Re: [PATCH 3/3] x86/apic: Use smaller array for __apicid_to_node[] mapping

2015-10-13 Thread Thomas Gleixner
On Tue, 13 Oct 2015, Jiang Liu wrote: > On 2015/10/12 18:25, Thomas Gleixner wrote: > > On Mon, 12 Oct 2015, Daniel J Blueman wrote: > >> Another approach which may be suitable without changing SRAT parsing to be > >> after the memory allocator is up, is to exploit the associativity of the > >> bot

Re: [PATCH 3/3] x86/apic: Use smaller array for __apicid_to_node[] mapping

2015-10-13 Thread Jiang Liu
On 2015/10/12 18:25, Thomas Gleixner wrote: > On Mon, 12 Oct 2015, Daniel J Blueman wrote: >> Another approach which may be suitable without changing SRAT parsing to be >> after the memory allocator is up, is to exploit the associativity of the >> bottom APIC ID bits. > > What's the problem with m

Re: [PATCH 3/3] x86/apic: Use smaller array for __apicid_to_node[] mapping

2015-10-12 Thread Daniel J Blueman
On Fri, Oct 9, 2015 at 11:35 PM, Jiang Liu wrote: On 2015/10/3 3:12, Denys Vlasenko wrote: From: Daniel J Blueman The Intel x2APIC spec states the upper 16-bits of APIC ID is the cluster ID [1, p2-12], intended for future distributed systems. Beyond the legacy 8-bit APIC ID, Numascale N

Re: [PATCH 3/3] x86/apic: Use smaller array for __apicid_to_node[] mapping

2015-10-12 Thread Thomas Gleixner
On Mon, 12 Oct 2015, Daniel J Blueman wrote: > Another approach which may be suitable without changing SRAT parsing to be > after the memory allocator is up, is to exploit the associativity of the > bottom APIC ID bits. What's the problem with moving (SRAT/ACPI/whatever) APIC parsing after the mem

Re: [PATCH 3/3] x86/apic: Use smaller array for __apicid_to_node[] mapping

2015-10-09 Thread Jiang Liu
On 2015/10/3 3:12, Denys Vlasenko wrote: > From: Daniel J Blueman > > The Intel x2APIC spec states the upper 16-bits of APIC ID is the > cluster ID [1, p2-12], intended for future distributed systems. Beyond > the legacy 8-bit APIC ID, Numascale NumaConnect uses 4-bits for the > position of a ser

Re: [PATCH 3/3] x86/apic: Use smaller array for __apicid_to_node[] mapping

2015-10-03 Thread Denys Vlasenko
On 10/03/2015 09:44 AM, Ingo Molnar wrote: > > * Denys Vlasenko wrote: > >> @@ -56,16 +56,34 @@ early_param("numa", numa_setup); >> /* >> * apicid, cpu, node mappings >> */ >> -s16 __apicid_to_node[MAX_LOCAL_APICID] = { >> -[0 ... MAX_LOCAL_APICID-1] = NUMA_NO_NODE >> + >> +struct apici

Re: [PATCH 3/3] x86/apic: Use smaller array for __apicid_to_node[] mapping

2015-10-03 Thread Ingo Molnar
* Denys Vlasenko wrote: > @@ -56,16 +56,34 @@ early_param("numa", numa_setup); > /* > * apicid, cpu, node mappings > */ > -s16 __apicid_to_node[MAX_LOCAL_APICID] = { > - [0 ... MAX_LOCAL_APICID-1] = NUMA_NO_NODE > + > +struct apicid_to_node __apicid_to_node[NR_CPUS] = { > + [0 ... N

[PATCH 3/3] x86/apic: Use smaller array for __apicid_to_node[] mapping

2015-10-02 Thread Denys Vlasenko
From: Daniel J Blueman The Intel x2APIC spec states the upper 16-bits of APIC ID is the cluster ID [1, p2-12], intended for future distributed systems. Beyond the legacy 8-bit APIC ID, Numascale NumaConnect uses 4-bits for the position of a server on each axis of a multi-dimension torus; SGI NUMA