Re: port-powerpc/57098: KASSERT firing on macppc: rw lock not held in trap?

[Martin Husemann <martin%duskware.de@localhost>]

The following reply was made to PR port-powerpc/57098; it has been noted by GNATS.

From: Martin Husemann <martin%duskware.de@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: port-powerpc/57098: KASSERT firing on macppc: rw lock not held
 in trap?
Date: Tue, 6 Dec 2022 17:55:45 +0100

 Additional info:
 
 (gdb) list *(uvm_fault_internal+0x1408)
 0x6ec64c is in uvm_fault_internal (../../../../uvm/uvm_fault.c:2047).
 2042    static void
 2043    uvm_fault_lower_neighbor(
 2044            struct uvm_faultinfo *ufi, const struct uvm_faultctx *flt,
 2045            vaddr_t currva, struct vm_page *pg)
 2046    {
 2047            const bool readonly = uvm_pagereadonly_p(pg) || pg->loan_count > 0;
 2048            UVMHIST_FUNC(__func__); UVMHIST_CALLED(maphist);
 2049    
 2050            /* locked: maps(read), amap(if there), uobj */
 2051    
 (gdb) list *(trap+0x164)
 0x138ad4 is in trap (../../../../arch/powerpc/powerpc/trap.c:325).
 320                             break;
 321                     }
 322     
 323                     ftype = VM_PROT_EXECUTE;
 324                     KASSERT(pcb->pcb_onfault == NULL);
 325                     rv = uvm_fault(map, trunc_page(tf->tf_srr0), ftype);
 326                     if (rv == 0) {
 327                             break;
 328                     }
 329                     ci->ci_ev_isi_fatal.ev_count++;
 
 Martin