git/gpg-interface.c

970 lines
26 KiB
C
Raw Normal View History

#include "cache.h"
#include "commit.h"
#include "config.h"
#include "run-command.h"
#include "strbuf.h"
#include "dir.h"
#include "gpg-interface.h"
#include "sigchain.h"
#include "tempfile.h"
#include "alias.h"
static char *configured_signing_key;
static const char *ssh_default_key_command, *ssh_allowed_signers, *ssh_revocation_file;
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
static enum signature_trust_level configured_min_trust_level = TRUST_UNDEFINED;
struct gpg_format {
const char *name;
const char *program;
const char **verify_args;
const char **sigs;
int (*verify_signed_buffer)(struct signature_check *sigc,
struct gpg_format *fmt, const char *payload,
size_t payload_size, const char *signature,
size_t signature_size);
int (*sign_buffer)(struct strbuf *buffer, struct strbuf *signature,
const char *signing_key);
const char *(*get_default_key)(void);
const char *(*get_key_id)(void);
};
static const char *openpgp_verify_args[] = {
"--keyid-format=long",
NULL
};
static const char *openpgp_sigs[] = {
"-----BEGIN PGP SIGNATURE-----",
"-----BEGIN PGP MESSAGE-----",
NULL
};
static const char *x509_verify_args[] = {
NULL
};
static const char *x509_sigs[] = {
"-----BEGIN SIGNED MESSAGE-----",
NULL
};
static const char *ssh_verify_args[] = { NULL };
static const char *ssh_sigs[] = {
"-----BEGIN SSH SIGNATURE-----",
NULL
};
static int verify_gpg_signed_buffer(struct signature_check *sigc,
struct gpg_format *fmt, const char *payload,
size_t payload_size, const char *signature,
size_t signature_size);
static int verify_ssh_signed_buffer(struct signature_check *sigc,
struct gpg_format *fmt, const char *payload,
size_t payload_size, const char *signature,
size_t signature_size);
static int sign_buffer_gpg(struct strbuf *buffer, struct strbuf *signature,
const char *signing_key);
static int sign_buffer_ssh(struct strbuf *buffer, struct strbuf *signature,
const char *signing_key);
static const char *get_default_ssh_signing_key(void);
static const char *get_ssh_key_id(void);
static struct gpg_format gpg_format[] = {
{
.name = "openpgp",
.program = "gpg",
.verify_args = openpgp_verify_args,
.sigs = openpgp_sigs,
.verify_signed_buffer = verify_gpg_signed_buffer,
.sign_buffer = sign_buffer_gpg,
.get_default_key = NULL,
.get_key_id = NULL,
},
{
.name = "x509",
.program = "gpgsm",
.verify_args = x509_verify_args,
.sigs = x509_sigs,
.verify_signed_buffer = verify_gpg_signed_buffer,
.sign_buffer = sign_buffer_gpg,
.get_default_key = NULL,
.get_key_id = NULL,
},
{
.name = "ssh",
.program = "ssh-keygen",
.verify_args = ssh_verify_args,
.sigs = ssh_sigs,
.verify_signed_buffer = verify_ssh_signed_buffer,
.sign_buffer = sign_buffer_ssh,
.get_default_key = get_default_ssh_signing_key,
.get_key_id = get_ssh_key_id,
},
};
static struct gpg_format *use_format = &gpg_format[0];
static struct gpg_format *get_format_by_name(const char *str)
{
int i;
for (i = 0; i < ARRAY_SIZE(gpg_format); i++)
if (!strcmp(gpg_format[i].name, str))
return gpg_format + i;
return NULL;
}
static struct gpg_format *get_format_by_sig(const char *sig)
{
int i, j;
for (i = 0; i < ARRAY_SIZE(gpg_format); i++)
for (j = 0; gpg_format[i].sigs[j]; j++)
if (starts_with(sig, gpg_format[i].sigs[j]))
return gpg_format + i;
return NULL;
}
void signature_check_clear(struct signature_check *sigc)
{
FREE_AND_NULL(sigc->payload);
FREE_AND_NULL(sigc->output);
FREE_AND_NULL(sigc->gpg_status);
FREE_AND_NULL(sigc->signer);
FREE_AND_NULL(sigc->key);
FREE_AND_NULL(sigc->fingerprint);
FREE_AND_NULL(sigc->primary_key_fingerprint);
}
/* An exclusive status -- only one of them can appear in output */
#define GPG_STATUS_EXCLUSIVE (1<<0)
/* The status includes key identifier */
#define GPG_STATUS_KEYID (1<<1)
/* The status includes user identifier */
#define GPG_STATUS_UID (1<<2)
/* The status includes key fingerprints */
#define GPG_STATUS_FINGERPRINT (1<<3)
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
/* The status includes trust level */
#define GPG_STATUS_TRUST_LEVEL (1<<4)
/* Short-hand for standard exclusive *SIG status with keyid & UID */
#define GPG_STATUS_STDSIG (GPG_STATUS_EXCLUSIVE|GPG_STATUS_KEYID|GPG_STATUS_UID)
static struct {
char result;
const char *check;
unsigned int flags;
} sigcheck_gpg_status[] = {
{ 'G', "GOODSIG ", GPG_STATUS_STDSIG },
{ 'B', "BADSIG ", GPG_STATUS_STDSIG },
{ 'E', "ERRSIG ", GPG_STATUS_EXCLUSIVE|GPG_STATUS_KEYID },
{ 'X', "EXPSIG ", GPG_STATUS_STDSIG },
{ 'Y', "EXPKEYSIG ", GPG_STATUS_STDSIG },
{ 'R', "REVKEYSIG ", GPG_STATUS_STDSIG },
{ 0, "VALIDSIG ", GPG_STATUS_FINGERPRINT },
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
{ 0, "TRUST_", GPG_STATUS_TRUST_LEVEL },
};
static struct {
const char *key;
enum signature_trust_level value;
} sigcheck_gpg_trust_level[] = {
{ "UNDEFINED", TRUST_UNDEFINED },
{ "NEVER", TRUST_NEVER },
{ "MARGINAL", TRUST_MARGINAL },
{ "FULLY", TRUST_FULLY },
{ "ULTIMATE", TRUST_ULTIMATE },
};
static void replace_cstring(char **field, const char *line, const char *next)
{
free(*field);
if (line && next)
*field = xmemdupz(line, next - line);
else
*field = NULL;
}
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
static int parse_gpg_trust_level(const char *level,
enum signature_trust_level *res)
{
size_t i;
for (i = 0; i < ARRAY_SIZE(sigcheck_gpg_trust_level); i++) {
if (!strcmp(sigcheck_gpg_trust_level[i].key, level)) {
*res = sigcheck_gpg_trust_level[i].value;
return 0;
}
}
return 1;
}
static void parse_gpg_output(struct signature_check *sigc)
{
const char *buf = sigc->gpg_status;
const char *line, *next;
int i, j;
int seen_exclusive_status = 0;
/* Iterate over all lines */
for (line = buf; *line; line = strchrnul(line+1, '\n')) {
while (*line == '\n')
line++;
if (!*line)
break;
/* Skip lines that don't start with GNUPG status */
if (!skip_prefix(line, "[GNUPG:] ", &line))
continue;
/* Iterate over all search strings */
for (i = 0; i < ARRAY_SIZE(sigcheck_gpg_status); i++) {
if (skip_prefix(line, sigcheck_gpg_status[i].check, &line)) {
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
/*
* GOODSIG, BADSIG etc. can occur only once for
* each signature. Therefore, if we had more
* than one then we're dealing with multiple
* signatures. We don't support them
* currently, and they're rather hard to
* create, so something is likely fishy and we
* should reject them altogether.
*/
if (sigcheck_gpg_status[i].flags & GPG_STATUS_EXCLUSIVE) {
if (seen_exclusive_status++)
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
goto error;
}
if (sigcheck_gpg_status[i].result)
sigc->result = sigcheck_gpg_status[i].result;
/* Do we have key information? */
if (sigcheck_gpg_status[i].flags & GPG_STATUS_KEYID) {
next = strchrnul(line, ' ');
replace_cstring(&sigc->key, line, next);
/* Do we have signer information? */
if (*next && (sigcheck_gpg_status[i].flags & GPG_STATUS_UID)) {
line = next + 1;
next = strchrnul(line, '\n');
replace_cstring(&sigc->signer, line, next);
}
}
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
/* Do we have trust level? */
if (sigcheck_gpg_status[i].flags & GPG_STATUS_TRUST_LEVEL) {
/*
* GPG v1 and v2 differs in how the
* TRUST_ lines are written. Some
* trust lines contain no additional
* space-separated information for v1.
*/
size_t trust_size = strcspn(line, " \n");
char *trust = xmemdupz(line, trust_size);
if (parse_gpg_trust_level(trust, &sigc->trust_level)) {
free(trust);
goto error;
}
free(trust);
}
/* Do we have fingerprint? */
if (sigcheck_gpg_status[i].flags & GPG_STATUS_FINGERPRINT) {
gpg-interface: limit search for primary key fingerprint The VALIDSIG status line from GnuPG with --status-fd is documented to have 9 required and 1 optional fields [1]. The final, and optional, field is used to specify the fingerprint of the primary key that made the signature in case it was made by a subkey. However, this field is only available for OpenPGP signatures; not for CMS/X.509. If the VALIDSIG status line does not have the optional 10th field, the current code will continue reading onto the next status line. And this is the case for non-OpenPGP signatures [1]. The consequence is that a subsequent status line may be considered as the "primary key" for signatures that does not have an actual primary key. Limit the search of these 9 or 10 fields to the single line to avoid this problem. If the 10th field is missing, report that there is no primary key fingerprint. [Reference] [1] GnuPG Details, General status codes https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/DETAILS;h=6ce340e8c04794add995e84308bb3091450bd28f;hb=HEAD#l483 The documentation says: VALIDSIG <args> The args are: - <fingerprint_in_hex> - <sig_creation_date> - <sig-timestamp> - <expire-timestamp> - <sig-version> - <reserved> - <pubkey-algo> - <hash-algo> - <sig-class> - [ <primary-key-fpr> ] This status indicates that the signature is cryptographically valid. [...] PRIMARY-KEY-FPR is the fingerprint of the primary key or identical to the first argument. The primary-key-fpr parameter is used for OpenPGP and not available for CMS signatures. [...] Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-11-23 04:23:12 +08:00
const char *limit;
char **field;
next = strchrnul(line, ' ');
replace_cstring(&sigc->fingerprint, line, next);
gpg-interface: limit search for primary key fingerprint The VALIDSIG status line from GnuPG with --status-fd is documented to have 9 required and 1 optional fields [1]. The final, and optional, field is used to specify the fingerprint of the primary key that made the signature in case it was made by a subkey. However, this field is only available for OpenPGP signatures; not for CMS/X.509. If the VALIDSIG status line does not have the optional 10th field, the current code will continue reading onto the next status line. And this is the case for non-OpenPGP signatures [1]. The consequence is that a subsequent status line may be considered as the "primary key" for signatures that does not have an actual primary key. Limit the search of these 9 or 10 fields to the single line to avoid this problem. If the 10th field is missing, report that there is no primary key fingerprint. [Reference] [1] GnuPG Details, General status codes https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/DETAILS;h=6ce340e8c04794add995e84308bb3091450bd28f;hb=HEAD#l483 The documentation says: VALIDSIG <args> The args are: - <fingerprint_in_hex> - <sig_creation_date> - <sig-timestamp> - <expire-timestamp> - <sig-version> - <reserved> - <pubkey-algo> - <hash-algo> - <sig-class> - [ <primary-key-fpr> ] This status indicates that the signature is cryptographically valid. [...] PRIMARY-KEY-FPR is the fingerprint of the primary key or identical to the first argument. The primary-key-fpr parameter is used for OpenPGP and not available for CMS signatures. [...] Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-11-23 04:23:12 +08:00
/*
* Skip interim fields. The search is
* limited to the same line since only
* OpenPGP signatures has a field with
* the primary fingerprint.
*/
limit = strchrnul(line, '\n');
for (j = 9; j > 0; j--) {
gpg-interface: limit search for primary key fingerprint The VALIDSIG status line from GnuPG with --status-fd is documented to have 9 required and 1 optional fields [1]. The final, and optional, field is used to specify the fingerprint of the primary key that made the signature in case it was made by a subkey. However, this field is only available for OpenPGP signatures; not for CMS/X.509. If the VALIDSIG status line does not have the optional 10th field, the current code will continue reading onto the next status line. And this is the case for non-OpenPGP signatures [1]. The consequence is that a subsequent status line may be considered as the "primary key" for signatures that does not have an actual primary key. Limit the search of these 9 or 10 fields to the single line to avoid this problem. If the 10th field is missing, report that there is no primary key fingerprint. [Reference] [1] GnuPG Details, General status codes https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/DETAILS;h=6ce340e8c04794add995e84308bb3091450bd28f;hb=HEAD#l483 The documentation says: VALIDSIG <args> The args are: - <fingerprint_in_hex> - <sig_creation_date> - <sig-timestamp> - <expire-timestamp> - <sig-version> - <reserved> - <pubkey-algo> - <hash-algo> - <sig-class> - [ <primary-key-fpr> ] This status indicates that the signature is cryptographically valid. [...] PRIMARY-KEY-FPR is the fingerprint of the primary key or identical to the first argument. The primary-key-fpr parameter is used for OpenPGP and not available for CMS signatures. [...] Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-11-23 04:23:12 +08:00
if (!*next || limit <= next)
break;
line = next + 1;
next = strchrnul(line, ' ');
}
gpg-interface: limit search for primary key fingerprint The VALIDSIG status line from GnuPG with --status-fd is documented to have 9 required and 1 optional fields [1]. The final, and optional, field is used to specify the fingerprint of the primary key that made the signature in case it was made by a subkey. However, this field is only available for OpenPGP signatures; not for CMS/X.509. If the VALIDSIG status line does not have the optional 10th field, the current code will continue reading onto the next status line. And this is the case for non-OpenPGP signatures [1]. The consequence is that a subsequent status line may be considered as the "primary key" for signatures that does not have an actual primary key. Limit the search of these 9 or 10 fields to the single line to avoid this problem. If the 10th field is missing, report that there is no primary key fingerprint. [Reference] [1] GnuPG Details, General status codes https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/DETAILS;h=6ce340e8c04794add995e84308bb3091450bd28f;hb=HEAD#l483 The documentation says: VALIDSIG <args> The args are: - <fingerprint_in_hex> - <sig_creation_date> - <sig-timestamp> - <expire-timestamp> - <sig-version> - <reserved> - <pubkey-algo> - <hash-algo> - <sig-class> - [ <primary-key-fpr> ] This status indicates that the signature is cryptographically valid. [...] PRIMARY-KEY-FPR is the fingerprint of the primary key or identical to the first argument. The primary-key-fpr parameter is used for OpenPGP and not available for CMS signatures. [...] Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-11-23 04:23:12 +08:00
field = &sigc->primary_key_fingerprint;
if (!j) {
next = strchrnul(line, '\n');
replace_cstring(field, line, next);
} else {
replace_cstring(field, NULL, NULL);
}
}
break;
}
}
}
return;
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
error:
sigc->result = 'E';
/* Clear partial data to avoid confusion */
FREE_AND_NULL(sigc->primary_key_fingerprint);
FREE_AND_NULL(sigc->fingerprint);
FREE_AND_NULL(sigc->signer);
FREE_AND_NULL(sigc->key);
}
static int verify_gpg_signed_buffer(struct signature_check *sigc,
struct gpg_format *fmt, const char *payload,
size_t payload_size, const char *signature,
size_t signature_size)
gpg-interface: prefer check_signature() for GPG verification This commit refactors the use of verify_signed_buffer() outside of gpg-interface.c to use check_signature() instead. It also turns verify_signed_buffer() into a file-local function since it's now only invoked internally by check_signature(). There were previously two globally scoped functions used in different parts of Git to perform GPG signature verification: verify_signed_buffer() and check_signature(). Now only check_signature() is used. The verify_signed_buffer() function doesn't guard against duplicate signatures as described by Michał Górny [1]. Instead it only ensures a non-erroneous exit code from GPG and the presence of at least one GOODSIG status field. This stands in contrast with check_signature() that returns an error if more than one signature is encountered. The lower degree of verification makes the use of verify_signed_buffer() problematic if callers don't parse and validate the various parts of the GPG status message themselves. And processing these messages seems like a task that should be reserved to gpg-interface.c with the function check_signature(). Furthermore, the use of verify_signed_buffer() makes it difficult to introduce new functionality that relies on the content of the GPG status lines. Now all operations that does signature verification share a single entry point to gpg-interface.c. This makes it easier to propagate changed or additional functionality in GPG signature verification to all parts of Git, without having odd edge-cases that don't perform the same degree of verification. [1] https://dev.gentoo.org/~mgorny/articles/attack-on-git-signature-verification.html Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-03-04 19:48:04 +08:00
{
struct child_process gpg = CHILD_PROCESS_INIT;
struct tempfile *temp;
int ret;
struct strbuf gpg_stdout = STRBUF_INIT;
struct strbuf gpg_stderr = STRBUF_INIT;
gpg-interface: prefer check_signature() for GPG verification This commit refactors the use of verify_signed_buffer() outside of gpg-interface.c to use check_signature() instead. It also turns verify_signed_buffer() into a file-local function since it's now only invoked internally by check_signature(). There were previously two globally scoped functions used in different parts of Git to perform GPG signature verification: verify_signed_buffer() and check_signature(). Now only check_signature() is used. The verify_signed_buffer() function doesn't guard against duplicate signatures as described by Michał Górny [1]. Instead it only ensures a non-erroneous exit code from GPG and the presence of at least one GOODSIG status field. This stands in contrast with check_signature() that returns an error if more than one signature is encountered. The lower degree of verification makes the use of verify_signed_buffer() problematic if callers don't parse and validate the various parts of the GPG status message themselves. And processing these messages seems like a task that should be reserved to gpg-interface.c with the function check_signature(). Furthermore, the use of verify_signed_buffer() makes it difficult to introduce new functionality that relies on the content of the GPG status lines. Now all operations that does signature verification share a single entry point to gpg-interface.c. This makes it easier to propagate changed or additional functionality in GPG signature verification to all parts of Git, without having odd edge-cases that don't perform the same degree of verification. [1] https://dev.gentoo.org/~mgorny/articles/attack-on-git-signature-verification.html Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-03-04 19:48:04 +08:00
temp = mks_tempfile_t(".git_vtag_tmpXXXXXX");
if (!temp)
return error_errno(_("could not create temporary file"));
if (write_in_full(temp->fd, signature, signature_size) < 0 ||
close_tempfile_gently(temp) < 0) {
error_errno(_("failed writing detached signature to '%s'"),
temp->filename.buf);
delete_tempfile(&temp);
return -1;
}
strvec_push(&gpg.args, fmt->program);
strvec_pushv(&gpg.args, fmt->verify_args);
strvec_pushl(&gpg.args,
"--status-fd=1",
"--verify", temp->filename.buf, "-",
NULL);
gpg-interface: prefer check_signature() for GPG verification This commit refactors the use of verify_signed_buffer() outside of gpg-interface.c to use check_signature() instead. It also turns verify_signed_buffer() into a file-local function since it's now only invoked internally by check_signature(). There were previously two globally scoped functions used in different parts of Git to perform GPG signature verification: verify_signed_buffer() and check_signature(). Now only check_signature() is used. The verify_signed_buffer() function doesn't guard against duplicate signatures as described by Michał Górny [1]. Instead it only ensures a non-erroneous exit code from GPG and the presence of at least one GOODSIG status field. This stands in contrast with check_signature() that returns an error if more than one signature is encountered. The lower degree of verification makes the use of verify_signed_buffer() problematic if callers don't parse and validate the various parts of the GPG status message themselves. And processing these messages seems like a task that should be reserved to gpg-interface.c with the function check_signature(). Furthermore, the use of verify_signed_buffer() makes it difficult to introduce new functionality that relies on the content of the GPG status lines. Now all operations that does signature verification share a single entry point to gpg-interface.c. This makes it easier to propagate changed or additional functionality in GPG signature verification to all parts of Git, without having odd edge-cases that don't perform the same degree of verification. [1] https://dev.gentoo.org/~mgorny/articles/attack-on-git-signature-verification.html Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-03-04 19:48:04 +08:00
sigchain_push(SIGPIPE, SIG_IGN);
ret = pipe_command(&gpg, payload, payload_size, &gpg_stdout, 0,
&gpg_stderr, 0);
gpg-interface: prefer check_signature() for GPG verification This commit refactors the use of verify_signed_buffer() outside of gpg-interface.c to use check_signature() instead. It also turns verify_signed_buffer() into a file-local function since it's now only invoked internally by check_signature(). There were previously two globally scoped functions used in different parts of Git to perform GPG signature verification: verify_signed_buffer() and check_signature(). Now only check_signature() is used. The verify_signed_buffer() function doesn't guard against duplicate signatures as described by Michał Górny [1]. Instead it only ensures a non-erroneous exit code from GPG and the presence of at least one GOODSIG status field. This stands in contrast with check_signature() that returns an error if more than one signature is encountered. The lower degree of verification makes the use of verify_signed_buffer() problematic if callers don't parse and validate the various parts of the GPG status message themselves. And processing these messages seems like a task that should be reserved to gpg-interface.c with the function check_signature(). Furthermore, the use of verify_signed_buffer() makes it difficult to introduce new functionality that relies on the content of the GPG status lines. Now all operations that does signature verification share a single entry point to gpg-interface.c. This makes it easier to propagate changed or additional functionality in GPG signature verification to all parts of Git, without having odd edge-cases that don't perform the same degree of verification. [1] https://dev.gentoo.org/~mgorny/articles/attack-on-git-signature-verification.html Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-03-04 19:48:04 +08:00
sigchain_pop(SIGPIPE);
delete_tempfile(&temp);
ret |= !strstr(gpg_stdout.buf, "\n[GNUPG:] GOODSIG ");
sigc->payload = xmemdupz(payload, payload_size);
sigc->output = strbuf_detach(&gpg_stderr, NULL);
sigc->gpg_status = strbuf_detach(&gpg_stdout, NULL);
parse_gpg_output(sigc);
strbuf_release(&gpg_stdout);
strbuf_release(&gpg_stderr);
gpg-interface: prefer check_signature() for GPG verification This commit refactors the use of verify_signed_buffer() outside of gpg-interface.c to use check_signature() instead. It also turns verify_signed_buffer() into a file-local function since it's now only invoked internally by check_signature(). There were previously two globally scoped functions used in different parts of Git to perform GPG signature verification: verify_signed_buffer() and check_signature(). Now only check_signature() is used. The verify_signed_buffer() function doesn't guard against duplicate signatures as described by Michał Górny [1]. Instead it only ensures a non-erroneous exit code from GPG and the presence of at least one GOODSIG status field. This stands in contrast with check_signature() that returns an error if more than one signature is encountered. The lower degree of verification makes the use of verify_signed_buffer() problematic if callers don't parse and validate the various parts of the GPG status message themselves. And processing these messages seems like a task that should be reserved to gpg-interface.c with the function check_signature(). Furthermore, the use of verify_signed_buffer() makes it difficult to introduce new functionality that relies on the content of the GPG status lines. Now all operations that does signature verification share a single entry point to gpg-interface.c. This makes it easier to propagate changed or additional functionality in GPG signature verification to all parts of Git, without having odd edge-cases that don't perform the same degree of verification. [1] https://dev.gentoo.org/~mgorny/articles/attack-on-git-signature-verification.html Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-03-04 19:48:04 +08:00
return ret;
}
static void parse_ssh_output(struct signature_check *sigc)
{
const char *line, *principal, *search;
char *to_free;
char *key = NULL;
/*
* ssh-keygen output should be:
* Good "git" signature for PRINCIPAL with RSA key SHA256:FINGERPRINT
*
* or for valid but unknown keys:
* Good "git" signature with RSA key SHA256:FINGERPRINT
*
* Note that "PRINCIPAL" can contain whitespace, "RSA" and
* "SHA256" part could be a different token that names of
* the algorithms used, and "FINGERPRINT" is a hexadecimal
* string. By finding the last occurence of " with ", we can
* reliably parse out the PRINCIPAL.
*/
sigc->result = 'B';
sigc->trust_level = TRUST_NEVER;
line = to_free = xmemdupz(sigc->output, strcspn(sigc->output, "\n"));
if (skip_prefix(line, "Good \"git\" signature for ", &line)) {
/* Valid signature and known principal */
sigc->result = 'G';
sigc->trust_level = TRUST_FULLY;
/* Search for the last "with" to get the full principal */
principal = line;
do {
search = strstr(line, " with ");
if (search)
line = search + 1;
} while (search != NULL);
sigc->signer = xmemdupz(principal, line - principal - 1);
} else if (skip_prefix(line, "Good \"git\" signature with ", &line)) {
/* Valid signature, but key unknown */
sigc->result = 'G';
sigc->trust_level = TRUST_UNDEFINED;
} else {
goto cleanup;
}
key = strstr(line, "key");
if (key) {
sigc->fingerprint = xstrdup(strstr(line, "key") + 4);
sigc->key = xstrdup(sigc->fingerprint);
} else {
/*
* Output did not match what we expected
* Treat the signature as bad
*/
sigc->result = 'B';
}
cleanup:
free(to_free);
}
static int verify_ssh_signed_buffer(struct signature_check *sigc,
struct gpg_format *fmt, const char *payload,
size_t payload_size, const char *signature,
size_t signature_size)
{
struct child_process ssh_keygen = CHILD_PROCESS_INIT;
struct tempfile *buffer_file;
int ret = -1;
const char *line;
size_t trust_size;
char *principal;
struct strbuf ssh_principals_out = STRBUF_INIT;
struct strbuf ssh_principals_err = STRBUF_INIT;
struct strbuf ssh_keygen_out = STRBUF_INIT;
struct strbuf ssh_keygen_err = STRBUF_INIT;
if (!ssh_allowed_signers) {
error(_("gpg.ssh.allowedSignersFile needs to be configured and exist for ssh signature verification"));
return -1;
}
buffer_file = mks_tempfile_t(".git_vtag_tmpXXXXXX");
if (!buffer_file)
return error_errno(_("could not create temporary file"));
if (write_in_full(buffer_file->fd, signature, signature_size) < 0 ||
close_tempfile_gently(buffer_file) < 0) {
error_errno(_("failed writing detached signature to '%s'"),
buffer_file->filename.buf);
delete_tempfile(&buffer_file);
return -1;
}
/* Find the principal from the signers */
strvec_pushl(&ssh_keygen.args, fmt->program,
"-Y", "find-principals",
"-f", ssh_allowed_signers,
"-s", buffer_file->filename.buf,
NULL);
ret = pipe_command(&ssh_keygen, NULL, 0, &ssh_principals_out, 0,
&ssh_principals_err, 0);
if (ret && strstr(ssh_principals_err.buf, "usage:")) {
error(_("ssh-keygen -Y find-principals/verify is needed for ssh signature verification (available in openssh version 8.2p1+)"));
goto out;
}
if (ret || !ssh_principals_out.len) {
/*
* We did not find a matching principal in the allowedSigners
* Check without validation
*/
child_process_init(&ssh_keygen);
strvec_pushl(&ssh_keygen.args, fmt->program,
"-Y", "check-novalidate",
"-n", "git",
"-s", buffer_file->filename.buf,
NULL);
pipe_command(&ssh_keygen, payload, payload_size,
&ssh_keygen_out, 0, &ssh_keygen_err, 0);
/*
* Fail on unknown keys
* we still call check-novalidate to display the signature info
*/
ret = -1;
} else {
/* Check every principal we found (one per line) */
for (line = ssh_principals_out.buf; *line;
line = strchrnul(line + 1, '\n')) {
while (*line == '\n')
line++;
if (!*line)
break;
trust_size = strcspn(line, "\n");
principal = xmemdupz(line, trust_size);
child_process_init(&ssh_keygen);
strbuf_release(&ssh_keygen_out);
strbuf_release(&ssh_keygen_err);
strvec_push(&ssh_keygen.args, fmt->program);
/*
* We found principals
* Try with each until we find a match
*/
strvec_pushl(&ssh_keygen.args, "-Y", "verify",
"-n", "git",
"-f", ssh_allowed_signers,
"-I", principal,
"-s", buffer_file->filename.buf,
NULL);
if (ssh_revocation_file) {
if (file_exists(ssh_revocation_file)) {
strvec_pushl(&ssh_keygen.args, "-r",
ssh_revocation_file, NULL);
} else {
warning(_("ssh signing revocation file configured but not found: %s"),
ssh_revocation_file);
}
}
sigchain_push(SIGPIPE, SIG_IGN);
ret = pipe_command(&ssh_keygen, payload, payload_size,
&ssh_keygen_out, 0, &ssh_keygen_err, 0);
sigchain_pop(SIGPIPE);
FREE_AND_NULL(principal);
if (!ret)
ret = !starts_with(ssh_keygen_out.buf, "Good");
if (!ret)
break;
}
}
sigc->payload = xmemdupz(payload, payload_size);
strbuf_stripspace(&ssh_keygen_out, 0);
strbuf_stripspace(&ssh_keygen_err, 0);
/* Add stderr outputs to show the user actual ssh-keygen errors */
strbuf_add(&ssh_keygen_out, ssh_principals_err.buf, ssh_principals_err.len);
strbuf_add(&ssh_keygen_out, ssh_keygen_err.buf, ssh_keygen_err.len);
sigc->output = strbuf_detach(&ssh_keygen_out, NULL);
sigc->gpg_status = xstrdup(sigc->output);
parse_ssh_output(sigc);
out:
if (buffer_file)
delete_tempfile(&buffer_file);
strbuf_release(&ssh_principals_out);
strbuf_release(&ssh_principals_err);
strbuf_release(&ssh_keygen_out);
strbuf_release(&ssh_keygen_err);
return ret;
}
int check_signature(const char *payload, size_t plen, const char *signature,
size_t slen, struct signature_check *sigc)
{
struct gpg_format *fmt;
int status;
sigc->result = 'N';
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
sigc->trust_level = -1;
fmt = get_format_by_sig(signature);
if (!fmt)
die(_("bad/incompatible signature '%s'"), signature);
status = fmt->verify_signed_buffer(sigc, fmt, payload, plen, signature,
slen);
if (status && !sigc->output)
return !!status;
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
status |= sigc->result != 'G';
status |= sigc->trust_level < configured_min_trust_level;
gpg-interface: propagate exit status from gpg back to the callers When gpg-interface API unified support for signature verification codepaths for signed tags and signed commits in mid 2015 at around v2.6.0-rc0~114, we accidentally loosened the GPG signature verification. Before that change, signed commits were verified by looking for "G"ood signature from GPG, while ignoring the exit status of "gpg --verify" process, while signed tags were verified by simply passing the exit status of "gpg --verify" through. The unified code we currently have ignores the exit status of "gpg --verify" and returns successful verification when the signature matches an unexpired key regardless of the trust placed on the key (i.e. in addition to "G"ood ones, we accept "U"ntrusted ones). Make these commands signal failure with their exit status when underlying "gpg --verify" (or the custom command specified by "gpg.program" configuration variable) does so. This essentially changes their behaviour in a backward incompatible way to reject signatures that have been made with untrusted keys even if they correctly verify, as that is how "gpg --verify" behaves. Note that the code still overrides a zero exit status obtained from "gpg" (or gpg.program) if the output does not say the signature is good or computes correctly but made with untrusted keys, to catch a poorly written wrapper around "gpg" the user may give us. We could exclude "U"ntrusted support from this fallback code, but that would be making two backward incompatible changes in a single commit, so let's avoid that for now. A follow-up change could do so if desired. Helped-by: Vojtech Myslivec <vojtech.myslivec@nic.cz> Helped-by: brian m. carlson <sandals@crustytoothpaste.net> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-08-10 02:40:27 +08:00
return !!status;
}
void print_signature_buffer(const struct signature_check *sigc, unsigned flags)
{
const char *output = flags & GPG_VERIFY_RAW ? sigc->gpg_status :
sigc->output;
if (flags & GPG_VERIFY_VERBOSE && sigc->payload)
fputs(sigc->payload, stdout);
if (output)
fputs(output, stderr);
}
size_t parse_signed_buffer(const char *buf, size_t size)
{
size_t len = 0;
size_t match = size;
while (len < size) {
const char *eol;
if (get_format_by_sig(buf + len))
match = len;
eol = memchr(buf + len, '\n', size - len);
len += eol ? eol - (buf + len) + 1 : size - len;
}
return match;
}
int parse_signature(const char *buf, size_t size, struct strbuf *payload, struct strbuf *signature)
{
size_t match = parse_signed_buffer(buf, size);
if (match != size) {
strbuf_add(payload, buf, match);
remove_signature(payload);
strbuf_add(signature, buf + match, size - match);
return 1;
}
return 0;
}
void set_signing_key(const char *key)
{
free(configured_signing_key);
configured_signing_key = xstrdup(key);
}
int git_gpg_config(const char *var, const char *value, void *cb)
{
struct gpg_format *fmt = NULL;
char *fmtname = NULL;
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
char *trust;
int ret;
if (!strcmp(var, "user.signingkey")) {
if (!value)
return config_error_nonbool(var);
set_signing_key(value);
return 0;
}
if (!strcmp(var, "gpg.format")) {
if (!value)
return config_error_nonbool(var);
fmt = get_format_by_name(value);
if (!fmt)
return error("unsupported value for %s: %s",
var, value);
use_format = fmt;
return 0;
}
gpg-interface: add minTrustLevel as a configuration option Previously, signature verification for merge and pull operations checked if the key had a trust-level of either TRUST_NEVER or TRUST_UNDEFINED in verify_merge_signature(). If that was the case, the process die()d. The other code paths that did signature verification relied entirely on the return code from check_commit_signature(). And signatures made with a good key, irregardless of its trust level, was considered valid by check_commit_signature(). This difference in behavior might induce users to erroneously assume that the trust level of a key in their keyring is always considered by Git, even for operations where it is not (e.g. during a verify-commit or verify-tag). The way it worked was by gpg-interface.c storing the result from the key/signature status *and* the lowest-two trust levels in the `result` member of the signature_check structure (the last of these status lines that were encountered got written to `result`). These are documented in GPG under the subsection `General status codes` and `Key related`, respectively [1]. The GPG documentation says the following on the TRUST_ status codes [1]: """ These are several similar status codes: - TRUST_UNDEFINED <error_token> - TRUST_NEVER <error_token> - TRUST_MARGINAL [0 [<validation_model>]] - TRUST_FULLY [0 [<validation_model>]] - TRUST_ULTIMATE [0 [<validation_model>]] For good signatures one of these status lines are emitted to indicate the validity of the key used to create the signature. The error token values are currently only emitted by gpgsm. """ My interpretation is that the trust level is conceptionally different from the validity of the key and/or signature. That seems to also have been the assumption of the old code in check_signature() where a result of 'G' (as in GOODSIG) and 'U' (as in TRUST_NEVER or TRUST_UNDEFINED) were both considered a success. The two cases where a result of 'U' had special meaning were in verify_merge_signature() (where this caused git to die()) and in format_commit_one() (where it affected the output of the %G? format specifier). I think it makes sense to refactor the processing of TRUST_ status lines such that users can configure a minimum trust level that is enforced globally, rather than have individual parts of git (e.g. merge) do it themselves (except for a grace period with backward compatibility). I also think it makes sense to not store the trust level in the same struct member as the key/signature status. While the presence of a TRUST_ status code does imply that the signature is good (see the first paragraph in the included snippet above), as far as I can tell, the order of the status lines from GPG isn't well-defined; thus it would seem plausible that the trust level could be overwritten with the key/signature status if they were stored in the same member of the signature_check structure. This patch introduces a new configuration option: gpg.minTrustLevel. It consolidates trust-level verification to gpg-interface.c and adds a new `trust_level` member to the signature_check structure. Backward-compatibility is maintained by introducing a special case in verify_merge_signature() such that if no user-configurable gpg.minTrustLevel is set, then the old behavior of rejecting TRUST_UNDEFINED and TRUST_NEVER is enforced. If, on the other hand, gpg.minTrustLevel is set, then that value overrides the old behavior. Similarly, the %G? format specifier will continue show 'U' for signatures made with a key that has a trust level of TRUST_UNDEFINED or TRUST_NEVER, even though the 'U' character no longer exist in the `result` member of the signature_check structure. A new format specifier, %GT, is also introduced for users that want to show all possible trust levels for a signature. Another approach would have been to simply drop the trust-level requirement in verify_merge_signature(). This would also have made the behavior consistent with other parts of git that perform signature verification. However, requiring a minimum trust level for signing keys does seem to have a real-world use-case. For example, the build system used by the Qubes OS project currently parses the raw output from verify-tag in order to assert a minimum trust level for keys used to sign git tags [2]. [1] https://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=blob;f=doc/doc/DETAILS;h=bd00006e933ac56719b1edd2478ecd79273eae72;hb=refs/heads/master [2] https://github.com/QubesOS/qubes-builder/blob/9674c1991deef45b1a1b1c71fddfab14ba50dccf/scripts/verify-git-tag#L43 Signed-off-by: Hans Jerry Illikainen <hji@dyntopia.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-12-27 21:55:57 +08:00
if (!strcmp(var, "gpg.mintrustlevel")) {
if (!value)
return config_error_nonbool(var);
trust = xstrdup_toupper(value);
ret = parse_gpg_trust_level(trust, &configured_min_trust_level);
free(trust);
if (ret)
return error("unsupported value for %s: %s", var,
value);
return 0;
}
if (!strcmp(var, "gpg.ssh.defaultkeycommand")) {
if (!value)
return config_error_nonbool(var);
return git_config_string(&ssh_default_key_command, var, value);
}
if (!strcmp(var, "gpg.ssh.allowedsignersfile")) {
if (!value)
return config_error_nonbool(var);
return git_config_pathname(&ssh_allowed_signers, var, value);
}
if (!strcmp(var, "gpg.ssh.revocationfile")) {
if (!value)
return config_error_nonbool(var);
return git_config_pathname(&ssh_revocation_file, var, value);
}
if (!strcmp(var, "gpg.program") || !strcmp(var, "gpg.openpgp.program"))
fmtname = "openpgp";
if (!strcmp(var, "gpg.x509.program"))
fmtname = "x509";
if (!strcmp(var, "gpg.ssh.program"))
fmtname = "ssh";
if (fmtname) {
fmt = get_format_by_name(fmtname);
return git_config_string(&fmt->program, var, value);
}
return 0;
}
static char *get_ssh_key_fingerprint(const char *signing_key)
{
struct child_process ssh_keygen = CHILD_PROCESS_INIT;
int ret = -1;
struct strbuf fingerprint_stdout = STRBUF_INIT;
struct strbuf **fingerprint;
char *fingerprint_ret;
/*
* With SSH Signing this can contain a filename or a public key
* For textual representation we usually want a fingerprint
*/
if (starts_with(signing_key, "ssh-")) {
strvec_pushl(&ssh_keygen.args, "ssh-keygen", "-lf", "-", NULL);
ret = pipe_command(&ssh_keygen, signing_key,
strlen(signing_key), &fingerprint_stdout, 0,
NULL, 0);
} else {
strvec_pushl(&ssh_keygen.args, "ssh-keygen", "-lf",
configured_signing_key, NULL);
ret = pipe_command(&ssh_keygen, NULL, 0, &fingerprint_stdout, 0,
NULL, 0);
}
if (!!ret)
die_errno(_("failed to get the ssh fingerprint for key '%s'"),
signing_key);
fingerprint = strbuf_split_max(&fingerprint_stdout, ' ', 3);
if (!fingerprint[1])
die_errno(_("failed to get the ssh fingerprint for key '%s'"),
signing_key);
fingerprint_ret = strbuf_detach(fingerprint[1], NULL);
strbuf_list_free(fingerprint);
strbuf_release(&fingerprint_stdout);
return fingerprint_ret;
}
/* Returns the first public key from an ssh-agent to use for signing */
static const char *get_default_ssh_signing_key(void)
{
struct child_process ssh_default_key = CHILD_PROCESS_INIT;
int ret = -1;
struct strbuf key_stdout = STRBUF_INIT, key_stderr = STRBUF_INIT;
struct strbuf **keys;
char *key_command = NULL;
const char **argv;
int n;
char *default_key = NULL;
if (!ssh_default_key_command)
die(_("either user.signingkey or gpg.ssh.defaultKeyCommand needs to be configured"));
key_command = xstrdup(ssh_default_key_command);
n = split_cmdline(key_command, &argv);
if (n < 0)
die("malformed build-time gpg.ssh.defaultKeyCommand: %s",
split_cmdline_strerror(n));
strvec_pushv(&ssh_default_key.args, argv);
ret = pipe_command(&ssh_default_key, NULL, 0, &key_stdout, 0,
&key_stderr, 0);
if (!ret) {
keys = strbuf_split_max(&key_stdout, '\n', 2);
if (keys[0] && starts_with(keys[0]->buf, "ssh-")) {
default_key = strbuf_detach(keys[0], NULL);
} else {
warning(_("gpg.ssh.defaultKeycommand succeeded but returned no keys: %s %s"),
key_stderr.buf, key_stdout.buf);
}
strbuf_list_free(keys);
} else {
warning(_("gpg.ssh.defaultKeyCommand failed: %s %s"),
key_stderr.buf, key_stdout.buf);
}
free(key_command);
free(argv);
strbuf_release(&key_stdout);
return default_key;
}
static const char *get_ssh_key_id(void) {
return get_ssh_key_fingerprint(get_signing_key());
}
/* Returns a textual but unique representation of the signing key */
const char *get_signing_key_id(void)
{
if (use_format->get_key_id) {
return use_format->get_key_id();
}
/* GPG/GPGSM only store a key id on this variable */
return get_signing_key();
}
const char *get_signing_key(void)
{
if (configured_signing_key)
return configured_signing_key;
if (use_format->get_default_key) {
return use_format->get_default_key();
}
return git_committer_info(IDENT_STRICT | IDENT_NO_DATE);
}
int sign_buffer(struct strbuf *buffer, struct strbuf *signature, const char *signing_key)
{
return use_format->sign_buffer(buffer, signature, signing_key);
}
/*
* Strip CR from the line endings, in case we are on Windows.
* NEEDSWORK: make it trim only CRs before LFs and rename
*/
static void remove_cr_after(struct strbuf *buffer, size_t offset)
{
size_t i, j;
for (i = j = offset; i < buffer->len; i++) {
if (buffer->buf[i] != '\r') {
if (i != j)
buffer->buf[j] = buffer->buf[i];
j++;
}
}
strbuf_setlen(buffer, j);
}
static int sign_buffer_gpg(struct strbuf *buffer, struct strbuf *signature,
const char *signing_key)
{
struct child_process gpg = CHILD_PROCESS_INIT;
int ret;
size_t bottom;
struct strbuf gpg_status = STRBUF_INIT;
strvec_pushl(&gpg.args,
use_format->program,
"--status-fd=2",
"-bsau", signing_key,
NULL);
bottom = signature->len;
/*
* When the username signingkey is bad, program could be terminated
* because gpg exits without reading and then write gets SIGPIPE.
*/
sigchain_push(SIGPIPE, SIG_IGN);
ret = pipe_command(&gpg, buffer->buf, buffer->len,
signature, 1024, &gpg_status, 0);
sigchain_pop(SIGPIPE);
ret |= !strstr(gpg_status.buf, "\n[GNUPG:] SIG_CREATED ");
strbuf_release(&gpg_status);
if (ret)
return error(_("gpg failed to sign the data"));
/* Strip CR from the line endings, in case we are on Windows. */
remove_cr_after(signature, bottom);
return 0;
}
static int sign_buffer_ssh(struct strbuf *buffer, struct strbuf *signature,
const char *signing_key)
{
struct child_process signer = CHILD_PROCESS_INIT;
int ret = -1;
size_t bottom, keylen;
struct strbuf signer_stderr = STRBUF_INIT;
struct tempfile *key_file = NULL, *buffer_file = NULL;
char *ssh_signing_key_file = NULL;
struct strbuf ssh_signature_filename = STRBUF_INIT;
if (!signing_key || signing_key[0] == '\0')
return error(
_("user.signingkey needs to be set for ssh signing"));
if (starts_with(signing_key, "ssh-")) {
/* A literal ssh key */
key_file = mks_tempfile_t(".git_signing_key_tmpXXXXXX");
if (!key_file)
return error_errno(
_("could not create temporary file"));
keylen = strlen(signing_key);
if (write_in_full(key_file->fd, signing_key, keylen) < 0 ||
close_tempfile_gently(key_file) < 0) {
error_errno(_("failed writing ssh signing key to '%s'"),
key_file->filename.buf);
goto out;
}
ssh_signing_key_file = strbuf_detach(&key_file->filename, NULL);
} else {
/* We assume a file */
ssh_signing_key_file = expand_user_path(signing_key, 1);
}
buffer_file = mks_tempfile_t(".git_signing_buffer_tmpXXXXXX");
if (!buffer_file) {
error_errno(_("could not create temporary file"));
goto out;
}
if (write_in_full(buffer_file->fd, buffer->buf, buffer->len) < 0 ||
close_tempfile_gently(buffer_file) < 0) {
error_errno(_("failed writing ssh signing key buffer to '%s'"),
buffer_file->filename.buf);
goto out;
}
strvec_pushl(&signer.args, use_format->program,
"-Y", "sign",
"-n", "git",
"-f", ssh_signing_key_file,
buffer_file->filename.buf,
NULL);
sigchain_push(SIGPIPE, SIG_IGN);
ret = pipe_command(&signer, NULL, 0, NULL, 0, &signer_stderr, 0);
sigchain_pop(SIGPIPE);
if (ret) {
if (strstr(signer_stderr.buf, "usage:"))
error(_("ssh-keygen -Y sign is needed for ssh signing (available in openssh version 8.2p1+)"));
error("%s", signer_stderr.buf);
goto out;
}
bottom = signature->len;
strbuf_addbuf(&ssh_signature_filename, &buffer_file->filename);
strbuf_addstr(&ssh_signature_filename, ".sig");
if (strbuf_read_file(signature, ssh_signature_filename.buf, 0) < 0) {
error_errno(
_("failed reading ssh signing data buffer from '%s'"),
ssh_signature_filename.buf);
}
unlink_or_warn(ssh_signature_filename.buf);
/* Strip CR from the line endings, in case we are on Windows. */
remove_cr_after(signature, bottom);
out:
if (key_file)
delete_tempfile(&key_file);
if (buffer_file)
delete_tempfile(&buffer_file);
strbuf_release(&signer_stderr);
strbuf_release(&ssh_signature_filename);
FREE_AND_NULL(ssh_signing_key_file);
return ret;
}