Functional reads in Ruby SE are a known and hard to solve problem, see https://gem5.atlassian.net/browse/GEM5-675 and https://gem5.atlassian.net/browse/GEM5-676
On Tue, Aug 18, 2020 at 6:33 PM VIPIN PATEL via gem5-users <gem5-users@gem5.org> wrote: > > Hi All, > > I using the GEM5 simulator to collect statistics of a micro-benchmark > program. I am encountering the functional read access failed for address > "0xXXXX". > > I have attached the source file of the micro-benchmark program. The > simulation is running fine for the case "1" and "4" in the switch construct. > The error is encountered for the cases "2" and "3" while I am using pthread > locks. > > I am using the MESI_Two_Level protocol with 4 core configuration( private L1I > and L1D cache for each core and shared L2 cache) in SE mode. > > I have read the discussion on google group > "https://groups.google.com/u/1/g/gem5-gpu-dev/c/Wt43jSYYXag" > and can infer the issue is generated when we have multiple copies of a block > in our system (might be in the transient state) and we tried to perform a > store or load to it. > > Can you point out the way to fix this? > Is pthread lock not supported in the gem5? Does the current release of gem5 > provide a workaround for this? > > Any help would be highly appreciated. Thanks in advance. > > Regards, > Vipin > Research Scholar > IIT Kanpur > _______________________________________________ > gem5-users mailing list -- gem5-users@gem5.org > To unsubscribe send an email to gem5-users-le...@gem5.org > %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s _______________________________________________ gem5-users mailing list -- gem5-users@gem5.org To unsubscribe send an email to gem5-users-le...@gem5.org %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s