Hi,
[please try to avoid top posting]
On Wed, 5 Apr 2023 at 19:20, deffo@gmx.de wrote:
Hi Simon,
even with CONFIG_BOOTSTAGE disabled I still get this data abort:
Starting kernel ...
data abort pc : [<fdf3b91a>] lr : [<00000000>] reloc pc : [<c010191a>] lr : [<c21c6000>] sp : fbbefafc ip : 00000000 fp : 00000001 r10: c20000d8 r9 : fbbffec0 r8 : 00000000 r7 : 00000000 r6 : 00000000 r5 : e28ff010 r4 : 00000000 r3 : 00000000 r2 : 00000a10 r1 : fdfab018 r0 : 2ffc0020 Flags: nzCv IRQs off FIQs off Mode SVC_32 (T) Code: f891 f03c f891 f05c (f891) f07c Resetting CPU ...
resetting ...
How can I debug this any further?
OK so it is not related to bootstage. Perhaps Linux is crashing due to something being in the wrong place?
I suggest copying some STM32 people here. You might be able to add a 'console=' or 'earlycon=' line to get a log from Linux.
Regards, Simon
Best regards
============================================================================================
Hi,
On Mon, 3 Apr 2023 at 00:43, deffo@gmx.de wrote:
Hi Simon,
it's an STM32MP1, here's a bdinfo and the full boot log. Btw, OPTEE OS is active:
STM32MP> bdinfo boot_params = 0x00000000 DRAM bank = 0x00000000 -> start = 0xc0000000 -> size = 0x40000000 flashstart = 0x00000000 flashsize = 0x00000000 flashoffset = 0x00000000 baudrate = 115200 bps relocaddr = 0xfdf3a000 reloc off = 0x3de3a000 Build = 32-bit current eth = ethernet@5800a000 ethaddr = 00:00:00:00:00:00 IP addr = <NULL> fdt_blob = 0xfbf27e10 new_fdt = 0xfbf27e10 fdt_size = 0x000100a0 lmb_dump_all: memory.cnt = 0x1 memory[0] [0xc0000000-0xffffffff], 0x40000000 bytes flags: 0 reserved.cnt = 0x5 reserved[0] [0x10000000-0x10047fff], 0x00048000 bytes flags: 4 reserved[1] [0x30000000-0x3003ffff], 0x00040000 bytes flags: 4 reserved[2] [0x38000000-0x3800ffff], 0x00010000 bytes flags: 4 reserved[3] [0xfbf23940-0xfdffffff], 0x020dc6c0 bytes flags: 0 reserved[4] [0xfe000000-0xffffffff], 0x02000000 bytes flags: 4 devicetree = board arch_number = 0x00000000 TLB addr = 0xfdff0000 irq_sp = 0xfbf27b80 sp start = 0xfbf27b70 Early malloc usage: 11f8 / 3000 STM32MP> ext4load mmc ${boot_instance}:${boot_part} ${kernel_addr_r} ${linux}; bootm ${kernel_addr_r} 10590706 bytes read in 542 ms (18.6 MiB/s) ## Loading kernel from FIT Image at c2000000 ... Using 'config' configuration Trying 'kernel' kernel subimage Description: Linux kernel - base Created: 2023-03-31 12:45:20 UTC Type: Kernel Image Compression: uncompressed Data Start: 0xc20000d8 Data Size: 7527552 Bytes = 7.2 MiB Architecture: ARM OS: Linux Load Address: 0xc0008000 Entry Point: 0xc0008000 Hash algo: sha256 Hash value: b0ca7e8de523721697b6990e67f142159845f1b5e2a52a4fef4da8f6754d49a7 Sign algo: sha256,rsa2048:board Sign value: unavailable Timestamp: unavailable Verifying Hash Integrity ... sha256+ OK kernel data at 0xc20000d8, len = 0x0072dc80 (7527552) ## Loading ramdisk from FIT Image at c2000000 ... Using 'config' configuration Trying 'ramdisk' ramdisk subimage Description: ramdisk Created: 2023-03-31 12:45:20 UTC Type: RAMDisk Image Compression: uncompressed Data Start: 0xc273d318 Data Size: 2998533 Bytes = 2.9 MiB Architecture: ARM OS: Linux Load Address: unavailable Entry Point: unavailable Hash algo: sha256 Hash value: 3d9201eee8e91a161719ea35b630d9d09dc6be204fcda3390ecd7a540e322048 Sign algo: sha256,rsa2048:board Sign value: unavailable Timestamp: unavailable Verifying Hash Integrity ... sha256+ OK ## Loading fdt from FIT Image at c2000000 ... Bootstage space exhasuted Using 'config' configuration Trying 'dtb' fdt subimage Description: DeviceTree blob - base Created: 2023-03-31 12:45:20 UTC Type: Flat Device Tree Compression: uncompressed Data Start: 0xc272de94 Data Size: 62326 Bytes = 60.9 KiB Architecture: ARM Load Address: 0xc4000000 Hash algo: sha256 Hash value: 2b1df931402fd642c8a1618aa140630e22fed4c095deed4df201637f008630ea Sign algo: sha256,rsa2048:board Sign value: unavailable Timestamp: unavailable Verifying Hash Integrity ... sha256+ OK Bootstage space exhasuted Bootstage space exhasuted Bootstage space exhasuted Bootstage space exhasuted Loading fdt from 0xc272de94 to 0xc4000000 Bootstage space exhasuted Booting using the fdt blob at 0xc4000000 Loading Kernel Image kernel loaded at 0xc0008000, end = 0xc0735c80 Bootstage space exhasuted Bootstage space exhasuted Loading Ramdisk to cfd23000, end cffff105 ... OK Loading Device Tree to cfd10000, end cfd22375 ... OK Bootstage space exhasuted Bootstage space exhasuted
Starting kernel ...
data abort pc : [<fdf3b8ba>] lr : [<00000000>] reloc pc : [<c01018ba>] lr : [<c21c6000>] sp : fbf2786c ip : 00000000 fp : 00000001 r10: c20000d8 r9 : fbf37eb0 r8 : 00000000 r7 : 00000000 r6 : 00000000 r5 : e28ff010 r4 : 00000000 r3 : 00000000 r2 : 00000a10 r1 : fdfabb14 r0 : 2ffc0020 Flags: nzCv IRQs off FIQs off Mode SVC_32 (T) Code: f891 f03c f891 f05c (f891) f07c Resetting CPU ...
resetting ...
Thanks and best regards
======================================
Hi,
On Sat, 1 Apr 2023 at 02:57, deffo@gmx.de wrote:
Hi,
I changed from v2020.10 to v2022.10 and suddenly I get a bunch of "Bootstage space exhasuted" messages during bootm.
reloc pc : [<c01018ba>] is memcpy()
It evens triggers a reset. Older version works just fine:
Sign value: unavailable Timestamp: unavailableVerifying Hash Integrity ... sha256+ OK Bootstage space exhasuted Bootstage space exhasuted Bootstage space exhasuted Bootstage space exhasuted Loading fdt from 0xc272de94 to 0xc4000000 Bootstage space exhasuted Booting using the fdt blob at 0xc4000000 Loading Kernel Image kernel loaded at 0xc0008000, end = 0xc0735c80 Bootstage space exhasuted Bootstage space exhasuted Loading Ramdisk to cfd23000, end cffff105 ... OK Loading Device Tree to cfd10000, end cfd22375 ... OK Bootstage space exhasuted Bootstage space exhasuted
Starting kernel ...
data abort pc : [<fdf3b8ba>] lr : [<00000000>] reloc pc : [<c01018ba>] lr : [<c21c6000>] sp : fbf2786c ip : 00000000 fp : 00000001 r10: c20000d8 r9 : fbf37eb0 r8 : 00000000 r7 : 00000000 r6 : 00000000 r5 : e28ff010 r4 : 00000000 r3 : 00000000 r2 : 00000a10 r1 : fdfabb14 r0 : 2ffc0020 Flags: nzCv IRQs off FIQs off Mode SVC_32 (T) Code: f891 f03c f891 f05c (f891) f07c Resetting CPU ...
resetting ...
Do you know which board it is? It would help to produce a full console trace.
If you disabled CONFIG_BOOTSTAGE does the problem go away? Running out of space should not cause any problems. Also, try disabling CONFIG_BOOTSTAGE and see if that makes a difference.
You can enable CONFIG_BOOTSTAGE_REPORT to see a report before the kernel boots.
Regards, Simon