| Message ID | 1450082498-27109-7-git-send-email-a.rigo@virtualopensystems.com |
|---|---|
| State | New |
| Headers | show |
On 14/12/2015 09:41, Alvise Rigo wrote: > Use the new slow path for atomic instruction translation when the > softmmu is enabled. > > Suggested-by: Jani Kokkonen <jani.kokkonen@huawei.com> > Suggested-by: Claudio Fontana <claudio.fontana@huawei.com> > Signed-off-by: Alvise Rigo <a.rigo@virtualopensystems.com> > --- > configure | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/configure b/configure > index b9552fd..cc3891a 100755 > --- a/configure > +++ b/configure > @@ -4794,6 +4794,7 @@ echo "Install blobs $blobs" > echo "KVM support $kvm" > echo "RDMA support $rdma" > echo "TCG interpreter $tcg_interpreter" > +echo "use ld/st excl $softmmu" > echo "fdt support $fdt" > echo "preadv support $preadv" > echo "fdatasync $fdatasync" > @@ -5186,6 +5187,9 @@ fi > if test "$tcg_interpreter" = "yes" ; then > echo "CONFIG_TCG_INTERPRETER=y" >> $config_host_mak > fi > +if test "$softmmu" = "yes" ; then > + echo "CONFIG_TCG_USE_LDST_EXCL=y" >> $config_host_mak > +fi Just use CONFIG_SOFTMMU in translate.c, no? A target other than ARM might need ll/sc in user-mode emulation as well. Paolo > if test "$fdatasync" = "yes" ; then > echo "CONFIG_FDATASYNC=y" >> $config_host_mak > fi >
On 14/12/2015 10:38, Paolo Bonzini wrote: >> > +if test "$softmmu" = "yes" ; then >> > + echo "CONFIG_TCG_USE_LDST_EXCL=y" >> $config_host_mak >> > +fi > Just use CONFIG_SOFTMMU in translate.c, no? > > A target other than ARM might need ll/sc in user-mode emulation as well. Sorry, that makes no sense. Couldn't hit cancel in time. :) Paolo
On 14/12/2015 09:41, Alvise Rigo wrote: > Use the new slow path for atomic instruction translation when the > softmmu is enabled. > > Suggested-by: Jani Kokkonen <jani.kokkonen@huawei.com> > Suggested-by: Claudio Fontana <claudio.fontana@huawei.com> > Signed-off-by: Alvise Rigo <a.rigo@virtualopensystems.com> > --- > configure | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/configure b/configure > index b9552fd..cc3891a 100755 > --- a/configure > +++ b/configure > @@ -4794,6 +4794,7 @@ echo "Install blobs $blobs" > echo "KVM support $kvm" > echo "RDMA support $rdma" > echo "TCG interpreter $tcg_interpreter" > +echo "use ld/st excl $softmmu" > echo "fdt support $fdt" > echo "preadv support $preadv" > echo "fdatasync $fdatasync" > @@ -5186,6 +5187,9 @@ fi > if test "$tcg_interpreter" = "yes" ; then > echo "CONFIG_TCG_INTERPRETER=y" >> $config_host_mak > fi > +if test "$softmmu" = "yes" ; then > + echo "CONFIG_TCG_USE_LDST_EXCL=y" >> $config_host_mak > +fi why is this "$softmmu" and not "$target_softmmu" ? > if test "$fdatasync" = "yes" ; then > echo "CONFIG_FDATASYNC=y" >> $config_host_mak > fi >
Hi, On Mon, Dec 14, 2015 at 11:14 AM, Laurent Vivier <lvivier@redhat.com> wrote: > > > On 14/12/2015 09:41, Alvise Rigo wrote: >> Use the new slow path for atomic instruction translation when the >> softmmu is enabled. >> >> Suggested-by: Jani Kokkonen <jani.kokkonen@huawei.com> >> Suggested-by: Claudio Fontana <claudio.fontana@huawei.com> >> Signed-off-by: Alvise Rigo <a.rigo@virtualopensystems.com> >> --- >> configure | 4 ++++ >> 1 file changed, 4 insertions(+) >> >> diff --git a/configure b/configure >> index b9552fd..cc3891a 100755 >> --- a/configure >> +++ b/configure >> @@ -4794,6 +4794,7 @@ echo "Install blobs $blobs" >> echo "KVM support $kvm" >> echo "RDMA support $rdma" >> echo "TCG interpreter $tcg_interpreter" >> +echo "use ld/st excl $softmmu" >> echo "fdt support $fdt" >> echo "preadv support $preadv" >> echo "fdatasync $fdatasync" >> @@ -5186,6 +5187,9 @@ fi >> if test "$tcg_interpreter" = "yes" ; then >> echo "CONFIG_TCG_INTERPRETER=y" >> $config_host_mak >> fi >> +if test "$softmmu" = "yes" ; then >> + echo "CONFIG_TCG_USE_LDST_EXCL=y" >> $config_host_mak >> +fi > > why is this "$softmmu" and not "$target_softmmu" ? I see now that is $target_softmmu setting CONFIG_SOFTMMU=y. So for my understanding, which are the cases where $softmmu is set while $target_softmmu is not? Thank you, alvise > >> if test "$fdatasync" = "yes" ; then >> echo "CONFIG_FDATASYNC=y" >> $config_host_mak >> fi >>
On 15/12/2015 15:23, alvise rigo wrote: > Hi, > > On Mon, Dec 14, 2015 at 11:14 AM, Laurent Vivier <lvivier@redhat.com> wrote: >> >> >> On 14/12/2015 09:41, Alvise Rigo wrote: >>> Use the new slow path for atomic instruction translation when the >>> softmmu is enabled. >>> >>> Suggested-by: Jani Kokkonen <jani.kokkonen@huawei.com> >>> Suggested-by: Claudio Fontana <claudio.fontana@huawei.com> >>> Signed-off-by: Alvise Rigo <a.rigo@virtualopensystems.com> >>> --- >>> configure | 4 ++++ >>> 1 file changed, 4 insertions(+) >>> >>> diff --git a/configure b/configure >>> index b9552fd..cc3891a 100755 >>> --- a/configure >>> +++ b/configure >>> @@ -4794,6 +4794,7 @@ echo "Install blobs $blobs" >>> echo "KVM support $kvm" >>> echo "RDMA support $rdma" >>> echo "TCG interpreter $tcg_interpreter" >>> +echo "use ld/st excl $softmmu" >>> echo "fdt support $fdt" >>> echo "preadv support $preadv" >>> echo "fdatasync $fdatasync" >>> @@ -5186,6 +5187,9 @@ fi >>> if test "$tcg_interpreter" = "yes" ; then >>> echo "CONFIG_TCG_INTERPRETER=y" >> $config_host_mak >>> fi >>> +if test "$softmmu" = "yes" ; then >>> + echo "CONFIG_TCG_USE_LDST_EXCL=y" >> $config_host_mak >>> +fi >> >> why is this "$softmmu" and not "$target_softmmu" ? > > I see now that is $target_softmmu setting CONFIG_SOFTMMU=y. > So for my understanding, which are the cases where $softmmu is set > while $target_softmmu is not? When compiling foo-linux-user. Paolo
On 15/12/2015 15:31, Paolo Bonzini wrote: > > > On 15/12/2015 15:23, alvise rigo wrote: >> Hi, >> >> On Mon, Dec 14, 2015 at 11:14 AM, Laurent Vivier <lvivier@redhat.com> wrote: >>> >>> >>> On 14/12/2015 09:41, Alvise Rigo wrote: >>>> Use the new slow path for atomic instruction translation when the >>>> softmmu is enabled. >>>> >>>> Suggested-by: Jani Kokkonen <jani.kokkonen@huawei.com> >>>> Suggested-by: Claudio Fontana <claudio.fontana@huawei.com> >>>> Signed-off-by: Alvise Rigo <a.rigo@virtualopensystems.com> >>>> --- >>>> configure | 4 ++++ >>>> 1 file changed, 4 insertions(+) >>>> >>>> diff --git a/configure b/configure >>>> index b9552fd..cc3891a 100755 >>>> --- a/configure >>>> +++ b/configure >>>> @@ -4794,6 +4794,7 @@ echo "Install blobs $blobs" >>>> echo "KVM support $kvm" >>>> echo "RDMA support $rdma" >>>> echo "TCG interpreter $tcg_interpreter" >>>> +echo "use ld/st excl $softmmu" >>>> echo "fdt support $fdt" >>>> echo "preadv support $preadv" >>>> echo "fdatasync $fdatasync" >>>> @@ -5186,6 +5187,9 @@ fi >>>> if test "$tcg_interpreter" = "yes" ; then >>>> echo "CONFIG_TCG_INTERPRETER=y" >> $config_host_mak >>>> fi >>>> +if test "$softmmu" = "yes" ; then >>>> + echo "CONFIG_TCG_USE_LDST_EXCL=y" >> $config_host_mak >>>> +fi >>> >>> why is this "$softmmu" and not "$target_softmmu" ? >> >> I see now that is $target_softmmu setting CONFIG_SOFTMMU=y. >> So for my understanding, which are the cases where $softmmu is set >> while $target_softmmu is not? > > When compiling foo-linux-user. In fact, after having asked the question, I've found that it is right to use softmmu which is a host parameter (while target_softmmu is guest parameter) and softmmu is set to true if there is at least one *-softmmu target. Laurent
diff --git a/configure b/configure index b9552fd..cc3891a 100755 --- a/configure +++ b/configure @@ -4794,6 +4794,7 @@ echo "Install blobs $blobs" echo "KVM support $kvm" echo "RDMA support $rdma" echo "TCG interpreter $tcg_interpreter" +echo "use ld/st excl $softmmu" echo "fdt support $fdt" echo "preadv support $preadv" echo "fdatasync $fdatasync" @@ -5186,6 +5187,9 @@ fi if test "$tcg_interpreter" = "yes" ; then echo "CONFIG_TCG_INTERPRETER=y" >> $config_host_mak fi +if test "$softmmu" = "yes" ; then + echo "CONFIG_TCG_USE_LDST_EXCL=y" >> $config_host_mak +fi if test "$fdatasync" = "yes" ; then echo "CONFIG_FDATASYNC=y" >> $config_host_mak fi
Use the new slow path for atomic instruction translation when the softmmu is enabled. Suggested-by: Jani Kokkonen <jani.kokkonen@huawei.com> Suggested-by: Claudio Fontana <claudio.fontana@huawei.com> Signed-off-by: Alvise Rigo <a.rigo@virtualopensystems.com> --- configure | 4 ++++ 1 file changed, 4 insertions(+)