Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH] bpf: fix misaligned access for BPF_PROG_TYPE_PERF_EVENT
 program type on x86_32 platform
To:     Alexei Starovoitov <alexei.starovoitov@gmail.com>,
        Wang YanQing <udknight@gmail.com>, ast@kernel.org,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20180426095749.GA29207@udknight>
 <20180427224854.2g7ximim7nwkgdpd@ast-mbp>
From:   Daniel Borkmann <daniel@iogearbox.net>
Message-ID: <c62438dd-2e2e-8c70-aa19-37d5173d76d1@iogearbox.net>
Date:   Sat, 28 Apr 2018 01:33:15 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.3.0
MIME-Version: 1.0
In-Reply-To: <20180427224854.2g7ximim7nwkgdpd@ast-mbp>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 04/28/2018 12:48 AM, Alexei Starovoitov wrote:
> On Thu, Apr 26, 2018 at 05:57:49PM +0800, Wang YanQing wrote:
>> All the testcases for BPF_PROG_TYPE_PERF_EVENT program type in
>> test_verifier(kselftest) report below errors on x86_32:
>> "
>> 172/p unpriv: spill/fill of different pointers ldx FAIL
>> Unexpected error message!
>> 0: (bf) r6 = r10
>> 1: (07) r6 += -8
>> 2: (15) if r1 == 0x0 goto pc+3
>> R1=ctx(id=0,off=0,imm=0) R6=fp-8,call_-1 R10=fp0,call_-1
>> 3: (bf) r2 = r10
>> 4: (07) r2 += -76
>> 5: (7b) *(u64 *)(r6 +0) = r2
>> 6: (55) if r1 != 0x0 goto pc+1
>> R1=ctx(id=0,off=0,imm=0) R2=fp-76,call_-1 R6=fp-8,call_-1 R10=fp0,call_-1 fp-8=fp
>> 7: (7b) *(u64 *)(r6 +0) = r1
>> 8: (79) r1 = *(u64 *)(r6 +0)
>> 9: (79) r1 = *(u64 *)(r1 +68)
>> invalid bpf_context access off=68 size=8
>>
>> 378/p check bpf_perf_event_data->sample_period byte load permitted FAIL
>> Failed to load prog 'Permission denied'!
>> 0: (b7) r0 = 0
>> 1: (71) r0 = *(u8 *)(r1 +68)
>> invalid bpf_context access off=68 size=1
>>
>> 379/p check bpf_perf_event_data->sample_period half load permitted FAIL
>> Failed to load prog 'Permission denied'!
>> 0: (b7) r0 = 0
>> 1: (69) r0 = *(u16 *)(r1 +68)
>> invalid bpf_context access off=68 size=2
>>
>> 380/p check bpf_perf_event_data->sample_period word load permitted FAIL
>> Failed to load prog 'Permission denied'!
>> 0: (b7) r0 = 0
>> 1: (61) r0 = *(u32 *)(r1 +68)
>> invalid bpf_context access off=68 size=4
>>
>> 381/p check bpf_perf_event_data->sample_period dword load permitted FAIL
>> Failed to load prog 'Permission denied'!
>> 0: (b7) r0 = 0
>> 1: (79) r0 = *(u64 *)(r1 +68)
>> invalid bpf_context access off=68 size=8
>> "
>>
>> This patch fix it, the fix isn't only necessary for x86_32, it will fix the
>> same problem for other platforms too, if their size of bpf_user_pt_regs_t
>> can't divide exactly into 8.
>>
>> Signed-off-by: Wang YanQing <udknight@gmail.com>
>> ---
>>  Hi all!
>>  After mainline accept this patch, then we need to submit a sync patch
>>  to update the tools/include/uapi/linux/bpf_perf_event.h.
>>
>>  Thanks.
>>
>>  include/uapi/linux/bpf_perf_event.h | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/include/uapi/linux/bpf_perf_event.h b/include/uapi/linux/bpf_perf_event.h
>> index eb1b9d2..ff4c092 100644
>> --- a/include/uapi/linux/bpf_perf_event.h
>> +++ b/include/uapi/linux/bpf_perf_event.h
>> @@ -12,7 +12,7 @@
>>  
>>  struct bpf_perf_event_data {
>>  	bpf_user_pt_regs_t regs;
>> -	__u64 sample_period;
>> +	__u64 sample_period __attribute__((aligned(8)));
> 
> I don't think this necessary.
> imo it's a bug in pe_prog_is_valid_access
> that should have allowed 8-byte access to 4-byte aligned sample_period.
> The access rewritten by pe_prog_convert_ctx_access anyway,
> no alignment issues as far as I can see.

Right, good point. Wang, could you give the below a test run:

diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 56ba0f2..95b9142 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -833,8 +833,14 @@ static bool pe_prog_is_valid_access(int off, int size, enum bpf_access_type type
 		return false;
 	if (type != BPF_READ)
 		return false;
-	if (off % size != 0)
-		return false;
+	if (off % size != 0) {
+		if (sizeof(long) != 4)
+			return false;
+		if (size != 8)
+			return false;
+		if (off % size != 4)
+			return false;
+	}

 	switch (off) {
 	case bpf_ctx_range(struct bpf_perf_event_data, sample_period):