This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
[review v3] Block signals during the initial part of dlopen
- From: "Florian Weimer (Code Review)" <gerrit at gnutoolchain-gerrit dot osci dot io>
- To: Florian Weimer <fweimer at redhat dot com>, Christian Brauner <christian dot brauner at ubuntu dot com>, libc-alpha at sourceware dot org
- Date: Fri, 15 Nov 2019 11:02:04 -0500
- Subject: [review v3] Block signals during the initial part of dlopen
- Auto-submitted: auto-generated
- References: <gerrit.1572549639000.Iad079080ebe7442c13313ba11dc2797953faef35@gnutoolchain-gerrit.osci.io>
- Reply-to: fweimer at redhat dot com, fweimer at redhat dot com, libc-alpha at sourceware dot org, christian dot brauner at ubuntu dot com
Change URL: https://gnutoolchain-gerrit.osci.io/r/c/glibc/+/472
......................................................................
Block signals during the initial part of dlopen
Lazy binding in a signal handler that interrupts a dlopen sees
intermediate dynamic linker state. This has likely been always
unsafe, but with the new pending NODELETE state, this is clearly
incorrect. Other threads are excluded via the loader lock, but the
current thread is not. Blocking signals until right before ELF
constructors run is the safe thing to do.
Change-Id: Iad079080ebe7442c13313ba11dc2797953faef35
---
M elf/dl-open.c
1 file changed, 26 insertions(+), 2 deletions(-)
diff --git a/elf/dl-open.c b/elf/dl-open.c
index db870ed..b9b4319 100644
--- a/elf/dl-open.c
+++ b/elf/dl-open.c
@@ -34,6 +34,7 @@
#include <atomic.h>
#include <libc-internal.h>
#include <array_length.h>
+#include <internal-signals.h>
#include <dl-dst.h>
#include <dl-prop.h>
@@ -52,6 +53,10 @@
/* Namespace ID. */
Lmid_t nsid;
+ /* Original signal mask. Used for unblocking signal handlers before
+ running ELF constructors. */
+ sigset_t original_signal_mask;
+
/* Original value of _ns_global_scope_pending_adds. Set by
dl_open_worker. Only valid if nsid is a real namespace
(non-negative). */
@@ -510,12 +515,16 @@
if (new == NULL)
{
assert (mode & RTLD_NOLOAD);
+ __libc_signal_restore_set (&args->original_signal_mask);
return;
}
if (__glibc_unlikely (mode & __RTLD_SPROF))
- /* This happens only if we load a DSO for 'sprof'. */
- return;
+ {
+ /* This happens only if we load a DSO for 'sprof'. */
+ __libc_signal_restore_set (&args->original_signal_mask);
+ return;
+ }
/* This object is directly loaded. */
++new->l_direct_opencount;
@@ -551,6 +560,7 @@
assert (_dl_debug_initialize (0, args->nsid)->r_state == RT_CONSISTENT);
+ __libc_signal_restore_set (&args->original_signal_mask);
return;
}
@@ -733,6 +743,10 @@
if (mode & RTLD_GLOBAL)
add_to_global_resize (new);
+ /* Unblock signals. Data structures are now consistent, and
+ application code may run. */
+ __libc_signal_restore_set (&args->original_signal_mask);
+
/* Run the initializer functions of new objects. Temporarily
disable the exception handler, so that lazy binding failures are
fatal. */
@@ -822,6 +836,10 @@
args.argv = argv;
args.env = env;
+ /* Recursive lazy binding during manipulation of the dynamic loader
+ structures may result in incorrect behavior. */
+ __libc_signal_block_all (&args.original_signal_mask);
+
struct dl_exception exception;
int errcode = _dl_catch_exception (&exception, dl_open_worker, &args);
@@ -862,10 +880,16 @@
_dl_close_worker (args.map, true);
+ /* Restore the signal mask. In the success case, this
+ happens inside dl_open_worker. */
+ __libc_signal_restore_set (&args.original_signal_mask);
+
/* All link_map_nodelete_pending objects should have been
deleted at this point, which is why it is not necessary
to reset the flag here. */
}
+ else
+ __libc_signal_restore_set (&args.original_signal_mask);
assert (_dl_debug_initialize (0, args.nsid)->r_state == RT_CONSISTENT);
--
Gerrit-Project: glibc
Gerrit-Branch: master
Gerrit-Change-Id: Iad079080ebe7442c13313ba11dc2797953faef35
Gerrit-Change-Number: 472
Gerrit-PatchSet: 3
Gerrit-Owner: Florian Weimer <fweimer@redhat.com>
Gerrit-Reviewer: Christian Brauner <christian.brauner@ubuntu.com>
Gerrit-Reviewer: Florian Weimer <fweimer@redhat.com>
Gerrit-MessageType: newpatchset