/* $NetBSD: vm_machdep.c,v 1.15 2023/12/20 06:13:58 thorpej Exp $ */ /*- * Copyright (c) 2014 The NetBSD Foundation, Inc. * All rights reserved. * * This code is derived from software contributed to The NetBSD Foundation * by Matt Thomas of 3am Software Foundry. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE * POSSIBILITY OF SUCH DAMAGE. */ #include "opt_compat_netbsd32.h" #include "opt_ddb.h" #include __KERNEL_RCSID(0, "$NetBSD: vm_machdep.c,v 1.15 2023/12/20 06:13:58 thorpej Exp $"); #include #include #include #include #include #include #include #include #include #include #include #include #include /* * Special compilation symbols: * * STACKCHECKS - Fill undefined and supervisor stacks with a known pattern * on forking and check the pattern on exit, reporting * the amount of stack used. */ void cpu_proc_fork(struct proc *p1, struct proc *p2) { } /* * Finish a fork operation, with LWP l2 nearly set up. * * Copy and update the pcb and trapframe, making the child ready to run. * * Rig the child's kernel stack so that it will start out in * lwp_trampoline() which will call the specified func with the argument arg. * * If an alternate user-level stack is requested (with non-zero values * in both the stack and stacksize args), set up the user stack pointer * accordingly. */ void cpu_lwp_fork(struct lwp *l1, struct lwp *l2, void *stack, size_t stacksize, void (*func)(void *), void *arg) { const struct pcb * const pcb1 = lwp_getpcb(l1); struct pcb * const pcb2 = lwp_getpcb(l2); #if 0 printf("cpu_lwp_fork: lwp1=%p, lwp2=%p, curlwp=%p, lwp0=%p\n", l1, l2, curlwp, &lwp0); #endif /* Copy the pcb */ *pcb2 = *pcb1; /* * Disable FP for a newly created LWP but remember if the * FP state is valid. */ l2->l_md.md_cpacr = CPACR_FPEN_NONE; KASSERT(l2->l_md.md_astpending == 0); #ifdef ARMV83_PAC /* * Temporary kern PAC key, not really strong, but better than zero. * The real key gets generated by the new thread in lwp_trampoline(). * * The reason we defer the generation, is because cpu_lwp_fork() can * be called early at boot time, when the CPRNG subsystem hasn't yet * been initialized. */ l2->l_md.md_ia_kern[0] = (uintptr_t)l2; l2->l_md.md_ia_kern[1] = (uintptr_t)stack; /* inherit user PAC key */ memcpy(l2->l_md.md_ia_user, l1->l_md.md_ia_user, sizeof(l2->l_md.md_ia_user)); memcpy(l2->l_md.md_ib_user, l1->l_md.md_ib_user, sizeof(l2->l_md.md_ib_user)); memcpy(l2->l_md.md_da_user, l1->l_md.md_da_user, sizeof(l2->l_md.md_da_user)); memcpy(l2->l_md.md_db_user, l1->l_md.md_db_user, sizeof(l2->l_md.md_db_user)); memcpy(l2->l_md.md_ga_user, l1->l_md.md_ga_user, sizeof(l2->l_md.md_ga_user)); #endif /* * Set up the kernel stack for the process. * Note: this stack is not in use if we are forking from p1 */ vaddr_t uv = uvm_lwp_getuarea(l2); #ifdef STACKCHECKS #define PCB_END(l) ((char *)lwp_getpcb((l)) + sizeof(struct pcb)) #define UAREA_END(l) ((char *)uvm_lwp_getuarea((l)) + USPACE) /* fill 0xdd for STACKCHECKS */ memset(PCB_END(l2), 0xdd, UAREA_END(l2) - PCB_END(l2)); printf("lwp %p: pcb=%p, stack=%p-%p\n", l2, lwp_getpcb(l2), PCB_END(l2), UAREA_END(l2)); #endif struct trapframe * const utf = (struct trapframe *)(uv + USPACE) - 1; l2->l_md.md_utf = utf; *utf = *l1->l_md.md_utf; /* * If specified, give the child a different stack (make sure it's * 16- or 8-byte aligned for 64- or 32-bit processes, respectively). */ if (stack != NULL) { utf->tf_sp = (vaddr_t)(stack) + stacksize; #ifdef COMPAT_NETBSD32 if (__predict_false(l2->l_proc->p_flag & PK_32)) { utf->tf_sp &= -8; utf->tf_reg[13] = utf->tf_sp; } else #endif utf->tf_sp &= -16; } /* build a new switchframe */ struct trapframe * const ktf = utf - 1; ktf->tf_reg[27] = (uint64_t)func; ktf->tf_reg[28] = (uint64_t)arg; ktf->tf_lr = (uintptr_t)lwp_trampoline; #ifdef DDB ktf->tf_reg[29] = (uint64_t)utf; ktf->tf_pc = (uint64_t)&&backtrace_here; ktf->tf_sp = 0; /* mark as switchframe */ backtrace_here: #endif pcb2->pcb_tf = ktf; } /* * cpu_exit is called as the last action during exit. * * We clean up a little and then call switch_exit() with the old proc as an * argument. switch_exit() first switches to lwp0's context, and finally * jumps into switch() to wait for another process to wake up. */ void cpu_lwp_free(struct lwp *l, int proc) { #ifdef STACKCHECKS /* Report how much stack has been used - debugging */ u_char *stop, *sbottom, *ptr; u_int cnt; stop = PCB_END(l); sbottom = UAREA_END(l); for (cnt = 0, ptr = stop; *ptr == 0xdd && ptr <= sbottom; cnt++, ptr++) ; log(LOG_INFO, "lwp %p: %u/%ld bytes are used for EL1 stack\n", l, cnt, sbottom - stop); #endif } void cpu_lwp_free2(struct lwp *l) { } /* * Map a user I/O request into kernel virtual address space. * Note: the pages are already locked by uvm_vslock(), so we * do not need to pass an access_type to pmap_enter(). */ int vmapbuf(struct buf *bp, vsize_t len) { vaddr_t faddr, taddr, off; paddr_t fpa; if ((bp->b_flags & B_PHYS) == 0) panic("vmapbuf"); bp->b_saveaddr = bp->b_data; faddr = trunc_page((vaddr_t)bp->b_data); off = (vaddr_t)bp->b_data - faddr; len = round_page(off + len); taddr = uvm_km_alloc(phys_map, len, atop(faddr) & uvmexp.colormask, UVM_KMF_VAONLY | UVM_KMF_WAITVA | UVM_KMF_COLORMATCH); bp->b_data = (void *)(taddr + off); /* * The region is locked, so we expect that pmap_extract() will return * true. */ while (len) { (void)pmap_extract(vm_map_pmap(&bp->b_proc->p_vmspace->vm_map), faddr, &fpa); pmap_enter(pmap_kernel(), taddr, fpa, VM_PROT_READ | VM_PROT_WRITE, VM_PROT_READ | VM_PROT_WRITE | PMAP_WIRED); faddr += PAGE_SIZE; taddr += PAGE_SIZE; len -= PAGE_SIZE; } pmap_update(pmap_kernel()); return 0; } /* * Unmap a previously-mapped user I/O request. */ void vunmapbuf(struct buf *bp, vsize_t len) { vaddr_t addr, off; if ((bp->b_flags & B_PHYS) == 0) panic("vunmapbuf"); /* * Make sure the cache does not have dirty data for the * pages we had mapped. */ addr = trunc_page((vaddr_t)bp->b_data); off = (vaddr_t)bp->b_data - addr; len = round_page(off + len); pmap_remove(pmap_kernel(), addr, addr + len); pmap_update(pmap_kernel()); uvm_km_free(phys_map, addr, len, UVM_KMF_VAONLY); bp->b_data = bp->b_saveaddr; bp->b_saveaddr = 0; }