Re: [PATCH v3] RISC-V: Check clint_time_val before use

2020-09-29 Thread Palmer Dabbelt
On Sat, 26 Sep 2020 22:52:19 PDT (-0700), Damien Le Moal wrote: On Sun, 2020-09-27 at 11:09 +0530, Anup Patel wrote: The NoMMU kernel is broken for QEMU virt machine from Linux-5.9-rc6 because clint_time_val is used even before CLINT driver is probed at following places: 1. rand_initialize() cal

Re: [PATCH v3] RISC-V: Check clint_time_val before use

2020-09-26 Thread Damien Le Moal
On Sun, 2020-09-27 at 11:09 +0530, Anup Patel wrote: > The NoMMU kernel is broken for QEMU virt machine from Linux-5.9-rc6 > because clint_time_val is used even before CLINT driver is probed > at following places: > 1. rand_initialize() calls get_cycles() which in-turn uses >clint_time_val > 2.

Re: [PATCH v3] RISC-V: Check clint_time_val before use

2020-09-26 Thread Damien Le Moal
On Sat, 2020-09-26 at 22:46 -0700, Palmer Dabbelt wrote: > On Sat, 26 Sep 2020 22:39:16 PDT (-0700), Anup Patel wrote: > > The NoMMU kernel is broken for QEMU virt machine from Linux-5.9-rc6 > > because clint_time_val is used even before CLINT driver is probed > > at following places: > > 1. rand_i

Re: [PATCH v3] RISC-V: Check clint_time_val before use

2020-09-26 Thread Palmer Dabbelt
On Sat, 26 Sep 2020 22:39:16 PDT (-0700), Anup Patel wrote: The NoMMU kernel is broken for QEMU virt machine from Linux-5.9-rc6 because clint_time_val is used even before CLINT driver is probed at following places: 1. rand_initialize() calls get_cycles() which in-turn uses clint_time_val 2. bo

[PATCH v3] RISC-V: Check clint_time_val before use

2020-09-26 Thread Anup Patel
The NoMMU kernel is broken for QEMU virt machine from Linux-5.9-rc6 because clint_time_val is used even before CLINT driver is probed at following places: 1. rand_initialize() calls get_cycles() which in-turn uses clint_time_val 2. boot_init_stack_canary() calls get_cycles() which in-turn use