Skip to contentSkip to conversation

cryptodisk: Allows UUIDs to be compared in a dash-insensitive manner

PATCH Superseded Glenn Washburn 2 participants 8 messages Jul 22, 2022 – Aug 3, 2022

v4 of 6 Superseded Latest is v6 →

lore
Share & tools
lore
#0 lore raw

v4 cover cryptodisk: Allows UUIDs to be compared in a dash-insensitive manner

Updates since v3:
* Rebase onto latest master
* Whitespace fixes

Glenn

Glenn Washburn (2):
luks, luks2: Force header.uuid to be NULL terminated
cryptodisk: Allows UUIDs to be compared in a dash-insensitive manner

 grub-core/disk/cryptodisk.c |  4 ++--
 grub-core/disk/geli.c       |  2 +-
 grub-core/disk/luks.c       | 24 +++++++-----------------
 grub-core/disk/luks2.c      | 18 +++++++-----------
 include/grub/misc.h         | 27 +++++++++++++++++++++++++++
 5 files changed, 44 insertions(+), 31 deletions(-)
Signature · 1 line
2.34.1
#1 lore raw

PATCH v4 1/2 luks, luks2: Force header.uuid to be NULL terminated

According to the LUKS specification the uuid header field is of data type "char[]", which is defined as "a string stored as null terminated sequence of 8-bit characters". So enforce this by adding a null byte as the last byte of the uuid. The LUKS2 specification defers to the LUKS1 specification in this regard.

---
 grub-core/disk/luks.c  | 3 +++
 grub-core/disk/luks2.c | 3 +++
 2 files changed, 6 insertions(+)
grub-core/disk/luks.c +3 -0
index 7f837d52c..ad72e5d1f 100644
@@ -87,6 +87,9 @@ luks_scan (grub_disk_t disk, grub_cryptomount_args_t cargs)
       return NULL;
     }
 
+  /* According to the spec the uuid field is NULL terminated, so enforce it. */
+  header.uuid[sizeof (header.uuid) - 1] = '\0';
+
   /* Look for LUKS magic sequence.  */
   if (grub_memcmp (header.magic, LUKS_MAGIC, sizeof (header.magic))
       || grub_be_to_cpu16 (header.version) != 1)
grub-core/disk/luks2.c +3 -0
index bf741d70f..1346c3821 100644
@@ -342,6 +342,9 @@ luks2_read_header (grub_disk_t disk, grub_luks2_header_t *outhdr)
       header = &secondary;
   grub_memcpy (outhdr, header, sizeof (*header));
 
+  /* According to the spec the uuid field is NULL terminated, so enforce it. */
+  header->uuid[sizeof (header->uuid) - 1] = '\0';
+
   return GRUB_ERR_NONE;
 }
 
Signature · 1 line
2.34.1
#2 lore raw

Re: [PATCH v4 1/2] luks, luks2: Force header.uuid to be NULL terminated

Quoted 7 lines · Glenn Washburn

According to the LUKS specification the uuid header field is of data type "char[]", which is defined as "a string stored as null terminated sequence of 8-bit characters". So enforce this by adding a null byte as the last byte of the uuid. The LUKS2 specification defers to the LUKS1 specification in this regard.

In general Reviewed-by: Daniel Kiper <[email protected]>...

Though I want to ask you to add a sentence to the commit message saying why current code works now. And a second sentence should say the part of existing code making it work will be removed by subsequent patch. Otherwise it is not really clear why this patch is needed...

Or maybe we should merge patch #1 and #2... However, I do not have strong opinion here.

Daniel

#3 lore raw

Re: [PATCH v4 1/2] luks, luks2: Force header.uuid to be NULL terminated

Quoted 7 lines · On Wed, 3 Aug 2022 15:30:58 +0200 Daniel Kiper <[email protected]> wrote:
Quoted 6 lines · Glenn Washburn

According to the LUKS specification the uuid header field is of data type "char[]", which is defined as "a string stored as null terminated sequence of 8-bit characters". So enforce this by adding a null byte as the last byte of the uuid. The LUKS2 specification defers to the LUKS1 specification in this regard.

Quoted 8 lines
Quoted 1 line

In general Reviewed-by: Daniel Kiper <[email protected]>...

Though I want to ask you to add a sentence to the commit message saying why current code works now. And a second sentence should say the part of existing code making it work will be removed by subsequent patch. Otherwise it is not really clear why this patch is needed...

Looking at this again, this patch really isn't needed in a functional sense, with or without the following patch. I think I might have seen this as desirable at one time because this was written before Patricks commit 1066336d (luks: Fix out-of-bounds copy of UUID).

Its not needed for the subsequent patch because UUID comparison is always bounded. I'll drop this change.

Glenn

Quoted 5 lines

Or maybe we should merge patch #1 and #2... However, I do not have strong opinion here.

Daniel

#4 lore raw

PATCH v4 2/2 cryptodisk: Allows UUIDs to be compared in a dash-insensitive manner

In reply to Glenn Washburn

A user can now specify UUID strings with dashes, instead of having to remove dashes. This is backwards-compatability preserving and also fixes a source of user confusion over the inconsistency with how UUIDs are specified between file system UUIDs and cryptomount UUIDs. Since cryptsetup, the reference implementation for LUKS, displays and generates UUIDs with dashes there has been additional confusion when using the UUID strings from cryptsetup as exact input into GRUB does not find the expected cryptodisk.

A new function grub_uuidcasecmp is added that is general enough to be used other places where UUIDs are being compared.

---
 grub-core/disk/cryptodisk.c |  4 ++--
 grub-core/disk/geli.c       |  2 +-
 grub-core/disk/luks.c       | 21 ++++-----------------
 grub-core/disk/luks2.c      | 15 ++++-----------
 include/grub/misc.h         | 27 +++++++++++++++++++++++++++
 5 files changed, 38 insertions(+), 31 deletions(-)
grub-core/disk/cryptodisk.c +2 -2
index f1fe0d390..00d66ab02 100644
@@ -701,7 +701,7 @@ grub_cryptodisk_open (const char *name, grub_disk_t disk)
   if (grub_memcmp (name, "cryptouuid/", sizeof ("cryptouuid/") - 1) == 0)
     {
       for (dev = cryptodisk_list; dev != NULL; dev = dev->next)
-	if (grub_strcasecmp (name + sizeof ("cryptouuid/") - 1, dev->uuid) == 0)
+	if (grub_uuidcasecmp (name + sizeof ("cryptouuid/") - 1, dev->uuid, sizeof (dev->uuid)) == 0)
 	  break;
     }
   else
@@ -928,7 +928,7 @@ grub_cryptodisk_get_by_uuid (const char *uuid)
 {
   grub_cryptodisk_t dev;
   for (dev = cryptodisk_list; dev != NULL; dev = dev->next)
-    if (grub_strcasecmp (dev->uuid, uuid) == 0)
+    if (grub_uuidcasecmp (dev->uuid, uuid, sizeof (dev->uuid)) == 0)
       return dev;
   return NULL;
 }
grub-core/disk/geli.c +1 -1
index e190066f9..722910d64 100644
@@ -305,7 +305,7 @@ geli_scan (grub_disk_t disk, grub_cryptomount_args_t cargs)
       return NULL;
     }
 
-  if (cargs->search_uuid != NULL && grub_strcasecmp (cargs->search_uuid, uuid) != 0)
+  if (cargs->search_uuid != NULL && grub_uuidcasecmp (cargs->search_uuid, uuid, sizeof (uuid)) != 0)
     {
       grub_dprintf ("geli", "%s != %s\n", uuid, cargs->search_uuid);
       return NULL;
grub-core/disk/luks.c +4 -17
index ad72e5d1f..3ca08b805 100644
@@ -66,10 +66,7 @@ static grub_cryptodisk_t
 luks_scan (grub_disk_t disk, grub_cryptomount_args_t cargs)
 {
   grub_cryptodisk_t newdev;
-  const char *iptr;
   struct grub_luks_phdr header;
-  char *optr;
-  char uuid[sizeof (header.uuid) + 1];
   char ciphername[sizeof (header.cipherName) + 1];
   char ciphermode[sizeof (header.cipherMode) + 1];
   char hashspec[sizeof (header.hashSpec) + 1];
@@ -95,19 +92,9 @@ luks_scan (grub_disk_t disk, grub_cryptomount_args_t cargs)
       || grub_be_to_cpu16 (header.version) != 1)
     return NULL;
 
-  grub_memset (uuid, 0, sizeof (uuid));
-  optr = uuid;
-  for (iptr = header.uuid; iptr < &header.uuid[ARRAY_SIZE (header.uuid)];
-       iptr++)
+  if (cargs->search_uuid != NULL && grub_uuidcasecmp (cargs->search_uuid, header.uuid, sizeof (header.uuid)) != 0)
     {
-      if (*iptr != '-')
-	*optr++ = *iptr;
-    }
-  *optr = 0;
-
-  if (cargs->search_uuid != NULL && grub_strcasecmp (cargs->search_uuid, uuid) != 0)
-    {
-      grub_dprintf ("luks", "%s != %s\n", uuid, cargs->search_uuid);
+      grub_dprintf ("luks", "%s != %s\n", header.uuid, cargs->search_uuid);
       return NULL;
     }
 
@@ -126,7 +113,7 @@ luks_scan (grub_disk_t disk, grub_cryptomount_args_t cargs)
   newdev->source_disk = NULL;
   newdev->log_sector_size = GRUB_LUKS1_LOG_SECTOR_SIZE;
   newdev->total_sectors = grub_disk_native_sectors (disk) - newdev->offset_sectors;
-  grub_memcpy (newdev->uuid, uuid, sizeof (uuid));
+  grub_memcpy (newdev->uuid, header.uuid, sizeof (header.uuid));
   newdev->modname = "luks";
 
   /* Configure the hash used for the AF splitter and HMAC.  */
@@ -146,7 +133,7 @@ luks_scan (grub_disk_t disk, grub_cryptomount_args_t cargs)
       return NULL;
     }
 
-  COMPILE_TIME_ASSERT (sizeof (newdev->uuid) >= sizeof (uuid));
+  COMPILE_TIME_ASSERT (sizeof (newdev->uuid) >= sizeof (header.uuid));
   return newdev;
 }
 
grub-core/disk/luks2.c +4 -11
index 1346c3821..2fb500c37 100644
@@ -353,8 +353,6 @@ luks2_scan (grub_disk_t disk, grub_cryptomount_args_t cargs)
 {
   grub_cryptodisk_t cryptodisk;
   grub_luks2_header_t header;
-  char uuid[sizeof (header.uuid) + 1];
-  grub_size_t i, j;
 
   if (cargs->check_boot)
     return NULL;
@@ -365,14 +363,9 @@ luks2_scan (grub_disk_t disk, grub_cryptomount_args_t cargs)
       return NULL;
     }
 
-  for (i = 0, j = 0; i < sizeof (header.uuid); i++)
-    if (header.uuid[i] != '-')
-      uuid[j++] = header.uuid[i];
-  uuid[j] = '\0';
-
-  if (cargs->search_uuid != NULL && grub_strcasecmp (cargs->search_uuid, uuid) != 0)
+  if (cargs->search_uuid != NULL && grub_uuidcasecmp (cargs->search_uuid, header.uuid, sizeof (header.uuid)) != 0)
     {
-      grub_dprintf ("luks2", "%s != %s\n", uuid, cargs->search_uuid);
+      grub_dprintf ("luks2", "%s != %s\n", header.uuid, cargs->search_uuid);
       return NULL;
     }
 
@@ -380,8 +373,8 @@ luks2_scan (grub_disk_t disk, grub_cryptomount_args_t cargs)
   if (!cryptodisk)
     return NULL;
 
-  COMPILE_TIME_ASSERT (sizeof (cryptodisk->uuid) >= sizeof (uuid));
-  grub_memcpy (cryptodisk->uuid, uuid, sizeof (uuid));
+  COMPILE_TIME_ASSERT (sizeof (cryptodisk->uuid) >= sizeof (header.uuid));
+  grub_memcpy (cryptodisk->uuid, header.uuid, sizeof (header.uuid));
 
   cryptodisk->modname = "luks2";
   return cryptodisk;
include/grub/misc.h +27 -0
index 776dbf8af..39957ecf9 100644
@@ -243,6 +243,33 @@ grub_strncasecmp (const char *s1, const char *s2, grub_size_t n)
     - (int) grub_tolower ((grub_uint8_t) *s2);
 }
 
+/* Do a case insensitive compare of two UUID strings by ignoring all dashes */
+static inline int
+grub_uuidcasecmp (const char *uuid1, const char *uuid2, grub_size_t n)
+{
+  if (n == 0)
+    return 0;
+
+  while (*s1 && *s2 && --n)
+    {
+      /* Skip forward to non-dash on both UUIDs. */
+      while ('-' == *s1)
+        ++s1;
+
+      while ('-' == *s2)
+        ++s2;
+
+      if (grub_tolower (*s1) != grub_tolower (*s2))
+	break;
+
+      s1++;
+      s2++;
+    }
+
+  return (int) grub_tolower ((grub_uint8_t) *s1)
+    - (int) grub_tolower ((grub_uint8_t) *s2);
+}
+
 /*
  * Note that these differ from the C standard's definitions of strtol,
  * strtoul(), and strtoull() by the addition of two const qualifiers on the end
Signature · 1 line
2.34.1
#5 lore raw
Quoted 7 lines · Glenn Washburn

A user can now specify UUID strings with dashes, instead of having to remove dashes. This is backwards-compatability preserving and also fixes a source of user confusion over the inconsistency with how UUIDs are specified between file system UUIDs and cryptomount UUIDs. Since cryptsetup, the reference implementation for LUKS, displays and generates UUIDs with dashes there has been additional confusion when using the UUID strings from cryptsetup as exact input into GRUB does not find the expected cryptodisk.

I expect the confusion comes from differences between older and newer cryptsetup implementations. AIUI older versions used UUIDs without dashes, and this is what GRUB expects today, but newer ones use UUIDs with dashes. Right? If it is true it took me some time to get it from the text above. So, I think it should be improved a bit to make it more clear...

Quoted 11 lines

A new function grub_uuidcasecmp is added that is general enough to be used other places where UUIDs are being compared.

---
 grub-core/disk/cryptodisk.c |  4 ++--
 grub-core/disk/geli.c       |  2 +-
 grub-core/disk/luks.c       | 21 ++++-----------------
 grub-core/disk/luks2.c      | 15 ++++-----------
 include/grub/misc.h         | 27 +++++++++++++++++++++++++++
 5 files changed, 38 insertions(+), 31 deletions(-)

[...]

Quoted 15 lines
include/grub/misc.h +7 -0
index 776dbf8af..39957ecf9 100644
@@ -243,6 +243,33 @@ grub_strncasecmp (const char *s1, const char *s2, grub_size_t n)
     - (int) grub_tolower ((grub_uint8_t) *s2);
 }
 
+/* Do a case insensitive compare of two UUID strings by ignoring all dashes */
+static inline int
+grub_uuidcasecmp (const char *uuid1, const char *uuid2, grub_size_t n)
+{
+  if (n == 0)
+    return 0;
+

include/grub/misc.h grub_strncasecmp (const char *s1, const char *s2, grub_size_t n) · line 262

Quoted 10 lines
include/grub/misc.h +10 -0
+  while (*s1 && *s2 && --n)
+    {
+      /* Skip forward to non-dash on both UUIDs. */
+      while ('-' == *s1)
+        ++s1;
+
+      while ('-' == *s2)
+        ++s2;
+
+      if (grub_tolower (*s1) != grub_tolower (*s2))

I think you took this code from grub_strncasecmp(). It is missing grub_uint8_t casts here too. However, if you take a look at grub_strcasecmp(), a few lines above, you can see casts in similar place. It is really confusing. I thing the casts has been added to drop sign. If I am right then grub_strncasecmp() and grub_uuidcasecmp() both should be fixed.

include/grub/misc.h grub_strncasecmp (const char *s1, const char *s2, grub_size_t n) · line 270

Quoted 8 lines
include/grub/misc.h +8 -0
+	break;
+
+      s1++;
+      s2++;
+    }
+
+  return (int) grub_tolower ((grub_uint8_t) *s1)
+    - (int) grub_tolower ((grub_uint8_t) *s2);

... and these casts suggests again something is wrong above too...

Daniel

#6 lore raw
Quoted 9 lines · On Tue, 2 Aug 2022 18:17:44 +0200 Daniel Kiper <[email protected]> wrote:
Quoted 7 lines · Glenn Washburn

A user can now specify UUID strings with dashes, instead of having to remove dashes. This is backwards-compatability preserving and also fixes a source of user confusion over the inconsistency with how UUIDs are specified between file system UUIDs and cryptomount UUIDs. Since cryptsetup, the reference implementation for LUKS, displays and generates UUIDs with dashes there has been additional confusion when using the UUID strings from cryptsetup as exact input into GRUB does not find the expected cryptodisk.

Quoted 6 lines

I expect the confusion comes from differences between older and newer cryptsetup implementations. AIUI older versions used UUIDs without dashes, and this is what GRUB expects today, but newer ones use UUIDs with dashes. Right? If it is true it took me some time to get it from the text above. So, I think it should be improved a bit to make it more clear...

This may be true, I honestly don't know nor recall that being old behavior of cryptsetup. Do you want me to assume that it is true, and to modify the commit message accordingly? I could hedge and add a sentence here like: "It is possible that very old versions of cryptsetup displayed the UUID without dashes, accounting for the LUKS handling of UUIDs without dashes".

Quoted 38 lines
Quoted 12 lines

A new function grub_uuidcasecmp is added that is general enough to be used other places where UUIDs are being compared.

---
 grub-core/disk/cryptodisk.c |  4 ++--
 grub-core/disk/geli.c       |  2 +-
 grub-core/disk/luks.c       | 21 ++++-----------------
 grub-core/disk/luks2.c      | 15 ++++-----------
 include/grub/misc.h         | 27 +++++++++++++++++++++++++++
 5 files changed, 38 insertions(+), 31 deletions(-)

[...]

Quoted 23 lines
include/grub/misc.h +15 -0
index 776dbf8af..39957ecf9 100644
@@ -243,6 +243,33 @@ grub_strncasecmp (const char *s1, const char *s2, grub_size_t n)
     - (int) grub_tolower ((grub_uint8_t) *s2);
 }
 
+/* Do a case insensitive compare of two UUID strings by ignoring all dashes */
+static inline int
+grub_uuidcasecmp (const char *uuid1, const char *uuid2, grub_size_t n)
+{
+  if (n == 0)
+    return 0;
+
+  while (*s1 && *s2 && --n)
+    {
+      /* Skip forward to non-dash on both UUIDs. */
+      while ('-' == *s1)
+        ++s1;
+
+      while ('-' == *s2)
+        ++s2;
Quoted 8 lines

include/grub/misc.h grub_strncasecmp (const char *s1, const char *s2, grub_size_t n) · line 262

Quoted 2 lines
include/grub/misc.h +2 -0
+
+      if (grub_tolower (*s1) != grub_tolower (*s2))

I think you took this code from grub_strncasecmp(). It is missing grub_uint8_t casts here too. However, if you take a look at grub_strcasecmp(), a few lines above, you can see casts in similar place. It is really confusing. I thing the casts has been added to drop sign. If I am right then grub_strncasecmp() and grub_uuidcasecmp() both should be fixed.

Yes, this part was just copied. So I'll add the casts here and to grub_strcasecmp(). Should I add the changes to grub_strcasecmp() in this patch and add a note in the commit message of the change?

Quoted 12 lines

include/grub/misc.h grub_strncasecmp (const char *s1, const char *s2, grub_size_t n) · line 270

Quoted 9 lines
include/grub/misc.h +8 -0
+	break;
+
+      s1++;
+      s2++;
+    }
+
+  return (int) grub_tolower ((grub_uint8_t) *s1)
+    - (int) grub_tolower ((grub_uint8_t) *s2);

... and these casts suggests again something is wrong above too...

Quoted 1 line

Daniel

Glenn

#7 lore raw
Quoted 19 lines · Glenn Washburn
Quoted 9 lines · On Tue, 2 Aug 2022 18:17:44 +0200 Daniel Kiper <[email protected]> wrote:
Quoted 7 lines · Glenn Washburn

A user can now specify UUID strings with dashes, instead of having to remove dashes. This is backwards-compatability preserving and also fixes a source of user confusion over the inconsistency with how UUIDs are specified between file system UUIDs and cryptomount UUIDs. Since cryptsetup, the reference implementation for LUKS, displays and generates UUIDs with dashes there has been additional confusion when using the UUID strings from cryptsetup as exact input into GRUB does not find the expected cryptodisk.

Quoted 6 lines

I expect the confusion comes from differences between older and newer cryptsetup implementations. AIUI older versions used UUIDs without dashes, and this is what GRUB expects today, but newer ones use UUIDs with dashes. Right? If it is true it took me some time to get it from the text above. So, I think it should be improved a bit to make it more clear...

Quoted 6 lines

This may be true, I honestly don't know nor recall that being old behavior of cryptsetup. Do you want me to assume that it is true, and to modify the commit message accordingly? I could hedge and add a sentence here like: "It is possible that very old versions of cryptsetup displayed the UUID without dashes, accounting for the LUKS handling of UUIDs without dashes".

I thought you know this. If not you can ignore my comment.

Quoted 46 lines
Quoted 37 lines
Quoted 11 lines

A new function grub_uuidcasecmp is added that is general enough to be used other places where UUIDs are being compared.

---
 grub-core/disk/cryptodisk.c |  4 ++--
 grub-core/disk/geli.c       |  2 +-
 grub-core/disk/luks.c       | 21 ++++-----------------
 grub-core/disk/luks2.c      | 15 ++++-----------
 include/grub/misc.h         | 27 +++++++++++++++++++++++++++
 5 files changed, 38 insertions(+), 31 deletions(-)

[...]

Quoted 23 lines
include/grub/misc.h +15 -0
index 776dbf8af..39957ecf9 100644
@@ -243,6 +243,33 @@ grub_strncasecmp (const char *s1, const char *s2, grub_size_t n)
     - (int) grub_tolower ((grub_uint8_t) *s2);
 }
 
+/* Do a case insensitive compare of two UUID strings by ignoring all dashes */
+static inline int
+grub_uuidcasecmp (const char *uuid1, const char *uuid2, grub_size_t n)
+{
+  if (n == 0)
+    return 0;
+
+  while (*s1 && *s2 && --n)
+    {
+      /* Skip forward to non-dash on both UUIDs. */
+      while ('-' == *s1)
+        ++s1;
+
+      while ('-' == *s2)
+        ++s2;
Quoted 8 lines

include/grub/misc.h grub_strncasecmp (const char *s1, const char *s2, grub_size_t n) · line 262

Quoted 2 lines
include/grub/misc.h +2 -0
+
+      if (grub_tolower (*s1) != grub_tolower (*s2))

I think you took this code from grub_strncasecmp(). It is missing grub_uint8_t casts here too. However, if you take a look at grub_strcasecmp(), a few lines above, you can see casts in similar place. It is really confusing. I thing the casts has been added to drop sign. If I am right then grub_strncasecmp() and grub_uuidcasecmp() both should be fixed.

Quoted 3 lines

Yes, this part was just copied. So I'll add the casts here and to grub_strcasecmp(). Should I add the changes to grub_strcasecmp() in this patch and add a note in the commit message of the change?

Please fix grub_strcasecmp() in separate patch. However, it can be a part of this patch series.

Daniel