Re: [gem5-dev] Arm fs test failure

Thank you for clearing that up, Gabe. I may have pressed ctrl+c with the wrong screen on focus by mistake. Att,Daniel Em quinta-feira, 22 de março de 2018 19:24:56 GMT+1, Gabe Black escreveu: The "Test failed: Unknown exit cause: user interrupt received"

Re: [gem5-dev] Help with X86 ISA (RDTSC/RDTSCP) to read the correct clock cycles

Hi Li, This patch (https://gem5-review.googlesource.com/c/public/gem5/+/9042/ ) and a few after it should solve the problem. Interestingly, the reason I found this issue was for exactly the same reason that you did. I was trying to run

[gem5-dev] Change in public/gem5[master]: hsail: Get rid of an inert private member of StorageSpace.

Gabe Black has submitted this change and it was merged. ( https://gem5-review.googlesource.com/9221 ) Change subject: hsail: Get rid of an inert private member of StorageSpace. .. hsail: Get rid of an inert private member of

[gem5-dev] Help with X86 ISA (RDTSC/RDTSCP) to read the correct clock cycles

Hi all (specifically Gabe and Jason), I'm trying to use gem5 to reproduce the results of Spectre code ( https://github.com/crozone/SpectrePoC) in FS mode. It uses RDTSC/RDTSCP to measure the timing of a load operation. I have exactly the same issue with one previous post, the gem5 simply returns

[gem5-dev] Change in public/gem5[master]: cpu: Make the protobuf inst tracer accept variable sized instructions.

Gabe Black has submitted this change and it was merged. ( https://gem5-review.googlesource.com/7561 ) Change subject: cpu: Make the protobuf inst tracer accept variable sized instructions. .. cpu: Make the protobuf inst

Re: [gem5-dev] Multicore ARM v8 KVM based simulation

Ok, thanks. We're deciding internally what approach to use to tackle this. Gabe On Wed, Mar 21, 2018 at 3:01 AM, Andreas Sandberg wrote: > Hi Gabe, > > There are issues with the IDE model that prevent it from working with > in-kernel GIC emulation. I believe the model

Re: [gem5-dev] Arm fs test failure

The "Test failed: Unknown exit cause: user interrupt received" ultimately comes from a signal handler which is installed for SIGINT, so it looks like the test failed because somebody killed it with Ctrl+C. Gabe On Thu, Mar 22, 2018 at 7:46 AM, Nikos Nikoleris wrote: >

[gem5-dev] Learning gem5 Ruby protocol

Hi all, I would like to push the new Ruby protocol I wrote into the mainline before my tutorial on Saturday (so, tomorrow). Let me know if you see any issues before tomorrow, otherwise, I'm going to push as is. If there are any problems, I'll hold off. This shouldn't affect anything outside of

Re: [gem5-dev] Arm fs test failure

Nikos, It was only once, so I don't know what might be the cause. If I ever experience it again, I will let you know. Regards,Daniel Em quinta-feira, 22 de março de 2018 15:46:16 GMT+1, Nikos Nikoleris escreveu: Hi Daniel, I am afraid, I can't reproduce the

[gem5-dev] Change in public/gem5[master]: mem-cache: Create MRU replacement policy

Daniel Carvalho has submitted this change and it was merged. ( https://gem5-review.googlesource.com/8889 ) Change subject: mem-cache: Create MRU replacement policy .. mem-cache: Create MRU replacement policy Implementation

[gem5-dev] Change in public/gem5[master]: mem-cache: Split array indexing and replacement policies.

Daniel Carvalho has submitted this change and it was merged. ( https://gem5-review.googlesource.com/8501 ) Change subject: mem-cache: Split array indexing and replacement policies. .. mem-cache: Split array indexing and

Re: [gem5-dev] Arm fs test failure

Hi Daniel, I am afraid, I can't reproduce the error you're experiencing, and the outputs from your runs are not very telling of the problem. I run the realview-simple-atomic-checkpoint regression at: * 68af229490fc811aebddf68b3e2e09e63a5fa475 and * 0473286ab1e9992a906eff38bf90c82eeccb and

[gem5-dev] Arm fs test failure

While running the regression tests while at master commit  68af229490fc811aebddf68b3e2e09e63a5fa475, one of the tests failed with the following output: --- quick/fs/10.linux-boot/arm/linux/realview-simple-atomic-checkpoint --- *** gem5: ERROR: gem5 exited with non-zero status: 1 gem5 exited with

[gem5-dev] Broken gpu-ruby-GPU_RfO

Hi gem5ers, I noticed SE gpu-ruby-GPU_RfO test is broken on upstream, with an assertion failure stopping simulation. This happens in /src/gpu-compute/gpu_tlb.cc: 1547, since the success flag doesn't get updated after the second lookup and it is using a stale success value. Hope maintainers

[gem5-dev] Cron <m5test@zizzer> /z/m5/regression/do-regression quick

* build/RISCV/tests/opt/quick/se/02.insttest/riscv/linux-rv64c/minor-timing: FAILED! * build/RISCV/tests/opt/quick/se/02.insttest/riscv/linux-rv64c/o3-timing: FAILED! * build/RISCV/tests/opt/quick/se/02.insttest/riscv/linux-rv64c/simple-timing-ruby: FAILED! *