0.
*/
bpf_skb_store_bytes(skb, xx, xx, a, 0);
}
In the code above, "a > 0" will be compiled to "jmp xxx if a == 0". In the
TRUE branch, the dst_reg will be marked as known to 0. However, in the
fallthrough(FALSE) branch, the dst_reg will not be handled, which makes
the [min, max] for a is [0, 99], not [1, 99].
For BPF_JNE, we can reduce the range of the dst reg if the src reg is a
const and is exactly the edge of the dst reg.
Signed-off-by: Menglong Dong <menglong8.dong@gmail.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Link: https://lore.kernel.org/r/20231219134800.1550388-2-menglong8.dong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Pull tracing fix from Steven Rostedt:
"While working on the ring buffer, I found one more bug with the
timestamp code, and the fix for this removed the need for the final
64-bit cmpxchg!
The ring buffer events hold a "delta" from the previous event. If it
is determined that the delta can not be calculated, it falls back to
adding an absolute timestamp value. The way to know if the delta can
be used is via two stored timestamps in the per-cpu buffer meta data:
before_stamp and write_stamp
The before_stamp is written by every event before it tries to allocate
its space on the ring buffer. The write_stamp is written after it
allocates its space and knows that nothing came in after it read the
previous before_stamp and write_stamp and the two matched.
A previous fix dd93942570 ("ring-buffer: Do not try to put back
write_stamp") removed putting back the write_stamp to match the
before_stamp so that the next event could use the delta, but races
were found where the two would match, but not be for of the previous
event.
It was determined to allow the event reservation to not have a valid
write_stamp when it is finished, and this fixed a lot of races.
The last use of the 64-bit timestamp cmpxchg depended on the
write_stamp being valid after an interruption. But this is no longer
the case, as if an event is interrupted by a softirq that writes an
event, and that event gets interrupted by a hardirq or NMI and that
writes an event, then the softirq could finish its reservation without
a valid write_stamp.
In the slow path of the event reservation, a delta can still be used
if the write_stamp is valid. Instead of using a cmpxchg against the
write stamp, the before_stamp needs to be read again to validate the
write_stamp. The cmpxchg is not needed.
This updates the slowpath to validate the write_stamp by comparing it
to the before_stamp and removes all rb_time_cmpxchg() as there are no
more users of that function.
The removal of the 32-bit updates of rb_time_t will be done in the
next merge window"
* tag 'trace-v6.7-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
ring-buffer: Fix slowpath of interrupted event
To synchronize the timestamps with the ring buffer reservation, there are
two timestamps that are saved in the buffer meta data.
1. before_stamp
2. write_stamp
When the two are equal, the write_stamp is considered valid, as in, it may
be used to calculate the delta of the next event as the write_stamp is the
timestamp of the previous reserved event on the buffer.
This is done by the following:
/*A*/ w = current position on the ring buffer
before = before_stamp
after = write_stamp
ts = read current timestamp
if (before != after) {
write_stamp is not valid, force adding an absolute
timestamp.
}
/*B*/ before_stamp = ts
/*C*/ write = local_add_return(event length, position on ring buffer)
if (w == write - event length) {
/* Nothing interrupted between A and C */
/*E*/ write_stamp = ts;
delta = ts - after
/*
* If nothing interrupted again,
* before_stamp == write_stamp and write_stamp
* can be used to calculate the delta for
* events that come in after this one.
*/
} else {
/*
* The slow path!
* Was interrupted between A and C.
*/
This is the place that there's a bug. We currently have:
after = write_stamp
ts = read current timestamp
/*F*/ if (write == current position on the ring buffer &&
after < ts && cmpxchg(write_stamp, after, ts)) {
delta = ts - after;
} else {
delta = 0;
}
The assumption is that if the current position on the ring buffer hasn't
moved between C and F, then it also was not interrupted, and that the last
event written has a timestamp that matches the write_stamp. That is the
write_stamp is valid.
But this may not be the case:
If a task context event was interrupted by softirq between B and C.
And the softirq wrote an event that got interrupted by a hard irq between
C and E.
and the hard irq wrote an event (does not need to be interrupted)
We have:
/*B*/ before_stamp = ts of normal context
---> interrupted by softirq
/*B*/ before_stamp = ts of softirq context
---> interrupted by hardirq
/*B*/ before_stamp = ts of hard irq context
/*E*/ write_stamp = ts of hard irq context
/* matches and write_stamp valid */
<----
/*E*/ write_stamp = ts of softirq context
/* No longer matches before_stamp, write_stamp is not valid! */
<---
w != write - length, go to slow path
// Right now the order of events in the ring buffer is:
//
// |-- softirq event --|-- hard irq event --|-- normal context event --|
//
after = write_stamp (this is the ts of softirq)
ts = read current timestamp
if (write == current position on the ring buffer [true] &&
after < ts [true] && cmpxchg(write_stamp, after, ts) [true]) {
delta = ts - after [Wrong!]
The delta is to be between the hard irq event and the normal context
event, but the above logic made the delta between the softirq event and
the normal context event, where the hard irq event is between the two. This
will shift all the remaining event timestamps on the sub-buffer
incorrectly.
The write_stamp is only valid if it matches the before_stamp. The cmpxchg
does nothing to help this.
Instead, the following logic can be done to fix this:
before = before_stamp
ts = read current timestamp
before_stamp = ts
after = write_stamp
if (write == current position on the ring buffer &&
after == before && after < ts) {
delta = ts - after
} else {
delta = 0;
}
The above will only use the write_stamp if it still matches before_stamp
and was tested to not have changed since C.
As a bonus, with this logic we do not need any 64-bit cmpxchg() at all!
This means the 32-bit rb_time_t workaround can finally be removed. But
that's for a later time.
Link: https://lore.kernel.org/linux-trace-kernel/20231218175229.58ec3daf@gandalf.local.home/
Link: https://lore.kernel.org/linux-trace-kernel/20231218230712.3a76b081@gandalf.local.home
Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Fixes: dd93942570 ("ring-buffer: Do not try to put back write_stamp")
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Alexei Starovoitov says:
====================
pull-request: bpf-next 2023-12-18
This PR is larger than usual and contains changes in various parts
of the kernel.
The main changes are:
1) Fix kCFI bugs in BPF, from Peter Zijlstra.
End result: all forms of indirect calls from BPF into kernel
and from kernel into BPF work with CFI enabled. This allows BPF
to work with CONFIG_FINEIBT=y.
2) Introduce BPF token object, from Andrii Nakryiko.
It adds an ability to delegate a subset of BPF features from privileged
daemon (e.g., systemd) through special mount options for userns-bound
BPF FS to a trusted unprivileged application. The design accommodates
suggestions from Christian Brauner and Paul Moore.
Example:
$ sudo mkdir -p /sys/fs/bpf/token
$ sudo mount -t bpf bpffs /sys/fs/bpf/token \
-o delegate_cmds=prog_load:MAP_CREATE \
-o delegate_progs=kprobe \
-o delegate_attachs=xdp
3) Various verifier improvements and fixes, from Andrii Nakryiko, Andrei Matei.
- Complete precision tracking support for register spills
- Fix verification of possibly-zero-sized stack accesses
- Fix access to uninit stack slots
- Track aligned STACK_ZERO cases as imprecise spilled registers.
It improves the verifier "instructions processed" metric from single
digit to 50-60% for some programs.
- Fix verifier retval logic
4) Support for VLAN tag in XDP hints, from Larysa Zaremba.
5) Allocate BPF trampoline via bpf_prog_pack mechanism, from Song Liu.
End result: better memory utilization and lower I$ miss for calls to BPF
via BPF trampoline.
6) Fix race between BPF prog accessing inner map and parallel delete,
from Hou Tao.
7) Add bpf_xdp_get_xfrm_state() kfunc, from Daniel Xu.
It allows BPF interact with IPSEC infra. The intent is to support
software RSS (via XDP) for the upcoming ipsec pcpu work.
Experiments on AWS demonstrate single tunnel pcpu ipsec reaching
line rate on 100G ENA nics.
8) Expand bpf_cgrp_storage to support cgroup1 non-attach, from Yafang Shao.
9) BPF file verification via fsverity, from Song Liu.
It allows BPF progs get fsverity digest.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (164 commits)
bpf: Ensure precise is reset to false in __mark_reg_const_zero()
selftests/bpf: Add more uprobe multi fail tests
bpf: Fail uprobe multi link with negative offset
selftests/bpf: Test the release of map btf
s390/bpf: Fix indirect trampoline generation
selftests/bpf: Temporarily disable dummy_struct_ops test on s390
x86/cfi,bpf: Fix bpf_exception_cb() signature
bpf: Fix dtor CFI
cfi: Add CFI_NOSEAL()
x86/cfi,bpf: Fix bpf_struct_ops CFI
x86/cfi,bpf: Fix bpf_callback_t CFI
x86/cfi,bpf: Fix BPF JIT call
cfi: Flip headers
selftests/bpf: Add test for abnormal cnt during multi-kprobe attachment
selftests/bpf: Don't use libbpf_get_error() in kprobe_multi_test
selftests/bpf: Add test for abnormal cnt during multi-uprobe attachment
bpf: Limit the number of kprobes when attaching program to multiple kprobes
bpf: Limit the number of uprobes when attaching program to multiple uprobes
bpf: xdp: Register generic_kfunc_set with XDP programs
selftests/bpf: utilize string values for delegate_xxx mount options
...
====================
Link: https://lore.kernel.org/r/20231219000520.34178-1-alexei.starovoitov@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
It is safe to always start with imprecise SCALAR_VALUE register.
Previously __mark_reg_const_zero() relied on caller to reset precise
mark, but it's very error prone and we already missed it in a few
places. So instead make __mark_reg_const_zero() reset precision always,
as it's a safe default for SCALAR_VALUE. Explanation is basically the
same as for why we are resetting (or rather not setting) precision in
current state. If necessary, precision propagation will set it to
precise correctly.
As such, also remove a big comment about forward precision propagation
in mark_reg_stack_read() and avoid unnecessarily setting precision to
true after reading from STACK_ZERO stack. Again, precision propagation
will correctly handle this, if that SCALAR_VALUE register will ever be
needed to be precise.
Reported-by: Maxim Mikityanskiy <maxtram95@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Acked-by: Maxim Mikityanskiy <maxtram95@gmail.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20231218173601.53047-1-andrii@kernel.org
Currently the __uprobe_register will return 0 (success) when called with
negative offset. The reason is that the call to register_for_each_vma and
then build_map_info won't return error for negative offset. They just won't
do anything - no matching vma is found so there's no registered breakpoint
for the uprobe.
I don't think we can change the behaviour of __uprobe_register and fail
for negative uprobe offset, because apps might depend on that already.
But I think we can still make the change and check for it on bpf multi
link syscall level.
Also moving the __get_user call and check for the offsets to the top of
loop, to fail early without extra __get_user calls for ref_ctr_offset
and cookie arrays.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Song Liu <song@kernel.org>
Link: https://lore.kernel.org/bpf/20231217215538.3361991-2-jolsa@kernel.org
Pull perf fix from Borislav Petkov:
- Avoid iterating over newly created group leader event's siblings
because there are none, and thus prevent a lockdep splat
* tag 'perf_urgent_for_v6.7_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf: Fix perf_event_validate_size() lockdep splat
Pull CXL (Compute Express Link) fixes from Dan Williams:
"A collection of CXL fixes.
The touch outside of drivers/cxl/ is for a helper that allocates
physical address space. Device hotplug tests showed that the driver
failed to utilize (skipped over) valid capacity when allocating a new
memory region. Outside of that, new tests uncovered a small crop of
lockdep reports.
There is also some miscellaneous error path and leak fixups that are
not urgent, but useful to cleanup now.
- Fix alloc_free_mem_region()'s scan for address space, prevent false
negative out-of-space events
- Fix sleeping lock acquisition from CXL trace event (atomic context)
- Fix put_device() like for the new CXL PMU driver
- Fix wrong pointer freed on error path
- Fixup several lockdep reports (missing lock hold) from new
assertion in cxl_num_decoders_committed() and new tests"
* tag 'cxl-fixes-6.7-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl:
cxl/pmu: Ensure put_device on pmu devices
cxl/cdat: Free correct buffer on checksum error
cxl/hdm: Fix dpa translation locking
kernel/resource: Increment by align value in get_free_mem_region()
cxl: Add cxl_num_decoders_committed() usage to cxl_test
cxl/memdev: Hold region_rwsem during inject and clear poison ops
cxl/core: Always hold region_rwsem while reading poison lists
cxl/hdm: Fix a benign lockdep splat
Pull tracing fixes from Steven Rostedt:
- Fix eventfs to check creating new files for events with names greater
than NAME_MAX. The eventfs lookup needs to check the return result of
simple_lookup().
- Fix the ring buffer to check the proper max data size. Events must be
able to fit on the ring buffer sub-buffer, if it cannot, then it
fails to be written and the logic to add the event is avoided. The
code to check if an event can fit failed to add the possible absolute
timestamp which may make the event not be able to fit. This causes
the ring buffer to go into an infinite loop trying to find a
sub-buffer that would fit the event. Luckily, there's a check that
will bail out if it looped over a 1000 times and it also warns.
The real fix is not to add the absolute timestamp to an event that is
starting at the beginning of a sub-buffer because it uses the
sub-buffer timestamp.
By avoiding the timestamp at the start of the sub-buffer allows
events that pass the first check to always find a sub-buffer that it
can fit on.
- Have large events that do not fit on a trace_seq to print "LINE TOO
BIG" like it does for the trace_pipe instead of what it does now
which is to silently drop the output.
- Fix a memory leak of forgetting to free the spare page that is saved
by a trace instance.
- Update the size of the snapshot buffer when the main buffer is
updated if the snapshot buffer is allocated.
- Fix ring buffer timestamp logic by removing all the places that tried
to put the before_stamp back to the write stamp so that the next
event doesn't add an absolute timestamp. But each of these updates
added a race where by making the two timestamp equal, it was
validating the write_stamp so that it can be incorrectly used for
calculating the delta of an event.
- There's a temp buffer used for printing the event that was using the
event data size for allocation when it needed to use the size of the
entire event (meta-data and payload data)
- For hardening, use "%.*s" for printing the trace_marker output, to
limit the amount that is printed by the size of the event. This was
discovered by development that added a bug that truncated the '\0'
and caused a crash.
- Fix a use-after-free bug in the use of the histogram files when an
instance is being removed.
- Remove a useless update in the rb_try_to_discard of the write_stamp.
The before_stamp was already changed to force the next event to add
an absolute timestamp that the write_stamp is not used. But the
write_stamp is modified again using an unneeded 64-bit cmpxchg.
- Fix several races in the 32-bit implementation of the
rb_time_cmpxchg() that does a 64-bit cmpxchg.
- While looking at fixing the 64-bit cmpxchg, I noticed that because
the ring buffer uses normal cmpxchg, and this can be done in NMI
context, there's some architectures that do not have a working
cmpxchg in NMI context. For these architectures, fail recording
events that happen in NMI context.
* tag 'trace-v6.7-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
ring-buffer: Do not record in NMI if the arch does not support cmpxchg in NMI
ring-buffer: Have rb_time_cmpxchg() set the msb counter too
ring-buffer: Fix 32-bit rb_time_read() race with rb_time_cmpxchg()
ring-buffer: Fix a race in rb_time_cmpxchg() for 32 bit archs
ring-buffer: Remove useless update to write_stamp in rb_try_to_discard()
ring-buffer: Do not try to put back write_stamp
tracing: Fix uaf issue when open the hist or hist_debug file
tracing: Add size check when printing trace_marker output
ring-buffer: Have saved event hold the entire event
ring-buffer: Do not update before stamp when switching sub-buffers
tracing: Update snapshot buffer on resize if it is allocated
ring-buffer: Fix memory leak of free page
eventfs: Fix events beyond NAME_MAX blocking tasks
tracing: Have large events show up as '[LINE TOO BIG]' instead of nothing
ring-buffer: Fix writing to the buffer with max_data_size
Ensure the various dtor functions match their prototype and retain
their CFI signatures, since they don't have their address taken, they
are prone to not getting CFI, making them impossible to call
indirectly.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20231215092707.799451071@infradead.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
BPF struct_ops uses __arch_prepare_bpf_trampoline() to write
trampolines for indirect function calls. These tramplines much have
matching CFI.
In order to obtain the correct CFI hash for the various methods, add a
matching structure that contains stub functions, the compiler will
generate correct CFI which we can pilfer for the trampolines.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20231215092707.566977112@infradead.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The current BPF call convention is __nocfi, except when it calls !JIT things,
then it calls regular C functions.
It so happens that with FineIBT the __nocfi and C calling conventions are
incompatible. Specifically __nocfi will call at func+0, while FineIBT will have
endbr-poison there, which is not a valid indirect target. Causing #CP.
Notably this only triggers on IBT enabled hardware, which is probably why this
hasn't been reported (also, most people will have JIT on anyway).
Implement proper CFI prologues for the BPF JIT codegen and drop __nocfi for
x86.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20231215092707.345270396@infradead.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This code is rarely (never?) enabled by distros, and it hasn't caught
anything in decades. Let's kill off this legacy debug code.
Suggested-by: Linus Torvalds <torvalds@linuxfoundation.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There are multiple ways to grab references to credentials, and the only
protection we have against overflowing it is the memory required to do
so.
With memory sizes only moving in one direction, let's bump the reference
count to 64-bit and move it outside the realm of feasibly overflowing.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
An abnormally big cnt may also be assigned to kprobe_multi.cnt when
attaching multiple kprobes. It will trigger the following warning in
kvmalloc_node():
if (unlikely(size > INT_MAX)) {
WARN_ON_ONCE(!(flags & __GFP_NOWARN));
return NULL;
}
Fix the warning by limiting the maximal number of kprobes in
bpf_kprobe_multi_link_attach(). If the number of kprobes is greater than
MAX_KPROBE_MULTI_CNT, the attachment will fail and return -E2BIG.
Fixes: 0dcac27254 ("bpf: Add multi kprobe link")
Signed-off-by: Hou Tao <houtao1@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20231215100708.2265609-3-houtao@huaweicloud.com
Pull misc fixes from Andrew Morton:
"17 hotfixes. 8 are cc:stable and the other 9 pertain to post-6.6
issues"
* tag 'mm-hotfixes-stable-2023-12-15-07-11' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
mm/mglru: reclaim offlined memcgs harder
mm/mglru: respect min_ttl_ms with memcgs
mm/mglru: try to stop at high watermarks
mm/mglru: fix underprotected page cache
mm/shmem: fix race in shmem_undo_range w/THP
Revert "selftests: error out if kernel header files are not yet built"
crash_core: fix the check for whether crashkernel is from high memory
x86, kexec: fix the wrong ifdeffery CONFIG_KEXEC
sh, kexec: fix the incorrect ifdeffery and dependency of CONFIG_KEXEC
mips, kexec: fix the incorrect ifdeffery and dependency of CONFIG_KEXEC
m68k, kexec: fix the incorrect ifdeffery and build dependency of CONFIG_KEXEC
loongarch, kexec: change dependency of object files
mm/damon/core: make damon_start() waits until kdamond_fn() starts
selftests/mm: cow: print ksft header before printing anything else
mm: fix VMA heap bounds checking
riscv: fix VMALLOC_START definition
kexec: drop dependency on ARCH_SUPPORTS_KEXEC from CRASH_DUMP
The following race can cause rb_time_read() to observe a corrupted time
stamp:
rb_time_cmpxchg()
[...]
if (!rb_time_read_cmpxchg(&t->msb, msb, msb2))
return false;
if (!rb_time_read_cmpxchg(&t->top, top, top2))
return false;
<interrupted before updating bottom>
__rb_time_read()
[...]
do {
c = local_read(&t->cnt);
top = local_read(&t->top);
bottom = local_read(&t->bottom);
msb = local_read(&t->msb);
} while (c != local_read(&t->cnt));
*cnt = rb_time_cnt(top);
/* If top and msb counts don't match, this interrupted a write */
if (*cnt != rb_time_cnt(msb))
return false;
^ this check fails to catch that "bottom" is still not updated.
So the old "bottom" value is returned, which is wrong.
Fix this by checking that all three of msb, top, and bottom 2-bit cnt
values match.
The reason to favor checking all three fields over requiring a specific
update order for both rb_time_set() and rb_time_cmpxchg() is because
checking all three fields is more robust to handle partial failures of
rb_time_cmpxchg() when interrupted by nested rb_time_set().
Link: https://lore.kernel.org/lkml/20231211201324.652870-1-mathieu.desnoyers@efficios.com/
Link: https://lore.kernel.org/linux-trace-kernel/20231212193049.680122-1-mathieu.desnoyers@efficios.com
Fixes: f458a14534 ("ring-buffer: Test last update in 32bit version of __rb_time_read()")
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Mathieu Desnoyers pointed out an issue in the rb_time_cmpxchg() for 32 bit
architectures. That is:
static bool rb_time_cmpxchg(rb_time_t *t, u64 expect, u64 set)
{
unsigned long cnt, top, bottom, msb;
unsigned long cnt2, top2, bottom2, msb2;
u64 val;
/* The cmpxchg always fails if it interrupted an update */
if (!__rb_time_read(t, &val, &cnt2))
return false;
if (val != expect)
return false;
<<<< interrupted here!
cnt = local_read(&t->cnt);
The problem is that the synchronization counter in the rb_time_t is read
*after* the value of the timestamp is read. That means if an interrupt
were to come in between the value being read and the counter being read,
it can change the value and the counter and the interrupted process would
be clueless about it!
The counter needs to be read first and then the value. That way it is easy
to tell if the value is stale or not. If the counter hasn't been updated,
then the value is still good.
Link: https://lore.kernel.org/linux-trace-kernel/20231211201324.652870-1-mathieu.desnoyers@efficios.com/
Link: https://lore.kernel.org/linux-trace-kernel/20231212115301.7a9c9a64@gandalf.local.home
Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Fixes: 10464b4aa6 ("ring-buffer: Add rb_time_t 64 bit operations for speeding up 32 bit")
Reported-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
When filtering is enabled, a temporary buffer is created to place the
content of the trace event output so that the filter logic can decide
from the trace event output if the trace event should be filtered out or
not. If it is to be filtered out, the content in the temporary buffer is
simply discarded, otherwise it is written into the trace buffer.
But if an interrupt were to come in while a previous event was using that
temporary buffer, the event written by the interrupt would actually go
into the ring buffer itself to prevent corrupting the data on the
temporary buffer. If the event is to be filtered out, the event in the
ring buffer is discarded, or if it fails to discard because another event
were to have already come in, it is turned into padding.
The update to the write_stamp in the rb_try_to_discard() happens after a
fix was made to force the next event after the discard to use an absolute
timestamp by setting the before_stamp to zero so it does not match the
write_stamp (which causes an event to use the absolute timestamp).
But there's an effort in rb_try_to_discard() to put back the write_stamp
to what it was before the event was added. But this is useless and
wasteful because nothing is going to be using that write_stamp for
calculations as it still will not match the before_stamp.
Remove this useless update, and in doing so, we remove another
cmpxchg64()!
Also update the comments to reflect this change as well as remove some
extra white space in another comment.
Link: https://lore.kernel.org/linux-trace-kernel/20231215081810.1f4f38fe@rorschach.local.home
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Joel Fernandes <joel@joelfernandes.org>
Cc: Vincent Donnefort <vdonnefort@google.com>
Fixes: b2dd797543 ("ring-buffer: Force absolute timestamp on discard of event")
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
If an update to an event is interrupted by another event between the time
the initial event allocated its buffer and where it wrote to the
write_stamp, the code try to reset the write stamp back to the what it had
just overwritten. It knows that it was overwritten via checking the
before_stamp, and if it didn't match what it wrote to the before_stamp
before it allocated its space, it knows it was overwritten.
To put back the write_stamp, it uses the before_stamp it read. The problem
here is that by writing the before_stamp to the write_stamp it makes the
two equal again, which means that the write_stamp can be considered valid
as the last timestamp written to the ring buffer. But this is not
necessarily true. The event that interrupted the event could have been
interrupted in a way that it was interrupted as well, and can end up
leaving with an invalid write_stamp. But if this happens and returns to
this context that uses the before_stamp to update the write_stamp again,
it can possibly incorrectly make it valid, causing later events to have in
correct time stamps.
As it is OK to leave this function with an invalid write_stamp (one that
doesn't match the before_stamp), there's no reason to try to make it valid
again in this case. If this race happens, then just leave with the invalid
write_stamp and the next event to come along will just add a absolute
timestamp and validate everything again.
Bonus points: This gets rid of another cmpxchg64!
Link: https://lore.kernel.org/linux-trace-kernel/20231214222921.193037a7@gandalf.local.home
Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Joel Fernandes <joel@joelfernandes.org>
Cc: Vincent Donnefort <vdonnefort@google.com>
Fixes: a389d86f7f ("ring-buffer: Have nested events still record running time stamp")
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
When lockdep is enabled, the for_each_sibling_event(sibling, event)
macro checks that event->ctx->mutex is held. When creating a new group
leader event, we call perf_event_validate_size() on a partially
initialized event where event->ctx is NULL, and so when
for_each_sibling_event() attempts to check event->ctx->mutex, we get a
splat, as reported by Lucas De Marchi:
WARNING: CPU: 8 PID: 1471 at kernel/events/core.c:1950 __do_sys_perf_event_open+0xf37/0x1080
This only happens for a new event which is its own group_leader, and in
this case there cannot be any sibling events. Thus it's safe to skip the
check for siblings, which avoids having to make invasive and ugly
changes to for_each_sibling_event().
Avoid the splat by bailing out early when the new event is its own
group_leader.
Fixes: 382c27f4ed ("perf: Fix perf_event_validate_size()")
Closes: https://lore.kernel.org/lkml/20231214000620.3081018-1-lucas.demarchi@intel.com/
Closes: https://lore.kernel.org/lkml/ZXpm6gQ%2Fd59jGsuW@xpf.sh.intel.com/
Reported-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reported-by: Pengfei Xu <pengfei.xu@intel.com>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20231215112450.3972309-1-mark.rutland@arm.com
Registering generic_kfunc_set with XDP programs enables some of the
newer BPF features inside XDP -- namely tree based data structures and
BPF exceptions.
The current motivation for this commit is to enable assertions inside
XDP bpf progs. Assertions are a standard and useful tool to encode
intent.
Signed-off-by: Daniel Xu <dxu@dxuuu.xyz>
Link: https://lore.kernel.org/r/d07d4614b81ca6aada44fcb89bb6b618fb66e4ca.1702594357.git.dxu@dxuuu.xyz
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Besides already supported special "any" value and hex bit mask, support
string-based parsing of delegation masks based on exact enumerator
names. Utilize BTF information of `enum bpf_cmd`, `enum bpf_map_type`,
`enum bpf_prog_type`, and `enum bpf_attach_type` types to find supported
symbolic names (ignoring __MAX_xxx guard values and stripping repetitive
prefixes like BPF_ for cmd and attach types, BPF_MAP_TYPE_ for maps, and
BPF_PROG_TYPE_ for prog types). The case doesn't matter, but it is
normalized to lower case in mount option output. So "PROG_LOAD",
"prog_load", and "MAP_create" are all valid values to specify for
delegate_cmds options, "array" is among supported for map types, etc.
Besides supporting string values, we also support multiple values
specified at the same time, using colon (':') separator.
There are corresponding changes on bpf_show_options side to use known
values to print them in human-readable format, falling back to hex mask
printing, if there are any unrecognized bits. This shouldn't be
necessary when enum BTF information is present, but in general we should
always be able to fall back to this even if kernel was built without BTF.
As mentioned, emitted symbolic names are normalized to be all lower case.
Example below shows various ways to specify delegate_cmds options
through mount command and how mount options are printed back:
12/14 14:39:07.604
vmuser@archvm:~/local/linux/tools/testing/selftests/bpf
$ mount | rg token
$ sudo mkdir -p /sys/fs/bpf/token
$ sudo mount -t bpf bpffs /sys/fs/bpf/token \
-o delegate_cmds=prog_load:MAP_CREATE \
-o delegate_progs=kprobe \
-o delegate_attachs=xdp
$ mount | grep token
bpffs on /sys/fs/bpf/token type bpf (rw,relatime,delegate_cmds=map_create:prog_load,delegate_progs=kprobe,delegate_attachs=xdp)
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20231214225016.1209867-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
When running `./test_progs -j` in my local vm with latest kernel,
I once hit a kasan error like below:
[ 1887.184724] BUG: KASAN: slab-use-after-free in bpf_rb_root_free+0x1f8/0x2b0
[ 1887.185599] Read of size 4 at addr ffff888106806910 by task kworker/u12:2/2830
[ 1887.186498]
[ 1887.186712] CPU: 3 PID: 2830 Comm: kworker/u12:2 Tainted: G OEL 6.7.0-rc3-00699-g90679706d486-dirty #494
[ 1887.188034] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014
[ 1887.189618] Workqueue: events_unbound bpf_map_free_deferred
[ 1887.190341] Call Trace:
[ 1887.190666] <TASK>
[ 1887.190949] dump_stack_lvl+0xac/0xe0
[ 1887.191423] ? nf_tcp_handle_invalid+0x1b0/0x1b0
[ 1887.192019] ? panic+0x3c0/0x3c0
[ 1887.192449] print_report+0x14f/0x720
[ 1887.192930] ? preempt_count_sub+0x1c/0xd0
[ 1887.193459] ? __virt_addr_valid+0xac/0x120
[ 1887.194004] ? bpf_rb_root_free+0x1f8/0x2b0
[ 1887.194572] kasan_report+0xc3/0x100
[ 1887.195085] ? bpf_rb_root_free+0x1f8/0x2b0
[ 1887.195668] bpf_rb_root_free+0x1f8/0x2b0
[ 1887.196183] ? __bpf_obj_drop_impl+0xb0/0xb0
[ 1887.196736] ? preempt_count_sub+0x1c/0xd0
[ 1887.197270] ? preempt_count_sub+0x1c/0xd0
[ 1887.197802] ? _raw_spin_unlock+0x1f/0x40
[ 1887.198319] bpf_obj_free_fields+0x1d4/0x260
[ 1887.198883] array_map_free+0x1a3/0x260
[ 1887.199380] bpf_map_free_deferred+0x7b/0xe0
[ 1887.199943] process_scheduled_works+0x3a2/0x6c0
[ 1887.200549] worker_thread+0x633/0x890
[ 1887.201047] ? __kthread_parkme+0xd7/0xf0
[ 1887.201574] ? kthread+0x102/0x1d0
[ 1887.202020] kthread+0x1ab/0x1d0
[ 1887.202447] ? pr_cont_work+0x270/0x270
[ 1887.202954] ? kthread_blkcg+0x50/0x50
[ 1887.203444] ret_from_fork+0x34/0x50
[ 1887.203914] ? kthread_blkcg+0x50/0x50
[ 1887.204397] ret_from_fork_asm+0x11/0x20
[ 1887.204913] </TASK>
[ 1887.204913] </TASK>
[ 1887.205209]
[ 1887.205416] Allocated by task 2197:
[ 1887.205881] kasan_set_track+0x3f/0x60
[ 1887.206366] __kasan_kmalloc+0x6e/0x80
[ 1887.206856] __kmalloc+0xac/0x1a0
[ 1887.207293] btf_parse_fields+0xa15/0x1480
[ 1887.207836] btf_parse_struct_metas+0x566/0x670
[ 1887.208387] btf_new_fd+0x294/0x4d0
[ 1887.208851] __sys_bpf+0x4ba/0x600
[ 1887.209292] __x64_sys_bpf+0x41/0x50
[ 1887.209762] do_syscall_64+0x4c/0xf0
[ 1887.210222] entry_SYSCALL_64_after_hwframe+0x63/0x6b
[ 1887.210868]
[ 1887.211074] Freed by task 36:
[ 1887.211460] kasan_set_track+0x3f/0x60
[ 1887.211951] kasan_save_free_info+0x28/0x40
[ 1887.212485] ____kasan_slab_free+0x101/0x180
[ 1887.213027] __kmem_cache_free+0xe4/0x210
[ 1887.213514] btf_free+0x5b/0x130
[ 1887.213918] rcu_core+0x638/0xcc0
[ 1887.214347] __do_softirq+0x114/0x37e
The error happens at bpf_rb_root_free+0x1f8/0x2b0:
00000000000034c0 <bpf_rb_root_free>:
; {
34c0: f3 0f 1e fa endbr64
34c4: e8 00 00 00 00 callq 0x34c9 <bpf_rb_root_free+0x9>
34c9: 55 pushq %rbp
34ca: 48 89 e5 movq %rsp, %rbp
...
; if (rec && rec->refcount_off >= 0 &&
36aa: 4d 85 ed testq %r13, %r13
36ad: 74 a9 je 0x3658 <bpf_rb_root_free+0x198>
36af: 49 8d 7d 10 leaq 0x10(%r13), %rdi
36b3: e8 00 00 00 00 callq 0x36b8 <bpf_rb_root_free+0x1f8>
<==== kasan function
36b8: 45 8b 7d 10 movl 0x10(%r13), %r15d
<==== use-after-free load
36bc: 45 85 ff testl %r15d, %r15d
36bf: 78 8c js 0x364d <bpf_rb_root_free+0x18d>
So the problem is at rec->refcount_off in the above.
I did some source code analysis and find the reason.
CPU A CPU B
bpf_map_put:
...
btf_put with rcu callback
...
bpf_map_free_deferred
with system_unbound_wq
... ... ...
... btf_free_rcu: ...
... ... bpf_map_free_deferred:
... ...
... ---------> btf_struct_metas_free()
... | race condition ...
... ---------> map->ops->map_free()
...
... btf->struct_meta_tab = NULL
In the above, map_free() corresponds to array_map_free() and eventually
calling bpf_rb_root_free() which calls:
...
__bpf_obj_drop_impl(obj, field->graph_root.value_rec, false);
...
Here, 'value_rec' is assigned in btf_check_and_fixup_fields() with following code:
meta = btf_find_struct_meta(btf, btf_id);
if (!meta)
return -EFAULT;
rec->fields[i].graph_root.value_rec = meta->record;
So basically, 'value_rec' is a pointer to the record in struct_metas_tab.
And it is possible that that particular record has been freed by
btf_struct_metas_free() and hence we have a kasan error here.
Actually it is very hard to reproduce the failure with current bpf/bpf-next
code, I only got the above error once. To increase reproducibility, I added
a delay in bpf_map_free_deferred() to delay map->ops->map_free(), which
significantly increased reproducibility.
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 5e43ddd1b83f..aae5b5213e93 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -695,6 +695,7 @@ static void bpf_map_free_deferred(struct work_struct *work)
struct bpf_map *map = container_of(work, struct bpf_map, work);
struct btf_record *rec = map->record;
+ mdelay(100);
security_bpf_map_free(map);
bpf_map_release_memcg(map);
/* implementation dependent freeing */
Hao also provided test cases ([1]) for easily reproducing the above issue.
There are two ways to fix the issue, the v1 of the patch ([2]) moving
btf_put() after map_free callback, and the v5 of the patch ([3]) using
a kptr style fix which tries to get a btf reference during
map_check_btf(). Each approach has its pro and cons. The first approach
delays freeing btf while the second approach needs to acquire reference
depending on context which makes logic not very elegant and may
complicate things with future new data structures. Alexei
suggested in [4] going back to v1 which is what this patch
tries to do.
Rerun './test_progs -j' with the above mdelay() hack for a couple
of times and didn't observe the error for the above rb_root test cases.
Running Hou's test ([1]) is also successful.
[1] https://lore.kernel.org/bpf/20231207141500.917136-1-houtao@huaweicloud.com/
[2] v1: https://lore.kernel.org/bpf/20231204173946.3066377-1-yonghong.song@linux.dev/
[3] v5: https://lore.kernel.org/bpf/20231208041621.2968241-1-yonghong.song@linux.dev/
[4] v4: https://lore.kernel.org/bpf/CAADnVQJ3FiXUhZJwX_81sjZvSYYKCFB3BT6P8D59RS2Gu+0Z7g@mail.gmail.com/
Cc: Hou Tao <houtao@huaweicloud.com>
Fixes: 958cf2e273 ("bpf: Introduce bpf_obj_new")
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20231214203815.1469107-1-yonghong.song@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
There is no rcu-read-lock requirement for ops->map_fd_get_ptr() or
ops->map_fd_put_ptr(), so doesn't use rcu-read-lock for these two
callbacks.
For bpf_fd_array_map_update_elem(), accessing array->ptrs doesn't need
rcu-read-lock because array->ptrs must still be allocated. For
bpf_fd_htab_map_update_elem(), htab_map_update_elem() only requires
rcu-read-lock to be held to avoid the WARN_ON_ONCE(), so only use
rcu_read_lock() during the invocation of htab_map_update_elem().
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Hou Tao <houtao1@huawei.com>
Link: https://lore.kernel.org/r/20231214043010.3458072-2-houtao@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
KASAN report following issue. The root cause is when opening 'hist'
file of an instance and accessing 'trace_event_file' in hist_show(),
but 'trace_event_file' has been freed due to the instance being removed.
'hist_debug' file has the same problem. To fix it, call
tracing_{open,release}_file_tr() in file_operations callback to have
the ref count and avoid 'trace_event_file' being freed.
BUG: KASAN: slab-use-after-free in hist_show+0x11e0/0x1278
Read of size 8 at addr ffff242541e336b8 by task head/190
CPU: 4 PID: 190 Comm: head Not tainted 6.7.0-rc5-g26aff849438c #133
Hardware name: linux,dummy-virt (DT)
Call trace:
dump_backtrace+0x98/0xf8
show_stack+0x1c/0x30
dump_stack_lvl+0x44/0x58
print_report+0xf0/0x5a0
kasan_report+0x80/0xc0
__asan_report_load8_noabort+0x1c/0x28
hist_show+0x11e0/0x1278
seq_read_iter+0x344/0xd78
seq_read+0x128/0x1c0
vfs_read+0x198/0x6c8
ksys_read+0xf4/0x1e0
__arm64_sys_read+0x70/0xa8
invoke_syscall+0x70/0x260
el0_svc_common.constprop.0+0xb0/0x280
do_el0_svc+0x44/0x60
el0_svc+0x34/0x68
el0t_64_sync_handler+0xb8/0xc0
el0t_64_sync+0x168/0x170
Allocated by task 188:
kasan_save_stack+0x28/0x50
kasan_set_track+0x28/0x38
kasan_save_alloc_info+0x20/0x30
__kasan_slab_alloc+0x6c/0x80
kmem_cache_alloc+0x15c/0x4a8
trace_create_new_event+0x84/0x348
__trace_add_new_event+0x18/0x88
event_trace_add_tracer+0xc4/0x1a0
trace_array_create_dir+0x6c/0x100
trace_array_create+0x2e8/0x568
instance_mkdir+0x48/0x80
tracefs_syscall_mkdir+0x90/0xe8
vfs_mkdir+0x3c4/0x610
do_mkdirat+0x144/0x200
__arm64_sys_mkdirat+0x8c/0xc0
invoke_syscall+0x70/0x260
el0_svc_common.constprop.0+0xb0/0x280
do_el0_svc+0x44/0x60
el0_svc+0x34/0x68
el0t_64_sync_handler+0xb8/0xc0
el0t_64_sync+0x168/0x170
Freed by task 191:
kasan_save_stack+0x28/0x50
kasan_set_track+0x28/0x38
kasan_save_free_info+0x34/0x58
__kasan_slab_free+0xe4/0x158
kmem_cache_free+0x19c/0x508
event_file_put+0xa0/0x120
remove_event_file_dir+0x180/0x320
event_trace_del_tracer+0xb0/0x180
__remove_instance+0x224/0x508
instance_rmdir+0x44/0x78
tracefs_syscall_rmdir+0xbc/0x140
vfs_rmdir+0x1cc/0x4c8
do_rmdir+0x220/0x2b8
__arm64_sys_unlinkat+0xc0/0x100
invoke_syscall+0x70/0x260
el0_svc_common.constprop.0+0xb0/0x280
do_el0_svc+0x44/0x60
el0_svc+0x34/0x68
el0t_64_sync_handler+0xb8/0xc0
el0t_64_sync+0x168/0x170
Link: https://lore.kernel.org/linux-trace-kernel/20231214012153.676155-1-zhengyejian1@huawei.com
Suggested-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Zheng Yejian <zhengyejian1@huawei.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Since commit 638e4b825d ("bpf: Allows per-cpu maps and map-in-map in
sleepable programs"), sleepable BPF program can also use map-in-map, but
maybe_wait_bpf_programs() doesn't handle it accordingly. The main reason
is that using synchronize_rcu_tasks_trace() to wait for the completions
of these sleepable BPF programs may incur a very long delay and
userspace may think it is hung, so the wait for sleepable BPF programs
is skipped. Update the comments in maybe_wait_bpf_programs() to reflect
the reason.
Signed-off-by: Hou Tao <houtao1@huawei.com>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/r/20231211083447.1921178-1-houtao@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
security_path_* based LSM hooks appear to be generally missing from
the sleepable_lsm_hooks list. Initially add a small subset of them to
the preexisting sleepable_lsm_hooks list so that sleepable BPF helpers
like bpf_d_path() can be used from sleepable BPF LSM based programs.
The security_path_* hooks added in this patch are similar to the
security_inode_* counterparts that already exist in the
sleepable_lsm_hooks list, and are called in roughly similar points and
contexts. Presumably, making them OK to be also annotated as
sleepable.
Building a kernel with DEBUG_ATOMIC_SLEEP options enabled and running
reasonable workloads stimulating activity that would be intercepted by
such security hooks didn't show any splats.
Notably, I haven't added all the security_path_* LSM hooks that are
available as I don't need them at this point in time.
Signed-off-by: Matt Bobrowski <mattbobrowski@google.com>
Acked-by: KP Singh <kpsingh@kernel.org>
Link: https://lore.kernel.org/r/ZXM3IHHXpNY9y82a@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
It's quite confusing in practice when it's possible to successfully
create a BPF token from BPF FS that didn't have any of delegate_xxx
mount options set up. While it's not wrong, it's actually more
meaningful to reject BPF_TOKEN_CREATE with specific error code (-ENOENT)
to let user-space know that no token delegation is setup up.
So, instead of creating empty BPF token that will be always ignored
because it doesn't have any of the allow_xxx bits set, reject it with
-ENOENT. If we ever need empty BPF token to be possible, we can support
that with extra flag passed into BPF_TOKEN_CREATE.
Acked-by: Christian Brauner <brauner@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20231213190842.3844987-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
