sysdeps/unix/sysv/linux/getdents.c

   1 /* Copyright (C) 1993-2017 Free Software Foundation, Inc.
   2    This file is part of the GNU C Library.
   3
   4    The GNU C Library is free software; you can redistribute it and/or
   5    modify it under the terms of the GNU Lesser General Public
   6    License as published by the Free Software Foundation; either
   7    version 2.1 of the License, or (at your option) any later version.
   8
   9    The GNU C Library is distributed in the hope that it will be useful,
  10    but WITHOUT ANY WARRANTY; without even the implied warranty of
  11    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  12    Lesser General Public License for more details.
  13
  14    You should have received a copy of the GNU Lesser General Public
  15    License along with the GNU C Library; if not, see
  16    <http://www.gnu.org/licenses/>.  */
  17
  18 #include <alloca.h>
  19 #include <assert.h>
  20 #include <errno.h>
  21 #include <dirent.h>
  22 #include <stddef.h>
  23 #include <stdint.h>
  24 #include <string.h>
  25 #include <unistd.h>
  26 #include <sys/param.h>
  27 #include <sys/types.h>
  28
  29 #include <sysdep.h>
  30 #include <sys/syscall.h>
  31
  32 #include <linux/posix_types.h>
  33
  34 #include <kernel-features.h>
  35
  36 /* For Linux we need a special version of this file since the
  37    definition of `struct dirent' is not the same for the kernel and
  38    the libc.  There is one additional field which might be introduced
  39    in the kernel structure in the future.
  40
  41    Here is the kernel definition of `struct dirent' as of 2.1.20:  */
  42
  43 struct kernel_dirent
  44   {
  45     long int d_ino;
  46     __kernel_off_t d_off;
  47     unsigned short int d_reclen;
  48     char d_name[256];
  49   };
  50
  51 struct kernel_dirent64
  52   {
  53     uint64_t            d_ino;
  54     int64_t             d_off;
  55     unsigned short int  d_reclen;
  56     unsigned char       d_type;
  57     char                d_name[256];
  58   };
  59
  60 #ifndef __GETDENTS
  61 # define __GETDENTS __getdents
  62 #endif
  63 #ifndef DIRENT_TYPE
  64 # define DIRENT_TYPE struct dirent
  65 #endif
  66 #ifndef DIRENT_SET_DP_INO
  67 # define DIRENT_SET_DP_INO(dp, value) (dp)->d_ino = (value)
  68 #endif
  69
  70 /* The problem here is that we cannot simply read the next NBYTES
  71    bytes.  We need to take the additional field into account.  We use
  72    some heuristic.  Assuming the directory contains names with 14
  73    characters on average we can compute an estimated number of entries
  74    which fit in the buffer.  Taking this number allows us to specify a
  75    reasonable number of bytes to read.  If we should be wrong, we can
  76    reset the file descriptor.  In practice the kernel is limiting the
  77    amount of data returned much more then the reduced buffer size.  */
  78 ssize_t
  79 internal_function
  80 __GETDENTS (int fd, char *buf, size_t nbytes)
  81 {
  82   ssize_t retval;
  83
  84   /* The d_ino and d_off fields in kernel_dirent and dirent must have
  85      the same sizes and alignments.  */
  86   if (sizeof (DIRENT_TYPE) == sizeof (struct dirent)
  87       && (sizeof (((struct kernel_dirent *) 0)->d_ino)
  88           == sizeof (((struct dirent *) 0)->d_ino))
  89       && (sizeof (((struct kernel_dirent *) 0)->d_off)
  90           == sizeof (((struct dirent *) 0)->d_off))
  91       && (offsetof (struct kernel_dirent, d_off)
  92           == offsetof (struct dirent, d_off))
  93       && (offsetof (struct kernel_dirent, d_reclen)
  94           == offsetof (struct dirent, d_reclen)))
  95     {
  96       retval = INLINE_SYSCALL (getdents, 3, fd, buf, nbytes);
  97
  98       /* The kernel added the d_type value after the name.  Change
  99          this now.  */
 100       if (retval != -1)
 101         {
 102           union
 103           {
 104             struct kernel_dirent k;
 105             struct dirent u;
 106           } *kbuf = (void *) buf;
 107
 108           while ((char *) kbuf < buf + retval)
 109             {
 110               char d_type = *((char *) kbuf + kbuf->k.d_reclen - 1);
 111               memmove (kbuf->u.d_name, kbuf->k.d_name,
 112                        strlen (kbuf->k.d_name) + 1);
 113               kbuf->u.d_type = d_type;
 114
 115               kbuf = (void *) ((char *) kbuf + kbuf->k.d_reclen);
 116             }
 117         }
 118
 119       return retval;
 120     }
 121
 122   off64_t last_offset = -1;
 123
 124 #ifdef __NR_getdents64
 125   {
 126     union
 127     {
 128       struct kernel_dirent64 k;
 129       DIRENT_TYPE u;
 130       char b[1];
 131     } *kbuf = (void *) buf, *outp, *inp;
 132     size_t kbytes = nbytes;
 133     if (offsetof (DIRENT_TYPE, d_name)
 134         < offsetof (struct kernel_dirent64, d_name)
 135         && nbytes <= sizeof (DIRENT_TYPE))
 136       {
 137         kbytes = (nbytes + offsetof (struct kernel_dirent64, d_name)
 138                   - offsetof (DIRENT_TYPE, d_name));
 139         kbuf = __alloca(kbytes);
 140       }
 141     retval = INLINE_SYSCALL (getdents64, 3, fd, kbuf, kbytes);
 142     const size_t size_diff = (offsetof (struct kernel_dirent64, d_name)
 143                               - offsetof (DIRENT_TYPE, d_name));
 144
 145     /* Return the error if encountered.  */
 146     if (retval == -1)
 147       return -1;
 148
 149     /* If the structure returned by the kernel is identical to what we
 150        need, don't do any conversions.  */
 151     if (offsetof (DIRENT_TYPE, d_name)
 152         == offsetof (struct kernel_dirent64, d_name)
 153         && sizeof (outp->u.d_ino) == sizeof (inp->k.d_ino)
 154         && sizeof (outp->u.d_off) == sizeof (inp->k.d_off))
 155       return retval;
 156
 157     /* These two pointers might alias the same memory buffer.
 158        Standard C requires that we always use the same type for them,
 159        so we must use the union type.  */
 160     inp = kbuf;
 161     outp = (void *) buf;
 162
 163     while (&inp->b < &kbuf->b + retval)
 164       {
 165         const size_t alignment = __alignof__ (DIRENT_TYPE);
 166         /* Since inp->k.d_reclen is already aligned for the kernel
 167            structure this may compute a value that is bigger
 168            than necessary.  */
 169         size_t old_reclen = inp->k.d_reclen;
 170         size_t new_reclen = ((old_reclen - size_diff + alignment - 1)
 171                              & ~(alignment - 1));
 172
 173         /* Copy the data out of the old structure into temporary space.
 174            Then copy the name, which may overlap if BUF == KBUF.  */
 175         const uint64_t d_ino = inp->k.d_ino;
 176         const int64_t d_off = inp->k.d_off;
 177         const uint8_t d_type = inp->k.d_type;
 178
 179         memmove (outp->u.d_name, inp->k.d_name,
 180                  old_reclen - offsetof (struct kernel_dirent64, d_name));
 181
 182         /* Now we have copied the data from INP and access only OUTP.  */
 183
 184         DIRENT_SET_DP_INO (&outp->u, d_ino);
 185         outp->u.d_off = d_off;
 186         if ((sizeof (outp->u.d_ino) != sizeof (inp->k.d_ino)
 187              && outp->u.d_ino != d_ino)
 188             || (sizeof (outp->u.d_off) != sizeof (inp->k.d_off)
 189                 && outp->u.d_off != d_off))
 190           {
 191             /* Overflow.  If there was at least one entry
 192                before this one, return them without error,
 193                otherwise signal overflow.  */
 194             if (last_offset != -1)
 195               {
 196                 __lseek64 (fd, last_offset, SEEK_SET);
 197                 return outp->b - buf;
 198               }
 199             __set_errno (EOVERFLOW);
 200             return -1;
 201           }
 202
 203         last_offset = d_off;
 204         outp->u.d_reclen = new_reclen;
 205         outp->u.d_type = d_type;
 206
 207         inp = (void *) inp + old_reclen;
 208         outp = (void *) outp + new_reclen;
 209       }
 210
 211     return outp->b - buf;
 212   }
 213 #endif
 214   {
 215     size_t red_nbytes;
 216     struct kernel_dirent *skdp, *kdp;
 217     const size_t size_diff = (offsetof (DIRENT_TYPE, d_name)
 218                               - offsetof (struct kernel_dirent, d_name));
 219
 220     red_nbytes = MIN (nbytes
 221                       - ((nbytes / (offsetof (DIRENT_TYPE, d_name) + 14))
 222                          * size_diff),
 223                       nbytes - size_diff);
 224
 225     skdp = kdp = __alloca (red_nbytes);
 226
 227     retval = INLINE_SYSCALL (getdents, 3, fd, (char *) kdp, red_nbytes);
 228
 229     if (retval == -1)
 230       return -1;
 231
 232     DIRENT_TYPE *dp = (DIRENT_TYPE *) buf;
 233     while ((char *) kdp < (char *) skdp + retval)
 234       {
 235         const size_t alignment = __alignof__ (DIRENT_TYPE);
 236         /* Since kdp->d_reclen is already aligned for the kernel structure
 237            this may compute a value that is bigger than necessary.  */
 238         size_t new_reclen = ((kdp->d_reclen + size_diff + alignment - 1)
 239                              & ~(alignment - 1));
 240         if ((char *) dp + new_reclen > buf + nbytes)
 241           {
 242             /* Our heuristic failed.  We read too many entries.  Reset
 243                the stream.  */
 244             assert (last_offset != -1);
 245             __lseek64 (fd, last_offset, SEEK_SET);
 246
 247             if ((char *) dp == buf)
 248               {
 249                 /* The buffer the user passed in is too small to hold even
 250                    one entry.  */
 251                 __set_errno (EINVAL);
 252                 return -1;
 253               }
 254
 255             break;
 256           }
 257
 258         last_offset = kdp->d_off;
 259         DIRENT_SET_DP_INO(dp, kdp->d_ino);
 260         dp->d_off = kdp->d_off;
 261         dp->d_reclen = new_reclen;
 262         dp->d_type = *((char *) kdp + kdp->d_reclen - 1);
 263         memcpy (dp->d_name, kdp->d_name,
 264                 kdp->d_reclen - offsetof (struct kernel_dirent, d_name));
 265
 266         dp = (DIRENT_TYPE *) ((char *) dp + new_reclen);
 267         kdp = (struct kernel_dirent *) (((char *) kdp) + kdp->d_reclen);
 268       }
 269
 270     return (char *) dp - buf;
 271   }
 272 }