[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CA+kUaCcaZCb4+zqQthsUvEApG4xFdGsSZ_vQ_1s=G49JHQQh9w@mail.gmail.com>
Date: Wed, 19 May 2021 11:29:16 +0800
From: Yao Haidong <yaohaidong369@...il.com>
To: Miklos Szeredi <miklos@...redi.hu>
Cc: Orson Zhai <orsonzhai@...il.com>,
Chunyan Zhang <zhang.lyra@...il.com>,
Chunyan Zhang <Chunyan.Zhang@...soc.com>,
Orson Zhai <Orson.Zhai@...soc.com>,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
Andrii Nakryiko <andrii@...nel.org>,
Martin KaFai Lau <kafai@...com>,
Song Liu <songliubraving@...com>, Yonghong Song <yhs@...com>,
John Fastabend <john.fastabend@...il.com>,
KP Singh <kpsingh@...nel.org>,
overlayfs <linux-unionfs@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
network dev <netdev@...r.kernel.org>,
bpf <bpf@...r.kernel.org>, stable <stable@...r.kernel.org>,
Haidong Yao <haidong.yao@...soc.com>
Subject: Re: [PATCH] ovl: useing ovl_revert_creds() instead of revert_creds(),
Hi Miklos:
Sorry,
Sometimes,old_cred is NULL.
Do you think this is a problem?
Can I add ovl_revert_creds()? or You have a better way?
Miklos Szeredi <miklos@...redi.hu> 于2021年5月18日周二 下午9:41写道:
>
> On Tue, 18 May 2021 at 14:21, Haidong Yao <yaohaidong369@...il.com> wrote:
> >
> > From: Haidong Yao <haidong.yao@...soc.com>
> >
> > After execution adb remout,happened crash.
> >
> > Fixes: 292f902a40c1 ("ovl: check permission to open real file")
> >
> > This fixes the warning below.
> >
> > [ 241.778266]c2 [T31619] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000004
> > [ 241.796757]c2 [T31619] Mem abort info:
> > [ 241.800475]c2 [T31619] ESR = 0x96000005
> > [ 241.804457]c2 [T31619] EC = 0x25: DABT (current EL), IL = 32 bits
> > [ 241.810686]c2 [T31619] SET = 0, FnV = 0
> > [ 241.814666]c2 [T31619] EA = 0, S1PTW = 0
> > [ 241.818733]c2 [T31619] Data abort info:
> > [ 241.822542]c2 [T31619] ISV = 0, ISS = 0x00000005
> > [ 241.827302]c2 [T31619] CM = 0, WnR = 0
> > [ 241.831198]c2 [T31619] user pgtable: 4k pages, 39-bit VAs, pgdp=000000017709f000
> > [ 241.838552]c2 [T31619] [0000000000000004] pgd=0000000000000000, pud=0000000000000000
> > [ 241.846255]c2 [T31619] Internal error: Oops: 96000005 [#1] PREEMPT SMP
> > [ 241.852746]c2 [T31619] sprd-sysdump: dump_die_cb save pregs_die_g ok .
> > [ 242.021424]c2 [T31619] CPU: 2 PID: 31619 Comm: ylog.opentcpdum Tainted: G S WC O 5.4.114-g23ed6b182f28-ab19500 #2
> > [ 242.032496]c2 [T31619] Hardware name: Spreadtrum UMS512-1H10 SoC (DT)
> > [ 242.038901]c2 [T31619] pstate: 20400085 (nzCv daIf +PAN -UAO)
> > [ 242.044620]c2 [T31619] pc : bpf_get_current_uid_gid+0x1c/0x34
> > [ 242.050331]c2 [T31619] lr : bpf_prog_4f08a086bc7e64fb_tracepoint_sche+0x5a0/0x2000
> > [ 242.057852]c2 [T31619] sp : ffffffc018fab4e0
> > [ 242.062093]c2 [T31619] x29: ffffffc018fab4e0 x28: 0000000005de0e82
> > [ 242.068323]c2 [T31619] x27: ffffffc01008ee04 x26: 0000000000000000
> > [ 242.074554]c2 [T31619] x25: ffffffc018fab640 x24: ffffffc000000000
> > [ 242.080785]c2 [T31619] x23: ffffffc0121cf868 x22: 000000383e0d5d53
> > [ 242.087015]c2 [T31619] x21: 0000000000000005 x20: 0000000000000007
> > [ 242.093246]c2 [T31619] x19: fffffffebf93cac0 x18: ffffffc016d7d0a0
> > [ 242.099477]c2 [T31619] x17: 0000007f477affdc x16: 0000000000002c26
> > [ 242.105707]c2 [T31619] x15: 000000000000007f x14: 000000000000007f
> > [ 242.111938]c2 [T31619] x13: 000000000001af6a x12: 0000000026762762
> > [ 242.118169]c2 [T31619] x11: 001a39dd8df5aa00 x10: ffffffc010093398
> > [ 242.124400]c2 [T31619] x9 : ffffffc0121bd018 x8 : 0000000000000000
> > [ 242.130630]c2 [T31619] x7 : ffffff80c45c0110 x6 : ffffffc0123b9dc6
> > [ 242.136862]c2 [T31619] x5 : ffffffc012399000 x4 : 0000000000000004
> > [ 242.143093]c2 [T31619] x3 : ffffff80f3412d00 x2 : ffffff80d5395a00
> > [ 242.149322]c2 [T31619] x1 : ffffffc018fab628 x0 : ffffff80c45c0110
> > [ 242.155554]c2 [T31619] Call trace:
> > [ 242.158931]c2 [T31619] bpf_get_current_uid_gid+0x1c/0x34
> > [ 242.164295]c2 [T31619] bpf_prog_4f08a086bc7e64fb_tracepoint_sche+0x5a0/0x2000
> > [ 242.171480]c2 [T31619] trace_call_bpf+0x1b0/0x3d4
> > [ 242.176240]c2 [T31619] perf_trace_sched_switch+0x1a8/0x204
> > [ 242.181779]c2 [T31619] __schedule+0x828/0x9ac
> > [ 242.186190]c2 [T31619] preempt_schedule_common+0x17c/0x2bc
> > [ 242.191730]c2 [T31619] vprintk_emit+0x738/0x7c8
> > [ 242.196316]c2 [T31619] vprintk_func+0x238/0x274
> > [ 242.200901]c2 [T31619] printk+0x64/0x90
> > [ 242.204799]c2 [T31619] die_kernel_fault+0x4c/0x80
> > [ 242.209556]c2 [T31619] __do_kernel_fault+0x24c/0x268
> > [ 242.214574]c2 [T31619] do_page_fault+0xa4/0x744
> > [ 242.219160]c2 [T31619] do_translation_fault+0x60/0x80
> > [ 242.224267]c2 [T31619] do_mem_abort+0x68/0xfc
> > [ 242.228681]c2 [T31619] el1_da+0x1c/0xc0
> > [ 242.232576]c2 [T31619] cap_vm_enough_memory+0x20/0x84
> > [ 242.237681]c2 [T31619] insert_vm_struct+0xf4/0x3cc
> > [ 242.242527]c2 [T31619] bprm_mm_init+0x188/0x2ac
> > [ 242.247113]c2 [T31619] __do_execve_file+0x3f8/0x904
> > [ 242.252045]c2 [T31619] __arm64_sys_execve+0x50/0x64
> > [ 242.256979]c2 [T31619] el0_svc_common+0xc0/0x22c
> > [ 242.261652]c2 [T31619] el0_svc_handler+0x2c/0x3c
> > [ 242.266323]c2 [T31619] el0_svc+0x8/0xc
> > [ 242.270137]c2 [T31619] Code: d503201f d5384108 b40000c8 f943a508 (f8404100)
> > [ 242.277143]c2 [T31619] ---[ end trace d2bf18208d1aac17 ]---
> >
> > Signed-off-by: Haidong Yao <haidong.yao@...soc.com>
> > ---
> > fs/overlayfs/file.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c
> > index 4d53d3b7e5fe..d9bc658f22ee 100644
> > --- a/fs/overlayfs/file.c
> > +++ b/fs/overlayfs/file.c
> > @@ -60,7 +60,7 @@ static struct file *ovl_open_realfile(const struct file *file,
> > realfile = open_with_fake_path(&file->f_path, flags, realinode,
> > current_cred());
> > }
> > - revert_creds(old_cred);
> > + ovl_revert_creds(inode->i_sb, old_cred);
>
> Upstream kernel doesn't have ovl_revert_creds().
>
> Thanks,
> Miklos
Powered by blists - more mailing lists