diff --git a/kexec-tools.spec b/kexec-tools.spec
index 92284f8..7b2b692 100644
--- a/kexec-tools.spec
+++ b/kexec-tools.spec
@@ -42,6 +42,7 @@ Source31: kdump-logger.sh
Source32: mkfadumprd
Source33: 92-crashkernel.install
Source34: crashkernel-howto.txt
+Source35: supported-kdump-targets.txt
#######################################
# These are sources for mkdumpramfs
@@ -152,6 +153,7 @@ cp %{SOURCE21} .
cp %{SOURCE26} .
cp %{SOURCE27} .
cp %{SOURCE34} .
+cp %{SOURCE35} .
make
%ifarch %{ix86} x86_64 ppc64 s390x ppc64le aarch64
@@ -374,6 +376,7 @@ done
%doc kdump-in-cluster-environment.txt
%doc live-image-kdump-howto.txt
%doc crashkernel-howto.txt
+%doc supported-kdump-targets.txt
%ifarch %{ix86} x86_64 ppc64 s390x ppc64le aarch64
%{_libdir}/eppic_makedumpfile.so
/usr/share/makedumpfile/
diff --git a/supported-kdump-targets.txt b/supported-kdump-targets.txt
new file mode 100644
index 0000000..f540f59
--- /dev/null
+++ b/supported-kdump-targets.txt
@@ -0,0 +1,128 @@
+Supported Kdump Targets
+
+This document try to list all supported kdump targets, and those supported
+or unknown/tech-preview targets, this can help users to decide whether a dump
+solution is available.
+
+Dump Target support status
+==========================
+This section tries to come up with some kind of guidelines in terms of
+what dump targets are supported/not supported. Whatever is listed here
+is not binding in any manner. It is just sharing of current understanding
+and if something is not right, this section needs to be edited.
+
+Following are 3 lists. First one contains supported targets. These are
+generic configurations which should work and some configuration most
+likely has worked in testing. Second list is known unsupported targets.
+These targets we know either don't work or we don't support. And third
+list is unknown/tech-preview. We either don't yet know the status of kdump
+on these targets or these are under tech-preview.
+
+Note, these lists are not set in stone and can be changed at any point of
+time. Also these lists might not be complete. We will add/remove items to
+it as we get more testing information. Also, there are many corner cases
+which can't possibly be listed. For example in general we might be
+supporting software iscsi but there might be some configurations of it
+which don't work.
+
+So if any target is listed in supported section, it does not mean it works
+in all possible configurations. It just means that in common configurations
+it should work but there can be issues with particular configurations which
+are not supported. As we come to know of particular issues, we will keep on
+updating lists accordingly.
+
+
+Supported Dump targets
+----------------------
+storage:
+ LVM volume (no thinp)
+ FC disks (qla2xxx, lpfc, bnx2fc, bfa)
+ software initiator based iSCSI
+ software RAID (mdraid)
+ hardware RAID (cciss, hpsa, megaraid_sas, mpt2sas, aacraid)
+ SCSI/SATA disks
+ iSCSI HBA (all offload)
+ hardware FCoE (qla2xxx, lpfc)
+ software FCoE (bnx2fc) (Extra configuration required,
+ please read "Note on FCoE" section below)
+
+network:
+ Hardware using kernel modules: (tg3, igb, ixgbe, sfc, e1000e, bna,
+ cnic, netxen_nic, qlge, bnx2x, bnx, qlcnic, be2net, enic,
+ virtio-net, ixgbevf, igbvf)
+ protocol: ipv4
+ bonding
+ vlan
+ bridge
+ team
+ vlan tagged bonding
+ bridge over bond/team/vlan
+
+hypervisor:
+ kvm
+ xen (Supported in select configurations only)
+
+filesystem:
+ ext[234]
+ xfs
+ nfs
+
+firmware:
+ BIOS
+ UEFI
+
+hypervisor:
+ VMWare ESXi 4.1 and 5.1
+ Hyper-V 2012 R2 (RHEL Gen1 UP Guest only)
+
+Unsupported Dump targets
+------------------------
+storage:
+ BIOS RAID
+ Thin provisioning volume
+ Software iSCSI with iBFT (bnx2i, cxgb3i, cxgb4i)
+ Software iSCSI with hybrid (be2iscsi)
+ FCoE
+ legacy IDE
+ glusterfs
+ gfs2/clvm/halvm
+
+network:
+ hardware using kernel modules: (sfc SRIOV, cxgb4vf, pch_gbe)
+ protocol: ipv6
+ wireless
+ Infiniband (IB)
+ vlan over bridge/team
+
+filesystem:
+ btrfs
+
+Unknown/tech-preview
+--------------------
+storage:
+ PCI Express based SSDs
+
+hypervisor:
+ Hyper-V 2008
+ Hyper-V 2012
+
+
+Note on FCoE
+=====================
+If you are trying to dump to a software FCoE target, you may encounter OOM
+issue, because some software FCoE requires more memory to work. In such case,
+you may need to increase the kdump reserved memory size in "crashkernel="
+kernel parameter.
+
+By default, RHEL systems have "crashkernel=auto" in kernel boot arguments.
+The auto reserved memory size is designed to balance the coverage of use cases
+and an acceptable memory overhead, so not every use case could fit in, software
+FCoE is one of the case.
+
+For hardware FCoE, kdump should work naturally as firmware will do the
+initialization job. The capture kernel and kdump tools will run just fine.
+
+Useful Links
+============
+[1] RHEL6: Enabling kdump for full-virt (HVM) Xen DomU
+ (https://access.redhat.com/knowledge/solutions/92943)