From: John Fastabend <john.fastabend@gmail.com>
To: Quentin Monnet <quentin@isovalent.com>,
John Fastabend <john.fastabend@gmail.com>,
Lam Thai <lamthai@arista.com>,
bpf@vger.kernel.org
Subject: Re: [PATCH] bpftool: fix a wrong type cast in btf_dumper_int
Date: Thu, 25 Aug 2022 08:54:07 -0700 [thread overview]
Message-ID: <63079b1f19592_12460b208f7@john.notmuch> (raw)
In-Reply-To: <e510b3f8-ed6c-55c7-3585-2e065324ae85@isovalent.com>
Quentin Monnet wrote:
> On 25/08/2022 07:29, John Fastabend wrote:
> > Lam Thai wrote:
> >> When `data` points to a boolean value, casting it to `int *` is problematic
> >> and could lead to a wrong value being passed to `jsonw_bool`. Change the
> >> cast to `bool *` instead.
> >
> > How is it problematic? Its from BTF_KIND_INT by my quick reading.
>
> Hi John, it's an INT but it also has a size of 1:
>
> struct map_value {
> int a;
> int b;
> short c;
> bool d;
> };
>
> # bpftool btf dump id 1107
> [...]
> [2] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED
> [...]
> [12] STRUCT 'map_value' size=12 vlen=4
> 'a' type_id=2 bits_offset=0
> 'b' type_id=2 bits_offset=32
> 'c' type_id=13 bits_offset=64
> 'd' type_id=14 bits_offset=80
> [13] INT 'short' size=2 bits_offset=0 nr_bits=16 encoding=SIGNED
> [14] INT '_Bool' size=1 bits_offset=0 nr_bits=8 encoding=BOOL
> [...]
>
> And Lam reported [0] that the pretty-print for the map does not display
> the correct boolean value, because it reads too many bytes from this
> *(int *)data.
>
> # bpftool map dump name my_map --pretty
> [{
> "key": ["0x00","0x00","0x00","0x00"
> ],
> "value":
> ["0x00","0x00","0x00","0x00","0x00","0x00","0x00","0x00","0x00","0x00","0x00","0x00"
> ],
> "formatted": {
> "key": 0,
> "value": {
> "a": 0,
> "b": 0,
> "c": 0,
> "d": true
> }
> }
> }
> ]
>
> The above is before the map gets any update. The bytes in "value" look
> correct, but "d" says "true" when it should be "false". So bpf tree
> would make sense to me.
>
> [0] https://github.com/libbpf/bpftool/issues/38
Thanks for the explanation. It would be nice to add the above in the
commit message.
Otherwise though.
Acked-by: John Fastabend <john.fastabend@gmail.com>
>
> >
> >>
> >> Fixes: b12d6ec09730 ("bpf: btf: add btf print functionality")
> >> Signed-off-by: Lam Thai <lamthai@arista.com>
> >> ---
> >
> > for bpf-next looks like a nice cleanup, I don't think its needed for bpf
> > tree?
> >
> >> tools/bpf/bpftool/btf_dumper.c | 2 +-
> >> 1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/tools/bpf/bpftool/btf_dumper.c b/tools/bpf/bpftool/btf_dumper.c
> >> index 125798b0bc5d..19924b6ce796 100644
> >> --- a/tools/bpf/bpftool/btf_dumper.c
> >> +++ b/tools/bpf/bpftool/btf_dumper.c
> >> @@ -452,7 +452,7 @@ static int btf_dumper_int(const struct btf_type *t, __u8 bit_offset,
> >> *(char *)data);
> >> break;
> >> case BTF_INT_BOOL:
> >> - jsonw_bool(jw, *(int *)data);
> >> + jsonw_bool(jw, *(bool *)data);
>
> Looks good, thanks
> Reviewed-by: Quentin Monnet <quentin@isovalent.com>
next prev parent reply other threads:[~2022-08-25 15:54 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-24 22:59 [PATCH] bpftool: fix a wrong type cast in btf_dumper_int Lam Thai
2022-08-25 6:29 ` John Fastabend
2022-08-25 8:56 ` Quentin Monnet
2022-08-25 15:54 ` John Fastabend [this message]
2022-08-25 18:44 ` Andrii Nakryiko
2022-08-25 18:50 ` patchwork-bot+netdevbpf
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=63079b1f19592_12460b208f7@john.notmuch \
--to=john.fastabend@gmail.com \
--cc=bpf@vger.kernel.org \
--cc=lamthai@arista.com \
--cc=quentin@isovalent.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox