Encounter: command "reportGateCount" doesn't seem to work pr

T

Thomas Popp

Guest
Hi,

I have been using various releases of (SOC) encounter in the last years,
namely 5.2, 7.1, and since recently 8.1. It seems to me that in all
these releases, the command "reportGateCount" doesn't work properly (=
as described in the documentation).

However, when searching support.cadence.com or this newsgroup, I
couldn't find any reports about others also experiencing this strange
behavior. So I want to actively ask people about their experience with
this command.


The details of the strange behavior I see are:

My aim is to report the area of the whole design AND the area of EACH
submodule. Thus, I call the command:

reportGateCount -level 9999 -limit 0 -outfile "area.log"

However, the file area.log never contains the area separate figures for
all submodules. It even changes between different designs and encounter
releases what the command exactly reports. E.g. for a specific (small)
design, encounter 7.1 reports:

Gate area 54.6000 um^2
[0] amba_multiplier Gates=331 Cells=164 Area=18072.6 um^2
[1] interface Gates=111 Cells=52 Area=6078.8 um^2
[1] multiplier Gates=214 Cells=110 Area=11702.6 um^2
[1] PADS Gates=0 Cells=0 Area=0.0 um^2

In this case, 2 submodules are missing at level [2] underneath instance
"multiplier", namely instances called "datapath" and "control".
Additionally, the area of the PADS module is reported as being 0. This
is not true because this module contains several periphery cells. I
think, their area should also be considered since I didn't give the
parameter "-stdCellOnly".

When I process exactly the same design (exactly the same control script)
with encounter 8.1, the content of area.log is only:

Gate area 54.6000 um^2
[0] amba_multiplier Gates=314 Cells=134 Area=17144.4 um^2


So what the heck is going on!? Am I really the only one seeing this
weired behavior, which would mean I'm a really unlucky guy ;-) Or is
this a known issue and no one really bothers to write about it or even
to correct it? I would be really interested in gaining some more
insights into this "mystery".

Thanks for any help or hint.

-Thomas
 
Hi,

Is there not a single encounter user in this forum who can give me a
hint whether he experiences the same behavior as I or not?

Thanks,
-Thomas


Am 09.04.2010 10:46, schrieb Thomas Popp:
Hi,

I have been using various releases of (SOC) encounter in the last years,
namely 5.2, 7.1, and since recently 8.1. It seems to me that in all
these releases, the command "reportGateCount" doesn't work properly (=
as described in the documentation).

However, when searching support.cadence.com or this newsgroup, I
couldn't find any reports about others also experiencing this strange
behavior. So I want to actively ask people about their experience with
this command.


The details of the strange behavior I see are:

My aim is to report the area of the whole design AND the area of EACH
submodule. Thus, I call the command:

reportGateCount -level 9999 -limit 0 -outfile "area.log"

However, the file area.log never contains the area separate figures for
all submodules. It even changes between different designs and encounter
releases what the command exactly reports. E.g. for a specific (small)
design, encounter 7.1 reports:

Gate area 54.6000 um^2
[0] amba_multiplier Gates=331 Cells=164 Area=18072.6 um^2
[1] interface Gates=111 Cells=52 Area=6078.8 um^2
[1] multiplier Gates=214 Cells=110 Area=11702.6 um^2
[1] PADS Gates=0 Cells=0 Area=0.0 um^2

In this case, 2 submodules are missing at level [2] underneath instance
"multiplier", namely instances called "datapath" and "control".
Additionally, the area of the PADS module is reported as being 0. This
is not true because this module contains several periphery cells. I
think, their area should also be considered since I didn't give the
parameter "-stdCellOnly".

When I process exactly the same design (exactly the same control script)
with encounter 8.1, the content of area.log is only:

Gate area 54.6000 um^2
[0] amba_multiplier Gates=314 Cells=134 Area=17144.4 um^2


So what the heck is going on!? Am I really the only one seeing this
weired behavior, which would mean I'm a really unlucky guy ;-) Or is
this a known issue and no one really bothers to write about it or even
to correct it? I would be really interested in gaining some more
insights into this "mystery".

Thanks for any help or hint.

-Thomas
 
Thomas Popp wrote, on 04/15/10 07:11:
Hi,

Is there not a single encounter user in this forum who can give me a
hint whether he experiences the same behavior as I or not?

Thanks,
-Thomas
There are occasionally people who respond to Encounter questions (not really my
field). Better place to ask could be:

http://www.cadence.com/community/forums/29.aspx

Or (if you're at a European University - I guess so from your email address),
you could ask Europractice for support - and they can ask Cadence in turn if
they can't answer it themselves.

Regards,

Andrew.
 
在 2010年4月9日星期五UTC+8下午4时46分40秒,Thomas Popp写道:
Hi,

I have been using various releases of (SOC) encounter in the last years,
namely 5.2, 7.1, and since recently 8.1. It seems to me that in all
these releases, the command "reportGateCount" doesn't work properly (> as described in the documentation).

However, when searching support.cadence.com or this newsgroup, I
couldn't find any reports about others also experiencing this strange
behavior. So I want to actively ask people about their experience with
this command.


The details of the strange behavior I see are:

My aim is to report the area of the whole design AND the area of EACH
submodule. Thus, I call the command:

reportGateCount -level 9999 -limit 0 -outfile "area.log"

However, the file area.log never contains the area separate figures for
all submodules. It even changes between different designs and encounter
releases what the command exactly reports. E.g. for a specific (small)
design, encounter 7.1 reports:

Gate area 54.6000 um^2
[0] amba_multiplier Gates=331 Cells=164 Area=18072.6 um^2
[1] interface Gates=111 Cells=52 Area=6078.8 um^2
[1] multiplier Gates=214 Cells=110 Area=11702.6 um^2
[1] PADS Gates=0 Cells=0 Area=0.0 um^2

In this case, 2 submodules are missing at level [2] underneath instance
"multiplier", namely instances called "datapath" and "control".
Additionally, the area of the PADS module is reported as being 0. This
is not true because this module contains several periphery cells. I
think, their area should also be considered since I didn't give the
parameter "-stdCellOnly".

When I process exactly the same design (exactly the same control script)
with encounter 8.1, the content of area.log is only:

Gate area 54.6000 um^2
[0] amba_multiplier Gates=314 Cells=134 Area=17144.4 um^2


So what the heck is going on!? Am I really the only one seeing this
weired behavior, which would mean I'm a really unlucky guy ;-) Or is
this a known issue and no one really bothers to write about it or even
to correct it? I would be really interested in gaining some more
insights into this "mystery".

Thanks for any help or hint.

-Thomas

My experience using reportGateCount of soc81 is more strange, it just only reports the NAND gate area, not any other instance or summary at all!

eg:
loading place ...
loading route ...
encounter 1> reportGateCount
Gate area 9.9792 um^2
encounter 2>
 

Welcome to EDABoard.com

Sponsor

Back
Top