All of lore.kernel.org
 help / color / mirror / Atom feed
* [Qemu-devel] [For 2.5?? PATCH 1/1] qemu_{real_}host_page_[size|mask] change types to ram_addr_t
@ 2015-12-02 10:43 Dr. David Alan Gilbert (git)
  2015-12-02 10:47 ` Juan Quintela
  0 siblings, 1 reply; 4+ messages in thread
From: Dr. David Alan Gilbert (git) @ 2015-12-02 10:43 UTC (permalink / raw)
  To: qemu-devel, pbonzini, anthony.perard; +Cc: quintela, stefano.stabellini

From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>

Anthony reported that >4GB guests on Xen with 32bit QEMU
broke after my 4ed023c (Round up RAMBlock sizes).

In that patch I mask sizes against qemu_host_page_size/mask
which are uintptr_t, and thus 32bit on a 32bit QEMU, even
though the ram space might be bigger than 4GB on Xen.

ram_addr_t is uintptr_t on KVM and uint64_t on Xen; so
there should be no change on KVM, and on Xen we always get
the larger size.

Remove the ~10 year old scary comment that the type of these
variables is probably wrong.

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reported-by: Anthony PERARD <anthony.perard@citrix.com>
Fixes: 4ed023ce2a39ab5812d33cf4d819def168965a7f
---
 include/exec/cpu-all.h | 9 ++++-----
 translate-all.c        | 4 ++--
 translate-common.c     | 4 ++--
 3 files changed, 8 insertions(+), 9 deletions(-)

diff --git a/include/exec/cpu-all.h b/include/exec/cpu-all.h
index f9998b9..e23b852 100644
--- a/include/exec/cpu-all.h
+++ b/include/exec/cpu-all.h
@@ -174,11 +174,10 @@ extern unsigned long reserved_va;
 #define TARGET_PAGE_MASK ~(TARGET_PAGE_SIZE - 1)
 #define TARGET_PAGE_ALIGN(addr) (((addr) + TARGET_PAGE_SIZE - 1) & TARGET_PAGE_MASK)
 
-/* ??? These should be the larger of uintptr_t and target_ulong.  */
-extern uintptr_t qemu_real_host_page_size;
-extern uintptr_t qemu_real_host_page_mask;
-extern uintptr_t qemu_host_page_size;
-extern uintptr_t qemu_host_page_mask;
+extern ram_addr_t qemu_real_host_page_size;
+extern ram_addr_t qemu_real_host_page_mask;
+extern ram_addr_t qemu_host_page_size;
+extern ram_addr_t qemu_host_page_mask;
 
 #define HOST_PAGE_ALIGN(addr) (((addr) + qemu_host_page_size - 1) & qemu_host_page_mask)
 #define REAL_HOST_PAGE_ALIGN(addr) (((addr) + qemu_real_host_page_size - 1) & \
diff --git a/translate-all.c b/translate-all.c
index a940bd2..317f251 100644
--- a/translate-all.c
+++ b/translate-all.c
@@ -117,8 +117,8 @@ typedef struct PageDesc {
 
 #define V_L1_SHIFT (L1_MAP_ADDR_SPACE_BITS - TARGET_PAGE_BITS - V_L1_BITS)
 
-uintptr_t qemu_host_page_size;
-uintptr_t qemu_host_page_mask;
+ram_addr_t qemu_host_page_size;
+ram_addr_t qemu_host_page_mask;
 
 /* The bottom level has pointers to PageDesc */
 static void *l1_map[V_L1_SIZE];
diff --git a/translate-common.c b/translate-common.c
index 619feb4..2fe2d1e 100644
--- a/translate-common.c
+++ b/translate-common.c
@@ -20,8 +20,8 @@
 #include "qemu-common.h"
 #include "qom/cpu.h"
 
-uintptr_t qemu_real_host_page_size;
-uintptr_t qemu_real_host_page_mask;
+ram_addr_t qemu_real_host_page_size;
+ram_addr_t qemu_real_host_page_mask;
 
 #ifndef CONFIG_USER_ONLY
 /* mask must never be zero, except for A20 change call */
-- 
2.5.0

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [Qemu-devel] [For 2.5?? PATCH 1/1] qemu_{real_}host_page_[size|mask] change types to ram_addr_t
  2015-12-02 10:43 [Qemu-devel] [For 2.5?? PATCH 1/1] qemu_{real_}host_page_[size|mask] change types to ram_addr_t Dr. David Alan Gilbert (git)
@ 2015-12-02 10:47 ` Juan Quintela
  2015-12-02 10:54   ` Paolo Bonzini
  2015-12-02 11:27   ` Dr. David Alan Gilbert
  0 siblings, 2 replies; 4+ messages in thread
From: Juan Quintela @ 2015-12-02 10:47 UTC (permalink / raw)
  To: Dr. David Alan Gilbert (git)
  Cc: anthony.perard, pbonzini, qemu-devel, stefano.stabellini

"Dr. David Alan Gilbert (git)" <dgilbert@redhat.com> wrote:
> From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
>
> Anthony reported that >4GB guests on Xen with 32bit QEMU
> broke after my 4ed023c (Round up RAMBlock sizes).
>
> In that patch I mask sizes against qemu_host_page_size/mask
> which are uintptr_t, and thus 32bit on a 32bit QEMU, even
> though the ram space might be bigger than 4GB on Xen.
>
> ram_addr_t is uintptr_t on KVM and uint64_t on Xen; so
> there should be no change on KVM, and on Xen we always get
> the larger size.
>
> Remove the ~10 year old scary comment that the type of these
> variables is probably wrong.
>
> Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> Reported-by: Anthony PERARD <anthony.perard@citrix.com>
> Fixes: 4ed023ce2a39ab5812d33cf4d819def168965a7f

Reviewed-by: Juan Quintela <quintela@redhat.com>

If nobody complains, I will sent this on next migration pull requset.

Paolo, I think that I preffer this that trusting the intptr_t sign
extension, but I can be convinced either way.  What do you think?

> ---
>  include/exec/cpu-all.h | 9 ++++-----
>  translate-all.c        | 4 ++--
>  translate-common.c     | 4 ++--
>  3 files changed, 8 insertions(+), 9 deletions(-)
>
> diff --git a/include/exec/cpu-all.h b/include/exec/cpu-all.h
> index f9998b9..e23b852 100644
> --- a/include/exec/cpu-all.h
> +++ b/include/exec/cpu-all.h
> @@ -174,11 +174,10 @@ extern unsigned long reserved_va;
>  #define TARGET_PAGE_MASK ~(TARGET_PAGE_SIZE - 1)
>  #define TARGET_PAGE_ALIGN(addr) (((addr) + TARGET_PAGE_SIZE - 1) &
> TARGET_PAGE_MASK)
>  
> -/* ??? These should be the larger of uintptr_t and target_ulong.  */
> -extern uintptr_t qemu_real_host_page_size;
> -extern uintptr_t qemu_real_host_page_mask;
> -extern uintptr_t qemu_host_page_size;
> -extern uintptr_t qemu_host_page_mask;
> +extern ram_addr_t qemu_real_host_page_size;
> +extern ram_addr_t qemu_real_host_page_mask;
> +extern ram_addr_t qemu_host_page_size;
> +extern ram_addr_t qemu_host_page_mask;
>  
>  #define HOST_PAGE_ALIGN(addr) (((addr) + qemu_host_page_size - 1) &
> qemu_host_page_mask)
>  #define REAL_HOST_PAGE_ALIGN(addr) (((addr) + qemu_real_host_page_size - 1) & \
> diff --git a/translate-all.c b/translate-all.c
> index a940bd2..317f251 100644
> --- a/translate-all.c
> +++ b/translate-all.c
> @@ -117,8 +117,8 @@ typedef struct PageDesc {
>  
>  #define V_L1_SHIFT (L1_MAP_ADDR_SPACE_BITS - TARGET_PAGE_BITS - V_L1_BITS)
>  
> -uintptr_t qemu_host_page_size;
> -uintptr_t qemu_host_page_mask;
> +ram_addr_t qemu_host_page_size;
> +ram_addr_t qemu_host_page_mask;
>  
>  /* The bottom level has pointers to PageDesc */
>  static void *l1_map[V_L1_SIZE];
> diff --git a/translate-common.c b/translate-common.c
> index 619feb4..2fe2d1e 100644
> --- a/translate-common.c
> +++ b/translate-common.c
> @@ -20,8 +20,8 @@
>  #include "qemu-common.h"
>  #include "qom/cpu.h"
>  
> -uintptr_t qemu_real_host_page_size;
> -uintptr_t qemu_real_host_page_mask;
> +ram_addr_t qemu_real_host_page_size;
> +ram_addr_t qemu_real_host_page_mask;
>  
>  #ifndef CONFIG_USER_ONLY
>  /* mask must never be zero, except for A20 change call */

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [Qemu-devel] [For 2.5?? PATCH 1/1] qemu_{real_}host_page_[size|mask] change types to ram_addr_t
  2015-12-02 10:47 ` Juan Quintela
@ 2015-12-02 10:54   ` Paolo Bonzini
  2015-12-02 11:27   ` Dr. David Alan Gilbert
  1 sibling, 0 replies; 4+ messages in thread
From: Paolo Bonzini @ 2015-12-02 10:54 UTC (permalink / raw)
  To: quintela, Dr. David Alan Gilbert (git)
  Cc: anthony.perard, qemu-devel, stefano.stabellini



On 02/12/2015 11:47, Juan Quintela wrote:
> Reviewed-by: Juan Quintela <quintela@redhat.com>
> 
> If nobody complains, I will sent this on next migration pull requset.
> 
> Paolo, I think that I preffer this that trusting the intptr_t sign
> extension, but I can be convinced either way.  What do you think?

It's not trusting, it's the way C works, but I'm okay either way.

Paolo

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [Qemu-devel] [For 2.5?? PATCH 1/1] qemu_{real_}host_page_[size|mask] change types to ram_addr_t
  2015-12-02 10:47 ` Juan Quintela
  2015-12-02 10:54   ` Paolo Bonzini
@ 2015-12-02 11:27   ` Dr. David Alan Gilbert
  1 sibling, 0 replies; 4+ messages in thread
From: Dr. David Alan Gilbert @ 2015-12-02 11:27 UTC (permalink / raw)
  To: Juan Quintela; +Cc: anthony.perard, pbonzini, qemu-devel, stefano.stabellini

* Juan Quintela (quintela@redhat.com) wrote:
> "Dr. David Alan Gilbert (git)" <dgilbert@redhat.com> wrote:
> > From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
> >
> > Anthony reported that >4GB guests on Xen with 32bit QEMU
> > broke after my 4ed023c (Round up RAMBlock sizes).
> >
> > In that patch I mask sizes against qemu_host_page_size/mask
> > which are uintptr_t, and thus 32bit on a 32bit QEMU, even
> > though the ram space might be bigger than 4GB on Xen.
> >
> > ram_addr_t is uintptr_t on KVM and uint64_t on Xen; so
> > there should be no change on KVM, and on Xen we always get
> > the larger size.
> >
> > Remove the ~10 year old scary comment that the type of these
> > variables is probably wrong.
> >
> > Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> > Reported-by: Anthony PERARD <anthony.perard@citrix.com>
> > Fixes: 4ed023ce2a39ab5812d33cf4d819def168965a7f
> 
> Reviewed-by: Juan Quintela <quintela@redhat.com>
> 
> If nobody complains, I will sent this on next migration pull requset.
> 
> Paolo, I think that I preffer this that trusting the intptr_t sign
> extension, but I can be convinced either way.  What do you think?

Hang on, Juan just pointed out this breaks builds of a bunch of stuff; checking....

Dave

> 
> > ---
> >  include/exec/cpu-all.h | 9 ++++-----
> >  translate-all.c        | 4 ++--
> >  translate-common.c     | 4 ++--
> >  3 files changed, 8 insertions(+), 9 deletions(-)
> >
> > diff --git a/include/exec/cpu-all.h b/include/exec/cpu-all.h
> > index f9998b9..e23b852 100644
> > --- a/include/exec/cpu-all.h
> > +++ b/include/exec/cpu-all.h
> > @@ -174,11 +174,10 @@ extern unsigned long reserved_va;
> >  #define TARGET_PAGE_MASK ~(TARGET_PAGE_SIZE - 1)
> >  #define TARGET_PAGE_ALIGN(addr) (((addr) + TARGET_PAGE_SIZE - 1) &
> > TARGET_PAGE_MASK)
> >  
> > -/* ??? These should be the larger of uintptr_t and target_ulong.  */
> > -extern uintptr_t qemu_real_host_page_size;
> > -extern uintptr_t qemu_real_host_page_mask;
> > -extern uintptr_t qemu_host_page_size;
> > -extern uintptr_t qemu_host_page_mask;
> > +extern ram_addr_t qemu_real_host_page_size;
> > +extern ram_addr_t qemu_real_host_page_mask;
> > +extern ram_addr_t qemu_host_page_size;
> > +extern ram_addr_t qemu_host_page_mask;
> >  
> >  #define HOST_PAGE_ALIGN(addr) (((addr) + qemu_host_page_size - 1) &
> > qemu_host_page_mask)
> >  #define REAL_HOST_PAGE_ALIGN(addr) (((addr) + qemu_real_host_page_size - 1) & \
> > diff --git a/translate-all.c b/translate-all.c
> > index a940bd2..317f251 100644
> > --- a/translate-all.c
> > +++ b/translate-all.c
> > @@ -117,8 +117,8 @@ typedef struct PageDesc {
> >  
> >  #define V_L1_SHIFT (L1_MAP_ADDR_SPACE_BITS - TARGET_PAGE_BITS - V_L1_BITS)
> >  
> > -uintptr_t qemu_host_page_size;
> > -uintptr_t qemu_host_page_mask;
> > +ram_addr_t qemu_host_page_size;
> > +ram_addr_t qemu_host_page_mask;
> >  
> >  /* The bottom level has pointers to PageDesc */
> >  static void *l1_map[V_L1_SIZE];
> > diff --git a/translate-common.c b/translate-common.c
> > index 619feb4..2fe2d1e 100644
> > --- a/translate-common.c
> > +++ b/translate-common.c
> > @@ -20,8 +20,8 @@
> >  #include "qemu-common.h"
> >  #include "qom/cpu.h"
> >  
> > -uintptr_t qemu_real_host_page_size;
> > -uintptr_t qemu_real_host_page_mask;
> > +ram_addr_t qemu_real_host_page_size;
> > +ram_addr_t qemu_real_host_page_mask;
> >  
> >  #ifndef CONFIG_USER_ONLY
> >  /* mask must never be zero, except for A20 change call */
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2015-12-02 11:27 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-12-02 10:43 [Qemu-devel] [For 2.5?? PATCH 1/1] qemu_{real_}host_page_[size|mask] change types to ram_addr_t Dr. David Alan Gilbert (git)
2015-12-02 10:47 ` Juan Quintela
2015-12-02 10:54   ` Paolo Bonzini
2015-12-02 11:27   ` Dr. David Alan Gilbert

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.