[gem5-users] Cannot locate m5op.h under util/m5 directory
Hi All, I would like to create checkpoints in my C++ application in se mode. I was following this <http://arm.ecs.soton.ac.uk/wp-content/uploads/2016/10/gem5_tutorial.pdf> tutorial where we need to include the header file m5op.h. In the latest version of gem5, I can't find the m5op.h or m5ops.h file under util/m5 directory. Am I missing something? Can anyone please guide me on that? Thanks and Regards, Mitali Sinha, Ph.D. Scholar, Advanced Multicore Systems Lab, Department of Computer Science and Engineering, Indraprastha Institute of Information Technology Delhi, India (IIITD), website: https://www.iiitd.edu.in/noc/ ___ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
Re: [gem5-users] [EXT] Full System Simulation complains about a missing "vmlinux.aarch32.ll_20131205.0-gem5"
You should check in the configs/common/FSConfig.py. The kernel and dtb names are mentioned there. The default names are set according to the 2014 image files. Since you are using the 2018 image, so gem5 is not able to load your kernels and images. You can do this: 1. Modify the names in FSConfig.py according to your downloaded images and kernels. 2. Download the default one. As you also mentioned that you have tried out the 2014 files, you should again go check if all the files mentioned in FSConfig.py are matching (kernel binaries and dtbs). Regards, Mitali Sinha, Ph.D. Scholar, Advanced Multicore Systems Lab, Department of Computer Science and Engineering, Indraprastha Institute of Information Technology Delhi, India On Wed, Oct 17, 2018 at 10:52 PM Vitorio Cargnini (lcargnini) < lcargn...@micron.com> wrote: > Amine, > > Two possible things: > > One: set M5_PATH environment variable > Two: modify configs/common/SysPath.py and edit the path and remove /m5 and > add your own path > > Try one of them or both fixes. > > > Regards, > Luis Vitorio. > > > From: gem5-users [mailto:gem5-users-boun...@gem5.org] On Behalf Of Amine > Marref > Sent: Wednesday, October 17, 2018 2:22 AM > To: m5-us...@m5sim.org > Subject: [EXT] [gem5-users] Full System Simulation complains about a > missing "vmlinux.aarch32.ll_20131205.0-gem5" > > Dear All, > > I am trying to use gem5 in FS mode. When I pass the disk image's name to > the simulation command, I get the error message " IOError: Can't find file > 'vmlinux.aarch32.ll_20131205.0-gem5' on path". > > The disk image I am running is the one from the archive " > http://www.gem5.org/dist/current/arm/aarch-system-20180409.tar.xz"; which > does not contain the file mentioned in the error message. > > Furthermore, I have used the flags --disk-image, --kernel, and --dtb to > specify exactly the images/binaries that need to be used; however, with no > luck. > > My M5_PATH is set correctly. > > In [1-5] below, I show relevant information to the issue. > > I also downloaded the system images contained in the archive " > http://www.gem5.org/dist/current/arm/aarch-system-2014-10.tar.xz"; (which > contains the file mentioned in the error message), set the M5_PATH to point > to the new location, and ran FS simulation. This time it did not produce > the previous error message, but simulation got stuck after producing the > message "Kernel panic in simulated kernel". > > I too am stuck. Any ideas? > > I am running gem version 2.0 inside a virtual machine of Ubuntu 16.04. > > Thanks, > Amine. > > [1] The variable M5_PATH is set correctly. > amine@ubuntu:~/Workspace/full-systems/aarch-system-20180409$ echo $M5_PATH > /home/amine/Workspace/full-systems/aarch-system-20180409/ > > [2] The content of the archive that contains the system images. > amine@ubuntu:~/Workspace/full-systems/aarch-system-20180409$ ls -R > .: > binaries disks m5out revisions > > ./binaries: > aarch64_gem5_server.dtb > armv7_gem5_v1_16cpu.dtb > armv7_gem5_v1_1cpu.dtb > armv7_gem5_v1_2cpu.dtb > armv7_gem5_v1_4cpu.dtb > armv7_gem5_v1_8cpu.dtb > armv8_gem5_v1_16cpu.dtb > armv8_gem5_v1_1cpu.dtb > armv8_gem5_v1_2cpu.dtb > armv8_gem5_v1_4cpu.dtb > armv8_gem5_v1_8cpu.dtb > armv8_gem5_v1_big_little_2_2.dtb > armv8_gem5_v1_big_little_2_4.dtb > binaries > boot.arm > boot_emm.arm > boot_emm.arm64 > m5.aarch32 > m5.aarch64 > vexpress-v2p-ca15-tc1-gem5_1cpus.dtb > vexpress-v2p-ca15-tc1-gem5_2cpus.dtb > vexpress-v2p-ca15-tc1-gem5_4cpus.dtb > vexpress-v2p-ca15-tc1-gem5_dvfs_1cpus.dtb > vexpress-v2p-ca15-tc1-gem5_dvfs_2cpus.dtb > vexpress-v2p-ca15-tc1-gem5_dvfs_4cpus.dtb > vexpress-v2p-ca15-tc1-gem5_dvfs_per_core_2cpus.dtb > vexpress-v2p-ca15-tc1-gem5_dvfs_per_core_4cpus.dtb > vmlinux.vexpress_emm > vmlinux.vexpress_emm64 > vmlinux.vexpress_gem5_v1 > vmlinux.vexpress_gem5_v1_64 > > ./disks: > aarch32-ubuntu-natty-headless.img linaro-minimal-aarch64.img > aarch64-ubuntu-trusty-headless.img linux-aarch32-ael.img > > ./m5out: > stats.txt > > ./revisions: > gem5 linux linux-arm64-legacy linux-arm-legacy > > [4] Running FS simulation fails. > amine@ubuntu:~/Workspace/full-systems/aarch-system-20180409$ > ~/Workspace/gem5/build/ARM/gem5.opt ~/Workspace/gem5/configs/example/fs.py > --disk-image=/home/amine/Workspace/full-systems/aarch-system-20180409/disks/aarch64-ubuntu-trusty-headless.img > gem5 Simulator System. http://gem5.org > gem5 is copyrighted software; use the --copyright option for details. > > gem5 compiled Oct 16 2018 21:43:32 > gem5 started Oct 17 2018 09:12:25 > gem5 executing on ubuntu, pid 57
Re: [gem5-users] Error in running Moby(Asimbench) benchmark in gem5
@Haeyoon Cho sorry for late reply.
I tried making the changes as follows:
gem5/configs/common/FSConfig.py:
default_dtbs = {
"RealViewEB": None,
"RealViewPBX": None,
"VExpress_EMM": "armv7_gem5_v1_%dcpu.20170616.dtb" % num_cpus,
"VExpress_EMM64": "vexpress.aarch64.20140821.dtb",
}
default_kernels = {
"RealViewEB": "vmlinux.arm.smp.fb.2.6.38.8",
"RealViewPBX": "vmlinux.smp.ics.arm.asimbench.2.6.35",
"VExpress_EMM64": "vmlinux.aarch64.20140821",
}
Then, I ran the following command:
build/ARM/gem5.opt configs/example/fs.py --os-type=android-ics
--machine-type=RealView_PBX --mem-size=512 --cpu-type=MinorCPU --caches
--disk-image=/home/sujay/gem5/full_system_asimbench/disks/ARMv7a-ICS-Android.SMP.Asimbench-v3.img
--kernel=/home/sujay/gem5/full_system_asimbench/binaries/vmlinux.smp.ics.arm.asimbench.2.6.35
--dtb-filename=/home/sujay/gem5/full_system_asimbench/binaries/armv7_gem5_v1_1cpu.dtb
--script=/home/sujay/gem5/full_system_asimbench/asimbench_boot_scripts/adobe.rcS
It gives the following error:
--- BEGIN LIBC BACKTRACE ---
build/ARM/gem5.opt(_Z15print_backtracev+0x28)[0xd99e78]
build/ARM/gem5.opt(_Z12abortHandleri+0x46)[0xdac9f6]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x11390)[0x7f5eb6c93390]
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0x38)[0x7f5eb569b428]
/lib/x86_64-linux-gnu/libc.so.6(abort+0x16a)[0x7f5eb569d02a]
build/ARM/gem5.opt[0x939dcf]
build/ARM/gem5.opt[0x123ff36]
build/ARM/gem5.opt(_ZN8DRAMCtrl4initEv+0x20)[0x12626e0]
build/ARM/gem5.opt[0x11c065a]
build/ARM/gem5.opt[0xa224b7]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x6f55)[0x7f5eb6f4fe65]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f5eb708704c]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x6ffd)[0x7f5eb6f4ff0d]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f5eb708704c]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x6ffd)[0x7f5eb6f4ff0d]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f5eb708704c]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCode+0x19)[0x7f5eb6f48d99]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x613b)[0x7f5eb6f4f04b]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f5eb708704c]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x6ffd)[0x7f5eb6f4ff0d]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x85c)[0x7f5eb708704c]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyEval_EvalCode+0x19)[0x7f5eb6f48d99]
/usr/lib/x86_64-linux-gnu/libpython2.7.so.1.0(PyRun_StringFlags+0x76)[0x7f5eb6fc31e6]
build/ARM/gem5.opt(_Z6m5MainiPPc+0x8f)[0xdab3ef]
build/ARM/gem5.opt(main+0x33)[0x8ba673]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0)[0x7f5eb5686830]
build/ARM/gem5.opt(_start+0x29)[0x8f6e19]
--- END LIBC BACKTRACE ---
Aborted (core dumped)
I don't know if I am doing it correctly. Also I am not sure if I am using
the correct dtb files and boot_emm.arm files that I have collected from
http://www.gem5.org/dist/current/arm/aarch-system-20180409.tar.xz.
<http://www.gem5.org/dist/current/arm/aarch-system-20180409.tar.xz> image.
Please help.
On Tue, May 29, 2018 at 8:43 PM, 조해윤 wrote:
> Dear Mitali Sinha.
>
> In my case, I run moby bench with RealView_PBX machine type.
> I don't know if it is the cause, but give it a try.
>
> Best Regard,
> Haeyoon Cho.
>
> 2018-05-29 20:39 GMT+09:00 Mitali Sinha :
>
>> I have followed the instructions provided in the gem5.org website to run
>> the Moby benchmarks on gem5 as follows:
>>
>> 1. Downloaded the Asimbench from https://bitbucket.org/yongbing
>> _huang/asimbench/downloads/ which contains the following folders and
>> files:
>> 1. asimbench_android_arm_kernel : vmlinux.smp.ics.arm.asimbench.
>> 2.6.35
>> 2. asimbench_boot_scripts : contains the rcs scripts for
>> different benchmarks
>> 3. asimbench_disk_image : ARMv7a-ICS-Android.SMP.Asimbench-v3.img;
>> sdcard-1g-mxplayer.img
>>
>> 2. I have used the "armv7_gem5_v1_1cpu.dtb" and "boot_emm.arm" binaries
>> downloaded from http://www.gem5.org/dist/current/arm/
>> aarch-system-20180409.tar.xz.
>> <http://www.gem5.org/dist/current/arm/aarch-system-20180409.tar.xz>
>>
>> 3. Made the following changes in the config/common/FSConfig.py :
>>
>> def makeArmSystem(..)
>> .
>> self.cf0 = CowIdeDisk(driveID='master')
>> self.cf2 = CowIdeDisk(driveID='master')
>> self.cf0.childImage(mde
[gem5-users] Error in running Moby(Asimbench) benchmark in gem5
I have followed the instructions provided in the gem5.org website to run
the Moby benchmarks on gem5 as follows:
1. Downloaded the Asimbench from
https://bitbucket.org/yongbing_huang/asimbench/downloads/ which contains
the following folders and files:
1. asimbench_android_arm_kernel :
vmlinux.smp.ics.arm.asimbench.2.6.35
2. asimbench_boot_scripts : contains the rcs scripts for
different benchmarks
3. asimbench_disk_image :
ARMv7a-ICS-Android.SMP.Asimbench-v3.img; sdcard-1g-mxplayer.img
2. I have used the "armv7_gem5_v1_1cpu.dtb" and "boot_emm.arm" binaries
downloaded from http://www.gem5.org/dist/current/arm/
aarch-system-20180409.tar.xz.
<http://www.gem5.org/dist/current/arm/aarch-system-20180409.tar.xz>
3. Made the following changes in the config/common/FSConfig.py :
def makeArmSystem(..)
.
self.cf0 = CowIdeDisk(driveID='master')
self.cf2 = CowIdeDisk(driveID='master')
self.cf0.childImage(mdesc.disk())
self.cf2.childImage(disk("sdcard-1g-mxplayer.img"))
# Old platforms have a built-in IDE or CF controller. Default to
# the IDE controller if both exist. New platforms expect the
# storage controller to be added from the config script.
if hasattr(self.realview, "ide"):
#self.realview.ide.disks = [self.cf0]
self.realview.ide.disks = [self.cf0, self.cf2]
elif hasattr(self.realview, "cf_ctrl"):
#self.realview.cf_ctrl.disks = [self.cf0]
self.realview.cf_ctrl.disks = [self.cf0, self.cf2]
else:
self.pci_ide = IdeController(disks=[self.cf0])
pci_devices.append(self.pci_ide)
...
4. Used the following command:
build/ARM/gem5.opt configs/example/fs.py --os-type=android-ics
--machine-type=VExpress_EMM --cpu-type=MinorCPU --caches
--disk-image=/home/sujay/gem5/full_system_asimbench/disks/ARMv7a-ICS-Android.SMP.Asimbench-v3.img
--kernel=/home/sujay/gem5/full_system_asimbench/binaries/vmlinux.smp.ics.arm.asimbench.2.6.35
--dtb-filename=/home/sujay/gem5/full_system_asimbench/binaries/armv7_gem5_v1_1cpu.dtb
--script=/home/sujay/gem5/full_system_asimbench/asimbench_boot_scripts/adobe.rcS
5. The above commands throw the following error:
..
Global frequency set at 1 ticks per second
warn: DRAM device capacity (8192 Mbytes) does not match the address range
assigned (512 Mbytes)
info: kernel located at:
/home/sujay/gem5/full_system_asimbench/binaries/vmlinux.smp.ics.arm.asimbench.2.6.35
Listening for system connection on port 5900
Listening for system connection on port 3456
0: system.remote_gdb: listening for remote gdb on port 7000
info: Using bootloader at address 0x10
info: Using kernel entry physical address at 0x80008000
warn: DTB file specified, but no device tree support in kernel
REAL SIMULATION
warn: Existing EnergyCtrl, but no enabled DVFSHandler found.
info: Entering event queue @ 0. Starting simulation...
*warn: Device system.membus.badaddr_responder accessed by read to address
0x10009018 size=4fatal: Received error response packet for inst:
0/13.7/35/140.143 pc: 0x8003aba0 (ldr)*
Memory Usage: 823816 KBytes
Looking for some suggestion regarding successfully running the asimbench on
gem5. Any help is appreciated. Thanks in advance.
--
Mitali Sinha,
PhD Scholar,
IIIT Delhi
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
[gem5-users] Kernel panic - not syncing: No working init found.
I am trying to run a simple hello world program in FS mode (ARM ISA)
following these steps:
$ wget http://www.m5sim.org/dist/current/arm/aarch-system-2014-10.tar.xz
$ tar xf aarch-system-2014-10.tar.xz
$ mkdir tmpdir
$ sudo mount -o loop,offset=32256 disks/aarch64-ubuntu-trusty-headless.img
tmpdir
$ cd tmpdir
$ cd etc/init
$ sudo cp /home/iiitd/Desktop/tty-gem5.conf .
$ cd..
$ cd sbin
$ sudo cp /home/iiitd/gem5/util/m5/m5 .
$ cd ..
$ sudo mkdir benchmark
$ sudo cp /home/iiitd/Desktop/mine/hello6 benchmark
$ cd..
$ sudo umount tmpdir
$ echo "export M5_PATH=/home/iiitd/full-system-2014" >> ~/.bashrc
$ source ~/.bashrc
$ cd gem5
$ build/ARM/gem5.opt configs/example/fs.py
--disk-image=/home/iiitd/full-system-2014/disks/aarch64-ubuntu-trusty-headless.img
--script=./home/iiitd/a.rcS
In another terminal:
$ cd gem5/util/term
$ ./m5term localhost 3457
m5 slave terminal: Terminal 0
Freeing unused kernel memory: 292K (806aa000 - 806f3000)
request_module: runaway loop modprobe binfmt-464c
Starting init: /sbin/init exists but couldn't execute it (error -8)
Starting init: /etc/init exists but couldn't execute it (error -13)
request_module: runaway loop modprobe binfmt-464c
Starting init: /bin/sh exists but couldn't execute it (error -8)
Kernel panic - not syncing: No working init found. Try passing init=
option to kernel. See Linux Documentation/init.txt for guidance.
CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.13.0-rc2 #1
[<80016954>] (unwind_backtrace+0x0/0xf4) from [<80012424>]
(show_stack+0x10/0x14)
[<80012424>] (show_stack+0x10/0x14) from [<804f9bd4>] (dump_stack+0x98/0xc4)
[<804f9bd4>] (dump_stack+0x98/0xc4) from [<804f6db4>] (panic+0xa0/0x1e4)
[<804f6db4>] (panic+0xa0/0x1e4) from [<804f4408>] (cpu_die+0x0/0x70)
[<804f4408>] (cpu_die+0x0/0x70) from [<>] ( (null))
I tried to use the recent images from "
http://www.m5sim.org/dist/current/arm/aarch-system-20170426.tar.xz and
followed the above steps.
:~/gem5$ build/ARM/gem5.opt configs/example/fs.py
--disk-image=/home/iiitd/full-system-2014/disks/aarch64-ubuntu-trusty-headless.img
--script=./home/iiitd/a.rcS
gem5 Simulator System. http://gem5.org
---
-
Traceback (most recent call last):
File "", line 1, in
File "/home/iiitd/gem5/src/python/m5/main.py", line 400, in main
exec filecode in scope
File "configs/example/fs.py", line 341, in
test_sys = build_test_system(np)
File "configs/example/fs.py", line 103, in build_test_system
ruby=options.ruby)
File "/home/iiitd/gem5/configs/common/FSConfig.py", line 304, in
makeArmSystem
self.kernel = binary(default_kernels[machine_type])
File "/home/iiitd/gem5/configs/common/SysPaths.py", line 49, in binary
return searchpath(binary.path, filename)
File "/home/iiitd/gem5/configs/common/SysPaths.py", line 41, in searchpath
raise IOError, "Can't find file '%s' on path." % filename
IOError: Can't find file 'vmlinux.aarch32.ll_20131205.0-gem5' on path.
In the newly downloaded binaries, there is no file called
"vmlinux.aarch32.ll_20131205.0-gem5". So, I think I have to change the dtb
and kernel names in FSConfig.py into appropriate one. But, I am not sure
which dtb and kernel shall we use. I tried with the following with no
success:
In FSConfig.py:
default_dtbs = {
"RealViewEB": None,
"RealViewPBX": None,
"VExpress_EMM": "armv7_gem5_v1_16cpu.20170426.dtb",
"VExpress_EMM64":
"vexpress-v2p-ca15-tc1-gem5_4cpus.20170426.dtb",
}
default_kernels = {
"RealViewEB": "vmlinux.arm.smp.fb.2.6.38.8",
"RealViewPBX": "vmlinux.arm.smp.fb.2.6.38.8",
"VExpress_EMM": "vmlinux.vexpress_gem5_v1.20170426",
"VExpress_EMM64": "vmlinux.vexpress_gem5_v1_64.20170426",
}
Thank you.
--
Mitali Sinha,
PhD Scholar,
IIIT Delhi
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
[gem5-users] Segmentation fault running in FS mode
I am trying to run a simple hello world C++ program in FS mode in ALPHA ISA. I am following the steps as follows: 1) Cross compiling the program using a pre-compiled cross-compiler for target ALPHA. $ wget http://www.m5sim.org/dist/current/alphaev67-unknown-linux-gnu.tar.bz2 $ tar -xjvf alphaev67-unknown-linux-gnu.tar.bz2 $ export CXX=/home/iiitd/Desktop/compiler/compiler/alphaev67-unknown-linux-gnu/bin $ export CC=/home/iiitd/Desktop/compiler/compiler/alphaev67-unknown-linux-gnu/bin $ export PATH=$PATH:/home/iiitd/Desktop/compiler/compiler/alphaev67-unknown-linux-gnu/bin $ alphaev67-unknown-linux-gnu-g++ -static new1.cpp -o hello6 2) Placing the program into the disk image. I have downloaded the Full system image file from http://gem5.org/Download. $ ls test-full-system binaries disks $ ls test-full-system/disks linux-bigswap2.img linux-latest.img $ cd test-full-system $ mkdir tmpdir $ ls tmpdir benchmarks bin dev etc iscsi lib linuxrc lost+found mnt modules proc sbin sys tmp usr var $ sudo mount -o loop,offset=32256 disks/linux-latest.img tmpdir $ cd tmpdir $ sudo cp /home/iiitd/Desktop/hello6 benchmark $ cd .. $ sudo umount tmpdir 3) Created a .rcS file: #! /bin/sh cd benchmarks /sbin/m5 dumpresetstats echo "executing hello world" ./hello6 echo "Done" /sbin/m5 exit 4) Run in Gem5 $ echo "export M5_PATH=/home/iiitd/test-full-system" >> ~/.bashrc $ source ~/.bashrc $ cd gem5 $ build/ALPHA/gem5.opt configs/example/fs.py --disk-image=/home/iiitd/test-full-system/disks/linux-latest.img --script=/home/iiitd/a.rcS In another terminal: $ cd util/term $ ./m5term localhost 3457 m5 slave terminal: Terminal 0 M5 console: m5AlphaAccess @ 0xFD02 Got Configuration 623 memsize 2000 pages 1 First free page after ROM 0xFC018000 - - EXT2-fs warning: checktime reached, running e2fsck is recommended loading script... executing hello world Segmentation fault Done Also, should I use the linux-bigswap2.img instead of linux-latest.img? When I try to mount the linux-bigswap2.img, I get the following error: $ sudo mount -o loop,offset=32256 disks/linux-bigswap2.img tmpdir [sudo] password for iiitd: NTFS signature is missing. Failed to mount '/dev/loop2': Invalid argument The device '/dev/loop2' doesn't seem to have a valid NTFS. Maybe the wrong device is used? Or the whole disk instead of a partition (e.g. /dev/sda, not /dev/sda1)? Or the other way around? -- Mitali Sinha, PhD Scholar, IIIT Delhi ___ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
[gem5-users] Problem in using m5threads for multithreaded program
I am trying to use the m5threads library for running a multithreaded program in gem5. I am also cross compiling my program. My host is "x64" and target is "arm". I have some doubts: 1. After running the makefile it creates a directory libpthread.a in the m5threads directory and gives me the following warning: :~/Desktop/m5thread/m5threads$ make gcc -O3 -static -c -o pthread.o pthread.c pthread.c: In function ‘pthread_create’: pthread.c:256:3: *warning: implicit declaration of function ‘clone’ [-Wimplicit-function-declaration]* * clone(__pthread_trampoline, tcb->stack_start_addr, CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHA* * ^* ar rv libpthread.a pthread.o ar: creating libpthread.a a - pthread.o *make: Warning: Archive 'libpthread.a' seems to have been created in deterministic mode. '' will always be updated. Please consider passing the U flag to ar to avoid the problem.* rm pthread.o Does the above warning needs to be considered? If so what should I do? 2. After making the m5threads, how to link my program to it? For example for using pthread library, I have to use "gcc -pthread". 3. Do I need to make any changes in my program to use the m5threads? Thank you. ___ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
Re: [gem5-users] Error in running a CNN benchmark in gem5 SE mode
Thanks for your response.
I will try out your suggestions.
On Tue, Apr 18, 2017 at 12:31 AM, Alberto Álvarez <681...@unizar.es> wrote:
> Hello,
>
> As you can see in the error("fatal: syscall set_tid_address (#256)
> unimplemented"), the emulation of system call "set_tid_address" is not
> emulated.
>
> You can apply this explanation in future problems with "syscall X
> unimplemmented".
>
> I propose you several options:
>
> 1. Use FS mode.
> 2. Try to change de benchmark code to avoid this syscall.
> 3. Try to implement the syscall emulation.
>
> I hope to help you.
>
> Best regards,
>
> Alberto.
>
>
>
>
>
>
> El 17 abr. 2017 20:48, Mitali Sinha escribió:
>
> Hi,
>
> I am trying to run tiny-dnn benchmarks (convolutional neural network
> benchmarks written in C++ ) in gem5 SE mode.
>
> Here are the steps I followed:
>
> 1. Cross-compiled some benchmark examples using a cross compiler (built
> using croosstool-ng) for ARM.
>
> 2. Run the executable in gem5 SE mode using following command:
> $ ./build/ARM/gem5.opt configs/example/se.py -c
> /home/iiitd/Desktop/tiny-dnn/examples/mnist/train -o
> /home/iiitd/Desktop/tiny-dnn/data
> ---
>
> Global frequency set at 1 ticks per second
> warn: DRAM device capacity (8192 Mbytes) does not match the
> address range assigned (512 Mbytes)
> 0: system.remote_gdb.listener: listening for remote gdb #0 on port
> 7000
> warn: ClockedObject: More than one power state change request
> encountered within the same simulation tick
> REAL SIMULATION
> info: Entering event queue @ 0. Starting simulation...
> fatal: syscall set_tid_address (#256) unimplemented.
>
> 3. I tried to ignore the error by making following changes in
> arch/arm/linux/process.cc
> SyscallDesc("set_tid_address", unimplementedFunc)
> to SyscallDesc("set_tid_address", ignoreFunc)
> SyscallDesc("set_robust_list", unimplementedFunc) to
> SyscallDesc("set_robust_list", ignoreFunc)
>
> 4. Run the following command
> $ scons build/ARM/gem5.opt -j4
> $ ./build/ARM/gem5.opt configs/example/se.py -c
> /home/iiitd/Desktop/tiny-dnn/examples/mnist/train -o
> /home/iiitd/Desktop/tiny-dnn/data
>
> ---
> Global frequency set at 1 ticks per second
> warn: DRAM device capacity (8192 Mbytes) does not match the address
> range assigned (512 Mbytes)
> 0: system.remote_gdb.listener: listening for remote gdb #0 on port
> 7000
> warn: ClockedObject: More than one power state change request
> encountered within the same simulation tick
> REAL SIMULATION
> info: Entering event queue @ 0. Starting simulation...
> warn: ignoring syscall set_tid_address(...)
> warn: ignoring syscall set_robust_list(...)
> warn: ignoring syscall rt_sigaction(...)
> (further warnings will be suppressed)
> warn: ignoring syscall rt_sigprocmask(...)
> (further warnings will be suppressed)
> warn: readlink() called on '/proc/self/exe' may yield unexpected
> results in various settings.
> Returning '/home/iiitd/Desktop/tiny-dnn-master2/examples/mnist/train'
> info: Increasing stack size by one page.
> warn: ignoring syscall futex(...)
> (further warnings will be suppressed)
> load models...
>
> The simulation freezes after the above steps. Any help is appreciated.
>
> Thanks.
>
>
>
>
> ___
> gem5-users mailing list
> gem5-users@gem5.org
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
[gem5-users] Error in running a CNN benchmark in gem5 SE mode
Hi,
I am trying to run tiny-dnn benchmarks (convolutional neural network
benchmarks written in C++ ) in gem5 SE mode.
Here are the steps I followed:
1. Cross-compiled some benchmark examples using a cross compiler (built
using croosstool-ng) for ARM.
2. Run the executable in gem5 SE mode using following command:
$ ./build/ARM/gem5.opt configs/example/se.py -c
/home/iiitd/Desktop/tiny-dnn/examples/mnist/train -o
/home/iiitd/Desktop/tiny-dnn/data
---
Global frequency set at 1 ticks per second
warn: DRAM device capacity (8192 Mbytes) does not match the address
range assigned (512 Mbytes)
0: system.remote_gdb.listener: listening for remote gdb #0 on port
7000
warn: ClockedObject: More than one power state change request
encountered within the same simulation tick
REAL SIMULATION
info: Entering event queue @ 0. Starting simulation...
fatal: syscall set_tid_address (#256) unimplemented.
3. I tried to ignore the error by making following changes in
arch/arm/linux/process.cc
SyscallDesc("set_tid_address", unimplementedFunc)
to SyscallDesc("set_tid_address", ignoreFunc)
SyscallDesc("set_robust_list", unimplementedFunc) to
SyscallDesc("set_robust_list", ignoreFunc)
4. Run the following command
$ scons build/ARM/gem5.opt -j4
$ ./build/ARM/gem5.opt configs/example/se.py -c
/home/iiitd/Desktop/tiny-dnn/examples/mnist/train -o
/home/iiitd/Desktop/tiny-dnn/data
---
Global frequency set at 1 ticks per second
warn: DRAM device capacity (8192 Mbytes) does not match the address
range assigned (512 Mbytes)
0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000
warn: ClockedObject: More than one power state change request
encountered within the same simulation tick
REAL SIMULATION
info: Entering event queue @ 0. Starting simulation...
warn: ignoring syscall set_tid_address(...)
warn: ignoring syscall set_robust_list(...)
warn: ignoring syscall rt_sigaction(...)
(further warnings will be suppressed)
warn: ignoring syscall rt_sigprocmask(...)
(further warnings will be suppressed)
warn: readlink() called on '/proc/self/exe' may yield unexpected
results in various settings.
Returning '/home/iiitd/Desktop/tiny-dnn-master2/examples/mnist/train'
info: Increasing stack size by one page.
warn: ignoring syscall futex(...)
(further warnings will be suppressed)
load models...
The simulation freezes after the above steps. Any help is appreciated.
Thanks.
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
