Re: [Gems-users] failed assertion 'id < m_nodes'

[Fengguang Song <song@xxxxxxxxxx>]

Mike,

This is my network file:
processors:8
procs_per_chip:2
L2banks:4
memories:2
bw_unit:10000

ext_node:L1Cache:0 int_node:0 link_latency:1 bw_multiplier:64
ext_node:L1Cache:1 int_node:1 link_latency:1 bw_multiplier:64
ext_node:L2Cache:0 int_node:3 link_latency:1 bw_multiplier:64
ext_node:Directory:0 int_node:4 link_latency:20 bw_multiplier:10
int_node:0 int_node:1 link_latency:1  bw_multiplier:16
int_node:0 int_node:2 link_latency:1  bw_multiplier:16
int_node:1 int_node:2 link_latency:1  bw_multiplier:16
int_node:3 int_node:2 link_latency:4  bw_multiplier:16
int_node:3 int_node:4 link_latency:20 bw_multiplier:10

ext_node:L1Cache:2 int_node:5 link_latency:1 bw_multiplier:64
ext_node:L1Cache:3 int_node:6 link_latency:1 bw_multiplier:64
ext_node:L2Cache:1 int_node:8 link_latency:1 bw_multiplier:64
int_node:5 int_node:6 link_latency:1 bw_multiplier:16
int_node:5 int_node:7 link_latency:1 bw_multiplier:16
int_node:6 int_node:7 link_latency:1 bw_multiplier:16
int_node:8 int_node:7 link_latency:4 bw_multiplier:16

int_node:3 int_node:4 link_latency:20 bw_multiplier:20
int_node:8 int_node:4 link_latency:20 bw_multiplier:20

ext_node:L1Cache:4 int_node:10 link_latency:1 bw_multiplier:64
ext_node:L1Cache:5 int_node:11 link_latency:1 bw_multiplier:64
ext_node:L2Cache:2 int_node:13 link_latency:1 bw_multiplier:64
ext_node:Directory:1 int_node:9 link_latency:20 bw_multiplier:10
int_node:10 int_node:11 link_latency:1 bw_multiplier:16
int_node:10 int_node:12 link_latency:1 bw_multiplier:16
int_node:11 int_node:12 link_latency:1 bw_multiplier:16
int_node:13 int_node:12 link_latency:4 bw_multiplier:16

ext_node:L1Cache:6 int_node:14 link_latency:1 bw_multiplier:64
ext_node:L1Cache:7 int_node:15 link_latency:1 bw_multiplier:64
ext_node:L2Cache:3 int_node:17 link_latency:1 bw_multiplier:64
int_node:14 int_node:15 link_latency:1 bw_multiplier:16
int_node:14 int_node:16 link_latency:1 bw_multiplier:16
int_node:15 int_node:16 link_latency:1 bw_multiplier:16
int_node:17 int_node:16 link_latency:4 bw_multiplier:16

int_node:13 int_node:9 link_latency:20 bw_multiplier:20
int_node:17 int_node:9 link_latency:20 bw_multiplier:20

int_node:4 int_node:9 link_latency:40 bw_multiplier:40


On Mon, 23 Apr 2007, Mike Marty wrote:

> Its likely something wrong with your network file.  The parser isn't the
> most robust in the world.  Post it here and I can take a quick look
>
>
>
> > Hi Mike,
> >
> > Yes, I'm using FILE_SPECIFIED and has defined my own network file.
> > Do you think it's related to # of memories? Because when I set
> > # of memories = 4, instead of 2, the error is gone.
> >
> > Now I am confused about how to configure the 4 directories.
> > I hope to have two distributed physical memories, how
> > should I configure them in the network file?
> >
> > Thanks,
> > Fengguang
> >
> > On Mon, 23 Apr 2007, Mike Marty wrote:
> >
> > > What NETWORK_TOPOLOGY were you using?  Your configuration is legal and
> > > should work for PT_TO_PT and other auto-generated topologies.  If you were
> > > using FILE_SPECIFIED, you will need to create network specification files
> > > to manage your configuration
> > >
> > > > Hi,
> > > >
> > > > I used the MOESI_CMP_directory protocol to build
> > > > Ruby. When I ran "ruby0.init", I got the following error:
> > > > ...
> > > > id = 14, m_nodes = 14
> > > > failed assertion 'id < m_nodes' at fn void
> > > > SimpleNetwork::checkNetworkAllocation(NodeID, bool, int) in
> > > > network/simple/SimpleNetwork.C:200
> > > > failed assertion 'id < m_nodes' at fn void
> > > > SimpleNetwork::checkNetworkAllocation(NodeID, bool, int) in
> > > > network/simple/SimpleNetwork.C:200processors:
> > > >
> > > > My file_specified network file:
> > > > processors:8
> > > > procs_per_chip:2
> > > > L2banks:4
> > > > memories:2
> > > > bw_unit:10000
> > > > ...
> > > >
> > > > The above system will have 4 chips, two of which compose a board
> > > > with private memory/directory. Is it a legal configuration?
> > > > How to choose a correct number of memories?
> > > >
> > > > I know that m_nodes = 14 makes sense. But how do you name those
> > > > id's (mapping object to id)?
> > > >
> > > > Thanks much,
> > > > Fengguang
> > > > _______________________________________________
> > > > Gems-users mailing list
> > > > Gems-users@xxxxxxxxxxx
> > > > https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> > > > Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.
> > > _______________________________________________
> > > Gems-users mailing list
> > > Gems-users@xxxxxxxxxxx
> > > https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> > > Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.
> > _______________________________________________
> > Gems-users mailing list
> > Gems-users@xxxxxxxxxxx
> > https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> > Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.
--
Fengguang Song                  Office: 321 Claxton Complex
Department of Computer Science           (865) 974-0517 (O)
University of Tennessee, Knoxville 37996-3450
URL: http://www.cs.utk.edu/~song