Date:   Thu, 3 Feb 2022 11:07:30 -0500
From:   Mauricio Vásquez Bernal <mauricio@...volk.io>
To:     Andrii Nakryiko <andrii.nakryiko@...il.com>
Cc:     Networking <netdev@...r.kernel.org>, bpf <bpf@...r.kernel.org>,
        Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Andrii Nakryiko <andrii@...nel.org>,
        Quentin Monnet <quentin@...valent.com>,
        Rafael David Tinoco <rafaeldtinoco@...il.com>,
        Lorenzo Fontana <lorenzo.fontana@...stic.co>,
        Leonardo Di Donato <leonardo.didonato@...stic.co>
Subject: Re: [PATCH bpf-next v5 2/9] bpftool: Add gen min_core_btf command

On Wed, Feb 2, 2022 at 12:58 PM Andrii Nakryiko
<andrii.nakryiko@...il.com> wrote:
>
> On Fri, Jan 28, 2022 at 2:33 PM Mauricio Vásquez <mauricio@...volk.io> wrote:
> >
> > This command is implemented under the "gen" command in bpftool and the
> > syntax is the following:
> >
> > $ bpftool gen min_core_btf INPUT OUTPUT OBJECT(S)
> >
> > INPUT can be either a single BTF file or a folder containing BTF files,
> > when it's a folder, a BTF file is generated for each BTF file contained
> > in this folder. OUTPUT is the file (or folder) where generated files are
> > stored and OBJECT(S) is the list of bpf objects we want to generate the
> > BTF file(s) for (each generated BTF file contains all the types needed
> > by all the objects).
> >
> > Signed-off-by: Mauricio Vásquez <mauricio@...volk.io>
> > Signed-off-by: Rafael David Tinoco <rafael.tinoco@...asec.com>
> > Signed-off-by: Lorenzo Fontana <lorenzo.fontana@...stic.co>
> > Signed-off-by: Leonardo Di Donato <leonardo.didonato@...stic.co>
> > ---
> >  tools/bpf/bpftool/bash-completion/bpftool |   6 +-
> >  tools/bpf/bpftool/gen.c                   | 112 +++++++++++++++++++++-
> >  2 files changed, 114 insertions(+), 4 deletions(-)
> >
> > diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
> > index 493753a4962e..958e1fd71b5c 100644
> > --- a/tools/bpf/bpftool/bash-completion/bpftool
> > +++ b/tools/bpf/bpftool/bash-completion/bpftool
> > @@ -1003,9 +1003,13 @@ _bpftool()
> >                              ;;
> >                      esac
> >                      ;;
> > +                min_core_btf)
> > +                    _filedir
> > +                    return 0
> > +                    ;;
> >                  *)
> >                      [[ $prev == $object ]] && \
> > -                        COMPREPLY=( $( compgen -W 'object skeleton help' -- "$cur" ) )
> > +                        COMPREPLY=( $( compgen -W 'object skeleton help min_core_btf' -- "$cur" ) )
> >                      ;;
> >              esac
> >              ;;
> > diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
> > index 8f78c27d41f0..7db31b0f265f 100644
> > --- a/tools/bpf/bpftool/gen.c
> > +++ b/tools/bpf/bpftool/gen.c
> > @@ -5,6 +5,7 @@
> >  #define _GNU_SOURCE
> >  #endif
> >  #include <ctype.h>
> > +#include <dirent.h>
> >  #include <errno.h>
> >  #include <fcntl.h>
> >  #include <linux/err.h>
> > @@ -1084,6 +1085,7 @@ static int do_help(int argc, char **argv)
> >         fprintf(stderr,
> >                 "Usage: %1$s %2$s object OUTPUT_FILE INPUT_FILE [INPUT_FILE...]\n"
> >                 "       %1$s %2$s skeleton FILE [name OBJECT_NAME]\n"
> > +               "       %1$s %2$s min_core_btf INPUT OUTPUT OBJECT(S)\n"
>
> OBJECTS(S) should be OBJECT... for this "CLI notation", no?

Updated it to be "min_core_btf INPUT OUTPUT OBJECT [OBJECT...]" like
the "bpftool object" command.

>
> >                 "       %1$s %2$s help\n"
> >                 "\n"
> >                 "       " HELP_SPEC_OPTIONS " |\n"
> > @@ -1094,10 +1096,114 @@ static int do_help(int argc, char **argv)
> >         return 0;
> >  }
> >
> > +/* Create BTF file for a set of BPF objects */
> > +static int btfgen(const char *src_btf, const char *dst_btf, const char *objspaths[])
> > +{
> > +       return -EOPNOTSUPP;
> > +}
> > +
> > +static int do_min_core_btf(int argc, char **argv)
> > +{
> > +       char src_btf_path[PATH_MAX], dst_btf_path[PATH_MAX];
> > +       bool input_is_file, output_is_file = true;
> > +       const char *input, *output;
> > +       const char **objs = NULL;
> > +       struct dirent *dir;
> > +       struct stat st;
> > +       DIR *d = NULL;
> > +       int i, err;
> > +
> > +       if (!REQ_ARGS(3)) {
> > +               usage();
> > +               return -1;
> > +       }
> > +
> > +       input = GET_ARG();
> > +       if (stat(input, &st) < 0) {
> > +               p_err("failed to stat %s: %s", input, strerror(errno));
> > +               return -errno;
> > +       }
> > +
> > +       if ((st.st_mode & S_IFMT) != S_IFDIR && (st.st_mode & S_IFMT) != S_IFREG) {
> > +               p_err("file type not valid: %s", input);
> > +               return -EINVAL;
> > +       }
> > +
> > +       input_is_file = (st.st_mode & S_IFMT) == S_IFREG;
>
> move before if and use input_is_file in the if itself instead of
> duplicating all the S_IFREG flags?
>
> > +
> > +       output = GET_ARG();
> > +       if (stat(output, &st) == 0 && (st.st_mode & S_IFMT) == S_IFDIR)
> > +               output_is_file = false;
>
> if stat() succeeds but it's neither directory or file, should be an
> error, right?
>
> > +
> > +       objs = (const char **) malloc((argc + 1) * sizeof(*objs));
>
> calloc() seems to be better suited for this (and zero-intialization is
> nice for safety and to avoid objs[argc] = NULL after the loop below)
>

You're right!

> > +       if (!objs) {
> > +               p_err("failed to allocate array for object names");
> > +               return -ENOMEM;
> > +       }
> > +
> > +       i = 0;
> > +       while (argc > 0)
> > +               objs[i++] = GET_ARG();
>
> for (i = 0; i < argc; i++) ?

GET_ARG() does argc--. I see this loop is usually written as while
(argc) in bpftool.

>
> > +
> > +       objs[i] = NULL;
> > +
> > +       /* single BTF file */
> > +       if (input_is_file) {
> > +               p_info("Processing source BTF file: %s", input);
> > +
> > +               if (output_is_file) {
> > +                       err = btfgen(input, output, objs);
> > +                       goto out;
> > +               }
> > +               snprintf(dst_btf_path, sizeof(dst_btf_path), "%s/%s", output,
> > +                        basename(input));
> > +               err = btfgen(input, dst_btf_path, objs);
> > +               goto out;
> > +       }
> > +
> > +       if (output_is_file) {
> > +               p_err("can't have just one file as output");
> > +               err = -EINVAL;
> > +               goto out;
> > +       }
> > +
> > +       /* directory with BTF files */
> > +       d = opendir(input);
> > +       if (!d) {
> > +               p_err("error opening input dir: %s", strerror(errno));
> > +               err = -errno;
> > +               goto out;
> > +       }
> > +
> > +       while ((dir = readdir(d)) != NULL) {
> > +               if (dir->d_type != DT_REG)
> > +                       continue;
> > +
> > +               if (strncmp(dir->d_name + strlen(dir->d_name) - 4, ".btf", 4))
> > +                       continue;
>
> this whole handling of input directory feels a bit icky, tbh... maybe
> we should require explicit listing of input files always. In CLI
> invocation those could be separated by "keywords", something like
> this:
>
> bpftool gen min_core_btf <output> inputs <file1> <file2> .... objects
> <obj1> <obj2> ...
>
> a bit of a downside is that you can't have a file named "inputs" or
> "objects", but that seems extremely unlikely? Quentin, any opinion as
> well?
>
> I'm mainly off put by a bit random ".btf" naming convention, the
> DT_REG skipping, etc.
>
> Another cleaner alternative from POV of bpftool (but might be less
> convenient for users) is to use @file convention to specify a file
> that contains a list of files. So
>
> bpftool gen min_core_btf <output> @btf_filelist.txt @obj_filelist.txt
>
> would take lists of inputs and outputs from respective files?
>
>
> But actually, let's take a step back again. Why should there be
> multiple inputs and outputs?

We're thinking about the use case when there are multiple source BTF
files in a folder and we want to generate a BTF for each one of them,
by supporting input and output folders we're able to avoid executing
bpftool multiple times. I agree that it complicates the implementation
and that the same can be done by using a script to run bpftool
multiple times, hence let's remove it. If we find out later on that
this is really important we can implement it.


> I can see why multiple objects are
> mandatory (you have an application that has multiple BPF objects used
> internally). But processing single vmlinux BTF at a time seems
> absolutely fine. I don't buy that CO-RE relo processing is that slow
> to require optimized batch processing.
>
> I might have asked this before, sorry, but the duration between each
> iteration of btfgen is pretty long and I'm losing the context.
>
> > +
> > +               snprintf(src_btf_path, sizeof(src_btf_path), "%s%s", input, dir->d_name);
> > +               snprintf(dst_btf_path, sizeof(dst_btf_path), "%s%s", output, dir->d_name);
> > +
> > +               p_info("Processing source BTF file: %s", src_btf_path);
> > +
> > +               err = btfgen(src_btf_path, dst_btf_path, objs);
> > +               if (err)
> > +                       goto out;
> > +       }
> > +
> > +out:
> > +       free(objs);
> > +       if (d)
> > +               closedir(d);
> > +       return err;
> > +}
> > +
> >  static const struct cmd cmds[] = {
> > -       { "object",     do_object },
> > -       { "skeleton",   do_skeleton },
> > -       { "help",       do_help },
> > +       { "object",             do_object },
> > +       { "skeleton",           do_skeleton },
> > +       { "min_core_btf",       do_min_core_btf},
> > +       { "help",               do_help },
> >         { 0 }
> >  };
> >
> > --
> > 2.25.1
> >

Powered by blists - more mailing lists