public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* [PATCH] Avoid collect2 calling signal unsafe functions and/or unlink, with uninitialized memory (for gcc-9 branch)
@ 2020-02-21 16:07 Bernd Edlinger
  2020-02-24 11:32 ` Richard Biener
  0 siblings, 1 reply; 2+ messages in thread
From: Bernd Edlinger @ 2020-02-21 16:07 UTC (permalink / raw)
  To: gcc-patches, Richard Biener, Jakub Jelinek

[-- Attachment #1: Type: text/plain, Size: 482 bytes --]

Hi,

this fixes the signal handler calling signal unsafe vfprintf and/or passing
uninitialized memory to unlink in signal handler.

This time it is the patch for the gcc-9 branch.

The difference to the gcc-8 branch is in tool_cleanup:
The variable that suppress the vfprintf is verbose = false;
not debug = false, to match the different logic in maybe_unlink.


Bootstrapped and reg-tested with x86_64-pc-linux-gnu.
Is it OK for the gcc-9 branch?


Thanks
Bernd.

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-Avoid-collect2-calling-signal-unsafe-functions-and-o.patch --]
[-- Type: text/x-patch; name="0001-Avoid-collect2-calling-signal-unsafe-functions-and-o.patch", Size: 1485 bytes --]

From d52ac2c0394f0432e183511f8a6d4f96b49f88a5 Mon Sep 17 00:00:00 2001
From: Bernd Edlinger <bernd.edlinger@hotmail.de>
Date: Mon, 17 Feb 2020 17:40:07 +0100
Subject: [PATCH] Avoid collect2 calling signal unsafe functions and/or unlink
 with uninitialized memory

2020-02-20  Bernd Edlinger  <bernd.edlinger@hotmail.de>

	* collect2.c (tool_cleanup): Avoid calling not signal-safe
	functions.
	(maybe_run_lto_and_relink): Avoid possible signal handler
	access to unintialzed memory (lto_o_files).
---
 gcc/collect2.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/gcc/collect2.c b/gcc/collect2.c
index eb84f84..8f092e7 100644
--- a/gcc/collect2.c
+++ b/gcc/collect2.c
@@ -384,6 +384,10 @@ static void scan_prog_file (const char *, scanpass, scanfilter);
 void
 tool_cleanup (bool from_signal)
 {
+  /* maybe_unlink may call notice, which is not signal safe.  */
+  if (from_signal)
+    verbose = false;
+
   if (c_file != 0 && c_file[0])
     maybe_unlink (c_file);
 
@@ -743,7 +747,10 @@ maybe_run_lto_and_relink (char **lto_ld_argv, char **object_lst,
 	      ++num_files;
 	  }
 
-	lto_o_files = XNEWVEC (char *, num_files + 1);
+	/* signal handler may access uninitialized memory
+	   and delete whatever it points to, if lto_o_files
+	   is not allocated with calloc.  */
+	lto_o_files = XCNEWVEC (char *, num_files + 1);
 	lto_o_files[num_files] = NULL;
 	start = XOBFINISH (&temporary_obstack, char *);
 	for (i = 0; i < num_files; ++i)
-- 
2.7.4


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH] Avoid collect2 calling signal unsafe functions and/or unlink, with uninitialized memory (for gcc-9 branch)
  2020-02-21 16:07 [PATCH] Avoid collect2 calling signal unsafe functions and/or unlink, with uninitialized memory (for gcc-9 branch) Bernd Edlinger
@ 2020-02-24 11:32 ` Richard Biener
  0 siblings, 0 replies; 2+ messages in thread
From: Richard Biener @ 2020-02-24 11:32 UTC (permalink / raw)
  To: Bernd Edlinger; +Cc: gcc-patches, Jakub Jelinek

On Fri, 21 Feb 2020, Bernd Edlinger wrote:

> Hi,
> 
> this fixes the signal handler calling signal unsafe vfprintf and/or passing
> uninitialized memory to unlink in signal handler.
> 
> This time it is the patch for the gcc-9 branch.
> 
> The difference to the gcc-8 branch is in tool_cleanup:
> The variable that suppress the vfprintf is verbose = false;
> not debug = false, to match the different logic in maybe_unlink.
> 
> 
> Bootstrapped and reg-tested with x86_64-pc-linux-gnu.
> Is it OK for the gcc-9 branch?

OK.

Richard.

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2020-02-24 11:32 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-02-21 16:07 [PATCH] Avoid collect2 calling signal unsafe functions and/or unlink, with uninitialized memory (for gcc-9 branch) Bernd Edlinger
2020-02-24 11:32 ` Richard Biener

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).