Re: [PATCH V3 1/2] MIPS: Loongson-3: Enable COP2 usage in kernel

[Thomas Bogendoerfer <tsbogend@alpha.franken.de>]

On Tue, Aug 11, 2020 at 02:45:05PM +0800, Huacai Chen wrote:
> Hi, Thomas and Jiaxun,
> 
> On Tue, Aug 11, 2020 at 10:18 AM Jiaxun Yang <jiaxun.yang@flygoat.com> wrote:
> >
> >
> >
> > 在 2020/8/10 22:12, Thomas Bogendoerfer 写道:
> > > On Sun, Aug 09, 2020 at 10:53:13PM +0800, Jiaxun Yang wrote:
> > >> Thus we still need to enable CU2 with exception for user space, and we can
> > >> always enable CU2 in
> > >> kernel since kernel won't be compiled with hard-float. :-)
> > > I see, how about the patch below
> > That looks fine for me.
> > Is it good with you, Huacai?
> 
> There are two problems:
> 1, zboot (arch/mips/boot/compressed/head.S) should be considered,
> because the initial value of Status may or may not contain CU2.

this comes with it's own memcpy/memset and stuff, I don't see a reason why
COP2 needs to be enabled there,

> 2, r4k_switch.S should set CU2 for the new process, otherwise it
> cannot use gslq/gssq while it in kernel (Because the new process
> doesn't contain CU2 in THERAD_STATUS.

which is correct for all user space process, otherwise the whole
cop2 exception thing wouldn't work. And if cop2 exception handling
has been run it's set in THREAD_STATUS.

> Though a process sets CU2 when it enters kernel, but it
> only sets CU2 in hardware, not in THREAD_STATUS).

A kernel thread will get THREAD_STATUS from current running kernel code,
at least that's how I read this code:

       if (unlikely(p->flags & PF_KTHREAD)) {
                /* kernel thread */
                unsigned long status = p->thread.cp0_status;
                memset(childregs, 0, sizeof(struct pt_regs));
                ti->addr_limit = KERNEL_DS;
                p->thread.reg16 = usp; /* fn */
                p->thread.reg17 = kthread_arg;
                p->thread.reg29 = childksp;
                p->thread.reg31 = (unsigned long) ret_from_kernel_thread;
#if defined(CONFIG_CPU_R3000) || defined(CONFIG_CPU_TX39XX)
                status = (status & ~(ST0_KUP | ST0_IEP | ST0_IEC)) |
                         ((status & (ST0_KUC | ST0_IEC)) << 2);
#else
                status |= ST0_EXL;
#endif
                childregs->cp0_status = status;
                return 0;
        }

If there is still something missing, I want to find the root cause
and not paper over it in r4k_switch.S and IMHO break COP2 handling for
loongsoon completely.

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]