On Wednesday 15 July 2015 15:13:05, Alison Wang wrote:
This patch addresses a problem mentioned recently on this mailing list: [1].
In that posting a LS1021 based system was locking up at about 5 minutes after boot, but the problem was mysteriously related to the toolchain used for building u-boot. Debugging the problem reveals a stuck interrupt 29 on the GIC.
It appears Freescale's LS1021 support in u-boot erroneously sets the 64-bit ARM generic PL1 physical time CompareValue register to all-ones with a 32-bit value. This causes the timer compare to fire 344 seconds after u-boot configures it. Depending on how fast u-boot gets the kernel booted, this amounts to about 5-minutes of Linux uptime before locking up.
Apparently the bug is masked by some toolchains. Perhaps this is explained by default compiler options, word sizes, or binutils versions. At any rate this patch makes the manipulation explicitly 64-bit which alleviates the issue.
initcall_run_list is the function "hiding" or not "hiding" this problem when calling timer_init. It is using r4 and r5 for it's loop variables. On gcc-4.8 and gcc-4.9 the usage of those two registers are switched. So a newer gcc uses a slightly different register allocation. While this function is perfectly fine, depending on the r4 register timer_init uses a different value for the upper 32-bit of CNTP_CVAL resulting in the different behavior. I've compared two u-boots objdumps showing and not showing this problem which _only_ differ in those 2 register usages.
Best regards, Alexander