RECENT BLOG NEWS

With the recent release of wolfSSL 4.1.0, the wolfSSL team has also updated the wolfSSL FIPS Ready library. This product features new, state of the art concepts and technology. In a single sentence, wolfSSL FIPS Ready is a testable and free to download open source embedded SSL/TLS library with support for FIPS validation, with FIPS enabled cryptography layer code included in the wolfSSL source tree. To further elaborate on what FIPS Ready really means, you do not get a FIPS certificate and you are not FIPS approved. FIPS Ready means that you have included the FIPS code into your build and that you are operating according to the FIPS enforced best practices of default entry point, and Power On Self Test (POST).

FIPS validation is a government certification for cryptographic modules that states the module in question has undergone thorough and rigorous testing to be certified. FIPS validation specifies that a software/encryption module is able to be used within or alongside government systems. The most recent FIPS specification is 140-2, with various levels of security offered (1-5). Currently, wolfCrypt has FIPS 140-2 validation with certificates #2425 and #3389. When trying to get software modules FIPS validated, this is often a costly and time-consuming effort and as such causes the FIPS validated modules to have high price tags.

Since the majority of wolfSSL products use the wolfCrypt encryption engine, this also means that if wolfSSH, wolfMQTT (with TLS support), wolfBoot, and other wolfSSL products are in place, they can be tested using FIPS validated code with their software before committing.

wolfSSL FIPS Ready can be downloaded from the wolfSSL download page, here: https://www.wolfssl.com/download/

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.

wolfTPM version 1.6 has been released (08/01/2019) and is available for download! This release includes improvements for compatibility, chip detection, initialization options and small stack support. This version adds new wrapper API’s for PCR extend as well as support for using HMAC with an existing key.

A full list of the release notes are listed below:

• Fix for wolfCrypt init/cleanup issue with reference count. (PR #75)
• Fix to restore existing TPM context after calling wolfTPM2_Test. (PR #74)
• Fix to resolve handling of unsupported ECC curves with the TPM module and ECDHE. (PR #69)
• Fix for wolfTPM2_SetCommand to ensure auth is cleared. (PR #69)
• Added --enable-smallstack build options for reducing stack usage. (PR #73)
• Added support for keeping an HMAC key loaded. (PR #72)
• Added API unit test framework. (PR #71)
• Added new wrapper API wolfTPM2_OpenExisting for accessing device that’s already started. (PR #71)
• Added new wolfTPM2_ExtendPCR wrapper. (PR #70)
• Added crypto callback flags for FIPS mode and Use Symmetric options. (PR #69)
• Added WOLFTPM_DEBUG_TIMEOUT macro for debugging the timeout checking. (PR #69)
• Added support for ST33 TPM2_SetMode command for disabling power saving. (PR #69)
• Improvements for chip detection, compatibility and startup performance (PR #67)
• Added support for XPRINTF.
• Fix printf type warnings.
• Moved the TPM hardware type build macro detection until after the user_settings.h include.
• Optimization to initialize Mutex and RNG only when use is required.
• Added missing stdio.h for printf in examples.
• Added new API’s TPM2_SetActiveCtx, TPM2_ChipStartup, TPM2_SetHalIoCb and TPM2_Init_ex.
• Allowed way to indicate BOOL type already defined.
• Added C++ support.
• Added new API wolfTPM2_Test for testing for TPM and optionally returning capabilities. (PR #66)
• Added way to include generated wolftpm/options.h (or customized one) using WOLFTPM_USER_SETTINGS. (PR #63)

You can download the latest release here: https://www.wolfssl.com/download/

Additional information on wolfTPM can be found on the wolfTPM product page.

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.

wolfMQTT v1.3.0 has been released (08/01/2019) and is available for download. With this release comes multithread support with the enable option --enable-mt, various fixes and improvements to examples, and more.

A full list of wolfMQTT 1.3.0 release notes are listed below:

• Fix fwpush example to use filename option -f. (PR #117)
• Added multiple thread support using --enable-mt or WOLFMQTT_MULTITHREAD. (PR #115)
• Fix for MQTT_DATA_TYPE_BINARY data length encoded twice. (PR #112)
• Fix to clear local structures for subscribe and unsubscribe ACK’s. (PR #112)
• Fix for SN_Encode_Unsubscribe using wrong data type for topic name ID. (PR #110)
• Add WOLFSSL_USER_SETTINGS to VS project files. (PR #109)
• Fixes for using RTCS in mqttnet.c socket example code. (PR #108)
• Fix MQTT-SN decode publish parsing and QoS2 response. (PR #107)
• Make MqttSocket_TlsSocket callbacks public. (PR #104)
• Improved the disconnect network error callback example. (PR #102)
• Add MQTT context information to socket callback examples. (PR #101)
• Initialize subscribe state to MQTT_MSG_BEGIN. (PR #99)
• Fix for Harmony possible circular include issue. (PR #98)

You can download the latest release here: https://www.wolfssl.com/download/

wolfMQTT Product Page

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.

Author: Daniel Stenberg

Without much fanfare or fireworks we put together and shipped a fresh new version of tiny-curl. We call it version 0.10 and it is based on the 7.65.3 curl tree.

tiny-curl is a patch set to build curl as tiny as possible while still being able to perform HTTPS GET requests and maintaining the libcurl API. Additionally, tiny-curl is ported to FreeRTOS.

Changes in 0.10

• The largest and primary change is that this version is based on curl 7.65.3, which brings more features and in particular more bug fixes compared to tiny-curl 0.9.
• Parts of the patches used for tiny-curl 0.9 was subsequently upstreamed and merged into curl proper, making the tiny-curl 0.10 patch much smaller.

Download

As before, tiny-curl is an effort that is on a separate track from the main curl. Download tiny-curl from wolfssl.com!

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.

The wolfSSL wolfBoot Secure Bootloader version 1.2 is now available! With this release comes various improvements and updates, including additional hardware support. Below lists the release notes associated with version 1.2:

–  Added support for multiple architectures
–  key generation and signing tools rewritten in python for portability
–  Added compile-time option to move flash-writing functions to RAM
–  Introduced the possibility for the bootloader to update itself
–  Fixed compile issues on macOS and WSL
–  Hardware support

•      Added RV32 RISC-V architecture
•      Added hardware-assisted dual-bank support on STM32F76x/77x
•      new HAL: RV32 FE310 (SiFive HiFive-1)
•      new HAL: STM32L0
•      new HAL: STM32G0
•      new HAL: STM32F7
•      new HAL: STM32WB

The latest version of wolfBoot is available for download here: https://wolfssl.com/download/

More information about boot loaders can be found here: https://en.wikipedia.org/wiki/Booting#BOOT-LOADER

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.

Microsoft has created a Linux subsystem that allows running Linux distributions on Windows 10. This blog describes the steps for building and running wolfSSL on WSL.

Installing WSL and Ubuntu 18.04:

1. Open PowerShell as Administrator (right-click and choose “Run as Administrator”)
2. Run the following commands:
   1. Enable-WindowsOptionalFeature -Online -FeatureName Microsoft-Windows-Subsystem-Linux
   2. Invoke-WebRequest -Uri https://aka.ms/wsl-ubuntu-1804 -OutFile Ubuntu.appx -UseBasicParsing
   3. Add-AppxPackage .\Ubuntu.appx
3. Start -> Ubuntu 18.04 (should be at the top of the start menu)
4. Installing, this may take a few minutes…
5. Choose username and password
6. Update Ubuntu:
   1. sudo apt update
   2. sudo apt upgrade

Building wolfSSL:

1. Install prerequisites
   1. sudo apt-get install autoconf libtool make execstack
2. git clone https://github.com/wolfSSL/wolfssl.git
3. cd wolfssl
4. ./autogen.sh
5. ./configure
6. make
7. make check

Enabling Intel Speedups and Running Benchmarks:

1. ./configure –enable-intelasm –enable-aesni –enable-sp –enable-sp-asm
2. make
3. WSL does not currently have support for the PROT_GROWSDOWN flag. Manually clear the executable stack flag in the library:
   1. execstack -c ./src/.libs/libwolfssl.so
4. ./wolfcrypt/benchmark/benchmark

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.

The wolfTPM library has now been tested on the Xilinx Zynq UltraScale with VxWorks. 

Testing was done using a Xilinx Zynq Pmod board with Infineon SLB9670 TPM 2.0 module. 

The TPM 2.0 module can be used for measured boot functionality and cryptographic security to extend the hardware root of trust enabled by Zynq and Zynq UltraScale+ devices.

The wolfTPM library and TPM 2.0 module can be accessed from VxWorks kernel mode and user space. 

More information on the Zynq Pmod module can be found here:

http://www.zedboard.org/product/trusted-platform-module-security-pmod

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.

Affected Users:

User of wolfSSL v4.0.0 are affected.

Summary of issue:

wolfSSL 4.0.0 has a Buffer Overflow in DoPreSharedKeys in tls13.c when a current identity size is greater than a client identity size. The identity data field of the PSK extension of the packet contains data beyond the buffer length to be stored in the undefined memory (RAM) of the server. The size of the data is about 65 kB. Attackers could write about 65 kB of data to the RAM space on affected servers.

Recommendation:

Users should upgrade to wolfSSL v4.1.0 

Research:

https://nvd.nist.gov/vuln/detail/CVE-2019-11873

Additional details:

Patch for this vulnerability can be viewed here: https://github.com/wolfSSL/wolfssl/pull/2239

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.

Related Items:
https://www.wolfssl.com/everything-wanted-know-wolfssl-support-handles-vulnerability-reports-afraid-ask/
https://www.wolfssl.com/docs/security-vulnerabilities/
https://github.com/wolfSSL/wolfssl/pull/2353

Settings that mitigate this vulnerability in affected versions (secp256r1 only, all other curves are affected):

• –enable-sp
• –enable-sp-asm
• –enable-fpecc

Affected Users:

Users with long-term private ECC ECDSA keys performing ECDSA sign operations with the USE_FAST_MATH setting (–enable-fastmath).

Users who have disabled the default enabled timing resistance while also using fastmath (–disable-harden) or using normal math (–disable-fastmath) will continue to be susceptible to timing attacks including this vulnerability.

Summary:

There is a potential leak of nonce sizes when performing ECDSA signing operations. The leak is considered to be difficult to exploit but it could potentially be used to recover private ECC keys in wolfSSL versions prior to release 4.1.0.

Recommendation:

Users with long-term private ECC ECDSA keys, performing ECDSA signing operations with the fastmath library should:

1.  update to wolfSSL version 4.1.0
2. replace any long-term private ECC ECDSA keys.
3. Not disable timing resistance

Research:

The research for this vulnerability is not yet publicly available, a public disclosure containing more details is currently scheduled for September 2nd 2019. CVE-2019-13628 has been reserved for when the public disclosure is made available.

Additional details:

More available upon public disclosure of research. The patch fixing this issue can be viewed at this link: https://github.com/wolfSSL/wolfssl/pull/2353/files

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.

Related Items:
https://www.wolfssl.com/everything-wanted-know-wolfssl-support-handles-vulnerability-reports-afraid-ask/
https://www.wolfssl.com/docs/security-vulnerabilities/
https://github.com/wolfSSL/wolfssl/pull/2353

wolfSSL is excited to announce its summer release of the wolfSSL embedded SSL/TLS library version 4.1.0! As with each release, wolfSSL 4.1.0 comes with many feature additions, bug fixes, and improvements to the wolfSSL library. 

The list below outlines some of the new features and notable fixes added for version 4.1.0:

• Fixes and updates for TLS 1.3:
  • Added additional sanity checks and alert messages for TLS 1.3
  • Major version TLS Draft is now ignored and an alert is sent if version negotiation occurs but no versions were matched
  • Added WOLFSSL_PSK_ONE_ID macro for indicating that only one identity in TLS 1.3 PSK is available and will be cached
  • Added XTIME_MS macro to simplify the tls13.c time requirement
  • Improved and refactored code related to parsing and creating TLS 1.3 client hello packets
  • TLS 1.3 version renegotiation now happens before interpreting ClientHello message
• Fixes and additions for PCKS7:
  • Added a fix for a check on the return value when verifying PKCS7 bundle signatures (users with applications using the function wc_PKCS7_VerifySignedData should update)
  • Added the function wc_PKCS7_GetSignerSID for PKCS7 firmware bundles
  • Added PKCS7 callback functions for unwrapping of CEK and for decryption
• Increased performance on ARM architecture:
  • Optimizations for Poly1305 and SHA-512/384 on ARM architecture using SIMD NEON extension
  • Optimizations for ChaCha20, Curve25519 and Ed 25519 on ARM architecture for performance increase
• Added Sniffer updates:
  • Added support for the null cipher and static ECDH key exchange and new SSLWatchCb callback
  • Added cipher suite TLS_RSA_WITH_NULL_MD5 (off by default)
  • Sniffer statistics print out with the macro WOLFSSL_SNIFFER_STATS defined
• Fixes, updates, and new functions added for OpenSSL Extra
• Added the build flag –enable-ecccustcurves=all to enable all curve types
• Added Java Secure Sockets Extension (JSSE) support
• Added additional TLS alert messages sent with the macro WOLFSSL_EXTRA_ALERTS defined
• Added CryptoCell-310 support on nRF52840
• Added SiFive HiFive E31 RISC?V core family port
• Added Telit IoT AppZone SDK port
• Added the build flag –enable-blake2s for 32-bit Blake2s support
• Added support for Ed25519ctx and Ed25519ph sign/verify algorithms as per RFC 8032

Stay tuned for more information regarding notable features and updates included with the wolfSSL 4.1.0 release. The following lists other various fixes and improvements that have been included with wolfSSL 4.1.0:

• Compile time fixes for build case with SP math and RSA only
• Fixes for Coverity static analysis report including explicit initialization of reported stack variables, as well as additional Coverity fixes thanks to Martin
• Fixes for scan build warnings (i.e possible null dereference in ecc.c)
• Resetting verify send value with a call to wolfSSL_clear function
• Fix for extern with sp_ModExp_2048 when building with –cpp option
• Fix for typo issue with –enable-sp=cortexm
• Adding #pragma warning disable 4127 for tfm.c when building with Visual Studio
• Improvements to the maximum ECC signature calculations
• Improvements to TLS write handling in error cases which helps user application not go through with a wolfSSL_write attempt after a wolfSSL_read failure
• Fix for read directory functions with Windows (wc_ReadDirFirst and wc_ReadDirNext)
• Sanity check on index before accessing domain component buffer in call to wolfSSL_X509_NAME_get_entry
• Sending fatal alert from client side on version error
• Fix for static RSA cipher suite with PK callback and no loaded private key
• Fix for potential memory leak in error case with the function wc_DsaKeyToDer, thanks to Chris H. for the report
• Adjusting STRING_USER macro to remove includes of standard lib <string.h> or <stdio.h>
• Bug fix for checking wrong allocation assignment in the function wc_PBKDF2 and handling potential leak on allocation failure. This case is only hit when the specific call to malloc fails in the function wc_PBKDF2. Thanks to Robert Altnoeder (Linbit) for the report
• Improved length checks when parsing ASN.1 certificates
• Improved checking of return values with TLS extension functions and error codes
• Removing redundant calls to the generate function when instantiating and reseeding DRBG
• Improvements for handling error return codes when reading input from transport layer
• Improvements to efficiency of SNI extension parsing and error checking with ALPN parsing

This release of wolfSSL also includes a fix for 2 security vulnerabilities. A full detailed list of additions and bug fixes can be found in the wolfSSL README. 

To download and view the most recent version of wolfSSL, the wolfSSL GitHub repository can be cloned from here: https://github.com/wolfssl/wolfssl.git, and the most recent stable release can be downloaded from the wolfSSL download page here: https://www.wolfssl.com/download/.

If you have any questions or run into any issues, contact us at facts@wolfssl.com, or call us at +1 425 245 8247.