drivers/xen/privcmd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
Fake flexible arrays (zero-length and one-element arrays) are deprecated,
and should be replaced by flexible-array members. So, replace
zero-length array with a flexible-array member in `struct
privcmd_kernel_ioreq`.
Also annotate array `ports` with `__counted_by()` to prepare for the
coming implementation by GCC and Clang of the `__counted_by` attribute.
Flexible array members annotated with `__counted_by` can have their
accesses bounds-checked at run-time via `CONFIG_UBSAN_BOUNDS` (for array
indexing) and `CONFIG_FORTIFY_SOURCE` (for strcpy/memcpy-family functions).
This fixes multiple -Warray-bounds warnings:
drivers/xen/privcmd.c:1239:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=]
drivers/xen/privcmd.c:1240:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=]
drivers/xen/privcmd.c:1241:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=]
drivers/xen/privcmd.c:1245:33: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=]
drivers/xen/privcmd.c:1258:67: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=]
This results in no differences in binary output.
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
---
drivers/xen/privcmd.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/xen/privcmd.c b/drivers/xen/privcmd.c
index 1ce7f3c7a950..0eb337a8ec0f 100644
--- a/drivers/xen/privcmd.c
+++ b/drivers/xen/privcmd.c
@@ -1115,7 +1115,7 @@ struct privcmd_kernel_ioreq {
spinlock_t lock; /* Protects ioeventfds list */
struct list_head ioeventfds;
struct list_head list;
- struct ioreq_port ports[0];
+ struct ioreq_port ports[] __counted_by(vcpus);
};
static irqreturn_t ioeventfd_interrupt(int irq, void *dev_id)
--
2.34.1
On 16.11.23 19:54, Gustavo A. R. Silva wrote: > Fake flexible arrays (zero-length and one-element arrays) are deprecated, > and should be replaced by flexible-array members. So, replace > zero-length array with a flexible-array member in `struct > privcmd_kernel_ioreq`. > > Also annotate array `ports` with `__counted_by()` to prepare for the > coming implementation by GCC and Clang of the `__counted_by` attribute. > Flexible array members annotated with `__counted_by` can have their > accesses bounds-checked at run-time via `CONFIG_UBSAN_BOUNDS` (for array > indexing) and `CONFIG_FORTIFY_SOURCE` (for strcpy/memcpy-family functions). > > This fixes multiple -Warray-bounds warnings: > drivers/xen/privcmd.c:1239:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > drivers/xen/privcmd.c:1240:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > drivers/xen/privcmd.c:1241:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > drivers/xen/privcmd.c:1245:33: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > drivers/xen/privcmd.c:1258:67: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > > This results in no differences in binary output. > > Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org> Reviewed-by: Juergen Gross <jgross@suse.com> Juergen
On Thu, Nov 16, 2023 at 12:54:59PM -0600, Gustavo A. R. Silva wrote: > Fake flexible arrays (zero-length and one-element arrays) are deprecated, > and should be replaced by flexible-array members. So, replace > zero-length array with a flexible-array member in `struct > privcmd_kernel_ioreq`. > > Also annotate array `ports` with `__counted_by()` to prepare for the > coming implementation by GCC and Clang of the `__counted_by` attribute. > Flexible array members annotated with `__counted_by` can have their > accesses bounds-checked at run-time via `CONFIG_UBSAN_BOUNDS` (for array > indexing) and `CONFIG_FORTIFY_SOURCE` (for strcpy/memcpy-family functions). > > This fixes multiple -Warray-bounds warnings: > drivers/xen/privcmd.c:1239:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > drivers/xen/privcmd.c:1240:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > drivers/xen/privcmd.c:1241:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > drivers/xen/privcmd.c:1245:33: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > drivers/xen/privcmd.c:1258:67: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] > > This results in no differences in binary output. > > Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org> Looks right to me. I can see the allocation: size = struct_size(kioreq, ports, ioeventfd->vcpus); kioreq = kzalloc(size, GFP_KERNEL); if (!kioreq) return ERR_PTR(-ENOMEM); kioreq->dom = ioeventfd->dom; kioreq->vcpus = ioeventfd->vcpus; Reviewed-by: Kees Cook <keescook@chromium.org> -- Kees Cook
On 11/16/23 15:08, Kees Cook wrote: > On Thu, Nov 16, 2023 at 12:54:59PM -0600, Gustavo A. R. Silva wrote: >> Fake flexible arrays (zero-length and one-element arrays) are deprecated, >> and should be replaced by flexible-array members. So, replace >> zero-length array with a flexible-array member in `struct >> privcmd_kernel_ioreq`. >> >> Also annotate array `ports` with `__counted_by()` to prepare for the >> coming implementation by GCC and Clang of the `__counted_by` attribute. >> Flexible array members annotated with `__counted_by` can have their >> accesses bounds-checked at run-time via `CONFIG_UBSAN_BOUNDS` (for array >> indexing) and `CONFIG_FORTIFY_SOURCE` (for strcpy/memcpy-family functions). >> >> This fixes multiple -Warray-bounds warnings: >> drivers/xen/privcmd.c:1239:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] >> drivers/xen/privcmd.c:1240:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] >> drivers/xen/privcmd.c:1241:30: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] >> drivers/xen/privcmd.c:1245:33: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] >> drivers/xen/privcmd.c:1258:67: warning: array subscript i is outside array bounds of 'struct ioreq_port[0]' [-Warray-bounds=] >> >> This results in no differences in binary output. >> >> Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org> > > Looks right to me. I can see the allocation: Yep, I always check for that; in particular, the 'counter' assignment. :) Do you want me to mention it in the changelog text? > > size = struct_size(kioreq, ports, ioeventfd->vcpus); > kioreq = kzalloc(size, GFP_KERNEL); > if (!kioreq) > return ERR_PTR(-ENOMEM); > > kioreq->dom = ioeventfd->dom; > kioreq->vcpus = ioeventfd->vcpus; > > > Reviewed-by: Kees Cook <keescook@chromium.org> Thanks! -- Gustavo
© 2016 - 2024 Red Hat, Inc.