* common/iobuf.c (iobuf_get_filelength): Change return type to
uint64_t and remove the overflow args. For Windows always use
GetFileSizeEx which is available since the long EOL-ed Windows XP.
* g10/sign.c (write_plaintext_packet): Adjust for changed
iobuf_get_filelength.
* g10/encrypt.c (encrypt_simple, encrypt_crypt): Ditto.
* g10/photoid.c (generate_photo_id): Ditto. Also add an upper limit.
* g10/filter.h (progress_filter_context_t): Change amount values to
use uint64_t.
* g10/progress.c (write_status_progress): Change accordingly.
--
GnuPG-bug-id: 6534
* g10/filter.h (armor_filter_context_t): New fields dearmor_mode and
dearmor_state.
* g10/dearmor.c (dearmor_file): Set dearmor_mode.
* g10/armor.c (is_armor_header): Magic to switch to generic dearmor
mode.
(parse_header_line): Treat non OpenPGP armor in a special way.
(check_input): Ditto.
(radix64_read): Detect non OpenPGP armor END lines.
* g10/armor.c (CRCINIT, CRCPOLY, CRCUPDATE, crc_table): Remove.
(new_armor_context): Open libgcrypt CRC24 context.
(release_armor_context): Close CRC24 context.
(initialize): Remove CRC table generation.
(get_afx_crc): New.
(check_input, fake_packet, radix64_read, armor_filter): Update to use
CRC24 context.
* g10/filter.h (armor_filter_context_t): Replace crc intermediate value
with libgcrypt md context pointer.
--
This patch changes armor filter to use optimized CRC24 implementation
from libgcrypt to speed up encryption and decryption.
Benchmark results below, tested on Intel Core i7-4790K (turbo off).
Encrypted 2 GiB through pipe to ramfs file using AES128. Decrypt
ramfs file out through pipe to /dev/null.
before patch-set
----------------
gpg process
armor: user time pipe transfer rate
encrypt-aead: 13.8 140 MB/s
decrypt-aead: 30.6 68 MB/s
encrypt-cfb: 17.4 114 MB/s
decrypt-cfb: 32.6 64 MB/s
after (decrypt+iobuf+crc opt)
-----------------------------
gpg process
armor: user time pipe transfer rate
encrypt-aead: 8.7 211 MB/s
decrypt-aead: 17.6 116 MB/s
encrypt-cfb: 12.6 153 MB/s
decrypt-cfb: 19.6 105 MB/s
Signed-off-by: Jussi Kivilinna <jussi.kivilinna@iki.fi>
* g10/compress.c (push_compress_filter2): Return an error if no filter
was pushed.
(push_compress_filter): Ditto.
(handle_compressed): Free CFX if no filter was pushed.
* g10/import.c (read_block): Ditto.
--
GnuPG-bug-id: 3898, 3930
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/openpgpdefs.h (PKT_ENCRYPTED_AEAD): New const.
* g10/dek.h (DEK): Increase size of use_aead to 4 bits.
* g10/filter.h (cipher_filter_context_t): Add new fields for AEAD.
* g10/packet.h (PKT_encrypted): Add fields aead_algo, cipher_algo, and
chunkbyte.
* g10/build-packet.c (do_encrypted_aead): New.
(build_packet): Call it.
* g10/parse-packet.c (dump_sig_subpkt): Handle SIGSUBPKT_PREF_AEAD.
(parse_one_sig_subpkt, can_handle_critical): Ditto.
(parse_encrypted): Clear new PKT_ENCRYPTED fields.
(parse_encrypted_aead): New.
(parse): Call it.
* g10/gpg.c (main): Take care of --rfc4880bis option when checking
compliance.
* g10/cipher-aead.c: Replace the stub by real code.
* g10/decrypt-data.c (decode_filter_ctx_t): Add fields for use with
AEAD.
(aead_set_nonce): New.
(aead_set_ad): New.
(decrypt_data): Support AEAD.
(aead_underflow): New.
(aead_decode_filter): New.
* g10/encrypt.c (use_aead): Make that new fucntion work.
(encrypt_simple): Use default_aead_algo() instead of EAX.
* g10/mainproc.c (proc_encrypted): Support AEAD.
(do_proc_packets): Support PKT_ENCRYPTED_AEAD.
--
This code has seen only a very few manual tests. Encrypting always
uses a 64k chunks and decryption has not been tested with larger
chunks. Those small chunks make debugging much faster.
Tests can be done using:
gpg --rfc4880bis --pinentry-mode=loopback --passphrase abc \
--force-aead --aead-algo ocb --s2k-mode 0 --cipher AES \
-v -z 0 --status-fd 2 -c <INFILE >OUTFILE
and
gpg --rfc4880bis --pinentry-mode=loopback --passphrase=abc \
--status-fd 2 -v -d <INFILE >OUTFILE
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/cipher.c (cipher_filter): Rename to cipher_filter_cfb.
* g10/cipher-aead.c: New. Right now only with a stub function.
* g10/Makefile.am (gpg_sources): Add file.
* g10/encrypt.c (encrypt_simple): Push either cipher_filter_cfb or
cipher_filter_aead.
(encrypt_crypt): Ditto.
(encrypt_filter): Ditto.
* g10/sign.c (sign_symencrypt_file): Ditto.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/filter.h (cipher_filter_context_t): Remove unused filed
'create_mdc'. Turn field 'header' into a bit field. Add new fields
'short_blklen_warn' and 'short_blklen_count'.
* g10/cipher.c (write_header): Print a warning if MDC is not used.
(cipher_filter): Print a warning for long messages encrypted with a
short block length algorithm.
--
Note that to test this warning in a reliable way compression needs to
be disabled.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/armor.c (parse_hash_header,carmor_filter): Ignore MD5 in hash
header.
(fake_packet): Remove pgp-2 workaround for white space stripping.
* g10/filter.h (armor_filter_context_t): Remove field pgp2mode.
* g10/options.h (opt): Remove field pgp2_workarounds.
* g10/gpg.c (main): Do not set this field.
* g10/gpgv.c (main): Ditto.
* g10/mainproc.c (proc_encrypted): Use SHA-1 as fallback s2k hash
algo. Using MD5 here is useless.
(proc_plaintext): Remove PGP-2 related woraround
(proc_tree): Remove another workaround but keep the one for PGP-5.
--
The removed code was either not anymore used or its use would have
caused an error message later anyway.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/build-packet.c (do_key): Remove support for building v3 keys.
* g10/parse-packet.c (read_protected_v3_mpi): Remove.
(parse_key): Remove support for v3-keys. Add dedicated warnings for
v3-key packets.
* g10/keyid.c (hash_public_key): Remove v3-key support.
(keyid_from_pk): Ditto.
(fingerprint_from_pk): Ditto.
* g10/options.h (opt): Remove fields force_v3_sigs and force_v4_certs.
* g10/gpg.c (cmd_and_opt_values): Remove oForceV3Sigs, oNoForceV3Sigs,
oForceV4Certs, oNoForceV4Certs.
(opts): Turn --force-v3-sigs, --no-force-v3-sigs, --force-v4-certs,
--no-force-v4-certs int dummy options.
(main): Remove setting of the force_v3_sigs force_v4_certs flags.
* g10/revoke.c (gen_revoke, create_revocation): Always create v4 certs.
* g10/sign.c (hash_uid): Remove support for v3-signatures
(hash_sigversion_to_magic): Ditto.
(only_old_style): Remove this v3-key function.
(write_signature_packets): Remove support for creating v3-signatures.
(sign_file): Ditto.
(sign_symencrypt_file): Ditto.
(clearsign_file): Ditto. Remove code to emit no Hash armor line if
only v3-keys are used.
(make_keysig_packet): Remove arg SIGVERSION and force using
v4-signatures. Change all callers to not pass a value for this arg.
Remove all v3-key related code.
(update_keysig_packet): Remove v3-signature support.
* g10/keyedit.c (sign_uids): Always create v4-signatures.
* g10/textfilter.c (copy_clearsig_text): Remove arg pgp2mode and
change caller.
--
v3 keys are deprecated for about 15 years and due the severe
weaknesses of MD5 it does not make any sense to keep code around to
use these old and broken keys. Users who need to decrypt old messages
should use gpg 1.4 and best re-encrypt them to modern standards.
verification of old (i.e. PGP2) created signatures is thus also not
anymore possible but such signatures have no values anyway - MD5 is
just too broken.
We have also kept support for v3 signatures until now. With the
removal of support for v3 keys it is questionable whether it makes any
sense to keep support for v3-signatures. What we do now is to keep
support for verification of v3-signatures but we force the use of
v4-signatures. The latter makes the --pgp6 and --pgp7 switch a bit
obsolete because those PGP versions require v3-signatures for
messages. These versions of PGP are also really old and not anymore
maintained so they have not received any bug fixes and should not be
used anyway.
Signed-off-by: Werner Koch <wk@gnupg.org>
We better do this once and for all instead of cluttering all future
commits with diffs of trailing white spaces. In the majority of cases
blank or single lines are affected and thus this change won't disturb
a git blame too much. For future commits the pre-commit scripts
checks that this won't happen again.
to libgcrypt functions, using shared error codes from libgpg-error,
replacing the old functions we used to have in ../util by those in
../jnlib and ../common, renaming the malloc functions and a couple of
types. Note, that not all changes are listed below becuause they are
too similar and done at far too many places. As of today the code
builds using the current libgcrypt from CVS but it is very unlikely
that it actually works.
* encode.c (encode_simple,encode_crypt): Use new style CTB for
compressssed packets when using MDC. We need to do this so that
concatenated messages are properly decrypted. Old style
compression assumes that it is the last packet; given that we
can't determine the length in advance, the uncompressor does not
know where to start. Actually we should use the new CTB always
but this would break PGP 2 compatibility.
* parse-packet.c (parse): Special treatment for new style CTB
compressed packets.
* build-packet.c (do_mdc): Removed. Was not used.
(do_encrypted_mdc): Count the version number and the MDC packet.
