* [PATCH -next] bpf: Add oversize check before call kvmalloc()
@ 2021-11-27 15:31 Bixuan Cui
2021-11-30 22:53 ` Andrii Nakryiko
0 siblings, 1 reply; 3+ messages in thread
From: Bixuan Cui @ 2021-11-27 15:31 UTC (permalink / raw)
To: linux-kernel, bpf, netdev
Cc: cuibixuan, ast, daniel, andrii, kafai, songliubraving, yhs,
john.fastabend, kpsingh
Commit 7661809d493b ("mm: don't allow oversized kvmalloc() calls") add
the oversize check. When the allocation is larger than what kvmalloc()
supports, the following warning triggered:
WARNING: CPU: 1 PID: 372 at mm/util.c:597 kvmalloc_node+0x111/0x120
mm/util.c:597
Modules linked in:
CPU: 1 PID: 372 Comm: syz-executor.4 Not tainted 5.15.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
Google 01/01/2011
RIP: 0010:kvmalloc_node+0x111/0x120 mm/util.c:597
Code: 01 00 00 00 4c 89 e7 e8 7d f7 0c 00 49 89 c5 e9 69 ff ff ff e8 60
20 d1 ff 41 89 ed 41 81 cd 00 20 01 00 eb 95 e8 4f 20 d1 ff <0f> 0b e9
4c ff ff ff 0f 1f 84 00 00 00 00 00 55 48 89 fd 53 e8 36
RSP: 0018:ffffc90002bf7c98 EFLAGS: 00010216
RAX: 00000000000000ec RBX: 1ffff9200057ef9f RCX: ffffc9000ac63000
RDX: 0000000000040000 RSI: ffffffff81a6a621 RDI: 0000000000000003
RBP: 0000000000102cc0 R08: 000000007fffffff R09: 00000000ffffffff
R10: ffffffff81a6a5de R11: 0000000000000000 R12: 00000000ffff9aaa
R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000000
FS: 00007f05f2573700(0000) GS:ffff8880b9d00000(0000)
knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b2f424000 CR3: 0000000027d2c000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
kvmalloc include/linux/slab.h:741 [inline]
map_lookup_elem kernel/bpf/syscall.c:1090 [inline]
__sys_bpf+0x3a5b/0x5f00 kernel/bpf/syscall.c:4603
__do_sys_bpf kernel/bpf/syscall.c:4722 [inline]
__se_sys_bpf kernel/bpf/syscall.c:4720 [inline]
__x64_sys_bpf+0x75/0xb0 kernel/bpf/syscall.c:4720
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
The type of 'value_size' is u32, its value may exceed INT_MAX.
Reported-by: syzbot+cecf5b7071a0dfb76530@syzkaller.appspotmail.com
Signed-off-by: Bixuan Cui <cuibixuan@linux.alibaba.com>
---
kernel/bpf/syscall.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 1033ee8..f5bc380 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1094,6 +1094,10 @@ static int map_lookup_elem(union bpf_attr *attr)
}
value_size = bpf_map_value_size(map);
+ if (value_size > INT_MAX) {
+ err = -EINVAL;
+ goto err_put;
+ }
err = -ENOMEM;
value = kvmalloc(value_size, GFP_USER | __GFP_NOWARN);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH -next] bpf: Add oversize check before call kvmalloc()
2021-11-27 15:31 [PATCH -next] bpf: Add oversize check before call kvmalloc() Bixuan Cui
@ 2021-11-30 22:53 ` Andrii Nakryiko
2021-12-01 10:29 ` Leon Romanovsky
0 siblings, 1 reply; 3+ messages in thread
From: Andrii Nakryiko @ 2021-11-30 22:53 UTC (permalink / raw)
To: Bixuan Cui
Cc: open list, bpf, Networking, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin Lau, Song Liu, Yonghong Song,
john fastabend, KP Singh
On Sat, Nov 27, 2021 at 7:32 AM Bixuan Cui <cuibixuan@linux.alibaba.com> wrote:
>
> Commit 7661809d493b ("mm: don't allow oversized kvmalloc() calls") add
> the oversize check. When the allocation is larger than what kvmalloc()
> supports, the following warning triggered:
>
> WARNING: CPU: 1 PID: 372 at mm/util.c:597 kvmalloc_node+0x111/0x120
> mm/util.c:597
> Modules linked in:
> CPU: 1 PID: 372 Comm: syz-executor.4 Not tainted 5.15.0-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
> Google 01/01/2011
> RIP: 0010:kvmalloc_node+0x111/0x120 mm/util.c:597
> Code: 01 00 00 00 4c 89 e7 e8 7d f7 0c 00 49 89 c5 e9 69 ff ff ff e8 60
> 20 d1 ff 41 89 ed 41 81 cd 00 20 01 00 eb 95 e8 4f 20 d1 ff <0f> 0b e9
> 4c ff ff ff 0f 1f 84 00 00 00 00 00 55 48 89 fd 53 e8 36
> RSP: 0018:ffffc90002bf7c98 EFLAGS: 00010216
> RAX: 00000000000000ec RBX: 1ffff9200057ef9f RCX: ffffc9000ac63000
> RDX: 0000000000040000 RSI: ffffffff81a6a621 RDI: 0000000000000003
> RBP: 0000000000102cc0 R08: 000000007fffffff R09: 00000000ffffffff
> R10: ffffffff81a6a5de R11: 0000000000000000 R12: 00000000ffff9aaa
> R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000000
> FS: 00007f05f2573700(0000) GS:ffff8880b9d00000(0000)
> knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 0000001b2f424000 CR3: 0000000027d2c000 CR4: 00000000003506e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
> <TASK>
> kvmalloc include/linux/slab.h:741 [inline]
> map_lookup_elem kernel/bpf/syscall.c:1090 [inline]
> __sys_bpf+0x3a5b/0x5f00 kernel/bpf/syscall.c:4603
> __do_sys_bpf kernel/bpf/syscall.c:4722 [inline]
> __se_sys_bpf kernel/bpf/syscall.c:4720 [inline]
> __x64_sys_bpf+0x75/0xb0 kernel/bpf/syscall.c:4720
> do_syscall_x64 arch/x86/entry/common.c:50 [inline]
> do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
> entry_SYSCALL_64_after_hwframe+0x44/0xae
>
> The type of 'value_size' is u32, its value may exceed INT_MAX.
>
> Reported-by: syzbot+cecf5b7071a0dfb76530@syzkaller.appspotmail.com
> Signed-off-by: Bixuan Cui <cuibixuan@linux.alibaba.com>
> ---
> kernel/bpf/syscall.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
> index 1033ee8..f5bc380 100644
> --- a/kernel/bpf/syscall.c
> +++ b/kernel/bpf/syscall.c
> @@ -1094,6 +1094,10 @@ static int map_lookup_elem(union bpf_attr *attr)
> }
>
> value_size = bpf_map_value_size(map);
> + if (value_size > INT_MAX) {
> + err = -EINVAL;
-E2BIG makes a bit more sense in this scenario?
> + goto err_put;
> + }
>
> err = -ENOMEM;
> value = kvmalloc(value_size, GFP_USER | __GFP_NOWARN);
> --
> 1.8.3.1
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH -next] bpf: Add oversize check before call kvmalloc()
2021-11-30 22:53 ` Andrii Nakryiko
@ 2021-12-01 10:29 ` Leon Romanovsky
0 siblings, 0 replies; 3+ messages in thread
From: Leon Romanovsky @ 2021-12-01 10:29 UTC (permalink / raw)
To: Andrii Nakryiko
Cc: Bixuan Cui, open list, bpf, Networking, Alexei Starovoitov,
Daniel Borkmann, Andrii Nakryiko, Martin Lau, Song Liu,
Yonghong Song, john fastabend, KP Singh
On Tue, Nov 30, 2021 at 02:53:16PM -0800, Andrii Nakryiko wrote:
> On Sat, Nov 27, 2021 at 7:32 AM Bixuan Cui <cuibixuan@linux.alibaba.com> wrote:
> >
> > Commit 7661809d493b ("mm: don't allow oversized kvmalloc() calls") add
> > the oversize check. When the allocation is larger than what kvmalloc()
> > supports, the following warning triggered:
> >
> > WARNING: CPU: 1 PID: 372 at mm/util.c:597 kvmalloc_node+0x111/0x120
> > mm/util.c:597
> > Modules linked in:
> > CPU: 1 PID: 372 Comm: syz-executor.4 Not tainted 5.15.0-syzkaller #0
> > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
> > Google 01/01/2011
> > RIP: 0010:kvmalloc_node+0x111/0x120 mm/util.c:597
> > Code: 01 00 00 00 4c 89 e7 e8 7d f7 0c 00 49 89 c5 e9 69 ff ff ff e8 60
> > 20 d1 ff 41 89 ed 41 81 cd 00 20 01 00 eb 95 e8 4f 20 d1 ff <0f> 0b e9
> > 4c ff ff ff 0f 1f 84 00 00 00 00 00 55 48 89 fd 53 e8 36
> > RSP: 0018:ffffc90002bf7c98 EFLAGS: 00010216
> > RAX: 00000000000000ec RBX: 1ffff9200057ef9f RCX: ffffc9000ac63000
> > RDX: 0000000000040000 RSI: ffffffff81a6a621 RDI: 0000000000000003
> > RBP: 0000000000102cc0 R08: 000000007fffffff R09: 00000000ffffffff
> > R10: ffffffff81a6a5de R11: 0000000000000000 R12: 00000000ffff9aaa
> > R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000000
> > FS: 00007f05f2573700(0000) GS:ffff8880b9d00000(0000)
> > knlGS:0000000000000000
> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > CR2: 0000001b2f424000 CR3: 0000000027d2c000 CR4: 00000000003506e0
> > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > Call Trace:
> > <TASK>
> > kvmalloc include/linux/slab.h:741 [inline]
> > map_lookup_elem kernel/bpf/syscall.c:1090 [inline]
> > __sys_bpf+0x3a5b/0x5f00 kernel/bpf/syscall.c:4603
> > __do_sys_bpf kernel/bpf/syscall.c:4722 [inline]
> > __se_sys_bpf kernel/bpf/syscall.c:4720 [inline]
> > __x64_sys_bpf+0x75/0xb0 kernel/bpf/syscall.c:4720
> > do_syscall_x64 arch/x86/entry/common.c:50 [inline]
> > do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
> > entry_SYSCALL_64_after_hwframe+0x44/0xae
> >
> > The type of 'value_size' is u32, its value may exceed INT_MAX.
> >
> > Reported-by: syzbot+cecf5b7071a0dfb76530@syzkaller.appspotmail.com
> > Signed-off-by: Bixuan Cui <cuibixuan@linux.alibaba.com>
> > ---
> > kernel/bpf/syscall.c | 4 ++++
> > 1 file changed, 4 insertions(+)
> >
> > diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
> > index 1033ee8..f5bc380 100644
> > --- a/kernel/bpf/syscall.c
> > +++ b/kernel/bpf/syscall.c
> > @@ -1094,6 +1094,10 @@ static int map_lookup_elem(union bpf_attr *attr)
> > }
> >
> > value_size = bpf_map_value_size(map);
> > + if (value_size > INT_MAX) {
> > + err = -EINVAL;
>
> -E2BIG makes a bit more sense in this scenario?
kvmalloc should be fixed do not print WARN_ON() on attempts to provide
such allocations sizes.
We are in RDMA, and everyone who receives this size as an input from the
user, seeing this type of error.
Thanks
>
> > + goto err_put;
> > + }
> >
> > err = -ENOMEM;
> > value = kvmalloc(value_size, GFP_USER | __GFP_NOWARN);
> > --
> > 1.8.3.1
> >
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2021-12-01 10:29 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-11-27 15:31 [PATCH -next] bpf: Add oversize check before call kvmalloc() Bixuan Cui
2021-11-30 22:53 ` Andrii Nakryiko
2021-12-01 10:29 ` Leon Romanovsky
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).