Thank you for the report and analysis. We will need to fix this everywhere where present. The refactoring was done due to need to fix RAND locking issues in the FIPS mode - deadlocks or no locking where needed. Unfortunately I made this mistake in the process. The refactoring was later dropped because upstream fixed the FIPS RNG locking in a different way.

The only thing I am curious about is why nobody reported the regression earlier - especially on RHEL.

I can only imagine that using openssl to implement multi-threaded TLS servers is somewhat of a rare use case. (Most of the off-the-shelf uses prefer to run multi-process single-threaded.)

(The other possibility is that even in that case, as I hinted above, the exact optimised form of the hash may be important. In the SSSE3 case the inner hash block loop is formulated as a loop from a start pointer to a specific end pointer value, so the negative length causes an unchecked wrap around the address space. Other forms may check the length value directly against an index, so under signed arithmetic we simply terminate the loop early. That might mean less good randomness is produced, but won't crash.)

This message is a reminder that Fedora 20 is nearing its end of life.
Approximately 4 (four) weeks from now Fedora will stop maintaining
and issuing updates for Fedora 20. It is Fedora's policy to close all
bug reports from releases that are no longer maintained. At that time
this bug will be closed as EOL if it remains open with a Fedora  'version'
of '20'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version' 
to a later Fedora version.

Thank you for reporting this issue and we are sorry that we were not 
able to fix it before Fedora 20 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora, you are encouraged  change the 'version' to a later Fedora 
version prior this bug is closed as described in the policy above.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events. Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.

Fedora 22 is not affected.

openssl-1.0.1e-43.fc20 has been submitted as an update for Fedora 20.
https://admin.fedoraproject.org/updates/openssl-1.0.1e-43.fc20

(In reply to John Sullivan from comment #0)
> Still present (I believe) in RH6/RH7/FC20. (Raising against Fedora because I
> currently use FC20 as my main desktop OS, though others will consider RH the
> more important packages.)

I'm trying to reproduce this with the stress tester below.  I don't see the issue.  Is there something else I need to configure to expose it?

I do see a data races with Helgrind, though, even on FC22, which supposedly has been fixed.


#include <openssl/crypto.h>
#include <openssl/rand.h>
#include <pthread.h>
#include <stdio.h>
#include <stdlib.h>
#include <stdint.h>
#include <stdbool.h>

static pthread_mutex_t *locks;

static void
lock_callback(int mode, int type, const char *file, int line)
{
  if (mode & CRYPTO_LOCK) {
    if (pthread_mutex_lock(locks + type) != 0) {
      abort();
    }
  } else {
    if (pthread_mutex_unlock(locks + type) != 0) {
      abort();
    }
  }
}

static void
init_locks(void)
{
  int count = CRYPTO_num_locks();
  locks = calloc(count, sizeof(*locks));
  for (int i = 0; i < count; ++i) {
    if (pthread_mutex_init(locks + i, NULL) != 0) {
      abort();
    }
  }
  CRYPTO_set_locking_callback(lock_callback);
}

static void *
runner(void *closure)
{
  uintptr_t size = (uintptr_t) closure;
  unsigned char *buffer = malloc(size);
  if (buffer == NULL) {
    abort();
  }
  while (true) {
    RAND_bytes(buffer, size);
  }
}



int
main(int argc, char **argv)
{
  int threads = atoi(argv[1]);
  int start_size = atoi(argv[2]);
  int size_increment = atoi(argv[3]);
  init_locks();

  for (int i = 1; i < threads; ++i) {
    pthread_t thread;
    if (pthread_create(&thread, NULL, runner,
		       (void *)(uintptr_t) start_size + i * size_increment) != 0) {
      abort();
    }
  }
  runner((void *)(uintptr_t) start_size);
}

(In reply to Florian Weimer from comment #7)
> (In reply to John Sullivan from comment #0)
> > Still present (I believe) in RH6/RH7/FC20. (Raising against Fedora because I
> > currently use FC20 as my main desktop OS, though others will consider RH the
> > more important packages.)
> 
> I'm trying to reproduce this with the stress tester below.  I don't see the
> issue.  Is there something else I need to configure to expose it?

I have been able to reproduce it with openssl-1.0.1e-30.el6.ppc64 and openssl-1.0.1e-30.el6.x86_64.  Thanks.

This issue has been classified as a security flaw and hence been assigned CVE-2015-3216

Package openssl-1.0.1e-43.fc20:
* should fix your issue,
* was pushed to the Fedora 20 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing openssl-1.0.1e-43.fc20'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2015-9371/openssl-1.0.1e-43.fc20
then log in and leave karma (feedback).

Package openssl-1.0.1e-44.fc20:
* should fix your issue,
* was pushed to the Fedora 20 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing openssl-1.0.1e-44.fc20'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2015-9371/openssl-1.0.1e-44.fc20
then log in and leave karma (feedback).

Package openssl-1.0.1e-45.fc20:
* should fix your issue,
* was pushed to the Fedora 20 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing openssl-1.0.1e-45.fc20'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2015-9371/openssl-1.0.1e-45.fc20
then log in and leave karma (feedback).

Fedora 20 changed to end-of-life (EOL) status on 2015-06-23. Fedora 20 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release. If you experience problems, please add a comment to this
bug.

Thank you for reporting this bug and we are sorry it could not be fixed.