Re: [PATCH v2] drivers/virt/fsl_hypervisor: Correcting error handling path

[Souptick Joarder]

On Thu, Jul 30, 2020 at 6:30 AM John Hubbard  wrote:
>
> On 7/29/20 12:01 PM, Souptick Joarder wrote:
> > First, when memory allocation for sg_list_unaligned failed, there
> > is no point of calling put_pages() as we haven't pinned any pages.
> >
> > Second, if get_user_pages_fast() failed we should unpinned num_pinned
> > pages, no point of checking till num_pages.
>
> Hi Souptick,
>
> For both of the above, the wording "no point" is so overly gentle as
> to be misleading. That's because calling put_page() on any pages beyond
> num_pinned is a *bug*.
>
> So let's reword that. And let's change the patch subject from "Correcting" to
> "fix".
>
> And probably good to add a Fixes: tag, too.

Is there any scripts/ settings to fetch Fixes: tag other than using git blame ?

>
> More:
>
> >
> > This will address both.
> >
> > Signed-off-by: Souptick Joarder 
> > Reviewed-by: Dan Carpenter 
> > Cc: John Hubbard 
> > ---
> > v2:
> >   Added review tag.
> >
> >   drivers/virt/fsl_hypervisor.c | 4 ++--
> >   1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/virt/fsl_hypervisor.c b/drivers/virt/fsl_hypervisor.c
> > index 1b0b11b..ea344d7 100644
> > --- a/drivers/virt/fsl_hypervisor.c
> > +++ b/drivers/virt/fsl_hypervisor.c
> > @@ -157,7 +157,7 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy 
> > __user *p)
> >
> >   unsigned int i;
> >   long ret = 0;
> > - int num_pinned; /* return value from get_user_pages() */
> > + int num_pinned = 0; /* return value from get_user_pages() */
> >   phys_addr_t remote_paddr; /* The next address in the remote buffer */
> >   uint32_t count; /* The number of bytes left to copy */
> >
> > @@ -293,7 +293,7 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy 
> > __user *p)
> >
> >   exit:
> >   if (pages) {
> > - for (i = 0; i < num_pages; i++)
> > + for (i = 0; i < num_pinned; i++)
> >   if (pages[i])
>
> I suspect that this "if" is unnecessary now.
>
> Either way, the diff itself looks good to me, so with the wording changes to
> the commit description, you can add:
>
>  Reviewed-by: John Hubbard 
>
> thanks,
> --
> John Hubbard
> NVIDIA
>
> >   put_page(pages[i]);
> >   }
> >
>

Re: [PATCH v2] drivers/virt/fsl_hypervisor: Correcting error handling path

[John Hubbard]

On 7/29/20 12:01 PM, Souptick Joarder wrote:

First, when memory allocation for sg_list_unaligned failed, there
is no point of calling put_pages() as we haven't pinned any pages.

Second, if get_user_pages_fast() failed we should unpinned num_pinned
pages, no point of checking till num_pages.


Hi Souptick,

For both of the above, the wording "no point" is so overly gentle as
to be misleading. That's because calling put_page() on any pages beyond
num_pinned is a *bug*.

So let's reword that. And let's change the patch subject from "Correcting" to
"fix".

And probably good to add a Fixes: tag, too.

More:



This will address both.

Signed-off-by: Souptick Joarder 
Reviewed-by: Dan Carpenter 
Cc: John Hubbard 
---
v2:
Added review tag.

  drivers/virt/fsl_hypervisor.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/virt/fsl_hypervisor.c b/drivers/virt/fsl_hypervisor.c
index 1b0b11b..ea344d7 100644
--- a/drivers/virt/fsl_hypervisor.c
+++ b/drivers/virt/fsl_hypervisor.c
@@ -157,7 +157,7 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy __user 
*p)
  
  	unsigned int i;

long ret = 0;
-   int num_pinned; /* return value from get_user_pages() */
+   int num_pinned = 0; /* return value from get_user_pages() */
phys_addr_t remote_paddr; /* The next address in the remote buffer */
uint32_t count; /* The number of bytes left to copy */
  
@@ -293,7 +293,7 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy __user *p)
  
  exit:

if (pages) {
-   for (i = 0; i < num_pages; i++)
+   for (i = 0; i < num_pinned; i++)
if (pages[i])


I suspect that this "if" is unnecessary now.

Either way, the diff itself looks good to me, so with the wording changes to
the commit description, you can add:

Reviewed-by: John Hubbard 

thanks,
--
John Hubbard
NVIDIA


put_page(pages[i]);
}

Re: [PATCH v2] drivers/virt/fsl_hypervisor: Correcting error handling path

[Souptick Joarder]

On Mon, Jun 22, 2020 at 10:05 PM Souptick Joarder  wrote:
>
> On Wed, Jun 17, 2020 at 8:22 AM Souptick Joarder  wrote:
> >
> > First, when memory allocation for sg_list_unaligned failed, there
> > is no point of calling put_pages() as we haven't pinned any pages.
> >
> > Second, if get_user_pages_fast() failed we should unpinned num_pinned
> > pages, no point of checking till num_pages.
> >
> > This will address both.
> >
> > Signed-off-by: Souptick Joarder 
> > Reviewed-by: Dan Carpenter 
> > Cc: John Hubbard 
>
> Any suggestion, what is the right tree to take this patch forward ?

If no further comment, can we get this patch in queue for 5.9 ?

>
> > ---
> > v2:
> > Added review tag.
> >
> >  drivers/virt/fsl_hypervisor.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/virt/fsl_hypervisor.c b/drivers/virt/fsl_hypervisor.c
> > index 1b0b11b..ea344d7 100644
> > --- a/drivers/virt/fsl_hypervisor.c
> > +++ b/drivers/virt/fsl_hypervisor.c
> > @@ -157,7 +157,7 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy 
> > __user *p)
> >
> > unsigned int i;
> > long ret = 0;
> > -   int num_pinned; /* return value from get_user_pages() */
> > +   int num_pinned = 0; /* return value from get_user_pages() */
> > phys_addr_t remote_paddr; /* The next address in the remote buffer 
> > */
> > uint32_t count; /* The number of bytes left to copy */
> >
> > @@ -293,7 +293,7 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy 
> > __user *p)
> >
> >  exit:
> > if (pages) {
> > -   for (i = 0; i < num_pages; i++)
> > +   for (i = 0; i < num_pinned; i++)
> > if (pages[i])
> > put_page(pages[i]);
> > }
> > --
> > 1.9.1
> >

Re: [PATCH v2] drivers/virt/fsl_hypervisor: Correcting error handling path

[Souptick Joarder]

On Wed, Jun 17, 2020 at 8:22 AM Souptick Joarder  wrote:
>
> First, when memory allocation for sg_list_unaligned failed, there
> is no point of calling put_pages() as we haven't pinned any pages.
>
> Second, if get_user_pages_fast() failed we should unpinned num_pinned
> pages, no point of checking till num_pages.
>
> This will address both.
>
> Signed-off-by: Souptick Joarder 
> Reviewed-by: Dan Carpenter 
> Cc: John Hubbard 

Any suggestion, what is the right tree to take this patch forward ?

> ---
> v2:
> Added review tag.
>
>  drivers/virt/fsl_hypervisor.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/virt/fsl_hypervisor.c b/drivers/virt/fsl_hypervisor.c
> index 1b0b11b..ea344d7 100644
> --- a/drivers/virt/fsl_hypervisor.c
> +++ b/drivers/virt/fsl_hypervisor.c
> @@ -157,7 +157,7 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy 
> __user *p)
>
> unsigned int i;
> long ret = 0;
> -   int num_pinned; /* return value from get_user_pages() */
> +   int num_pinned = 0; /* return value from get_user_pages() */
> phys_addr_t remote_paddr; /* The next address in the remote buffer */
> uint32_t count; /* The number of bytes left to copy */
>
> @@ -293,7 +293,7 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy 
> __user *p)
>
>  exit:
> if (pages) {
> -   for (i = 0; i < num_pages; i++)
> +   for (i = 0; i < num_pinned; i++)
> if (pages[i])
> put_page(pages[i]);
> }
> --
> 1.9.1
>