SchOS/ldls/host/inspect.c

// SPDX-License-Identifier: Apache-2.0
//
// ldls-inspect -- what is in a relocatable object, and what it wants.
//
// docs/plan/ldls.md milestone L0, run on the machine doing the building rather
// than inside SchOS. That is the same arrangement `mkschfs` has with the
// kernel, for the same reason: the interesting failures here are format
// failures, and a format failure found on the host costs a second instead of a
// boot.
//
// It loads and relocates for real -- into a buffer at a pretended address --
// so that "it parses" and "it can actually be laid out and relocated" are not
// confused with each other. Nothing is executed.
//
//     ldls-inspect 8139too.ko
//
// **No module is committed to this tree**, and none ever will be; see
// docs/plan/ldls.md rule 1. `make ldls-check` runs this against an object
// built from ldls/test/refmod.c, which is ours.

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#include "../ko.h"

static unsigned char *
slurp(const char *path, unsigned long long *len)
{
  FILE *f = fopen(path, "rb");
  unsigned char *buf;
  long n;

  if (!f)
    return 0;
  if (fseek(f, 0, SEEK_END) != 0) { fclose(f); return 0; }
  n = ftell(f);
  if (n <= 0) { fclose(f); return 0; }
  rewind(f);

  buf = malloc((size_t)n);
  if (!buf) { fclose(f); return 0; }
  if (fread(buf, 1, (size_t)n, f) != (size_t)n) {
    free(buf); fclose(f); return 0;
  }
  fclose(f);
  *len = (unsigned long long)n;
  return buf;
}

// At L0 nothing is known, so the resolver knows nothing -- and that is the
// point rather than a placeholder. Every name it refuses becomes a line in the
// work queue.
static int
resolve_nothing(void *ctx, const char *name, unsigned long long *out)
{
  (void)ctx; (void)name; (void)out;
  return -1;
}

// Resolve every name to one plausible address, so relocation runs to the end.
//
// This answers a different question from the list of undefined symbols, and it
// is the one that decides whether the loader is finished: does this module
// contain a relocation type the loader does not handle? With nothing resolved
// that can never be reached, because the first unknown name stops everything.
//
// The address is near the module rather than anywhere, because PC32 and PLT32
// have to reach it -- a fake symbol two gigabytes away would fail the range
// check and say something true about the fake rather than about the loader.
static int
resolve_to_stub(void *ctx, const char *name, unsigned long long *out)
{
  (void)name;
  *out = *(unsigned long long *)ctx;
  return 0;
}

static const char *
flags_of(const struct ko_section *s)
{
  static char b[8];
  int n = 0;
  b[n++] = (s->flags & KO_SHF_ALLOC)     ? 'a' : '-';
  b[n++] = (s->flags & KO_SHF_WRITE)     ? 'w' : '-';
  b[n++] = (s->flags & KO_SHF_EXECINSTR) ? 'x' : '-';
  b[n]   = 0;
  return b;
}

int
main(int argc, char **argv)
{
  struct ko k;
  unsigned char *img, *mem;
  unsigned long long len, span;
  const char *lic;
  int i, quiet = 0, stub = 0, nsec;
  struct ko_section *sec;

  if (argc < 2) {
    fprintf(stderr, "usage: ldls-inspect [-q] [-r] <object.ko>\n");
    return 2;
  }
  if (strcmp(argv[1], "-q") == 0) { quiet = 1; argv++; argc--; }
  if (argc > 1 && strcmp(argv[1], "-r") == 0) { stub = 1; argv++; argc--; }
  if (argc < 2) { fprintf(stderr, "usage: ldls-inspect [-q] <object>\n"); return 2; }

  img = slurp(argv[1], &len);
  if (!img) {
    fprintf(stderr, "ldls-inspect: cannot read %s\n", argv[1]);
    return 1;
  }

  // Exactly what this object needs, so there is no limit to run into.
  nsec = ko_sections_needed(img, len);
  if (nsec < 0) {
    fprintf(stderr, "ldls-inspect: %s: not an ELF64 relocatable object\n",
            argv[1]);
    return 1;
  }
  sec = calloc((size_t)(nsec ? nsec : 1), sizeof(struct ko_section));
  if (!sec) { fprintf(stderr, "ldls-inspect: out of memory\n"); return 1; }

  if (ko_open(&k, img, len, sec, nsec) < 0) {
    fprintf(stderr, "ldls-inspect: %s: %s\n", argv[1], ko_err(&k));
    return 1;
  }

  printf("%s\n", argv[1]);
  printf("  %llu bytes, %d sections, %d symbols\n",
         (unsigned long long)len, k.nsh, k.nsym);

  // The licence first, and deliberately so. It is the one property that
  // decides whether this object may be on this machine at all, it is a string
  // in the file, and nothing has to be loaded to read it.
  lic = ko_modinfo(&k, "license");
  printf("  module    %s\n", ko_name(&k) ? ko_name(&k) : "(not a kernel module)");
  printf("  license   %s\n", lic ? lic : "(none declared)");
  if (ko_modinfo(&k, "description"))
    printf("  describes %s\n", ko_modinfo(&k, "description"));

  span = ko_span(&k);
  printf("  loads to  %llu bytes\n", span);

  if (!quiet) {
    printf("\n  sections that take space at run time\n");
    for (i = 0; i < k.nsh; i++) {
      const struct ko_section *s = &k.sec[i];
      if (!s->alloc || s->size == 0)
        continue;
      printf("    %-24s %s %8llu  align %llu%s\n",
             s->name, flags_of(s), s->size, s->align,
             s->type == KO_SHT_NOBITS ? "  (zeroed)" : "");
    }
  }

  // Lay it out and relocate at a plausible address, so that the report covers
  // the parts that only fail once something has an address.
  mem = malloc(span ? (size_t)span : 1);
  if (!mem) { fprintf(stderr, "ldls-inspect: out of memory\n"); return 1; }

  if (ko_layout(&k, mem, span, 0x40000000ULL) < 0) {
    fprintf(stderr, "ldls-inspect: layout: %s\n", ko_err(&k));
    return 1;
  }

  // What the loader is being asked to handle, exactly.
  ko_reltypes(&k);
  printf("\n  relocations: %u\n", k.nrel);
  {
    static const struct { unsigned int t; const char *n; } known[] = {
      { KO_R_X86_64_64,    "R_X86_64_64"    },
      { KO_R_X86_64_PC32,  "R_X86_64_PC32"  },
      { KO_R_X86_64_PLT32, "R_X86_64_PLT32" },
      { KO_R_X86_64_32,    "R_X86_64_32"    },
      { KO_R_X86_64_32S,   "R_X86_64_32S"   },
      { KO_R_X86_64_PC64,  "R_X86_64_PC64"  },
      { KO_R_X86_64_NONE,  "R_X86_64_NONE"  },
    };
    unsigned int accounted = 0;
    for (i = 0; i < (int)(sizeof(known) / sizeof(known[0])); i++) {
      if (!k.reltype[known[i].t])
        continue;
      printf("    %-16s %u\n", known[i].n, k.reltype[known[i].t]);
      accounted += k.reltype[known[i].t];
    }
    for (i = 0; i < 64; i++) {
      unsigned int t = (unsigned int)i;
      if (!k.reltype[i] || t == KO_R_X86_64_64 || t == KO_R_X86_64_PC32 ||
          t == KO_R_X86_64_PLT32 || t == KO_R_X86_64_32 ||
          t == KO_R_X86_64_32S || t == KO_R_X86_64_PC64 ||
          t == KO_R_X86_64_NONE)
        continue;
      printf("    type %-11d %u   <-- the loader does not handle this\n",
             i, k.reltype[i]);
    }
    if (k.reltype_other)
      printf("    type >= 64       %u   <-- the loader does not handle this\n",
             k.reltype_other);
    (void)accounted;
  }

  printf("\n  undefined symbols: %d%s\n", k.nundef,
         k.undef_lost ? " (and more than the loader will hold)" : "");
  for (i = 0; i < k.nundef; i++)
    printf("    %s\n", k.undef[i]);

  // With a resolver that knows nothing this must fail, and the first name it
  // names is the first thing L1 has to provide. Reaching the end instead means
  // the object needed nothing at all, which a real module never does.
  if (stub) {
    unsigned long long at = 0x40000000ULL + 0x200000ULL;
    if (ko_relocate(&k, resolve_to_stub, &at) == 0)
      printf("\n  relocated in full against stubs -- every relocation type in\n"
             "  this object is one the loader handles\n");
    else
      printf("\n  relocation failed: %s\n", ko_err(&k));
  } else if (ko_relocate(&k, resolve_nothing, 0) == 0) {
    printf("\n  relocated with an empty symbol table -- it needs nothing\n");
  } else {
    printf("\n  stops at: %s\n", ko_err(&k));
    printf("  (try -r to resolve everything to a stub and reach the end)\n");
  }

  free(mem);
  free(sec);
  free(img);
  return 0;
}