Skip to content
This repository has been archived by the owner on Jan 30, 2023. It is now read-only.

Releases: mozilla/libprio

Release 1.6

09 May 00:06
52643ce
Compare
Choose a tag to compare

commit 52643ce
Author: Robert Helmer rhelmer@rhelmer.org
Date: Sun Apr 26 12:23:09 2020 -0700

upgrade Travis CI distro to Ubuntu Bionic (#86)

commit f72b2b2
Author: Henry Corrigan-Gibbs 163574+henrycg@users.noreply.github.com
Date: Tue Apr 21 19:14:17 2020 +0200

Fix serialization bug in `prio/serial.c` (#81)

* In `serial_read_packet_client`, we should check that the type
of the incoming packet matches the type of the `PrioPacketClient`
passed into `serial_read_packet_client` and fail if the two don't
match.

* Add a test for this behavior.

Co-authored-by: Robert Helmer <rhelmer@rhelmer.org>

commit ed4142a
Author: Christian Holler (:decoder) choller@mozilla.com
Date: Tue Apr 21 19:07:56 2020 +0200

Fix memory leak in serial_read_packet_client (#80)

* Fix memory leak in serial_read_packet_client

* Fixing more memory leaks in `prio/serial.c`

The `UP_CHECK` macro just returned on failure, rather than
jumping to the `cleanup` label. This commit replaces `UP_CHECK`
with `UP_CHECKC`, which jumps to `cleanup` on failure.

* Replace `P_CHECK` with `P_CHECKC` in prio/serial.c

Every function in `prio/serial.c` has a `cleanup` label, but
much of the code in these functions won't actually jump to
cleanup on failure. As far as I can tell, this doesn't cause
any additional memory leaks, but in case someone adds cleanup
code at the `cleanup` label in the future, it seems prudent to
always jump to `cleanup` on failure.

Co-authored-by: Henry Corrigan-Gibbs <henrycg@csail.mit.edu>

commit 6efb51c
Author: tititiou36 679652+tititiou36@users.noreply.github.com
Date: Sun Apr 19 12:15:31 2020 +0200

No need to memset memory allocated with calloc

commit 9a30c74
Author: Gabriel Ganne gabriel.ganne@gmail.com
Date: Wed Aug 7 16:12:05 2019 +0200

remove unnecessary runtime call to strlen()

PRIO_TAG is a fixed-size string, we can use sizeof to the same effect.
This also has the nice side-effect of removing a variable-length-array
since AAD_LEN could not be known at compile-time

commit 854d87d
Author: Anthony Miyaguchi amiyaguchi@mozilla.com
Date: Tue Apr 30 14:44:59 2019 -0700

Update SConstruct to modify include path based on sys.platform (#74)

* Format SConstruct with black

* Update SConstruct to modify include path based on OS

* Remove unneeded exports

commit 662191a
Author: Henry Corrigan-Gibbs henrycg-git@cs.stanford.edu
Date: Sat Mar 30 08:57:09 2019 -0700

Remove comment in CODE_OF_CONDUCT.md

commit bbca24f
Author: Mozilla-GitHub-Standards 48073334+Mozilla-GitHub-Standards@users.noreply.github.com
Date: Fri Mar 29 23:57:03 2019 -0700

Add Mozilla Code of Conduct file

Fixes #72.

_(Message COC002)_

Remove precomputed table of roots

28 Mar 04:46
Compare
Choose a tag to compare
Remove precomputed table of roots (#68) (#69)

* Remove precomputed table of roots (#68)

There is no need to ship the large precomputed table of roots of unity;
it's possible to compute the table on the fly in linear time (in the
table size). This commit:
* removes the table from `prio/params.h`,
* removes the script that generates it (`scripts/gen_params.py`), and
* adds code to generate the table in `prio/config.c`.

* Point Travis to latest version of NSS

1.4

10 Jan 05:48
a95cfdd
Compare
Choose a tag to compare
1.4
use `unsigned long long` to support 64-bit ints on 32-bit platforms (…

1.3: Fix security bug in proof-checking code (#60)

09 Jan 06:46
Compare
Choose a tag to compare
* Add PRG_get_int_range() function

* Fix bug in proof-checking code.

Per Appendix D.2 of the full version of the Prio paper, the servers
should evaluate the polynomials f, g, and h at a random point r in the
set {M+1, ..., modulus-1}, where M is the number of multiplication gates
in the "Valid" circuit. The implementation previously sampled the point
r from the larger set {0, ..., modulus-1}.

* Remove extraneous printf() in test code

* Fixes per clang-format