| Message ID | 391d524c496acc97a8801d8bea80976f58485810.1700676682.git.dxu@dxuuu.xyz |
|---|---|
| State | New |
| Headers | show |
| Series | None | expand |
On Sat, Nov 25, 2023 at 4:52 PM Yonghong Song <yonghong.song@linux.dev> wrote: > > > > > diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c > > index 3065a716544d..ec7e04e012ae 100644 > > --- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c > > +++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c > > @@ -6,6 +6,7 @@ > > * modify it under the terms of version 2 of the GNU General Public > > * License as published by the Free Software Foundation. > > */ > > +#define BPF_NO_PRESERVE_ACCESS_INDEX > > This is a temporary workaround and hopefully we can lift it in the > near future. Please add a comment here with prefix 'Workaround' to > explain why this is needed and later on we can earliy search the > keyword and remember to tackle this. I suspect we will forget to remove this "workaround" and people will start copy pasting it. Let's change the test instead to avoid bitfield access.
On Sat, 2023-11-25 at 20:22 -0800, Yonghong Song wrote: [...] > --- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c > +++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c > @@ -6,7 +6,10 @@ > * modify it under the terms of version 2 of the GNU General Public > * License as published by the Free Software Foundation. > */ > -#define BPF_NO_PRESERVE_ACCESS_INDEX > +#if __has_attribute(preserve_static_offset) > +struct __attribute__((preserve_static_offset)) erspan_md2; > +struct __attribute__((preserve_static_offset)) erspan_metadata; > +#endif > #include "vmlinux.h" [...] > int bpf_skb_get_fou_encap(struct __sk_buff *skb_ctx, > @@ -174,9 +177,13 @@ int erspan_set_tunnel(struct __sk_buff *skb) > __u8 hwid = 7; > > md.version = 2; > +#if __has_attribute(preserve_static_offset) > md.u.md2.dir = direction; > md.u.md2.hwid = hwid & 0xf; > md.u.md2.hwid_upper = (hwid >> 4) & 0x3; > +#else > + /* Change bit-field store to byte(s)-level stores. */ > +#endif > #endif > > ret = bpf_skb_set_tunnel_opt(skb, &md, sizeof(md)); > > ==== > > Eduard, could you double check whether this is a valid use case > to solve this kind of issue with preserve_static_offset attribute? Tbh I'm not sure. This test passes with preserve_static_offset because it suppresses preserve_access_index. In general clang translates bitfield access to a set of IR statements like: C: struct foo { unsigned _; unsigned a:1; ... }; ... foo->a ... IR: %a = getelementptr inbounds %struct.foo, ptr %0, i32 0, i32 1 %bf.load = load i8, ptr %a, align 4 %bf.clear = and i8 %bf.load, 1 %bf.cast = zext i8 %bf.clear to i32 With preserve_static_offset the getelementptr+load are replaced by a single statement which is preserved as-is till code generation, thus load with align 4 is preserved. On the other hand, I'm not sure that clang guarantees that load or stores used for bitfield access would be always aligned according to verifier expectations. I think we should check if there are some clang knobs that prevent generation of unaligned memory access. I'll take a look.
On 11/28/23 11:17 AM, Daniel Xu wrote: > On Tue, Nov 28, 2023 at 10:13:50AM -0600, Daniel Xu wrote: >> On Mon, Nov 27, 2023 at 08:06:01PM -0800, Yonghong Song wrote: >>> On 11/27/23 7:01 PM, Daniel Xu wrote: >>>> On Mon, Nov 27, 2023 at 02:45:11PM -0600, Daniel Xu wrote: >>>>> On Sun, Nov 26, 2023 at 09:53:04PM -0800, Yonghong Song wrote: >>>>>> On 11/27/23 12:44 AM, Yonghong Song wrote: >>>>>>> On 11/26/23 8:52 PM, Eduard Zingerman wrote: >>>>>>>> On Sun, 2023-11-26 at 18:04 -0600, Daniel Xu wrote: >>>>>>>> [...] >>>>>>>>>> Tbh I'm not sure. This test passes with preserve_static_offset >>>>>>>>>> because it suppresses preserve_access_index. In general clang >>>>>>>>>> translates bitfield access to a set of IR statements like: >>>>>>>>>> >>>>>>>>>> C: >>>>>>>>>> struct foo { >>>>>>>>>> unsigned _; >>>>>>>>>> unsigned a:1; >>>>>>>>>> ... >>>>>>>>>> }; >>>>>>>>>> ... foo->a ... >>>>>>>>>> >>>>>>>>>> IR: >>>>>>>>>> %a = getelementptr inbounds %struct.foo, ptr %0, i32 0, i32 1 >>>>>>>>>> %bf.load = load i8, ptr %a, align 4 >>>>>>>>>> %bf.clear = and i8 %bf.load, 1 >>>>>>>>>> %bf.cast = zext i8 %bf.clear to i32 >>>>>>>>>> >>>>>>>>>> With preserve_static_offset the getelementptr+load are replaced by a >>>>>>>>>> single statement which is preserved as-is till code generation, >>>>>>>>>> thus load with align 4 is preserved. >>>>>>>>>> >>>>>>>>>> On the other hand, I'm not sure that clang guarantees that load or >>>>>>>>>> stores used for bitfield access would be always aligned according to >>>>>>>>>> verifier expectations. >>>>>>>>>> >>>>>>>>>> I think we should check if there are some clang knobs that prevent >>>>>>>>>> generation of unaligned memory access. I'll take a look. >>>>>>>>> Is there a reason to prefer fixing in compiler? I'm not opposed to it, >>>>>>>>> but the downside to compiler fix is it takes years to propagate and >>>>>>>>> sprinkles ifdefs into the code. >>>>>>>>> >>>>>>>>> Would it be possible to have an analogue of BPF_CORE_READ_BITFIELD()? >>>>>>>> Well, the contraption below passes verification, tunnel selftest >>>>>>>> appears to work. I might have messed up some shifts in the macro, >>>>>>>> though. >>>>>>> I didn't test it. But from high level it should work. >>>>>>> >>>>>>>> Still, if clang would peek unlucky BYTE_{OFFSET,SIZE} for a particular >>>>>>>> field access might be unaligned. >>>>>>> clang should pick a sensible BYTE_SIZE/BYTE_OFFSET to meet >>>>>>> alignment requirement. This is also required for BPF_CORE_READ_BITFIELD. >>>>>>> >>>>>>>> --- >>>>>>>> >>>>>>>> diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c >>>>>>>> b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c >>>>>>>> index 3065a716544d..41cd913ac7ff 100644 >>>>>>>> --- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c >>>>>>>> +++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c >>>>>>>> @@ -9,6 +9,7 @@ >>>>>>>> #include "vmlinux.h" >>>>>>>> #include <bpf/bpf_helpers.h> >>>>>>>> #include <bpf/bpf_endian.h> >>>>>>>> +#include <bpf/bpf_core_read.h> >>>>>>>> #include "bpf_kfuncs.h" >>>>>>>> #include "bpf_tracing_net.h" >>>>>>>> @@ -144,6 +145,38 @@ int ip6gretap_get_tunnel(struct __sk_buff *skb) >>>>>>>> return TC_ACT_OK; >>>>>>>> } >>>>>>>> +#define BPF_CORE_WRITE_BITFIELD(s, field, new_val) ({ \ >>>>>>>> + void *p = (void *)s + __CORE_RELO(s, field, BYTE_OFFSET); \ >>>>>>>> + unsigned byte_size = __CORE_RELO(s, field, BYTE_SIZE); \ >>>>>>>> + unsigned lshift = __CORE_RELO(s, field, LSHIFT_U64); \ >>>>>>>> + unsigned rshift = __CORE_RELO(s, field, RSHIFT_U64); \ >>>>>>>> + unsigned bit_size = (rshift - lshift); \ >>>>>>>> + unsigned long long nval, val, hi, lo; \ >>>>>>>> + \ >>>>>>>> + asm volatile("" : "=r"(p) : "0"(p)); \ >>>>>>> Use asm volatile("" : "+r"(p)) ? >>>>>>> >>>>>>>> + \ >>>>>>>> + switch (byte_size) { \ >>>>>>>> + case 1: val = *(unsigned char *)p; break; \ >>>>>>>> + case 2: val = *(unsigned short *)p; break; \ >>>>>>>> + case 4: val = *(unsigned int *)p; break; \ >>>>>>>> + case 8: val = *(unsigned long long *)p; break; \ >>>>>>>> + } \ >>>>>>>> + hi = val >> (bit_size + rshift); \ >>>>>>>> + hi <<= bit_size + rshift; \ >>>>>>>> + lo = val << (bit_size + lshift); \ >>>>>>>> + lo >>= bit_size + lshift; \ >>>>>>>> + nval = new_val; \ >>>>>>>> + nval <<= lshift; \ >>>>>>>> + nval >>= rshift; \ >>>>>>>> + val = hi | nval | lo; \ >>>>>>>> + switch (byte_size) { \ >>>>>>>> + case 1: *(unsigned char *)p = val; break; \ >>>>>>>> + case 2: *(unsigned short *)p = val; break; \ >>>>>>>> + case 4: *(unsigned int *)p = val; break; \ >>>>>>>> + case 8: *(unsigned long long *)p = val; break; \ >>>>>>>> + } \ >>>>>>>> +}) >>>>>>> I think this should be put in libbpf public header files but not sure >>>>>>> where to put it. bpf_core_read.h although it is core write? >>>>>>> >>>>>>> But on the other hand, this is a uapi struct bitfield write, >>>>>>> strictly speaking, CORE write is really unnecessary here. It >>>>>>> would be great if we can relieve users from dealing with >>>>>>> such unnecessary CORE writes. In that sense, for this particular >>>>>>> case, I would prefer rewriting the code by using byte-level >>>>>>> stores... >>>>>> or preserve_static_offset to clearly mean to undo bitfield CORE ... >>>>> Ok, I will do byte-level rewrite for next revision. >>>> [...] >>>> >>>> This patch seems to work: https://pastes.dxuuu.xyz/0glrf9 . >>>> >>>> But I don't think it's very pretty. Also I'm seeing on the internet that >>>> people are saying the exact layout of bitfields is compiler dependent. >>> Any reference for this (exact layout of bitfields is compiler dependent)? >>> >>>> So I am wondering if these byte sized writes are correct. For that >>>> matter, I am wondering how the GCC generated bitfield accesses line up >>>> with clang generated BPF bytecode. Or why uapi contains a bitfield. >>> One thing for sure is memory layout of bitfields should be the same >>> for both clang and gcc as it is determined by C standard. Register >>> representation and how to manipulate could be different for different >>> compilers. >> I was reading this thread: >> https://github.com/Lora-net/LoRaMac-node/issues/697. It's obviously not >> authoritative, but they sure sound confident! >> >> I think I've also heard it before a long time ago when I was working on >> adding bitfield support to bpftrace. > Wikipedia [0] also claims this: > > The layout of bit fields in a C struct is > implementation-defined. For behavior that remains predictable > across compilers, it may be preferable to emulate bit fields > with a primitive and bit operators: > > [0]: https://en.wikipedia.org/wiki/Bit_field#C_programming_language Thanks for the informaiton. I am truely not aware of bit field layout could be different for different compilers. Does this mean source level bitfield manipulation may not work? uapi has bitfield is okay. compiler should do the right thing to do load/store in bitfields. Also, the networking bitfields are related memory layout transferring on the wire. Its memory layout is determined (although little/big endian interpresentation is different). BPF_CORE_WRITE_BITFIELD 'should' also be okay since the offset/size etc. is gotten from the compiler internals (from dwarf in more precise term). So looks like BPF_CORE_WRITE_BITFIELD is the way to go. Please use it then.
diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c index 3065a716544d..ec7e04e012ae 100644 --- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c +++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c @@ -6,6 +6,7 @@ * modify it under the terms of version 2 of the GNU General Public * License as published by the Free Software Foundation. */ +#define BPF_NO_PRESERVE_ACCESS_INDEX #include "vmlinux.h" #include <bpf/bpf_helpers.h> #include <bpf/bpf_endian.h>
Switching to vmlinux.h definitions seems to make the verifier very unhappy with bitfield accesses. The error is: ; md.u.md2.dir = direction; 33: (69) r1 = *(u16 *)(r2 +11) misaligned stack access off (0x0; 0x0)+-64+11 size 2 It looks like disabling CO-RE relocations seem to make the error go away. Co-developed-by: Antony Antony <antony.antony@secunet.com> Signed-off-by: Antony Antony <antony.antony@secunet.com> Signed-off-by: Daniel Xu <dxu@dxuuu.xyz> --- tools/testing/selftests/bpf/progs/test_tunnel_kern.c | 1 + 1 file changed, 1 insertion(+)