public inbox for glibc-cvs@sourceware.org help / color / mirror / Atom feed
From: Adhemerval Zanella <azanella@sourceware.org> To: glibc-cvs@sourceware.org Subject: [glibc] elf: Ignore LD_PROFILE for setuid binaries Date: Tue, 21 Nov 2023 20:49:22 +0000 (GMT) [thread overview] Message-ID: <20231121204922.725303858415@sourceware.org> (raw) https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=4a133885a7c8ae7ebe34e36fcdb353f8e94c810f commit 4a133885a7c8ae7ebe34e36fcdb353f8e94c810f Author: Adhemerval Zanella <adhemerval.zanella@linaro.org> Date: Mon Nov 6 17:25:45 2023 -0300 elf: Ignore LD_PROFILE for setuid binaries Loader does not ignore LD_PROFILE in secure-execution mode (different than man-page states [1]), rather it uses a different path (/var/profile) and ignore LD_PROFILE_OUTPUT. Allowing secure-execution profiling is already a non good security boundary, since it enables different code paths and extra OS access by the process. But by ignoring LD_PROFILE_OUTPUT, the resulting profile file might also be acceded in a racy manner since the file name does not use any process-specific information (such as pid, timing, etc.). Another side-effect is it forces lazy binding even on libraries that might be with DF_BIND_NOW. [1] https://man7.org/linux/man-pages/man8/ld.so.8.html Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org> Diff: --- elf/Makefile | 3 +++ elf/rtld.c | 8 +++----- elf/tst-env-setuid.c | 12 +++++++++++- 3 files changed, 17 insertions(+), 6 deletions(-) diff --git a/elf/Makefile b/elf/Makefile index 1af8ca4f84..414fdbdec8 100644 --- a/elf/Makefile +++ b/elf/Makefile @@ -3002,3 +3002,6 @@ $(objpfx)tst-non-directory-path.out: tst-non-directory-path.sh \ $(evaluate-test) tst-env-setuid-ARGS = -- $(host-test-program-cmd) + +# Reuse a module with a SONAME, to specific as the LD_PROFILE. +$(objpfx)tst-env-setuid: $(objpfx)tst-sonamemove-runmod2.so diff --git a/elf/rtld.c b/elf/rtld.c index 51b6d9f326..a09cf2a9df 100644 --- a/elf/rtld.c +++ b/elf/rtld.c @@ -361,6 +361,7 @@ struct rtld_global_ro _rtld_global_ro attribute_relro = ._dl_fpu_control = _FPU_DEFAULT, ._dl_pagesize = EXEC_PAGESIZE, ._dl_inhibit_cache = 0, + ._dl_profile_output = "/var/tmp", /* Function pointers. */ ._dl_debug_printf = _dl_debug_printf, @@ -2534,10 +2535,6 @@ process_envvars (struct dl_main_state *state) char *envline; char *debug_output = NULL; - /* This is the default place for profiling data file. */ - GLRO(dl_profile_output) - = &"/var/tmp\0/var/profile"[__libc_enable_secure ? 9 : 0]; - while ((envline = _dl_next_ld_env_entry (&runp)) != NULL) { size_t len = 0; @@ -2586,7 +2583,8 @@ process_envvars (struct dl_main_state *state) } /* Which shared object shall be profiled. */ - if (memcmp (envline, "PROFILE", 7) == 0 && envline[8] != '\0') + if (!__libc_enable_secure + && memcmp (envline, "PROFILE", 7) == 0 && envline[8] != '\0') GLRO(dl_profile) = &envline[8]; break; diff --git a/elf/tst-env-setuid.c b/elf/tst-env-setuid.c index ba295a6a14..76b8e1fb45 100644 --- a/elf/tst-env-setuid.c +++ b/elf/tst-env-setuid.c @@ -34,6 +34,9 @@ static char SETGID_CHILD[] = "setgid-child"; #define FILTERED_VALUE "some-filtered-value" #define UNFILTERED_VALUE "some-unfiltered-value" +/* It assumes no other programs is being profile with a library with same + SONAME using the default folder. */ +#define PROFILE_LIB "tst-sonamemove-runmod2.so" struct envvar_t { @@ -50,7 +53,7 @@ static const struct envvar_t filtered_envvars[] = { "LD_HWCAP_MASK", FILTERED_VALUE }, { "LD_LIBRARY_PATH", FILTERED_VALUE }, { "LD_PRELOAD", FILTERED_VALUE }, - { "LD_PROFILE", FILTERED_VALUE }, + { "LD_PROFILE", "tst-sonamemove-runmod2.so" }, { "MALLOC_ARENA_MAX", FILTERED_VALUE }, { "MALLOC_PERTURB_", FILTERED_VALUE }, { "MALLOC_TRACE", FILTERED_VALUE }, @@ -87,6 +90,13 @@ test_child (void) ret |= !(env != NULL && strcmp (env, e->value) == 0); } + /* Also check if no profile file was created. */ + { + char *profilepath = xasprintf ("/var/tmp/%s.profile", PROFILE_LIB); + ret |= !access (profilepath, R_OK); + free (profilepath); + } + return ret; }
reply other threads:[~2023-11-21 20:49 UTC|newest] Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20231121204922.725303858415@sourceware.org \ --to=azanella@sourceware.org \ --cc=glibc-cvs@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: linkBe 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).