https://sourceware.org/bugzilla/show_bug.cgi?id=23657

https://sourceware.org/bugzilla/show_bug.cgi?id=23657

[Umesh Kalappa]

[-- Attachment #1: Type: text/plain, Size: 244 bytes --]

Hi Kuster ,

Attached patch, fix the subjected issue (where we discard these weird
sections from the final object file) and verified the same in the
master.

Please let us know your comments/suggestions on the same.

Thank you
~Umesh

[-- Attachment #2: 0001-CVE-2019-1010180.patch --]
[-- Type: application/octet-stream, Size: 2585 bytes --]

From 17dd66f13a7f91af17d636288a257a2fa0ea8e93 Mon Sep 17 00:00:00 2001
From: Vinay Kumar <vinay.kumar@blackfigtech.com>
Date: Tue, 29 Oct 2019 12:53:37 +0000
Subject: [PATCH] CVE-2019-1010180 fix

---
 meta/recipes-devtools/gdb/gdb-8.3.1.inc       |  1 +
 .../gdb/gdb/CVE-2019-1010180.patch            | 34 +++++++++++++++++++
 2 files changed, 35 insertions(+)
 create mode 100644 meta/recipes-devtools/gdb/gdb/CVE-2019-1010180.patch

diff --git a/meta/recipes-devtools/gdb/gdb-8.3.1.inc b/meta/recipes-devtools/gdb/gdb-8.3.1.inc
index 39f1c48..aec913f 100644
--- a/meta/recipes-devtools/gdb/gdb-8.3.1.inc
+++ b/meta/recipes-devtools/gdb/gdb-8.3.1.inc
@@ -16,6 +16,7 @@ SRC_URI = "${GNU_MIRROR}/gdb/gdb-${PV}.tar.xz \
            file://0009-Change-order-of-CFLAGS.patch \
            file://0010-resolve-restrict-keyword-conflict.patch \
            file://0011-Fix-invalid-sigprocmask-call.patch \
+           file://CVE-2019-1010180.patch \
            "
 SRC_URI[md5sum] = "73b6a5d8141672c62bf851cd34c4aa83"
 SRC_URI[sha256sum] = "1e55b4d7cdca7b34be12f4ceae651623aa73b2fd640152313f9f66a7149757c4"
diff --git a/meta/recipes-devtools/gdb/gdb/CVE-2019-1010180.patch b/meta/recipes-devtools/gdb/gdb/CVE-2019-1010180.patch
new file mode 100644
index 0000000..6e52024
--- /dev/null
+++ b/meta/recipes-devtools/gdb/gdb/CVE-2019-1010180.patch
@@ -0,0 +1,34 @@
+diff --git a/gdb/ChangeLog b/gdb/ChangeLog
+index 961a2e2..1dfb0cb 100644
+--- a/gdb/ChangeLog
++++ b/gdb/ChangeLog
+@@ -1,3 +1,9 @@
++2019-10-16  Keith Seitz  <keiths@redhat.com>
++
++	PR gdb/23567
++	* dwarf2read.c (dwarf2_per_objfile::locate_sections): Discard
++	sections whose size is greater than the file size.
++
+ 2019-09-20  Joel Brobecker  <brobecker@adacore.com>
+ 
+ 	* version.in: Set GDB version number to 8.3.1.
+diff --git a/gdb/dwarf2read.c b/gdb/dwarf2read.c
+index 4251ed0..fafdb89 100644
+--- a/gdb/dwarf2read.c
++++ b/gdb/dwarf2read.c
+@@ -2378,6 +2378,15 @@ dwarf2_per_objfile::locate_sections (bfd *abfd, asection *sectp,
+   if ((aflag & SEC_HAS_CONTENTS) == 0)
+     {
+     }
++  else if (elf_section_data (sectp)->this_hdr.sh_size
++	   > bfd_get_file_size (abfd))
++    {
++      bfd_size_type size = elf_section_data (sectp)->this_hdr.sh_size;
++      warning (_("Discarding section %s which has a section size (%s"
++		 ") larger than the file size [in module %s]"),
++	       bfd_section_name (abfd, sectp), phex_nz (size, sizeof (size)),
++	       bfd_get_filename (abfd));
++    }
+   else if (section_is_p (sectp->name, &names.info))
+     {
+       this->info.s.section = sectp;
-- 
2.18.1

Re: https://sourceware.org/bugzilla/show_bug.cgi?id=23657

[Randy MacLeod]

On 10/30/19 5:25 AM, Umesh Kalappa wrote:
> Hi Kuster ,
> 
> Attached patch, fix the subjected issue (where we discard these weird
> sections from the final object file) and verified the same in the
> master.
> 
> Please let us know your comments/suggestions on the same.
> 
> Thank you
> ~Umesh
> 
> 

Hi Umesh,

Thanks for submitting the patch as I requested.
The patch content itself looks fine but since I think that
you will be involved in additional oe-core work involving
the toolchain, please read:
    https://www.openembedded.org/wiki/How_to_submit_a_patch_to_OpenEmbedded

Trevor and I will be happy to help you get the format right but we
should do that off the list initially.

In this case for example the bug should be fixed on the master branch
initially, then a separate email sent for the zeus branch.
There are examples of this on the oe-core list, such as:
    master:
 
http://lists.openembedded.org/pipermail/openembedded-core/2019-October/288405.html

    zeus:
 
http://lists.openembedded.org/pipermail/openembedded-core/2019-October/288400.html

Good luck and let me know if you have any questions.

-- 
# Randy MacLeod
# Wind River Linux