config.c: NULL check when reading protected config

PATCH Superseded Glen Choo via GitGitGadget 6 participants 14 messages Jul 26, 2022 – Jul 27, 2022

lore

Share & tools

lore

Glen Choo via GitGitGadget

author

2022-07-26 17:09 UTC #0 lore raw

PATCH 1/1 config.c: NULL check when reading protected config

From: Glen Choo <[email protected]>

In read_protected_config(), check whether each file name is NULL before attempting to read it. This mirrors do_git_config_sequence() (which read_protected_config() is modelled after).

Without these NULL checks,

make SANITIZE=address test T=t0410*.sh

fails because xdg_config is NULL, causing us to call fopen(NULL).

Reported-by Ævar Arnfjörð Bjarmason <[email protected]>

Signed-off-by Glen Choo <[email protected]>

---
config.c: NULL check when reading protected config

This fixes the SANITIZE=address failure on master, That was introduced by gc/bare-repo-discovery. Thanks again to Ævar for the original report
[1] and for proposing a way to catch this in CI [2].

[1]
https://lore.kernel.org/git/[email protected]
[2]
https://lore.kernel.org/git/[email protected]

Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-git-1299%2Fchooglen%2Fconfig%2Ffix-sanitize-address-v1
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-git-1299/chooglen/config/fix-sanitize-address-v1
Pull-Request: https://github.com/git/git/pull/1299

 config.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

config.c +6 -3

index 015bec360f5..b0ba7f439a4 100644
@@ -2645,9 +2645,12 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);
 
-	git_configset_add_file(&protected_config, system_config);
-	git_configset_add_file(&protected_config, xdg_config);
-	git_configset_add_file(&protected_config, user_config);
+	if (system_config)
+		git_configset_add_file(&protected_config, system_config);
+	if (xdg_config)
+		git_configset_add_file(&protected_config, xdg_config);
+	if (user_config)
+		git_configset_add_file(&protected_config, user_config);
 	git_configset_add_parameters(&protected_config);
 
 	free(system_config);

base-commit: 6a475b71f8c4ce708d69fdc9317aefbde3769e25

Signature · 1 line

gitgitgadget

Taylor Blau

2022-07-26 17:27 UTC #1 lore raw

Quoted 5 lines · Glen Choo via GitGitGadget

From: Glen Choo <[email protected]>
In read_protected_config(), check whether each file name is NULL before attempting to read it. This mirrors do_git_config_sequence() (which read_protected_config() is modelled after).

s/modelled/modeled

Quoted 3 lines

Without these NULL checks,
make SANITIZE=address test T=t0410*.sh

I'm glad that t0410 was catching this for us already, though it is too bad we didn't see it outside of the ASan builds, or I think we could have potentially caught this earlier.

Either way, I think the test coverage here is sufficient, so what you wrote makes sense.

Quoted 8 lines

config.c +0 -0

index 015bec360f5..b0ba7f439a4 100644
@@ -2645,9 +2645,12 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);

config.c static void read_protected_config(void) · line 2654

Quoted 10 lines

config.c +6 -3

-	git_configset_add_file(&protected_config, system_config);
-	git_configset_add_file(&protected_config, xdg_config);
-	git_configset_add_file(&protected_config, user_config);
+	if (system_config)
+		git_configset_add_file(&protected_config, system_config);
+	if (xdg_config)
+		git_configset_add_file(&protected_config, xdg_config);
+	if (user_config)
+		git_configset_add_file(&protected_config, user_config);
 	git_configset_add_parameters(&protected_config);

I wonder: should it become a BUG() to call git_configset_add_file() with a NULL filename? That would have elevated the test failure outside of just the ASAn builds, I'd think.

There's certainty a risk of being too defensive, but elevating this error beyond just the ASan builds indicates that this would be an appropriate layer of defense IMHO.

Thanks,
Taylor

Glen Choo

2022-07-26 17:40 UTC #2 lore raw

Quoted 8 lines · Taylor Blau

Quoted 5 lines · Glen Choo via GitGitGadget
From: Glen Choo <[email protected]>
In read_protected_config(), check whether each file name is NULL before attempting to read it. This mirrors do_git_config_sequence() (which read_protected_config() is modelled after).
s/modelled/modeled

Ah, thanks.

Quoted 34 lines

Quoted 3 lines
Without these NULL checks,
make SANITIZE=address test T=t0410*.sh
I'm glad that t0410 was catching this for us already, though it is too bad we didn't see it outside of the ASan builds, or I think we could have potentially caught this earlier.
Either way, I think the test coverage here is sufficient, so what you wrote makes sense.
Quoted 8 lines
config.c +0 -0
index 015bec360f5..b0ba7f439a4 100644
@@ -2645,9 +2645,12 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);
 
config.c static void read_protected_config(void) · line 2654
Quoted 10 lines
config.c +6 -3
-	git_configset_add_file(&protected_config, system_config);
-	git_configset_add_file(&protected_config, xdg_config);
-	git_configset_add_file(&protected_config, user_config);
+	if (system_config)
+		git_configset_add_file(&protected_config, system_config);
+	if (xdg_config)
+		git_configset_add_file(&protected_config, xdg_config);
+	if (user_config)
+		git_configset_add_file(&protected_config, user_config);
 	git_configset_add_parameters(&protected_config);
I wonder: should it become a BUG() to call git_configset_add_file() with a NULL filename? That would have elevated the test failure outside of just the ASAn builds, I'd think.

Quoted 3 lines

There's certainty a risk of being too defensive, but elevating this error beyond just the ASan builds indicates that this would be an appropriate layer of defense IMHO.

Hm, if we're going in this direction, what if we made it a BUG() to call fopen_or_warn() with a NULL filename? Then we wouldn't have to reimplement this BUG() check in all of its callers.

Quoted 3 lines

Thanks,
Taylor

Taylor Blau

2022-07-26 17:44 UTC #3 lore raw

Quoted 11 lines · Glen Choo

Quoted 7 lines
I wonder: should it become a BUG() to call git_configset_add_file() with a NULL filename? That would have elevated the test failure outside of just the ASAn builds, I'd think.
There's certainty a risk of being too defensive, but elevating this error beyond just the ASan builds indicates that this would be an appropriate layer of defense IMHO.
Hm, if we're going in this direction, what if we made it a BUG() to call fopen_or_warn() with a NULL filename? Then we wouldn't have to reimplement this BUG() check in all of its callers.

That may be too low-level of a place to put this check, but I don't have a strong opinion about it either way (including whether we should have such a BUG() *anywhere* in this series, including git_configset_add_file()).

Thanks,
Taylor

Derrick Stolee

2022-07-26 17:52 UTC #4 lore raw

Quoted 9 lines · Taylor Blau

Quoted 8 lines · Glen Choo
Quoted 7 lines
I wonder: should it become a BUG() to call git_configset_add_file() with a NULL filename? That would have elevated the test failure outside of just the ASAn builds, I'd think.
There's certainty a risk of being too defensive, but elevating this error beyond just the ASan builds indicates that this would be an appropriate layer of defense IMHO.

Quoted 8 lines

Quoted 3 lines
Hm, if we're going in this direction, what if we made it a BUG() to call fopen_or_warn() with a NULL filename? Then we wouldn't have to reimplement this BUG() check in all of its callers.
That may be too low-level of a place to put this check, but I don't have a strong opinion about it either way (including whether we should have such a BUG() *anywhere* in this series, including git_configset_add_file()).

Since git_configset_add_file() returns an 'int', could we return -1 if the supplied 'filename' was null? (The correct place to check would be down in git_config_from_file_with_options().)

It would save all these checks here.

(Also: do we care that we are ignoring the return values in read_protected_config()?

Thanks,
-Stolee

Glen Choo

2022-07-26 19:43 UTC #5 lore raw

Quoted 19 lines · Derrick Stolee

Quoted 9 lines · Taylor Blau
Quoted 8 lines · Glen Choo
Quoted 7 lines
I wonder: should it become a BUG() to call git_configset_add_file() with a NULL filename? That would have elevated the test failure outside of just the ASAn builds, I'd think.
There's certainty a risk of being too defensive, but elevating this error beyond just the ASan builds indicates that this would be an appropriate layer of defense IMHO.
Quoted 8 lines
Quoted 3 lines
Hm, if we're going in this direction, what if we made it a BUG() to call fopen_or_warn() with a NULL filename? Then we wouldn't have to reimplement this BUG() check in all of its callers.
That may be too low-level of a place to put this check, but I don't have a strong opinion about it either way (including whether we should have such a BUG() *anywhere* in this series, including git_configset_add_file()).

Quoted 5 lines

Since git_configset_add_file() returns an 'int', could we return -1 if the supplied 'filename' was null? (The correct place to check would be down in git_config_from_file_with_options().)
It would save all these checks here.

Hm, IIUC you are suggesting that git_configset_add_file() returns -1 instead of BUG()-ing?

BUG() sounds better IMO, since there really is nothing useful that git_configset_add_file() (and later functions) can do with a NULL file name. Plus, git_configset_add_file() has already reserved -1 to mean "a file was specified but could not be read".

Quoted 3 lines

(Also: do we care that we are ignoring the return values in read_protected_config()?

I don't think we care (unless this is a style issue). "git config" succeeds even if it encounters non-repo files that can't be read. In a similar vein, I don't think it matters for protected config if we can't read one of the files (e.g. xdg_config) or even all of the files; all that matters is that we've read everything that we can.

Quoted 3 lines

Thanks,
-Stolee

Ævar Arnfjörð Bjarmason

2022-07-26 19:14 UTC #6 lore raw

In reply to Glen Choo via GitGitGadget

Quoted 11 lines · Glen Choo via GitGitGadget

From: Glen Choo <[email protected]>
In read_protected_config(), check whether each file name is NULL before attempting to read it. This mirrors do_git_config_sequence() (which read_protected_config() is modelled after).
Without these NULL checks,
make SANITIZE=address test T=t0410*.sh
fails because xdg_config is NULL, causing us to call fopen(NULL).

FWIW a lot more than that fails, that's just the test I focused on for the bug report, the others ones (I didn't check out all of them) all variants of that.

See https://github.com/avar/git/runs/7519070124?check_suite_focus=true
for the current failing run with that "[2]" patch you quoted. We fail a total of 14 test files (and many more tests within those files).

Quoted 42 lines

Reported-by Ævar Arnfjörð Bjarmason <[email protected]>
Signed-off-by Glen Choo <[email protected]>
---
config.c: NULL check when reading protected config
This fixes the SANITIZE=address failure on master, That was introduced by gc/bare-repo-discovery. Thanks again to Ævar for the original report
[1] and for proposing a way to catch this in CI [2].
[1]
https://lore.kernel.org/git/[email protected]
[2]
https://lore.kernel.org/git/[email protected]
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-git-1299%2Fchooglen%2Fconfig%2Ffix-sanitize-address-v1
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-git-1299/chooglen/config/fix-sanitize-address-v1
Pull-Request: https://github.com/git/git/pull/1299
 config.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
config.c +6 -3
index 015bec360f5..b0ba7f439a4 100644
@@ -2645,9 +2645,12 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);
 
-	git_configset_add_file(&protected_config, system_config);
-	git_configset_add_file(&protected_config, xdg_config);
-	git_configset_add_file(&protected_config, user_config);
+	if (system_config)
+		git_configset_add_file(&protected_config, system_config);
+	if (xdg_config)
+		git_configset_add_file(&protected_config, xdg_config);
+	if (user_config)
+		git_configset_add_file(&protected_config, user_config);
 	git_configset_add_parameters(&protected_config);
 
 	free(system_config);

Quoted 1 line

base-commit: 6a475b71f8c4ce708d69fdc9317aefbde3769e25

Re your claim in
https://lore.kernel.org/git/[email protected]/
I tried testing this, and came up with the below.

I wonder if we should work in here for general paranoia, but I'm not too familiar with the this part of the config reading, maybe we're confident enough that these are invariants within the process.

This will BUG() out if these variables change within the process, which would mean that our caching assumptions are no longer true, which would cause you to return the wrong data here.

Of course you'd have segfaulted or similar before, but this should demonstrate that not only are these sometimes NULL, but that they stay that way.

config.c +28 -0

index 015bec360f5..cdd665c1cc8 100644
@@ -2102,6 +2102,30 @@ int git_config_system(void)
 	return !git_env_bool("GIT_CONFIG_NOSYSTEM", 0);
 }
 
+static char *last_system_config;
+static char *last_xdg_config;
+static char *last_user_config;
+
+static void sanity_check_config_1(const char *name, char **last,
+				  const char *now)
+{
+	if (*last && now && strcmp(*last, now))
+		BUG("%s_config: had '%s', now '%s'", name, *last, now);
+	else if (*last && !now)
+		BUG("%s_config: had '%s', now NULL", name, *last);
+	free(*last);
+	*last = xstrdup_or_null(now);
+}
+
+static void sanity_check_config(const char *system_config,
+				const char *xdg_config,
+				const char *user_config)
+{
+	sanity_check_config_1("system", &last_system_config, system_config);
+	sanity_check_config_1("xdg", &last_xdg_config, xdg_config);
+	sanity_check_config_1("user", &last_user_config, user_config);
+}
+
 static int do_git_config_sequence(const struct config_options *opts,
 				  config_fn_t fn, void *data)
 {

@@ -2134,6 +2158,8 @@ static int do_git_config_sequence(const struct config_options *opts,
 	if (user_config && !access_or_die(user_config, R_OK, ACCESS_EACCES_OK))
 		ret += git_config_from_file(fn, user_config, data);
 
+	sanity_check_config(system_config, xdg_config, user_config);
+
 	current_parsing_scope = CONFIG_SCOPE_LOCAL;
 	if (!opts->ignore_repo && repo_config &&
 	    !access_or_die(repo_config, R_OK, 0))

@@ -2645,6 +2671,8 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);
 
+	sanity_check_config(system_config, xdg_config, user_config);
+
 	git_configset_add_file(&protected_config, system_config);
 	git_configset_add_file(&protected_config, xdg_config);
 	git_configset_add_file(&protected_config, user_config);

Glen Choo

2022-07-26 20:00 UTC #7 lore raw

Quoted 14 lines · Ævar Arnfjörð Bjarmason

Quoted 11 lines · Glen Choo via GitGitGadget
From: Glen Choo <[email protected]>
In read_protected_config(), check whether each file name is NULL before attempting to read it. This mirrors do_git_config_sequence() (which read_protected_config() is modelled after).
Without these NULL checks,
make SANITIZE=address test T=t0410*.sh
fails because xdg_config is NULL, causing us to call fopen(NULL).

Quoted 7 lines

FWIW a lot more than that fails, that's just the test I focused on for the bug report, the others ones (I didn't check out all of them) all variants of that.
See https://github.com/avar/git/runs/7519070124?check_suite_focus=true
for the current failing run with that "[2]" patch you quoted. We fail a total of 14 test files (and many more tests within those files).

Ah thanks, I'll amend the message accordingly.

Quoted 52 lines

Quoted 42 lines
Reported-by Ævar Arnfjörð Bjarmason <[email protected]>
Signed-off-by Glen Choo <[email protected]>
---
config.c: NULL check when reading protected config
This fixes the SANITIZE=address failure on master, That was introduced by gc/bare-repo-discovery. Thanks again to Ævar for the original report
[1] and for proposing a way to catch this in CI [2].
[1]
https://lore.kernel.org/git/[email protected]
[2]
https://lore.kernel.org/git/[email protected]
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-git-1299%2Fchooglen%2Fconfig%2Ffix-sanitize-address-v1
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-git-1299/chooglen/config/fix-sanitize-address-v1
Pull-Request: https://github.com/git/git/pull/1299
 config.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
config.c +6 -3
index 015bec360f5..b0ba7f439a4 100644
@@ -2645,9 +2645,12 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);
 
-	git_configset_add_file(&protected_config, system_config);
-	git_configset_add_file(&protected_config, xdg_config);
-	git_configset_add_file(&protected_config, user_config);
+	if (system_config)
+		git_configset_add_file(&protected_config, system_config);
+	if (xdg_config)
+		git_configset_add_file(&protected_config, xdg_config);
+	if (user_config)
+		git_configset_add_file(&protected_config, user_config);
 	git_configset_add_parameters(&protected_config);
 
 	free(system_config);
Quoted 1 line
base-commit: 6a475b71f8c4ce708d69fdc9317aefbde3769e25
Re your claim in
https://lore.kernel.org/git/[email protected]/
I tried testing this, and came up with the below.
I wonder if we should work in here for general paranoia, but I'm not too familiar with the this part of the config reading, maybe we're confident enough that these are invariants within the process.

Quoted 7 lines

This will BUG() out if these variables change within the process, which would mean that our caching assumptions are no longer true, which would cause you to return the wrong data here.
Of course you'd have segfaulted or similar before, but this should demonstrate that not only are these sometimes NULL, but that they stay that way.

Interesting, this is worth proposing, but I suspect that the conversation will be long enough for this to be its own thread. Surely someone must have given some thought to this, especially for long-lived processes (git-daemon?).

There's also the general question of config cache freshness, e.g. what if another git process writes to a shared config file? (We don't worry about the single process case because the process will invalidate its own cache).

Perhaps we should also worry about that (probably more common) case in addition to this one? At any rate, that seems like a bigger topic than this fix here.

Quoted 54 lines

config.c +28 -0

index 015bec360f5..cdd665c1cc8 100644
@@ -2102,6 +2102,30 @@ int git_config_system(void)
 	return !git_env_bool("GIT_CONFIG_NOSYSTEM", 0);
 }
 
+static char *last_system_config;
+static char *last_xdg_config;
+static char *last_user_config;
+
+static void sanity_check_config_1(const char *name, char **last,
+				  const char *now)
+{
+	if (*last && now && strcmp(*last, now))
+		BUG("%s_config: had '%s', now '%s'", name, *last, now);
+	else if (*last && !now)
+		BUG("%s_config: had '%s', now NULL", name, *last);
+	free(*last);
+	*last = xstrdup_or_null(now);
+}
+
+static void sanity_check_config(const char *system_config,
+				const char *xdg_config,
+				const char *user_config)
+{
+	sanity_check_config_1("system", &last_system_config, system_config);
+	sanity_check_config_1("xdg", &last_xdg_config, xdg_config);
+	sanity_check_config_1("user", &last_user_config, user_config);
+}
+
 static int do_git_config_sequence(const struct config_options *opts,
 				  config_fn_t fn, void *data)
 {

@@ -2134,6 +2158,8 @@ static int do_git_config_sequence(const struct config_options *opts,
 	if (user_config && !access_or_die(user_config, R_OK, ACCESS_EACCES_OK))
 		ret += git_config_from_file(fn, user_config, data);
 
+	sanity_check_config(system_config, xdg_config, user_config);
+
 	current_parsing_scope = CONFIG_SCOPE_LOCAL;
 	if (!opts->ignore_repo && repo_config &&
 	    !access_or_die(repo_config, R_OK, 0))

@@ -2645,6 +2671,8 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);
 
+	sanity_check_config(system_config, xdg_config, user_config);
+
 	git_configset_add_file(&protected_config, system_config);
 	git_configset_add_file(&protected_config, xdg_config);
 	git_configset_add_file(&protected_config, user_config);

Ævar Arnfjörð Bjarmason

2022-07-27 09:14 UTC #8 lore raw

Quoted 91 lines · Glen Choo

Quoted 14 lines · Ævar Arnfjörð Bjarmason
Quoted 11 lines · Glen Choo via GitGitGadget
From: Glen Choo <[email protected]>
In read_protected_config(), check whether each file name is NULL before attempting to read it. This mirrors do_git_config_sequence() (which read_protected_config() is modelled after).
Without these NULL checks,
make SANITIZE=address test T=t0410*.sh
fails because xdg_config is NULL, causing us to call fopen(NULL).
Quoted 7 lines
FWIW a lot more than that fails, that's just the test I focused on for the bug report, the others ones (I didn't check out all of them) all variants of that.
See https://github.com/avar/git/runs/7519070124?check_suite_focus=true
for the current failing run with that "[2]" patch you quoted. We fail a total of 14 test files (and many more tests within those files).
Ah thanks, I'll amend the message accordingly.
Quoted 52 lines
Quoted 42 lines
Reported-by Ævar Arnfjörð Bjarmason <[email protected]>
Signed-off-by Glen Choo <[email protected]>
---
config.c: NULL check when reading protected config
This fixes the SANITIZE=address failure on master, That was introduced by gc/bare-repo-discovery. Thanks again to Ævar for the original report
[1] and for proposing a way to catch this in CI [2].
[1]
https://lore.kernel.org/git/[email protected]
[2]
https://lore.kernel.org/git/[email protected]
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-git-1299%2Fchooglen%2Fconfig%2Ffix-sanitize-address-v1
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-git-1299/chooglen/config/fix-sanitize-address-v1
Pull-Request: https://github.com/git/git/pull/1299
 config.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
config.c +6 -3
index 015bec360f5..b0ba7f439a4 100644
@@ -2645,9 +2645,12 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);
 
-	git_configset_add_file(&protected_config, system_config);
-	git_configset_add_file(&protected_config, xdg_config);
-	git_configset_add_file(&protected_config, user_config);
+	if (system_config)
+		git_configset_add_file(&protected_config, system_config);
+	if (xdg_config)
+		git_configset_add_file(&protected_config, xdg_config);
+	if (user_config)
+		git_configset_add_file(&protected_config, user_config);
 	git_configset_add_parameters(&protected_config);
 
 	free(system_config);
Quoted 1 line
base-commit: 6a475b71f8c4ce708d69fdc9317aefbde3769e25
Re your claim in
https://lore.kernel.org/git/[email protected]/
I tried testing this, and came up with the below.
I wonder if we should work in here for general paranoia, but I'm not too familiar with the this part of the config reading, maybe we're confident enough that these are invariants within the process.
Quoted 7 lines
This will BUG() out if these variables change within the process, which would mean that our caching assumptions are no longer true, which would cause you to return the wrong data here.
Of course you'd have segfaulted or similar before, but this should demonstrate that not only are these sometimes NULL, but that they stay that way.
Interesting, this is worth proposing, but I suspect that the conversation will be long enough for this to be its own thread. Surely someone must have given some thought to this, especially for long-lived processes (git-daemon?).

Quoted 8 lines

There's also the general question of config cache freshness, e.g. what if another git process writes to a shared config file? (We don't worry about the single process case because the process will invalidate its own cache).
Perhaps we should also worry about that (probably more common) case in addition to this one? At any rate, that seems like a bigger topic than this fix here.

We can leave it for later, I've run it as a one-off and didn't have any failures.

But FWIW I think it's tied up in this fix here, i.e. your original code both added caching, and implicitly assumed that these were never NULL, so it was "obvious" that it didn't need such assertions.

But now we have 3x if's in a code path that's cached, and the cache is *not* guarded by the same 3x checks.

So we can leave it for later, but it really seems worth adding some self-documentation here sooner than later.

This BUG() method I came up with is one way, another would be to strdup it and use a "static" variable in the function, i.e. stick with whatever value(s) we start out with.

But in any case, this fix seems correct, and fixes the current issues SANITIZE=address is spotting for us, thanks!

Glen Choo via GitGitGadget

author

2022-07-26 22:21 UTC #9 lore raw

PATCH v2 1/1 config.c: NULL check when reading protected config

In reply to Glen Choo via GitGitGadget

From: Glen Choo <[email protected]>

In read_protected_config(), check whether each file name is NULL before attempting to read it, and add a BUG() call to git_config_from_file_with_options() to make this error easier to catch in the future.

The NULL checks mirror what do_git_config_sequence() does (which read_protected_config() is modeled after). Without these NULL checks, multiple tests fail with "make SANITIZE=address", e.g. in the final test of t4010, xdg_config is NULL causing us to call fopen(NULL).

Reported-by Ævar Arnfjörð Bjarmason <[email protected]>

Signed-off-by Glen Choo <[email protected]>

---
config.c: NULL check when reading protected config

This fixes the SANITIZE=address failure on master, That was introduced by gc/bare-repo-discovery. Thanks again to Ævar for the original report
[1] and for proposing a way to catch this in CI [2].

Changes in v2:

* Fix typo
* Add BUG() to git_config_from_file_with_options()

[1]
https://lore.kernel.org/git/[email protected]
[2]
https://lore.kernel.org/git/[email protected]

Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-git-1299%2Fchooglen%2Fconfig%2Ffix-sanitize-address-v2
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-git-1299/chooglen/config/fix-sanitize-address-v2
Pull-Request: https://github.com/git/git/pull/1299

Range-diff vs v1:

1: 17b4a489c69 ! 1: ba51078418a config.c: NULL check when reading protected config
@@ Commit message
config.c: NULL check when reading protected config

In read_protected_config(), check whether each file name is NULL before
- attempting to read it. This mirrors do_git_config_sequence() (which
- read_protected_config() is modelled after).
+ attempting to read it, and add a BUG() call to
+ git_config_from_file_with_options() to make this error easier to catch
+ in the future.

- Without these NULL checks,
-
- make SANITIZE=address test T=t0410*.sh
-
- fails because xdg_config is NULL, causing us to call fopen(NULL).
+ The NULL checks mirror what do_git_config_sequence() does (which
+ read_protected_config() is modeled after). Without these NULL checks,
+ multiple tests fail with "make SANITIZE=address", e.g. in the final test
+ of t4010, xdg_config is NULL causing us to call fopen(NULL).

Reported-by Ævar Arnfjörð Bjarmason <[email protected]>

Signed-off-by Glen Choo <[email protected]>

## config.c ##
+@@ config.c: int git_config_from_file_with_options(config_fn_t fn, const char *filename,
+ int ret = -1;
+ FILE *f;
+
++ if (!filename)
++ BUG("filename cannot be NULL");
+ f = fopen_or_warn(filename, "r");
+ if (f) {
+ ret = do_config_from_file(fn, CONFIG_ORIGIN_FILE, filename,
@@ config.c: static void read_protected_config(void)
system_config = git_system_config();
git_global_config(&user_config, &xdg_config);

 config.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

config.c +8 -3

index 015bec360f5..e8ebef77d5c 100644
@@ -1979,6 +1979,8 @@ int git_config_from_file_with_options(config_fn_t fn, const char *filename,
 	int ret = -1;
 	FILE *f;
 
+	if (!filename)
+		BUG("filename cannot be NULL");
 	f = fopen_or_warn(filename, "r");
 	if (f) {
 		ret = do_config_from_file(fn, CONFIG_ORIGIN_FILE, filename,

@@ -2645,9 +2647,12 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);
 
-	git_configset_add_file(&protected_config, system_config);
-	git_configset_add_file(&protected_config, xdg_config);
-	git_configset_add_file(&protected_config, user_config);
+	if (system_config)
+		git_configset_add_file(&protected_config, system_config);
+	if (xdg_config)
+		git_configset_add_file(&protected_config, xdg_config);
+	if (user_config)
+		git_configset_add_file(&protected_config, user_config);
 	git_configset_add_parameters(&protected_config);
 
 	free(system_config);

base-commit: 6a475b71f8c4ce708d69fdc9317aefbde3769e25

Signature · 1 line

gitgitgadget

Ævar Arnfjörð Bjarmason

2022-07-27 09:15 UTC #10 lore raw

nonnull v.s. BUG() if !x (was: [PATCH v2] config.c: NULL check when reading protected config)

config.c int git_config_from_file_with_options(config_fn_t fn, const char *filename, · line 1983

Quoted 4 lines · Glen Choo via GitGitGadget

From: Glen Choo <[email protected]>
config.c +2 -0
+	if (!filename)
+		BUG("filename cannot be NULL");

Looks good, but as an aside I wonder if we wouldn't get better code analysis with "nonnull" for this sort of thing, but we can leave this for now:
https://gcc.gnu.org/onlinedocs/gcc-12.1.0/gcc/Common-Function-Attributes.html#Common-Function-Attributes

Glen Choo

2022-07-27 18:03 UTC #11 lore raw

Re: nonnull v.s. BUG() if !x (was: [PATCH v2] config.c: NULL check when reading protected config)

Quoted 11 lines · Ævar Arnfjörð Bjarmason

config.c int git_config_from_file_with_options(config_fn_t fn, const char *filename, · line 1983
Quoted 4 lines · Glen Choo via GitGitGadget
From: Glen Choo <[email protected]>
config.c +2 -0
+	if (!filename)
+		BUG("filename cannot be NULL");
Looks good, but as an aside I wonder if we wouldn't get better code analysis with "nonnull" for this sort of thing, but we can leave this for now:
https://gcc.gnu.org/onlinedocs/gcc-12.1.0/gcc/Common-Function-Attributes.html#Common-Function-Attributes

Interesting. I wonder how good the analysis is vs the cost, e.g. it's useful if it detects _maybe_ NULL variables, but it might be too expensive if it requires us to mark all of our variables as non-NULL.

Junio C Hamano

2022-07-27 15:01 UTC #12 lore raw

In reply to Glen Choo via GitGitGadget

Quoted 8 lines · Glen Choo via GitGitGadget

config.c +0 -0

index 015bec360f5..e8ebef77d5c 100644
@@ -2645,9 +2647,12 @@ static void read_protected_config(void)
 	system_config = git_system_config();
 	git_global_config(&user_config, &xdg_config);

config.c static void read_protected_config(void) · line 2656

Quoted 10 lines

config.c +6 -3

-	git_configset_add_file(&protected_config, system_config);
-	git_configset_add_file(&protected_config, xdg_config);
-	git_configset_add_file(&protected_config, user_config);
+	if (system_config)
+		git_configset_add_file(&protected_config, system_config);
+	if (xdg_config)
+		git_configset_add_file(&protected_config, xdg_config);
+	if (user_config)
+		git_configset_add_file(&protected_config, user_config);
 	git_configset_add_parameters(&protected_config);

This does make it similar to the way how the primary "config sequence" reader calls git_config_from_file(), so I do prefer the patch as posted as a short-term "oops, we merged a buggy code that segfaults and here is a fix" patch.

It however makes me wonder if it is simpler to allow passing NULL to git_config_from_file_with_options() and make it silently turn into a no-op. I.e. instead of ...

config.c int git_config_from_file_with_options(config_fn_t fn, const char *filename, · line 1983

Quoted 6 lines

config.c +2 -0

@@ -1979,6 +1979,8 @@ int git_config_from_file_with_options(config_fn_t fn, const char *filename,
 	int ret = -1;
 	FILE *f;
 
+	if (!filename)
+		BUG("filename cannot be NULL");

... we could do

if (!filename)
return 0; /* successful no-op */

Even if there are codepaths that feed arbitrary pathnames given by the end user, they wouldn't be passing NULL (they may pass an empty string, or a filename that causes fopen() to fail), would they?

But that is something we should leave to a follow-up series, not "oops, we need to fix it now" fix.

Thanks, will queue.

Quoted 3 lines

config.c +0 -0

 	f = fopen_or_warn(filename, "r");
 	if (f) {
 		ret = do_config_from_file(fn, CONFIG_ORIGIN_FILE, filename,

Glen Choo

2022-07-27 17:43 UTC #13 lore raw

Quoted 13 lines · Junio C Hamano

It however makes me wonder if it is simpler to allow passing NULL to git_config_from_file_with_options() and make it silently turn into a no-op. I.e. instead of ...
config.c int git_config_from_file_with_options(config_fn_t fn, const char *filename, · line 1983
Quoted 6 lines
config.c +2 -0
@@ -1979,6 +1979,8 @@ int git_config_from_file_with_options(config_fn_t fn, const char *filename,
 	int ret = -1;
 	FILE *f;
 
+	if (!filename)
+		BUG("filename cannot be NULL");
... we could do

Quoted 6 lines

if (!filename)
return 0; /* successful no-op */
Even if there are codepaths that feed arbitrary pathnames given by the end user, they wouldn't be passing NULL (they may pass an empty string, or a filename that causes fopen() to fail), would they?

Yeah, that's worth considering. I'm not sure how I feel about it yet, but hopefully I find some time to dig around and form an opinion.

Quoted 5 lines

But that is something we should leave to a follow-up series, not "oops, we need to fix it now" fix.
Thanks, will queue.

Thanks :)

Quoted 4 lines

config.c +0 -0

 	f = fopen_or_warn(filename, "r");
 	if (f) {
 		ret = do_config_from_file(fn, CONFIG_ORIGIN_FILE, filename,