linux-trace-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* WARNING: kmalloc bug in bpf_uprobe_multi_link_attach
@ 2023-12-11  8:12 xingwei lee
  2023-12-11 11:29 ` Hou Tao
  0 siblings, 1 reply; 8+ messages in thread
From: xingwei lee @ 2023-12-11  8:12 UTC (permalink / raw)
  To: ast
  Cc: daniel, andrii, martin.lau, song, yonghong.song, john.fastabend,
	kpsingh, sdf, haoluo, jolsa, rostedt, mhiramat, mathieu.desnoyers,
	bpf, linux-kernel, linux-trace-kernel

Sorry for containing HTML part, repeat the mail
Hello I found a bug in net/bpf in the lastest upstream linux and
lastest net tree.
WARNING: kmalloc bug in bpf_uprobe_multi_link_attach

kernel: net 28a7cb045ab700de5554193a1642917602787784
Kernel config: https://github.com/google/syzkaller/commits/fc59b78e3174009510ed15f20665e7ab2435ebee

in the lastest net tree, the crash like:

[   68.363836][ T8223] ------------[ cut here ]------------
[   68.364967][ T8223] WARNING: CPU: 2 PID: 8223 at mm/util.c:632
kvmalloc_node+0x18a/0x1a0
[   68.366527][ T8223] Modules linked in:
[   68.367882][ T8223] CPU: 2 PID: 8223 Comm: 36d Not tainted
6.7.0-rc4-00146-g28a7cb045ab7 #2
[   68.369260][ T8223] Hardware name: QEMU Standard PC (i440FX + PIIX,
1996), BIOS 1.16.2-1.fc38 04/014
[   68.370811][ T8223] RIP: 0010:kvmalloc_node+0x18a/0x1a0
[   68.371689][ T8223] Code: dc 1c 00 eb aa e8 86 33 c6 ff 41 81 e4 00
20 00 00 31 ff 44 89 e6 e8 e5 20
[   68.375001][ T8223] RSP: 0018:ffffc9001088fb68 EFLAGS: 00010293
[   68.375989][ T8223] RAX: 0000000000000000 RBX: 00000037ffffcec8
RCX: ffffffff81c1a32b
[   68.377154][ T8223] RDX: ffff88802cc00040 RSI: ffffffff81c1a339
RDI: 0000000000000005
[   68.377950][ T8223] RBP: 0000000000000400 R08: 0000000000000005
R09: 0000000000000000
[   68.378744][ T8223] R10: 0000000000000000 R11: 0000000000000000
R12: 0000000000000000
[   68.379523][ T8223] R13: 00000000ffffffff R14: ffff888017eb4a28
R15: 0000000000000000
[   68.380307][ T8223] FS:  0000000000827380(0000)
GS:ffff8880b9900000(0000) knlGS:0000000000000000
[   68.381185][ T8223] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   68.381843][ T8223] CR2: 0000000020000140 CR3: 00000000204d2000
CR4: 0000000000750ef0
[   68.382624][ T8223] PKRU: 55555554
[   68.382978][ T8223] Call Trace:
[   68.383312][ T8223]  <TASK>
[   68.383608][ T8223]  ? show_regs+0x8f/0xa0
[   68.384052][ T8223]  ? __warn+0xe6/0x390
[   68.384470][ T8223]  ? kvmalloc_node+0x18a/0x1a0
[   68.385111][ T8223]  ? report_bug+0x3b9/0x580
[   68.385585][ T8223]  ? handle_bug+0x67/0x90
[   68.386032][ T8223]  ? exc_invalid_op+0x17/0x40
[   68.386503][ T8223]  ? asm_exc_invalid_op+0x1a/0x20
[   68.387065][ T8223]  ? kvmalloc_node+0x17b/0x1a0
[   68.387551][ T8223]  ? kvmalloc_node+0x189/0x1a0
[   68.388051][ T8223]  ? kvmalloc_node+0x18a/0x1a0
[   68.388537][ T8223]  ? kvmalloc_node+0x189/0x1a0
[   68.389038][ T8223]  bpf_uprobe_multi_link_attach+0x436/0xfb0
[   68.389633][ T8223]  ? __might_fault+0x13f/0x1a0
[   68.390129][ T8223]  ? bpf_kprobe_multi_link_attach+0x10/0x10
[   68.390731][ T8223]  ? __fget_light+0x1fc/0x260
[   68.391206][ T8223]  ? __sanitizer_cov_trace_switch+0x54/0x90
[   68.391812][ T8223]  __sys_bpf+0x3ea0/0x4840
[   68.392267][ T8223]  ? slab_free_freelist_hook+0x114/0x1e0
[   68.393032][ T8223]  ? bpf_perf_link_attach+0x540/0x540
[   68.393580][ T8223]  ? putname+0x12e/0x170
[   68.394015][ T8223]  ? kmem_cache_free+0xf8/0x350
[   68.394509][ T8223]  ? putname+0x12e/0x170
[   68.394948][ T8223]  ? do_sys_openat2+0xb1/0x1e0
[   68.395442][ T8223]  ? __x64_sys_creat+0xcd/0x120
[   68.395945][ T8223]  __x64_sys_bpf+0x78/0xc0
[   68.396393][ T8223]  ? syscall_enter_from_user_mode+0x7f/0x120
[   68.397040][ T8223]  do_syscall_64+0x41/0x110
[   68.397502][ T8223]  entry_SYSCALL_64_after_hwframe+0x63/0x6b
[   68.398098][ T8223] RIP: 0033:0x410ead
[   68.398498][ T8223] Code: b3 66 2e 0f 1f 84 00 00 00 00 00 66 90 f3
0f 1e fa 48 89 f8 48 89 f7 48 88
[   68.400432][ T8223] RSP: 002b:00007ffdbabd7098 EFLAGS: 00000246
ORIG_RAX: 0000000000000141
[   68.401271][ T8223] RAX: ffffffffffffffda RBX: 00007ffdbabd7298
RCX: 0000000000410ead
[   68.402063][ T8223] RDX: 0000000000000040 RSI: 0000000020000340
RDI: 000000000000001c
[   68.402864][ T8223] RBP: 00007ffdbabd70b0 R08: 0000000000000000
R09: 0000000000000000
[   68.403649][ T8223] R10: 0000000000000000 R11: 0000000000000246
R12: 000000000049be68
[   68.404786][ T8223] R13: 0000000000000001 R14: 0000000000000001
R15: 0000000000000001
[   68.405574][ T8223]  </TASK>
[   68.405890][ T8223] Kernel panic - not syncing: kernel: panic_on_warn set ...
[   68.406611][ T8223] CPU: 2 PID: 8223 Comm: 36d Not tainted
6.7.0-rc4-00146-g28a7cb045ab7 #2
[   68.407453][ T8223] Hardware name: QEMU Standard PC (i440FX + PIIX,
1996), BIOS 1.16.2-1.fc38 04/014
[   68.408386][ T8223] Call Trace:
[   68.408723][ T8223]  <TASK>
[   68.409023][ T8223]  dump_stack_lvl+0xd3/0x1b0
[   68.409484][ T8223]  panic+0x6dc/0x790
[   68.409884][ T8223]  ? panic_smp_self_stop+0xa0/0xa0
[   68.410408][ T8223]  ? show_trace_log_lvl+0x363/0x4f0
[   68.410947][ T8223]  ? check_panic_on_warn+0x1f/0xb0
[   68.411458][ T8223]  ? kvmalloc_node+0x18a/0x1a0
[   68.411955][ T8223]  check_panic_on_warn+0xab/0xb0
[   68.412453][ T8223]  __warn+0xf2/0x390
[   68.412855][ T8223]  ? kvmalloc_node+0x18a/0x1a0
[   68.413334][ T8223]  report_bug+0x3b9/0x580
[   68.413778][ T8223]  handle_bug+0x67/0x90
[   68.414195][ T8223]  exc_invalid_op+0x17/0x40
[   68.414651][ T8223]  asm_exc_invalid_op+0x1a/0x20
[   68.415150][ T8223] RIP: 0010:kvmalloc_node+0x18a/0x1a0
[   68.415693][ T8223] Code: dc 1c 00 eb aa e8 86 33 c6 ff 41 81 e4 00
20 00 00 31 ff 44 89 e6 e8 e5 20
[   68.417651][ T8223] RSP: 0018:ffffc9001088fb68 EFLAGS: 00010293
[   68.418279][ T8223] RAX: 0000000000000000 RBX: 00000037ffffcec8
RCX: ffffffff81c1a32b
[   68.419090][ T8223] RDX: ffff88802cc00040 RSI: ffffffff81c1a339
RDI: 0000000000000005
[   68.419884][ T8223] RBP: 0000000000000400 R08: 0000000000000005
R09: 0000000000000000
[   68.420678][ T8223] R10: 0000000000000000 R11: 0000000000000000
R12: 0000000000000000
[   68.421474][ T8223] R13: 00000000ffffffff R14: ffff888017eb4a28
R15: 0000000000000000
[   68.422287][ T8223]  ? kvmalloc_node+0x17b/0x1a0
[   68.422780][ T8223]  ? kvmalloc_node+0x189/0x1a0
[   68.423270][ T8223]  ? kvmalloc_node+0x189/0x1a0
[   68.423769][ T8223]  bpf_uprobe_multi_link_attach+0x436/0xfb0
[   68.424372][ T8223]  ? __might_fault+0x13f/0x1a0
[   68.424860][ T8223]  ? bpf_kprobe_multi_link_attach+0x10/0x10
[   68.425462][ T8223]  ? __fget_light+0x1fc/0x260
[   68.425951][ T8223]  ? __sanitizer_cov_trace_switch+0x54/0x90
[   68.426545][ T8223]  __sys_bpf+0x3ea0/0x4840
[   68.427005][ T8223]  ? slab_free_freelist_hook+0x114/0x1e0
[   68.427583][ T8223]  ? bpf_perf_link_attach+0x540/0x540
[   68.428133][ T8223]  ? putname+0x12e/0x170
[   68.428564][ T8223]  ? kmem_cache_free+0xf8/0x350
[   68.429079][ T8223]  ? putname+0x12e/0x170
[   68.429519][ T8223]  ? do_sys_openat2+0xb1/0x1e0
[   68.430029][ T8223]  ? __x64_sys_creat+0xcd/0x120
[   68.430536][ T8223]  __x64_sys_bpf+0x78/0xc0
[   68.431000][ T8223]  ? syscall_enter_from_user_mode+0x7f/0x120
[   68.431623][ T8223]  do_syscall_64+0x41/0x110
[   68.432099][ T8223]  entry_SYSCALL_64_after_hwframe+0x63/0x6b
[   68.432708][ T8223] RIP: 0033:0x410ead
[   68.433117][ T8223] Code: b3 66 2e 0f 1f 84 00 00 00 00 00 66 90 f3
0f 1e fa 48 89 f8 48 89 f7 48 88
[   68.435240][ T8223] RSP: 002b:00007ffdbabd7098 EFLAGS: 00000246
ORIG_RAX: 0000000000000141
[   68.436087][ T8223] RAX: ffffffffffffffda RBX: 00007ffdbabd7298
RCX: 0000000000410ead
[   68.436898][ T8223] RDX: 0000000000000040 RSI: 0000000020000340
RDI: 000000000000001c
[   68.437697][ T8223] RBP: 00007ffdbabd70b0 R08: 0000000000000000
R09: 0000000000000000
[   68.438499][ T8223] R10: 0000000000000000 R11: 0000000000000246
R12: 000000000049be68
[   68.439305][ T8223] R13: 0000000000000001 R14: 0000000000000001
R15: 0000000000000001
[   68.440115][ T8223]  </TASK>
[   68.440773][ T8223] Kernel Offset: disabled
[   68.441251][ T8223] Rebooting in 86400 seconds..

=* repro.c =*
// autogenerated by syzkaller (https://github.com/google/syzkaller)

#define _GNU_SOURCE

#include <endian.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/syscall.h>
#include <sys/types.h>
#include <unistd.h>

#ifndef __NR_bpf
#define __NR_bpf 321
#endif

#define BITMASK(bf_off, bf_len) (((1ull << (bf_len)) - 1) << (bf_off))
#define STORE_BY_BITMASK(type, htobe, addr, val, bf_off, bf_len)     \
  *(type*)(addr) =                                                   \
      htobe((htobe(*(type*)(addr)) & ~BITMASK((bf_off), (bf_len))) | \
            (((type)(val) << (bf_off)) & BITMASK((bf_off), (bf_len))))

uint64_t r[1] = {0xffffffffffffffff};

int main(void) {
  syscall(__NR_mmap, /*addr=*/0x1ffff000ul, /*len=*/0x1000ul, /*prot=*/0ul,
          /*flags=*/0x32ul, /*fd=*/-1, /*offset=*/0ul);
  syscall(__NR_mmap, /*addr=*/0x20000000ul, /*len=*/0x1000000ul, /*prot=*/7ul,
          /*flags=*/0x32ul, /*fd=*/-1, /*offset=*/0ul);
  syscall(__NR_mmap, /*addr=*/0x21000000ul, /*len=*/0x1000ul, /*prot=*/0ul,
          /*flags=*/0x32ul, /*fd=*/-1, /*offset=*/0ul);
  intptr_t res = 0;
  *(uint32_t*)0x20000140 = 2;
  *(uint32_t*)0x20000144 = 3;
  *(uint64_t*)0x20000148 = 0x20000200;
  *(uint8_t*)0x20000200 = 0x18;
  STORE_BY_BITMASK(uint8_t, , 0x20000201, 0, 0, 4);
  STORE_BY_BITMASK(uint8_t, , 0x20000201, 0, 4, 4);
  *(uint16_t*)0x20000202 = 0;
  *(uint32_t*)0x20000204 = 0;
  *(uint8_t*)0x20000208 = 0;
  *(uint8_t*)0x20000209 = 0;
  *(uint16_t*)0x2000020a = 0;
  *(uint32_t*)0x2000020c = 0;
  *(uint8_t*)0x20000210 = 0x95;
  *(uint8_t*)0x20000211 = 0;
  *(uint16_t*)0x20000212 = 0;
  *(uint32_t*)0x20000214 = 0;
  *(uint64_t*)0x20000150 = 0x20000240;
  memcpy((void*)0x20000240, "GPL\000", 4);
  *(uint32_t*)0x20000158 = 0;
  *(uint32_t*)0x2000015c = 0;
  *(uint64_t*)0x20000160 = 0;
  *(uint32_t*)0x20000168 = 0;
  *(uint32_t*)0x2000016c = 0;
  memset((void*)0x20000170, 0, 16);
  *(uint32_t*)0x20000180 = 0;
  *(uint32_t*)0x20000184 = 0x30;
  *(uint32_t*)0x20000188 = 0;
  *(uint32_t*)0x2000018c = 0;
  *(uint64_t*)0x20000190 = 0;
  *(uint32_t*)0x20000198 = 0;
  *(uint32_t*)0x2000019c = 0;
  *(uint64_t*)0x200001a0 = 0;
  *(uint32_t*)0x200001a8 = 0;
  *(uint32_t*)0x200001ac = 0;
  *(uint32_t*)0x200001b0 = 0;
  *(uint32_t*)0x200001b4 = 0;
  *(uint64_t*)0x200001b8 = 0;
  *(uint64_t*)0x200001c0 = 0;
  *(uint32_t*)0x200001c8 = 0;
  *(uint32_t*)0x200001cc = 0;
  res = syscall(__NR_bpf, /*cmd=*/5ul, /*arg=*/0x20000140ul, /*size=*/0x90ul);
  if (res != -1) r[0] = res;
  memcpy((void*)0x20000000, "./file0\000", 8);
  syscall(__NR_creat, /*file=*/0x20000000ul, /*mode=*/0ul);
  *(uint32_t*)0x20000340 = r[0];
  *(uint32_t*)0x20000344 = 0;
  *(uint32_t*)0x20000348 = 0x30;
  *(uint32_t*)0x2000034c = 0;
  *(uint64_t*)0x20000350 = 0x20000080;
  memcpy((void*)0x20000080, "./file0\000", 8);
  *(uint64_t*)0x20000358 = 0x200000c0;
  *(uint64_t*)0x200000c0 = 0;
  *(uint64_t*)0x20000360 = 0;
  *(uint64_t*)0x20000368 = 0;
  *(uint32_t*)0x20000370 = 0xffffff1f;
  *(uint32_t*)0x20000374 = 0;
  *(uint32_t*)0x20000378 = 0;
  syscall(__NR_bpf, /*cmd=*/0x1cul, /*arg=*/0x20000340ul, /*size=*/0x40ul);
  return 0;
}

=* repro.txt =*
r0 = bpf$PROG_LOAD(0x5, &(0x7f0000000140)={0x2, 0x3,
&(0x7f0000000200)=@framed, &(0x7f0000000240)='GPL\x00', 0x0, 0x0, 0x0,
0x0, 0x0, '\x00', 0x0, 0x30, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0, 0x0}, 0x90)
creat(&(0x7f0000000000)='./file0\x00', 0x0)
bpf$BPF_LINK_CREATE_XDP(0x1c, &(0x7f0000000340)={r0, 0x0, 0x30, 0x0,
@val=@uprobe_multi={&(0x7f0000000080)='./file0\x00',
&(0x7f00000000c0)=[0x0], 0x0, 0x0, 0xffffff1f}}, 0x40)


See aslo https://gist.github.com/xrivendell7/15d43946c73aa13247b4b20b68798aaa

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: WARNING: kmalloc bug in bpf_uprobe_multi_link_attach
       [not found] <CABOYnLz2e+_0P88RgoDy6epWz9xrM2zhfMQdVrcjNiPqrFcBeQ@mail.gmail.com>
@ 2023-12-11 11:22 ` Masami Hiramatsu
  0 siblings, 0 replies; 8+ messages in thread
From: Masami Hiramatsu @ 2023-12-11 11:22 UTC (permalink / raw)
  To: xingwei lee, Jiri Olsa
  Cc: ast, daniel, andrii, martin.lau, song, yonghong.song,
	john.fastabend, kpsingh, sdf, haoluo, jolsa, rostedt, mhiramat,
	mathieu.desnoyers, bpf, linux-kernel, linux-trace-kernel

On Mon, 11 Dec 2023 16:10:32 +0800
xingwei lee <xrivendell7@gmail.com> wrote:

> Hello I found a bug in net/bpf in the lastest upstream linux and lastest
> net tree.
> WARNING: kmalloc bug in bpf_uprobe_multi_link_attach

Hmm, uprobe_multi is recently introduced and it seems a normal
uprobes unlike kprobe_multi (which uses fprobe instead of kprobe). 

The warning warns that the required size is bigger than INT_MAX. Maybe
too many links or uprobes were going to be allocated?

Thanks,

> 
> kernel: net 28a7cb045ab700de5554193a1642917602787784
> Kernel config:
> https://github.com/google/syzkaller/commits/fc59b78e3174009510ed15f20665e7ab2435ebee
> 
> in the lastestcrash like:
> 
> [   68.363836][ T8223] ------------[ cut here ]------------
> [   68.364967][ T8223] WARNING: CPU: 2 PID: 8223 at mm/util.c:632
> kvmalloc_node+0x18a/0x1a0
> [   68.366527][ T8223] Modules linked in:
> [   68.367882][ T8223] CPU: 2 PID: 8223 Comm: 36d Not tainted
> 6.7.0-rc4-00146-g28a7cb045ab7 #2
> [   68.369260][ T8223] Hardware name: QEMU Standard PC (i440FX + PIIX,
> 1996), BIOS 1.16.2-1.fc38 04/014
> [   68.370811][ T8223] RIP: 0010:kvmalloc_node+0x18a/0x1a0
> [   68.371689][ T8223] Code: dc 1c 00 eb aa e8 86 33 c6 ff 41 81 e4 00 20
> 00 00 31 ff 44 89 e6 e8 e5 20
> [   68.375001][ T8223] RSP: 0018:ffffc9001088fb68 EFLAGS: 00010293
> [   68.375989][ T8223] RAX: 0000000000000000 RBX: 00000037ffffcec8 RCX:
> ffffffff81c1a32b
> [   68.377154][ T8223] RDX: ffff88802cc00040 RSI: ffffffff81c1a339 RDI:
> 0000000000000005
> [   68.377950][ T8223] RBP: 0000000000000400 R08: 0000000000000005 R09:
> 0000000000000000
> [   68.378744][ T8223] R10: 0000000000000000 R11: 0000000000000000 R12:
> 0000000000000000
> [   68.379523][ T8223] R13: 00000000ffffffff R14: ffff888017eb4a28 R15:
> 0000000000000000
> [   68.380307][ T8223] FS:  0000000000827380(0000)
> GS:ffff8880b9900000(0000) knlGS:0000000000000000
> [   68.381185][ T8223] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   68.381843][ T8223] CR2: 0000000020000140 CR3: 00000000204d2000 CR4:
> 0000000000750ef0
> [   68.382624][ T8223] PKRU: 55555554
> [   68.382978][ T8223] Call Trace:
> [   68.383312][ T8223]  <TASK>
> [   68.383608][ T8223]  ? show_regs+0x8f/0xa0
> [   68.384052][ T8223]  ? __warn+0xe6/0x390
> [   68.384470][ T8223]  ? kvmalloc_node+0x18a/0x1a0
> [   68.385111][ T8223]  ? report_bug+0x3b9/0x580
> [   68.385585][ T8223]  ? handle_bug+0x67/0x90
> [   68.386032][ T8223]  ? exc_invalid_op+0x17/0x40
> [   68.386503][ T8223]  ? asm_exc_invalid_op+0x1a/0x20
> [   68.387065][ T8223]  ? kvmalloc_node+0x17b/0x1a0
> [   68.387551][ T8223]  ? kvmalloc_node+0x189/0x1a0
> [   68.388051][ T8223]  ? kvmalloc_node+0x18a/0x1a0
> [   68.388537][ T8223]  ? kvmalloc_node+0x189/0x1a0
> [   68.389038][ T8223]  bpf_uprobe_multi_link_attach+0x436/0xfb0
> [   68.389633][ T8223]  ? __might_fault+0x13f/0x1a0
> [   68.390129][ T8223]  ? bpf_kprobe_multi_link_attach+0x10/0x10
> [   68.390731][ T8223]  ? __fget_light+0x1fc/0x260
> [   68.391206][ T8223]  ? __sanitizer_cov_trace_switch+0x54/0x90
> [   68.391812][ T8223]  __sys_bpf+0x3ea0/0x4840
> [   68.392267][ T8223]  ? slab_free_freelist_hook+0x114/0x1e0
> [   68.393032][ T8223]  ? bpf_perf_link_attach+0x540/0x540
> [   68.393580][ T8223]  ? putname+0x12e/0x170
> [   68.394015][ T8223]  ? kmem_cache_free+0xf8/0x350
> [   68.394509][ T8223]  ? putname+0x12e/0x170
> [   68.394948][ T8223]  ? do_sys_openat2+0xb1/0x1e0
> [   68.395442][ T8223]  ? __x64_sys_creat+0xcd/0x120
> [   68.395945][ T8223]  __x64_sys_bpf+0x78/0xc0
> [   68.396393][ T8223]  ? syscall_enter_from_user_mode+0x7f/0x120
> [   68.397040][ T8223]  do_syscall_64+0x41/0x110
> [   68.397502][ T8223]  entry_SYSCALL_64_after_hwframe+0x63/0x6b
> [   68.398098][ T8223] RIP: 0033:0x410ead
> [   68.398498][ T8223] Code: b3 66 2e 0f 1f 84 00 00 00 00 00 66 90 f3 0f
> 1e fa 48 89 f8 48 89 f7 48 88
> [   68.400432][ T8223] RSP: 002b:00007ffdbabd7098 EFLAGS: 00000246
> ORIG_RAX: 0000000000000141
> [   68.401271][ T8223] RAX: ffffffffffffffda RBX: 00007ffdbabd7298 RCX:
> 0000000000410ead
> [   68.402063][ T8223] RDX: 0000000000000040 RSI: 0000000020000340 RDI:
> 000000000000001c
> [   68.402864][ T8223] RBP: 00007ffdbabd70b0 R08: 0000000000000000 R09:
> 0000000000000000
> [   68.403649][ T8223] R10: 0000000000000000 R11: 0000000000000246 R12:
> 000000000049be68
> [   68.404786][ T8223] R13: 0000000000000001 R14: 0000000000000001 R15:
> 0000000000000001
> [   68.405574][ T8223]  </TASK>
> [   68.405890][ T8223] Kernel panic - not syncing: kernel: panic_on_warn
> set ...
> [   68.406611][ T8223] CPU: 2 PID: 8223 Comm: 36d Not tainted
> 6.7.0-rc4-00146-g28a7cb045ab7 #2
> [   68.407453][ T8223] Hardware name: QEMU Standard PC (i440FX + PIIX,
> 1996), BIOS 1.16.2-1.fc38 04/014
> [   68.408386][ T8223] Call Trace:
> [   68.408723][ T8223]  <TASK>
> [   68.409023][ T8223]  dump_stack_lvl+0xd3/0x1b0
> [   68.409484][ T8223]  panic+0x6dc/0x790
> [   68.409884][ T8223]  ? panic_smp_self_stop+0xa0/0xa0
> [   68.410408][ T8223]  ? show_trace_log_lvl+0x363/0x4f0
> [   68.410947][ T8223]  ? check_panic_on_warn+0x1f/0xb0
> [   68.411458][ T8223]  ? kvmalloc_node+0x18a/0x1a0
> [   68.411955][ T8223]  check_panic_on_warn+0xab/0xb0
> [   68.412453][ T8223]  __warn+0xf2/0x390
> [   68.412855][ T8223]  ? kvmalloc_node+0x18a/0x1a0
> [   68.413334][ T8223]  report_bug+0x3b9/0x580
> [   68.413778][ T8223]  handle_bug+0x67/0x90
> [   68.414195][ T8223]  exc_invalid_op+0x17/0x40
> [   68.414651][ T8223]  asm_exc_invalid_op+0x1a/0x20
> [   68.415150][ T8223] RIP: 0010:kvmalloc_node+0x18a/0x1a0
> [   68.415693][ T8223] Code: dc 1c 00 eb aa e8 86 33 c6 ff 41 81 e4 00 20
> 00 00 31 ff 44 89 e6 e8 e5 20
> [   68.417651][ T8223] RSP: 0018:ffffc9001088fb68 EFLAGS: 00010293
> [   68.418279][ T8223] RAX: 0000000000000000 RBX: 00000037ffffcec8 RCX:
> ffffffff81c1a32b
> [   68.419090][ T8223] RDX: ffff88802cc00040 RSI: ffffffff81c1a339 RDI:
> 0000000000000005
> [   68.419884][ T8223] RBP: 0000000000000400 R08: 0000000000000005 R09:
> 0000000000000000
> [   68.420678][ T8223] R10: 0000000000000000 R11: 0000000000000000 R12:
> 0000000000000000
> [   68.421474][ T8223] R13: 00000000ffffffff R14: ffff888017eb4a28 R15:
> 0000000000000000
> [   68.422287][ T8223]  ? kvmalloc_node+0x17b/0x1a0
> [   68.422780][ T8223]  ? kvmalloc_node+0x189/0x1a0
> [   68.423270][ T8223]  ? kvmalloc_node+0x189/0x1a0
> [   68.423769][ T8223]  bpf_uprobe_multi_link_attach+0x436/0xfb0
> [   68.424372][ T8223]  ? __might_fault+0x13f/0x1a0
> [   68.424860][ T8223]  ? bpf_kprobe_multi_link_attach+0x10/0x10
> [   68.425462][ T8223]  ? __fget_light+0x1fc/0x260
> [   68.425951][ T8223]  ? __sanitizer_cov_trace_switch+0x54/0x90
> [   68.426545][ T8223]  __sys_bpf+0x3ea0/0x4840
> [   68.427005][ T8223]  ? slab_free_freelist_hook+0x114/0x1e0
> [   68.427583][ T8223]  ? bpf_perf_link_attach+0x540/0x540
> [   68.428133][ T8223]  ? putname+0x12e/0x170
> [   68.428564][ T8223]  ? kmem_cache_free+0xf8/0x350
> [   68.429079][ T8223]  ? putname+0x12e/0x170
> [   68.429519][ T8223]  ? do_sys_openat2+0xb1/0x1e0
> [   68.430029][ T8223]  ? __x64_sys_creat+0xcd/0x120
> [   68.430536][ T8223]  __x64_sys_bpf+0x78/0xc0
> [   68.431000][ T8223]  ? syscall_enter_from_user_mode+0x7f/0x120
> [   68.431623][ T8223]  do_syscall_64+0x41/0x110
> [   68.432099][ T8223]  entry_SYSCALL_64_after_hwframe+0x63/0x6b
> [   68.432708][ T8223] RIP: 0033:0x410ead
> [   68.433117][ T8223] Code: b3 66 2e 0f 1f 84 00 00 00 00 00 66 90 f3 0f
> 1e fa 48 89 f8 48 89 f7 48 88
> [   68.435240][ T8223] RSP: 002b:00007ffdbabd7098 EFLAGS: 00000246
> ORIG_RAX: 0000000000000141
> [   68.436087][ T8223] RAX: ffffffffffffffda RBX: 00007ffdbabd7298 RCX:
> 0000000000410ead
> [   68.436898][ T8223] RDX: 0000000000000040 RSI: 0000000020000340 RDI:
> 000000000000001c
> [   68.437697][ T8223] RBP: 00007ffdbabd70b0 R08: 0000000000000000 R09:
> 0000000000000000
> [   68.438499][ T8223] R10: 0000000000000000 R11: 0000000000000246 R12:
> 000000000049be68
> [   68.439305][ T8223] R13: 0000000000000001 R14: 0000000000000001 R15:
> 0000000000000001
> [   68.440115][ T8223]  </TASK>
> [   68.440773][ T8223] Kernel Offset: disabled
> [   68.441251][ T8223] Rebooting in 86400 seconds..
> 
> =* repro.c =*
> // autogenerated by syzkaller (https://github.com/google/syzkaller)
> 
> #define _GNU_SOURCE
> 
> #include <endian.h>
> #include <stdint.h>
> #include <stdio.h>
> #include <stdlib.h>
> #include <string.h>
> #include <sys/syscall.h>
> #include <sys/types.h>
> #include <unistd.h>
> 
> #ifndef __NR_bpf
> #define __NR_bpf 321
> #endif
> 
> #define BITMASK(bf_off, bf_len) (((1ull << (bf_len)) - 1) << (bf_off))
> #define STORE_BY_BITMASK(type, htobe, addr, val, bf_off, bf_len)     \
>   *(type*)(addr) =                                                   \
>       htobe((htobe(*(type*)(addr)) & ~BITMASK((bf_off), (bf_len))) | \
>             (((type)(val) << (bf_off)) & BITMASK((bf_off), (bf_len))))
> 
> uint64_t r[1] = {0xffffffffffffffff};
> 
> int main(void) {
>   syscall(__NR_mmap, /*addr=*/0x1ffff000ul, /*len=*/0x1000ul, /*prot=*/0ul,
>           /*flags=*/0x32ul, /*fd=*/-1, /*offset=*/0ul);
>   syscall(__NR_mmap, /*addr=*/0x20000000ul, /*len=*/0x1000000ul,
> /*prot=*/7ul,
>           /*flags=*/0x32ul, /*fd=*/-1, /*offset=*/0ul);
>   syscall(__NR_mmap, /*addr=*/0x21000000ul, /*len=*/0x1000ul, /*prot=*/0ul,
>           /*flags=*/0x32ul, /*fd=*/-1, /*offset=*/0ul);
>   intptr_t res = 0;
>   *(uint32_t*)0x20000140 = 2;
>   *(uint32_t*)0x20000144 = 3;
>   *(uint64_t*)0x20000148 = 0x20000200;
>   *(uint8_t*)0x20000200 = 0x18;
>   STORE_BY_BITMASK(uint8_t, , 0x20000201, 0, 0, 4);
>   STORE_BY_BITMASK(uint8_t, , 0x20000201, 0, 4, 4);
>   *(uint16_t*)0x20000202 = 0;
>   *(uint32_t*)0x20000204 = 0;
>   *(uint8_t*)0x20000208 = 0;
>   *(uint8_t*)0x20000209 = 0;
>   *(uint16_t*)0x2000020a = 0;
>   *(uint32_t*)0x2000020c = 0;
>   *(uint8_t*)0x20000210 = 0x95;
>   *(uint8_t*)0x20000211 = 0;
>   *(uint16_t*)0x20000212 = 0;
>   *(uint32_t*)0x20000214 = 0;
>   *(uint64_t*)0x20000150 = 0x20000240;
>   memcpy((void*)0x20000240, "GPL\000", 4);
>   *(uint32_t*)0x20000158 = 0;
>   *(uint32_t*)0x2000015c = 0;
>   *(uint64_t*)0x20000160 = 0;
>   *(uint32_t*)0x20000168 = 0;
>   *(uint32_t*)0x2000016c = 0;
>   memset((void*)0x20000170, 0, 16);
>   *(uint32_t*)0x20000180 = 0;
>   *(uint32_t*)0x20000184 = 0x30;
>   *(uint32_t*)0x20000188 = 0;
>   *(uint32_t*)0x2000018c = 0;
>   *(uint64_t*)0x20000190 = 0;
>   *(uint32_t*)0x20000198 = 0;
>   *(uint32_t*)0x2000019c = 0;
>   *(uint64_t*)0x200001a0 = 0;
>   *(uint32_t*)0x200001a8 = 0;
>   *(uint32_t*)0x200001ac = 0;
>   *(uint32_t*)0x200001b0 = 0;
>   *(uint32_t*)0x200001b4 = 0;
>   *(uint64_t*)0x200001b8 = 0;
>   *(uint64_t*)0x200001c0 = 0;
>   *(uint32_t*)0x200001c8 = 0;
>   *(uint32_t*)0x200001cc = 0;
>   res = syscall(__NR_bpf, /*cmd=*/5ul, /*arg=*/0x20000140ul,
> /*size=*/0x90ul);
>   if (res != -1) r[0] = res;
>   memcpy((void*)0x20000000, "./file0\000", 8);
>   syscall(__NR_creat, /*file=*/0x20000000ul, /*mode=*/0ul);
>   *(uint32_t*)0x20000340 = r[0];
>   *(uint32_t*)0x20000344 = 0;
>   *(uint32_t*)0x20000348 = 0x30;
>   *(uint32_t*)0x2000034c = 0;
>   *(uint64_t*)0x20000350 = 0x20000080;
>   memcpy((void*)0x20000080, "./file0\000", 8);
>   *(uint64_t*)0x20000358 = 0x200000c0;
>   *(uint64_t*)0x200000c0 = 0;
>   *(uint64_t*)0x20000360 = 0;
>   *(uint64_t*)0x20000368 = 0;
>   *(uint32_t*)0x20000370 = 0xffffff1f;
>   *(uint32_t*)0x20000374 = 0;
>   *(uint32_t*)0x20000378 = 0;
>   syscall(__NR_bpf, /*cmd=*/0x1cul, /*arg=*/0x20000340ul, /*size=*/0x40ul);
>   return 0;
> }
> 
> =* repro.txt =*
> r0 = bpf$PROG_LOAD(0x5, &(0x7f0000000140)={0x2, 0x3,
> &(0x7f0000000200)=@framed, &(0x7f0000000240)='GPL\x00', 0x0, 0x0, 0x0, 0x0,
> 0x0, '\x00', 0x0, 0x30, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
> 0x0, 0x0, 0x0}, 0x90)
> creat(&(0x7f0000000000)='./file0\x00', 0x0)
> bpf$BPF_LINK_CREATE_XDP(0x1c, &(0x7f0000000340)={r0, 0x0, 0x30, 0x0,
> @val=@uprobe_multi={&(0x7f0000000080)='./file0\x00',
> &(0x7f00000000c0)=[0x0], 0x0, 0x0, 0xffffff1f}}, 0x40)
> 
> 
> See aslo
> https://gist.github.com/xrivendell7/15d43946c73aa13247b4b20b68798aaa


-- 
Masami Hiramatsu (Google) <mhiramat@kernel.org>

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: WARNING: kmalloc bug in bpf_uprobe_multi_link_attach
  2023-12-11  8:12 WARNING: kmalloc bug in bpf_uprobe_multi_link_attach xingwei lee
@ 2023-12-11 11:29 ` Hou Tao
  2023-12-11 13:01   ` Jiri Olsa
  0 siblings, 1 reply; 8+ messages in thread
From: Hou Tao @ 2023-12-11 11:29 UTC (permalink / raw)
  To: xingwei lee, ast
  Cc: daniel, andrii, martin.lau, song, yonghong.song, john.fastabend,
	kpsingh, sdf, haoluo, jolsa, rostedt, mhiramat, mathieu.desnoyers,
	bpf, linux-kernel, linux-trace-kernel

Hi,

On 12/11/2023 4:12 PM, xingwei lee wrote:
> Sorry for containing HTML part, repeat the mail
> Hello I found a bug in net/bpf in the lastest upstream linux and
> lastest net tree.
> WARNING: kmalloc bug in bpf_uprobe_multi_link_attach
>
> kernel: net 28a7cb045ab700de5554193a1642917602787784
> Kernel config: https://github.com/google/syzkaller/commits/fc59b78e3174009510ed15f20665e7ab2435ebee
>
> in the lastest net tree, the crash like:
>
> [   68.363836][ T8223] ------------[ cut here ]------------
> [   68.364967][ T8223] WARNING: CPU: 2 PID: 8223 at mm/util.c:632
> kvmalloc_node+0x18a/0x1a0
> [   68.366527][ T8223] Modules linked in:
> [   68.367882][ T8223] CPU: 2 PID: 8223 Comm: 36d Not tainted
> 6.7.0-rc4-00146-g28a7cb045ab7 #2
> [   68.369260][ T8223] Hardware name: QEMU Standard PC (i440FX + PIIX,
> 1996), BIOS 1.16.2-1.fc38 04/014
> [   68.370811][ T8223] RIP: 0010:kvmalloc_node+0x18a/0x1a0
> [   68.371689][ T8223] Code: dc 1c 00 eb aa e8 86 33 c6 ff 41 81 e4 00
> 20 00 00 31 ff 44 89 e6 e8 e5 20
> [   68.375001][ T8223] RSP: 0018:ffffc9001088fb68 EFLAGS: 00010293
> [   68.375989][ T8223] RAX: 0000000000000000 RBX: 00000037ffffcec8
> RCX: ffffffff81c1a32b
> [   68.377154][ T8223] RDX: ffff88802cc00040 RSI: ffffffff81c1a339
> RDI: 0000000000000005
> [   68.377950][ T8223] RBP: 0000000000000400 R08: 0000000000000005
> R09: 0000000000000000
> [   68.378744][ T8223] R10: 0000000000000000 R11: 0000000000000000
> R12: 0000000000000000
> [   68.379523][ T8223] R13: 00000000ffffffff R14: ffff888017eb4a28
> R15: 0000000000000000
> [   68.380307][ T8223] FS:  0000000000827380(0000)
> GS:ffff8880b9900000(0000) knlGS:0000000000000000
> [   68.381185][ T8223] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   68.381843][ T8223] CR2: 0000000020000140 CR3: 00000000204d2000
> CR4: 0000000000750ef0
> [   68.382624][ T8223] PKRU: 55555554
> [   68.382978][ T8223] Call Trace:
> [   68.383312][ T8223]  <TASK>
> [   68.383608][ T8223]  ? show_regs+0x8f/0xa0
> [   68.384052][ T8223]  ? __warn+0xe6/0x390
> [   68.384470][ T8223]  ? kvmalloc_node+0x18a/0x1a0
> [   68.385111][ T8223]  ? report_bug+0x3b9/0x580
> [   68.385585][ T8223]  ? handle_bug+0x67/0x90
> [   68.386032][ T8223]  ? exc_invalid_op+0x17/0x40
> [   68.386503][ T8223]  ? asm_exc_invalid_op+0x1a/0x20
> [   68.387065][ T8223]  ? kvmalloc_node+0x17b/0x1a0
> [   68.387551][ T8223]  ? kvmalloc_node+0x189/0x1a0
> [   68.388051][ T8223]  ? kvmalloc_node+0x18a/0x1a0
> [   68.388537][ T8223]  ? kvmalloc_node+0x189/0x1a0
> [   68.389038][ T8223]  bpf_uprobe_multi_link_attach+0x436/0xfb0

It seems a big attr->link_create.uprobe_multi.cnt is passed to
bpf_uprobe_multi_link_attach(). Could you please try the first patch in
the following patch set ?

https://lore.kernel.org/bpf/20231211112843.4147157-1-houtao@huaweicloud.com/T/#t
> [   68.389633][ T8223]  ? __might_fault+0x13f/0x1a0
> [   68.390129][ T8223]  ? bpf_kprobe_multi_link_attach+0x10/0x10

SNIP
>   res = syscall(__NR_bpf, /*cmd=*/5ul, /*arg=*/0x20000140ul, /*size=*/0x90ul);
>   if (res != -1) r[0] = res;
>   memcpy((void*)0x20000000, "./file0\000", 8);
>   syscall(__NR_creat, /*file=*/0x20000000ul, /*mode=*/0ul);
>   *(uint32_t*)0x20000340 = r[0];
>   *(uint32_t*)0x20000344 = 0;
>   *(uint32_t*)0x20000348 = 0x30;
>   *(uint32_t*)0x2000034c = 0;
>   *(uint64_t*)0x20000350 = 0x20000080;
>   memcpy((void*)0x20000080, "./file0\000", 8);

0x20000350 is the address of attr->link_create.uprobe_multi.path.
>   *(uint64_t*)0x20000358 = 0x200000c0;
>   *(uint64_t*)0x200000c0 = 0;
>   *(uint64_t*)0x20000360 = 0;
>   *(uint64_t*)0x20000368 = 0;
>   *(uint32_t*)0x20000370 = 0xffffff1f;

The value of attr->link_create.uprobe_multi.cnt is 0xffffff1f, so 
0xffffff1f * sizeof(bpf_uprobe) will be greater than INT_MAX, and
triggers the warning in mm/util.c:

        /* Don't even allow crazy sizes */
        if (unlikely(size > INT_MAX)) {
                WARN_ON_ONCE(!(flags & __GFP_NOWARN));
                return NULL;
        }

Adding __GFP_NOWARN when doing kvcalloc() can fix the warning.
>   *(uint32_t*)0x20000374 = 0;
>   *(uint32_t*)0x20000378 = 0;
>   syscall(__NR_bpf, /*cmd=*/0x1cul, /*arg=*/0x20000340ul, /*size=*/0x40ul);
>   return 0;
> }
>
> =* repro.txt =*
> r0 = bpf$PROG_LOAD(0x5, &(0x7f0000000140)={0x2, 0x3,
> &(0x7f0000000200)=@framed, &(0x7f0000000240)='GPL\x00', 0x0, 0x0, 0x0,
> 0x0, 0x0, '\x00', 0x0, 0x30, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
> 0x0, 0x0, 0x0, 0x0, 0x0}, 0x90)
> creat(&(0x7f0000000000)='./file0\x00', 0x0)
> bpf$BPF_LINK_CREATE_XDP(0x1c, &(0x7f0000000340)={r0, 0x0, 0x30, 0x0,
> @val=@uprobe_multi={&(0x7f0000000080)='./file0\x00',
> &(0x7f00000000c0)=[0x0], 0x0, 0x0, 0xffffff1f}}, 0x40
>
>
> See aslo https://gist.github.com/xrivendell7/15d43946c73aa13247b4b20b68798aaa
>
> .


^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: WARNING: kmalloc bug in bpf_uprobe_multi_link_attach
  2023-12-11 11:29 ` Hou Tao
@ 2023-12-11 13:01   ` Jiri Olsa
  2023-12-11 14:34     ` Jiri Olsa
  0 siblings, 1 reply; 8+ messages in thread
From: Jiri Olsa @ 2023-12-11 13:01 UTC (permalink / raw)
  To: Hou Tao, mhiramat
  Cc: xingwei lee, ast, daniel, andrii, martin.lau, song, yonghong.song,
	john.fastabend, kpsingh, sdf, haoluo, rostedt, mathieu.desnoyers,
	bpf, linux-kernel, linux-trace-kernel

On Mon, Dec 11, 2023 at 07:29:40PM +0800, Hou Tao wrote:

SNIP

> 
> It seems a big attr->link_create.uprobe_multi.cnt is passed to
> bpf_uprobe_multi_link_attach(). Could you please try the first patch in
> the following patch set ?
> 
> https://lore.kernel.org/bpf/20231211112843.4147157-1-houtao@huaweicloud.com/T/#t
> > [   68.389633][ T8223]  ? __might_fault+0x13f/0x1a0
> > [   68.390129][ T8223]  ? bpf_kprobe_multi_link_attach+0x10/0x10
> 
> SNIP
> >   res = syscall(__NR_bpf, /*cmd=*/5ul, /*arg=*/0x20000140ul, /*size=*/0x90ul);
> >   if (res != -1) r[0] = res;
> >   memcpy((void*)0x20000000, "./file0\000", 8);
> >   syscall(__NR_creat, /*file=*/0x20000000ul, /*mode=*/0ul);
> >   *(uint32_t*)0x20000340 = r[0];
> >   *(uint32_t*)0x20000344 = 0;
> >   *(uint32_t*)0x20000348 = 0x30;
> >   *(uint32_t*)0x2000034c = 0;
> >   *(uint64_t*)0x20000350 = 0x20000080;
> >   memcpy((void*)0x20000080, "./file0\000", 8);
> 
> 0x20000350 is the address of attr->link_create.uprobe_multi.path.
> >   *(uint64_t*)0x20000358 = 0x200000c0;
> >   *(uint64_t*)0x200000c0 = 0;
> >   *(uint64_t*)0x20000360 = 0;
> >   *(uint64_t*)0x20000368 = 0;
> >   *(uint32_t*)0x20000370 = 0xffffff1f;
> 
> The value of attr->link_create.uprobe_multi.cnt is 0xffffff1f, so 
> 0xffffff1f * sizeof(bpf_uprobe) will be greater than INT_MAX, and
> triggers the warning in mm/util.c:
> 
>         /* Don't even allow crazy sizes */
>         if (unlikely(size > INT_MAX)) {
>                 WARN_ON_ONCE(!(flags & __GFP_NOWARN));
>                 return NULL;
>         }
> 
> Adding __GFP_NOWARN when doing kvcalloc() can fix the warning.

hi,
looks like that's the case.. thanks for fixing that

btw while checking on that I found kprobe_multi bench attach test
takes forever on latest bpf-next/master

	test_kprobe_multi_bench_attach:PASS:bpf_program__attach_kprobe_multi_opts 0 nsec
	test_kprobe_multi_bench_attach: found 56140 functions
	test_kprobe_multi_bench_attach: attached in  89.174s
	test_kprobe_multi_bench_attach: detached in  13.245s
	#113/1   kprobe_multi_bench_attach/kernel:OK

Masami,
any idea of any change on fprobe/ftrace side recently? I'm going to check ;-)

thanks,
jirka

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: WARNING: kmalloc bug in bpf_uprobe_multi_link_attach
  2023-12-11 13:01   ` Jiri Olsa
@ 2023-12-11 14:34     ` Jiri Olsa
  0 siblings, 0 replies; 8+ messages in thread
From: Jiri Olsa @ 2023-12-11 14:34 UTC (permalink / raw)
  To: Jiri Olsa
  Cc: Hou Tao, mhiramat, xingwei lee, ast, daniel, andrii, martin.lau,
	song, yonghong.song, john.fastabend, kpsingh, sdf, haoluo,
	rostedt, mathieu.desnoyers, bpf, linux-kernel, linux-trace-kernel

On Mon, Dec 11, 2023 at 02:01:43PM +0100, Jiri Olsa wrote:
> On Mon, Dec 11, 2023 at 07:29:40PM +0800, Hou Tao wrote:
> 
> SNIP
> 
> > 
> > It seems a big attr->link_create.uprobe_multi.cnt is passed to
> > bpf_uprobe_multi_link_attach(). Could you please try the first patch in
> > the following patch set ?
> > 
> > https://lore.kernel.org/bpf/20231211112843.4147157-1-houtao@huaweicloud.com/T/#t
> > > [   68.389633][ T8223]  ? __might_fault+0x13f/0x1a0
> > > [   68.390129][ T8223]  ? bpf_kprobe_multi_link_attach+0x10/0x10
> > 
> > SNIP
> > >   res = syscall(__NR_bpf, /*cmd=*/5ul, /*arg=*/0x20000140ul, /*size=*/0x90ul);
> > >   if (res != -1) r[0] = res;
> > >   memcpy((void*)0x20000000, "./file0\000", 8);
> > >   syscall(__NR_creat, /*file=*/0x20000000ul, /*mode=*/0ul);
> > >   *(uint32_t*)0x20000340 = r[0];
> > >   *(uint32_t*)0x20000344 = 0;
> > >   *(uint32_t*)0x20000348 = 0x30;
> > >   *(uint32_t*)0x2000034c = 0;
> > >   *(uint64_t*)0x20000350 = 0x20000080;
> > >   memcpy((void*)0x20000080, "./file0\000", 8);
> > 
> > 0x20000350 is the address of attr->link_create.uprobe_multi.path.
> > >   *(uint64_t*)0x20000358 = 0x200000c0;
> > >   *(uint64_t*)0x200000c0 = 0;
> > >   *(uint64_t*)0x20000360 = 0;
> > >   *(uint64_t*)0x20000368 = 0;
> > >   *(uint32_t*)0x20000370 = 0xffffff1f;
> > 
> > The value of attr->link_create.uprobe_multi.cnt is 0xffffff1f, so 
> > 0xffffff1f * sizeof(bpf_uprobe) will be greater than INT_MAX, and
> > triggers the warning in mm/util.c:
> > 
> >         /* Don't even allow crazy sizes */
> >         if (unlikely(size > INT_MAX)) {
> >                 WARN_ON_ONCE(!(flags & __GFP_NOWARN));
> >                 return NULL;
> >         }
> > 
> > Adding __GFP_NOWARN when doing kvcalloc() can fix the warning.
> 
> hi,
> looks like that's the case.. thanks for fixing that
> 
> btw while checking on that I found kprobe_multi bench attach test
> takes forever on latest bpf-next/master
> 
> 	test_kprobe_multi_bench_attach:PASS:bpf_program__attach_kprobe_multi_opts 0 nsec
> 	test_kprobe_multi_bench_attach: found 56140 functions
> 	test_kprobe_multi_bench_attach: attached in  89.174s
> 	test_kprobe_multi_bench_attach: detached in  13.245s
> 	#113/1   kprobe_multi_bench_attach/kernel:OK
> 
> Masami,
> any idea of any change on fprobe/ftrace side recently? I'm going to check ;-)

nah sry, I had IBT enabled.. forgot the reason, but it's slow ;-)

jirka

^ permalink raw reply	[flat|nested] 8+ messages in thread

* WARNING: kmalloc bug in bpf_uprobe_multi_link_attach
@ 2024-05-14  7:27 Ubisectech Sirius
  2024-05-15 21:30 ` Alexei Starovoitov
  0 siblings, 1 reply; 8+ messages in thread
From: Ubisectech Sirius @ 2024-05-14  7:27 UTC (permalink / raw)
  To: linux-trace-kernel, linux-kernel; +Cc: ast, daniel, andrii

[-- Attachment #1: Type: text/plain, Size: 2911 bytes --]

Hello.
We are Ubisectech Sirius Team, the vulnerability lab of China ValiantSec. Recently, our team has discovered a issue in Linux kernel 6.7.  Attached to the email were a PoC file of the issue.

Stack dump:

loop3: detected capacity change from 0 to 8
MTD: Attempt to mount non-MTD device "/dev/loop3"
------------[ cut here ]------------
WARNING: CPU: 1 PID: 10075 at mm/util.c:632 kvmalloc_node+0x199/0x1b0 mm/util.c:632
Modules linked in:
CPU: 1 PID: 10075 Comm: syz-executor.3 Not tainted 6.7.0 #2
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
RIP: 0010:kvmalloc_node+0x199/0x1b0 mm/util.c:632
Code: 02 1d 00 eb aa e8 a7 49 c6 ff 41 81 e5 00 20 00 00 31 ff 44 89 ee e8 36 45 c6 ff 45 85 ed 0f 85 1b ff ff ff e8 88 49 c6 ff 90 <0f> 0b 90 e9 dd fe ff ff 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40
RSP: 0018:ffffc90002007b60 EFLAGS: 00010212
RAX: 00000000000023e4 RBX: 0000000000000400 RCX: ffffc90003aaa000
RDX: 0000000000040000 RSI: ffffffff81c3acc8 RDI: 0000000000000005
RBP: 00000037ffffcec8 R08: 0000000000000005 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
R13: 0000000000000000 R14: 00000000ffffffff R15: ffff88805ff6e1b8
FS:  00007fc62205f640(0000) GS:ffff88807ec00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b2e026000 CR3: 000000005f338000 CR4: 0000000000750ef0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
PKRU: 55555554
Call Trace:
 <TASK>
 kvmalloc include/linux/slab.h:738 [inline]
 kvmalloc_array include/linux/slab.h:756 [inline]
 kvcalloc include/linux/slab.h:761 [inline]
 bpf_uprobe_multi_link_attach+0x3fe/0xf60 kernel/trace/bpf_trace.c:3239
 link_create kernel/bpf/syscall.c:5012 [inline]
 __sys_bpf+0x2e85/0x4e00 kernel/bpf/syscall.c:5453
 __do_sys_bpf kernel/bpf/syscall.c:5487 [inline]
 __se_sys_bpf kernel/bpf/syscall.c:5485 [inline]
 __x64_sys_bpf+0x78/0xc0 kernel/bpf/syscall.c:5485
 do_syscall_x64 arch/x86/entry/common.c:52 [inline]
 do_syscall_64+0x43/0x120 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x6f/0x77
RIP: 0033:0x7fc62128fd6d
Code: c3 e8 97 2b 00 00 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fc62205f028 EFLAGS: 00000246 ORIG_RAX: 0000000000000141
RAX: ffffffffffffffda RBX: 00007fc6213cbf80 RCX: 00007fc62128fd6d
RDX: 0000000000000040 RSI: 00000000200001c0 RDI: 000000000000001c
RBP: 00007fc6212f14cd R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 000000000000000b R14: 00007fc6213cbf80 R15: 00007fc62203f000
 </TASK>

Thank you for taking the time to read this email and we look forward to working with you further.




[-- Attachment #2: poc.c --]
[-- Type: application/octet-stream, Size: 18890 bytes --]

// autogenerated by syzkaller (https://github.com/google/syzkaller)

#define _GNU_SOURCE

#include <endian.h>
#include <errno.h>
#include <fcntl.h>
#include <setjmp.h>
#include <stdbool.h>
#include <stddef.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/ioctl.h>
#include <sys/mman.h>
#include <sys/mount.h>
#include <sys/stat.h>
#include <sys/syscall.h>
#include <sys/types.h>
#include <unistd.h>

#include <linux/loop.h>

#ifndef __NR_bpf
#define __NR_bpf 321
#endif
#ifndef __NR_memfd_create
#define __NR_memfd_create 319
#endif

static unsigned long long procid;

#define BITMASK(bf_off, bf_len) (((1ull << (bf_len)) - 1) << (bf_off))
#define STORE_BY_BITMASK(type, htobe, addr, val, bf_off, bf_len)               \
  *(type*)(addr) =                                                             \
      htobe((htobe(*(type*)(addr)) & ~BITMASK((bf_off), (bf_len))) |           \
            (((type)(val) << (bf_off)) & BITMASK((bf_off), (bf_len))))

//% This code is derived from puff.{c,h}, found in the zlib development. The
//% original files come with the following copyright notice:

//% Copyright (C) 2002-2013 Mark Adler, all rights reserved
//% version 2.3, 21 Jan 2013
//% This software is provided 'as-is', without any express or implied
//% warranty.  In no event will the author be held liable for any damages
//% arising from the use of this software.
//% Permission is granted to anyone to use this software for any purpose,
//% including commercial applications, and to alter it and redistribute it
//% freely, subject to the following restrictions:
//% 1. The origin of this software must not be misrepresented; you must not
//%    claim that you wrote the original software. If you use this software
//%    in a product, an acknowledgment in the product documentation would be
//%    appreciated but is not required.
//% 2. Altered source versions must be plainly marked as such, and must not be
//%    misrepresented as being the original software.
//% 3. This notice may not be removed or altered from any source distribution.
//% Mark Adler    madler@alumni.caltech.edu

//% BEGIN CODE DERIVED FROM puff.{c,h}

#define MAXBITS 15
#define MAXLCODES 286
#define MAXDCODES 30
#define MAXCODES (MAXLCODES + MAXDCODES)
#define FIXLCODES 288

struct puff_state {
  unsigned char* out;
  unsigned long outlen;
  unsigned long outcnt;
  const unsigned char* in;
  unsigned long inlen;
  unsigned long incnt;
  int bitbuf;
  int bitcnt;
  jmp_buf env;
};
static int puff_bits(struct puff_state* s, int need)
{
  long val = s->bitbuf;
  while (s->bitcnt < need) {
    if (s->incnt == s->inlen)
      longjmp(s->env, 1);
    val |= (long)(s->in[s->incnt++]) << s->bitcnt;
    s->bitcnt += 8;
  }
  s->bitbuf = (int)(val >> need);
  s->bitcnt -= need;
  return (int)(val & ((1L << need) - 1));
}
static int puff_stored(struct puff_state* s)
{
  s->bitbuf = 0;
  s->bitcnt = 0;
  if (s->incnt + 4 > s->inlen)
    return 2;
  unsigned len = s->in[s->incnt++];
  len |= s->in[s->incnt++] << 8;
  if (s->in[s->incnt++] != (~len & 0xff) ||
      s->in[s->incnt++] != ((~len >> 8) & 0xff))
    return -2;
  if (s->incnt + len > s->inlen)
    return 2;
  if (s->outcnt + len > s->outlen)
    return 1;
  for (; len--; s->outcnt++, s->incnt++) {
    if (s->in[s->incnt])
      s->out[s->outcnt] = s->in[s->incnt];
  }
  return 0;
}
struct puff_huffman {
  short* count;
  short* symbol;
};
static int puff_decode(struct puff_state* s, const struct puff_huffman* h)
{
  int first = 0;
  int index = 0;
  int bitbuf = s->bitbuf;
  int left = s->bitcnt;
  int code = first = index = 0;
  int len = 1;
  short* next = h->count + 1;
  while (1) {
    while (left--) {
      code |= bitbuf & 1;
      bitbuf >>= 1;
      int count = *next++;
      if (code - count < first) {
        s->bitbuf = bitbuf;
        s->bitcnt = (s->bitcnt - len) & 7;
        return h->symbol[index + (code - first)];
      }
      index += count;
      first += count;
      first <<= 1;
      code <<= 1;
      len++;
    }
    left = (MAXBITS + 1) - len;
    if (left == 0)
      break;
    if (s->incnt == s->inlen)
      longjmp(s->env, 1);
    bitbuf = s->in[s->incnt++];
    if (left > 8)
      left = 8;
  }
  return -10;
}
static int puff_construct(struct puff_huffman* h, const short* length, int n)
{
  int len;
  for (len = 0; len <= MAXBITS; len++)
    h->count[len] = 0;
  int symbol;
  for (symbol = 0; symbol < n; symbol++)
    (h->count[length[symbol]])++;
  if (h->count[0] == n)
    return 0;
  int left = 1;
  for (len = 1; len <= MAXBITS; len++) {
    left <<= 1;
    left -= h->count[len];
    if (left < 0)
      return left;
  }
  short offs[MAXBITS + 1];
  offs[1] = 0;
  for (len = 1; len < MAXBITS; len++)
    offs[len + 1] = offs[len] + h->count[len];
  for (symbol = 0; symbol < n; symbol++)
    if (length[symbol] != 0)
      h->symbol[offs[length[symbol]]++] = symbol;
  return left;
}
static int puff_codes(struct puff_state* s, const struct puff_huffman* lencode,
                      const struct puff_huffman* distcode)
{
  static const short lens[29] = {3,  4,  5,  6,   7,   8,   9,   10,  11, 13,
                                 15, 17, 19, 23,  27,  31,  35,  43,  51, 59,
                                 67, 83, 99, 115, 131, 163, 195, 227, 258};
  static const short lext[29] = {0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2,
                                 2, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 0};
  static const short dists[30] = {
      1,    2,    3,    4,    5,    7,    9,    13,    17,    25,
      33,   49,   65,   97,   129,  193,  257,  385,   513,   769,
      1025, 1537, 2049, 3073, 4097, 6145, 8193, 12289, 16385, 24577};
  static const short dext[30] = {0, 0, 0,  0,  1,  1,  2,  2,  3,  3,
                                 4, 4, 5,  5,  6,  6,  7,  7,  8,  8,
                                 9, 9, 10, 10, 11, 11, 12, 12, 13, 13};
  int symbol;
  do {
    symbol = puff_decode(s, lencode);
    if (symbol < 0)
      return symbol;
    if (symbol < 256) {
      if (s->outcnt == s->outlen)
        return 1;
      if (symbol)
        s->out[s->outcnt] = symbol;
      s->outcnt++;
    } else if (symbol > 256) {
      symbol -= 257;
      if (symbol >= 29)
        return -10;
      int len = lens[symbol] + puff_bits(s, lext[symbol]);
      symbol = puff_decode(s, distcode);
      if (symbol < 0)
        return symbol;
      unsigned dist = dists[symbol] + puff_bits(s, dext[symbol]);
      if (dist > s->outcnt)
        return -11;
      if (s->outcnt + len > s->outlen)
        return 1;
      while (len--) {
        if (dist <= s->outcnt && s->out[s->outcnt - dist])
          s->out[s->outcnt] = s->out[s->outcnt - dist];
        s->outcnt++;
      }
    }
  } while (symbol != 256);
  return 0;
}
static int puff_fixed(struct puff_state* s)
{
  static int virgin = 1;
  static short lencnt[MAXBITS + 1], lensym[FIXLCODES];
  static short distcnt[MAXBITS + 1], distsym[MAXDCODES];
  static struct puff_huffman lencode, distcode;
  if (virgin) {
    lencode.count = lencnt;
    lencode.symbol = lensym;
    distcode.count = distcnt;
    distcode.symbol = distsym;
    short lengths[FIXLCODES];
    int symbol;
    for (symbol = 0; symbol < 144; symbol++)
      lengths[symbol] = 8;
    for (; symbol < 256; symbol++)
      lengths[symbol] = 9;
    for (; symbol < 280; symbol++)
      lengths[symbol] = 7;
    for (; symbol < FIXLCODES; symbol++)
      lengths[symbol] = 8;
    puff_construct(&lencode, lengths, FIXLCODES);
    for (symbol = 0; symbol < MAXDCODES; symbol++)
      lengths[symbol] = 5;
    puff_construct(&distcode, lengths, MAXDCODES);
    virgin = 0;
  }
  return puff_codes(s, &lencode, &distcode);
}
static int puff_dynamic(struct puff_state* s)
{
  static const short order[19] = {16, 17, 18, 0, 8,  7, 9,  6, 10, 5,
                                  11, 4,  12, 3, 13, 2, 14, 1, 15};
  int nlen = puff_bits(s, 5) + 257;
  int ndist = puff_bits(s, 5) + 1;
  int ncode = puff_bits(s, 4) + 4;
  if (nlen > MAXLCODES || ndist > MAXDCODES)
    return -3;
  short lengths[MAXCODES];
  int index;
  for (index = 0; index < ncode; index++)
    lengths[order[index]] = puff_bits(s, 3);
  for (; index < 19; index++)
    lengths[order[index]] = 0;
  short lencnt[MAXBITS + 1], lensym[MAXLCODES];
  struct puff_huffman lencode = {lencnt, lensym};
  int err = puff_construct(&lencode, lengths, 19);
  if (err != 0)
    return -4;
  index = 0;
  while (index < nlen + ndist) {
    int symbol;
    int len;
    symbol = puff_decode(s, &lencode);
    if (symbol < 0)
      return symbol;
    if (symbol < 16)
      lengths[index++] = symbol;
    else {
      len = 0;
      if (symbol == 16) {
        if (index == 0)
          return -5;
        len = lengths[index - 1];
        symbol = 3 + puff_bits(s, 2);
      } else if (symbol == 17)
        symbol = 3 + puff_bits(s, 3);
      else
        symbol = 11 + puff_bits(s, 7);
      if (index + symbol > nlen + ndist)
        return -6;
      while (symbol--)
        lengths[index++] = len;
    }
  }
  if (lengths[256] == 0)
    return -9;
  err = puff_construct(&lencode, lengths, nlen);
  if (err && (err < 0 || nlen != lencode.count[0] + lencode.count[1]))
    return -7;
  short distcnt[MAXBITS + 1], distsym[MAXDCODES];
  struct puff_huffman distcode = {distcnt, distsym};
  err = puff_construct(&distcode, lengths + nlen, ndist);
  if (err && (err < 0 || ndist != distcode.count[0] + distcode.count[1]))
    return -8;
  return puff_codes(s, &lencode, &distcode);
}
static int puff(unsigned char* dest, unsigned long* destlen,
                const unsigned char* source, unsigned long sourcelen)
{
  struct puff_state s = {
      .out = dest,
      .outlen = *destlen,
      .outcnt = 0,
      .in = source,
      .inlen = sourcelen,
      .incnt = 0,
      .bitbuf = 0,
      .bitcnt = 0,
  };
  int err;
  if (setjmp(s.env) != 0)
    err = 2;
  else {
    int last;
    do {
      last = puff_bits(&s, 1);
      int type = puff_bits(&s, 2);
      err = type == 0 ? puff_stored(&s)
                      : (type == 1 ? puff_fixed(&s)
                                   : (type == 2 ? puff_dynamic(&s) : -1));
      if (err != 0)
        break;
    } while (!last);
  }
  *destlen = s.outcnt;
  return err;
}

//% END CODE DERIVED FROM puff.{c,h}

#define ZLIB_HEADER_WIDTH 2

static int puff_zlib_to_file(const unsigned char* source,
                             unsigned long sourcelen, int dest_fd)
{
  if (sourcelen < ZLIB_HEADER_WIDTH)
    return 0;
  source += ZLIB_HEADER_WIDTH;
  sourcelen -= ZLIB_HEADER_WIDTH;
  const unsigned long max_destlen = 132 << 20;
  void* ret = mmap(0, max_destlen, PROT_WRITE | PROT_READ,
                   MAP_PRIVATE | MAP_ANON, -1, 0);
  if (ret == MAP_FAILED)
    return -1;
  unsigned char* dest = (unsigned char*)ret;
  unsigned long destlen = max_destlen;
  int err = puff(dest, &destlen, source, sourcelen);
  if (err) {
    munmap(dest, max_destlen);
    errno = -err;
    return -1;
  }
  if (write(dest_fd, dest, destlen) != (ssize_t)destlen) {
    munmap(dest, max_destlen);
    return -1;
  }
  return munmap(dest, max_destlen);
}

static int setup_loop_device(unsigned char* data, unsigned long size,
                             const char* loopname, int* loopfd_p)
{
  int err = 0, loopfd = -1;
  int memfd = syscall(__NR_memfd_create, "syzkaller", 0);
  if (memfd == -1) {
    err = errno;
    goto error;
  }
  if (puff_zlib_to_file(data, size, memfd)) {
    err = errno;
    goto error_close_memfd;
  }
  loopfd = open(loopname, O_RDWR);
  if (loopfd == -1) {
    err = errno;
    goto error_close_memfd;
  }
  if (ioctl(loopfd, LOOP_SET_FD, memfd)) {
    if (errno != EBUSY) {
      err = errno;
      goto error_close_loop;
    }
    ioctl(loopfd, LOOP_CLR_FD, 0);
    usleep(1000);
    if (ioctl(loopfd, LOOP_SET_FD, memfd)) {
      err = errno;
      goto error_close_loop;
    }
  }
  close(memfd);
  *loopfd_p = loopfd;
  return 0;

error_close_loop:
  close(loopfd);
error_close_memfd:
  close(memfd);
error:
  errno = err;
  return -1;
}

static void reset_loop_device(const char* loopname)
{
  int loopfd = open(loopname, O_RDWR);
  if (loopfd == -1) {
    return;
  }
  if (ioctl(loopfd, LOOP_CLR_FD, 0)) {
  }
  close(loopfd);
}

static long syz_mount_image(volatile long fsarg, volatile long dir,
                            volatile long flags, volatile long optsarg,
                            volatile long change_dir,
                            volatile unsigned long size, volatile long image)
{
  unsigned char* data = (unsigned char*)image;
  int res = -1, err = 0, need_loop_device = !!size;
  char* mount_opts = (char*)optsarg;
  char* target = (char*)dir;
  char* fs = (char*)fsarg;
  char* source = NULL;
  char loopname[64];
  if (need_loop_device) {
    int loopfd;
    memset(loopname, 0, sizeof(loopname));
    snprintf(loopname, sizeof(loopname), "/dev/loop%llu", procid);
    if (setup_loop_device(data, size, loopname, &loopfd) == -1)
      return -1;
    close(loopfd);
    source = loopname;
  }
  mkdir(target, 0777);
  char opts[256];
  memset(opts, 0, sizeof(opts));
  if (strlen(mount_opts) > (sizeof(opts) - 32)) {
  }
  strncpy(opts, mount_opts, sizeof(opts) - 32);
  if (strcmp(fs, "iso9660") == 0) {
    flags |= MS_RDONLY;
  } else if (strncmp(fs, "ext", 3) == 0) {
    bool has_remount_ro = false;
    char* remount_ro_start = strstr(opts, "errors=remount-ro");
    if (remount_ro_start != NULL) {
      char after = *(remount_ro_start + strlen("errors=remount-ro"));
      char before = remount_ro_start == opts ? '\0' : *(remount_ro_start - 1);
      has_remount_ro = ((before == '\0' || before == ',') &&
                        (after == '\0' || after == ','));
    }
    if (strstr(opts, "errors=panic") || !has_remount_ro)
      strcat(opts, ",errors=continue");
  } else if (strcmp(fs, "xfs") == 0) {
    strcat(opts, ",nouuid");
  }
  res = mount(source, target, fs, flags, opts);
  if (res == -1) {
    err = errno;
    goto error_clear_loop;
  }
  res = open(target, O_RDONLY | O_DIRECTORY);
  if (res == -1) {
    err = errno;
    goto error_clear_loop;
  }
  if (change_dir) {
    res = chdir(target);
    if (res == -1) {
      err = errno;
    }
  }

error_clear_loop:
  if (need_loop_device)
    reset_loop_device(loopname);
  errno = err;
  return res;
}

uint64_t r[1] = {0xffffffffffffffff};

int main(void)
{
  syscall(__NR_mmap, /*addr=*/0x1ffff000ul, /*len=*/0x1000ul, /*prot=*/0ul,
          /*flags=MAP_FIXED|MAP_ANONYMOUS|MAP_PRIVATE*/ 0x32ul, /*fd=*/-1,
          /*offset=*/0ul);
  syscall(__NR_mmap, /*addr=*/0x20000000ul, /*len=*/0x1000000ul,
          /*prot=PROT_WRITE|PROT_READ|PROT_EXEC*/ 7ul,
          /*flags=MAP_FIXED|MAP_ANONYMOUS|MAP_PRIVATE*/ 0x32ul, /*fd=*/-1,
          /*offset=*/0ul);
  syscall(__NR_mmap, /*addr=*/0x21000000ul, /*len=*/0x1000ul, /*prot=*/0ul,
          /*flags=MAP_FIXED|MAP_ANONYMOUS|MAP_PRIVATE*/ 0x32ul, /*fd=*/-1,
          /*offset=*/0ul);
  intptr_t res = 0;
  memcpy((void*)0x20000140, "cramfs\000", 7);
  memcpy((void*)0x20000180, "./file0\000", 8);
  memcpy(
      (void*)0x20000400,
      "\x78\x9c\xec\xd0\xbf\x6b\x1a\x61\x1c\xc7\xf1\xf7\xa3\x56\xed\x4f\x2d\xb5"
      "\xd0\x16\x5a\x0b\x1d\x7a\x28\xe2\x79\x62\xb7\x0e\x5a\x2a\x3d\xa8\x3d\x68"
      "\xb7\x4e\x82\x5e\x69\xc1\x1f\x25\x42\xc8\x18\x03\xd9\x32\xe4\x0f\x70\x88"
      "\x09\x64\x12\x87\x90\x31\x43\x62\x26\x13\x85\x60\xfe\x8d\x1c\x64\x0b\x64"
      "\x31\x3c\x77\x47\xe2\x9c\xf9\x79\x2d\x77\xf7\xfd\x7c\xf9\x1c\xcf\xf3\xf5"
      "\xf3\x54\x23\x0e\x41\x3c\x5f\xda\xcd\xff\x4b\x76\xa7\x63\xd7\xdf\xff\xb4"
      "\x2a\xe5\x5f\x7b\xfb\x07\xcf\x48\x42\x04\x88\xfa\xf9\xf5\xdc\x23\xf7\x0f"
      "\x8b\xf0\x57\x3e\x43\xe0\x74\xa1\x0e\x1c\x3f\xf4\xcb\x6a\xed\x86\xfc\x76"
      "\x8a\xa0\x01\xa5\xc7\xf0\xe7\x5f\xc3\xd6\xf1\x76\x1f\xc9\x59\x82\x5b\x72"
      "\xa6\x7d\x80\xd1\x4b\x6f\xcf\x80\x80\xd3\x25\xca\xc2\x2c\xef\xef\xbd\x09"
      "\xcd\x29\xc5\xee\xfa\xe6\x7d\xf8\x28\xfb\x9e\x7b\xb3\x9c\xf0\x4b\x7b\xe9"
      "\xe1\xe0\xe4\xc7\x24\x60\x66\x52\xbb\x6f\x83\xac\x9b\xe9\xe4\x0b\x01\x2b"
      "\xfe\x7c\x6c\x66\x52\xef\xec\x0d\xa3\xfc\x5a\x2e\xb7\xfa\xb0\x4d\xc9\xcd"
      "\xcf\xb3\xe3\xcc\x59\x76\x38\x98\x4d\x27\x95\xef\x56\xc5\x9a\xe6\x0d\xe3"
      "\x53\x5e\xcf\xe9\x7a\x61\x66\x9d\x4e\xcc\xc2\xea\x26\xa1\x6f\x4f\x96\xe1"
      "\xb7\xd7\x17\xf6\xff\x98\x0a\x73\xe1\xbe\xac\x09\xe8\x09\x18\xb8\xb9\x73"
      "\x24\x9e\x02\xa3\xad\x2b\xab\x89\x90\xc7\xbe\x6c\xc5\x41\xb8\x89\xbc\x5b"
      "\x37\x89\x45\x16\x93\x5a\xf5\xd5\x83\x9d\x6a\x22\x1e\x20\xa8\x21\x50\x14"
      "\x45\x51\x14\x45\x51\x14\x45\x51\x94\x7b\xba\x09\x00\x00\xff\xff\x86\xf5"
      "\x68\x06",
      344);
  syz_mount_image(/*fs=*/0x20000140, /*dir=*/0x20000180, /*flags=*/0,
                  /*opts=*/0x20000340, /*chdir=*/1, /*size=*/0x158,
                  /*img=*/0x20000400);
  memcpy((void*)0x20000040, "./file0\000", 8);
  syscall(__NR_chdir, /*dir=*/0x20000040ul);
  *(uint32_t*)0x20000680 = 2;
  *(uint32_t*)0x20000684 = 3;
  *(uint64_t*)0x20000688 = 0x20000080;
  *(uint8_t*)0x20000080 = 0x18;
  STORE_BY_BITMASK(uint8_t, , 0x20000081, 0, 0, 4);
  STORE_BY_BITMASK(uint8_t, , 0x20000081, 0, 4, 4);
  *(uint16_t*)0x20000082 = 0;
  *(uint32_t*)0x20000084 = 0;
  *(uint8_t*)0x20000088 = 0;
  *(uint8_t*)0x20000089 = 0;
  *(uint16_t*)0x2000008a = 0;
  *(uint32_t*)0x2000008c = 0;
  *(uint8_t*)0x20000090 = 0x95;
  *(uint8_t*)0x20000091 = 0;
  *(uint16_t*)0x20000092 = 0;
  *(uint32_t*)0x20000094 = 0;
  *(uint64_t*)0x20000690 = 0x20000000;
  memcpy((void*)0x20000000, "syzkaller\000", 10);
  *(uint32_t*)0x20000698 = 0;
  *(uint32_t*)0x2000069c = 0;
  *(uint64_t*)0x200006a0 = 0;
  *(uint32_t*)0x200006a8 = 0;
  *(uint32_t*)0x200006ac = 0;
  memset((void*)0x200006b0, 0, 16);
  *(uint32_t*)0x200006c0 = 0;
  *(uint32_t*)0x200006c4 = 0x30;
  *(uint32_t*)0x200006c8 = 0;
  *(uint32_t*)0x200006cc = 0;
  *(uint64_t*)0x200006d0 = 0;
  *(uint32_t*)0x200006d8 = 0;
  *(uint32_t*)0x200006dc = 0;
  *(uint64_t*)0x200006e0 = 0;
  *(uint32_t*)0x200006e8 = 0;
  *(uint32_t*)0x200006ec = 0;
  *(uint32_t*)0x200006f0 = 0;
  *(uint32_t*)0x200006f4 = 0;
  *(uint64_t*)0x200006f8 = 0;
  *(uint64_t*)0x20000700 = 0;
  *(uint32_t*)0x20000708 = 0;
  *(uint32_t*)0x2000070c = 0;
  res = syscall(__NR_bpf, /*cmd=*/5ul, /*arg=*/0x20000680ul, /*size=*/0x90ul);
  if (res != -1)
    r[0] = res;
  *(uint32_t*)0x200001c0 = r[0];
  *(uint32_t*)0x200001c4 = 0;
  *(uint32_t*)0x200001c8 = 0x30;
  *(uint32_t*)0x200001cc = 0;
  *(uint64_t*)0x200001d0 = 0x20000040;
  memcpy((void*)0x20000040, "./file0\000", 8);
  *(uint64_t*)0x200001d8 = 0x200000c0;
  *(uint64_t*)0x200000c0 = 0;
  *(uint64_t*)0x200001e0 = 0;
  *(uint64_t*)0x200001e8 = 0;
  *(uint32_t*)0x200001f0 = 0xffffff1f;
  *(uint32_t*)0x200001f4 = 1;
  *(uint32_t*)0x200001f8 = 0;
  syscall(__NR_bpf, /*cmd=*/0x1cul, /*arg=*/0x200001c0ul, /*size=*/0x40ul);
  return 0;
}

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: WARNING: kmalloc bug in bpf_uprobe_multi_link_attach
  2024-05-14  7:27 Ubisectech Sirius
@ 2024-05-15 21:30 ` Alexei Starovoitov
  2024-05-15 21:47   ` Jiri Olsa
  0 siblings, 1 reply; 8+ messages in thread
From: Alexei Starovoitov @ 2024-05-15 21:30 UTC (permalink / raw)
  To: Ubisectech Sirius, Jiri Olsa
  Cc: linux-trace-kernel, linux-kernel, ast, daniel, andrii

On Tue, May 14, 2024 at 12:33 AM Ubisectech Sirius
<bugreport@ubisectech.com> wrote:
>
> Hello.
> We are Ubisectech Sirius Team, the vulnerability lab of China ValiantSec. Recently, our team has discovered a issue in Linux kernel 6.7.  Attached to the email were a PoC file of the issue.

Jiri,

please take a look.

> Stack dump:
>
> loop3: detected capacity change from 0 to 8
> MTD: Attempt to mount non-MTD device "/dev/loop3"
> ------------[ cut here ]------------
> WARNING: CPU: 1 PID: 10075 at mm/util.c:632 kvmalloc_node+0x199/0x1b0 mm/util.c:632
> Modules linked in:
> CPU: 1 PID: 10075 Comm: syz-executor.3 Not tainted 6.7.0 #2
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
> RIP: 0010:kvmalloc_node+0x199/0x1b0 mm/util.c:632
> Code: 02 1d 00 eb aa e8 a7 49 c6 ff 41 81 e5 00 20 00 00 31 ff 44 89 ee e8 36 45 c6 ff 45 85 ed 0f 85 1b ff ff ff e8 88 49 c6 ff 90 <0f> 0b 90 e9 dd fe ff ff 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40
> RSP: 0018:ffffc90002007b60 EFLAGS: 00010212
> RAX: 00000000000023e4 RBX: 0000000000000400 RCX: ffffc90003aaa000
> RDX: 0000000000040000 RSI: ffffffff81c3acc8 RDI: 0000000000000005
> RBP: 00000037ffffcec8 R08: 0000000000000005 R09: 0000000000000000
> R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
> R13: 0000000000000000 R14: 00000000ffffffff R15: ffff88805ff6e1b8
> FS:  00007fc62205f640(0000) GS:ffff88807ec00000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 0000001b2e026000 CR3: 000000005f338000 CR4: 0000000000750ef0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> PKRU: 55555554
> Call Trace:
>  <TASK>
>  kvmalloc include/linux/slab.h:738 [inline]
>  kvmalloc_array include/linux/slab.h:756 [inline]
>  kvcalloc include/linux/slab.h:761 [inline]
>  bpf_uprobe_multi_link_attach+0x3fe/0xf60 kernel/trace/bpf_trace.c:3239
>  link_create kernel/bpf/syscall.c:5012 [inline]
>  __sys_bpf+0x2e85/0x4e00 kernel/bpf/syscall.c:5453
>  __do_sys_bpf kernel/bpf/syscall.c:5487 [inline]
>  __se_sys_bpf kernel/bpf/syscall.c:5485 [inline]
>  __x64_sys_bpf+0x78/0xc0 kernel/bpf/syscall.c:5485
>  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
>  do_syscall_64+0x43/0x120 arch/x86/entry/common.c:83
>  entry_SYSCALL_64_after_hwframe+0x6f/0x77
> RIP: 0033:0x7fc62128fd6d
> Code: c3 e8 97 2b 00 00 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b0 ff ff ff f7 d8 64 89 01 48
> RSP: 002b:00007fc62205f028 EFLAGS: 00000246 ORIG_RAX: 0000000000000141
> RAX: ffffffffffffffda RBX: 00007fc6213cbf80 RCX: 00007fc62128fd6d
> RDX: 0000000000000040 RSI: 00000000200001c0 RDI: 000000000000001c
> RBP: 00007fc6212f14cd R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
> R13: 000000000000000b R14: 00007fc6213cbf80 R15: 00007fc62203f000
>  </TASK>
>
> Thank you for taking the time to read this email and we look forward to working with you further.
>
>
>

^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: WARNING: kmalloc bug in bpf_uprobe_multi_link_attach
  2024-05-15 21:30 ` Alexei Starovoitov
@ 2024-05-15 21:47   ` Jiri Olsa
  0 siblings, 0 replies; 8+ messages in thread
From: Jiri Olsa @ 2024-05-15 21:47 UTC (permalink / raw)
  To: Alexei Starovoitov
  Cc: Ubisectech Sirius, linux-trace-kernel, linux-kernel, ast, daniel,
	andrii

On Wed, May 15, 2024 at 02:30:37PM -0700, Alexei Starovoitov wrote:
> On Tue, May 14, 2024 at 12:33 AM Ubisectech Sirius
> <bugreport@ubisectech.com> wrote:
> >
> > Hello.
> > We are Ubisectech Sirius Team, the vulnerability lab of China ValiantSec. Recently, our team has discovered a issue in Linux kernel 6.7.  Attached to the email were a PoC file of the issue.
> 
> Jiri,
> 
> please take a look.
> 
> > Stack dump:
> >
> > loop3: detected capacity change from 0 to 8
> > MTD: Attempt to mount non-MTD device "/dev/loop3"
> > ------------[ cut here ]------------
> > WARNING: CPU: 1 PID: 10075 at mm/util.c:632 kvmalloc_node+0x199/0x1b0 mm/util.c:632

hi,
this should be already fixed via:
  https://lore.kernel.org/bpf/20231215100708.2265609-2-houtao@huaweicloud.com/

original report was in here:
  https://lore.kernel.org/bpf/CABOYnLwwJY=yFAGie59LFsUsBAgHfroVqbzZ5edAXbFE3YiNVA@mail.gmail.com/

the fix should be in v6.7, can you check if your kernel has:
  8b2efe51ba85 bpf: Limit the number of uprobes when attaching program to multiple uprobes

thanks,
jirka


> > Modules linked in:
> > CPU: 1 PID: 10075 Comm: syz-executor.3 Not tainted 6.7.0 #2
> > Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
> > RIP: 0010:kvmalloc_node+0x199/0x1b0 mm/util.c:632
> > Code: 02 1d 00 eb aa e8 a7 49 c6 ff 41 81 e5 00 20 00 00 31 ff 44 89 ee e8 36 45 c6 ff 45 85 ed 0f 85 1b ff ff ff e8 88 49 c6 ff 90 <0f> 0b 90 e9 dd fe ff ff 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40
> > RSP: 0018:ffffc90002007b60 EFLAGS: 00010212
> > RAX: 00000000000023e4 RBX: 0000000000000400 RCX: ffffc90003aaa000
> > RDX: 0000000000040000 RSI: ffffffff81c3acc8 RDI: 0000000000000005
> > RBP: 00000037ffffcec8 R08: 0000000000000005 R09: 0000000000000000
> > R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
> > R13: 0000000000000000 R14: 00000000ffffffff R15: ffff88805ff6e1b8
> > FS:  00007fc62205f640(0000) GS:ffff88807ec00000(0000) knlGS:0000000000000000
> > CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > CR2: 0000001b2e026000 CR3: 000000005f338000 CR4: 0000000000750ef0
> > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > PKRU: 55555554
> > Call Trace:
> >  <TASK>
> >  kvmalloc include/linux/slab.h:738 [inline]
> >  kvmalloc_array include/linux/slab.h:756 [inline]
> >  kvcalloc include/linux/slab.h:761 [inline]
> >  bpf_uprobe_multi_link_attach+0x3fe/0xf60 kernel/trace/bpf_trace.c:3239
> >  link_create kernel/bpf/syscall.c:5012 [inline]
> >  __sys_bpf+0x2e85/0x4e00 kernel/bpf/syscall.c:5453
> >  __do_sys_bpf kernel/bpf/syscall.c:5487 [inline]
> >  __se_sys_bpf kernel/bpf/syscall.c:5485 [inline]
> >  __x64_sys_bpf+0x78/0xc0 kernel/bpf/syscall.c:5485
> >  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
> >  do_syscall_64+0x43/0x120 arch/x86/entry/common.c:83
> >  entry_SYSCALL_64_after_hwframe+0x6f/0x77
> > RIP: 0033:0x7fc62128fd6d
> > Code: c3 e8 97 2b 00 00 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b0 ff ff ff f7 d8 64 89 01 48
> > RSP: 002b:00007fc62205f028 EFLAGS: 00000246 ORIG_RAX: 0000000000000141
> > RAX: ffffffffffffffda RBX: 00007fc6213cbf80 RCX: 00007fc62128fd6d
> > RDX: 0000000000000040 RSI: 00000000200001c0 RDI: 000000000000001c
> > RBP: 00007fc6212f14cd R08: 0000000000000000 R09: 0000000000000000
> > R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
> > R13: 000000000000000b R14: 00007fc6213cbf80 R15: 00007fc62203f000
> >  </TASK>
> >
> > Thank you for taking the time to read this email and we look forward to working with you further.
> >
> >
> >

^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2024-05-15 21:47 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-12-11  8:12 WARNING: kmalloc bug in bpf_uprobe_multi_link_attach xingwei lee
2023-12-11 11:29 ` Hou Tao
2023-12-11 13:01   ` Jiri Olsa
2023-12-11 14:34     ` Jiri Olsa
     [not found] <CABOYnLz2e+_0P88RgoDy6epWz9xrM2zhfMQdVrcjNiPqrFcBeQ@mail.gmail.com>
2023-12-11 11:22 ` Masami Hiramatsu
  -- strict thread matches above, loose matches on Subject: below --
2024-05-14  7:27 Ubisectech Sirius
2024-05-15 21:30 ` Alexei Starovoitov
2024-05-15 21:47   ` Jiri Olsa

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).