| Message ID | 1392652187-28381-4-git-send-email-imammedo@redhat.com |
|---|---|
| State | New |
| Headers | show |
On Mon, Feb 17, 2014 at 04:49:44PM +0100, Igor Mammedov wrote: > From: Wanlong Gao <gaowanlong@cn.fujitsu.com> > > If the total number of the assigned numa nodes memory is not > equal to the assigned ram size, it will write the wrong data > to ACPI table, then the guest will ignore the wrong ACPI table > and recognize all memory to one node. It's buggy, we should > check it to ensure that we write the right data to ACPI table. > > Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com> > Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> > Signed-off-by: Igor Mammedov <imammedo@redhat.com> > --- > numa.c | 10 ++++++++++ > 1 files changed, 10 insertions(+), 0 deletions(-) > > diff --git a/numa.c b/numa.c > index 7845036..d12a4f2 100644 > --- a/numa.c > +++ b/numa.c > @@ -151,6 +151,16 @@ void set_numa_nodes(void) > node_mem[i] = ram_size - usedmem; > } > > + uint64_t numa_total = 0; I was going to point out that variable declarations in the middle of blocks goes against coding style (at least I was told so), but my patch to amend CODING_STYLE to document it was ignored for 2 weeks, already. So, I am not sure we really have that rule. (Personally I am not against declaring variables in the middle of blocks, I think it makes the code more readable, and it is perfectly valid C99 code.) Reviewed-by: Eduardo Habkost <ehabkost@redhat.com> > + for (i = 0; i < nb_numa_nodes; i++) { > + numa_total += node_mem[i]; > + } > + if (numa_total != ram_size) { > + fprintf(stderr, "qemu: numa nodes total memory size " > + "should equal ram_size\n"); > + exit(1); > + } > + > for (i = 0; i < nb_numa_nodes; i++) { > if (!bitmap_empty(node_cpumask[i], MAX_CPUMASK_BITS)) { > break; > -- > 1.7.1 > >
Il 26/02/2014 18:15, Eduardo Habkost ha scritto: > On Mon, Feb 17, 2014 at 04:49:44PM +0100, Igor Mammedov wrote: >> From: Wanlong Gao <gaowanlong@cn.fujitsu.com> >> >> If the total number of the assigned numa nodes memory is not >> equal to the assigned ram size, it will write the wrong data >> to ACPI table, then the guest will ignore the wrong ACPI table >> and recognize all memory to one node. It's buggy, we should >> check it to ensure that we write the right data to ACPI table. >> >> Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com> >> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> >> Signed-off-by: Igor Mammedov <imammedo@redhat.com> >> --- >> numa.c | 10 ++++++++++ >> 1 files changed, 10 insertions(+), 0 deletions(-) >> >> diff --git a/numa.c b/numa.c >> index 7845036..d12a4f2 100644 >> --- a/numa.c >> +++ b/numa.c >> @@ -151,6 +151,16 @@ void set_numa_nodes(void) >> node_mem[i] = ram_size - usedmem; >> } >> >> + uint64_t numa_total = 0; > > I was going to point out that variable declarations in the middle of > blocks goes against coding style (at least I was told so), but my patch > to amend CODING_STYLE to document it was ignored for 2 weeks, already. > So, I am not sure we really have that rule. > > (Personally I am not against declaring variables in the middle of > blocks, I think it makes the code more readable, and it is perfectly > valid C99 code.) > > Reviewed-by: Eduardo Habkost <ehabkost@redhat.com> I agree, but I fixed it already. Paolo
diff --git a/numa.c b/numa.c index 7845036..d12a4f2 100644 --- a/numa.c +++ b/numa.c @@ -151,6 +151,16 @@ void set_numa_nodes(void) node_mem[i] = ram_size - usedmem; } + uint64_t numa_total = 0; + for (i = 0; i < nb_numa_nodes; i++) { + numa_total += node_mem[i]; + } + if (numa_total != ram_size) { + fprintf(stderr, "qemu: numa nodes total memory size " + "should equal ram_size\n"); + exit(1); + } + for (i = 0; i < nb_numa_nodes; i++) { if (!bitmap_empty(node_cpumask[i], MAX_CPUMASK_BITS)) { break;