libburnia/libisofs
2
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Made development macro Libisofs_hardlink_prooF unconditional.

Browse Source
This commit is contained in:
Thomas Schmitt
2009-08-22 22:33:04 +02:00
parent b1c4571a95
commit d0996450c7
2 changed files with 0 additions and 117 deletions
Download Patch File Download Diff File Expand all files Collapse all files

103
libisofs/fs_image.c
View File

@@ -1095,33 +1095,6 @@ char *get_name(_ImageFsData *fsdata, const char *str, size_t len)
}
#ifndef Libisofs_hardlink_prooF
/**
* A global counter for default inode numbers for the ISO image filesystem.
* @param fs The filesystem where the number shall be used
* @param flag bit0= reset count
*/
static
ino_t fs_give_ino_number(IsoImageFilesystem *fs, int flag)
{
_ImageFsData *fsdata;
fsdata = (_ImageFsData*)fs->data;
if (flag & 1)
fsdata->inode_counter = 0;
fsdata->inode_counter++;
if (fsdata->inode_counter == 0) {
/* >>> raise alert because of inode rollover */;
}
return fsdata->inode_counter;
}
#endif /* ! Libisofs_hardlink_prooF */
/**
*
* @param src
@@ -1646,56 +1619,11 @@ int iso_file_source_new_ifs(IsoImageFilesystem *fs, IsoFileSource *parent,
return ISO_SUCCESS;
}
#ifdef Libisofs_hardlink_prooF
/* Production of missing inode numbers is delayed until the image is
complete. Then all nodes which shall get a new inode number will
be served.
*/
#else /* Libisofs_hardlink_prooF */
#ifdef Libisofs_new_fs_image_inO
if (fsdata->rr != RR_EXT_112) {
if (fsdata->rr == 0) {
atts.st_nlink = 1;
}
}
atts.st_ino = fs_give_ino_number(fs, 0);
#else /* Libisofs_new_fs_image_inO */
/* ts Nov 25 2008: TODO
This seems not fully consistent with read_rr_PX() which decides
by (px->len_sue[0] == 44) whether an inode number is present or not.
What if read_rr_PX finds a PX of length 36 in a IEEE_1282 image ?
It is illegal but could confuse the image by duplicate inode numbers.
Regrettably it is not enough to just use single default numbers.
If only one number misses in the image, then all would need to be
defaulted by the following iso_global_inode code.
Why do duplicate inode numbers confuse the file lengths, anyway ?
(See ticket 144)
*/
if (fsdata->rr != RR_EXT_112) {
/*
* Only RRIP 1.12 provides valid inode numbers. If not, it is not easy
* to generate those serial numbers, and we use extend block instead.
* It BREAKS POSIX SEMANTICS, but its suitable for our needs
*/
atts.st_ino = fs_give_ino_number(fs, 0);
if (fsdata->rr == 0) {
atts.st_nlink = 1;
}
}
#endif /* ! Libisofs_new_fs_image_inO */
#endif /* ! Libisofs_hardlink_prooF */
/*
* if we haven't RR extensions, or a needed TF time stamp is not present,
* we use plain iso recording time
@@ -2865,8 +2793,6 @@ int image_builder_create_node(IsoNodeBuilder *builder, IsoImage *image,
goto failure;
}
#ifdef Libisofs_hardlink_prooF
/* Attach ino as xinfo if valid and no IsoStream is involved */
if (info.st_ino != 0 && (info.st_mode & S_IFMT) != S_IFREG &&
!fsdata->make_new_ino) {
@@ -2875,8 +2801,6 @@ int image_builder_create_node(IsoNodeBuilder *builder, IsoImage *image,
goto failure;
}
#endif /* Libisofs_hardlink_prooF */
*node = new;
return ISO_SUCCESS;
@@ -2939,25 +2863,7 @@ int create_boot_img_filesrc(IsoImageFilesystem *fs, IsoImage *image,
memset(&atts, 0, sizeof(struct stat));
atts.st_mode = S_IFREG;
#ifdef Libisofs_hardlink_prooF
atts.st_ino = img_give_ino_number(image, 0);
#else /* Libisofs_hardlink_prooF */
#ifdef Libisofs_new_fs_image_inO
atts.st_ino = fs_give_ino_number(fs, 0);
#else /* Libisofs_new_fs_image_inO */
atts.st_ino = fsdata->imgblock; /* not the best solution, but... */
#endif /* ! Libisofs_new_fs_image_inO */
#endif /* ! Libisofs_hardlink_prooF */
atts.st_nlink = 1;
/*
@@ -3104,17 +3010,12 @@ int iso_image_import(IsoImage *image, IsoDataSource *src,
if (ret < 0)
goto import_revert;
#ifdef Libisofs_hardlink_prooF
/* Attach ino as xinfo if valid */
if (info.st_ino != 0 && !data->make_new_ino) {
ret = iso_node_set_ino(&(image->root->node), info.st_ino, 0);
if (ret < 0)
goto import_revert;
}
#endif /* Libisofs_hardlink_prooF */
}
/* if old image has el-torito, add a new catalog */
@@ -3151,8 +3052,6 @@ int iso_image_import(IsoImage *image, IsoDataSource *src,
goto import_revert;
}
#ifdef Libisofs_hardlink_prooF
/* Take over inode management from IsoImageFilesystem.
data->inode_counter is supposed to hold the maximum PX inode number.
*/
@@ -3173,8 +3072,6 @@ int iso_image_import(IsoImage *image, IsoDataSource *src,
}
}
#endif /* ! Libisofs_hardlink_prooF */
if (data->eltorito) {
/* if catalog and image nodes were not filled, we create them here */
if (image->bootcat->image->image == NULL) {

14
libisofs/libisofs.h
View File

@@ -5731,12 +5731,6 @@ struct burn_source {
/* ---------------------------- Improvements --------------------------- */
/* Hardlinks : Override Libisofs_new_fs_image_inO and preserve inode numbers
from session to session.
*/
#define Libisofs_hardlink_prooF yes
/* Checksums : During image writing equip IsoFile objects with MD5 checksums
and compute an overall checksum of the session. Store them in
a separate checksum block area after the data area of the
@@ -5747,14 +5741,6 @@ struct burn_source {
/* ---------------------------- Experiments ---------------------------- */
/* <<< on its way out
Experiment: Ignore PX inode numbers,
have boot image inode number counted by fs_give_ino_number()
Overridden if Libisofs_hardlink_prooF is defined.
#define Libisofs_new_fs_image_inO yes
*/
/* Experiment: Write obsolete RR entries with Rock Ridge.
I suspect Solaris wants to see them.