From: Siddhesh Poyarekar <siddhesh@sourceware.org>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>,
libc-alpha@sourceware.org
Subject: Re: [PATCH v2 06/19] elf: Do not parse ill-formatted strings
Date: Fri, 27 Oct 2023 06:25:34 -0400 [thread overview]
Message-ID: <29b98533-e9b9-4144-b09c-2e513edfeef7@sourceware.org> (raw)
In-Reply-To: <20231017130526.2216827-7-adhemerval.zanella@linaro.org>
On 2023-10-17 09:05, Adhemerval Zanella wrote:
> Instead of ignoring ill-formatted tunable strings, first, check all the
> tunable definitions are correct and then set each tunable value. It
> means that partially invalid strings, like "key1=value1:key2=key2=value'
> or 'key1=value':key2=value2=value2' do not enable 'key1=value1'. It
> avoids possible user-defined errors in tunable definitions.
>
> Checked on x86_64-linux-gnu.
> ---
Harsher than 5/19, but fair I guess. Please send v3 with a tiny nit
fixup I've mentioned below.
Thanks,
Sid
> elf/dl-tunables.c | 50 +++++++++++++++++++++++++++++++++++-----------
> elf/tst-tunables.c | 13 ++++++++----
> 2 files changed, 47 insertions(+), 16 deletions(-)
>
> diff --git a/elf/dl-tunables.c b/elf/dl-tunables.c
> index 59bee61124..5d4b8c5bc0 100644
> --- a/elf/dl-tunables.c
> +++ b/elf/dl-tunables.c
> @@ -154,17 +154,29 @@ __tunable_set_val (tunable_id_t id, tunable_val_t *valp, tunable_num_t *minp,
> do_tunable_update_val (cur, valp, minp, maxp);
> }
>
> -/* Parse the tunable string VALSTRING. VALSTRING is a duplicated values,
> - where delimiters ':' are replaced with '\0', so string tunables are null
> - terminated. */
> -static void
> -parse_tunables (char *valstring)
> +struct tunable_toset_t
> +{
> + tunable_t *t;
> + const char *value;
> +};
> +
> +enum { tunables_list_size = array_length (tunable_list) };
> +
> +/* Parse the tunable string VALSTRING and set TUNABLES with the found tunables
> + and their respectibles values. VALSTRING is a duplicated values, where
> + delimiters ':' are replaced with '\0', so string tunables are null
> + terminated.
> + Return the number of tunables found (including 0 if the string is empty)
> + or -1 if for a ill-formatted definition. */
> +static int
> +parse_tunables_string (char *valstring, struct tunable_toset_t *tunables)
> {
> if (valstring == NULL || *valstring == '\0')
> - return;
> + return 0;
>
> char *p = valstring;
> bool done = false;
> + int ntunables = 0;
>
> while (!done)
> {
> @@ -177,7 +189,7 @@ parse_tunables (char *valstring)
> /* If we reach the end of the string before getting a valid name-value
> pair, bail out. */
> if (*p == '\0')
> - break;
> + return -1;
>
> /* We did not find a valid name-value pair before encountering the
> colon. */
> @@ -190,30 +202,44 @@ parse_tunables (char *valstring)
> /* Skip the ':' or '='. */
> p++;
>
> - const char *value = p;
> + char *value = p;
>
> while (*p != '=' && *p != ':' && *p != '\0')
> p++;
>
> if (*p == '=')
> - break;
> + return -1;
> else if (*p == '\0')
> done = true;
> else
> *p++ = '\0';
>
> /* Add the tunable if it exists. */
> - for (size_t i = 0; i < sizeof (tunable_list) / sizeof (tunable_t); i++)
> + for (size_t i = 0; i < tunables_list_size; i++)
> {
> tunable_t *cur = &tunable_list[i];
>
> if (tunable_is_name (cur->name, name))
> {
> - tunable_initialize (cur, value);
> + tunables[ntunables++] = (struct tunable_toset_t) { cur, value };
> break;
> }
> }
> }
> +
> + return ntunables;
> +}
> +
> +static void
> +parse_tunables (char *valstring)
> +{
> + struct tunable_toset_t tunables[tunables_list_size];
> + int ntunables = parse_tunables_string (valstring, tunables);
> + if (ntunables == -1)
> + return;
You don't actually need this; the for loop below will return without
doing anything if ntunables == -1.
> +
> + for (int i = 0; i < ntunables; i++)
> + tunable_initialize (tunables[i].t, tunables[i].value);
> }
>
> /* Initialize the tunables list from the environment. For now we only use the
> @@ -240,7 +266,7 @@ __tunables_init (char **envp)
> continue;
> }
>
> - for (int i = 0; i < sizeof (tunable_list) / sizeof (tunable_t); i++)
> + for (int i = 0; i < tunables_list_size; i++)
> {
> tunable_t *cur = &tunable_list[i];
>
> diff --git a/elf/tst-tunables.c b/elf/tst-tunables.c
> index 03039b5260..e124fa4c6d 100644
> --- a/elf/tst-tunables.c
> +++ b/elf/tst-tunables.c
> @@ -161,7 +161,7 @@ static const struct test_t
> 0,
> 0,
> },
> - /* If there is a ill-formatted key=value, everything after is also ignored. */
> + /* Ill-formatted tunables string is not parsed. */
> {
> "glibc.malloc.mmap_threshold=glibc.malloc.mmap_threshold=4096:glibc.malloc.check=2",
> 0,
> @@ -186,13 +186,18 @@ static const struct test_t
> 0,
> 0,
> },
> - /* Valid tunables set before ill-formatted ones are set. */
> {
> "glibc.malloc.check=2:glibc.malloc.mmap_threshold=4096=4096",
> - 2,
> 0,
> 0,
> - }
> + 0,
> + },
> + {
> + "glibc.malloc.check=2:glibc.malloc.mmap_threshold=4096=4096",
> + 0,
> + 0,
> + 0,
> + },
> };
>
> static int
next prev parent reply other threads:[~2023-10-27 10:25 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-17 13:05 [PATCH v2 00/19] Improve loader environment variable handling Adhemerval Zanella
2023-10-17 13:05 ` [PATCH v2 01/19] elf: Remove /etc/suid-debug support Adhemerval Zanella
2023-10-18 12:31 ` Siddhesh Poyarekar
2023-10-18 18:27 ` Adhemerval Zanella Netto
2023-10-17 13:05 ` [PATCH v2 02/19] elf: Add GLIBC_TUNABLES to unsecvars Adhemerval Zanella
2023-10-18 12:52 ` Siddhesh Poyarekar
2023-10-17 13:05 ` [PATCH v2 03/19] elf: Ignore GLIBC_TUNABLES for setuid/setgid binaries Adhemerval Zanella
2023-10-18 13:04 ` Siddhesh Poyarekar
2023-10-27 10:38 ` Siddhesh Poyarekar
2023-10-28 2:14 ` DJ Delorie
2023-10-30 16:51 ` Adhemerval Zanella Netto
2023-10-17 13:05 ` [PATCH v2 04/19] elf: Add all malloc tunable to unsecvars Adhemerval Zanella
2023-10-18 13:04 ` Siddhesh Poyarekar
2023-10-27 10:38 ` Siddhesh Poyarekar
2023-10-28 3:45 ` DJ Delorie
2023-10-17 13:05 ` [PATCH v2 05/19] elf: Do not process invalid tunable format Adhemerval Zanella
2023-10-18 13:42 ` Siddhesh Poyarekar
2023-10-17 13:05 ` [PATCH v2 06/19] elf: Do not parse ill-formatted strings Adhemerval Zanella
2023-10-27 10:25 ` Siddhesh Poyarekar [this message]
2023-10-27 12:51 ` Adhemerval Zanella Netto
2023-10-17 13:05 ` [PATCH v2 07/19] elf: Fix _dl_debug_vdprintf to work before self-relocation Adhemerval Zanella
2023-10-27 10:27 ` Siddhesh Poyarekar
2023-10-17 13:05 ` [PATCH v2 08/19] elf: Emit warning if tunable is ill-formatted Adhemerval Zanella
2023-10-27 10:28 ` Siddhesh Poyarekar
2023-10-27 12:51 ` Adhemerval Zanella Netto
2023-10-17 13:05 ` [PATCH v2 09/19] x86: Use dl-symbol-redir-ifunc.h on cpu-tunables Adhemerval Zanella
2023-10-27 10:32 ` Siddhesh Poyarekar
2023-10-17 13:05 ` [PATCH v2 10/19] s390: " Adhemerval Zanella
2023-10-27 10:34 ` Siddhesh Poyarekar
2023-10-17 13:05 ` [PATCH v2 11/19] elf: Do not duplicate the GLIBC_TUNABLES string Adhemerval Zanella
2023-10-27 11:54 ` Siddhesh Poyarekar
2023-10-30 18:24 ` Adhemerval Zanella Netto
2023-10-17 13:05 ` [PATCH v2 12/19] elf: Ignore LD_PROFILE for setuid binaries Adhemerval Zanella
2023-10-27 14:25 ` Siddhesh Poyarekar
2023-12-12 10:11 ` Florian Weimer
2023-10-17 13:05 ` [PATCH v2 13/19] elf: Remove LD_PROFILE for static binaries Adhemerval Zanella
2023-10-17 13:05 ` [PATCH v2 14/19] elf: Ignore loader debug env vars for setuid Adhemerval Zanella
2023-10-17 13:05 ` [PATCH v2 15/19] elf: Remove any_debug from dl_main_state Adhemerval Zanella
2023-10-27 14:26 ` Siddhesh Poyarekar
2023-10-30 19:03 ` Adhemerval Zanella Netto
2023-10-17 13:05 ` [PATCH v2 16/19] elf: Ignore LD_LIBRARY_PATH and debug env var for setuid for static Adhemerval Zanella
2023-10-17 13:05 ` [PATCH v2 17/19] elf: Add comments on how LD_AUDIT and LD_PRELOAD handle __libc_enable_secure Adhemerval Zanella
2023-10-27 14:31 ` Siddhesh Poyarekar
2023-10-17 13:05 ` [PATCH v2 18/19] elf: Ignore LD_BIND_NOW and LD_BIND_NOT for setuid binaries Adhemerval Zanella
2023-10-17 13:05 ` [PATCH v2 19/19] elf: Refactor process_envvars Adhemerval Zanella
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=29b98533-e9b9-4144-b09c-2e513edfeef7@sourceware.org \
--to=siddhesh@sourceware.org \
--cc=adhemerval.zanella@linaro.org \
--cc=libc-alpha@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).