[PATCH v2 0/2] Remove PAGE_SIZE from public headers

[PATCH v2 0/2] Remove PAGE_SIZE from public headers

[stefano]

Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
ARM64 is broken by this. PAGE_SIZE should not be used because it could
be undefined or it could be defined differently on different operating
systems.

Stefano Stabellini (2):
      usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
      vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE

 xen/include/public/io/usbif.h   | 5 +++--
 xen/include/public/io/vscsiif.h | 3 ++-
 2 files changed, 5 insertions(+), 3 deletions(-)

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

[PATCH v2 1/2] usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE

[Stefano Stabellini]

Do not reference PAGE_SIZE directly: it could be undefined, or it could
have different values in the frontend or in the backend.

Define USBIF_RING_SIZE as 4096, assuming all users of usbif.h have 4K
page granularity. Replace PAGE_SIZE with USBIF_RING_SIZE.

Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
---
 xen/include/public/io/usbif.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/xen/include/public/io/usbif.h b/xen/include/public/io/usbif.h
index 4053c24..c6a5863 100644
--- a/xen/include/public/io/usbif.h
+++ b/xen/include/public/io/usbif.h
@@ -170,6 +170,7 @@ enum usb_spec_version {
 
 #define USBIF_MAX_SEGMENTS_PER_REQUEST (16)
 #define USBIF_MAX_PORTNR	31
+#define USBIF_RING_SIZE	4096
 
 /*
  * RING for transferring urbs.
@@ -226,7 +227,7 @@ struct usbif_urb_response {
 typedef struct usbif_urb_response usbif_urb_response_t;
 
 DEFINE_RING_TYPES(usbif_urb, struct usbif_urb_request, struct usbif_urb_response);
-#define USB_URB_RING_SIZE __CONST_RING_SIZE(usbif_urb, PAGE_SIZE)
+#define USB_URB_RING_SIZE __CONST_RING_SIZE(usbif_urb, USBIF_RING_SIZE)
 
 /*
  * RING for notifying connect/disconnect events to frontend
@@ -248,6 +249,6 @@ struct usbif_conn_response {
 typedef struct usbif_conn_response usbif_conn_response_t;
 
 DEFINE_RING_TYPES(usbif_conn, struct usbif_conn_request, struct usbif_conn_response);
-#define USB_CONN_RING_SIZE __CONST_RING_SIZE(usbif_conn, PAGE_SIZE)
+#define USB_CONN_RING_SIZE __CONST_RING_SIZE(usbif_conn, USBIF_RING_SIZE)
 
 #endif /* __XEN_PUBLIC_IO_USBIF_H__ */
-- 
1.9.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

[PATCH v2 2/2] vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE

[Stefano Stabellini]

Do not reference PAGE_SIZE directly: it could be undefined, or it could
have different values in the frontend or in the backend.

Define VSCSIIF_PAGE_SIZE as 4096, assuming all users of vscsiif.h have
4K page granularity. Replace PAGE_SIZE with VSCSIIF_PAGE_SIZE.

Signed-off-by: Stefano Stabellini <sstabellini@kernel.org>
---
 xen/include/public/io/vscsiif.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/xen/include/public/io/vscsiif.h b/xen/include/public/io/vscsiif.h
index 0a4709f..d0bd3b5 100644
--- a/xen/include/public/io/vscsiif.h
+++ b/xen/include/public/io/vscsiif.h
@@ -248,6 +248,7 @@
  */
 #define VSCSIIF_MAX_COMMAND_SIZE         16
 #define VSCSIIF_SENSE_BUFFERSIZE         96
+#define VSCSIIF_PAGE_SIZE              4096
 
 struct scsiif_request_segment {
     grant_ref_t gref;
@@ -256,7 +257,7 @@ struct scsiif_request_segment {
 };
 typedef struct scsiif_request_segment vscsiif_segment_t;
 
-#define VSCSIIF_SG_PER_PAGE (PAGE_SIZE / sizeof(struct scsiif_request_segment))
+#define VSCSIIF_SG_PER_PAGE (VSCSIIF_PAGE_SIZE / sizeof(struct scsiif_request_segment))
 
 /* Size of one request is 252 bytes */
 struct vscsiif_request {
-- 
1.9.1


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Juergen Gross]

On 19/10/16 21:21, stefano@aporeto.com wrote:
> Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
> ARM64 is broken by this. PAGE_SIZE should not be used because it could
> be undefined or it could be defined differently on different operating
> systems.
> 
> Stefano Stabellini (2):
>       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
>       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
> 
>  xen/include/public/io/usbif.h   | 5 +++--
>  xen/include/public/io/vscsiif.h | 3 ++-
>  2 files changed, 5 insertions(+), 3 deletions(-)
> 

In principle I'm okay with these changes. But wouldn't it be
better to add

#define XEN_RING_PAGE_SIZE 4096

to xen/include/public/io/ring.h instead of having each interface
file its own definition?


Juergen

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Stefano Stabellini]

On Thu, 20 Oct 2016, Juergen Gross wrote:
> On 19/10/16 21:21, stefano@aporeto.com wrote:
> > Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
> > ARM64 is broken by this. PAGE_SIZE should not be used because it could
> > be undefined or it could be defined differently on different operating
> > systems.
> > 
> > Stefano Stabellini (2):
> >       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
> >       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
> > 
> >  xen/include/public/io/usbif.h   | 5 +++--
> >  xen/include/public/io/vscsiif.h | 3 ++-
> >  2 files changed, 5 insertions(+), 3 deletions(-)
> > 
> 
> In principle I'm okay with these changes. But wouldn't it be
> better to add
> 
> #define XEN_RING_PAGE_SIZE 4096
> 
> to xen/include/public/io/ring.h instead of having each interface
> file its own definition?

I would prefer to avoid having to spell out the page size for all the io
interfaces, because the same interfaces should work fine with other page
granularities, if the grant table hypercalls supported them.

However in reality frontend and backend would have to agree on the page
granularity to use beforehand. Some sort of consensus protocol is
required. When that protocol is introduced, we could get rid of
#define XEN_RING_PAGE_SIZE 4096.

What do people think about this?

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Andrew Cooper]

On 20/10/16 20:09, Stefano Stabellini wrote:
> On Thu, 20 Oct 2016, Juergen Gross wrote:
>> On 19/10/16 21:21, stefano@aporeto.com wrote:
>>> Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
>>> ARM64 is broken by this. PAGE_SIZE should not be used because it could
>>> be undefined or it could be defined differently on different operating
>>> systems.
>>>
>>> Stefano Stabellini (2):
>>>       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
>>>       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
>>>
>>>  xen/include/public/io/usbif.h   | 5 +++--
>>>  xen/include/public/io/vscsiif.h | 3 ++-
>>>  2 files changed, 5 insertions(+), 3 deletions(-)
>>>
>> In principle I'm okay with these changes. But wouldn't it be
>> better to add
>>
>> #define XEN_RING_PAGE_SIZE 4096
>>
>> to xen/include/public/io/ring.h instead of having each interface
>> file its own definition?
> I would prefer to avoid having to spell out the page size for all the io
> interfaces, because the same interfaces should work fine with other page
> granularities, if the grant table hypercalls supported them.
>
> However in reality frontend and backend would have to agree on the page
> granularity to use beforehand. Some sort of consensus protocol is
> required. When that protocol is introduced, we could get rid of
> #define XEN_RING_PAGE_SIZE 4096.
>
> What do people think about this?

Not all current rings are 4k in size, and not all future rings should be
implicitly restricted to 4k.

-1 to this introducing XEN_RING_PAGE_SIZE.  It will be more confusing
than helpful.

~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Stefano Stabellini]

On Thu, 20 Oct 2016, Andrew Cooper wrote:
> On 20/10/16 20:09, Stefano Stabellini wrote:
> > On Thu, 20 Oct 2016, Juergen Gross wrote:
> >> On 19/10/16 21:21, stefano@aporeto.com wrote:
> >>> Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
> >>> ARM64 is broken by this. PAGE_SIZE should not be used because it could
> >>> be undefined or it could be defined differently on different operating
> >>> systems.
> >>>
> >>> Stefano Stabellini (2):
> >>>       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
> >>>       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
> >>>
> >>>  xen/include/public/io/usbif.h   | 5 +++--
> >>>  xen/include/public/io/vscsiif.h | 3 ++-
> >>>  2 files changed, 5 insertions(+), 3 deletions(-)
> >>>
> >> In principle I'm okay with these changes. But wouldn't it be
> >> better to add
> >>
> >> #define XEN_RING_PAGE_SIZE 4096
> >>
> >> to xen/include/public/io/ring.h instead of having each interface
> >> file its own definition?
> > I would prefer to avoid having to spell out the page size for all the io
> > interfaces, because the same interfaces should work fine with other page
> > granularities, if the grant table hypercalls supported them.
> >
> > However in reality frontend and backend would have to agree on the page
> > granularity to use beforehand. Some sort of consensus protocol is
> > required. When that protocol is introduced, we could get rid of
> > #define XEN_RING_PAGE_SIZE 4096.
> >
> > What do people think about this?
> 
> Not all current rings are 4k in size, and not all future rings should be
> implicitly restricted to 4k.
> 
> -1 to this introducing XEN_RING_PAGE_SIZE.  It will be more confusing
> than helpful.

Indeed, you are right. Since you are at it, feel like acking the current
patches? :-)

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Andrew Cooper]

On 20/10/16 21:04, Stefano Stabellini wrote:
> On Thu, 20 Oct 2016, Andrew Cooper wrote:
>> On 20/10/16 20:09, Stefano Stabellini wrote:
>>> On Thu, 20 Oct 2016, Juergen Gross wrote:
>>>> On 19/10/16 21:21, stefano@aporeto.com wrote:
>>>>> Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
>>>>> ARM64 is broken by this. PAGE_SIZE should not be used because it could
>>>>> be undefined or it could be defined differently on different operating
>>>>> systems.
>>>>>
>>>>> Stefano Stabellini (2):
>>>>>       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
>>>>>       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
>>>>>
>>>>>  xen/include/public/io/usbif.h   | 5 +++--
>>>>>  xen/include/public/io/vscsiif.h | 3 ++-
>>>>>  2 files changed, 5 insertions(+), 3 deletions(-)
>>>>>
>>>> In principle I'm okay with these changes. But wouldn't it be
>>>> better to add
>>>>
>>>> #define XEN_RING_PAGE_SIZE 4096
>>>>
>>>> to xen/include/public/io/ring.h instead of having each interface
>>>> file its own definition?
>>> I would prefer to avoid having to spell out the page size for all the io
>>> interfaces, because the same interfaces should work fine with other page
>>> granularities, if the grant table hypercalls supported them.
>>>
>>> However in reality frontend and backend would have to agree on the page
>>> granularity to use beforehand. Some sort of consensus protocol is
>>> required. When that protocol is introduced, we could get rid of
>>> #define XEN_RING_PAGE_SIZE 4096.
>>>
>>> What do people think about this?
>> Not all current rings are 4k in size, and not all future rings should be
>> implicitly restricted to 4k.
>>
>> -1 to this introducing XEN_RING_PAGE_SIZE.  It will be more confusing
>> than helpful.
> Indeed, you are right. Since you are at it, feel like acking the current
> patches? :-)

Good point.  Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>,
although you also need a release ack at this point.

~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Stefano Stabellini]

On Thu, 20 Oct 2016, Andrew Cooper wrote:
> On 20/10/16 21:04, Stefano Stabellini wrote:
> > On Thu, 20 Oct 2016, Andrew Cooper wrote:
> >> On 20/10/16 20:09, Stefano Stabellini wrote:
> >>> On Thu, 20 Oct 2016, Juergen Gross wrote:
> >>>> On 19/10/16 21:21, stefano@aporeto.com wrote:
> >>>>> Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
> >>>>> ARM64 is broken by this. PAGE_SIZE should not be used because it could
> >>>>> be undefined or it could be defined differently on different operating
> >>>>> systems.
> >>>>>
> >>>>> Stefano Stabellini (2):
> >>>>>       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
> >>>>>       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
> >>>>>
> >>>>>  xen/include/public/io/usbif.h   | 5 +++--
> >>>>>  xen/include/public/io/vscsiif.h | 3 ++-
> >>>>>  2 files changed, 5 insertions(+), 3 deletions(-)
> >>>>>
> >>>> In principle I'm okay with these changes. But wouldn't it be
> >>>> better to add
> >>>>
> >>>> #define XEN_RING_PAGE_SIZE 4096
> >>>>
> >>>> to xen/include/public/io/ring.h instead of having each interface
> >>>> file its own definition?
> >>> I would prefer to avoid having to spell out the page size for all the io
> >>> interfaces, because the same interfaces should work fine with other page
> >>> granularities, if the grant table hypercalls supported them.
> >>>
> >>> However in reality frontend and backend would have to agree on the page
> >>> granularity to use beforehand. Some sort of consensus protocol is
> >>> required. When that protocol is introduced, we could get rid of
> >>> #define XEN_RING_PAGE_SIZE 4096.
> >>>
> >>> What do people think about this?
> >> Not all current rings are 4k in size, and not all future rings should be
> >> implicitly restricted to 4k.
> >>
> >> -1 to this introducing XEN_RING_PAGE_SIZE.  It will be more confusing
> >> than helpful.
> > Indeed, you are right. Since you are at it, feel like acking the current
> > patches? :-)
> 
> Good point.  Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>,
> although you also need a release ack at this point.

Wei, I also need a release ack for

http://marc.info/?l=xen-devel&m=147690490302553

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Wei Liu]

On Thu, Oct 20, 2016 at 01:38:35PM -0700, Stefano Stabellini wrote:
> On Thu, 20 Oct 2016, Andrew Cooper wrote:
> > On 20/10/16 21:04, Stefano Stabellini wrote:
> > > On Thu, 20 Oct 2016, Andrew Cooper wrote:
> > >> On 20/10/16 20:09, Stefano Stabellini wrote:
> > >>> On Thu, 20 Oct 2016, Juergen Gross wrote:
> > >>>> On 19/10/16 21:21, stefano@aporeto.com wrote:
> > >>>>> Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
> > >>>>> ARM64 is broken by this. PAGE_SIZE should not be used because it could
> > >>>>> be undefined or it could be defined differently on different operating
> > >>>>> systems.
> > >>>>>
> > >>>>> Stefano Stabellini (2):
> > >>>>>       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
> > >>>>>       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
> > >>>>>
> > >>>>>  xen/include/public/io/usbif.h   | 5 +++--
> > >>>>>  xen/include/public/io/vscsiif.h | 3 ++-
> > >>>>>  2 files changed, 5 insertions(+), 3 deletions(-)
> > >>>>>
> > >>>> In principle I'm okay with these changes. But wouldn't it be
> > >>>> better to add
> > >>>>
> > >>>> #define XEN_RING_PAGE_SIZE 4096
> > >>>>
> > >>>> to xen/include/public/io/ring.h instead of having each interface
> > >>>> file its own definition?
> > >>> I would prefer to avoid having to spell out the page size for all the io
> > >>> interfaces, because the same interfaces should work fine with other page
> > >>> granularities, if the grant table hypercalls supported them.
> > >>>
> > >>> However in reality frontend and backend would have to agree on the page
> > >>> granularity to use beforehand. Some sort of consensus protocol is
> > >>> required. When that protocol is introduced, we could get rid of
> > >>> #define XEN_RING_PAGE_SIZE 4096.
> > >>>
> > >>> What do people think about this?
> > >> Not all current rings are 4k in size, and not all future rings should be
> > >> implicitly restricted to 4k.
> > >>
> > >> -1 to this introducing XEN_RING_PAGE_SIZE.  It will be more confusing
> > >> than helpful.
> > > Indeed, you are right. Since you are at it, feel like acking the current
> > > patches? :-)
> > 
> > Good point.  Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>,
> > although you also need a release ack at this point.
> 
> Wei, I also need a release ack for
> 
> http://marc.info/?l=xen-devel&m=147690490302553

Presumably you want to apply that to our QEMU tree? If so:

Release-acked-by: Wei Liu <wei.liu2@citrix.com>

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Stefano Stabellini]

On Fri, 21 Oct 2016, Wei Liu wrote:
> On Thu, Oct 20, 2016 at 01:38:35PM -0700, Stefano Stabellini wrote:
> > On Thu, 20 Oct 2016, Andrew Cooper wrote:
> > > On 20/10/16 21:04, Stefano Stabellini wrote:
> > > > On Thu, 20 Oct 2016, Andrew Cooper wrote:
> > > >> On 20/10/16 20:09, Stefano Stabellini wrote:
> > > >>> On Thu, 20 Oct 2016, Juergen Gross wrote:
> > > >>>> On 19/10/16 21:21, stefano@aporeto.com wrote:
> > > >>>>> Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
> > > >>>>> ARM64 is broken by this. PAGE_SIZE should not be used because it could
> > > >>>>> be undefined or it could be defined differently on different operating
> > > >>>>> systems.
> > > >>>>>
> > > >>>>> Stefano Stabellini (2):
> > > >>>>>       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
> > > >>>>>       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
> > > >>>>>
> > > >>>>>  xen/include/public/io/usbif.h   | 5 +++--
> > > >>>>>  xen/include/public/io/vscsiif.h | 3 ++-
> > > >>>>>  2 files changed, 5 insertions(+), 3 deletions(-)
> > > >>>>>
> > > >>>> In principle I'm okay with these changes. But wouldn't it be
> > > >>>> better to add
> > > >>>>
> > > >>>> #define XEN_RING_PAGE_SIZE 4096
> > > >>>>
> > > >>>> to xen/include/public/io/ring.h instead of having each interface
> > > >>>> file its own definition?
> > > >>> I would prefer to avoid having to spell out the page size for all the io
> > > >>> interfaces, because the same interfaces should work fine with other page
> > > >>> granularities, if the grant table hypercalls supported them.
> > > >>>
> > > >>> However in reality frontend and backend would have to agree on the page
> > > >>> granularity to use beforehand. Some sort of consensus protocol is
> > > >>> required. When that protocol is introduced, we could get rid of
> > > >>> #define XEN_RING_PAGE_SIZE 4096.
> > > >>>
> > > >>> What do people think about this?
> > > >> Not all current rings are 4k in size, and not all future rings should be
> > > >> implicitly restricted to 4k.
> > > >>
> > > >> -1 to this introducing XEN_RING_PAGE_SIZE.  It will be more confusing
> > > >> than helpful.
> > > > Indeed, you are right. Since you are at it, feel like acking the current
> > > > patches? :-)
> > > 
> > > Good point.  Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>,
> > > although you also need a release ack at this point.
> > 
> > Wei, I also need a release ack for
> > 
> > http://marc.info/?l=xen-devel&m=147690490302553
> 
> Presumably you want to apply that to our QEMU tree? If so:
> 
> Release-acked-by: Wei Liu <wei.liu2@citrix.com>

I sent a pull request to qemu with the patch and applied it to staging.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Wei Liu]

On Thu, Oct 20, 2016 at 09:05:26PM +0100, Andrew Cooper wrote:
> On 20/10/16 21:04, Stefano Stabellini wrote:
> > On Thu, 20 Oct 2016, Andrew Cooper wrote:
> >> On 20/10/16 20:09, Stefano Stabellini wrote:
> >>> On Thu, 20 Oct 2016, Juergen Gross wrote:
> >>>> On 19/10/16 21:21, stefano@aporeto.com wrote:
> >>>>> Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
> >>>>> ARM64 is broken by this. PAGE_SIZE should not be used because it could
> >>>>> be undefined or it could be defined differently on different operating
> >>>>> systems.
> >>>>>
> >>>>> Stefano Stabellini (2):
> >>>>>       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
> >>>>>       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
> >>>>>
> >>>>>  xen/include/public/io/usbif.h   | 5 +++--
> >>>>>  xen/include/public/io/vscsiif.h | 3 ++-
> >>>>>  2 files changed, 5 insertions(+), 3 deletions(-)
> >>>>>
> >>>> In principle I'm okay with these changes. But wouldn't it be
> >>>> better to add
> >>>>
> >>>> #define XEN_RING_PAGE_SIZE 4096
> >>>>
> >>>> to xen/include/public/io/ring.h instead of having each interface
> >>>> file its own definition?
> >>> I would prefer to avoid having to spell out the page size for all the io
> >>> interfaces, because the same interfaces should work fine with other page
> >>> granularities, if the grant table hypercalls supported them.
> >>>
> >>> However in reality frontend and backend would have to agree on the page
> >>> granularity to use beforehand. Some sort of consensus protocol is
> >>> required. When that protocol is introduced, we could get rid of
> >>> #define XEN_RING_PAGE_SIZE 4096.
> >>>
> >>> What do people think about this?
> >> Not all current rings are 4k in size, and not all future rings should be
> >> implicitly restricted to 4k.
> >>
> >> -1 to this introducing XEN_RING_PAGE_SIZE.  It will be more confusing
> >> than helpful.
> > Indeed, you are right. Since you are at it, feel like acking the current
> > patches? :-)
> 
> Good point.  Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>,
> although you also need a release ack at this point.
> 

Acked-by: Wei Liu <wei.liu2@citrix.com>
Release-acked-by: Wei Liu <wei.liu2@citrix.com>

And applied.


> ~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH v2 0/2] Remove PAGE_SIZE from public headers

[Stefano Stabellini]

On Fri, 21 Oct 2016, Wei Liu wrote:
> On Thu, Oct 20, 2016 at 09:05:26PM +0100, Andrew Cooper wrote:
> > On 20/10/16 21:04, Stefano Stabellini wrote:
> > > On Thu, 20 Oct 2016, Andrew Cooper wrote:
> > >> On 20/10/16 20:09, Stefano Stabellini wrote:
> > >>> On Thu, 20 Oct 2016, Juergen Gross wrote:
> > >>>> On 19/10/16 21:21, stefano@aporeto.com wrote:
> > >>>>> Reference to PAGE_SIZE slipped in to two public header files. QEMU build on
> > >>>>> ARM64 is broken by this. PAGE_SIZE should not be used because it could
> > >>>>> be undefined or it could be defined differently on different operating
> > >>>>> systems.
> > >>>>>
> > >>>>> Stefano Stabellini (2):
> > >>>>>       usbif.h: replace PAGE_SIZE with USBIF_RING_SIZE
> > >>>>>       vscsiif.h: replace PAGE_SIZE with VSCSIIF_PAGE_SIZE
> > >>>>>
> > >>>>>  xen/include/public/io/usbif.h   | 5 +++--
> > >>>>>  xen/include/public/io/vscsiif.h | 3 ++-
> > >>>>>  2 files changed, 5 insertions(+), 3 deletions(-)
> > >>>>>
> > >>>> In principle I'm okay with these changes. But wouldn't it be
> > >>>> better to add
> > >>>>
> > >>>> #define XEN_RING_PAGE_SIZE 4096
> > >>>>
> > >>>> to xen/include/public/io/ring.h instead of having each interface
> > >>>> file its own definition?
> > >>> I would prefer to avoid having to spell out the page size for all the io
> > >>> interfaces, because the same interfaces should work fine with other page
> > >>> granularities, if the grant table hypercalls supported them.
> > >>>
> > >>> However in reality frontend and backend would have to agree on the page
> > >>> granularity to use beforehand. Some sort of consensus protocol is
> > >>> required. When that protocol is introduced, we could get rid of
> > >>> #define XEN_RING_PAGE_SIZE 4096.
> > >>>
> > >>> What do people think about this?
> > >> Not all current rings are 4k in size, and not all future rings should be
> > >> implicitly restricted to 4k.
> > >>
> > >> -1 to this introducing XEN_RING_PAGE_SIZE.  It will be more confusing
> > >> than helpful.
> > > Indeed, you are right. Since you are at it, feel like acking the current
> > > patches? :-)
> > 
> > Good point.  Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>,
> > although you also need a release ack at this point.
> > 
> 
> Acked-by: Wei Liu <wei.liu2@citrix.com>
> Release-acked-by: Wei Liu <wei.liu2@citrix.com>
> 
> And applied.

Thanks. Please backport to stable trees.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel