Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1321806 > unrolled thread
| Started by | Eric Auger <eric.auger@linaro.org> |
|---|---|
| First post | 2016-01-29 15:50 +0100 |
| Last post | 2016-02-01 22:40 +0100 |
| Articles | 4 — 2 participants |
Back to article view | Back to linux.kernel
[PATCH] vfio: pci: fix oops in case of vfio_msi_set_vector_signal failure Eric Auger <eric.auger@linaro.org> - 2016-01-29 15:50 +0100
Re: [PATCH] vfio: pci: fix oops in case of vfio_msi_set_vector_signal failure Alex Williamson <alex.williamson@redhat.com> - 2016-01-29 22:50 +0100
Re: [PATCH] vfio: pci: fix oops in case of vfio_msi_set_vector_signal failure Eric Auger <eric.auger@linaro.org> - 2016-02-01 18:30 +0100
Re: [PATCH] vfio: pci: fix oops in case of vfio_msi_set_vector_signal failure Alex Williamson <alex.williamson@redhat.com> - 2016-02-01 22:40 +0100
| From | Eric Auger <eric.auger@linaro.org> |
|---|---|
| Date | 2016-01-29 15:50 +0100 |
| Subject | [PATCH] vfio: pci: fix oops in case of vfio_msi_set_vector_signal failure |
| Message-ID | <qWix5-4ym-31@gated-at.bofh.it> |
In case vfio_msi_set_vector_signal fails we tear down everything.
In the tear down loop we compare int j against unsigned start. Given
the arithmetic conversion I think it is converted into an unsigned and
becomes 0xffffffff, leading to the loop being entered again and things
turn bad when accessing vdev->msix[vector].vector. So let's use int
parameters instead.
Signed-off-by: Eric Auger <eric.auger@linaro.org>
---
drivers/vfio/pci/vfio_pci_intrs.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c
index 3b3ba15..510c48d 100644
--- a/drivers/vfio/pci/vfio_pci_intrs.c
+++ b/drivers/vfio/pci/vfio_pci_intrs.c
@@ -374,8 +374,8 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
return 0;
}
-static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
- unsigned count, int32_t *fds, bool msix)
+static int vfio_msi_set_block(struct vfio_pci_device *vdev, int start,
+ int count, int32_t *fds, bool msix)
{
int i, j, ret = 0;
--
1.9.1
[toc] | [next] | [standalone]
| From | Alex Williamson <alex.williamson@redhat.com> |
|---|---|
| Date | 2016-01-29 22:50 +0100 |
| Subject | Re: [PATCH] vfio: pci: fix oops in case of vfio_msi_set_vector_signal failure |
| Message-ID | <qWp5x-Re-9@gated-at.bofh.it> |
| In reply to | #1321806 |
On Fri, 2016-01-29 at 14:43 +0000, Eric Auger wrote:
> In case vfio_msi_set_vector_signal fails we tear down everything.
> In the tear down loop we compare int j against unsigned start. Given
> the arithmetic conversion I think it is converted into an unsigned and
> becomes 0xffffffff, leading to the loop being entered again and things
> turn bad when accessing vdev->msix[vector].vector. So let's use int
> parameters instead.
>
> Signed-off-by: Eric Auger <eric.auger@linaro.org>
> ---
> drivers/vfio/pci/vfio_pci_intrs.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c
> index 3b3ba15..510c48d 100644
> --- a/drivers/vfio/pci/vfio_pci_intrs.c
> +++ b/drivers/vfio/pci/vfio_pci_intrs.c
> @@ -374,8 +374,8 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
> return 0;
> }
>
> -static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
> - unsigned count, int32_t *fds, bool msix)
> +static int vfio_msi_set_block(struct vfio_pci_device *vdev, int start,
> + int count, int32_t *fds, bool msix)
> {
> int i, j, ret = 0;
>
Nice find, I don't think that's the only bug there though. If @start is
-1 (UINT32_MAX) and @count is 1, then @j gets set to -1 in the setup and
we hit the same index dereference problem. What if we did this instead:
diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c
index 3b3ba15..2ae84ad 100644
--- a/drivers/vfio/pci/vfio_pci_intrs.c
+++ b/drivers/vfio/pci/vfio_pci_intrs.c
@@ -309,14 +309,14 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
int vector, int fd, bool msix)
{
struct pci_dev *pdev = vdev->pdev;
- int irq = msix ? vdev->msix[vector].vector : pdev->irq + vector;
- char *name = msix ? "vfio-msix" : "vfio-msi";
struct eventfd_ctx *trigger;
- int ret;
+ int irq, ret;
- if (vector >= vdev->num_ctx)
+ if (vector < 0 || vector >= vdev->num_ctx)
return -EINVAL;
+ irq = msix ? vdev->msix[vector].vector : pdev->irq + vector;
+
if (vdev->ctx[vector].trigger) {
free_irq(irq, vdev->ctx[vector].trigger);
irq_bypass_unregister_producer(&vdev->ctx[vector].producer);
@@ -328,8 +328,9 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
if (fd < 0)
return 0;
- vdev->ctx[vector].name = kasprintf(GFP_KERNEL, "%s[%d](%s)",
- name, vector, pci_name(pdev));
+ vdev->ctx[vector].name = kasprintf(GFP_KERNEL, "vfio-msi%s[%d](%s)",
+ msix ? "x" : "", vector,
+ pci_name(pdev));
if (!vdev->ctx[vector].name)
return -ENOMEM;
@@ -379,7 +380,7 @@ static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
{
int i, j, ret = 0;
- if (start + count > vdev->num_ctx)
+ if (start >= vdev->num_ctx || start + count > vdev->num_ctx)
return -EINVAL;
for (i = 0, j = start; i < count && !ret; i++, j++) {
@@ -388,7 +389,7 @@ static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
}
if (ret) {
- for (--j; j >= start; j--)
+ for (--j; j >= 0 && j >= start; j--)
vfio_msi_set_vector_signal(vdev, j, -1, msix);
}
So we fix the problem with vfio_msi_set_vector_signal() dereferencing
the array before it validates the index (even though it shouldn't be
able to get there anymore), and then we do a better job of verifying
start and count (comparing to num_ctx will use unsigned even though
num_ctx itself is signed) and finally explicitly test the <0 case, which
I suppose we could also do by casting start at that point (we know it's
within the bounds of a signed integer given the previous tests).
Thanks,
Alex
[toc] | [prev] | [next] | [standalone]
| From | Eric Auger <eric.auger@linaro.org> |
|---|---|
| Date | 2016-02-01 18:30 +0100 |
| Subject | Re: [PATCH] vfio: pci: fix oops in case of vfio_msi_set_vector_signal failure |
| Message-ID | <qXqsy-6bL-21@gated-at.bofh.it> |
| In reply to | #1322116 |
Hi Alex,
On 01/29/2016 10:41 PM, Alex Williamson wrote:
> On Fri, 2016-01-29 at 14:43 +0000, Eric Auger wrote:
>> In case vfio_msi_set_vector_signal fails we tear down everything.
>> In the tear down loop we compare int j against unsigned start. Given
>> the arithmetic conversion I think it is converted into an unsigned and
>> becomes 0xffffffff, leading to the loop being entered again and things
>> turn bad when accessing vdev->msix[vector].vector. So let's use int
>> parameters instead.
>>
>> Signed-off-by: Eric Auger <eric.auger@linaro.org>
>> ---
>> drivers/vfio/pci/vfio_pci_intrs.c | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c
>> index 3b3ba15..510c48d 100644
>> --- a/drivers/vfio/pci/vfio_pci_intrs.c
>> +++ b/drivers/vfio/pci/vfio_pci_intrs.c
>> @@ -374,8 +374,8 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
>> return 0;
>> }
>>
>> -static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
>> - unsigned count, int32_t *fds, bool msix)
>> +static int vfio_msi_set_block(struct vfio_pci_device *vdev, int start,
>> + int count, int32_t *fds, bool msix)
>> {
>> int i, j, ret = 0;
>>
>
> Nice find, I don't think that's the only bug there though. If @start is
> -1 (UINT32_MAX) and @count is 1, then @j gets set to -1 in the setup and
> we hit the same index dereference problem. What if we did this instead:
>
> diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c
> index 3b3ba15..2ae84ad 100644
> --- a/drivers/vfio/pci/vfio_pci_intrs.c
> +++ b/drivers/vfio/pci/vfio_pci_intrs.c
> @@ -309,14 +309,14 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
> int vector, int fd, bool msix)
> {
> struct pci_dev *pdev = vdev->pdev;
> - int irq = msix ? vdev->msix[vector].vector : pdev->irq + vector;
> - char *name = msix ? "vfio-msix" : "vfio-msi";
> struct eventfd_ctx *trigger;
> - int ret;
> + int irq, ret;
>
> - if (vector >= vdev->num_ctx)
> + if (vector < 0 || vector >= vdev->num_ctx)
> return -EINVAL;
>
> + irq = msix ? vdev->msix[vector].vector : pdev->irq + vector;
> +
> if (vdev->ctx[vector].trigger) {
> free_irq(irq, vdev->ctx[vector].trigger);
> irq_bypass_unregister_producer(&vdev->ctx[vector].producer);
> @@ -328,8 +328,9 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
> if (fd < 0)
> return 0;
>
> - vdev->ctx[vector].name = kasprintf(GFP_KERNEL, "%s[%d](%s)",
> - name, vector, pci_name(pdev));
> + vdev->ctx[vector].name = kasprintf(GFP_KERNEL, "vfio-msi%s[%d](%s)",
> + msix ? "x" : "", vector,
> + pci_name(pdev));
> if (!vdev->ctx[vector].name)
> return -ENOMEM;
>
> @@ -379,7 +380,7 @@ static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
> {
> int i, j, ret = 0;
>
> - if (start + count > vdev->num_ctx)
> + if (start >= vdev->num_ctx || start + count > vdev->num_ctx)
> return -EINVAL;
>
> for (i = 0, j = start; i < count && !ret; i++, j++) {
> @@ -388,7 +389,7 @@ static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
> }
>
> if (ret) {
> - for (--j; j >= start; j--)
> + for (--j; j >= 0 && j >= start; j--)
> vfio_msi_set_vector_signal(vdev, j, -1, msix);
> }
>
>
> So we fix the problem with vfio_msi_set_vector_signal() dereferencing
> the array before it validates the index (even though it shouldn't be
> able to get there anymore), and then we do a better job of verifying
> start and count (comparing to num_ctx will use unsigned even though
> num_ctx itself is signed) and finally explicitly test the <0 case, which
> I suppose we could also do by casting start at that point (we know it's
> within the bounds of a signed integer given the previous tests).
Yes it looks OK to me.
I guess you submit? I will test it.
Best Regards
Eric
> Thanks,
>
> Alex
>
[toc] | [prev] | [next] | [standalone]
| From | Alex Williamson <alex.williamson@redhat.com> |
|---|---|
| Date | 2016-02-01 22:40 +0100 |
| Subject | Re: [PATCH] vfio: pci: fix oops in case of vfio_msi_set_vector_signal failure |
| Message-ID | <qXumu-vl-11@gated-at.bofh.it> |
| In reply to | #1323342 |
On Mon, 2016-02-01 at 18:27 +0100, Eric Auger wrote:
> Hi Alex,
> On 01/29/2016 10:41 PM, Alex Williamson wrote:
> > On Fri, 2016-01-29 at 14:43 +0000, Eric Auger wrote:
> > > In case vfio_msi_set_vector_signal fails we tear down everything.
> > > In the tear down loop we compare int j against unsigned start. Given
> > > the arithmetic conversion I think it is converted into an unsigned and
> > > becomes 0xffffffff, leading to the loop being entered again and things
> > > turn bad when accessing vdev->msix[vector].vector. So let's use int
> > > parameters instead.
> > >
> > > Signed-off-by: Eric Auger <eric.auger@linaro.org>
> > > ---
> > > drivers/vfio/pci/vfio_pci_intrs.c | 4 ++--
> > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c
> > > index 3b3ba15..510c48d 100644
> > > --- a/drivers/vfio/pci/vfio_pci_intrs.c
> > > +++ b/drivers/vfio/pci/vfio_pci_intrs.c
> > > @@ -374,8 +374,8 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
> > > return 0;
> > > }
> > >
> > > -static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
> > > - unsigned count, int32_t *fds, bool msix)
> > > +static int vfio_msi_set_block(struct vfio_pci_device *vdev, int start,
> > > + int count, int32_t *fds, bool msix)
> > > {
> > > int i, j, ret = 0;
> > >
> >
> > Nice find, I don't think that's the only bug there though. If @start is
> > -1 (UINT32_MAX) and @count is 1, then @j gets set to -1 in the setup and
> > we hit the same index dereference problem. What if we did this instead:
> >
> > diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c
> > index 3b3ba15..2ae84ad 100644
> > --- a/drivers/vfio/pci/vfio_pci_intrs.c
> > +++ b/drivers/vfio/pci/vfio_pci_intrs.c
> > @@ -309,14 +309,14 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
> > int vector, int fd, bool msix)
> > {
> > struct pci_dev *pdev = vdev->pdev;
> > - int irq = msix ? vdev->msix[vector].vector : pdev->irq + vector;
> > - char *name = msix ? "vfio-msix" : "vfio-msi";
> > struct eventfd_ctx *trigger;
> > - int ret;
> > + int irq, ret;
> >
> > - if (vector >= vdev->num_ctx)
> > + if (vector < 0 || vector >= vdev->num_ctx)
> > return -EINVAL;
> >
> > + irq = msix ? vdev->msix[vector].vector : pdev->irq + vector;
> > +
> > if (vdev->ctx[vector].trigger) {
> > free_irq(irq, vdev->ctx[vector].trigger);
> > irq_bypass_unregister_producer(&vdev->ctx[vector].producer);
> > @@ -328,8 +328,9 @@ static int vfio_msi_set_vector_signal(struct vfio_pci_device *vdev,
> > if (fd < 0)
> > return 0;
> >
> > - vdev->ctx[vector].name = kasprintf(GFP_KERNEL, "%s[%d](%s)",
> > - name, vector, pci_name(pdev));
> > + vdev->ctx[vector].name = kasprintf(GFP_KERNEL, "vfio-msi%s[%d](%s)",
> > + msix ? "x" : "", vector,
> > + pci_name(pdev));
> > if (!vdev->ctx[vector].name)
> > return -ENOMEM;
> >
> > @@ -379,7 +380,7 @@ static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
> > {
> > int i, j, ret = 0;
> >
> > - if (start + count > vdev->num_ctx)
> > + if (start >= vdev->num_ctx || start + count > vdev->num_ctx)
> > return -EINVAL;
> >
> > for (i = 0, j = start; i < count && !ret; i++, j++) {
> > @@ -388,7 +389,7 @@ static int vfio_msi_set_block(struct vfio_pci_device *vdev, unsigned start,
> > }
> >
> > if (ret) {
> > - for (--j; j >= start; j--)
> > + for (--j; j >= 0 && j >= start; j--)
> > vfio_msi_set_vector_signal(vdev, j, -1, msix);
> > }
> >
> >
> > So we fix the problem with vfio_msi_set_vector_signal() dereferencing
> > the array before it validates the index (even though it shouldn't be
> > able to get there anymore), and then we do a better job of verifying
> > start and count (comparing to num_ctx will use unsigned even though
> > num_ctx itself is signed) and finally explicitly test the <0 case, which
> > I suppose we could also do by casting start at that point (we know it's
> > within the bounds of a signed integer given the previous tests).
>
> Yes it looks OK to me.
>
> I guess you submit? I will test it.
Yep, I'll post a real patch. Thanks,
Alex
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web