Re: [Gems-users] A question on ruby tester

["Lei Yang" <lya755@xxxxxxxxxxxxxxxxxxxx>]

That sounds reasonable. But in this line,

12689   0  -1        Seq                Done       >       [0x34ea, line 0x34c0] 1867 cycles NULL No

it shows first the actual address, then the aligned block address. So the traces does print word address too. It doesn't seem to me that this is a code consistency problem. Is it possible that there is a bug in the sequencer? I'm just curious.

Thanks,
Lei

----- Original Message -----
From: "Mike Marty" <mikem@xxxxxxxxxxx>
To: "Lei Yang" <lya755@xxxxxxxxxxxxxxxxxxxx>
Sent: Thursday, February 15, 2007 10:21 PM
Subject: Re: [Gems-users] A question on ruby tester


> It is the same block cache block.  It is probably just inconsistent code
> that prints the word address in one spot and the block address in another.
>
> --Mike
>
> Lei Yang wrote:
>> Thank you Mike. Did you mean g_SEQUENCER_OUTSTANDING_REQUESTS? I made
>> that 1 but still have the same problem. I guess I don't understand in the
>> first place why in cycle 10624  the request address is 0xdc0 but the
>> deadlock complains about 0xdee?
>>  Thanks a lot!
>> Lei
>>  ----- Original Message -----
>>
>>     *From:* Mike Marty <mailto:mikem@xxxxxxxxxxx>
>>     *To:* 'Lei Yang' <mailto:lya755@xxxxxxxxxxxxxxxxxxxx> ; 'Gems
>>     Users' <mailto:gems-users@xxxxxxxxxxx>
>>     *Sent:* Thursday, February 15, 2007 9:22 PM
>>     *Subject:* RE: [Gems-users] A question on ruby tester
>>
>>     A couple more tips:
>>
>>
>>     Take a look at the PROTOCOL_DEBUG_TRACE.  See if there is another
>>     block that has failed.  If too many requests are outstanding, the
>>     sequencer will start the request but there are no available
>>     resources to handle it.
>>
>>     There is a section on the Wiki on using the trace output.
>>
>>
>>     When first debugging a protocol, change g_OUTSTANDING_REQUESTS to
>>     1 until that works.
>>
>>
>>     --Mike
>>
>>
>>     -----Original Message-----
>>     *From:* gems-users-bounces@xxxxxxxxxxx
>>     [mailto:gems-users-bounces@xxxxxxxxxxx] *On Behalf Of *Lei Yang
>>     *Sent:* Thursday, February 15, 2007 4:18 PM
>>     *To:* Gems-users@xxxxxxxxxxx
>>     *Subject:* [Gems-users] A question on ruby tester
>>
>>
>>     Hello everyone,
>>
>>
>>     I am using Ruby tester to verify my cache coherence protocol and
>>     found a weird problem. Basically I got this error message of
>>     possible deadlock:
>>
>>
>>     Warning: in fn virtual void Sequencer::wakeup() in
>>     system/Sequencer.C:121: Possible Deadlock detected
>>     Warning: in fn virtual void Sequencer::wakeup() in
>>     system/Sequencer.C:122: request is [CacheMsg: Address=[0xdee, line
>>     0xdc0]
>>     Type=ATOMIC ProgramCounter=[0xf0, line 0xc0]
>>     AccessMode=SupervisorMode Size=1 Prefetch=No Version=0 Aborted=0
>>     Time=10624 ]
>>     Warning: in fn virtual void Sequencer::wakeup() in
>>     system/Sequencer.C:123: m_chip_ptr->getID() is 2
>>     Warning: in fn virtual void Sequencer::wakeup() in
>>     system/Sequencer.C:124: m_version is 0
>>     Warning: in fn virtual void Sequencer::wakeup() in
>>     system/Sequencer.C:125: current_time is 1000002
>>     Warning: in fn virtual void Sequencer::wakeup() in
>>     system/Sequencer.C:126: request.getTime() is 10624
>>     Warning: in fn virtual void Sequencer::wakeup() in
>>     system/Sequencer.C:127: current_time - request.getTime() is 989378
>>     Warning: in fn virtual void Sequencer::wakeup() in
>>     system/Sequencer.C:128: keys.size() is 16
>>     Warning: in fn virtual void Sequencer::wakeup() in
>>     system/Sequencer.C:129: *m_writeRequestTable_ptr is [ [0xdc0, line
>>     0xdc0]=
>>
>>
>>
>>     I searched for address 0xdee and it appears NO where in the output
>>     trace. Then I went to request time 10624 and saw this:
>>
>>     10624   2  -1        Seq               Begin       >       [0xdc0,
>>     line 0xdc0]
>>
>>
>>     After this line there is no occurance of 0xdc0. It appears to me
>>     that after the request on 0xdc0 is issued by the sequencer,
>>     nothing else was done. I don't even know wheter it is instruction
>>     read or load/store.  Then somehow it leads to a deadlock. Could
>>     anyone please give me a hint on how might I find out the problem?
>>
>>
>>     Thank you very much!
>>
>>
>>     Lei
>>
>>
>
>