Hi All,
I'm new to TF-A. While working on enabling Linux on A-520 FVP, I ran into an issue that I wanted to ask about. My stack looks like this:TF-A->U-boot->kernel. Initially I was able to boot on Base FVP(FVP_Base_RevC-2xAEMvA) and get to the linux console. Now we need to enable the same for A-520 FVP as our soc will be A-520. I saw the crash happening very early in "cpu_helpers.S. I made the change in plat/arm/board/fvp/platform.mk by adding lib/cpus/aarch64/cortex_a520.S in one of the FVP_CPU_LIBS and enabling the ERRATA_A520_2938996, ERRATA_A520_2858100 and ERRATA_A520_2630792. These changes helped me get past the TF-A and now i ' m seeing a crash inside kernel in pci.
16.720973] pci-host-generic 40000000.pci: host bridge /pci@40000000 ranges: [ 16.721375] pci-host-generic 40000000.pci: MEM 0x0050000000..0x005fffffff -> 0x0050000000 [ 16.722013] pci-host-generic 40000000.pci: ECAM at [mem 0x40000000-0x4fffffff] for [bus 00-ff] [ 16.723128] pci-host-generic 40000000.pci: PCI host bridge to bus 0000:00 [ 16.723461] pci_bus 0000:00: root bus resource [bus 00-ff] [ 16.723761] pci_bus 0000:00: root bus resource [mem 0x50000000-0x5fffffff] [ 16.724101] Internal error: synchronous external abort: 0000000096000010 [#1] SMP [ 16.724216] Modules linked in: [ 16.724302] CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.15.0-rc7-dirty #12 PREEMPT [ 16.724449] Hardware name: FVP Base RevC (DT) [ 16.724531] pstate: 214000c9 (nzCv daIF +PAN -UAO -TCO +DIT -SSBS BTYPE=--) [ 16.724666] pc : pci_generic_config_read+0x38/0xb8 [ 16.724802] lr : pci_generic_config_read+0x24/0xb8 [ 16.724939] sp : ffff80008272b940 [ 16.725012] x29: ffff80008272b940 x28: 0000000000000000 x27: ffff800081dc00b0 [ 16.725205] x26: ffff800081ec9060 x25: ffff800081ec9078 x24: ffff80008266a9a0 [ 16.725401] x23: 0000000000000000 x22: ffff80008272b9f4 x21: ffff000800fa8000 [ 16.725594] x20: ffff80008272b964 x19: 0000000000000004 x18: 0000000000000006 [ 16.725781] x17: 6666666666663478 x16: 302d303030303030 x15: ffff800082edbbbf [ 16.725980] x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000 [ 16.726166] x11: 0000000000000001 x10: 3ea1f6d484b7e318 x9 : 69631046f78aed23 [ 16.726355] x8 : ffff000800169108 x7 : ffff800082250960 x6 : 00000000000000ff [ 16.726547] x5 : 0000000000000000 x4 : 0000000000000000 x3 : ffff800090000000 [ 16.726738] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff800090000000
Do i need to do enable somethign else in TF-A? I tried adding : pci: pci@40000000 { compatible = "pci-host-ecam-generic"; device_type = "pci"; bus-range = <0x0 0xff>; reg = <0x0 0x40000000 0x0 0x10000000>; ranges = <0x2000000 0x0 0x50000000 0x0 0x50000000 0x0 0x10000000>; }; in plat/arm/board/fvp/fdts/fvp_soc_fw_config.dts but it does not help.
Also I see the kernel is not able to bring all the cpu's online: [ 0.040364] smp: Bringing up secondary CPUs ... [ 5.116993] CPU1: failed to come online [ 5.117063] CPU1: failed in unknown state : 0x0 [ 10.266076] CPU2: failed to come online [ 10.266152] CPU2: failed in unknown state : 0x0 [ 15.415178] CPU3: failed to come online [ 15.415247] CPU3: failed in unknown state : 0x0 [ 15.421021] psci: failed to boot CPU4 (-22) [ 15.421099] CPU4: failed to boot: -22 [ 15.425045] psci: failed to boot CPU5 (-22) [ 15.425123] CPU5: failed to boot: -22 [ 15.429079] psci: failed to boot CPU6 (-22) [ 15.429157] CPU6: failed to boot: -22 [ 15.433153] psci: failed to boot CPU7 (-22) [ 15.433227] CPU7: failed to boot: -22 [ 15.433587] smp: Brought up 1 node, 1 CPU [ 15.433672] SMP: Total of 1 processors activated. [ 15.433746] CPU: All CPU(s) started at EL2
I’m still getting familiar with the codebase and community, so apologies if this has already been addressed. I’d really appreciate any guidance or pointers, and if this is a known issue or good for a first-time contributor, I’d be happy to help.
Regards, Shaunak
Hi Shaunak Looks like the PCI space is not correctly being simulated by FVP_Base_Cortex-A520 or the PCI space is not present at the location mentioned by DT. Try removing all the PCI related nodes in Linux DT and Linux will not try to access the PCI space.
I am not sure about the secondary CPUs not booting. If you are using the debug build of TF-A , see if there is any Error output on the other UART consoles.
Best Regards Soby Mathew
From: shaunak saha via TF-A tf-a@lists.trustedfirmware.org Sent: Tuesday, June 3, 2025 5:48 PM To: tf-a@lists.trustedfirmware.org Subject: [TF-A] FVP_Base_Cortex-A520 crash
Hi All,
I'm new to TF-A. While working on enabling Linux on A-520 FVP, I ran into an issue that I wanted to ask about. My stack looks like this:TF-A->U-boot->kernel. Initially I was able to boot on Base FVP(FVP_Base_RevC-2xAEMvA) and get to the linux console. Now we need to enable the same for A-520 FVP as our soc will be A-520. I saw the crash happening very early in "cpu_helpers.S. I made the change in plat/arm/board/fvp/platform.mkhttp://platform.mk by adding lib/cpus/aarch64/cortex_a520.S in one of the FVP_CPU_LIBS and enabling the ERRATA_A520_2938996, ERRATA_A520_2858100 and ERRATA_A520_2630792. These changes helped me get past the TF-A and now i ' m seeing a crash inside kernel in pci.
16.720973] pci-host-generic 40000000.pci: host bridge /pci@40000000 ranges: [ 16.721375] pci-host-generic 40000000.pci: MEM 0x0050000000..0x005fffffff -> 0x0050000000 [ 16.722013] pci-host-generic 40000000.pci: ECAM at [mem 0x40000000-0x4fffffff] for [bus 00-ff] [ 16.723128] pci-host-generic 40000000.pci: PCI host bridge to bus 0000:00 [ 16.723461] pci_bus 0000:00: root bus resource [bus 00-ff] [ 16.723761] pci_bus 0000:00: root bus resource [mem 0x50000000-0x5fffffff] [ 16.724101] Internal error: synchronous external abort: 0000000096000010 [#1] SMP [ 16.724216] Modules linked in: [ 16.724302] CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.15.0-rc7-dirty #12 PREEMPT [ 16.724449] Hardware name: FVP Base RevC (DT) [ 16.724531] pstate: 214000c9 (nzCv daIF +PAN -UAO -TCO +DIT -SSBS BTYPE=--) [ 16.724666] pc : pci_generic_config_read+0x38/0xb8 [ 16.724802] lr : pci_generic_config_read+0x24/0xb8 [ 16.724939] sp : ffff80008272b940 [ 16.725012] x29: ffff80008272b940 x28: 0000000000000000 x27: ffff800081dc00b0 [ 16.725205] x26: ffff800081ec9060 x25: ffff800081ec9078 x24: ffff80008266a9a0 [ 16.725401] x23: 0000000000000000 x22: ffff80008272b9f4 x21: ffff000800fa8000 [ 16.725594] x20: ffff80008272b964 x19: 0000000000000004 x18: 0000000000000006 [ 16.725781] x17: 6666666666663478 x16: 302d303030303030 x15: ffff800082edbbbf [ 16.725980] x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000 [ 16.726166] x11: 0000000000000001 x10: 3ea1f6d484b7e318 x9 : 69631046f78aed23 [ 16.726355] x8 : ffff000800169108 x7 : ffff800082250960 x6 : 00000000000000ff [ 16.726547] x5 : 0000000000000000 x4 : 0000000000000000 x3 : ffff800090000000 [ 16.726738] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff800090000000
Do i need to do enable somethign else in TF-A? I tried adding : pci: pci@40000000 { compatible = "pci-host-ecam-generic"; device_type = "pci"; bus-range = <0x0 0xff>; reg = <0x0 0x40000000 0x0 0x10000000>; ranges = <0x2000000 0x0 0x50000000 0x0 0x50000000 0x0 0x10000000>; }; in plat/arm/board/fvp/fdts/fvp_soc_fw_config.dts but it does not help.
Also I see the kernel is not able to bring all the cpu's online: [ 0.040364] smp: Bringing up secondary CPUs ... [ 5.116993] CPU1: failed to come online [ 5.117063] CPU1: failed in unknown state : 0x0 [ 10.266076] CPU2: failed to come online [ 10.266152] CPU2: failed in unknown state : 0x0 [ 15.415178] CPU3: failed to come online [ 15.415247] CPU3: failed in unknown state : 0x0 [ 15.421021] psci: failed to boot CPU4 (-22) [ 15.421099] CPU4: failed to boot: -22 [ 15.425045] psci: failed to boot CPU5 (-22) [ 15.425123] CPU5: failed to boot: -22 [ 15.429079] psci: failed to boot CPU6 (-22) [ 15.429157] CPU6: failed to boot: -22 [ 15.433153] psci: failed to boot CPU7 (-22) [ 15.433227] CPU7: failed to boot: -22 [ 15.433587] smp: Brought up 1 node, 1 CPU [ 15.433672] SMP: Total of 1 processors activated. [ 15.433746] CPU: All CPU(s) started at EL2
I’m still getting familiar with the codebase and community, so apologies if this has already been addressed. I’d really appreciate any guidance or pointers, and if this is a known issue or good for a first-time contributor, I’d be happy to help.
Regards, Shaunak
Hi Soby,
Thanks a lot for your reply. Yes, I was able to boot now disabling the pci and smmu in the kernel dts file. Will update here if i find a solution for the PCI part.
If you are using the debug build of TF-A , see if there is any Error
output on the other UART consoles. I am using debug TF-A. I'm using the bl1.bin and fip.bin file from the TF-A build. I didn't see any specific logs. Will check again.
Regards, Shaunak
On Tue, Jun 3, 2025 at 9:50 PM Soby Mathew via TF-A < tf-a@lists.trustedfirmware.org> wrote:
Hi Shaunak
Looks like the PCI space is not correctly being simulated by FVP_Base_Cortex-A520 or the PCI space is not present at the location mentioned by DT. Try removing all the PCI related nodes in Linux DT and Linux will not try to access the PCI space.
I am not sure about the secondary CPUs not booting. If you are using the debug build of TF-A , see if there is any Error output on the other UART consoles.
Best Regards
Soby Mathew
*From:* shaunak saha via TF-A tf-a@lists.trustedfirmware.org *Sent:* Tuesday, June 3, 2025 5:48 PM *To:* tf-a@lists.trustedfirmware.org *Subject:* [TF-A] FVP_Base_Cortex-A520 crash
Hi All,
I'm new to TF-A. While working on enabling Linux on A-520 FVP, I ran into an issue that I wanted to ask about. My stack looks like this:TF-A->U-boot->kernel. Initially I was able to boot on Base FVP(FVP_Base_RevC-2xAEMvA) and get to the linux console. Now we need to enable the same for A-520 FVP as our soc will be A-520. I saw the crash happening very early in "cpu_helpers.S. I made the change in plat/arm/board/fvp/platform.mk by adding lib/cpus/aarch64/cortex_a520.S in one of the FVP_CPU_LIBS and enabling the ERRATA_A520_2938996, ERRATA_A520_2858100 and ERRATA_A520_2630792. These changes helped me get past the TF-A and now i ' m seeing a crash inside kernel in pci.
16.720973] pci-host-generic 40000000.pci: host bridge /pci@40000000 ranges: [ 16.721375] pci-host-generic 40000000.pci: MEM 0x0050000000..0x005fffffff -> 0x0050000000 [ 16.722013] pci-host-generic 40000000.pci: ECAM at [mem 0x40000000-0x4fffffff] for [bus 00-ff] [ 16.723128] pci-host-generic 40000000.pci: PCI host bridge to bus 0000:00 [ 16.723461] pci_bus 0000:00: root bus resource [bus 00-ff] [ 16.723761] pci_bus 0000:00: root bus resource [mem 0x50000000-0x5fffffff] [ 16.724101] Internal error: synchronous external abort: 0000000096000010 [#1] SMP [ 16.724216] Modules linked in: [ 16.724302] CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.15.0-rc7-dirty #12 PREEMPT [ 16.724449] Hardware name: FVP Base RevC (DT) [ 16.724531] pstate: 214000c9 (nzCv daIF +PAN -UAO -TCO +DIT -SSBS BTYPE=--) [ 16.724666] pc : pci_generic_config_read+0x38/0xb8 [ 16.724802] lr : pci_generic_config_read+0x24/0xb8 [ 16.724939] sp : ffff80008272b940 [ 16.725012] x29: ffff80008272b940 x28: 0000000000000000 x27: ffff800081dc00b0 [ 16.725205] x26: ffff800081ec9060 x25: ffff800081ec9078 x24: ffff80008266a9a0 [ 16.725401] x23: 0000000000000000 x22: ffff80008272b9f4 x21: ffff000800fa8000 [ 16.725594] x20: ffff80008272b964 x19: 0000000000000004 x18: 0000000000000006 [ 16.725781] x17: 6666666666663478 x16: 302d303030303030 x15: ffff800082edbbbf [ 16.725980] x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000 [ 16.726166] x11: 0000000000000001 x10: 3ea1f6d484b7e318 x9 : 69631046f78aed23 [ 16.726355] x8 : ffff000800169108 x7 : ffff800082250960 x6 : 00000000000000ff [ 16.726547] x5 : 0000000000000000 x4 : 0000000000000000 x3 : ffff800090000000 [ 16.726738] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff800090000000
Do i need to do enable somethign else in TF-A? I tried adding : pci: pci@40000000 { compatible = "pci-host-ecam-generic"; device_type = "pci"; bus-range = <0x0 0xff>; reg = <0x0 0x40000000 0x0 0x10000000>; ranges = <0x2000000 0x0 0x50000000 0x0 0x50000000 0x0 0x10000000>; }; in plat/arm/board/fvp/fdts/fvp_soc_fw_config.dts but it does not help.
Also I see the kernel is not able to bring all the cpu's online: [ 0.040364] smp: Bringing up secondary CPUs ... [ 5.116993] CPU1: failed to come online [ 5.117063] CPU1: failed in unknown state : 0x0 [ 10.266076] CPU2: failed to come online [ 10.266152] CPU2: failed in unknown state : 0x0 [ 15.415178] CPU3: failed to come online [ 15.415247] CPU3: failed in unknown state : 0x0 [ 15.421021] psci: failed to boot CPU4 (-22) [ 15.421099] CPU4: failed to boot: -22 [ 15.425045] psci: failed to boot CPU5 (-22) [ 15.425123] CPU5: failed to boot: -22 [ 15.429079] psci: failed to boot CPU6 (-22) [ 15.429157] CPU6: failed to boot: -22 [ 15.433153] psci: failed to boot CPU7 (-22) [ 15.433227] CPU7: failed to boot: -22 [ 15.433587] smp: Brought up 1 node, 1 CPU [ 15.433672] SMP: Total of 1 processors activated. [ 15.433746] CPU: All CPU(s) started at EL2
I’m still getting familiar with the codebase and community, so apologies if this has already been addressed. I’d really appreciate any guidance or pointers, and if this is a known issue or good for a first-time contributor, I’d be happy to help.
Regards, Shaunak -- TF-A mailing list -- tf-a@lists.trustedfirmware.org To unsubscribe send an email to tf-a-leave@lists.trustedfirmware.org
tf-a@lists.trustedfirmware.org
-
shaunak saha -
Soby Mathew