> -----原始邮件----- > 发件人: gem5-users-requ...@gem5.org > 发送时间: 2021-06-26 15:30:05 (星期六) > 收件人: gem5-users@gem5.org > 抄送: > 主题: gem5-users Digest, Vol 179, Issue 42 > > Send gem5-users mailing list submissions to > gem5-users@gem5.org > > To subscribe or unsubscribe via email, send a message with subject or > body 'help' to > gem5-users-requ...@gem5.org > > You can reach the person managing the list at > gem5-users-ow...@gem5.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of gem5-users digest..." > > Today's Topics: > > 1. Fatal error while running multithreaded program in SE mode > (hissa alshamsi) > 2. gem5 RISCV, issue on boot when mounting filesystem (Truan David) > 3. Re: gem5 RISCV, issue on boot when mounting filesystem > (Νικόλαος Ταμπουρατζής) > 4. Re: gem5 RISCV, issue on boot when mounting filesystem > (Jason Lowe-Power) > 5. Re: Fatal error while running multithreaded program in SE mode > (Jason Lowe-Power) > 6. Issue in using HMC in SE mode (hissa alshamsi) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Fri, 25 Jun 2021 10:30:26 +0000 > From: hissa alshamsi <hissa.alsha...@hotmail.com> > Subject: [gem5-users] Fatal error while running multithreaded program > in SE mode > To: "gem5-users@gem5.org" <gem5-users@gem5.org> > Message-ID: <db8p191mb055503c7749ad3d535748b3281...@db8p191mb0555.eur> P191.PROD.OUTLOOK.COM> > Content-Type: multipart/alternative; boundary="_000_DB8P191MB055503 > C7749AD3D535748B3281069DB8P191MB0555EURP_" > > Hi everyone, > > I am trying to run a multithreaded program in SE mode, after installing m5thread and following the steps in https://github.com/WeijingShi/playground/blob/master/Run-openmp-code-in-gem5.md > > But I keep getting that syscall getdents unimplemented. > > This is the line I have used: > build/X86/gem5.opt configs/example/se.py --cpu-type=TimingSimpleCPU –cpu-cycle=2GH-n 8 --cmd=m5threads/tests/test_omp -o '8 8' > > > **** REAL SIMULATION **** > info: Entering event queue @ 0. Starting simulation... > warn: Attempting to open special file: /sys/devices/system/cpu. Ignoring. Simulation may take un-expected code path or be non-deterministic until proper handling is implemented. > fatal: syscall getdents (#78) unimplemented. > Memory Usage: 829528 Kbytes > > Any clue on how to run a multithreaded program in SE mode without using ruby? > > Thank you in advance, > Hessa. > -------------- next part -------------- > A message part incompatible with plain text digests has been removed ... > Name: not available > Type: text/html > Size: 2781 bytes > Desc: not available > > ------------------------------ > > Message: 2 > Date: Fri, 25 Jun 2021 14:00:54 +0000 > From: Truan David <david.tr...@heig-vd.ch> > Subject: [gem5-users] gem5 RISCV, issue on boot when mounting > filesystem > To: "gem5-users@gem5.org" <gem5-users@gem5.org> > Cc: Medwed Gregory <gregory.med...@heig-vd.ch>, Zapater Sancho Marina > <marina.zapa...@heig-vd.ch> > Message-ID: <fe300b8909964d5b8f621e8362207...@heig-vd.ch> > Content-Type: text/plain; charset="iso-8859-1" > > Hello, > We are a team working on gem5 RISCV. We are trying to use gem5 full-system but we are encountering some issues when mounting the rootfs/initramfs. > > This is our setup: > - gem5 on branch develop from https://gem5.googlesource.com/public/gem5 (c493d2c4ad) > - BBL on branche master from https://github.com/riscv/riscv-pk.git (e8e6b3aa) > - Linux 5.10, checkout on v5.10 tag (2c85ebc57) > > This is what we tried so far: > > -Compiling BBL with and without specifying the DTS: No changes > -Compiling BBL with Linux vmlinux OR Image as payload: No changes > -Compiling Linux with an minimalistic initramfs which only prints a "Hello World" from the init script: No changes > -Using fs_linux.py OR run_riscv.py as the entry point: No changes > > Here are the boot logs from different experiments, with only the last lines of the boot log to keep this mail short: > > ===================== > This is the boot log when specifying a disk-image param, using fs_linux.py: > > Command: > $GEM5_FAST_BIN -v \ > -d $GEM5_OUTPUT \ > $GEM5_HOME/configs/example/riscv/fs_linux.py \ > --cpu-type=AtomicSimpleCPU \ > --cpu-clock=1GHz \ > -n 1 \ > --disk-image=$DISK \ > --kernel=<bbl> \ > --mem-type=DDR4_2400_4x16 \ > --mem-size=4GB \ > --command-line="root=/dev/vda ro console=ttyS0" > > > ==== m5 terminal: Terminal 0 ==== > ... > [ 1.123014] [drm] radeon kernel modesetting enabled. > [ 1.146199] loop: module loaded > [ 1.147219] virtio_blk virtio0: [vda] 6821 512-byte logical blocks (3.49 MB/3.33 MiB) > [ 1.147423] vda: detected capacity change from 0 to 3492352 > > ===================== > This is the boot log when NOT specifying a disk-image param, using fs_linux.py: > > Command: > $GEM5_FAST_BIN -v \ > -d $GEM5_OUTPUT \ > $GEM5_HOME/configs/example/riscv/fs_linux.py \ > --cpu-type=AtomicSimpleCPU \ > --cpu-clock=1GHz \ > -n 1 \ > --kernel=<bbl> \ > --mem-type=DDR4_2400_4x16 \ > --mem-size=4GB \ > --command-line="root=/dev/vda ro console=ttyS0" > > ==== m5 terminal: Terminal 0 ==== > ... > [ 1.121246] [drm] radeon kernel modesetting enabled. > [ 1.144593] loop: module loaded > [ 1.148461] libphy: Fixed MDIO Bus: probed > [ 1.152291] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver > [ 1.152426] ehci-pci: EHCI PCI platform driver > [ 1.152703] ehci-platform: EHCI generic platform driver > [ 1.152981] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver > [ 1.153111] ohci-pci: OHCI PCI platform driver > [ 1.153376] ohci-platform: OHCI generic platform driver > [ 1.154497] usbcore: registered new interface driver uas > [ 1.154842] usbcore: registered new interface driver usb-storage > [ 1.155422] mousedev: PS/2 mouse device common for all mice > [ 1.156642] usbcore: registered new interface driver usbhid > [ 1.156760] usbhid: USB HID core driver > [ 1.161109] NET: Registered protocol family 10 > [ 1.164410] Segment Routing with IPv6 > [ 1.164841] sit: IPv6, IPv4 and MPLS over IPv4 tunneling driver > [ 1.167720] NET: Registered protocol family 17 > [ 1.168261] Key type dns_resolver registered > [ 1.170789] Freeing unused kernel memory: 184K > > > ===================== > This is the boot log using run_riscv.py: > > Command: > $GEM5_BIN -v \ > -d $GEM5_OUTPUT \ > $HOME/riscv-fs/configs-riscv-fs/run_riscv.py \ > $BOOTLOADER \ > $DISK \ > simple \ > 1 > > ==== m5 terminal: Terminal 0 ==== > ... > [ 1.040529] [drm] radeon kernel modesetting enabled. > [ 1.083453] loop: module loaded > [ 1.085896] virtio_blk virtio0: [vda] 1048576 512-byte logical blocks (537 MB/512 MiB) > [ 1.086433] vda: detected capacity change from 0 to 536870912 > [ 1.097333] libphy: Fixed MDIO Bus: probed > [ 1.099885] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver > [ 1.100302] ehci-pci: EHCI PCI platform driver > [ 1.101024] ehci-platform: EHCI generic platform driver > [ 1.101590] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver > [ 1.101965] ohci-pci: OHCI PCI platform driver > [ 1.102395] ohci-platform: OHCI generic platform driver > [ 1.104650] usbcore: registered new interface driver uas > [ 1.105608] usbcore: registered new interface driver usb-storage > [ 1.106993] mousedev: PS/2 mouse device common for all mice > [ 1.108385] usbcore: registered new interface driver usbhid > [ 1.109064] usbhid: USB HID core driver > [ 1.113051] NET: Registered protocol family 10 > [ 1.118537] Segment Routing with IPv6 > [ 1.119142] sit: IPv6, IPv4 and MPLS over IPv4 tunneling driver > [ 1.122581] NET: Registered protocol family 17 > [ 1.123375] Key type dns_resolver registered > [ 1.129869] Freeing unused kernel memory: 184K > ===================== > > > Everything works fine using QEMU, either using the full rootfs or only booting on the minimalistic initramfs: > > qemu-system-riscv64 -machine virt -m 512 -nographic \ > -kernel <bbl.bin> \ > -device loader,file=<linux 5.10="" with="" or="" without="" initramfs="">,addr=0x80200000 \ > -drive file=<full rootfs="">,format=raw,id=hd0 \ > -device virtio-blk-device,drive=hd0 \ > -netdev user,id=net0,hostfwd=tcp::2323-10.0.2.20:23,hostfwd=tcp::8080-10.0.2.20:80 \ > -device virtio-net-device,netdev=net0 \ > -append "root=/dev/vda ro console=ttyS0" > > Do you have any hint or do you spot something that is missing in our setup? I'll gladly provide more informations if needed. > > Thanks in advance for your answers! > > > > > David TRUAN > > Ingénieur Ra&D > Institut Reconfigurable & Embedded Digital Systems (REDS) > > david.tr...@heig-vd.ch > > Bureau A23 > > Haute École d'Ingénierie et de Gestion du Canton de Vaud > Route de Cheseaux 1 - CP 521 - 1401 Yverdon-les-Bains > > ------------------------------ > > Message: 3 > Date: Fri, 25 Jun 2021 18:20:22 +0300 > From: Νικόλαος Ταμπουρατζής <ntampourat...@ece.auth.gr> > Subject: [gem5-users] Re: gem5 RISCV, issue on boot when mounting > filesystem > To: gem5 users mailing list <gem5-users@gem5.org> > Message-ID: > <20210625182022.horde.fqjx7bfnkcd2drc6qwdi...@webmail.auth.gr> > Content-Type: text/plain; charset=utf-8; format=flowed; DelSp=Yes > > Dear David, > > I have used the bbl, Linux Kernel, and riscv_disk from the following > GitHub: https://github.com/ppeetteerrs/gem5-RISC-V-FS-Linux/ .You can > use either the files from prebuild directory or you can create your > own according to instructions of this Github. Specifically, I have > used the following configurations which are working properly (5 of 6): > > ------------ RISCV 1 core ------------ > > AtomicSimpleCPU (Working) > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=AtomicSimpleCPU > --disk-image=$OUT/riscv_disk -n 1 > > > TimingSimpleCPU (Working) > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=TimingSimpleCPU > --disk-image=$OUT/riscv_disk -n 1 > > DerivO3CPU (Working) > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=DerivO3CPU > --disk-image=$OUT/riscv_disk -n 1 > > > ------------ RISCV 2 cores ------------ > > AtomicSimpleCPU (Working) > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=AtomicSimpleCPU > --disk-image=$OUT/riscv_disk -n 2 > > TimingSimpleCPU (Working) > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=TimingSimpleCPU > --disk-image=$OUT/riscv_disk -n 2 > > DerivO3CPU (NOT Working): void BaseDynInst< <template-parameter-1-1> > >::initVars() [with Impl = O3CPUImpl]: Assertion `cpu->instcount <= > 1500' failed. > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=DerivO3CPU > --disk-image=$OUT/riscv_disk -n 2 > > > To be noticed that when I use --mem-size more than 512MB, gem5 is not booted. > > > Finally, when I use the last configuration I get the above error > (please let me know if you/anyone can resolve it :)) > > Best regards, > Nikolaos Tampouratzis > > > Quoting Truan David via gem5-users <gem5-users@gem5.org>: > > > Hello, > > We are a team working on gem5 RISCV. We are trying to use gem5 > > full-system but we are encountering some issues when mounting the > > rootfs/initramfs. > > > > This is our setup: > > - gem5 on branch develop from > > https://gem5.googlesource.com/public/gem5 (c493d2c4ad) > > - BBL on branche master from https://github.com/riscv/riscv-pk.git (e8e6b3aa) > > - Linux 5.10, checkout on v5.10 tag (2c85ebc57) > > > > This is what we tried so far: > > > > -Compiling BBL with and without specifying the DTS: No changes > > -Compiling BBL with Linux vmlinux OR Image as payload: No changes > > -Compiling Linux with an minimalistic initramfs which only prints a > > "Hello World" from the init script: No changes > > -Using fs_linux.py OR run_riscv.py as the entry point: No changes > > > > Here are the boot logs from different experiments, with only the > > last lines of the boot log to keep this mail short: > > > > ===================== > > This is the boot log when specifying a disk-image param, using fs_linux.py: > > > > Command: > > $GEM5_FAST_BIN -v \ > > -d $GEM5_OUTPUT \ > > $GEM5_HOME/configs/example/riscv/fs_linux.py \ > > --cpu-type=AtomicSimpleCPU \ > > --cpu-clock=1GHz \ > > -n 1 \ > > --disk-image=$DISK \ > > --kernel=<bbl> \ > > --mem-type=DDR4_2400_4x16 \ > > --mem-size=4GB \ > > --command-line="root=/dev/vda ro console=ttyS0" > > > > > > ==== m5 terminal: Terminal 0 ==== > > ... > > [ 1.123014] [drm] radeon kernel modesetting enabled. > > [ 1.146199] loop: module loaded > > [ 1.147219] virtio_blk virtio0: [vda] 6821 512-byte logical > > blocks (3.49 MB/3.33 MiB) > > [ 1.147423] vda: detected capacity change from 0 to 3492352 > > > > ===================== > > This is the boot log when NOT specifying a disk-image param, using > > fs_linux.py: > > > > Command: > > $GEM5_FAST_BIN -v \ > > -d $GEM5_OUTPUT \ > > $GEM5_HOME/configs/example/riscv/fs_linux.py \ > > --cpu-type=AtomicSimpleCPU \ > > --cpu-clock=1GHz \ > > -n 1 \ > > --kernel=<bbl> \ > > --mem-type=DDR4_2400_4x16 \ > > --mem-size=4GB \ > > --command-line="root=/dev/vda ro console=ttyS0" > > > > ==== m5 terminal: Terminal 0 ==== > > ... > > [ 1.121246] [drm] radeon kernel modesetting enabled. > > [ 1.144593] loop: module loaded > > [ 1.148461] libphy: Fixed MDIO Bus: probed > > [ 1.152291] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver > > [ 1.152426] ehci-pci: EHCI PCI platform driver > > [ 1.152703] ehci-platform: EHCI generic platform driver > > [ 1.152981] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver > > [ 1.153111] ohci-pci: OHCI PCI platform driver > > [ 1.153376] ohci-platform: OHCI generic platform driver > > [ 1.154497] usbcore: registered new interface driver uas > > [ 1.154842] usbcore: registered new interface driver usb-storage > > [ 1.155422] mousedev: PS/2 mouse device common for all mice > > [ 1.156642] usbcore: registered new interface driver usbhid > > [ 1.156760] usbhid: USB HID core driver > > [ 1.161109] NET: Registered protocol family 10 > > [ 1.164410] Segment Routing with IPv6 > > [ 1.164841] sit: IPv6, IPv4 and MPLS over IPv4 tunneling driver > > [ 1.167720] NET: Registered protocol family 17 > > [ 1.168261] Key type dns_resolver registered > > [ 1.170789] Freeing unused kernel memory: 184K > > > > > > ===================== > > This is the boot log using run_riscv.py: > > > > Command: > > $GEM5_BIN -v \ > > -d $GEM5_OUTPUT \ > > $HOME/riscv-fs/configs-riscv-fs/run_riscv.py \ > > $BOOTLOADER \ > > $DISK \ > > simple \ > > 1 > > > > ==== m5 terminal: Terminal 0 ==== > > ... > > [ 1.040529] [drm] radeon kernel modesetting enabled. > > [ 1.083453] loop: module loaded > > [ 1.085896] virtio_blk virtio0: [vda] 1048576 512-byte logical > > blocks (537 MB/512 MiB) > > [ 1.086433] vda: detected capacity change from 0 to 536870912 > > [ 1.097333] libphy: Fixed MDIO Bus: probed > > [ 1.099885] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver > > [ 1.100302] ehci-pci: EHCI PCI platform driver > > [ 1.101024] ehci-platform: EHCI generic platform driver > > [ 1.101590] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver > > [ 1.101965] ohci-pci: OHCI PCI platform driver > > [ 1.102395] ohci-platform: OHCI generic platform driver > > [ 1.104650] usbcore: registered new interface driver uas > > [ 1.105608] usbcore: registered new interface driver usb-storage > > [ 1.106993] mousedev: PS/2 mouse device common for all mice > > [ 1.108385] usbcore: registered new interface driver usbhid > > [ 1.109064] usbhid: USB HID core driver > > [ 1.113051] NET: Registered protocol family 10 > > [ 1.118537] Segment Routing with IPv6 > > [ 1.119142] sit: IPv6, IPv4 and MPLS over IPv4 tunneling driver > > [ 1.122581] NET: Registered protocol family 17 > > [ 1.123375] Key type dns_resolver registered > > [ 1.129869] Freeing unused kernel memory: 184K > > ===================== > > > > > > Everything works fine using QEMU, either using the full rootfs or > > only booting on the minimalistic initramfs: > > > > qemu-system-riscv64 -machine virt -m 512 -nographic \ > > -kernel <bbl.bin> \ > > -device loader,file=<linux 5.10="" with="" or="" without="" initramfs="">,addr=0x80200000 \ > > -drive file=<full rootfs="">,format=raw,id=hd0 \ > > -device virtio-blk-device,drive=hd0 \ > > -netdev > > user,id=net0,hostfwd=tcp::2323-10.0.2.20:23,hostfwd=tcp::8080-10.0.2.20:80 > > \ > > -device virtio-net-device,netdev=net0 \ > > -append "root=/dev/vda ro console=ttyS0" > > > > Do you have any hint or do you spot something that is missing in our > > setup? I'll gladly provide more informations if needed. > > > > Thanks in advance for your answers! > > > > > > > > > > David TRUAN > > > > Ingénieur Ra&D > > Institut Reconfigurable & Embedded Digital Systems (REDS) > > > > david.tr...@heig-vd.ch > > > > Bureau A23 > > > > Haute École d'Ingénierie et de Gestion du Canton de Vaud > > Route de Cheseaux 1 - CP 521 - 1401 Yverdon-les-Bains > > _______________________________________________ > > 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 > > > > ------------------------------ > > Message: 4 > Date: Fri, 25 Jun 2021 08:41:54 -0700 > From: Jason Lowe-Power <ja...@lowepower.com> > Subject: [gem5-users] Re: gem5 RISCV, issue on boot when mounting > filesystem > To: gem5 users mailing list <gem5-users@gem5.org> > Cc: Νικόλαος Ταμπουρατζής <ntampourat...@ece.auth.gr> > Message-ID: > <cafehpuf2zc9bnfzyqv8ecxclt03uriubbystzbczlr78dcy...@mail.gmail.com> > Content-Type: multipart/alternative; > boundary="000000000000ce11ba05c598fc7d" > > Hi everyone, > > These details on gem5-resources have also been tested multiple times. We > have also gotten unmodified OpenSBI working with gem5 as well. Ayaz can > provide more details if you need. > > https://gem5.googlesource.com/public/gem5-resources/+/refs/heads/stable/src/riscv-fs/ > > Cheers, > Jason > > On Fri, Jun 25, 2021 at 8:22 AM Νικόλαος Ταμπουρατζής via gem5-users < > gem5-users@gem5.org> wrote: > > > Dear David, > > > > I have used the bbl, Linux Kernel, and riscv_disk from the following > > GitHub: https://github.com/ppeetteerrs/gem5-RISC-V-FS-Linux/ .You can > > use either the files from prebuild directory or you can create your > > own according to instructions of this Github. Specifically, I have > > used the following configurations which are working properly (5 of 6): > > > > ------------ RISCV 1 core ------------ > > > > AtomicSimpleCPU (Working) > > > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=AtomicSimpleCPU > > --disk-image=$OUT/riscv_disk -n 1 > > > > > > TimingSimpleCPU (Working) > > > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=TimingSimpleCPU > > --disk-image=$OUT/riscv_disk -n 1 > > > > DerivO3CPU (Working) > > > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=DerivO3CPU > > --disk-image=$OUT/riscv_disk -n 1 > > > > > > ------------ RISCV 2 cores ------------ > > > > AtomicSimpleCPU (Working) > > > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=AtomicSimpleCPU > > --disk-image=$OUT/riscv_disk -n 2 > > > > TimingSimpleCPU (Working) > > > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=TimingSimpleCPU > > --disk-image=$OUT/riscv_disk -n 2 > > > > DerivO3CPU (NOT Working): void BaseDynInst< <template-parameter-1-1> > > >::initVars() [with Impl = O3CPUImpl]: Assertion `cpu->instcount <= > > 1500' failed. > > > > $GEM5/build/RISCV/gem5.opt -d $GEM5/node0 > > $GEM5/configs/example/riscv/fs_linux.py --kernel=$OUT/bbl --caches > > --mem-size=512MB --mem-type=DDR4_2400_8x8 --cpu-type=DerivO3CPU > > --disk-image=$OUT/riscv_disk -n 2 > > > > > > To be noticed that when I use --mem-size more than 512MB, gem5 is not > > booted. > > > > > > Finally, when I use the last configuration I get the above error > > (please let me know if you/anyone can resolve it :)) > > > > Best regards, > > Nikolaos Tampouratzis > > > > > > Quoting Truan David via gem5-users <gem5-users@gem5.org>: > > > > > Hello, > > > We are a team working on gem5 RISCV. We are trying to use gem5 > > > full-system but we are encountering some issues when mounting the > > > rootfs/initramfs. > > > > > > This is our setup: > > > - gem5 on branch develop from > > > https://gem5.googlesource.com/public/gem5 (c493d2c4ad) > > > - BBL on branche master from https://github.com/riscv/riscv-pk.git > > (e8e6b3aa) > > > - Linux 5.10, checkout on v5.10 tag (2c85ebc57) > > > > > > This is what we tried so far: > > > > > > -Compiling BBL with and without specifying the DTS: No changes > > > -Compiling BBL with Linux vmlinux OR Image as payload: No changes > > > -Compiling Linux with an minimalistic initramfs which only prints a > > > "Hello World" from the init script: No changes > > > -Using fs_linux.py OR run_riscv.py as the entry point: No changes > > > > > > Here are the boot logs from different experiments, with only the > > > last lines of the boot log to keep this mail short: > > > > > > ===================== > > > This is the boot log when specifying a disk-image param, using > > fs_linux.py: > > > > > > Command: > > > $GEM5_FAST_BIN -v \ > > > -d $GEM5_OUTPUT \ > > > $GEM5_HOME/configs/example/riscv/fs_linux.py \ > > > --cpu-type=AtomicSimpleCPU \ > > > --cpu-clock=1GHz \ > > > -n 1 \ > > > --disk-image=$DISK \ > > > --kernel=<bbl> \ > > > --mem-type=DDR4_2400_4x16 \ > > > --mem-size=4GB \ > > > --command-line="root=/dev/vda ro console=ttyS0" > > > > > > > > > ==== m5 terminal: Terminal 0 ==== > > > ... > > > [ 1.123014] [drm] radeon kernel modesetting enabled. > > > [ 1.146199] loop: module loaded > > > [ 1.147219] virtio_blk virtio0: [vda] 6821 512-byte logical > > > blocks (3.49 MB/3.33 MiB) > > > [ 1.147423] vda: detected capacity change from 0 to 3492352 > > > > > > ===================== > > > This is the boot log when NOT specifying a disk-image param, using > > > fs_linux.py: > > > > > > Command: > > > $GEM5_FAST_BIN -v \ > > > -d $GEM5_OUTPUT \ > > > $GEM5_HOME/configs/example/riscv/fs_linux.py \ > > > --cpu-type=AtomicSimpleCPU \ > > > --cpu-clock=1GHz \ > > > -n 1 \ > > > --kernel=<bbl> \ > > > --mem-type=DDR4_2400_4x16 \ > > > --mem-size=4GB \ > > > --command-line="root=/dev/vda ro console=ttyS0" > > > > > > ==== m5 terminal: Terminal 0 ==== > > > ... > > > [ 1.121246] [drm] radeon kernel modesetting enabled. > > > [ 1.144593] loop: module loaded > > > [ 1.148461] libphy: Fixed MDIO Bus: probed > > > [ 1.152291] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver > > > [ 1.152426] ehci-pci: EHCI PCI platform driver > > > [ 1.152703] ehci-platform: EHCI generic platform driver > > > [ 1.152981] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver > > > [ 1.153111] ohci-pci: OHCI PCI platform driver > > > [ 1.153376] ohci-platform: OHCI generic platform driver > > > [ 1.154497] usbcore: registered new interface driver uas > > > [ 1.154842] usbcore: registered new interface driver usb-storage > > > [ 1.155422] mousedev: PS/2 mouse device common for all mice > > > [ 1.156642] usbcore: registered new interface driver usbhid > > > [ 1.156760] usbhid: USB HID core driver > > > [ 1.161109] NET: Registered protocol family 10 > > > [ 1.164410] Segment Routing with IPv6 > > > [ 1.164841] sit: IPv6, IPv4 and MPLS over IPv4 tunneling driver > > > [ 1.167720] NET: Registered protocol family 17 > > > [ 1.168261] Key type dns_resolver registered > > > [ 1.170789] Freeing unused kernel memory: 184K > > > > > > > > > ===================== > > > This is the boot log using run_riscv.py: > > > > > > Command: > > > $GEM5_BIN -v \ > > > -d $GEM5_OUTPUT \ > > > $HOME/riscv-fs/configs-riscv-fs/run_riscv.py \ > > > $BOOTLOADER \ > > > $DISK \ > > > simple \ > > > 1 > > > > > > ==== m5 terminal: Terminal 0 ==== > > > ... > > > [ 1.040529] [drm] radeon kernel modesetting enabled. > > > [ 1.083453] loop: module loaded > > > [ 1.085896] virtio_blk virtio0: [vda] 1048576 512-byte logical > > > blocks (537 MB/512 MiB) > > > [ 1.086433] vda: detected capacity change from 0 to 536870912 > > > [ 1.097333] libphy: Fixed MDIO Bus: probed > > > [ 1.099885] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver > > > [ 1.100302] ehci-pci: EHCI PCI platform driver > > > [ 1.101024] ehci-platform: EHCI generic platform driver > > > [ 1.101590] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver > > > [ 1.101965] ohci-pci: OHCI PCI platform driver > > > [ 1.102395] ohci-platform: OHCI generic platform driver > > > [ 1.104650] usbcore: registered new interface driver uas > > > [ 1.105608] usbcore: registered new interface driver usb-storage > > > [ 1.106993] mousedev: PS/2 mouse device common for all mice > > > [ 1.108385] usbcore: registered new interface driver usbhid > > > [ 1.109064] usbhid: USB HID core driver > > > [ 1.113051] NET: Registered protocol family 10 > > > [ 1.118537] Segment Routing with IPv6 > > > [ 1.119142] sit: IPv6, IPv4 and MPLS over IPv4 tunneling driver > > > [ 1.122581] NET: Registered protocol family 17 > > > [ 1.123375] Key type dns_resolver registered > > > [ 1.129869] Freeing unused kernel memory: 184K > > > ===================== > > > > > > > > > Everything works fine using QEMU, either using the full rootfs or > > > only booting on the minimalistic initramfs: > > > > > > qemu-system-riscv64 -machine virt -m 512 -nographic \ > > > -kernel <bbl.bin> \ > > > -device loader,file=<linux 5.10="" with="" or="" without=""> > initramfs>,addr=0x80200000 \ > > > -drive file=<full rootfs="">,format=raw,id=hd0 \ > > > -device virtio-blk-device,drive=hd0 \ > > > -netdev > > > > > user,id=net0,hostfwd=tcp::2323-10.0.2.20:23,hostfwd=tcp::8080-10.0.2.20:80 > > > \ > > > -device virtio-net-device,netdev=net0 \ > > > -append "root=/dev/vda ro console=ttyS0" > > > > > > Do you have any hint or do you spot something that is missing in our > > > setup? I'll gladly provide more informations if needed. > > > > > > Thanks in advance for your answers! > > > > > > > > > > > > > > > David TRUAN > > > > > > Ingénieur Ra&D > > > Institut Reconfigurable & Embedded Digital Systems (REDS) > > > > > > david.tr...@heig-vd.ch > > > > > > Bureau A23 > > > > > > Haute École d'Ingénierie et de Gestion du Canton de Vaud > > > Route de Cheseaux 1 - CP 521 - 1401 Yverdon-les-Bains > > > _______________________________________________ > > > 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 > -------------- next part -------------- > A message part incompatible with plain text digests has been removed ... > Name: not available > Type: text/html > Size: 11753 bytes > Desc: not available > > ------------------------------ > > Message: 5 > Date: Fri, 25 Jun 2021 08:58:19 -0700 > From: Jason Lowe-Power <ja...@lowepower.com> > Subject: [gem5-users] Re: Fatal error while running multithreaded > program in SE mode > To: gem5 users mailing list <gem5-users@gem5.org> > Cc: hissa alshamsi <hissa.alsha...@hotmail.com> > Message-ID: > <cafehpuehp+q7amp6dv-l9tqnt4wmcl0vln+qtu55vmwxxru...@mail.gmail.com> > Content-Type: multipart/alternative; > boundary="0000000000007a911d05c5993730" > > What version of gem5 are you using? I believe gem5-20.0+ has the getdents > syscall implemented. I'm sure that 21.0 has the syscall implemented. > Whether you're using Ruby or classic caches it shouldn't make > any difference on whether the syscalls are implemented. > > Cheers, > Jason > > On Fri, Jun 25, 2021 at 3:32 AM hissa alshamsi via gem5-users < > gem5-users@gem5.org> wrote: > > > Hi everyone, > > > > > > > > I am trying to run a multithreaded program in SE mode, after installing > > m5thread and following the steps in > > https://github.com/WeijingShi/playground/blob/master/Run-openmp-code-in-gem5.md > > > > > > > > But I keep getting that syscall getdents unimplemented. > > > > > > > > This is the line I have used: > > > > build/X86/gem5.opt configs/example/se.py --cpu-type=TimingSimpleCPU > > –cpu-cycle=2GH-n 8 --cmd=m5threads/tests/test_omp -o '8 8' > > > > > > > > > > > > **** REAL SIMULATION **** > > > > info: Entering event queue @ 0. Starting simulation... > > > > warn: Attempting to open special file: /sys/devices/system/cpu. Ignoring. > > Simulation may take un-expected code path or be non-deterministic until > > proper handling is implemented. > > > > fatal: syscall getdents (#78) unimplemented. > > > > Memory Usage: 829528 Kbytes > > > > > > > > Any clue on how to run a multithreaded program in SE mode without using > > ruby? > > > > > > > > Thank you in advance, > > > > Hessa. > > _______________________________________________ > > 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 > -------------- next part -------------- > A message part incompatible with plain text digests has been removed ... > Name: not available > Type: text/html > Size: 2775 bytes > Desc: not available > > ------------------------------ > > Message: 6 > Date: Sat, 26 Jun 2021 06:16:32 +0000 > From: hissa alshamsi <hissa.alsha...@hotmail.com> > Subject: [gem5-users] Issue in using HMC in SE mode > To: "gem5-users@gem5.org" <gem5-users@gem5.org> > Message-ID: <db8p191mb05557016c4e644b7b7e1942881...@db8p191mb0555.eur> P191.PROD.OUTLOOK.COM> > Content-Type: multipart/alternative; boundary="_000_DB8P191MB055570 > 16C4E644B7B7E1942881059DB8P191MB0555EURP_" > > Hello, > > I am using gem5.20 and trying to run the hello world program in se.py using HMC, but I’m getting an error saying I’m trying to initiate an orphan node. > > > build/X86/gem5.opt configs/example/se.py --cmd=tests/test-progs/hello/bin/x86/linux/hello --mem-type=HMC_2500_1x32 > > Traceback (most recent call last): > File "<string>", line 1, in <module> > File "build/X86/python/m5/main.py", line 457, in main > exec(filecode, scope) > File "configs/example/se.py", line 287, in <module> > Simulation.run(options, root, system, FutureClass) > File "/home/Hessa/gem5_20/gem5-20.1.0.0/configs/common/Simulation.py", line 622, in run > m5.instantiate(checkpoint_dir) > File "build/X86/python/m5/simulate.py", line 119, in instantiate > for obj in root.descendants(): obj.connectPorts() > File "build/X86/python/m5/SimObject.py", line 1701, in connectPorts > portRef.ccConnect() > File "build/X86/python/m5/params.py", line 1984, in ccConnect > peer_port = peer.simobj.getPort(peer.name, peer.index) > File "build/X86/python/m5/SimObject.py", line 1002, in cxx_call > ccobj = self.getCCClass() if static else self.getCCObject() > File "build/X86/python/m5/SimObject.py", line 1662, in getCCObject > raise RuntimeError("Attempt to instantiate orphan node") > RuntimeError: Attempt to instantiate orphan node > > Does anyone know what might be the cause and the solution for this? > Thank you, > Hessa. > -------------- next part -------------- > A message part incompatible with plain text digests has been removed ... > Name: not available > Type: text/html > Size: 3868 bytes > Desc: not available > > ------------------------------ > > Subject: Digest Footer > > _______________________________________________ > 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 > > ------------------------------ > > End of gem5-users Digest, Vol 179, Issue 42 > ******************************************* </module></module></string></db8p191mb05557016c4e644b7b7e1942881...@db8p191mb0555.eur></gem5-users@gem5.org></hissa.alsha...@hotmail.com></cafehpuehp+q7amp6dv-l9tqnt4wmcl0vln+qtu55vmwxxru...@mail.gmail.com></hissa.alsha...@hotmail.com></gem5-users@gem5.org></ja...@lowepower.com></full></linux></bbl.bin></bbl></bbl></gem5-users@gem5.org></template-parameter-1-1></cafehpuf2zc9bnfzyqv8ecxclt03uriubbystzbczlr78dcy...@mail.gmail.com></ntampourat...@ece.auth.gr></gem5-users@gem5.org></ja...@lowepower.com></full></linux></bbl.bin></bbl></bbl></gem5-users@gem5.org></template-parameter-1-1></gem5-users@gem5.org></ntampourat...@ece.auth.gr></full></linux></bbl.bin></bbl></bbl></fe300b8909964d5b8f621e8362207...@heig-vd.ch></marina.zapa...@heig-vd.ch></gregory.med...@heig-vd.ch></gem5-users@gem5.org></david.tr...@heig-vd.ch></db8p191mb055503c7749ad3d535748b3281...@db8p191mb0555.eur></gem5-users@gem5.org></hissa.alsha...@hotmail.com> _______________________________________________ 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
