I'm really not sure why this string has disappeared from /usr/bin/gpg1.

It was present in 1.4.22-1, and it's still present in the source code.

But it's no longer visible post-compilation of the as-yet-unreleased
1.4.22-2, for whatever reason.

autoreconf should be automatic now, so no need for additional explicit
dependency.

with Debian dir e380fc83ccd168618d12b87d14a99231f97e2fb2

Signed-off-by: Werner Koch <wk@gnupg.org>

--

Signed-off-by: Werner Koch <wk@gnupg.org>

--

* g10/textfilter.c (copy_clearsig_text): Free the buffer.

--

Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>

(backport from master commit:
6b9a89e4c7d6f19de62e0a908a8d80c98bf99819)

* cipher/rsa.c (secret): Don't keep secmem.

Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>

* cipher/Makefile.am (AM_CPPFLAGS): Add mpi dirs.
* cipher/rsa.c: Change include file.

Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>

Marcus Brinkmann authored Jul 07, 2017 and NIIBE Yutaka committed Jul 07, 2017

* cipher/rsa.c (secret_core_crt): Blind secret D with randomized
nonce R for mpi_powm computation.

--

Backport of libgcrypt 8725c99ffa41778f382ca97233183bcd687bb0ce.

Signed-off-by: Marcus Brinkmann <mb@g10code.com>

* mpi/mpi-pow.c (mpi_powm): Fix allocation size.

Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>

* mpi/mpi-pow.c (_gcry_mpi_powm): Compare msize for max_u_size.  Move
the assignment to base_u into the loop.  Copy content refered by RP to
BASE_U except the last of the loop.

--

Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>

(backport commit of libgcrypt master:
78130828e9a140a9de4dafadbc844dbb64cb709a)

* mpi/mpi-pow.c (_gcry_mpi_powm): Simplify the loop.

--

(backport of libgcrypt master commit:
 719468e53133d3bdf12156c5bfdea2bf15f9f6f1)

Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>

* mpi/longlong.h [__arm__] (add_ssaaaa, sub_ddmmss): Add __CLOBBER_CC.
[__arm__][__ARM_ARCH <= 3] (umul_ppmm): Add __AND_CLOBBER_CC.

--
This is a backport of libgcrypt 8aa4f2161 and 3b1cc9e6c.

Signed-off-by: Marcus Brinkmann <mb@g10code.com>
GnuPG-bug-id: 3182

Signed-off-by: Marcus Brinkmann <mb@g10code.com>
GnuPG-bug-id: 3243

Ineiev authored Apr 14, 2016 and NIIBE Yutaka committed May 10, 2017

* g10/keygen.c (proc_parameter_file): Fix secmem leak.

--

proc_parameter_file adds certain parameters to the list in the PARA
argument; however, these new entries are leaked because they
are added to head, while the PARA list is released by the caller
of proc_parameter_file.

GnuPG-bug-id: 1371
Signed-off-by: Ineiev <ineiev@gnu.org>

* g10/build-packet.c (do_user_id): Avoid indeterminate length header.
--

We are able to import such user ids but when exporting them the
exported data could not be imported again because the parser bails out
on invalid keyrings.  This is now fixed and should be backported.

Note that in 1.4 and 2.0 this is only an issue for attribute packets.
In 2.1 user IDs were also affected.a

Signed-off-by: Werner Koch <wk@gnupg.org>