Re: [PATCH] drm: vmwgfx: constify vmw_fence_ops

[Sinclair Yeh]

On Wed, Aug 30, 2017 at 10:30:24AM +0200, Daniel Vetter wrote:
> On Wed, Aug 30, 2017 at 08:21:46AM +0200, Thomas Hellstrom wrote:
> > On 08/30/2017 07:47 AM, Arvind Yadav wrote:
> > > vmw_fence_ops are not supposed to change at runtime. Functions
> > > "dma_fence_init" working with const vmw_fence_ops provided
> > > by . So mark the non-const structs as const.
> > > 
> > > Signed-off-by: Arvind Yadav 
> > > ---
> > >   drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
> > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
> > > b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > > index b8bc5bc..abc5f03 100644
> > > --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > > +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > > @@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool 
> > > intr, signed long timeout)
> > >   return ret;
> > >   }
> > > -static struct dma_fence_ops vmw_fence_ops = {
> > > +static const struct dma_fence_ops vmw_fence_ops = {
> > >   .get_driver_name = vmw_fence_get_driver_name,
> > >   .get_timeline_name = vmw_fence_get_timeline_name,
> > >   .enable_signaling = vmw_fence_enable_signaling,
> > 
> > Reviewed-by: Thomas Hellstrom 
> 
> Does this mean you'll merge it, or does this mean you'll expect someone
> else to merge this?

Yes, we have this queued and will go out in the next pull request.

Re: [PATCH] drm: vmwgfx: constify vmw_fence_ops

[Sinclair Yeh]

On Wed, Aug 30, 2017 at 10:30:24AM +0200, Daniel Vetter wrote:
> On Wed, Aug 30, 2017 at 08:21:46AM +0200, Thomas Hellstrom wrote:
> > On 08/30/2017 07:47 AM, Arvind Yadav wrote:
> > > vmw_fence_ops are not supposed to change at runtime. Functions
> > > "dma_fence_init" working with const vmw_fence_ops provided
> > > by . So mark the non-const structs as const.
> > > 
> > > Signed-off-by: Arvind Yadav 
> > > ---
> > >   drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
> > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
> > > b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > > index b8bc5bc..abc5f03 100644
> > > --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > > +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > > @@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool 
> > > intr, signed long timeout)
> > >   return ret;
> > >   }
> > > -static struct dma_fence_ops vmw_fence_ops = {
> > > +static const struct dma_fence_ops vmw_fence_ops = {
> > >   .get_driver_name = vmw_fence_get_driver_name,
> > >   .get_timeline_name = vmw_fence_get_timeline_name,
> > >   .enable_signaling = vmw_fence_enable_signaling,
> > 
> > Reviewed-by: Thomas Hellstrom 
> 
> Does this mean you'll merge it, or does this mean you'll expect someone
> else to merge this?

Yes, we have this queued and will go out in the next pull request.

Re: [PATCH] drm: vmwgfx: constify vmw_fence_ops

[Thomas Hellstrom]

On 08/30/2017 10:30 AM, Daniel Vetter wrote:

On Wed, Aug 30, 2017 at 08:21:46AM +0200, Thomas Hellstrom wrote:

On 08/30/2017 07:47 AM, Arvind Yadav wrote:

vmw_fence_ops are not supposed to change at runtime. Functions
"dma_fence_init" working with const vmw_fence_ops provided
by . So mark the non-const structs as const.

Signed-off-by: Arvind Yadav 
---
   drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
   1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
index b8bc5bc..abc5f03 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
@@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool intr, 
signed long timeout)
return ret;
   }
-static struct dma_fence_ops vmw_fence_ops = {
+static const struct dma_fence_ops vmw_fence_ops = {
.get_driver_name = vmw_fence_get_driver_name,
.get_timeline_name = vmw_fence_get_timeline_name,
.enable_signaling = vmw_fence_enable_signaling,

Reviewed-by: Thomas Hellstrom 

Does this mean you'll merge it, or does this mean you'll expect someone
else to merge this?

I'm always confused when maintainers reply with an r-b/ack for a patch
only touching their driver, and no further information at all.
-Daniel


For patches only touching our driver, I'd say we're always responsible 
for sorting out how it's going to be merged.


Since Sinclair is maintaining the vmwgfx trees I thought I'd give him a 
chance to comment on how he wanted it merged.


/Thomas

Re: [PATCH] drm: vmwgfx: constify vmw_fence_ops

[Thomas Hellstrom]

On 08/30/2017 10:30 AM, Daniel Vetter wrote:

On Wed, Aug 30, 2017 at 08:21:46AM +0200, Thomas Hellstrom wrote:

On 08/30/2017 07:47 AM, Arvind Yadav wrote:

vmw_fence_ops are not supposed to change at runtime. Functions
"dma_fence_init" working with const vmw_fence_ops provided
by . So mark the non-const structs as const.

Signed-off-by: Arvind Yadav 
---
   drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
   1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
index b8bc5bc..abc5f03 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
@@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool intr, 
signed long timeout)
return ret;
   }
-static struct dma_fence_ops vmw_fence_ops = {
+static const struct dma_fence_ops vmw_fence_ops = {
.get_driver_name = vmw_fence_get_driver_name,
.get_timeline_name = vmw_fence_get_timeline_name,
.enable_signaling = vmw_fence_enable_signaling,

Reviewed-by: Thomas Hellstrom 

Does this mean you'll merge it, or does this mean you'll expect someone
else to merge this?

I'm always confused when maintainers reply with an r-b/ack for a patch
only touching their driver, and no further information at all.
-Daniel


For patches only touching our driver, I'd say we're always responsible 
for sorting out how it's going to be merged.


Since Sinclair is maintaining the vmwgfx trees I thought I'd give him a 
chance to comment on how he wanted it merged.


/Thomas

Re: [PATCH] drm: vmwgfx: constify vmw_fence_ops

[Daniel Vetter]

On Wed, Aug 30, 2017 at 08:21:46AM +0200, Thomas Hellstrom wrote:
> On 08/30/2017 07:47 AM, Arvind Yadav wrote:
> > vmw_fence_ops are not supposed to change at runtime. Functions
> > "dma_fence_init" working with const vmw_fence_ops provided
> > by . So mark the non-const structs as const.
> > 
> > Signed-off-by: Arvind Yadav 
> > ---
> >   drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
> >   1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
> > b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > index b8bc5bc..abc5f03 100644
> > --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > @@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool 
> > intr, signed long timeout)
> > return ret;
> >   }
> > -static struct dma_fence_ops vmw_fence_ops = {
> > +static const struct dma_fence_ops vmw_fence_ops = {
> > .get_driver_name = vmw_fence_get_driver_name,
> > .get_timeline_name = vmw_fence_get_timeline_name,
> > .enable_signaling = vmw_fence_enable_signaling,
> 
> Reviewed-by: Thomas Hellstrom 

Does this mean you'll merge it, or does this mean you'll expect someone
else to merge this?

I'm always confused when maintainers reply with an r-b/ack for a patch
only touching their driver, and no further information at all.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch

Re: [PATCH] drm: vmwgfx: constify vmw_fence_ops

[Daniel Vetter]

On Wed, Aug 30, 2017 at 08:21:46AM +0200, Thomas Hellstrom wrote:
> On 08/30/2017 07:47 AM, Arvind Yadav wrote:
> > vmw_fence_ops are not supposed to change at runtime. Functions
> > "dma_fence_init" working with const vmw_fence_ops provided
> > by . So mark the non-const structs as const.
> > 
> > Signed-off-by: Arvind Yadav 
> > ---
> >   drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
> >   1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
> > b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > index b8bc5bc..abc5f03 100644
> > --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
> > @@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool 
> > intr, signed long timeout)
> > return ret;
> >   }
> > -static struct dma_fence_ops vmw_fence_ops = {
> > +static const struct dma_fence_ops vmw_fence_ops = {
> > .get_driver_name = vmw_fence_get_driver_name,
> > .get_timeline_name = vmw_fence_get_timeline_name,
> > .enable_signaling = vmw_fence_enable_signaling,
> 
> Reviewed-by: Thomas Hellstrom 

Does this mean you'll merge it, or does this mean you'll expect someone
else to merge this?

I'm always confused when maintainers reply with an r-b/ack for a patch
only touching their driver, and no further information at all.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch

Re: [PATCH] drm: vmwgfx: constify vmw_fence_ops

[Thomas Hellstrom]

On 08/30/2017 07:47 AM, Arvind Yadav wrote:

vmw_fence_ops are not supposed to change at runtime. Functions
"dma_fence_init" working with const vmw_fence_ops provided
by . So mark the non-const structs as const.

Signed-off-by: Arvind Yadav 
---
  drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
index b8bc5bc..abc5f03 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
@@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool intr, 
signed long timeout)
return ret;
  }
  
-static struct dma_fence_ops vmw_fence_ops = {

+static const struct dma_fence_ops vmw_fence_ops = {
.get_driver_name = vmw_fence_get_driver_name,
.get_timeline_name = vmw_fence_get_timeline_name,
.enable_signaling = vmw_fence_enable_signaling,


Reviewed-by: Thomas Hellstrom 

Re: [PATCH] drm: vmwgfx: constify vmw_fence_ops

[Thomas Hellstrom]

On 08/30/2017 07:47 AM, Arvind Yadav wrote:

vmw_fence_ops are not supposed to change at runtime. Functions
"dma_fence_init" working with const vmw_fence_ops provided
by . So mark the non-const structs as const.

Signed-off-by: Arvind Yadav 
---
  drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
index b8bc5bc..abc5f03 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
@@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool intr, 
signed long timeout)
return ret;
  }
  
-static struct dma_fence_ops vmw_fence_ops = {

+static const struct dma_fence_ops vmw_fence_ops = {
.get_driver_name = vmw_fence_get_driver_name,
.get_timeline_name = vmw_fence_get_timeline_name,
.enable_signaling = vmw_fence_enable_signaling,


Reviewed-by: Thomas Hellstrom 

[PATCH] drm: vmwgfx: constify vmw_fence_ops

[Arvind Yadav]

vmw_fence_ops are not supposed to change at runtime. Functions
"dma_fence_init" working with const vmw_fence_ops provided
by . So mark the non-const structs as const.

Signed-off-by: Arvind Yadav 
---
 drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
index b8bc5bc..abc5f03 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
@@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool intr, 
signed long timeout)
return ret;
 }
 
-static struct dma_fence_ops vmw_fence_ops = {
+static const struct dma_fence_ops vmw_fence_ops = {
.get_driver_name = vmw_fence_get_driver_name,
.get_timeline_name = vmw_fence_get_timeline_name,
.enable_signaling = vmw_fence_enable_signaling,
-- 
1.9.1

[PATCH] drm: vmwgfx: constify vmw_fence_ops

[Arvind Yadav]

vmw_fence_ops are not supposed to change at runtime. Functions
"dma_fence_init" working with const vmw_fence_ops provided
by . So mark the non-const structs as const.

Signed-off-by: Arvind Yadav 
---
 drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c 
b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
index b8bc5bc..abc5f03 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
@@ -225,7 +225,7 @@ static long vmw_fence_wait(struct dma_fence *f, bool intr, 
signed long timeout)
return ret;
 }
 
-static struct dma_fence_ops vmw_fence_ops = {
+static const struct dma_fence_ops vmw_fence_ops = {
.get_driver_name = vmw_fence_get_driver_name,
.get_timeline_name = vmw_fence_get_timeline_name,
.enable_signaling = vmw_fence_enable_signaling,
-- 
1.9.1