##// END OF EJS Templates
revlog: add a mechanism to verify expected file position before appending...
revlog: add a mechanism to verify expected file position before appending If someone uses `hg debuglocks`, or some non-hg process writes to the .hg directory without respecting the locks, or if the repo's on a networked filesystem, it's possible for the revlog code to write out corrupted data. The form of this corruption can vary depending on what data was written and how that happened. We are in the "networked filesystem" case (though I've had users also do this to themselves with the "`hg debuglocks`" scenario), and most often see this with the changelog. What ends up happening is we produce two items (let's call them rev1 and rev2) in the .i file that have the same linkrev, baserev, and offset into the .d file, while the data in the .d file is appended properly. rev2's compressed_size is accurate for rev2, but when we go to decompress the data in the .d file, we use the offset that's recorded in the index file, which is the same as rev1, and attempt to decompress rev2.compressed_size bytes of rev1's data. This usually does not succeed. :) When using inline data, this also fails, though I haven't investigated why too closely. This shows up as a "patch decode" error. I believe what's happening there is that we're basically ignoring the offset field, getting the data properly, but since baserev != rev, it thinks this is a delta based on rev (instead of a full text) and can't actually apply it as such. For now, I'm going to make this an optional component and default it to entirely off. I may increase the default severity of this in the future, once I've enabled it for my users and we gain more experience with it. Luckily, most of my users have a versioned filesystem and can roll back to before the corruption has been written, it's just a hassle to do so and not everyone knows how (so it's a support burden). Users on other filesystems will not have that luxury, and this can cause them to have a corrupted repository that they are unlikely to know how to resolve, and they'll see this as a data-loss event. Refusing to create the corruption is a much better user experience. This mechanism is not perfect. There may be false-negatives (racy writes that are not detected). There should not be any false-positives (non-racy writes that are detected as such). This is not a mechanism that makes putting a repo on a networked filesystem "safe" or "supported", just *less* likely to cause corruption. Differential Revision: https://phab.mercurial-scm.org/D9952
Kyle Lippincott -
r47349:e9901d01 default
Show More
Name Size Modified Last Commit Author
/ mercurial / thirdparty / sha1dc
lib
LICENSE.txt Loading ...
README.md Loading ...
cext.c Loading ...

sha1collisiondetection

Library and command line tool to detect SHA-1 collisions in files

Copyright 2017 Marc Stevens marc@marc-stevens.nl

Distributed under the MIT Software License.

See accompanying file LICENSE.txt or copy at https://opensource.org/licenses/MIT.

Developers

About

This library and command line tool were designed as near drop-in replacements for common SHA-1 libraries and sha1sum.
They will compute the SHA-1 hash of any given file and additionally will detect cryptanalytic collision attacks against SHA-1 present in each file. It is very fast and takes less than twice the amount of time as regular SHA-1.

More specifically they will detect any cryptanalytic collision attack against SHA-1 using any of the top 32 SHA-1 disturbance vectors with probability 1:

    I(43,0), I(44,0), I(45,0), I(46,0), I(47,0), I(48,0), I(49,0), I(50,0), I(51,0), I(52,0),
    I(46,2), I(47,2), I(48,2), I(49,2), I(50,2), I(51,2),
    II(45,0), II(46,0), II(47,0), II(48,0), II(49,0), II(50,0), II(51,0), II(52,0), II(53,0), II(54,0), II(55,0), II(56,0),
    II(46,2), II(49,2), II(50,2), II(51,2)

The possibility of false positives can be neglected as the probability is smaller than 2^-90.

The library supports both an indicator flag that applications can check and act on, as well as a special safe-hash mode that returns the real SHA-1 hash when no collision was detected and a different safe hash when a collision was detected.
Colliding files will have the same SHA-1 hash, but will have different unpredictable safe-hashes.
This essentially enables protection of applications against SHA-1 collisions with no further changes in the application, e.g., digital signature forgeries based on SHA-1 collisions automatically become invalid.

For the theoretical explanation of collision detection see the award-winning paper on Counter-Cryptanalysis:

Counter-cryptanalysis, Marc Stevens, CRYPTO 2013, Lecture Notes in Computer Science, vol. 8042, Springer, 2013, pp. 129-146,
https://marc-stevens.nl/research/papers/C13-S.pdf

Compiling

Run:

make

Command-line usage

There are two programs bin/sha1dcsum and bin/sha1dcsum_partialcoll.
The first program bin/sha1dcsum will detect and warn for files that were generated with a cryptanalytic SHA-1 collision attack like the one documented at https://shattered.io/.
The second program bin/sha1dcsum_partialcoll will detect and warn for files that were generated with a cryptanalytic collision attack against reduced-round SHA-1 (of which there are a few examples so far).

Examples:

bin/sha1dcsum test/sha1_reducedsha_coll.bin test/shattered-1.pdf
bin/sha1dcsum_partialcoll test/sha1reducedsha_coll.bin test/shattered-1.pdf
pipe_data | bin/sha1dcsum -

Library usage

See the documentation in lib/sha1.h. Here is a simple example code snippet:

#include <sha1dc/sha1.h>

SHA1_CTX ctx;
unsigned char hash[20];
SHA1DCInit(&ctx);

/** disable safe-hash mode (safe-hash mode is enabled by default) **/
// SHA1DCSetSafeHash(&ctx, 0);
/** disable use of unavoidable attack conditions to speed up detection (enabled by default) **/
// SHA1DCSetUseUBC(&ctx, 0);

SHA1DCUpdate(&ctx, buffer, (unsigned)(size));

int iscoll = SHA1DCFinal(hash,&ctx);
if (iscoll)
    printf("collision detected");
else
    printf("no collision detected");

Inclusion in other programs

In order to make it easier to include these sources in other project
there are several preprocessor macros that the code uses. Rather than
copy/pasting and customizing or specializing the code, first see if
setting any of these defines appropriately will allow you to avoid
modifying the code yourself.

  • SHA1DC_NO_STANDARD_INCLUDES

Skips including standard headers. Use this if your project for
whatever reason wishes to do its own header includes.

  • SHA1DC_CUSTOM_INCLUDE_SHA1_C

    Includes a custom header at the top of sha1.c. Usually this would be
    set in conjunction with SHA1DC_NO_STANDARD_INCLUDES to point to a
    header file which includes various standard headers.

  • SHA1DC_INIT_SAFE_HASH_DEFAULT

    Sets the default for safe_hash in SHA1DCInit(). Valid values are 0
    and 1. If unset 1 is the default.

  • SHA1DC_CUSTOM_TRAILING_INCLUDE_SHA1_C

    Includes a custom trailer in sha1.c. Useful for any extra utility
    functions that make use of the functions already defined in sha1.c.

  • SHA1DC_CUSTOM_TRAILING_INCLUDE_SHA1_H

    Includes a custom trailer in sha1.h. Useful for defining the
    prototypes of the functions or code included by
    SHA1DC_CUSTOM_TRAILING_INCLUDE_SHA1_C.

  • SHA1DC_CUSTOM_INCLUDE_UBC_CHECK_C

    Includes a custom header at the top of ubc_check.c.

  • SHA1DC_CUSTOM_TRAILING_INCLUDE_UBC_CHECK_C

    Includes a custom trailer in ubc_check.c.

  • SHA1DC_CUSTOM_TRAILING_INCLUDE_UBC_CHECK_H

    Includes a custom trailer in ubc_check.H.

This code will try to auto-detect certain things based on
CPU/platform. Unless you're running on some really obscure CPU or
porting to a new platform you should not need to tweak this. If you do
please open an issue at
https://github.com/cr-marcstevens/sha1collisiondetection

  • SHA1DC_FORCE_LITTLEENDIAN / SHA1DC_FORCE_BIGENDIAN

    Override the check for processor endianenss and force either
    Little-Endian or Big-Endian.

  • SHA1DC_FORCE_UNALIGNED_ACCESS

    Permit unaligned access. This will fail on e.g. SPARC processors, so
    it's only permitted on a whitelist of processors. If your CPU isn't
    detected as allowing this, and allows unaligned access, setting this
    may improve performance (or make it worse, if the kernel has to
    catch and emulate such access on its own).