public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Kirill Radkin <kirill.radkin@syntacore.com>
To: "gdb-patches@sourceware.org" <gdb-patches@sourceware.org>
Cc: Konstantin Vladimirov <konstantin.vladimirov@syntacore.com>,
	"Ivan Tetyushkin" <ivan.tetyushkin@syntacore.com>
Subject: [PATCH] [gdbserver] Fix overflow detection in gdbserver
Date: Thu, 21 Dec 2023 08:38:49 +0000	[thread overview]
Message-ID: <f4e42985699247b2ab9c2635770063d6@syntacore.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 6037 bytes --]

Hi,


Recently I have encountered an issue with gdbserver not being able to send debug information from large files (> 2Gb). I found that this issue is connected with parsing offsets (integer) from vFile::pread packets. There is already a patch (https://sourceware.org/bugzilla/show_bug.cgi?id=23198) that allows to parse integers up to 0x7fffffff (32-bit integer), but it doesn't fix the problem at all, because offsets can have a off_t type which can be larger than 32-bit integer (it depends on system).

This patch allows require_int() function to parse offset up to the maximum value implied by the off_t type.


From 01117d757a0eb4f632aded4b15a06dbcccc7adf7 Mon Sep 17 00:00:00 2001
From: Kirill Radkin <kirill.radkin@syntacore.com>
Date: Mon, 13 Nov 2023 16:27:15 +0300
Subject: [PATCH] gdbserver: Fix overflow detection in gdbserver

Currently gdbserver use require_int() function to
parse offset which we get, for example, in vFile::pread
packet. This function allows integers up to 0x7fffffff,
(to fit in 32-bit int) but actually offset (for pread
system call) have a off_t type which can be larger than
32-bit.

This patch allows require_int() function to parse offset
up to the maximum value implied by the off_t type.
---
 gdb/testsuite/gdb.server/pread-offset-size.S  | 26 ++++++++++
 .../gdb.server/pread-offset-size.exp          | 47 +++++++++++++++++++
 gdbserver/hostio.cc                           | 16 +++++--
 3 files changed, 85 insertions(+), 4 deletions(-)
 create mode 100644 gdb/testsuite/gdb.server/pread-offset-size.S
 create mode 100644 gdb/testsuite/gdb.server/pread-offset-size.exp

diff --git a/gdb/testsuite/gdb.server/pread-offset-size.S b/gdb/testsuite/gdb.server/pread-offset-size.S
new file mode 100644
index 00000000000..31748090ac3
--- /dev/null
+++ b/gdb/testsuite/gdb.server/pread-offset-size.S
@@ -0,0 +1,26 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2023-2023 Free Software Foundation, Inc.
+
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 3 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
+
+ .text
+ .globl _start
+_start:
+ .skip 3742415472
+   ret
+ .globl f
+ .type f, @function
+f:
+   ret
diff --git a/gdb/testsuite/gdb.server/pread-offset-size.exp b/gdb/testsuite/gdb.server/pread-offset-size.exp
new file mode 100644
index 00000000000..a4b648b29fc
--- /dev/null
+++ b/gdb/testsuite/gdb.server/pread-offset-size.exp
@@ -0,0 +1,47 @@
+# Copyright (C) 2023-2023 Free Software Foundation, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+load_lib gdbserver-support.exp
+
+if {[skip_gdbserver_tests]} {
+    return
+}
+
+standard_testfile .S
+
+if { [prepare_for_testing ${testfile}.exp $testfile \
+      $srcfile {debug additional_flags=-nostdlib} ] } {
+    return -1
+}
+
+gdb_exit
+gdb_start
+
+gdb_test_no_output "set remote exec-file $binfile" \
+"set remote exec-file"
+
+# Make sure we're disconnected, in case we're testing with an
+# extended-remote board, therefore already connected.
+gdb_test "disconnect" ".*"
+
+set res [gdbserver_spawn ""]
+set gdbserver_protocol [lindex $res 0]
+set gdbserver_gdbport [lindex $res 1]
+
+gdb_test "target $gdbserver_protocol $gdbserver_gdbport" \
+"Remote debugging using .*" \
+"target $gdbserver_protocol $gdbserver_gdbport"
+
+gdb_test "break f" "Breakpoint 1.*"
diff --git a/gdbserver/hostio.cc b/gdbserver/hostio.cc
index ea70c26da0f..068771428f9 100644
--- a/gdbserver/hostio.cc
+++ b/gdbserver/hostio.cc
@@ -90,12 +90,16 @@ require_filename (char **pp, char *filename)
   return 0;
 }

+template <typename T>
 static int
-require_int (char **pp, int *value)
+require_int (char **pp, T *value)
 {
   char *p;
   int count, firstdigit;

+  /* Max count of hexadecimal digits in off_t (1 hex digit is 4 bits) */
+  int max_count = sizeof(T) * CHAR_BIT / 4;
+
   p = *pp;
   *value = 0;
   count = 0;
@@ -112,7 +116,9 @@ require_int (char **pp, int *value)
  firstdigit = nib;

       /* Don't allow overflow.  */
-      if (count >= 8 || (count == 7 && firstdigit >= 0x8))
+      if (count >= max_count || (static_cast<T>(-1) < 0
+ && count == (max_count - 1)
+ && firstdigit >= 0x8))
  return -1;

       *value = *value * 16 + nib;
@@ -344,7 +350,8 @@ handle_open (char *own_buf)
 static void
 handle_pread (char *own_buf, int *new_packet_len)
 {
-  int fd, ret, len, offset, bytes_sent;
+  int fd, ret, len, bytes_sent;
+  off_t offset;
   char *p, *data;
   static int max_reply_size = -1;

@@ -411,7 +418,8 @@ handle_pread (char *own_buf, int *new_packet_len)
 static void
 handle_pwrite (char *own_buf, int packet_len)
 {
-  int fd, ret, len, offset;
+  int fd, ret, len;
+  off_t offset;
   char *p, *data;

   p = own_buf + strlen ("vFile:pwrite:");
--
2.34.1

             reply	other threads:[~2023-12-21  8:38 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-21  8:38 Kirill Radkin [this message]
2023-12-21 19:42 ` Simon Marchi
2023-12-22 15:41   ` Tom Tromey
2024-01-10 14:57     ` [PATCH] gdbserver: " Kirill Radkin
2024-02-08 16:19       ` Tom Tromey
2024-02-19 14:19         ` [PATCH v2] " Kirill Radkin

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=f4e42985699247b2ab9c2635770063d6@syntacore.com \
    --to=kirill.radkin@syntacore.com \
    --cc=gdb-patches@sourceware.org \
    --cc=ivan.tetyushkin@syntacore.com \
    --cc=konstantin.vladimirov@syntacore.com \
    /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).