| Message ID | 1485868446-10587-4-git-send-email-mst@redhat.com |
|---|---|
| State | New |
| Headers | show |
On Tue, Jan 31, 2017 at 03:14:29PM +0200, Michael S. Tsirkin wrote: > We are switching BUILD_BUG_ON to verify that it's parameter is a > compile-time constant, and it turns out that some gcc versions > (specifically gcc (Ubuntu 5.4.0-6ubuntu1~16.04.4) 5.4.0 20160609) are > not smart enough to figure it out for expressions involving local > variables. This is harmless but means that the check is ineffective for > these platforms. To fix, replace the variable with macros. > > Reported-by: Peter Maydell <peter.maydell@linaro.org> > Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Ugh.. I pulled this into my tree already (I've been having trouble with my test setup, which is why I haven't sent my own pullreq yet). However, it needs a s/%u/%llu/ to compile correctly on some platforms. > --- > hw/ppc/spapr.c | 14 ++++++++------ > 1 file changed, 8 insertions(+), 6 deletions(-) > > diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c > index a642e66..b81f2ac 100644 > --- a/hw/ppc/spapr.c > +++ b/hw/ppc/spapr.c > @@ -2630,8 +2630,8 @@ static void spapr_phb_placement(sPAPRMachineState *spapr, uint32_t index, > * 1TiB 64-bit MMIO windows for each PHB. > */ > const uint64_t base_buid = 0x800000020000000ULL; > - const int max_phbs = > - (SPAPR_PCI_LIMIT - SPAPR_PCI_BASE) / SPAPR_PCI_MEM64_WIN_SIZE - 1; > +#define SPAPR_MAX_PHBS ((SPAPR_PCI_LIMIT - SPAPR_PCI_BASE) / \ > + SPAPR_PCI_MEM64_WIN_SIZE - 1) > int i; > > /* Sanity check natural alignments */ > @@ -2640,12 +2640,14 @@ static void spapr_phb_placement(sPAPRMachineState *spapr, uint32_t index, > QEMU_BUILD_BUG_ON((SPAPR_PCI_MEM64_WIN_SIZE % SPAPR_PCI_MEM32_WIN_SIZE) != 0); > QEMU_BUILD_BUG_ON((SPAPR_PCI_MEM32_WIN_SIZE % SPAPR_PCI_IO_WIN_SIZE) != 0); > /* Sanity check bounds */ > - QEMU_BUILD_BUG_ON((max_phbs * SPAPR_PCI_IO_WIN_SIZE) > SPAPR_PCI_MEM32_WIN_SIZE); > - QEMU_BUILD_BUG_ON((max_phbs * SPAPR_PCI_MEM32_WIN_SIZE) > SPAPR_PCI_MEM64_WIN_SIZE); > + QEMU_BUILD_BUG_ON((SPAPR_MAX_PHBS * SPAPR_PCI_IO_WIN_SIZE) > > + SPAPR_PCI_MEM32_WIN_SIZE); > + QEMU_BUILD_BUG_ON((SPAPR_MAX_PHBS * SPAPR_PCI_MEM32_WIN_SIZE) > > + SPAPR_PCI_MEM64_WIN_SIZE); > > - if (index >= max_phbs) { > + if (index >= SPAPR_MAX_PHBS) { > error_setg(errp, "\"index\" for PAPR PHB is too large (max %u)", > - max_phbs - 1); > + SPAPR_MAX_PHBS - 1); > return; > } >
On Wed, Feb 01, 2017 at 09:16:28AM +1100, David Gibson wrote: > On Tue, Jan 31, 2017 at 03:14:29PM +0200, Michael S. Tsirkin wrote: > > We are switching BUILD_BUG_ON to verify that it's parameter is a > > compile-time constant, and it turns out that some gcc versions > > (specifically gcc (Ubuntu 5.4.0-6ubuntu1~16.04.4) 5.4.0 20160609) are > > not smart enough to figure it out for expressions involving local > > variables. This is harmless but means that the check is ineffective for > > these platforms. To fix, replace the variable with macros. > > > > Reported-by: Peter Maydell <peter.maydell@linaro.org> > > Signed-off-by: Michael S. Tsirkin <mst@redhat.com> > > Ugh.. I pulled this into my tree already (I've been having trouble > with my test setup, which is why I haven't sent my own pullreq yet). > However, it needs a s/%u/%llu/ to compile correctly on some platforms. I need it to avoid breaking build. I fixed it up - want to drop it from you tree? Or leave it there - git will resolve the trivial conflict. > > --- > > hw/ppc/spapr.c | 14 ++++++++------ > > 1 file changed, 8 insertions(+), 6 deletions(-) > > > > diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c > > index a642e66..b81f2ac 100644 > > --- a/hw/ppc/spapr.c > > +++ b/hw/ppc/spapr.c > > @@ -2630,8 +2630,8 @@ static void spapr_phb_placement(sPAPRMachineState *spapr, uint32_t index, > > * 1TiB 64-bit MMIO windows for each PHB. > > */ > > const uint64_t base_buid = 0x800000020000000ULL; > > - const int max_phbs = > > - (SPAPR_PCI_LIMIT - SPAPR_PCI_BASE) / SPAPR_PCI_MEM64_WIN_SIZE - 1; > > +#define SPAPR_MAX_PHBS ((SPAPR_PCI_LIMIT - SPAPR_PCI_BASE) / \ > > + SPAPR_PCI_MEM64_WIN_SIZE - 1) > > int i; > > > > /* Sanity check natural alignments */ > > @@ -2640,12 +2640,14 @@ static void spapr_phb_placement(sPAPRMachineState *spapr, uint32_t index, > > QEMU_BUILD_BUG_ON((SPAPR_PCI_MEM64_WIN_SIZE % SPAPR_PCI_MEM32_WIN_SIZE) != 0); > > QEMU_BUILD_BUG_ON((SPAPR_PCI_MEM32_WIN_SIZE % SPAPR_PCI_IO_WIN_SIZE) != 0); > > /* Sanity check bounds */ > > - QEMU_BUILD_BUG_ON((max_phbs * SPAPR_PCI_IO_WIN_SIZE) > SPAPR_PCI_MEM32_WIN_SIZE); > > - QEMU_BUILD_BUG_ON((max_phbs * SPAPR_PCI_MEM32_WIN_SIZE) > SPAPR_PCI_MEM64_WIN_SIZE); > > + QEMU_BUILD_BUG_ON((SPAPR_MAX_PHBS * SPAPR_PCI_IO_WIN_SIZE) > > > + SPAPR_PCI_MEM32_WIN_SIZE); > > + QEMU_BUILD_BUG_ON((SPAPR_MAX_PHBS * SPAPR_PCI_MEM32_WIN_SIZE) > > > + SPAPR_PCI_MEM64_WIN_SIZE); > > > > - if (index >= max_phbs) { > > + if (index >= SPAPR_MAX_PHBS) { > > error_setg(errp, "\"index\" for PAPR PHB is too large (max %u)", > > - max_phbs - 1); > > + SPAPR_MAX_PHBS - 1); > > return; > > } > > > > -- > David Gibson | I'll have my music baroque, and my code > david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ > | _way_ _around_! > http://www.ozlabs.org/~dgibson
On Wed, Feb 01, 2017 at 03:37:54AM +0200, Michael S. Tsirkin wrote: > On Wed, Feb 01, 2017 at 09:16:28AM +1100, David Gibson wrote: > > On Tue, Jan 31, 2017 at 03:14:29PM +0200, Michael S. Tsirkin wrote: > > > We are switching BUILD_BUG_ON to verify that it's parameter is a > > > compile-time constant, and it turns out that some gcc versions > > > (specifically gcc (Ubuntu 5.4.0-6ubuntu1~16.04.4) 5.4.0 20160609) are > > > not smart enough to figure it out for expressions involving local > > > variables. This is harmless but means that the check is ineffective for > > > these platforms. To fix, replace the variable with macros. > > > > > > Reported-by: Peter Maydell <peter.maydell@linaro.org> > > > Signed-off-by: Michael S. Tsirkin <mst@redhat.com> > > > > Ugh.. I pulled this into my tree already (I've been having trouble > > with my test setup, which is why I haven't sent my own pullreq yet). > > However, it needs a s/%u/%llu/ to compile correctly on some platforms. > > I need it to avoid breaking build. I fixed it up - want to drop it > from you tree? Or leave it there - git will resolve the trivial > conflict. Since you've already sent a pull, I'm guessing it will be merged by the time I'm able to get my pullreq together. Since I'll rebase before that, I expect it will drop out of my tree naturally. > > > > --- > > > hw/ppc/spapr.c | 14 ++++++++------ > > > 1 file changed, 8 insertions(+), 6 deletions(-) > > > > > > diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c > > > index a642e66..b81f2ac 100644 > > > --- a/hw/ppc/spapr.c > > > +++ b/hw/ppc/spapr.c > > > @@ -2630,8 +2630,8 @@ static void spapr_phb_placement(sPAPRMachineState *spapr, uint32_t index, > > > * 1TiB 64-bit MMIO windows for each PHB. > > > */ > > > const uint64_t base_buid = 0x800000020000000ULL; > > > - const int max_phbs = > > > - (SPAPR_PCI_LIMIT - SPAPR_PCI_BASE) / SPAPR_PCI_MEM64_WIN_SIZE - 1; > > > +#define SPAPR_MAX_PHBS ((SPAPR_PCI_LIMIT - SPAPR_PCI_BASE) / \ > > > + SPAPR_PCI_MEM64_WIN_SIZE - 1) > > > int i; > > > > > > /* Sanity check natural alignments */ > > > @@ -2640,12 +2640,14 @@ static void spapr_phb_placement(sPAPRMachineState *spapr, uint32_t index, > > > QEMU_BUILD_BUG_ON((SPAPR_PCI_MEM64_WIN_SIZE % SPAPR_PCI_MEM32_WIN_SIZE) != 0); > > > QEMU_BUILD_BUG_ON((SPAPR_PCI_MEM32_WIN_SIZE % SPAPR_PCI_IO_WIN_SIZE) != 0); > > > /* Sanity check bounds */ > > > - QEMU_BUILD_BUG_ON((max_phbs * SPAPR_PCI_IO_WIN_SIZE) > SPAPR_PCI_MEM32_WIN_SIZE); > > > - QEMU_BUILD_BUG_ON((max_phbs * SPAPR_PCI_MEM32_WIN_SIZE) > SPAPR_PCI_MEM64_WIN_SIZE); > > > + QEMU_BUILD_BUG_ON((SPAPR_MAX_PHBS * SPAPR_PCI_IO_WIN_SIZE) > > > > + SPAPR_PCI_MEM32_WIN_SIZE); > > > + QEMU_BUILD_BUG_ON((SPAPR_MAX_PHBS * SPAPR_PCI_MEM32_WIN_SIZE) > > > > + SPAPR_PCI_MEM64_WIN_SIZE); > > > > > > - if (index >= max_phbs) { > > > + if (index >= SPAPR_MAX_PHBS) { > > > error_setg(errp, "\"index\" for PAPR PHB is too large (max %u)", > > > - max_phbs - 1); > > > + SPAPR_MAX_PHBS - 1); > > > return; > > > } > > > > > > >
diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c index a642e66..b81f2ac 100644 --- a/hw/ppc/spapr.c +++ b/hw/ppc/spapr.c @@ -2630,8 +2630,8 @@ static void spapr_phb_placement(sPAPRMachineState *spapr, uint32_t index, * 1TiB 64-bit MMIO windows for each PHB. */ const uint64_t base_buid = 0x800000020000000ULL; - const int max_phbs = - (SPAPR_PCI_LIMIT - SPAPR_PCI_BASE) / SPAPR_PCI_MEM64_WIN_SIZE - 1; +#define SPAPR_MAX_PHBS ((SPAPR_PCI_LIMIT - SPAPR_PCI_BASE) / \ + SPAPR_PCI_MEM64_WIN_SIZE - 1) int i; /* Sanity check natural alignments */ @@ -2640,12 +2640,14 @@ static void spapr_phb_placement(sPAPRMachineState *spapr, uint32_t index, QEMU_BUILD_BUG_ON((SPAPR_PCI_MEM64_WIN_SIZE % SPAPR_PCI_MEM32_WIN_SIZE) != 0); QEMU_BUILD_BUG_ON((SPAPR_PCI_MEM32_WIN_SIZE % SPAPR_PCI_IO_WIN_SIZE) != 0); /* Sanity check bounds */ - QEMU_BUILD_BUG_ON((max_phbs * SPAPR_PCI_IO_WIN_SIZE) > SPAPR_PCI_MEM32_WIN_SIZE); - QEMU_BUILD_BUG_ON((max_phbs * SPAPR_PCI_MEM32_WIN_SIZE) > SPAPR_PCI_MEM64_WIN_SIZE); + QEMU_BUILD_BUG_ON((SPAPR_MAX_PHBS * SPAPR_PCI_IO_WIN_SIZE) > + SPAPR_PCI_MEM32_WIN_SIZE); + QEMU_BUILD_BUG_ON((SPAPR_MAX_PHBS * SPAPR_PCI_MEM32_WIN_SIZE) > + SPAPR_PCI_MEM64_WIN_SIZE); - if (index >= max_phbs) { + if (index >= SPAPR_MAX_PHBS) { error_setg(errp, "\"index\" for PAPR PHB is too large (max %u)", - max_phbs - 1); + SPAPR_MAX_PHBS - 1); return; }
We are switching BUILD_BUG_ON to verify that it's parameter is a compile-time constant, and it turns out that some gcc versions (specifically gcc (Ubuntu 5.4.0-6ubuntu1~16.04.4) 5.4.0 20160609) are not smart enough to figure it out for expressions involving local variables. This is harmless but means that the check is ineffective for these platforms. To fix, replace the variable with macros. Reported-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> --- hw/ppc/spapr.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)