[ Upstream commit 3fb0fdb3bbe7aed495109b3296b06c2409734023 ]
On 32-bit kernels, the stackprotector canary is quite nasty -- it is
stored at %gs:(20), which is nasty because 32-bit kernels use %fs for
percpu storage. It's even nastier because it means that whether %gs
contains userspace state or kernel state while running kernel code
depends on whether stackprotector is enabled (this is
CONFIG_X86_32_LAZY_GS), and this setting radically changes the way
that segment selectors work. Supporting both variants is a
maintenance and testing mess.
Merely rearranging so that percpu and the stack canary
share the same segment would be messy as the 32-bit percpu address
layout isn't currently compatible with putting a variable at a fixed
offset.
Fortunately, GCC 8.1 added options that allow the stack canary to be
accessed as %fs:__stack_chk_guard, effectively turning it into an ordinary
percpu variable. This lets us get rid of all of the code to manage the
stack canary GDT descriptor and the CONFIG_X86_32_LAZY_GS mess.
(That name is special. We could use any symbol we want for the
%fs-relative mode, but for CONFIG_SMP=n, gcc refuses to let us use any
name other than __stack_chk_guard.)
Forcibly disable stackprotector on older compilers that don't support
the new options and turn the stack canary into a percpu variable. The
"lazy GS" approach is now used for all 32-bit configurations.
Also makes load_gs_index() work on 32-bit kernels. On 64-bit kernels,
it loads the GS selector and updates the user GSBASE accordingly. (This
is unchanged.) On 32-bit kernels, it loads the GS selector and updates
GSBASE, which is now always the user base. This means that the overall
effect is the same on 32-bit and 64-bit, which avoids some ifdeffery.
[ bp: Massage commit message. ]
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/c0ff7dba14041c7e5d1cae5d4df052f03759bef3.1613243844.git.luto@kernel.org
Stable-dep-of: e3f269ed0acc ("x86/pm: Work around false positive kmemleak report in msr_build_context()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
This involves initializing the boot params EFI related fields and the
efi global variable.
Without this fix a PVH dom0 doesn't detect when booted from EFI, and
thus doesn't support accessing any of the EFI related data.
Reported-by: PGNet Dev <pgnet.dev@gmail.com>
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: stable@vger.kernel.org # 4.19+
For certain applications it is desirable to rapidly boot a KVM virtual
machine. In cases where legacy hardware and software support within the
guest is not needed, Qemu should be able to boot directly into the
uncompressed Linux kernel binary without the need to run firmware.
There already exists an ABI to allow this for Xen PVH guests and the ABI
is supported by Linux and FreeBSD:
https://xenbits.xen.org/docs/unstable/misc/pvh.html
This patch enables Qemu to use that same entry point for booting KVM
guests.
Signed-off-by: Maran Wilson <maran.wilson@oracle.com>
Suggested-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Suggested-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Tested-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
We need to refactor PVH entry code so that support for other hypervisors
like Qemu/KVM can be added more easily.
The original design for PVH entry in Xen guests relies on being able to
obtain the memory map from the hypervisor using a hypercall. When we
extend the PVH entry ABI to support other hypervisors like Qemu/KVM,
a new mechanism will be added that allows the guest to get the memory
map without needing to use hypercalls.
For Xen guests, the hypercall approach will still be supported. In
preparation for adding support for other hypervisors, we can move the
code that uses hypercalls into the Xen specific file. This will allow us
to compile kernels in the future without CONFIG_XEN that are still capable
of being booted as a Qemu/KVM guest via the PVH entry point.
Signed-off-by: Maran Wilson <maran.wilson@oracle.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
We need to refactor PVH entry code so that support for other hypervisors
like Qemu/KVM can be added more easily.
This patch moves the small block of code used for initializing Xen PVH
virtual machines into the Xen specific file. This initialization is not
going to be needed for Qemu/KVM guests. Moving it out of the common file
is going to allow us to compile kernels in the future without CONFIG_XEN
that are still capable of being booted as a Qemu/KVM guest via the PVH
entry point.
Signed-off-by: Maran Wilson <maran.wilson@oracle.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
We need to refactor PVH entry code so that support for other hypervisors
like Qemu/KVM can be added more easily.
The first step in that direction is to create a new file that will
eventually hold the Xen specific routines.
Signed-off-by: Maran Wilson <maran.wilson@oracle.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Once hypervisors other than Xen start using the PVH entry point for
starting VMs, we would like the option of being able to compile PVH entry
capable kernels without enabling CONFIG_XEN and all the code that comes
along with that. To allow that, we are moving the PVH code out of Xen and
into files sitting at a higher level in the tree.
This patch is not introducing any code or functional changes, just moving
files from one location to another.
Signed-off-by: Maran Wilson <maran.wilson@oracle.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
