| Message ID | 20191023195515.13168-4-rcampbell@nvidia.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | HMM tests and minor fixes | expand |
On Wed, Oct 23, 2019 at 12:55:15PM -0700, Ralph Campbell wrote: > Add self tests for HMM. > > Signed-off-by: Ralph Campbell <rcampbell@nvidia.com> You can add my signoff Signed-off-by: Jérôme Glisse <jglisse@redhat.com> > --- > MAINTAINERS | 3 + > drivers/char/Kconfig | 11 + > drivers/char/Makefile | 1 + > drivers/char/hmm_dmirror.c | 1566 ++++++++++++++++++++++++ > include/Kbuild | 1 + > include/uapi/linux/hmm_dmirror.h | 74 ++ > tools/testing/selftests/vm/.gitignore | 1 + > tools/testing/selftests/vm/Makefile | 3 + > tools/testing/selftests/vm/config | 3 + > tools/testing/selftests/vm/hmm-tests.c | 1311 ++++++++++++++++++++ > tools/testing/selftests/vm/run_vmtests | 16 + > tools/testing/selftests/vm/test_hmm.sh | 97 ++ > 12 files changed, 3087 insertions(+) > create mode 100644 drivers/char/hmm_dmirror.c > create mode 100644 include/uapi/linux/hmm_dmirror.h > create mode 100644 tools/testing/selftests/vm/hmm-tests.c > create mode 100755 tools/testing/selftests/vm/test_hmm.sh > > diff --git a/MAINTAINERS b/MAINTAINERS > index 296de2b51c83..9890b6b8eea0 100644 > --- a/MAINTAINERS > +++ b/MAINTAINERS > @@ -7427,8 +7427,11 @@ M: Jérôme Glisse <jglisse@redhat.com> > L: linux-mm@kvack.org > S: Maintained > F: mm/hmm* > +F: drivers/char/hmm* > F: include/linux/hmm* > +F: include/uapi/linux/hmm* > F: Documentation/vm/hmm.rst > +F: tools/testing/selftests/vm/*hmm* > > HOST AP DRIVER > M: Jouni Malinen <j@w1.fi> > diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig > index df0fc997dc3e..cc8ddb99550d 100644 > --- a/drivers/char/Kconfig > +++ b/drivers/char/Kconfig > @@ -535,6 +535,17 @@ config ADI > and SSM (Silicon Secured Memory). Intended consumers of this > driver include crash and makedumpfile. > > +config HMM_DMIRROR > + tristate "HMM driver for testing Heterogeneous Memory Management" > + depends on HMM_MIRROR > + depends on DEVICE_PRIVATE > + help > + This is a pseudo device driver solely for testing HMM. > + Say Y here if you want to build the HMM test driver. > + Doing so will allow you to run tools/testing/selftest/vm/hmm-tests. > + > + If in doubt, say "N". > + > endmenu > > config RANDOM_TRUST_CPU > diff --git a/drivers/char/Makefile b/drivers/char/Makefile > index 7c5ea6f9df14..d4a168c0c138 100644 > --- a/drivers/char/Makefile > +++ b/drivers/char/Makefile > @@ -52,3 +52,4 @@ js-rtc-y = rtc.o > obj-$(CONFIG_XILLYBUS) += xillybus/ > obj-$(CONFIG_POWERNV_OP_PANEL) += powernv-op-panel.o > obj-$(CONFIG_ADI) += adi.o > +obj-$(CONFIG_HMM_DMIRROR) += hmm_dmirror.o > diff --git a/drivers/char/hmm_dmirror.c b/drivers/char/hmm_dmirror.c > new file mode 100644 > index 000000000000..5a1ed34e72e1 > --- /dev/null > +++ b/drivers/char/hmm_dmirror.c > @@ -0,0 +1,1566 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Copyright 2013 Red Hat Inc. > + * > + * This program is free software; you can redistribute it and/or > + * modify it under the terms of the GNU General Public License as > + * published by the Free Software Foundation; either version 2 of > + * the License, or (at your option) any later version. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + * > + * Authors: Jérôme Glisse <jglisse@redhat.com> > + */ > +/* > + * This is a driver to exercice the HMM (heterogeneous memory management) > + * mirror and zone device private memory migration APIs of the kernel. > + * Userspace programs can register with the driver to mirror their own address > + * space and can use the device to read/write any valid virtual address. > + * > + * In some ways it can also serve as an example driver for people wanting to use > + * HMM inside their own device driver. > + */ > +#include <linux/init.h> > +#include <linux/fs.h> > +#include <linux/mm.h> > +#include <linux/module.h> > +#include <linux/kernel.h> > +#include <linux/cdev.h> > +#include <linux/device.h> > +#include <linux/mutex.h> > +#include <linux/rwsem.h> > +#include <linux/sched.h> > +#include <linux/slab.h> > +#include <linux/highmem.h> > +#include <linux/delay.h> > +#include <linux/pagemap.h> > +#include <linux/hmm.h> > +#include <linux/vmalloc.h> > +#include <linux/swap.h> > +#include <linux/swapops.h> > +#include <linux/sched/mm.h> > +#include <linux/platform_device.h> > + > +#include <uapi/linux/hmm_dmirror.h> > + > +#define DMIRROR_NDEVICES 2 > +#define DMIRROR_RANGE_FAULT_TIMEOUT 1000 > +#define DEVMEM_CHUNK_SIZE (256 * 1024 * 1024U) > +#define DEVMEM_CHUNKS_RESERVE 16 > + > +static const struct dev_pagemap_ops dmirror_devmem_ops; > +static dev_t dmirror_dev; > +static struct platform_device *dmirror_platform_devices[DMIRROR_NDEVICES]; > +static struct page *dmirror_zero_page; > + > +struct dmirror_device; > + > +struct dmirror_bounce { > + void *ptr; > + unsigned long size; > + unsigned long addr; > + unsigned long cpages; > +}; > + > +#define DPT_SHIFT PAGE_SHIFT > +#define DPT_VALID (1UL << 0) > +#define DPT_WRITE (1UL << 1) > +#define DPT_DPAGE (1UL << 2) > +#define DPT_ZPAGE 0x20UL > + > +const uint64_t dmirror_hmm_flags[HMM_PFN_FLAG_MAX] = { > + [HMM_PFN_VALID] = DPT_VALID, > + [HMM_PFN_WRITE] = DPT_WRITE, > + [HMM_PFN_DEVICE_PRIVATE] = DPT_DPAGE, > +}; > + > +static const uint64_t dmirror_hmm_values[HMM_PFN_VALUE_MAX] = { > + [HMM_PFN_NONE] = 0, > + [HMM_PFN_ERROR] = 0x10, > + [HMM_PFN_SPECIAL] = 0x10, > +}; > + > +struct dmirror_pt { > + u64 pgd[PTRS_PER_PGD]; > + struct rw_semaphore lock; > +}; > + > +/* > + * Data attached to the open device file. > + * Note that it might be shared after a fork(). > + */ > +struct dmirror { > + struct hmm_mirror mirror; > + struct dmirror_device *mdevice; > + struct dmirror_pt pt; > + struct mutex mutex; > +}; > + > +/* > + * ZONE_DEVICE pages for migration and simulating device memory. > + */ > +struct dmirror_chunk { > + struct dev_pagemap pagemap; > + struct dmirror_device *mdevice; > +}; > + > +/* > + * Per device data. > + */ > +struct dmirror_device { > + struct cdev cdevice; > + struct hmm_devmem *devmem; > + struct platform_device *pdevice; > + > + unsigned int devmem_capacity; > + unsigned int devmem_count; > + struct dmirror_chunk **devmem_chunks; > + struct mutex devmem_lock; /* protects the above */ > + > + unsigned long calloc; > + unsigned long cfree; > + struct page *free_pages; > + spinlock_t lock; /* protects the above */ > +}; > + > +static inline unsigned long dmirror_pt_pgd(unsigned long addr) > +{ > + return (addr >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1); > +} > + > +static inline unsigned long dmirror_pt_pud(unsigned long addr) > +{ > + return (addr >> PUD_SHIFT) & (PTRS_PER_PUD - 1); > +} > + > +static inline unsigned long dmirror_pt_pmd(unsigned long addr) > +{ > + return (addr >> PMD_SHIFT) & (PTRS_PER_PMD - 1); > +} > + > +static inline unsigned long dmirror_pt_pte(unsigned long addr) > +{ > + return (addr >> PAGE_SHIFT) & (PTRS_PER_PTE - 1); > +} > + > +static inline struct page *dmirror_pt_page(u64 *dptep) > +{ > + u64 dpte = *dptep; > + > + if (dpte == DPT_ZPAGE) > + return dmirror_zero_page; > + if (!(dpte & DPT_VALID)) > + return NULL; > + return pfn_to_page((u64)dpte >> DPT_SHIFT); > +} > + > +static inline struct page *dmirror_pt_page_write(u64 *dptep) > +{ > + u64 dpte = *dptep; > + > + if (!(dpte & DPT_VALID) || !(dpte & DPT_WRITE)) > + return NULL; > + return pfn_to_page((u64)dpte >> DPT_SHIFT); > +} > + > +static inline u64 dmirror_pt_from_page(struct page *page) > +{ > + if (!page) > + return 0; > + return (page_to_pfn(page) << DPT_SHIFT) | DPT_VALID; > +} > + > +static struct page *populate_pt(struct dmirror *dmirror, u64 *dptep) > +{ > + struct page *page; > + > + /* > + * Since we don't free page tables until the process exits, > + * we can unlock and relock without the page table being freed > + * from under us. > + */ > + mutex_unlock(&dmirror->mutex); > + page = alloc_page(GFP_HIGHUSER | __GFP_ZERO); > + mutex_lock(&dmirror->mutex); > + if (page) { > + if (unlikely(*dptep)) { > + __free_page(page); > + page = dmirror_pt_page(dptep); > + } else > + *dptep = dmirror_pt_from_page(page); > + } else if (*dptep) > + page = dmirror_pt_page(dptep); > + return page; > +} > + > +static inline unsigned long dmirror_pt_pud_end(unsigned long addr) > +{ > + return (addr & PGDIR_MASK) + ((unsigned long)PTRS_PER_PUD << PUD_SHIFT); > +} > + > +static inline unsigned long dmirror_pt_pmd_end(unsigned long addr) > +{ > + return (addr & PUD_MASK) + ((unsigned long)PTRS_PER_PMD << PMD_SHIFT); > +} > + > +static inline unsigned long dmirror_pt_pte_end(unsigned long addr) > +{ > + return (addr & PMD_MASK) + ((unsigned long)PTRS_PER_PTE << PAGE_SHIFT); > +} > + > +typedef int (*dmirror_walk_cb_t)(struct dmirror *dmirror, > + unsigned long start, > + unsigned long end, > + u64 *dptep, > + void *private); > + > +static int dmirror_pt_walk(struct dmirror *dmirror, > + dmirror_walk_cb_t cb, > + unsigned long start, > + unsigned long end, > + void *private, > + bool populate) > +{ > + u64 *dpgdp = &dmirror->pt.pgd[dmirror_pt_pgd(start)]; > + unsigned long addr; > + int ret = -ENOENT; > + > + for (addr = start; addr < end; dpgdp++) { > + u64 *dpudp; > + unsigned long pud_end; > + struct page *pud_page; > + > + pud_end = min(end, dmirror_pt_pud_end(addr)); > + pud_page = dmirror_pt_page(dpgdp); > + if (!pud_page) { > + if (!populate) { > + addr = pud_end; > + continue; > + } > + pud_page = populate_pt(dmirror, dpgdp); > + if (!pud_page) > + return -ENOMEM; > + } > + dpudp = kmap(pud_page); > + dpudp += dmirror_pt_pud(addr); > + for (; addr != pud_end; dpudp++) { > + u64 *dpmdp; > + unsigned long pmd_end; > + struct page *pmd_page; > + > + pmd_end = min(end, dmirror_pt_pmd_end(addr)); > + pmd_page = dmirror_pt_page(dpudp); > + if (!pmd_page) { > + if (!populate) { > + addr = pmd_end; > + continue; > + } > + pmd_page = populate_pt(dmirror, dpudp); > + if (!pmd_page) { > + kunmap(pud_page); > + return -ENOMEM; > + } > + } > + dpmdp = kmap(pmd_page); > + dpmdp += dmirror_pt_pmd(addr); > + for (; addr != pmd_end; dpmdp++) { > + u64 *dptep; > + unsigned long pte_end; > + struct page *pte_page; > + > + pte_end = min(end, dmirror_pt_pte_end(addr)); > + pte_page = dmirror_pt_page(dpmdp); > + if (!pte_page) { > + if (!populate) { > + addr = pte_end; > + continue; > + } > + pte_page = populate_pt(dmirror, dpmdp); > + if (!pte_page) { > + kunmap(pmd_page); > + kunmap(pud_page); > + return -ENOMEM; > + } > + } > + if (!cb) { > + addr = pte_end; > + continue; > + } > + dptep = kmap(pte_page); > + dptep += dmirror_pt_pte(addr); > + ret = cb(dmirror, addr, pte_end, dptep, > + private); > + kunmap(pte_page); > + if (ret) { > + kunmap(pmd_page); > + kunmap(pud_page); > + return ret; > + } > + addr = pte_end; > + } > + kunmap(pmd_page); > + addr = pmd_end; > + } > + kunmap(pud_page); > + addr = pud_end; > + } > + > + return ret; > +} > + > +static void dmirror_pt_free(struct dmirror *dmirror) > +{ > + u64 *dpgdp = dmirror->pt.pgd; > + > + for (; dpgdp != dmirror->pt.pgd + PTRS_PER_PGD; dpgdp++) { > + u64 *dpudp, *dpudp_orig; > + u64 *dpudp_end; > + struct page *pud_page; > + > + pud_page = dmirror_pt_page(dpgdp); > + if (!pud_page) > + continue; > + > + dpudp_orig = kmap_atomic(pud_page); > + dpudp = dpudp_orig; > + dpudp_end = dpudp + PTRS_PER_PUD; > + for (; dpudp != dpudp_end; dpudp++) { > + u64 *dpmdp, *dpmdp_orig; > + u64 *dpmdp_end; > + struct page *pmd_page; > + > + pmd_page = dmirror_pt_page(dpudp); > + if (!pmd_page) > + continue; > + > + dpmdp_orig = kmap_atomic(pmd_page); > + dpmdp = dpmdp_orig; > + dpmdp_end = dpmdp + PTRS_PER_PMD; > + for (; dpmdp != dpmdp_end; dpmdp++) { > + struct page *pte_page; > + > + pte_page = dmirror_pt_page(dpmdp); > + if (!pte_page) > + continue; > + > + *dpmdp = 0; > + __free_page(pte_page); > + } > + kunmap_atomic(dpmdp_orig); > + *dpudp = 0; > + __free_page(pmd_page); > + } > + kunmap_atomic(dpudp_orig); > + *dpgdp = 0; > + __free_page(pud_page); > + } > +} > + > +static int dmirror_bounce_init(struct dmirror_bounce *bounce, > + unsigned long addr, > + unsigned long size) > +{ > + bounce->addr = addr; > + bounce->size = size; > + bounce->cpages = 0; > + bounce->ptr = vmalloc(size); > + if (!bounce->ptr) > + return -ENOMEM; > + return 0; > +} > + > +static int dmirror_bounce_copy_from(struct dmirror_bounce *bounce, > + unsigned long addr) > +{ > + unsigned long end = addr + bounce->size; > + char __user *uptr = (void __user *)addr; > + void *ptr = bounce->ptr; > + > + for (; addr < end; addr += PAGE_SIZE, ptr += PAGE_SIZE, > + uptr += PAGE_SIZE) { > + int ret; > + > + ret = copy_from_user(ptr, uptr, PAGE_SIZE); > + if (ret) > + return ret; > + } > + > + return 0; > +} > + > +static int dmirror_bounce_copy_to(struct dmirror_bounce *bounce, > + unsigned long addr) > +{ > + unsigned long end = addr + bounce->size; > + char __user *uptr = (void __user *)addr; > + void *ptr = bounce->ptr; > + > + for (; addr < end; addr += PAGE_SIZE, ptr += PAGE_SIZE, > + uptr += PAGE_SIZE) { > + int ret; > + > + ret = copy_to_user(uptr, ptr, PAGE_SIZE); > + if (ret) > + return ret; > + } > + > + return 0; > +} > + > +static void dmirror_bounce_fini(struct dmirror_bounce *bounce) > +{ > + vfree(bounce->ptr); > +} > + > +static int dmirror_do_update(struct dmirror *dmirror, > + unsigned long addr, > + unsigned long end, > + u64 *dptep, > + void *private) > +{ > + /* > + * The page table doesn't hold references to pages since it relies on > + * the mmu notifier to clear pointers when they become stale. > + * Therefore, it is OK to just clear the pte. > + */ > + for (; addr < end; addr += PAGE_SIZE, ++dptep) > + *dptep = 0; > + > + return 0; > +} > + > +static int dmirror_update(struct hmm_mirror *mirror, > + const struct mmu_notifier_range *update) > +{ > + struct dmirror *dmirror = container_of(mirror, struct dmirror, mirror); > + > + if (mmu_notifier_range_blockable(update)) > + mutex_lock(&dmirror->mutex); > + else if (!mutex_trylock(&dmirror->mutex)) > + return -EAGAIN; > + > + dmirror_pt_walk(dmirror, dmirror_do_update, update->start, > + update->end, NULL, false); > + mutex_unlock(&dmirror->mutex); > + return 0; > +} > + > +static const struct hmm_mirror_ops dmirror_ops = { > + .sync_cpu_device_pagetables = &dmirror_update, > +}; > + > +/* > + * dmirror_new() - allocate and initialize dmirror struct. > + * > + * @mdevice: The device this mirror is associated with. > + * @filp: The active device file descriptor this mirror is associated with. > + */ > +static struct dmirror *dmirror_new(struct dmirror_device *mdevice) > +{ > + struct mm_struct *mm = get_task_mm(current); > + struct dmirror *dmirror; > + int ret; > + > + if (!mm) > + goto err; > + > + /* Mirror this process address space */ > + dmirror = kzalloc(sizeof(*dmirror), GFP_KERNEL); > + if (dmirror == NULL) > + goto err_mmput; > + > + dmirror->mdevice = mdevice; > + dmirror->mirror.ops = &dmirror_ops; > + mutex_init(&dmirror->mutex); > + > + down_write(&mm->mmap_sem); > + ret = hmm_mirror_register(&dmirror->mirror, mm); > + up_write(&mm->mmap_sem); > + if (ret) > + goto err_free; > + > + mmput(mm); > + return dmirror; > + > +err_free: > + kfree(dmirror); > +err_mmput: > + mmput(mm); > +err: > + return NULL; > +} > + > +static void dmirror_del(struct dmirror *dmirror) > +{ > + hmm_mirror_unregister(&dmirror->mirror); > + dmirror_pt_free(dmirror); > + kfree(dmirror); > +} > + > +/* > + * Below are the file operation for the dmirror device file. Only ioctl matters. > + * > + * Note this is highly specific to the dmirror device driver and should not be > + * construed as an example on how to design the API a real device driver would > + * expose to userspace. > + */ > +static ssize_t dmirror_fops_read(struct file *filp, > + char __user *buf, > + size_t count, > + loff_t *ppos) > +{ > + return -EINVAL; > +} > + > +static ssize_t dmirror_fops_write(struct file *filp, > + const char __user *buf, > + size_t count, > + loff_t *ppos) > +{ > + return -EINVAL; > +} > + > +static int dmirror_fops_mmap(struct file *filp, struct vm_area_struct *vma) > +{ > + /* Forbid mmap of the dmirror device file. */ > + return -EINVAL; > +} > + > +static int dmirror_fops_open(struct inode *inode, struct file *filp) > +{ > + struct cdev *cdev = inode->i_cdev; > + struct dmirror_device *mdevice; > + struct dmirror *dmirror; > + > + /* No exclusive opens. */ > + if (filp->f_flags & O_EXCL) > + return -EINVAL; > + > + mdevice = container_of(cdev, struct dmirror_device, cdevice); > + dmirror = dmirror_new(mdevice); > + if (!dmirror) > + return -ENOMEM; > + > + /* Only the first open registers the address space. */ > + mutex_lock(&mdevice->devmem_lock); > + if (filp->private_data) > + goto err_busy; > + filp->private_data = dmirror; > + mutex_unlock(&mdevice->devmem_lock); > + > + return 0; > + > +err_busy: > + mutex_unlock(&mdevice->devmem_lock); > + dmirror_del(dmirror); > + return -EBUSY; > +} > + > +static int dmirror_fops_release(struct inode *inode, struct file *filp) > +{ > + struct dmirror *dmirror = filp->private_data; > + > + if (!dmirror) > + return 0; > + > + dmirror_del(dmirror); > + filp->private_data = NULL; > + > + return 0; > +} > + > +static inline struct dmirror_device *dmirror_page_to_device(struct page *page) > + > +{ > + struct dmirror_chunk *devmem; > + > + devmem = container_of(page->pgmap, struct dmirror_chunk, pagemap); > + return devmem->mdevice; > +} > + > +static bool dmirror_device_is_mine(struct dmirror_device *mdevice, > + struct page *page) > +{ > + if (!is_zone_device_page(page)) > + return false; > + return page->pgmap->ops == &dmirror_devmem_ops && > + dmirror_page_to_device(page) == mdevice; > +} > + > +static int dmirror_do_fault(struct dmirror *dmirror, > + unsigned long addr, > + unsigned long end, > + u64 *dptep, > + void *private) > +{ > + struct hmm_range *range = private; > + unsigned long idx = (addr - range->start) >> PAGE_SHIFT; > + uint64_t *pfns = range->pfns; > + > + for (; addr < end; addr += PAGE_SIZE, ++dptep, ++idx) { > + struct page *page; > + u64 dpte; > + > + /* > + * HMM_PFN_ERROR is returned if it is accessing invalid memory > + * either because of memory error (hardware detected memory > + * corruption) or more likely because of truncate on mmap > + * file. > + */ > + if (pfns[idx] == range->values[HMM_PFN_ERROR]) > + return -EFAULT; > + /* > + * The only special PFN HMM returns is the read-only zero page > + * which doesn't have a matching struct page. > + */ > + if (pfns[idx] == range->values[HMM_PFN_SPECIAL]) { > + *dptep = DPT_ZPAGE; > + continue; > + } > + if (!(pfns[idx] & range->flags[HMM_PFN_VALID])) > + return -EFAULT; > + page = hmm_device_entry_to_page(range, pfns[idx]); > + /* We asked for pages to be populated but check anyway. */ > + if (!page) > + return -EFAULT; > + dpte = dmirror_pt_from_page(page); > + if (is_zone_device_page(page)) { > + if (!dmirror_device_is_mine(dmirror->mdevice, page)) > + continue; > + dpte |= DPT_DPAGE; > + } > + if (pfns[idx] & range->flags[HMM_PFN_WRITE]) > + dpte |= DPT_WRITE; > + else if (range->default_flags & range->flags[HMM_PFN_WRITE]) > + return -EFAULT; > + *dptep = dpte; > + } > + > + return 0; > +} > + > +static int dmirror_fault(struct dmirror *dmirror, > + unsigned long start, > + unsigned long end, > + bool write) > +{ > + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; > + unsigned long addr; > + unsigned long next; > + uint64_t pfns[64]; > + struct hmm_range range = { > + .pfns = pfns, > + .flags = dmirror_hmm_flags, > + .values = dmirror_hmm_values, > + .pfn_shift = DPT_SHIFT, > + .pfn_flags_mask = ~(dmirror_hmm_flags[HMM_PFN_VALID] | > + dmirror_hmm_flags[HMM_PFN_WRITE]), > + .default_flags = dmirror_hmm_flags[HMM_PFN_VALID] | > + (write ? dmirror_hmm_flags[HMM_PFN_WRITE] : 0), > + }; > + int ret = 0; > + > + for (addr = start; addr < end; ) { > + long count; > + > + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); > + range.start = addr; > + range.end = next; > + > + down_read(&mm->mmap_sem); > + > + ret = hmm_range_register(&range, &dmirror->mirror); > + if (ret) { > + up_read(&mm->mmap_sem); > + break; > + } > + > + if (!hmm_range_wait_until_valid(&range, > + DMIRROR_RANGE_FAULT_TIMEOUT)) { > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + continue; > + } > + > + count = hmm_range_fault(&range, 0); > + if (count < 0) { > + ret = count; > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + break; > + } > + > + if (!hmm_range_valid(&range)) { > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + continue; > + } > + mutex_lock(&dmirror->mutex); > + ret = dmirror_pt_walk(dmirror, dmirror_do_fault, > + addr, next, &range, true); > + mutex_unlock(&dmirror->mutex); > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + if (ret) > + break; > + > + addr = next; > + } > + > + return ret; > +} > + > +static int dmirror_do_read(struct dmirror *dmirror, > + unsigned long addr, > + unsigned long end, > + u64 *dptep, > + void *private) > +{ > + struct dmirror_bounce *bounce = private; > + void *ptr; > + > + ptr = bounce->ptr + ((addr - bounce->addr) & PAGE_MASK); > + > + for (; addr < end; addr += PAGE_SIZE, ++dptep) { > + struct page *page; > + void *tmp; > + > + page = dmirror_pt_page(dptep); > + if (!page) > + return -ENOENT; > + > + tmp = kmap(page); > + memcpy(ptr, tmp, PAGE_SIZE); > + kunmap(page); > + > + ptr += PAGE_SIZE; > + bounce->cpages++; > + } > + > + return 0; > +} > + > +static int dmirror_read(struct dmirror *dmirror, > + struct hmm_dmirror_cmd *cmd) > +{ > + struct dmirror_bounce bounce; > + unsigned long start, end; > + unsigned long size = cmd->npages << PAGE_SHIFT; > + int ret; > + > + start = cmd->addr; > + end = start + size; > + if (end < start) > + return -EINVAL; > + > + ret = dmirror_bounce_init(&bounce, start, size); > + if (ret) > + return ret; > + > +again: > + mutex_lock(&dmirror->mutex); > + ret = dmirror_pt_walk(dmirror, dmirror_do_read, start, end, &bounce, > + false); > + mutex_unlock(&dmirror->mutex); > + if (ret == 0) > + ret = dmirror_bounce_copy_to(&bounce, cmd->ptr); > + else if (ret == -ENOENT) { > + start = cmd->addr + (bounce.cpages << PAGE_SHIFT); > + ret = dmirror_fault(dmirror, start, end, false); > + if (ret == 0) { > + cmd->faults++; > + goto again; > + } > + } > + > + cmd->cpages = bounce.cpages; > + dmirror_bounce_fini(&bounce); > + return ret; > +} > + > +static int dmirror_do_write(struct dmirror *dmirror, > + unsigned long addr, > + unsigned long end, > + u64 *dptep, > + void *private) > +{ > + struct dmirror_bounce *bounce = private; > + void *ptr; > + > + ptr = bounce->ptr + ((addr - bounce->addr) & PAGE_MASK); > + > + for (; addr < end; addr += PAGE_SIZE, ++dptep) { > + struct page *page; > + void *tmp; > + > + page = dmirror_pt_page_write(dptep); > + if (!page) > + return -ENOENT; > + > + tmp = kmap(page); > + memcpy(tmp, ptr, PAGE_SIZE); > + kunmap(page); > + > + ptr += PAGE_SIZE; > + bounce->cpages++; > + } > + > + return 0; > +} > + > +static int dmirror_write(struct dmirror *dmirror, > + struct hmm_dmirror_cmd *cmd) > +{ > + struct dmirror_bounce bounce; > + unsigned long start, end; > + unsigned long size = cmd->npages << PAGE_SHIFT; > + int ret; > + > + start = cmd->addr; > + end = start + size; > + if (end < start) > + return -EINVAL; > + > + ret = dmirror_bounce_init(&bounce, start, size); > + if (ret) > + return ret; > + ret = dmirror_bounce_copy_from(&bounce, cmd->ptr); > + if (ret) > + return ret; > + > +again: > + mutex_lock(&dmirror->mutex); > + ret = dmirror_pt_walk(dmirror, dmirror_do_write, > + start, end, &bounce, false); > + mutex_unlock(&dmirror->mutex); > + if (ret == -ENOENT) { > + start = cmd->addr + (bounce.cpages << PAGE_SHIFT); > + ret = dmirror_fault(dmirror, start, end, true); > + if (ret == 0) { > + cmd->faults++; > + goto again; > + } > + } > + > + cmd->cpages = bounce.cpages; > + dmirror_bounce_fini(&bounce); > + return ret; > +} > + > +static bool dmirror_allocate_chunk(struct dmirror_device *mdevice, > + struct page **ppage) > +{ > + struct dmirror_chunk *devmem; > + struct resource *res; > + unsigned long pfn; > + unsigned long pfn_first; > + unsigned long pfn_last; > + void *ptr; > + > + mutex_lock(&mdevice->devmem_lock); > + > + if (mdevice->devmem_count == mdevice->devmem_capacity) { > + struct dmirror_chunk **new_chunks; > + unsigned int new_capacity; > + > + new_capacity = mdevice->devmem_capacity + > + DEVMEM_CHUNKS_RESERVE; > + new_chunks = krealloc(mdevice->devmem_chunks, > + sizeof(new_chunks[0]) * new_capacity, > + GFP_KERNEL); > + if (!new_chunks) > + goto err; > + mdevice->devmem_capacity = new_capacity; > + mdevice->devmem_chunks = new_chunks; > + } > + > + res = devm_request_free_mem_region(&mdevice->pdevice->dev, > + &iomem_resource, DEVMEM_CHUNK_SIZE); > + if (IS_ERR(res)) > + goto err; > + > + devmem = kzalloc(sizeof(*devmem), GFP_KERNEL); > + if (!devmem) > + goto err; > + > + devmem->pagemap.type = MEMORY_DEVICE_PRIVATE; > + devmem->pagemap.res = *res; > + devmem->pagemap.ops = &dmirror_devmem_ops; > + ptr = devm_memremap_pages(&mdevice->pdevice->dev, &devmem->pagemap); > + if (IS_ERR(ptr)) > + goto err_free; > + > + devmem->mdevice = mdevice; > + pfn_first = devmem->pagemap.res.start >> PAGE_SHIFT; > + pfn_last = pfn_first + > + (resource_size(&devmem->pagemap.res) >> PAGE_SHIFT); > + mdevice->devmem_chunks[mdevice->devmem_count++] = devmem; > + > + mutex_unlock(&mdevice->devmem_lock); > + > + pr_info("added new %u MB chunk (total %u chunks, %u MB) PFNs [0x%lx 0x%lx)\n", > + DEVMEM_CHUNK_SIZE / (1024 * 1024), > + mdevice->devmem_count, > + mdevice->devmem_count * (DEVMEM_CHUNK_SIZE / (1024 * 1024)), > + pfn_first, pfn_last); > + > + spin_lock(&mdevice->lock); > + for (pfn = pfn_first; pfn < pfn_last; pfn++) { > + struct page *page = pfn_to_page(pfn); > + > + page->zone_device_data = mdevice->free_pages; > + mdevice->free_pages = page; > + } > + if (ppage) { > + *ppage = mdevice->free_pages; > + mdevice->free_pages = (*ppage)->zone_device_data; > + mdevice->calloc++; > + } > + spin_unlock(&mdevice->lock); > + > + return true; > + > +err_free: > + kfree(devmem); > +err: > + mutex_unlock(&mdevice->devmem_lock); > + return false; > +} > + > +static struct page *dmirror_devmem_alloc_page(struct dmirror_device *mdevice) > +{ > + struct page *dpage = NULL; > + struct page *rpage; > + > + /* > + * This is a fake device so we alloc real system memory to store > + * our device memory. > + */ > + rpage = alloc_page(GFP_HIGHUSER); > + if (!rpage) > + return NULL; > + > + spin_lock(&mdevice->lock); > + > + if (mdevice->free_pages) { > + dpage = mdevice->free_pages; > + mdevice->free_pages = dpage->zone_device_data; > + mdevice->calloc++; > + spin_unlock(&mdevice->lock); > + } else { > + spin_unlock(&mdevice->lock); > + if (!dmirror_allocate_chunk(mdevice, &dpage)) > + goto error; > + } > + > + dpage->zone_device_data = rpage; > + get_page(dpage); > + lock_page(dpage); > + return dpage; > + > +error: > + __free_page(rpage); > + return NULL; > +} > + > +static void dmirror_migrate_alloc_and_copy(struct migrate_vma *args, > + struct dmirror *dmirror) > +{ > + struct dmirror_device *mdevice = dmirror->mdevice; > + const unsigned long *src = args->src; > + unsigned long *dst = args->dst; > + unsigned long addr; > + > + for (addr = args->start; addr < args->end; addr += PAGE_SIZE, > + src++, dst++) { > + struct page *spage; > + struct page *dpage; > + struct page *rpage; > + > + if (!(*src & MIGRATE_PFN_MIGRATE)) > + continue; > + > + /* > + * Note that spage might be NULL which is OK since it is an > + * unallocated pte_none() or read-only zero page. > + */ > + spage = migrate_pfn_to_page(*src); > + > + /* > + * Don't migrate device private pages from our own driver or > + * others. For our own we would do a device private memory copy > + * not a migration and for others, we would need to fault the > + * other device's page into system memory first. > + */ > + if (spage && is_zone_device_page(spage)) > + continue; > + > + dpage = dmirror_devmem_alloc_page(mdevice); > + if (!dpage) > + continue; > + > + rpage = dpage->zone_device_data; > + if (spage) > + copy_highpage(rpage, spage); > + else > + clear_highpage(rpage); > + > + /* > + * Normally, a device would use the page->zone_device_data to > + * point to the mirror but here we use it to hold the page for > + * the simulated device memory and that page holds the pointer > + * to the mirror. > + */ > + rpage->zone_device_data = dmirror; > + > + *dst = migrate_pfn(page_to_pfn(dpage)) | > + MIGRATE_PFN_LOCKED; > + if ((*src & MIGRATE_PFN_WRITE) || > + (!spage && args->vma->vm_flags & VM_WRITE)) > + *dst |= MIGRATE_PFN_WRITE; > + } > + /* Try to pre-allocate page tables. */ > + mutex_lock(&dmirror->mutex); > + dmirror_pt_walk(dmirror, NULL, args->start, args->end, NULL, true); > + mutex_unlock(&dmirror->mutex); > +} > + > +struct dmirror_migrate { > + struct hmm_dmirror_cmd *cmd; > + const unsigned long *src; > + const unsigned long *dst; > + unsigned long start; > +}; > + > +static int dmirror_do_migrate(struct dmirror *dmirror, > + unsigned long addr, > + unsigned long end, > + u64 *dptep, > + void *private) > +{ > + struct dmirror_migrate *migrate = private; > + const unsigned long *src = migrate->src; > + const unsigned long *dst = migrate->dst; > + unsigned long idx = (addr - migrate->start) >> PAGE_SHIFT; > + > + for (; addr < end; addr += PAGE_SIZE, ++dptep, ++idx) { > + struct page *page; > + u64 dpte; > + > + if (!(src[idx] & MIGRATE_PFN_MIGRATE)) > + continue; > + > + page = migrate_pfn_to_page(dst[idx]); > + if (!page) > + continue; > + > + /* > + * Map the page that holds the data so dmirror_pt_walk() > + * doesn't have to deal with ZONE_DEVICE private pages. > + */ > + page = page->zone_device_data; > + dpte = dmirror_pt_from_page(page) | DPT_DPAGE; > + if (dst[idx] & MIGRATE_PFN_WRITE) > + dpte |= DPT_WRITE; > + *dptep = dpte; > + } > + > + return 0; > +} > + > +static void dmirror_migrate_finalize_and_map(struct migrate_vma *args, > + struct dmirror *dmirror, > + struct hmm_dmirror_cmd *cmd) > +{ > + struct dmirror_migrate migrate; > + > + migrate.cmd = cmd; > + migrate.src = args->src; > + migrate.dst = args->dst; > + migrate.start = args->start; > + > + /* Map the migrated pages into the device's page tables. */ > + mutex_lock(&dmirror->mutex); > + dmirror_pt_walk(dmirror, dmirror_do_migrate, args->start, args->end, > + &migrate, true); > + mutex_unlock(&dmirror->mutex); > +} > + > +static int dmirror_migrate(struct dmirror *dmirror, > + struct hmm_dmirror_cmd *cmd) > +{ > + unsigned long start, end, addr; > + unsigned long size = cmd->npages << PAGE_SHIFT; > + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; > + struct vm_area_struct *vma; > + unsigned long src_pfns[64]; > + unsigned long dst_pfns[64]; > + struct dmirror_bounce bounce; > + struct migrate_vma args; > + unsigned long next; > + int ret; > + > + start = cmd->addr; > + end = start + size; > + if (end < start) > + return -EINVAL; > + > + down_read(&mm->mmap_sem); > + for (addr = start; addr < end; addr = next) { > + next = min(end, addr + (ARRAY_SIZE(src_pfns) << PAGE_SHIFT)); > + > + vma = find_vma(mm, addr); > + if (!vma || addr < vma->vm_start) { > + ret = -EINVAL; > + goto out; > + } > + if (next > vma->vm_end) > + next = vma->vm_end; > + > + args.vma = vma; > + args.src = src_pfns; > + args.dst = dst_pfns; > + args.start = addr; > + args.end = next; > + ret = migrate_vma_setup(&args); > + if (ret) > + goto out; > + > + dmirror_migrate_alloc_and_copy(&args, dmirror); > + migrate_vma_pages(&args); > + dmirror_migrate_finalize_and_map(&args, dmirror, cmd); > + migrate_vma_finalize(&args); > + } > + up_read(&mm->mmap_sem); > + > + /* Return the migrated data for verification. */ > + ret = dmirror_bounce_init(&bounce, start, size); > + if (ret) > + return ret; > + mutex_lock(&dmirror->mutex); > + ret = dmirror_pt_walk(dmirror, dmirror_do_read, start, end, &bounce, > + false); > + mutex_unlock(&dmirror->mutex); > + if (ret == 0) > + ret = dmirror_bounce_copy_to(&bounce, cmd->ptr); > + cmd->cpages = bounce.cpages; > + dmirror_bounce_fini(&bounce); > + return ret; > + > +out: > + up_read(&mm->mmap_sem); > + return ret; > +} > + > +static void dmirror_mkentry(struct dmirror *dmirror, > + struct hmm_range *range, > + unsigned char *perm, > + uint64_t entry) > +{ > + struct page *page; > + > + if (entry == range->values[HMM_PFN_ERROR]) { > + *perm = HMM_DMIRROR_PROT_ERROR; > + return; > + } > + page = hmm_device_entry_to_page(range, entry); > + if (!page) { > + *perm = HMM_DMIRROR_PROT_NONE; > + return; > + } > + if (entry & range->flags[HMM_PFN_DEVICE_PRIVATE]) { > + /* Is the page migrated to this device or some other? */ > + if (dmirror->mdevice == dmirror_page_to_device(page)) > + *perm = HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL; > + else > + *perm = HMM_DMIRROR_PROT_DEV_PRIVATE_REMOTE; > + } else if (is_zero_pfn(page_to_pfn(page))) > + *perm = HMM_DMIRROR_PROT_ZERO; > + else > + *perm = HMM_DMIRROR_PROT_NONE; > + if (entry & range->flags[HMM_PFN_WRITE]) > + *perm |= HMM_DMIRROR_PROT_WRITE; > + else > + *perm |= HMM_DMIRROR_PROT_READ; > +} > + > +static int dmirror_snapshot(struct dmirror *dmirror, > + struct hmm_dmirror_cmd *cmd) > +{ > + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; > + unsigned long start, end; > + unsigned long size = cmd->npages << PAGE_SHIFT; > + unsigned long addr; > + unsigned long next; > + uint64_t pfns[64]; > + unsigned char perm[64]; > + char __user *uptr; > + struct hmm_range range = { > + .pfns = pfns, > + .flags = dmirror_hmm_flags, > + .values = dmirror_hmm_values, > + .pfn_shift = DPT_SHIFT, > + .pfn_flags_mask = ~0ULL, > + }; > + int ret = 0; > + > + start = cmd->addr; > + end = start + size; > + uptr = (void __user *)cmd->ptr; > + > + for (addr = start; addr < end; ) { > + long count; > + unsigned long i; > + unsigned long n; > + > + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); > + range.start = addr; > + range.end = next; > + > + down_read(&mm->mmap_sem); > + > + ret = hmm_range_register(&range, &dmirror->mirror); > + if (ret) { > + up_read(&mm->mmap_sem); > + break; > + } > + > + if (!hmm_range_wait_until_valid(&range, > + DMIRROR_RANGE_FAULT_TIMEOUT)) { > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + continue; > + } > + > + count = hmm_range_fault(&range, HMM_FAULT_SNAPSHOT); > + if (count < 0) { > + ret = count; > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + if (ret == -EBUSY) > + continue; > + break; > + } > + > + if (!hmm_range_valid(&range)) { > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + continue; > + } > + > + n = (next - addr) >> PAGE_SHIFT; > + for (i = 0; i < n; i++) > + dmirror_mkentry(dmirror, &range, perm + i, pfns[i]); > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + > + ret = copy_to_user(uptr, perm, n); > + if (ret) > + break; > + > + cmd->cpages += n; > + uptr += n; > + addr = next; > + } > + > + return ret; > +} > + > +static long dmirror_fops_unlocked_ioctl(struct file *filp, > + unsigned int command, > + unsigned long arg) > +{ > + void __user *uarg = (void __user *)arg; > + struct hmm_dmirror_cmd cmd; > + struct dmirror *dmirror; > + int ret; > + > + dmirror = filp->private_data; > + if (!dmirror) > + return -EINVAL; > + > + ret = copy_from_user(&cmd, uarg, sizeof(cmd)); > + if (ret) > + return ret; > + > + if (cmd.addr & ~PAGE_MASK) > + return -EINVAL; > + if (cmd.addr >= (cmd.addr + (cmd.npages << PAGE_SHIFT))) > + return -EINVAL; > + > + cmd.cpages = 0; > + cmd.faults = 0; > + > + switch (command) { > + case HMM_DMIRROR_READ: > + ret = dmirror_read(dmirror, &cmd); > + break; > + > + case HMM_DMIRROR_WRITE: > + ret = dmirror_write(dmirror, &cmd); > + break; > + > + case HMM_DMIRROR_MIGRATE: > + ret = dmirror_migrate(dmirror, &cmd); > + break; > + > + case HMM_DMIRROR_SNAPSHOT: > + ret = dmirror_snapshot(dmirror, &cmd); > + break; > + > + default: > + return -EINVAL; > + } > + if (ret) > + return ret; > + > + return copy_to_user(uarg, &cmd, sizeof(cmd)); > +} > + > +static const struct file_operations dmirror_fops = { > + .read = dmirror_fops_read, > + .write = dmirror_fops_write, > + .mmap = dmirror_fops_mmap, > + .open = dmirror_fops_open, > + .release = dmirror_fops_release, > + .unlocked_ioctl = dmirror_fops_unlocked_ioctl, > + .llseek = default_llseek, > + .owner = THIS_MODULE, > +}; > + > +static void dmirror_devmem_free(struct page *page) > +{ > + struct page *rpage = page->zone_device_data; > + struct dmirror_device *mdevice; > + > + if (rpage) > + __free_page(rpage); > + > + mdevice = dmirror_page_to_device(page); > + > + spin_lock(&mdevice->lock); > + mdevice->cfree++; > + page->zone_device_data = mdevice->free_pages; > + mdevice->free_pages = page; > + spin_unlock(&mdevice->lock); > +} > + > +static vm_fault_t dmirror_devmem_fault_alloc_and_copy(struct migrate_vma *args, > + struct dmirror_device *mdevice) > +{ > + struct vm_area_struct *vma = args->vma; > + const unsigned long *src = args->src; > + unsigned long *dst = args->dst; > + unsigned long start = args->start; > + unsigned long end = args->end; > + unsigned long addr; > + > + for (addr = start; addr < end; addr += PAGE_SIZE, > + src++, dst++) { > + struct page *dpage, *spage; > + > + spage = migrate_pfn_to_page(*src); > + if (!spage || !(*src & MIGRATE_PFN_MIGRATE)) > + continue; > + if (!dmirror_device_is_mine(mdevice, spage)) > + continue; > + spage = spage->zone_device_data; > + > + dpage = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, addr); > + if (!dpage) > + continue; > + > + lock_page(dpage); > + copy_highpage(dpage, spage); > + *dst = migrate_pfn(page_to_pfn(dpage)) | > + MIGRATE_PFN_LOCKED; > + if (*src & MIGRATE_PFN_WRITE) > + *dst |= MIGRATE_PFN_WRITE; > + } > + return 0; > +} > + > +static void dmirror_devmem_fault_finalize_and_map(struct migrate_vma *args, > + struct dmirror *dmirror) > +{ > + /* Invalidate the device's page table mapping. */ > + mutex_lock(&dmirror->mutex); > + dmirror_pt_walk(dmirror, dmirror_do_update, args->start, args->end, > + NULL, false); > + mutex_unlock(&dmirror->mutex); > +} > + > +static vm_fault_t dmirror_devmem_fault(struct vm_fault *vmf) > +{ > + struct migrate_vma args; > + unsigned long src_pfns; > + unsigned long dst_pfns; > + struct page *rpage; > + struct dmirror *dmirror; > + vm_fault_t ret; > + > + /* FIXME demonstrate how we can adjust migrate range */ > + args.vma = vmf->vma; > + args.start = vmf->address; > + args.end = args.start + PAGE_SIZE; > + args.src = &src_pfns; > + args.dst = &dst_pfns; > + > + if (migrate_vma_setup(&args)) > + return VM_FAULT_SIGBUS; > + > + /* > + * Normally, a device would use the page->zone_device_data to point to > + * the mirror but here we use it to hold the page for the simulated > + * device memory and that page holds the pointer to the mirror. > + */ > + rpage = vmf->page->zone_device_data; > + dmirror = rpage->zone_device_data; > + > + ret = dmirror_devmem_fault_alloc_and_copy(&args, dmirror->mdevice); > + if (ret) > + return ret; > + migrate_vma_pages(&args); > + dmirror_devmem_fault_finalize_and_map(&args, dmirror); > + migrate_vma_finalize(&args); > + return 0; > +} > + > +static const struct dev_pagemap_ops dmirror_devmem_ops = { > + .page_free = dmirror_devmem_free, > + .migrate_to_ram = dmirror_devmem_fault, > +}; > + > +static void dmirror_pdev_del(void *arg) > +{ > + struct dmirror_device *mdevice = arg; > + unsigned int i; > + > + if (mdevice->devmem_chunks) { > + for (i = 0; i < mdevice->devmem_count; i++) > + kfree(mdevice->devmem_chunks[i]); > + kfree(mdevice->devmem_chunks); > + } > + > + cdev_del(&mdevice->cdevice); > + kfree(mdevice); > +} > + > +static int dmirror_probe(struct platform_device *pdev) > +{ > + struct dmirror_device *mdevice; > + int ret; > + > + mdevice = kzalloc(sizeof(*mdevice), GFP_KERNEL); > + if (!mdevice) > + return -ENOMEM; > + > + mdevice->pdevice = pdev; > + mutex_init(&mdevice->devmem_lock); > + spin_lock_init(&mdevice->lock); > + > + cdev_init(&mdevice->cdevice, &dmirror_fops); > + ret = cdev_add(&mdevice->cdevice, pdev->dev.devt, 1); > + if (ret) { > + kfree(mdevice); > + return ret; > + } > + > + platform_set_drvdata(pdev, mdevice); > + ret = devm_add_action_or_reset(&pdev->dev, dmirror_pdev_del, mdevice); > + if (ret) > + return ret; > + > + /* Build list of free struct page */ > + dmirror_allocate_chunk(mdevice, NULL); > + > + return 0; > +} > + > +static int dmirror_remove(struct platform_device *pdev) > +{ > + /* all probe actions are unwound by devm */ > + return 0; > +} > + > +static struct platform_driver dmirror_device_driver = { > + .probe = dmirror_probe, > + .remove = dmirror_remove, > + .driver = { > + .name = "HMM_DMIRROR", > + }, > +}; > + > +static int __init hmm_dmirror_init(void) > +{ > + int ret; > + int id; > + > + ret = platform_driver_register(&dmirror_device_driver); > + if (ret) > + return ret; > + > + ret = alloc_chrdev_region(&dmirror_dev, 0, DMIRROR_NDEVICES, > + "HMM_DMIRROR"); > + if (ret) > + goto err_unreg; > + > + for (id = 0; id < DMIRROR_NDEVICES; id++) { > + struct platform_device *pd; > + > + pd = platform_device_alloc("HMM_DMIRROR", id); > + if (!pd) { > + ret = -ENOMEM; > + goto err_chrdev; > + } > + pd->dev.devt = MKDEV(MAJOR(dmirror_dev), id); > + ret = platform_device_add(pd); > + if (ret) { > + platform_device_put(pd); > + goto err_chrdev; > + } > + dmirror_platform_devices[id] = pd; > + } > + > + /* > + * Allocate a zero page to simulate a reserved page of device private > + * memory which is always zero. The zero_pfn page isn't used just to > + * make the code here simpler (i.e., we need a struct page for it). > + */ > + dmirror_zero_page = alloc_page(GFP_HIGHUSER | __GFP_ZERO); > + if (!dmirror_zero_page) > + goto err_chrdev; > + > + pr_info("hmm_dmirror loaded. This is only for testing HMM.\n"); > + return 0; > + > +err_chrdev: > + while (--id >= 0) { > + platform_device_unregister(dmirror_platform_devices[id]); > + dmirror_platform_devices[id] = NULL; > + } > + unregister_chrdev_region(dmirror_dev, 1); > +err_unreg: > + platform_driver_unregister(&dmirror_device_driver); > + return ret; > +} > + > +static void __exit hmm_dmirror_exit(void) > +{ > + int id; > + > + if (dmirror_zero_page) > + __free_page(dmirror_zero_page); > + for (id = 0; id < DMIRROR_NDEVICES; id++) > + platform_device_unregister(dmirror_platform_devices[id]); > + unregister_chrdev_region(dmirror_dev, DMIRROR_NDEVICES); > + platform_driver_unregister(&dmirror_device_driver); > + mmu_notifier_synchronize(); > +} > + > +module_init(hmm_dmirror_init); > +module_exit(hmm_dmirror_exit); > +MODULE_LICENSE("GPL"); > diff --git a/include/Kbuild b/include/Kbuild > index ffba79483cc5..6ffb44a45957 100644 > --- a/include/Kbuild > +++ b/include/Kbuild > @@ -1063,6 +1063,7 @@ header-test- += uapi/linux/coda_psdev.h > header-test- += uapi/linux/errqueue.h > header-test- += uapi/linux/eventpoll.h > header-test- += uapi/linux/hdlc/ioctl.h > +header-test- += uapi/linux/hmm_dmirror.h > header-test- += uapi/linux/input.h > header-test- += uapi/linux/kvm.h > header-test- += uapi/linux/kvm_para.h > diff --git a/include/uapi/linux/hmm_dmirror.h b/include/uapi/linux/hmm_dmirror.h > new file mode 100644 > index 000000000000..61d3643aff95 > --- /dev/null > +++ b/include/uapi/linux/hmm_dmirror.h > @@ -0,0 +1,74 @@ > +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ > +/* > + * Copyright 2013 Red Hat Inc. > + * > + * This program is free software; you can redistribute it and/or > + * modify it under the terms of the GNU General Public License as > + * published by the Free Software Foundation; either version 2 of > + * the License, or (at your option) any later version. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + * > + * Authors: Jérôme Glisse <jglisse@redhat.com> > + */ > +/* > + * This is a dummy driver to exercise the HMM (heterogeneous memory management) > + * API of the kernel. It allows a userspace program to expose its entire address > + * space through the HMM dummy driver file. > + */ > +#ifndef _UAPI_LINUX_HMM_DMIRROR_H > +#define _UAPI_LINUX_HMM_DMIRROR_H > + > +#include <linux/types.h> > +#include <linux/ioctl.h> > + > +/* > + * Structure to pass to the HMM test driver to mimic a device accessing > + * system memory and ZONE_DEVICE private memory through device page tables. > + * > + * @addr: (in) user address the device will read/write > + * @ptr: (in) user address where device data is copied to/from > + * @npages: (in) number of pages to read/write > + * @cpages: (out) number of pages copied > + * @faults: (out) number of device page faults seen > + */ > +struct hmm_dmirror_cmd { > + __u64 addr; > + __u64 ptr; > + __u64 npages; > + __u64 cpages; > + __u64 faults; > +}; > + > +/* Expose the address space of the calling process through hmm dummy dev file */ > +#define HMM_DMIRROR_READ _IOWR('H', 0x00, struct hmm_dmirror_cmd) > +#define HMM_DMIRROR_WRITE _IOWR('H', 0x01, struct hmm_dmirror_cmd) > +#define HMM_DMIRROR_MIGRATE _IOWR('H', 0x02, struct hmm_dmirror_cmd) > +#define HMM_DMIRROR_SNAPSHOT _IOWR('H', 0x03, struct hmm_dmirror_cmd) > + > +/* > + * Values returned in hmm_dmirror_cmd.ptr for HMM_DMIRROR_SNAPSHOT. > + * HMM_DMIRROR_PROT_ERROR: no valid mirror PTE for this page > + * HMM_DMIRROR_PROT_NONE: unpopulated PTE or PTE with no access > + * HMM_DMIRROR_PROT_READ: read-only PTE > + * HMM_DMIRROR_PROT_WRITE: read/write PTE > + * HMM_DMIRROR_PROT_ZERO: special read-only zero page > + * HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL: Migrated device private page on the > + * device the ioctl() is made > + * HMM_DMIRROR_PROT_DEV_PRIVATE_REMOTE: Migrated device private page on some > + * other device > + */ > +enum { > + HMM_DMIRROR_PROT_ERROR = 0xFF, > + HMM_DMIRROR_PROT_NONE = 0x00, > + HMM_DMIRROR_PROT_READ = 0x01, > + HMM_DMIRROR_PROT_WRITE = 0x02, > + HMM_DMIRROR_PROT_ZERO = 0x10, > + HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL = 0x20, > + HMM_DMIRROR_PROT_DEV_PRIVATE_REMOTE = 0x30, > +}; > + > +#endif /* _UAPI_LINUX_HMM_DMIRROR_H */ > diff --git a/tools/testing/selftests/vm/.gitignore b/tools/testing/selftests/vm/.gitignore > index 31b3c98b6d34..3054565b3f07 100644 > --- a/tools/testing/selftests/vm/.gitignore > +++ b/tools/testing/selftests/vm/.gitignore > @@ -14,3 +14,4 @@ virtual_address_range > gup_benchmark > va_128TBswitch > map_fixed_noreplace > +hmm-tests > diff --git a/tools/testing/selftests/vm/Makefile b/tools/testing/selftests/vm/Makefile > index 9534dc2bc929..5643cfb5e3d6 100644 > --- a/tools/testing/selftests/vm/Makefile > +++ b/tools/testing/selftests/vm/Makefile > @@ -5,6 +5,7 @@ CFLAGS = -Wall -I ../../../../usr/include $(EXTRA_CFLAGS) > LDLIBS = -lrt > TEST_GEN_FILES = compaction_test > TEST_GEN_FILES += gup_benchmark > +TEST_GEN_FILES += hmm-tests > TEST_GEN_FILES += hugepage-mmap > TEST_GEN_FILES += hugepage-shm > TEST_GEN_FILES += map_hugetlb > @@ -26,6 +27,8 @@ TEST_FILES := test_vmalloc.sh > KSFT_KHDR_INSTALL := 1 > include ../lib.mk > > +$(OUTPUT)/hmm-tests: LDLIBS += -lhugetlbfs -lpthread > + > $(OUTPUT)/userfaultfd: LDLIBS += -lpthread > > $(OUTPUT)/mlock-random-test: LDLIBS += -lcap > diff --git a/tools/testing/selftests/vm/config b/tools/testing/selftests/vm/config > index 1c0d76cb5adf..34cfab18e737 100644 > --- a/tools/testing/selftests/vm/config > +++ b/tools/testing/selftests/vm/config > @@ -1,2 +1,5 @@ > CONFIG_SYSVIPC=y > CONFIG_USERFAULTFD=y > +CONFIG_HMM_MIRROR=y > +CONFIG_DEVICE_PRIVATE=y > +CONFIG_HMM_DMIRROR=m > diff --git a/tools/testing/selftests/vm/hmm-tests.c b/tools/testing/selftests/vm/hmm-tests.c > new file mode 100644 > index 000000000000..f4ae6188fd0e > --- /dev/null > +++ b/tools/testing/selftests/vm/hmm-tests.c > @@ -0,0 +1,1311 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Copyright 2013 Red Hat Inc. > + * > + * This program is free software; you can redistribute it and/or > + * modify it under the terms of the GNU General Public License as > + * published by the Free Software Foundation; either version 2 of > + * the License, or (at your option) any later version. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + * > + * Authors: Jérôme Glisse <jglisse@redhat.com> > + */ > + > +/* > + * HMM stands for Heterogeneous Memory Management, it is a helper layer inside > + * the linux kernel to help device drivers mirror a process address space in > + * the device. This allows the device to use the same address space which > + * makes communication and data exchange a lot easier. > + * > + * This framework's sole purpose is to exercise various code paths inside > + * the kernel to make sure that HMM performs as expected and to flush out any > + * bugs. > + */ > + > +#include "../kselftest_harness.h" > + > +#include <errno.h> > +#include <fcntl.h> > +#include <stdio.h> > +#include <stdlib.h> > +#include <stdint.h> > +#include <unistd.h> > +#include <strings.h> > +#include <time.h> > +#include <pthread.h> > +#include <hugetlbfs.h> > +#include <sys/types.h> > +#include <sys/stat.h> > +#include <sys/mman.h> > +#include <sys/ioctl.h> > +#include <linux/hmm_dmirror.h> > + > +struct hmm_buffer { > + void *ptr; > + void *mirror; > + unsigned long size; > + int fd; > + uint64_t cpages; > + uint64_t faults; > +}; > + > +#define TWOMEG (1 << 21) > +#define HMM_BUFFER_SIZE (1024 << 12) > +#define HMM_PATH_MAX 64 > +#define NTIMES 256 > + > +#define ALIGN(x, a) (((x) + (a - 1)) & (~((a) - 1))) > + > +FIXTURE(hmm) > +{ > + int fd; > + unsigned int page_size; > + unsigned int page_shift; > +}; > + > +FIXTURE(hmm2) > +{ > + int fd0; > + int fd1; > + unsigned int page_size; > + unsigned int page_shift; > +}; > + > +static int hmm_open(int unit) > +{ > + char pathname[HMM_PATH_MAX]; > + int fd; > + > + snprintf(pathname, sizeof(pathname), "/dev/hmm_dmirror%d", unit); > + fd = open(pathname, O_RDWR, 0); > + if (fd < 0) > + fprintf(stderr, "could not open hmm dmirror driver (%s)\n", > + pathname); > + return fd; > +} > + > +FIXTURE_SETUP(hmm) > +{ > + self->page_size = sysconf(_SC_PAGE_SIZE); > + self->page_shift = ffs(self->page_size) - 1; > + > + self->fd = hmm_open(0); > + ASSERT_GE(self->fd, 0); > +} > + > +FIXTURE_SETUP(hmm2) > +{ > + self->page_size = sysconf(_SC_PAGE_SIZE); > + self->page_shift = ffs(self->page_size) - 1; > + > + self->fd0 = hmm_open(0); > + ASSERT_GE(self->fd0, 0); > + self->fd1 = hmm_open(1); > + ASSERT_GE(self->fd1, 0); > +} > + > +FIXTURE_TEARDOWN(hmm) > +{ > + int ret = close(self->fd); > + > + ASSERT_EQ(ret, 0); > + self->fd = -1; > +} > + > +FIXTURE_TEARDOWN(hmm2) > +{ > + int ret = close(self->fd0); > + > + ASSERT_EQ(ret, 0); > + self->fd0 = -1; > + > + ret = close(self->fd1); > + ASSERT_EQ(ret, 0); > + self->fd1 = -1; > +} > + > +static int hmm_dmirror_cmd(int fd, > + unsigned long request, > + struct hmm_buffer *buffer, > + unsigned long npages) > +{ > + struct hmm_dmirror_cmd cmd; > + int ret; > + > + /* Simulate a device reading system memory. */ > + cmd.addr = (__u64)buffer->ptr; > + cmd.ptr = (__u64)buffer->mirror; > + cmd.npages = npages; > + > + for (;;) { > + ret = ioctl(fd, request, &cmd); > + if (ret == 0) > + break; > + if (errno == EINTR) > + continue; > + return -errno; > + } > + buffer->cpages = cmd.cpages; > + buffer->faults = cmd.faults; > + > + return 0; > +} > + > +static void hmm_buffer_free(struct hmm_buffer *buffer) > +{ > + if (buffer == NULL) > + return; > + > + if (buffer->ptr) > + munmap(buffer->ptr, buffer->size); > + free(buffer->mirror); > + free(buffer); > +} > + > +/* > + * Create a temporary file that will be deleted on close. > + */ > +static int hmm_create_file(unsigned long size) > +{ > + char path[HMM_PATH_MAX]; > + int fd; > + > + strcpy(path, "/tmp"); > + fd = open(path, O_TMPFILE | O_EXCL | O_RDWR, 0600); > + if (fd >= 0) { > + int r; > + > + do { > + r = ftruncate(fd, size); > + } while (r == -1 && errno == EINTR); > + if (!r) > + return fd; > + close(fd); > + } > + return -1; > +} > + > +/* > + * Return a random unsigned number. > + */ > +static unsigned int hmm_random(void) > +{ > + static int fd = -1; > + unsigned int r; > + > + if (fd < 0) { > + fd = open("/dev/urandom", O_RDONLY); > + if (fd < 0) { > + fprintf(stderr, "%s:%d failed to open /dev/urandom\n", > + __FILE__, __LINE__); > + return ~0U; > + } > + } > + read(fd, &r, sizeof(r)); > + return r; > +} > + > +static void hmm_nanosleep(unsigned int n) > +{ > + struct timespec t; > + > + t.tv_sec = 0; > + t.tv_nsec = n; > + nanosleep(&t, NULL); > +} > + > +/* > + * Simple NULL test of device open/close. > + */ > +TEST_F(hmm, open_close) > +{ > +} > + > +/* > + * Read private anonymous memory. > + */ > +TEST_F(hmm, anon_read) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + int ret; > + int val; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* > + * Initialize buffer in system memory but leave the first two pages > + * zero (pte_none and pfn_zero). > + */ > + i = 2 * self->page_size / sizeof(*ptr); > + for (ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Set buffer permission to read-only. */ > + ret = mprotect(buffer->ptr, size, PROT_READ); > + ASSERT_EQ(ret, 0); > + > + /* Populate the CPU page table with a special zero page. */ > + val = *(int *)(buffer->ptr + self->page_size); > + ASSERT_EQ(val, 0); > + > + /* Simulate a device reading system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device read. */ > + ptr = buffer->mirror; > + for (i = 0; i < 2 * self->page_size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], 0); > + for (; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + hmm_buffer_free(buffer); > +} > + > +/* > + * Read private anonymous memory which has been protected with > + * mprotect() PROT_NONE. > + */ > +TEST_F(hmm, anon_read_prot) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + int ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize buffer in system memory. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Initialize mirror buffer so we can verify it isn't written. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ptr[i] = -i; > + > + /* Protect buffer from reading. */ > + ret = mprotect(buffer->ptr, size, PROT_NONE); > + ASSERT_EQ(ret, 0); > + > + /* Simulate a device reading system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, npages); > + ASSERT_EQ(ret, -EFAULT); > + > + /* Allow CPU to read the buffer so we can check it. */ > + ret = mprotect(buffer->ptr, size, PROT_READ); > + ASSERT_EQ(ret, 0); > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + /* Check what the device read. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], -i); > + > + hmm_buffer_free(buffer); > +} > + > +/* > + * Write private anonymous memory. > + */ > +TEST_F(hmm, anon_write) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + int ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize data that the device will write to buffer->ptr. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Simulate a device writing system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device wrote. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + hmm_buffer_free(buffer); > +} > + > +/* > + * Write private anonymous memory which has been protected with > + * mprotect() PROT_READ. > + */ > +TEST_F(hmm, anon_write_prot) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + int ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Simulate a device reading a zero page of memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, 1); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, 1); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Initialize data that the device will write to buffer->ptr. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Simulate a device writing system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); > + ASSERT_EQ(ret, -EPERM); > + > + /* Check what the device wrote. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], 0); > + > + /* Now allow writing and see that the zero page is replaced. */ > + ret = mprotect(buffer->ptr, size, PROT_WRITE | PROT_READ); > + ASSERT_EQ(ret, 0); > + > + /* Simulate a device writing system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device wrote. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + hmm_buffer_free(buffer); > +} > + > +/* > + * Check that a device writing an anonymous private mapping > + * will copy-on-write if a child process inherits the mapping. > + */ > +TEST_F(hmm, anon_write_child) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + pid_t pid; > + int child_fd; > + int ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize buffer->ptr so we can tell if it is written. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Initialize data that the device will write to buffer->ptr. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ptr[i] = -i; > + > + pid = fork(); > + if (pid == -1) > + ASSERT_EQ(pid, 0); > + if (pid != 0) { > + waitpid(pid, &ret, 0); > + ASSERT_EQ(WIFEXITED(ret), 1); > + > + /* Check that the parent's buffer did not change. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + return; > + } > + > + /* Check that we see the parent's values. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], -i); > + > + /* The child process needs its own mirror to its own mm. */ > + child_fd = hmm_open(0); > + ASSERT_GE(child_fd, 0); > + > + /* Simulate a device writing system memory. */ > + ret = hmm_dmirror_cmd(child_fd, HMM_DMIRROR_WRITE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device wrote. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], -i); > + > + close(child_fd); > + exit(0); > +} > + > +/* > + * Check that a device writing an anonymous shared mapping > + * will not copy-on-write if a child process inherits the mapping. > + */ > +TEST_F(hmm, anon_write_child_shared) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + pid_t pid; > + int child_fd; > + int ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_SHARED | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize buffer->ptr so we can tell if it is written. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Initialize data that the device will write to buffer->ptr. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ptr[i] = -i; > + > + pid = fork(); > + if (pid == -1) > + ASSERT_EQ(pid, 0); > + if (pid != 0) { > + waitpid(pid, &ret, 0); > + ASSERT_EQ(WIFEXITED(ret), 1); > + > + /* Check that the parent's buffer did change. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], -i); > + return; > + } > + > + /* Check that we see the parent's values. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], -i); > + > + /* The child process needs its own mirror to its own mm. */ > + child_fd = hmm_open(0); > + ASSERT_GE(child_fd, 0); > + > + /* Simulate a device writing system memory. */ > + ret = hmm_dmirror_cmd(child_fd, HMM_DMIRROR_WRITE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device wrote. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], -i); > + > + close(child_fd); > + exit(0); > +} > + > +/* > + * Write private anonymous huge page. > + */ > +TEST_F(hmm, anon_write_huge) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + void *old_ptr; > + void *map; > + int *ptr; > + int ret; > + > + size = 2 * TWOMEG; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + size = TWOMEG; > + npages = size >> self->page_shift; > + map = (void *)ALIGN((uintptr_t)buffer->ptr, size); > + ret = madvise(map, size, MADV_HUGEPAGE); > + ASSERT_EQ(ret, 0); > + old_ptr = buffer->ptr; > + buffer->ptr = map; > + > + /* Initialize data that the device will write to buffer->ptr. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Simulate a device writing system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device wrote. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + buffer->ptr = old_ptr; > + hmm_buffer_free(buffer); > +} > + > +/* > + * Write huge TLBFS page. > + */ > +TEST_F(hmm, anon_write_hugetlbfs) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + int ret; > + long pagesizes[4]; > + int n, idx; > + > + /* Skip test if we can't allocate a hugetlbfs page. */ > + > + n = gethugepagesizes(pagesizes, 4); > + if (n <= 0) > + return; > + for (idx = 0; --n > 0; ) { > + if (pagesizes[n] < pagesizes[idx]) > + idx = n; > + } > + size = ALIGN(TWOMEG, pagesizes[idx]); > + npages = size >> self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->ptr = get_hugepage_region(size, GHR_STRICT); > + if (buffer->ptr == NULL) { > + free(buffer); > + return; > + } > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + /* Initialize data that the device will write to buffer->ptr. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Simulate a device writing system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device wrote. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + free_hugepage_region(buffer->ptr); > + buffer->ptr = NULL; > + hmm_buffer_free(buffer); > +} > + > +/* > + * Read mmap'ed file memory. > + */ > +TEST_F(hmm, file_read) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + int ret; > + int fd; > + off_t off; > + ssize_t len; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + fd = hmm_create_file(size); > + ASSERT_GE(fd, 0); > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = fd; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + /* Write initial contents of the file. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + off = lseek(fd, 0, SEEK_SET); > + ASSERT_EQ(off, 0); > + len = write(fd, buffer->mirror, size); > + ASSERT_EQ(len, size); > + memset(buffer->mirror, 0, size); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ, > + MAP_SHARED, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Simulate a device reading system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device read. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + hmm_buffer_free(buffer); > +} > + > +/* > + * Write mmap'ed file memory. > + */ > +TEST_F(hmm, file_write) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + int ret; > + int fd; > + off_t off; > + ssize_t len; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + fd = hmm_create_file(size); > + ASSERT_GE(fd, 0); > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = fd; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_SHARED, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize data that the device will write to buffer->ptr. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Simulate a device writing system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device wrote. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + /* Check that the device also wrote the file. */ > + off = lseek(fd, 0, SEEK_SET); > + ASSERT_EQ(off, 0); > + len = read(fd, buffer->mirror, size); > + ASSERT_EQ(len, size); > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + hmm_buffer_free(buffer); > +} > + > +/* > + * Migrate anonymous memory to device private memory. > + */ > +TEST_F(hmm, migrate) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + int ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize buffer in system memory. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Migrate memory to device. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_MIGRATE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + > + /* Check what the device read. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + hmm_buffer_free(buffer); > +} > + > +/* > + * Migrate anonymous memory to device private memory and fault it back to system > + * memory. > + */ > +TEST_F(hmm, migrate_fault) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + int *ptr; > + int ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize buffer in system memory. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Migrate memory to device. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_MIGRATE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + > + /* Check what the device read. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + /* Fault pages back to system memory and check them. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + hmm_buffer_free(buffer); > +} > + > +/* > + * Try to migrate various memory types to device private memory. > + */ > +TEST_F(hmm2, migrate_mixed) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + int *ptr; > + unsigned char *p; > + int ret; > + int val; > + > + npages = 6; > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + /* Reserve a range of addresses. */ > + buffer->ptr = mmap(NULL, size, > + PROT_NONE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + p = buffer->ptr; > + > + /* Now try to migrate everything to device 1. */ > + ret = hmm_dmirror_cmd(self->fd1, HMM_DMIRROR_MIGRATE, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, 6); > + > + /* Punch a hole after the first page address. */ > + ret = munmap(buffer->ptr + self->page_size, self->page_size); > + ASSERT_EQ(ret, 0); > + > + /* We expect an error if the vma doesn't cover the range. */ > + ret = hmm_dmirror_cmd(self->fd1, HMM_DMIRROR_MIGRATE, buffer, 3); > + ASSERT_EQ(ret, -EINVAL); > + > + /* Page 2 will be a read-only zero page. */ > + ret = mprotect(buffer->ptr + 2 * self->page_size, self->page_size, > + PROT_READ); > + ASSERT_EQ(ret, 0); > + ptr = (int *)(buffer->ptr + 2 * self->page_size); > + val = *ptr + 3; > + ASSERT_EQ(val, 3); > + > + /* Page 3 will be read-only. */ > + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, > + PROT_READ | PROT_WRITE); > + ASSERT_EQ(ret, 0); > + ptr = (int *)(buffer->ptr + 3 * self->page_size); > + *ptr = val; > + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, > + PROT_READ); > + ASSERT_EQ(ret, 0); > + > + /* Page 4 will be read-write. */ > + ret = mprotect(buffer->ptr + 4 * self->page_size, self->page_size, > + PROT_READ | PROT_WRITE); > + ASSERT_EQ(ret, 0); > + ptr = (int *)(buffer->ptr + 4 * self->page_size); > + *ptr = val; > + > + /* Page 5 won't be migrated to device 0 because it's on device 1. */ > + buffer->ptr = p + 5 * self->page_size; > + ret = hmm_dmirror_cmd(self->fd0, HMM_DMIRROR_MIGRATE, buffer, 1); > + ASSERT_EQ(ret, -ENOENT); > + buffer->ptr = p; > + > + /* Now try to migrate pages 2-3 to device 1. */ > + buffer->ptr = p + 2 * self->page_size; > + ret = hmm_dmirror_cmd(self->fd1, HMM_DMIRROR_MIGRATE, buffer, 2); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, 2); > + buffer->ptr = p; > + > + hmm_buffer_free(buffer); > +} > + > +/* > + * Migrate anonymous memory to device private memory and fault it back to system > + * memory multiple times. > + */ > +TEST_F(hmm, migrate_multiple) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + unsigned long c; > + int *ptr; > + int ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + for (c = 0; c < NTIMES; c++) { > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize buffer in system memory. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ptr[i] = i; > + > + /* Migrate memory to device. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_MIGRATE, buffer, > + npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + > + /* Check what the device read. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + /* Fault pages back to system memory and check them. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i); > + > + hmm_buffer_free(buffer); > + } > +} > + > +/* > + * Read anonymous memory multiple times. > + */ > +TEST_F(hmm, anon_read_multiple) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + unsigned long i; > + unsigned long c; > + int *ptr; > + int ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + for (c = 0; c < NTIMES; c++) { > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize buffer in system memory. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ptr[i] = i + c; > + > + /* Simulate a device reading system memory. */ > + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, > + npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device read. */ > + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) > + ASSERT_EQ(ptr[i], i + c); > + > + hmm_buffer_free(buffer); > + } > +} > + > +void *unmap_buffer(void *p) > +{ > + struct hmm_buffer *buffer = p; > + > + /* Delay for a bit and then unmap buffer while it is being read. */ > + hmm_nanosleep(hmm_random() % 32000); > + munmap(buffer->ptr + buffer->size / 2, buffer->size / 2); > + buffer->ptr = NULL; > + > + return NULL; > +} > + > +/* > + * Try reading anonymous memory while it is being unmapped. > + */ > +TEST_F(hmm, anon_teardown) > +{ > + unsigned long npages; > + unsigned long size; > + unsigned long c; > + void *ret; > + > + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; > + ASSERT_NE(npages, 0); > + size = npages << self->page_shift; > + > + for (c = 0; c < NTIMES; ++c) { > + pthread_t thread; > + struct hmm_buffer *buffer; > + unsigned long i; > + int *ptr; > + int rc; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(size); > + ASSERT_NE(buffer->mirror, NULL); > + > + buffer->ptr = mmap(NULL, size, > + PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + > + /* Initialize buffer in system memory. */ > + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) > + ptr[i] = i + c; > + > + rc = pthread_create(&thread, NULL, unmap_buffer, buffer); > + ASSERT_EQ(rc, 0); > + > + /* Simulate a device reading system memory. */ > + rc = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, > + npages); > + if (rc == 0) { > + ASSERT_EQ(buffer->cpages, npages); > + ASSERT_EQ(buffer->faults, 1); > + > + /* Check what the device read. */ > + for (i = 0, ptr = buffer->mirror; > + i < size / sizeof(*ptr); > + ++i) > + ASSERT_EQ(ptr[i], i + c); > + } > + > + pthread_join(thread, &ret); > + hmm_buffer_free(buffer); > + } > +} > + > +/* > + * Test memory snapshot without faulting in pages accessed by the device. > + */ > +TEST_F(hmm2, snapshot) > +{ > + struct hmm_buffer *buffer; > + unsigned long npages; > + unsigned long size; > + int *ptr; > + unsigned char *p; > + unsigned char *m; > + int ret; > + int val; > + > + npages = 7; > + size = npages << self->page_shift; > + > + buffer = malloc(sizeof(*buffer)); > + ASSERT_NE(buffer, NULL); > + > + buffer->fd = -1; > + buffer->size = size; > + buffer->mirror = malloc(npages); > + ASSERT_NE(buffer->mirror, NULL); > + > + /* Reserve a range of addresses. */ > + buffer->ptr = mmap(NULL, size, > + PROT_NONE, > + MAP_PRIVATE | MAP_ANONYMOUS, > + buffer->fd, 0); > + ASSERT_NE(buffer->ptr, MAP_FAILED); > + p = buffer->ptr; > + > + /* Punch a hole after the first page address. */ > + ret = munmap(buffer->ptr + self->page_size, self->page_size); > + ASSERT_EQ(ret, 0); > + > + /* Page 2 will be read-only zero page. */ > + ret = mprotect(buffer->ptr + 2 * self->page_size, self->page_size, > + PROT_READ); > + ASSERT_EQ(ret, 0); > + ptr = (int *)(buffer->ptr + 2 * self->page_size); > + val = *ptr + 3; > + ASSERT_EQ(val, 3); > + > + /* Page 3 will be read-only. */ > + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, > + PROT_READ | PROT_WRITE); > + ASSERT_EQ(ret, 0); > + ptr = (int *)(buffer->ptr + 3 * self->page_size); > + *ptr = val; > + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, > + PROT_READ); > + ASSERT_EQ(ret, 0); > + > + /* Page 4-6 will be read-write. */ > + ret = mprotect(buffer->ptr + 4 * self->page_size, 3 * self->page_size, > + PROT_READ | PROT_WRITE); > + ASSERT_EQ(ret, 0); > + ptr = (int *)(buffer->ptr + 4 * self->page_size); > + *ptr = val; > + > + /* Page 5 will be migrated to device 0. */ > + buffer->ptr = p + 5 * self->page_size; > + ret = hmm_dmirror_cmd(self->fd0, HMM_DMIRROR_MIGRATE, buffer, 1); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, 1); > + > + /* Page 6 will be migrated to device 1. */ > + buffer->ptr = p + 6 * self->page_size; > + ret = hmm_dmirror_cmd(self->fd1, HMM_DMIRROR_MIGRATE, buffer, 1); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, 1); > + > + /* Simulate a device snapshotting CPU pagetables. */ > + buffer->ptr = p; > + ret = hmm_dmirror_cmd(self->fd0, HMM_DMIRROR_SNAPSHOT, buffer, npages); > + ASSERT_EQ(ret, 0); > + ASSERT_EQ(buffer->cpages, npages); > + > + /* Check what the device saw. */ > + m = buffer->mirror; > + ASSERT_EQ(m[0], HMM_DMIRROR_PROT_NONE); > + ASSERT_EQ(m[1], HMM_DMIRROR_PROT_NONE); > + ASSERT_EQ(m[2], HMM_DMIRROR_PROT_ZERO | HMM_DMIRROR_PROT_READ); > + ASSERT_EQ(m[3], HMM_DMIRROR_PROT_READ); > + ASSERT_EQ(m[4], HMM_DMIRROR_PROT_WRITE); > + ASSERT_EQ(m[5], HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL | > + HMM_DMIRROR_PROT_WRITE); > + ASSERT_EQ(m[6], HMM_DMIRROR_PROT_DEV_PRIVATE_REMOTE | > + HMM_DMIRROR_PROT_WRITE); > + > + hmm_buffer_free(buffer); > +} > + > +TEST_HARNESS_MAIN > diff --git a/tools/testing/selftests/vm/run_vmtests b/tools/testing/selftests/vm/run_vmtests > index 951c507a27f7..634cfefdaffd 100755 > --- a/tools/testing/selftests/vm/run_vmtests > +++ b/tools/testing/selftests/vm/run_vmtests > @@ -227,4 +227,20 @@ else > exitcode=1 > fi > > +echo "------------------------------------" > +echo "running HMM smoke test" > +echo "------------------------------------" > +./test_hmm.sh smoke > +ret_val=$? > + > +if [ $ret_val -eq 0 ]; then > + echo "[PASS]" > +elif [ $ret_val -eq $ksft_skip ]; then > + echo "[SKIP]" > + exitcode=$ksft_skip > +else > + echo "[FAIL]" > + exitcode=1 > +fi > + > exit $exitcode > diff --git a/tools/testing/selftests/vm/test_hmm.sh b/tools/testing/selftests/vm/test_hmm.sh > new file mode 100755 > index 000000000000..268d32752045 > --- /dev/null > +++ b/tools/testing/selftests/vm/test_hmm.sh > @@ -0,0 +1,97 @@ > +#!/bin/bash > +# SPDX-License-Identifier: GPL-2.0 > +# > +# Copyright (C) 2018 Uladzislau Rezki (Sony) <urezki@gmail.com> > +# > +# This is a test script for the kernel test driver to analyse vmalloc > +# allocator. Therefore it is just a kernel module loader. You can specify > +# and pass different parameters in order to: > +# a) analyse performance of vmalloc allocations; > +# b) stressing and stability check of vmalloc subsystem. > + > +TEST_NAME="test_hmm" > +DRIVER="hmm_dmirror" > + > +# 1 if fails > +exitcode=1 > + > +# Kselftest framework requirement - SKIP code is 4. > +ksft_skip=4 > + > +check_test_requirements() > +{ > + uid=$(id -u) > + if [ $uid -ne 0 ]; then > + echo "$0: Must be run as root" > + exit $ksft_skip > + fi > + > + if ! which modprobe > /dev/null 2>&1; then > + echo "$0: You need modprobe installed" > + exit $ksft_skip > + fi > + > + if ! modinfo $DRIVER > /dev/null 2>&1; then > + echo "$0: You must have the following enabled in your kernel:" > + echo "CONFIG_HMM_DMIRROR=m" > + exit $ksft_skip > + fi > +} > + > +load_driver() > +{ > + modprobe $DRIVER > /dev/null 2>&1 > + if [ $? == 0 ]; then > + major=$(awk "\$2==\"HMM_DMIRROR\" {print \$1}" /proc/devices) > + mknod /dev/hmm_dmirror0 c $major 0 > + mknod /dev/hmm_dmirror1 c $major 1 > + fi > +} > + > +unload_driver() > +{ > + modprobe -r $DRIVER > /dev/null 2>&1 > + rm -f /dev/hmm_dmirror? > +} > + > +run_smoke() > +{ > + echo "Running smoke test. Note, this test provides basic coverage." > + > + load_driver > + ./hmm-tests > + unload_driver > +} > + > +usage() > +{ > + echo -n "Usage: $0" > + echo > + echo "Example usage:" > + echo > + echo "# Shows help message" > + echo "./${TEST_NAME}.sh" > + echo > + echo "# Smoke testing" > + echo "./${TEST_NAME}.sh smoke" > + echo > + exit 0 > +} > + > +function run_test() > +{ > + if [ $# -eq 0 ]; then > + usage > + else > + if [ "$1" = "smoke" ]; then > + run_smoke > + else > + usage > + fi > + fi > +} > + > +check_test_requirements > +run_test $@ > + > +exit 0 > -- > 2.20.1 >
On 10/23/19 1:28 PM, Jerome Glisse wrote: > On Wed, Oct 23, 2019 at 12:55:15PM -0700, Ralph Campbell wrote: >> Add self tests for HMM. >> >> Signed-off-by: Ralph Campbell <rcampbell@nvidia.com> > > You can add my signoff > > Signed-off-by: Jérôme Glisse <jglisse@redhat.com> Thanks!
On Wed, Oct 23, 2019 at 12:55:15PM -0700, Ralph Campbell wrote: > Add self tests for HMM. > > Signed-off-by: Ralph Campbell <rcampbell@nvidia.com> > --- > MAINTAINERS | 3 + > drivers/char/Kconfig | 11 + > drivers/char/Makefile | 1 + > drivers/char/hmm_dmirror.c | 1566 ++++++++++++++++++++++++ > include/Kbuild | 1 + > include/uapi/linux/hmm_dmirror.h | 74 ++ > tools/testing/selftests/vm/.gitignore | 1 + > tools/testing/selftests/vm/Makefile | 3 + > tools/testing/selftests/vm/config | 3 + > tools/testing/selftests/vm/hmm-tests.c | 1311 ++++++++++++++++++++ > tools/testing/selftests/vm/run_vmtests | 16 + > tools/testing/selftests/vm/test_hmm.sh | 97 ++ > 12 files changed, 3087 insertions(+) > create mode 100644 drivers/char/hmm_dmirror.c > create mode 100644 include/uapi/linux/hmm_dmirror.h > create mode 100644 tools/testing/selftests/vm/hmm-tests.c > create mode 100755 tools/testing/selftests/vm/test_hmm.sh This is really big, it would be nice to get a comment from the various kernel testing folks if this approach makes sense with the test frameworks. Do we have other drivers that are only intended to be used by selftests? Frankly, I'm not super excited about the idea of a 'test driver', it seems more logical for testing to have some way for a test harness to call hmm_range_fault() under various conditions and check the results? It seems especially over-complicated to use a full page table layout for this, wouldn't something simple like an xarray be good enough for test purposes? > +/* > + * Below are the file operation for the dmirror device file. Only ioctl matters. > + * > + * Note this is highly specific to the dmirror device driver and should not be > + * construed as an example on how to design the API a real device driver would > + * expose to userspace. > + */ > +static ssize_t dmirror_fops_read(struct file *filp, > + char __user *buf, > + size_t count, > + loff_t *ppos) > +{ > + return -EINVAL; > +} > + > +static ssize_t dmirror_fops_write(struct file *filp, > + const char __user *buf, > + size_t count, > + loff_t *ppos) > +{ > + return -EINVAL; > +} > + > +static int dmirror_fops_mmap(struct file *filp, struct vm_area_struct *vma) > +{ > + /* Forbid mmap of the dmirror device file. */ > + return -EINVAL; > +} I'm pretty sure these can just be left as NULL in the fops? > +static int dmirror_fault(struct dmirror *dmirror, > + unsigned long start, > + unsigned long end, > + bool write) > +{ > + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; > + unsigned long addr; > + unsigned long next; > + uint64_t pfns[64]; > + struct hmm_range range = { > + .pfns = pfns, > + .flags = dmirror_hmm_flags, > + .values = dmirror_hmm_values, > + .pfn_shift = DPT_SHIFT, > + .pfn_flags_mask = ~(dmirror_hmm_flags[HMM_PFN_VALID] | > + dmirror_hmm_flags[HMM_PFN_WRITE]), > + .default_flags = dmirror_hmm_flags[HMM_PFN_VALID] | > + (write ? dmirror_hmm_flags[HMM_PFN_WRITE] : 0), > + }; > + int ret = 0; > + > + for (addr = start; addr < end; ) { > + long count; > + > + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); > + range.start = addr; > + range.end = next; > + > + down_read(&mm->mmap_sem); > + > + ret = hmm_range_register(&range, &dmirror->mirror); > + if (ret) { > + up_read(&mm->mmap_sem); > + break; > + } > + > + if (!hmm_range_wait_until_valid(&range, > + DMIRROR_RANGE_FAULT_TIMEOUT)) { > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + continue; > + } > + > + count = hmm_range_fault(&range, 0); > + if (count < 0) { > + ret = count; > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + break; > + } > + > + if (!hmm_range_valid(&range)) { There is no 'driver lock' being held here, how does this work? Shouldn't it hold dmirror->mutex for this sequence? > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + continue; > + } > + mutex_lock(&dmirror->mutex); > + ret = dmirror_pt_walk(dmirror, dmirror_do_fault, > + addr, next, &range, true); > + mutex_unlock(&dmirror->mutex); Ie move it down into this block > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + if (ret) > + break; > + > + addr = next; > + } > + > + return ret; > +} > +static int dmirror_read(struct dmirror *dmirror, > + struct hmm_dmirror_cmd *cmd) > +{ Why not just use pread()/pwrite() for this instead of an ioctl? > + struct dmirror_bounce bounce; > + unsigned long start, end; > + unsigned long size = cmd->npages << PAGE_SHIFT; > + int ret; > + > + start = cmd->addr; > + end = start + size; > + if (end < start) > + return -EINVAL; > + > + ret = dmirror_bounce_init(&bounce, start, size); > + if (ret) > + return ret; > + > +static int dmirror_snapshot(struct dmirror *dmirror, > + struct hmm_dmirror_cmd *cmd) > +{ > + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; > + unsigned long start, end; > + unsigned long size = cmd->npages << PAGE_SHIFT; > + unsigned long addr; > + unsigned long next; > + uint64_t pfns[64]; > + unsigned char perm[64]; > + char __user *uptr; > + struct hmm_range range = { > + .pfns = pfns, > + .flags = dmirror_hmm_flags, > + .values = dmirror_hmm_values, > + .pfn_shift = DPT_SHIFT, > + .pfn_flags_mask = ~0ULL, > + }; > + int ret = 0; > + > + start = cmd->addr; > + end = start + size; > + uptr = (void __user *)cmd->ptr; > + > + for (addr = start; addr < end; ) { > + long count; > + unsigned long i; > + unsigned long n; > + > + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); > + range.start = addr; > + range.end = next; > + > + down_read(&mm->mmap_sem); > + > + ret = hmm_range_register(&range, &dmirror->mirror); > + if (ret) { > + up_read(&mm->mmap_sem); > + break; > + } > + > + if (!hmm_range_wait_until_valid(&range, > + DMIRROR_RANGE_FAULT_TIMEOUT)) { > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + continue; > + } > + > + count = hmm_range_fault(&range, HMM_FAULT_SNAPSHOT); > + if (count < 0) { > + ret = count; > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + if (ret == -EBUSY) > + continue; > + break; > + } > + > + if (!hmm_range_valid(&range)) { Same as for dmirror_fault > + hmm_range_unregister(&range); > + up_read(&mm->mmap_sem); > + continue; > + } > + > + n = (next - addr) >> PAGE_SHIFT; > + for (i = 0; i < n; i++) > + dmirror_mkentry(dmirror, &range, perm + i, pfns[i]); Is this missing locking too? > +static int dmirror_remove(struct platform_device *pdev) > +{ > + /* all probe actions are unwound by devm */ > + return 0; > +} > + > +static struct platform_driver dmirror_device_driver = { > + .probe = dmirror_probe, > + .remove = dmirror_remove, > + .driver = { > + .name = "HMM_DMIRROR", > + }, > +}; This presence of a platform_driver and device is very confusing. I'm sure Greg KH would object to this as a misuse of platform drivers. A platform device isn't needed to create a char dev, so what is this for? > diff --git a/include/Kbuild b/include/Kbuild > index ffba79483cc5..6ffb44a45957 100644 > --- a/include/Kbuild > +++ b/include/Kbuild > @@ -1063,6 +1063,7 @@ header-test- += uapi/linux/coda_psdev.h > header-test- += uapi/linux/errqueue.h > header-test- += uapi/linux/eventpoll.h > header-test- += uapi/linux/hdlc/ioctl.h > +header-test- += uapi/linux/hmm_dmirror.h Why? This list should only be updated if the header is broken in some way. > diff --git a/tools/testing/selftests/vm/hmm-tests.c b/tools/testing/selftests/vm/hmm-tests.c > new file mode 100644 > index 000000000000..f4ae6188fd0e > --- /dev/null > +++ b/tools/testing/selftests/vm/hmm-tests.c > @@ -0,0 +1,1311 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * Copyright 2013 Red Hat Inc. > + * > + * This program is free software; you can redistribute it and/or > + * modify it under the terms of the GNU General Public License as > + * published by the Free Software Foundation; either version 2 of > + * the License, or (at your option) any later version. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. btw, I think if a SPDX is present I don't think the license text is required, just the copyright. I think these tests should also study the various case of invoke pte_hole, ie faulting/snappshotting before/after a vma, or across a vma range with a hole, etc, etc. Jason
On 10/29/19 10:58 AM, Jason Gunthorpe wrote: > On Wed, Oct 23, 2019 at 12:55:15PM -0700, Ralph Campbell wrote: >> Add self tests for HMM. >> >> Signed-off-by: Ralph Campbell <rcampbell@nvidia.com> > >> --- >> MAINTAINERS | 3 + >> drivers/char/Kconfig | 11 + >> drivers/char/Makefile | 1 + >> drivers/char/hmm_dmirror.c | 1566 ++++++++++++++++++++++++ >> include/Kbuild | 1 + >> include/uapi/linux/hmm_dmirror.h | 74 ++ >> tools/testing/selftests/vm/.gitignore | 1 + >> tools/testing/selftests/vm/Makefile | 3 + >> tools/testing/selftests/vm/config | 3 + >> tools/testing/selftests/vm/hmm-tests.c | 1311 ++++++++++++++++++++ >> tools/testing/selftests/vm/run_vmtests | 16 + >> tools/testing/selftests/vm/test_hmm.sh | 97 ++ >> 12 files changed, 3087 insertions(+) >> create mode 100644 drivers/char/hmm_dmirror.c >> create mode 100644 include/uapi/linux/hmm_dmirror.h >> create mode 100644 tools/testing/selftests/vm/hmm-tests.c >> create mode 100755 tools/testing/selftests/vm/test_hmm.sh > > This is really big, it would be nice to get a comment from the various > kernel testing folks if this approach makes sense with the test > frameworks. Do we have other drivers that are only intended to be used > by selftests? > > Frankly, I'm not super excited about the idea of a 'test driver', it > seems more logical for testing to have some way for a test harness to > call hmm_range_fault() under various conditions and check the results? test_vmalloc.sh at least uses a test module(s). > It seems especially over-complicated to use a full page table layout > for this, wouldn't something simple like an xarray be good enough for > test purposes? Possibly. A page table is really just a lookup table from virtual address to pfn/page. Part of the rationale was to mimic what a real device might do. >> +/* >> + * Below are the file operation for the dmirror device file. Only ioctl matters. >> + * >> + * Note this is highly specific to the dmirror device driver and should not be >> + * construed as an example on how to design the API a real device driver would >> + * expose to userspace. >> + */ >> +static ssize_t dmirror_fops_read(struct file *filp, >> + char __user *buf, >> + size_t count, >> + loff_t *ppos) >> +{ >> + return -EINVAL; >> +} >> + >> +static ssize_t dmirror_fops_write(struct file *filp, >> + const char __user *buf, >> + size_t count, >> + loff_t *ppos) >> +{ >> + return -EINVAL; >> +} >> + >> +static int dmirror_fops_mmap(struct file *filp, struct vm_area_struct *vma) >> +{ >> + /* Forbid mmap of the dmirror device file. */ >> + return -EINVAL; >> +} > > I'm pretty sure these can just be left as NULL in the fops? I think so. >> +static int dmirror_fault(struct dmirror *dmirror, >> + unsigned long start, >> + unsigned long end, >> + bool write) >> +{ >> + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; >> + unsigned long addr; >> + unsigned long next; >> + uint64_t pfns[64]; >> + struct hmm_range range = { >> + .pfns = pfns, >> + .flags = dmirror_hmm_flags, >> + .values = dmirror_hmm_values, >> + .pfn_shift = DPT_SHIFT, >> + .pfn_flags_mask = ~(dmirror_hmm_flags[HMM_PFN_VALID] | >> + dmirror_hmm_flags[HMM_PFN_WRITE]), >> + .default_flags = dmirror_hmm_flags[HMM_PFN_VALID] | >> + (write ? dmirror_hmm_flags[HMM_PFN_WRITE] : 0), >> + }; >> + int ret = 0; >> + >> + for (addr = start; addr < end; ) { >> + long count; >> + >> + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); >> + range.start = addr; >> + range.end = next; >> + >> + down_read(&mm->mmap_sem); >> + >> + ret = hmm_range_register(&range, &dmirror->mirror); >> + if (ret) { >> + up_read(&mm->mmap_sem); >> + break; >> + } >> + >> + if (!hmm_range_wait_until_valid(&range, >> + DMIRROR_RANGE_FAULT_TIMEOUT)) { >> + hmm_range_unregister(&range); >> + up_read(&mm->mmap_sem); >> + continue; >> + } >> + >> + count = hmm_range_fault(&range, 0); >> + if (count < 0) { >> + ret = count; >> + hmm_range_unregister(&range); >> + up_read(&mm->mmap_sem); >> + break; >> + } >> + >> + if (!hmm_range_valid(&range)) { > > There is no 'driver lock' being held here, how does this work? > Shouldn't it hold dmirror->mutex for this sequence? I have a modified version of this driver that's based on your series removing hmm_mirror_register() which uses a mutex. Otherwise, it looks similar to the changes in nouveau. >> + hmm_range_unregister(&range); >> + up_read(&mm->mmap_sem); >> + continue; >> + } >> + mutex_lock(&dmirror->mutex); >> + ret = dmirror_pt_walk(dmirror, dmirror_do_fault, >> + addr, next, &range, true); >> + mutex_unlock(&dmirror->mutex); > > Ie move it down into this block > >> + hmm_range_unregister(&range); >> + up_read(&mm->mmap_sem); >> + if (ret) >> + break; >> + >> + addr = next; >> + } >> + >> + return ret; >> +} > >> +static int dmirror_read(struct dmirror *dmirror, >> + struct hmm_dmirror_cmd *cmd) >> +{ > > Why not just use pread()/pwrite() for this instead of an ioctl? pread()/pwrite() could certainly be implemented. I think the idea was that the read/write is actually the "device" doing read/write and making that clearly different from a program reading/writing the device. Also, the ioctl() allows information about what faults or events happened during the operation. I only have number of pages and number of page faults returned at the moment, but one of Jerome's version of this driver had other counters being returned. >> + struct dmirror_bounce bounce; >> + unsigned long start, end; >> + unsigned long size = cmd->npages << PAGE_SHIFT; >> + int ret; >> + >> + start = cmd->addr; >> + end = start + size; >> + if (end < start) >> + return -EINVAL; >> + >> + ret = dmirror_bounce_init(&bounce, start, size); >> + if (ret) >> + return ret; >> + >> +static int dmirror_snapshot(struct dmirror *dmirror, >> + struct hmm_dmirror_cmd *cmd) >> +{ >> + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; >> + unsigned long start, end; >> + unsigned long size = cmd->npages << PAGE_SHIFT; >> + unsigned long addr; >> + unsigned long next; >> + uint64_t pfns[64]; >> + unsigned char perm[64]; >> + char __user *uptr; >> + struct hmm_range range = { >> + .pfns = pfns, >> + .flags = dmirror_hmm_flags, >> + .values = dmirror_hmm_values, >> + .pfn_shift = DPT_SHIFT, >> + .pfn_flags_mask = ~0ULL, >> + }; >> + int ret = 0; >> + >> + start = cmd->addr; >> + end = start + size; >> + uptr = (void __user *)cmd->ptr; >> + >> + for (addr = start; addr < end; ) { >> + long count; >> + unsigned long i; >> + unsigned long n; >> + >> + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); >> + range.start = addr; >> + range.end = next; >> + >> + down_read(&mm->mmap_sem); >> + >> + ret = hmm_range_register(&range, &dmirror->mirror); >> + if (ret) { >> + up_read(&mm->mmap_sem); >> + break; >> + } >> + >> + if (!hmm_range_wait_until_valid(&range, >> + DMIRROR_RANGE_FAULT_TIMEOUT)) { >> + hmm_range_unregister(&range); >> + up_read(&mm->mmap_sem); >> + continue; >> + } >> + >> + count = hmm_range_fault(&range, HMM_FAULT_SNAPSHOT); >> + if (count < 0) { >> + ret = count; >> + hmm_range_unregister(&range); >> + up_read(&mm->mmap_sem); >> + if (ret == -EBUSY) >> + continue; >> + break; >> + } >> + >> + if (!hmm_range_valid(&range)) { > > Same as for dmirror_fault > >> + hmm_range_unregister(&range); >> + up_read(&mm->mmap_sem); >> + continue; >> + } >> + >> + n = (next - addr) >> PAGE_SHIFT; >> + for (i = 0; i < n; i++) >> + dmirror_mkentry(dmirror, &range, perm + i, pfns[i]); > > Is this missing locking too? Yes. It's in the updated version as mentioned above. >> +static int dmirror_remove(struct platform_device *pdev) >> +{ >> + /* all probe actions are unwound by devm */ >> + return 0; >> +} >> + >> +static struct platform_driver dmirror_device_driver = { >> + .probe = dmirror_probe, >> + .remove = dmirror_remove, >> + .driver = { >> + .name = "HMM_DMIRROR", >> + }, >> +}; > > This presence of a platform_driver and device is very confusing. I'm > sure Greg KH would object to this as a misuse of platform drivers. > > A platform device isn't needed to create a char dev, so what is this for? The devm_request_free_mem_region() and devm_memremap_pages() calls for creating the ZONE_DEVICE private pages tie into the devm* clean up framework. I thought a platform_driver was the simplest way to also be able to call devm_add_action_or_reset() to clean up on module unload and be compatible with the private page clean up. >> diff --git a/include/Kbuild b/include/Kbuild >> index ffba79483cc5..6ffb44a45957 100644 >> --- a/include/Kbuild >> +++ b/include/Kbuild >> @@ -1063,6 +1063,7 @@ header-test- += uapi/linux/coda_psdev.h >> header-test- += uapi/linux/errqueue.h >> header-test- += uapi/linux/eventpoll.h >> header-test- += uapi/linux/hdlc/ioctl.h >> +header-test- += uapi/linux/hmm_dmirror.h > > Why? This list should only be updated if the header is broken in some > way. Should this be in include/linux/ instead? I wasn't sure where the "right" place was to put the header. > >> diff --git a/tools/testing/selftests/vm/hmm-tests.c b/tools/testing/selftests/vm/hmm-tests.c >> new file mode 100644 >> index 000000000000..f4ae6188fd0e >> --- /dev/null >> +++ b/tools/testing/selftests/vm/hmm-tests.c >> @@ -0,0 +1,1311 @@ >> +// SPDX-License-Identifier: GPL-2.0 >> +/* >> + * Copyright 2013 Red Hat Inc. >> + * >> + * This program is free software; you can redistribute it and/or >> + * modify it under the terms of the GNU General Public License as >> + * published by the Free Software Foundation; either version 2 of >> + * the License, or (at your option) any later version. >> + * >> + * This program is distributed in the hope that it will be useful, >> + * but WITHOUT ANY WARRANTY; without even the implied warranty of >> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the >> + * GNU General Public License for more details. > > btw, I think if a SPDX is present I don't think the license text is > required, just the copyright. Since I was starting from Jerome's HMM test driver, I didn't want to delete any of the original copyright text. If Jerome is OK with just the SPDX header, that's OK with me. > I think these tests should also study the various case of invoke > pte_hole, ie faulting/snappshotting before/after a vma, or across a > vma range with a hole, etc, etc. > > Jason > There are tests for vma hole, pte_none(), zero page, and normal page. Nothing stress testing races, just set up the mmap() and test once. I can add more test cases if you have something specific in mind.
On Tue, Oct 29, 2019 at 02:16:05PM -0700, Ralph Campbell wrote: > > Frankly, I'm not super excited about the idea of a 'test driver', it > > seems more logical for testing to have some way for a test harness to > > call hmm_range_fault() under various conditions and check the results? > > test_vmalloc.sh at least uses a test module(s). Well, that is good, is it also under drivers/char? It kind feels like it should not be there... > > It seems especially over-complicated to use a full page table layout > > for this, wouldn't something simple like an xarray be good enough for > > test purposes? > > Possibly. A page table is really just a lookup table from virtual address > to pfn/page. Part of the rationale was to mimic what a real device > might do. Well, but the details of the page table layout don't see really important to this testing, IMHO. > > > + for (addr = start; addr < end; ) { > > > + long count; > > > + > > > + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); > > > + range.start = addr; > > > + range.end = next; > > > + > > > + down_read(&mm->mmap_sem); Also, did we get a mmget() before doing this down_read? > > > + > > > + ret = hmm_range_register(&range, &dmirror->mirror); > > > + if (ret) { > > > + up_read(&mm->mmap_sem); > > > + break; > > > + } > > > + > > > + if (!hmm_range_wait_until_valid(&range, > > > + DMIRROR_RANGE_FAULT_TIMEOUT)) { > > > + hmm_range_unregister(&range); > > > + up_read(&mm->mmap_sem); > > > + continue; > > > + } > > > + > > > + count = hmm_range_fault(&range, 0); > > > + if (count < 0) { > > > + ret = count; > > > + hmm_range_unregister(&range); > > > + up_read(&mm->mmap_sem); > > > + break; > > > + } > > > + > > > + if (!hmm_range_valid(&range)) { > > > > There is no 'driver lock' being held here, how does this work? > > Shouldn't it hold dmirror->mutex for this sequence? > > I have a modified version of this driver that's based on your series > removing hmm_mirror_register() which uses a mutex. > Otherwise, it looks similar to the changes in nouveau. Well, that locking pattern is required even for original hmm calls.. > > > +static int dmirror_read(struct dmirror *dmirror, > > > + struct hmm_dmirror_cmd *cmd) > > > +{ > > > > Why not just use pread()/pwrite() for this instead of an ioctl? > > pread()/pwrite() could certainly be implemented. > I think the idea was that the read/write is actually the "device" > doing read/write and making that clearly different from a program > reading/writing the device. Also, the ioctl() allows information > about what faults or events happened during the operation. I only > have number of pages and number of page faults returned at the moment, > but one of Jerome's version of this driver had other counters being > returned. Makes sense I guess > > > +static struct platform_driver dmirror_device_driver = { > > > + .probe = dmirror_probe, > > > + .remove = dmirror_remove, > > > + .driver = { > > > + .name = "HMM_DMIRROR", > > > + }, > > > +}; > > > > This presence of a platform_driver and device is very confusing. I'm > > sure Greg KH would object to this as a misuse of platform drivers. > > > > A platform device isn't needed to create a char dev, so what is this for? > > The devm_request_free_mem_region() and devm_memremap_pages() calls for > creating the ZONE_DEVICE private pages tie into the devm* clean up framework. > I thought a platform_driver was the simplest way to also be able to call > devm_add_action_or_reset() to clean up on module unload and be compatible > with the private page clean up. IIRC Christoph recently fixed things so there was a non devm version of these functions. Certainly we should not be making fake platform_devices just to call devm. There is also a struct device inside the cdev, maybe that could be arrange to be devm compatible if it was *really* needed. > > > diff --git a/include/Kbuild b/include/Kbuild > > > index ffba79483cc5..6ffb44a45957 100644 > > > +++ b/include/Kbuild > > > @@ -1063,6 +1063,7 @@ header-test- += uapi/linux/coda_psdev.h > > > header-test- += uapi/linux/errqueue.h > > > header-test- += uapi/linux/eventpoll.h > > > header-test- += uapi/linux/hdlc/ioctl.h > > > +header-test- += uapi/linux/hmm_dmirror.h > > > > Why? This list should only be updated if the header is broken in some > > way. > > Should this be in include/linux/ instead? > I wasn't sure where the "right" place was to put the header. No, it is right, it just shouldn't be in this makefile. Jason
On Tue, 2019-10-29 at 17:58 +0000, Jason Gunthorpe wrote: > On Wed, Oct 23, 2019 at 12:55:15PM -0700, Ralph Campbell wrote: > > Add self tests for HMM. > > > > Signed-off-by: Ralph Campbell <rcampbell@nvidia.com> > > --- > > MAINTAINERS | 3 + > > drivers/char/Kconfig | 11 + > > drivers/char/Makefile | 1 + > > drivers/char/hmm_dmirror.c | 1566 ++++++++++++++++++++++++ > > include/Kbuild | 1 + > > include/uapi/linux/hmm_dmirror.h | 74 ++ > > tools/testing/selftests/vm/.gitignore | 1 + > > tools/testing/selftests/vm/Makefile | 3 + > > tools/testing/selftests/vm/config | 3 + > > tools/testing/selftests/vm/hmm-tests.c | 1311 ++++++++++++++++++++ > > tools/testing/selftests/vm/run_vmtests | 16 + > > tools/testing/selftests/vm/test_hmm.sh | 97 ++ > > 12 files changed, 3087 insertions(+) > > create mode 100644 drivers/char/hmm_dmirror.c > > create mode 100644 include/uapi/linux/hmm_dmirror.h > > create mode 100644 tools/testing/selftests/vm/hmm-tests.c > > create mode 100755 tools/testing/selftests/vm/test_hmm.sh > > This is really big, it would be nice to get a comment from the various > kernel testing folks if this approach makes sense with the test > frameworks. Do we have other drivers that are only intended to be used > by selftests? > > Frankly, I'm not super excited about the idea of a 'test driver', it > seems more logical for testing to have some way for a test harness to > call hmm_range_fault() under various conditions and check the results? Not a big fan of those selftests either. Could it be saner to use the new KUnit framework for those instead?
On 10/29/19 4:12 PM, Jason Gunthorpe wrote: > On Tue, Oct 29, 2019 at 02:16:05PM -0700, Ralph Campbell wrote: > >>> Frankly, I'm not super excited about the idea of a 'test driver', it >>> seems more logical for testing to have some way for a test harness to >>> call hmm_range_fault() under various conditions and check the results? >> >> test_vmalloc.sh at least uses a test module(s). > > Well, that is good, is it also under drivers/char? It kind feels like > it should not be there... I think most of the test modules live in lib/ but I wasn't sure that was the right place for the HMM test driver. If you think that is better, I can easily move it. >>> It seems especially over-complicated to use a full page table layout >>> for this, wouldn't something simple like an xarray be good enough for >>> test purposes? >> >> Possibly. A page table is really just a lookup table from virtual address >> to pfn/page. Part of the rationale was to mimic what a real device >> might do. > > Well, but the details of the page table layout don't see really > important to this testing, IMHO. One problem with XArray is that on 32-bit machines the value would need to be u64 to hold a pfn which won't fit in a ULONG_MAX. I guess we could make the driver 64-bit only. >>>> + for (addr = start; addr < end; ) { >>>> + long count; >>>> + >>>> + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); >>>> + range.start = addr; >>>> + range.end = next; >>>> + >>>> + down_read(&mm->mmap_sem); > > Also, did we get a mmget() before doing this down_read? > >>>> + >>>> + ret = hmm_range_register(&range, &dmirror->mirror); >>>> + if (ret) { >>>> + up_read(&mm->mmap_sem); >>>> + break; >>>> + } >>>> + >>>> + if (!hmm_range_wait_until_valid(&range, >>>> + DMIRROR_RANGE_FAULT_TIMEOUT)) { >>>> + hmm_range_unregister(&range); >>>> + up_read(&mm->mmap_sem); >>>> + continue; >>>> + } >>>> + >>>> + count = hmm_range_fault(&range, 0); >>>> + if (count < 0) { >>>> + ret = count; >>>> + hmm_range_unregister(&range); >>>> + up_read(&mm->mmap_sem); >>>> + break; >>>> + } >>>> + >>>> + if (!hmm_range_valid(&range)) { >>> >>> There is no 'driver lock' being held here, how does this work? >>> Shouldn't it hold dmirror->mutex for this sequence? >> >> I have a modified version of this driver that's based on your series >> removing hmm_mirror_register() which uses a mutex. >> Otherwise, it looks similar to the changes in nouveau. > > Well, that locking pattern is required even for original hmm calls.. Will be fixed in v4. > >>>> +static int dmirror_read(struct dmirror *dmirror, >>>> + struct hmm_dmirror_cmd *cmd) >>>> +{ >>> >>> Why not just use pread()/pwrite() for this instead of an ioctl? >> >> pread()/pwrite() could certainly be implemented. >> I think the idea was that the read/write is actually the "device" >> doing read/write and making that clearly different from a program >> reading/writing the device. Also, the ioctl() allows information >> about what faults or events happened during the operation. I only >> have number of pages and number of page faults returned at the moment, >> but one of Jerome's version of this driver had other counters being >> returned. > > Makes sense I guess > >>>> +static struct platform_driver dmirror_device_driver = { >>>> + .probe = dmirror_probe, >>>> + .remove = dmirror_remove, >>>> + .driver = { >>>> + .name = "HMM_DMIRROR", >>>> + }, >>>> +}; >>> >>> This presence of a platform_driver and device is very confusing. I'm >>> sure Greg KH would object to this as a misuse of platform drivers. >>> >>> A platform device isn't needed to create a char dev, so what is this for? >> >> The devm_request_free_mem_region() and devm_memremap_pages() calls for >> creating the ZONE_DEVICE private pages tie into the devm* clean up framework. >> I thought a platform_driver was the simplest way to also be able to call >> devm_add_action_or_reset() to clean up on module unload and be compatible >> with the private page clean up. > > IIRC Christoph recently fixed things so there was a non devm version > of these functions. Certainly we should not be making fake > platform_devices just to call devm. > > There is also a struct device inside the cdev, maybe that could be > arrange to be devm compatible if it was *really* needed. Will be fixed in v4. >>>> diff --git a/include/Kbuild b/include/Kbuild >>>> index ffba79483cc5..6ffb44a45957 100644 >>>> +++ b/include/Kbuild >>>> @@ -1063,6 +1063,7 @@ header-test- += uapi/linux/coda_psdev.h >>>> header-test- += uapi/linux/errqueue.h >>>> header-test- += uapi/linux/eventpoll.h >>>> header-test- += uapi/linux/hdlc/ioctl.h >>>> +header-test- += uapi/linux/hmm_dmirror.h >>> >>> Why? This list should only be updated if the header is broken in some >>> way. >> >> Should this be in include/linux/ instead? >> I wasn't sure where the "right" place was to put the header. > > No, it is right, it just shouldn't be in this makefile. > > Jason Will be fixed in v4. Thanks for the review, the code is much simpler now.
On Wed, Oct 30, 2019 at 05:14:30PM -0700, Ralph Campbell wrote: > > Well, that is good, is it also under drivers/char? It kind feels like > > it should not be there... > > I think most of the test modules live in lib/ but I wasn't sure that > was the right place for the HMM test driver. > If you think that is better, I can easily move it. It would be good to get the various test people involved in this, I really don't know. > > > > It seems especially over-complicated to use a full page table layout > > > > for this, wouldn't something simple like an xarray be good enough for > > > > test purposes? > > > > > > Possibly. A page table is really just a lookup table from virtual address > > > to pfn/page. Part of the rationale was to mimic what a real device > > > might do. > > > > Well, but the details of the page table layout don't see really > > important to this testing, IMHO. > > One problem with XArray is that on 32-bit machines the value would > need to be u64 to hold a pfn which won't fit in a ULONG_MAX. > I guess we could make the driver 64-bit only. Why would a 32 bit machine need a 64 bit pfn? Jason
On 10/31/19 5:42 AM, Jason Gunthorpe wrote: > On Wed, Oct 30, 2019 at 05:14:30PM -0700, Ralph Campbell wrote: > >>> Well, that is good, is it also under drivers/char? It kind feels like >>> it should not be there... >> >> I think most of the test modules live in lib/ but I wasn't sure that >> was the right place for the HMM test driver. >> If you think that is better, I can easily move it. > > It would be good to get the various test people involved in this, I > really don't know. OK. >>>>> It seems especially over-complicated to use a full page table layout >>>>> for this, wouldn't something simple like an xarray be good enough for >>>>> test purposes? >>>> >>>> Possibly. A page table is really just a lookup table from virtual address >>>> to pfn/page. Part of the rationale was to mimic what a real device >>>> might do. >>> >>> Well, but the details of the page table layout don't see really >>> important to this testing, IMHO. >> >> One problem with XArray is that on 32-bit machines the value would >> need to be u64 to hold a pfn which won't fit in a ULONG_MAX. >> I guess we could make the driver 64-bit only. > > Why would a 32 bit machine need a 64 bit pfn? > > Jason > On x86, Physical Address Extension (PAE) uses a 64 bit PTE. See arch/x86/include/asm/pgtable_32_types.h which includes arch/x86/include/asm/pgtable-3level_types.h.
On Thu, Oct 31, 2019 at 10:28:12AM -0700, Ralph Campbell wrote: > > > > > > It seems especially over-complicated to use a full page table layout > > > > > > for this, wouldn't something simple like an xarray be good enough for > > > > > > test purposes? > > > > > > > > > > Possibly. A page table is really just a lookup table from virtual address > > > > > to pfn/page. Part of the rationale was to mimic what a real device > > > > > might do. > > > > > > > > Well, but the details of the page table layout don't see really > > > > important to this testing, IMHO. > > > > > > One problem with XArray is that on 32-bit machines the value would > > > need to be u64 to hold a pfn which won't fit in a ULONG_MAX. > > > I guess we could make the driver 64-bit only. > > > > Why would a 32 bit machine need a 64 bit pfn? > > > > On x86, Physical Address Extension (PAE) uses a 64 bit PTE. > See arch/x86/include/asm/pgtable_32_types.h which includes > arch/x86/include/asm/pgtable-3level_types.h. That is the content of the PTE, not the address of the PTE. In this case the xarray index is the 'virtual' address of the fictional device and it can easily be 32 bits with no problem Jason
On 10/31/19 10:34 AM, Jason Gunthorpe wrote: > On Thu, Oct 31, 2019 at 10:28:12AM -0700, Ralph Campbell wrote: >>>>>>> It seems especially over-complicated to use a full page table layout >>>>>>> for this, wouldn't something simple like an xarray be good enough for >>>>>>> test purposes? >>>>>> >>>>>> Possibly. A page table is really just a lookup table from virtual address >>>>>> to pfn/page. Part of the rationale was to mimic what a real device >>>>>> might do. >>>>> >>>>> Well, but the details of the page table layout don't see really >>>>> important to this testing, IMHO. >>>> >>>> One problem with XArray is that on 32-bit machines the value would >>>> need to be u64 to hold a pfn which won't fit in a ULONG_MAX. >>>> I guess we could make the driver 64-bit only. >>> >>> Why would a 32 bit machine need a 64 bit pfn? >>> >> >> On x86, Physical Address Extension (PAE) uses a 64 bit PTE. >> See arch/x86/include/asm/pgtable_32_types.h which includes >> arch/x86/include/asm/pgtable-3level_types.h. > > That is the content of the PTE, not the address of the PTE. In this > case the xarray index is the 'virtual' address of the fictional device > and it can easily be 32 bits with no problem > > Jason > Oh, I see. You mean use a 32-bit user virtual address for the index and store a pointer to the 64-bit PTE which of course would be 32 bit. That should work. I was stuck on thinking the PTE needed to be stored.
diff --git a/MAINTAINERS b/MAINTAINERS index 296de2b51c83..9890b6b8eea0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7427,8 +7427,11 @@ M: Jérôme Glisse <jglisse@redhat.com> L: linux-mm@kvack.org S: Maintained F: mm/hmm* +F: drivers/char/hmm* F: include/linux/hmm* +F: include/uapi/linux/hmm* F: Documentation/vm/hmm.rst +F: tools/testing/selftests/vm/*hmm* HOST AP DRIVER M: Jouni Malinen <j@w1.fi> diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index df0fc997dc3e..cc8ddb99550d 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -535,6 +535,17 @@ config ADI and SSM (Silicon Secured Memory). Intended consumers of this driver include crash and makedumpfile. +config HMM_DMIRROR + tristate "HMM driver for testing Heterogeneous Memory Management" + depends on HMM_MIRROR + depends on DEVICE_PRIVATE + help + This is a pseudo device driver solely for testing HMM. + Say Y here if you want to build the HMM test driver. + Doing so will allow you to run tools/testing/selftest/vm/hmm-tests. + + If in doubt, say "N". + endmenu config RANDOM_TRUST_CPU diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 7c5ea6f9df14..d4a168c0c138 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -52,3 +52,4 @@ js-rtc-y = rtc.o obj-$(CONFIG_XILLYBUS) += xillybus/ obj-$(CONFIG_POWERNV_OP_PANEL) += powernv-op-panel.o obj-$(CONFIG_ADI) += adi.o +obj-$(CONFIG_HMM_DMIRROR) += hmm_dmirror.o diff --git a/drivers/char/hmm_dmirror.c b/drivers/char/hmm_dmirror.c new file mode 100644 index 000000000000..5a1ed34e72e1 --- /dev/null +++ b/drivers/char/hmm_dmirror.c @@ -0,0 +1,1566 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright 2013 Red Hat Inc. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2 of + * the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Authors: Jérôme Glisse <jglisse@redhat.com> + */ +/* + * This is a driver to exercice the HMM (heterogeneous memory management) + * mirror and zone device private memory migration APIs of the kernel. + * Userspace programs can register with the driver to mirror their own address + * space and can use the device to read/write any valid virtual address. + * + * In some ways it can also serve as an example driver for people wanting to use + * HMM inside their own device driver. + */ +#include <linux/init.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/cdev.h> +#include <linux/device.h> +#include <linux/mutex.h> +#include <linux/rwsem.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/highmem.h> +#include <linux/delay.h> +#include <linux/pagemap.h> +#include <linux/hmm.h> +#include <linux/vmalloc.h> +#include <linux/swap.h> +#include <linux/swapops.h> +#include <linux/sched/mm.h> +#include <linux/platform_device.h> + +#include <uapi/linux/hmm_dmirror.h> + +#define DMIRROR_NDEVICES 2 +#define DMIRROR_RANGE_FAULT_TIMEOUT 1000 +#define DEVMEM_CHUNK_SIZE (256 * 1024 * 1024U) +#define DEVMEM_CHUNKS_RESERVE 16 + +static const struct dev_pagemap_ops dmirror_devmem_ops; +static dev_t dmirror_dev; +static struct platform_device *dmirror_platform_devices[DMIRROR_NDEVICES]; +static struct page *dmirror_zero_page; + +struct dmirror_device; + +struct dmirror_bounce { + void *ptr; + unsigned long size; + unsigned long addr; + unsigned long cpages; +}; + +#define DPT_SHIFT PAGE_SHIFT +#define DPT_VALID (1UL << 0) +#define DPT_WRITE (1UL << 1) +#define DPT_DPAGE (1UL << 2) +#define DPT_ZPAGE 0x20UL + +const uint64_t dmirror_hmm_flags[HMM_PFN_FLAG_MAX] = { + [HMM_PFN_VALID] = DPT_VALID, + [HMM_PFN_WRITE] = DPT_WRITE, + [HMM_PFN_DEVICE_PRIVATE] = DPT_DPAGE, +}; + +static const uint64_t dmirror_hmm_values[HMM_PFN_VALUE_MAX] = { + [HMM_PFN_NONE] = 0, + [HMM_PFN_ERROR] = 0x10, + [HMM_PFN_SPECIAL] = 0x10, +}; + +struct dmirror_pt { + u64 pgd[PTRS_PER_PGD]; + struct rw_semaphore lock; +}; + +/* + * Data attached to the open device file. + * Note that it might be shared after a fork(). + */ +struct dmirror { + struct hmm_mirror mirror; + struct dmirror_device *mdevice; + struct dmirror_pt pt; + struct mutex mutex; +}; + +/* + * ZONE_DEVICE pages for migration and simulating device memory. + */ +struct dmirror_chunk { + struct dev_pagemap pagemap; + struct dmirror_device *mdevice; +}; + +/* + * Per device data. + */ +struct dmirror_device { + struct cdev cdevice; + struct hmm_devmem *devmem; + struct platform_device *pdevice; + + unsigned int devmem_capacity; + unsigned int devmem_count; + struct dmirror_chunk **devmem_chunks; + struct mutex devmem_lock; /* protects the above */ + + unsigned long calloc; + unsigned long cfree; + struct page *free_pages; + spinlock_t lock; /* protects the above */ +}; + +static inline unsigned long dmirror_pt_pgd(unsigned long addr) +{ + return (addr >> PGDIR_SHIFT) & (PTRS_PER_PGD - 1); +} + +static inline unsigned long dmirror_pt_pud(unsigned long addr) +{ + return (addr >> PUD_SHIFT) & (PTRS_PER_PUD - 1); +} + +static inline unsigned long dmirror_pt_pmd(unsigned long addr) +{ + return (addr >> PMD_SHIFT) & (PTRS_PER_PMD - 1); +} + +static inline unsigned long dmirror_pt_pte(unsigned long addr) +{ + return (addr >> PAGE_SHIFT) & (PTRS_PER_PTE - 1); +} + +static inline struct page *dmirror_pt_page(u64 *dptep) +{ + u64 dpte = *dptep; + + if (dpte == DPT_ZPAGE) + return dmirror_zero_page; + if (!(dpte & DPT_VALID)) + return NULL; + return pfn_to_page((u64)dpte >> DPT_SHIFT); +} + +static inline struct page *dmirror_pt_page_write(u64 *dptep) +{ + u64 dpte = *dptep; + + if (!(dpte & DPT_VALID) || !(dpte & DPT_WRITE)) + return NULL; + return pfn_to_page((u64)dpte >> DPT_SHIFT); +} + +static inline u64 dmirror_pt_from_page(struct page *page) +{ + if (!page) + return 0; + return (page_to_pfn(page) << DPT_SHIFT) | DPT_VALID; +} + +static struct page *populate_pt(struct dmirror *dmirror, u64 *dptep) +{ + struct page *page; + + /* + * Since we don't free page tables until the process exits, + * we can unlock and relock without the page table being freed + * from under us. + */ + mutex_unlock(&dmirror->mutex); + page = alloc_page(GFP_HIGHUSER | __GFP_ZERO); + mutex_lock(&dmirror->mutex); + if (page) { + if (unlikely(*dptep)) { + __free_page(page); + page = dmirror_pt_page(dptep); + } else + *dptep = dmirror_pt_from_page(page); + } else if (*dptep) + page = dmirror_pt_page(dptep); + return page; +} + +static inline unsigned long dmirror_pt_pud_end(unsigned long addr) +{ + return (addr & PGDIR_MASK) + ((unsigned long)PTRS_PER_PUD << PUD_SHIFT); +} + +static inline unsigned long dmirror_pt_pmd_end(unsigned long addr) +{ + return (addr & PUD_MASK) + ((unsigned long)PTRS_PER_PMD << PMD_SHIFT); +} + +static inline unsigned long dmirror_pt_pte_end(unsigned long addr) +{ + return (addr & PMD_MASK) + ((unsigned long)PTRS_PER_PTE << PAGE_SHIFT); +} + +typedef int (*dmirror_walk_cb_t)(struct dmirror *dmirror, + unsigned long start, + unsigned long end, + u64 *dptep, + void *private); + +static int dmirror_pt_walk(struct dmirror *dmirror, + dmirror_walk_cb_t cb, + unsigned long start, + unsigned long end, + void *private, + bool populate) +{ + u64 *dpgdp = &dmirror->pt.pgd[dmirror_pt_pgd(start)]; + unsigned long addr; + int ret = -ENOENT; + + for (addr = start; addr < end; dpgdp++) { + u64 *dpudp; + unsigned long pud_end; + struct page *pud_page; + + pud_end = min(end, dmirror_pt_pud_end(addr)); + pud_page = dmirror_pt_page(dpgdp); + if (!pud_page) { + if (!populate) { + addr = pud_end; + continue; + } + pud_page = populate_pt(dmirror, dpgdp); + if (!pud_page) + return -ENOMEM; + } + dpudp = kmap(pud_page); + dpudp += dmirror_pt_pud(addr); + for (; addr != pud_end; dpudp++) { + u64 *dpmdp; + unsigned long pmd_end; + struct page *pmd_page; + + pmd_end = min(end, dmirror_pt_pmd_end(addr)); + pmd_page = dmirror_pt_page(dpudp); + if (!pmd_page) { + if (!populate) { + addr = pmd_end; + continue; + } + pmd_page = populate_pt(dmirror, dpudp); + if (!pmd_page) { + kunmap(pud_page); + return -ENOMEM; + } + } + dpmdp = kmap(pmd_page); + dpmdp += dmirror_pt_pmd(addr); + for (; addr != pmd_end; dpmdp++) { + u64 *dptep; + unsigned long pte_end; + struct page *pte_page; + + pte_end = min(end, dmirror_pt_pte_end(addr)); + pte_page = dmirror_pt_page(dpmdp); + if (!pte_page) { + if (!populate) { + addr = pte_end; + continue; + } + pte_page = populate_pt(dmirror, dpmdp); + if (!pte_page) { + kunmap(pmd_page); + kunmap(pud_page); + return -ENOMEM; + } + } + if (!cb) { + addr = pte_end; + continue; + } + dptep = kmap(pte_page); + dptep += dmirror_pt_pte(addr); + ret = cb(dmirror, addr, pte_end, dptep, + private); + kunmap(pte_page); + if (ret) { + kunmap(pmd_page); + kunmap(pud_page); + return ret; + } + addr = pte_end; + } + kunmap(pmd_page); + addr = pmd_end; + } + kunmap(pud_page); + addr = pud_end; + } + + return ret; +} + +static void dmirror_pt_free(struct dmirror *dmirror) +{ + u64 *dpgdp = dmirror->pt.pgd; + + for (; dpgdp != dmirror->pt.pgd + PTRS_PER_PGD; dpgdp++) { + u64 *dpudp, *dpudp_orig; + u64 *dpudp_end; + struct page *pud_page; + + pud_page = dmirror_pt_page(dpgdp); + if (!pud_page) + continue; + + dpudp_orig = kmap_atomic(pud_page); + dpudp = dpudp_orig; + dpudp_end = dpudp + PTRS_PER_PUD; + for (; dpudp != dpudp_end; dpudp++) { + u64 *dpmdp, *dpmdp_orig; + u64 *dpmdp_end; + struct page *pmd_page; + + pmd_page = dmirror_pt_page(dpudp); + if (!pmd_page) + continue; + + dpmdp_orig = kmap_atomic(pmd_page); + dpmdp = dpmdp_orig; + dpmdp_end = dpmdp + PTRS_PER_PMD; + for (; dpmdp != dpmdp_end; dpmdp++) { + struct page *pte_page; + + pte_page = dmirror_pt_page(dpmdp); + if (!pte_page) + continue; + + *dpmdp = 0; + __free_page(pte_page); + } + kunmap_atomic(dpmdp_orig); + *dpudp = 0; + __free_page(pmd_page); + } + kunmap_atomic(dpudp_orig); + *dpgdp = 0; + __free_page(pud_page); + } +} + +static int dmirror_bounce_init(struct dmirror_bounce *bounce, + unsigned long addr, + unsigned long size) +{ + bounce->addr = addr; + bounce->size = size; + bounce->cpages = 0; + bounce->ptr = vmalloc(size); + if (!bounce->ptr) + return -ENOMEM; + return 0; +} + +static int dmirror_bounce_copy_from(struct dmirror_bounce *bounce, + unsigned long addr) +{ + unsigned long end = addr + bounce->size; + char __user *uptr = (void __user *)addr; + void *ptr = bounce->ptr; + + for (; addr < end; addr += PAGE_SIZE, ptr += PAGE_SIZE, + uptr += PAGE_SIZE) { + int ret; + + ret = copy_from_user(ptr, uptr, PAGE_SIZE); + if (ret) + return ret; + } + + return 0; +} + +static int dmirror_bounce_copy_to(struct dmirror_bounce *bounce, + unsigned long addr) +{ + unsigned long end = addr + bounce->size; + char __user *uptr = (void __user *)addr; + void *ptr = bounce->ptr; + + for (; addr < end; addr += PAGE_SIZE, ptr += PAGE_SIZE, + uptr += PAGE_SIZE) { + int ret; + + ret = copy_to_user(uptr, ptr, PAGE_SIZE); + if (ret) + return ret; + } + + return 0; +} + +static void dmirror_bounce_fini(struct dmirror_bounce *bounce) +{ + vfree(bounce->ptr); +} + +static int dmirror_do_update(struct dmirror *dmirror, + unsigned long addr, + unsigned long end, + u64 *dptep, + void *private) +{ + /* + * The page table doesn't hold references to pages since it relies on + * the mmu notifier to clear pointers when they become stale. + * Therefore, it is OK to just clear the pte. + */ + for (; addr < end; addr += PAGE_SIZE, ++dptep) + *dptep = 0; + + return 0; +} + +static int dmirror_update(struct hmm_mirror *mirror, + const struct mmu_notifier_range *update) +{ + struct dmirror *dmirror = container_of(mirror, struct dmirror, mirror); + + if (mmu_notifier_range_blockable(update)) + mutex_lock(&dmirror->mutex); + else if (!mutex_trylock(&dmirror->mutex)) + return -EAGAIN; + + dmirror_pt_walk(dmirror, dmirror_do_update, update->start, + update->end, NULL, false); + mutex_unlock(&dmirror->mutex); + return 0; +} + +static const struct hmm_mirror_ops dmirror_ops = { + .sync_cpu_device_pagetables = &dmirror_update, +}; + +/* + * dmirror_new() - allocate and initialize dmirror struct. + * + * @mdevice: The device this mirror is associated with. + * @filp: The active device file descriptor this mirror is associated with. + */ +static struct dmirror *dmirror_new(struct dmirror_device *mdevice) +{ + struct mm_struct *mm = get_task_mm(current); + struct dmirror *dmirror; + int ret; + + if (!mm) + goto err; + + /* Mirror this process address space */ + dmirror = kzalloc(sizeof(*dmirror), GFP_KERNEL); + if (dmirror == NULL) + goto err_mmput; + + dmirror->mdevice = mdevice; + dmirror->mirror.ops = &dmirror_ops; + mutex_init(&dmirror->mutex); + + down_write(&mm->mmap_sem); + ret = hmm_mirror_register(&dmirror->mirror, mm); + up_write(&mm->mmap_sem); + if (ret) + goto err_free; + + mmput(mm); + return dmirror; + +err_free: + kfree(dmirror); +err_mmput: + mmput(mm); +err: + return NULL; +} + +static void dmirror_del(struct dmirror *dmirror) +{ + hmm_mirror_unregister(&dmirror->mirror); + dmirror_pt_free(dmirror); + kfree(dmirror); +} + +/* + * Below are the file operation for the dmirror device file. Only ioctl matters. + * + * Note this is highly specific to the dmirror device driver and should not be + * construed as an example on how to design the API a real device driver would + * expose to userspace. + */ +static ssize_t dmirror_fops_read(struct file *filp, + char __user *buf, + size_t count, + loff_t *ppos) +{ + return -EINVAL; +} + +static ssize_t dmirror_fops_write(struct file *filp, + const char __user *buf, + size_t count, + loff_t *ppos) +{ + return -EINVAL; +} + +static int dmirror_fops_mmap(struct file *filp, struct vm_area_struct *vma) +{ + /* Forbid mmap of the dmirror device file. */ + return -EINVAL; +} + +static int dmirror_fops_open(struct inode *inode, struct file *filp) +{ + struct cdev *cdev = inode->i_cdev; + struct dmirror_device *mdevice; + struct dmirror *dmirror; + + /* No exclusive opens. */ + if (filp->f_flags & O_EXCL) + return -EINVAL; + + mdevice = container_of(cdev, struct dmirror_device, cdevice); + dmirror = dmirror_new(mdevice); + if (!dmirror) + return -ENOMEM; + + /* Only the first open registers the address space. */ + mutex_lock(&mdevice->devmem_lock); + if (filp->private_data) + goto err_busy; + filp->private_data = dmirror; + mutex_unlock(&mdevice->devmem_lock); + + return 0; + +err_busy: + mutex_unlock(&mdevice->devmem_lock); + dmirror_del(dmirror); + return -EBUSY; +} + +static int dmirror_fops_release(struct inode *inode, struct file *filp) +{ + struct dmirror *dmirror = filp->private_data; + + if (!dmirror) + return 0; + + dmirror_del(dmirror); + filp->private_data = NULL; + + return 0; +} + +static inline struct dmirror_device *dmirror_page_to_device(struct page *page) + +{ + struct dmirror_chunk *devmem; + + devmem = container_of(page->pgmap, struct dmirror_chunk, pagemap); + return devmem->mdevice; +} + +static bool dmirror_device_is_mine(struct dmirror_device *mdevice, + struct page *page) +{ + if (!is_zone_device_page(page)) + return false; + return page->pgmap->ops == &dmirror_devmem_ops && + dmirror_page_to_device(page) == mdevice; +} + +static int dmirror_do_fault(struct dmirror *dmirror, + unsigned long addr, + unsigned long end, + u64 *dptep, + void *private) +{ + struct hmm_range *range = private; + unsigned long idx = (addr - range->start) >> PAGE_SHIFT; + uint64_t *pfns = range->pfns; + + for (; addr < end; addr += PAGE_SIZE, ++dptep, ++idx) { + struct page *page; + u64 dpte; + + /* + * HMM_PFN_ERROR is returned if it is accessing invalid memory + * either because of memory error (hardware detected memory + * corruption) or more likely because of truncate on mmap + * file. + */ + if (pfns[idx] == range->values[HMM_PFN_ERROR]) + return -EFAULT; + /* + * The only special PFN HMM returns is the read-only zero page + * which doesn't have a matching struct page. + */ + if (pfns[idx] == range->values[HMM_PFN_SPECIAL]) { + *dptep = DPT_ZPAGE; + continue; + } + if (!(pfns[idx] & range->flags[HMM_PFN_VALID])) + return -EFAULT; + page = hmm_device_entry_to_page(range, pfns[idx]); + /* We asked for pages to be populated but check anyway. */ + if (!page) + return -EFAULT; + dpte = dmirror_pt_from_page(page); + if (is_zone_device_page(page)) { + if (!dmirror_device_is_mine(dmirror->mdevice, page)) + continue; + dpte |= DPT_DPAGE; + } + if (pfns[idx] & range->flags[HMM_PFN_WRITE]) + dpte |= DPT_WRITE; + else if (range->default_flags & range->flags[HMM_PFN_WRITE]) + return -EFAULT; + *dptep = dpte; + } + + return 0; +} + +static int dmirror_fault(struct dmirror *dmirror, + unsigned long start, + unsigned long end, + bool write) +{ + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; + unsigned long addr; + unsigned long next; + uint64_t pfns[64]; + struct hmm_range range = { + .pfns = pfns, + .flags = dmirror_hmm_flags, + .values = dmirror_hmm_values, + .pfn_shift = DPT_SHIFT, + .pfn_flags_mask = ~(dmirror_hmm_flags[HMM_PFN_VALID] | + dmirror_hmm_flags[HMM_PFN_WRITE]), + .default_flags = dmirror_hmm_flags[HMM_PFN_VALID] | + (write ? dmirror_hmm_flags[HMM_PFN_WRITE] : 0), + }; + int ret = 0; + + for (addr = start; addr < end; ) { + long count; + + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); + range.start = addr; + range.end = next; + + down_read(&mm->mmap_sem); + + ret = hmm_range_register(&range, &dmirror->mirror); + if (ret) { + up_read(&mm->mmap_sem); + break; + } + + if (!hmm_range_wait_until_valid(&range, + DMIRROR_RANGE_FAULT_TIMEOUT)) { + hmm_range_unregister(&range); + up_read(&mm->mmap_sem); + continue; + } + + count = hmm_range_fault(&range, 0); + if (count < 0) { + ret = count; + hmm_range_unregister(&range); + up_read(&mm->mmap_sem); + break; + } + + if (!hmm_range_valid(&range)) { + hmm_range_unregister(&range); + up_read(&mm->mmap_sem); + continue; + } + mutex_lock(&dmirror->mutex); + ret = dmirror_pt_walk(dmirror, dmirror_do_fault, + addr, next, &range, true); + mutex_unlock(&dmirror->mutex); + hmm_range_unregister(&range); + up_read(&mm->mmap_sem); + if (ret) + break; + + addr = next; + } + + return ret; +} + +static int dmirror_do_read(struct dmirror *dmirror, + unsigned long addr, + unsigned long end, + u64 *dptep, + void *private) +{ + struct dmirror_bounce *bounce = private; + void *ptr; + + ptr = bounce->ptr + ((addr - bounce->addr) & PAGE_MASK); + + for (; addr < end; addr += PAGE_SIZE, ++dptep) { + struct page *page; + void *tmp; + + page = dmirror_pt_page(dptep); + if (!page) + return -ENOENT; + + tmp = kmap(page); + memcpy(ptr, tmp, PAGE_SIZE); + kunmap(page); + + ptr += PAGE_SIZE; + bounce->cpages++; + } + + return 0; +} + +static int dmirror_read(struct dmirror *dmirror, + struct hmm_dmirror_cmd *cmd) +{ + struct dmirror_bounce bounce; + unsigned long start, end; + unsigned long size = cmd->npages << PAGE_SHIFT; + int ret; + + start = cmd->addr; + end = start + size; + if (end < start) + return -EINVAL; + + ret = dmirror_bounce_init(&bounce, start, size); + if (ret) + return ret; + +again: + mutex_lock(&dmirror->mutex); + ret = dmirror_pt_walk(dmirror, dmirror_do_read, start, end, &bounce, + false); + mutex_unlock(&dmirror->mutex); + if (ret == 0) + ret = dmirror_bounce_copy_to(&bounce, cmd->ptr); + else if (ret == -ENOENT) { + start = cmd->addr + (bounce.cpages << PAGE_SHIFT); + ret = dmirror_fault(dmirror, start, end, false); + if (ret == 0) { + cmd->faults++; + goto again; + } + } + + cmd->cpages = bounce.cpages; + dmirror_bounce_fini(&bounce); + return ret; +} + +static int dmirror_do_write(struct dmirror *dmirror, + unsigned long addr, + unsigned long end, + u64 *dptep, + void *private) +{ + struct dmirror_bounce *bounce = private; + void *ptr; + + ptr = bounce->ptr + ((addr - bounce->addr) & PAGE_MASK); + + for (; addr < end; addr += PAGE_SIZE, ++dptep) { + struct page *page; + void *tmp; + + page = dmirror_pt_page_write(dptep); + if (!page) + return -ENOENT; + + tmp = kmap(page); + memcpy(tmp, ptr, PAGE_SIZE); + kunmap(page); + + ptr += PAGE_SIZE; + bounce->cpages++; + } + + return 0; +} + +static int dmirror_write(struct dmirror *dmirror, + struct hmm_dmirror_cmd *cmd) +{ + struct dmirror_bounce bounce; + unsigned long start, end; + unsigned long size = cmd->npages << PAGE_SHIFT; + int ret; + + start = cmd->addr; + end = start + size; + if (end < start) + return -EINVAL; + + ret = dmirror_bounce_init(&bounce, start, size); + if (ret) + return ret; + ret = dmirror_bounce_copy_from(&bounce, cmd->ptr); + if (ret) + return ret; + +again: + mutex_lock(&dmirror->mutex); + ret = dmirror_pt_walk(dmirror, dmirror_do_write, + start, end, &bounce, false); + mutex_unlock(&dmirror->mutex); + if (ret == -ENOENT) { + start = cmd->addr + (bounce.cpages << PAGE_SHIFT); + ret = dmirror_fault(dmirror, start, end, true); + if (ret == 0) { + cmd->faults++; + goto again; + } + } + + cmd->cpages = bounce.cpages; + dmirror_bounce_fini(&bounce); + return ret; +} + +static bool dmirror_allocate_chunk(struct dmirror_device *mdevice, + struct page **ppage) +{ + struct dmirror_chunk *devmem; + struct resource *res; + unsigned long pfn; + unsigned long pfn_first; + unsigned long pfn_last; + void *ptr; + + mutex_lock(&mdevice->devmem_lock); + + if (mdevice->devmem_count == mdevice->devmem_capacity) { + struct dmirror_chunk **new_chunks; + unsigned int new_capacity; + + new_capacity = mdevice->devmem_capacity + + DEVMEM_CHUNKS_RESERVE; + new_chunks = krealloc(mdevice->devmem_chunks, + sizeof(new_chunks[0]) * new_capacity, + GFP_KERNEL); + if (!new_chunks) + goto err; + mdevice->devmem_capacity = new_capacity; + mdevice->devmem_chunks = new_chunks; + } + + res = devm_request_free_mem_region(&mdevice->pdevice->dev, + &iomem_resource, DEVMEM_CHUNK_SIZE); + if (IS_ERR(res)) + goto err; + + devmem = kzalloc(sizeof(*devmem), GFP_KERNEL); + if (!devmem) + goto err; + + devmem->pagemap.type = MEMORY_DEVICE_PRIVATE; + devmem->pagemap.res = *res; + devmem->pagemap.ops = &dmirror_devmem_ops; + ptr = devm_memremap_pages(&mdevice->pdevice->dev, &devmem->pagemap); + if (IS_ERR(ptr)) + goto err_free; + + devmem->mdevice = mdevice; + pfn_first = devmem->pagemap.res.start >> PAGE_SHIFT; + pfn_last = pfn_first + + (resource_size(&devmem->pagemap.res) >> PAGE_SHIFT); + mdevice->devmem_chunks[mdevice->devmem_count++] = devmem; + + mutex_unlock(&mdevice->devmem_lock); + + pr_info("added new %u MB chunk (total %u chunks, %u MB) PFNs [0x%lx 0x%lx)\n", + DEVMEM_CHUNK_SIZE / (1024 * 1024), + mdevice->devmem_count, + mdevice->devmem_count * (DEVMEM_CHUNK_SIZE / (1024 * 1024)), + pfn_first, pfn_last); + + spin_lock(&mdevice->lock); + for (pfn = pfn_first; pfn < pfn_last; pfn++) { + struct page *page = pfn_to_page(pfn); + + page->zone_device_data = mdevice->free_pages; + mdevice->free_pages = page; + } + if (ppage) { + *ppage = mdevice->free_pages; + mdevice->free_pages = (*ppage)->zone_device_data; + mdevice->calloc++; + } + spin_unlock(&mdevice->lock); + + return true; + +err_free: + kfree(devmem); +err: + mutex_unlock(&mdevice->devmem_lock); + return false; +} + +static struct page *dmirror_devmem_alloc_page(struct dmirror_device *mdevice) +{ + struct page *dpage = NULL; + struct page *rpage; + + /* + * This is a fake device so we alloc real system memory to store + * our device memory. + */ + rpage = alloc_page(GFP_HIGHUSER); + if (!rpage) + return NULL; + + spin_lock(&mdevice->lock); + + if (mdevice->free_pages) { + dpage = mdevice->free_pages; + mdevice->free_pages = dpage->zone_device_data; + mdevice->calloc++; + spin_unlock(&mdevice->lock); + } else { + spin_unlock(&mdevice->lock); + if (!dmirror_allocate_chunk(mdevice, &dpage)) + goto error; + } + + dpage->zone_device_data = rpage; + get_page(dpage); + lock_page(dpage); + return dpage; + +error: + __free_page(rpage); + return NULL; +} + +static void dmirror_migrate_alloc_and_copy(struct migrate_vma *args, + struct dmirror *dmirror) +{ + struct dmirror_device *mdevice = dmirror->mdevice; + const unsigned long *src = args->src; + unsigned long *dst = args->dst; + unsigned long addr; + + for (addr = args->start; addr < args->end; addr += PAGE_SIZE, + src++, dst++) { + struct page *spage; + struct page *dpage; + struct page *rpage; + + if (!(*src & MIGRATE_PFN_MIGRATE)) + continue; + + /* + * Note that spage might be NULL which is OK since it is an + * unallocated pte_none() or read-only zero page. + */ + spage = migrate_pfn_to_page(*src); + + /* + * Don't migrate device private pages from our own driver or + * others. For our own we would do a device private memory copy + * not a migration and for others, we would need to fault the + * other device's page into system memory first. + */ + if (spage && is_zone_device_page(spage)) + continue; + + dpage = dmirror_devmem_alloc_page(mdevice); + if (!dpage) + continue; + + rpage = dpage->zone_device_data; + if (spage) + copy_highpage(rpage, spage); + else + clear_highpage(rpage); + + /* + * Normally, a device would use the page->zone_device_data to + * point to the mirror but here we use it to hold the page for + * the simulated device memory and that page holds the pointer + * to the mirror. + */ + rpage->zone_device_data = dmirror; + + *dst = migrate_pfn(page_to_pfn(dpage)) | + MIGRATE_PFN_LOCKED; + if ((*src & MIGRATE_PFN_WRITE) || + (!spage && args->vma->vm_flags & VM_WRITE)) + *dst |= MIGRATE_PFN_WRITE; + } + /* Try to pre-allocate page tables. */ + mutex_lock(&dmirror->mutex); + dmirror_pt_walk(dmirror, NULL, args->start, args->end, NULL, true); + mutex_unlock(&dmirror->mutex); +} + +struct dmirror_migrate { + struct hmm_dmirror_cmd *cmd; + const unsigned long *src; + const unsigned long *dst; + unsigned long start; +}; + +static int dmirror_do_migrate(struct dmirror *dmirror, + unsigned long addr, + unsigned long end, + u64 *dptep, + void *private) +{ + struct dmirror_migrate *migrate = private; + const unsigned long *src = migrate->src; + const unsigned long *dst = migrate->dst; + unsigned long idx = (addr - migrate->start) >> PAGE_SHIFT; + + for (; addr < end; addr += PAGE_SIZE, ++dptep, ++idx) { + struct page *page; + u64 dpte; + + if (!(src[idx] & MIGRATE_PFN_MIGRATE)) + continue; + + page = migrate_pfn_to_page(dst[idx]); + if (!page) + continue; + + /* + * Map the page that holds the data so dmirror_pt_walk() + * doesn't have to deal with ZONE_DEVICE private pages. + */ + page = page->zone_device_data; + dpte = dmirror_pt_from_page(page) | DPT_DPAGE; + if (dst[idx] & MIGRATE_PFN_WRITE) + dpte |= DPT_WRITE; + *dptep = dpte; + } + + return 0; +} + +static void dmirror_migrate_finalize_and_map(struct migrate_vma *args, + struct dmirror *dmirror, + struct hmm_dmirror_cmd *cmd) +{ + struct dmirror_migrate migrate; + + migrate.cmd = cmd; + migrate.src = args->src; + migrate.dst = args->dst; + migrate.start = args->start; + + /* Map the migrated pages into the device's page tables. */ + mutex_lock(&dmirror->mutex); + dmirror_pt_walk(dmirror, dmirror_do_migrate, args->start, args->end, + &migrate, true); + mutex_unlock(&dmirror->mutex); +} + +static int dmirror_migrate(struct dmirror *dmirror, + struct hmm_dmirror_cmd *cmd) +{ + unsigned long start, end, addr; + unsigned long size = cmd->npages << PAGE_SHIFT; + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; + struct vm_area_struct *vma; + unsigned long src_pfns[64]; + unsigned long dst_pfns[64]; + struct dmirror_bounce bounce; + struct migrate_vma args; + unsigned long next; + int ret; + + start = cmd->addr; + end = start + size; + if (end < start) + return -EINVAL; + + down_read(&mm->mmap_sem); + for (addr = start; addr < end; addr = next) { + next = min(end, addr + (ARRAY_SIZE(src_pfns) << PAGE_SHIFT)); + + vma = find_vma(mm, addr); + if (!vma || addr < vma->vm_start) { + ret = -EINVAL; + goto out; + } + if (next > vma->vm_end) + next = vma->vm_end; + + args.vma = vma; + args.src = src_pfns; + args.dst = dst_pfns; + args.start = addr; + args.end = next; + ret = migrate_vma_setup(&args); + if (ret) + goto out; + + dmirror_migrate_alloc_and_copy(&args, dmirror); + migrate_vma_pages(&args); + dmirror_migrate_finalize_and_map(&args, dmirror, cmd); + migrate_vma_finalize(&args); + } + up_read(&mm->mmap_sem); + + /* Return the migrated data for verification. */ + ret = dmirror_bounce_init(&bounce, start, size); + if (ret) + return ret; + mutex_lock(&dmirror->mutex); + ret = dmirror_pt_walk(dmirror, dmirror_do_read, start, end, &bounce, + false); + mutex_unlock(&dmirror->mutex); + if (ret == 0) + ret = dmirror_bounce_copy_to(&bounce, cmd->ptr); + cmd->cpages = bounce.cpages; + dmirror_bounce_fini(&bounce); + return ret; + +out: + up_read(&mm->mmap_sem); + return ret; +} + +static void dmirror_mkentry(struct dmirror *dmirror, + struct hmm_range *range, + unsigned char *perm, + uint64_t entry) +{ + struct page *page; + + if (entry == range->values[HMM_PFN_ERROR]) { + *perm = HMM_DMIRROR_PROT_ERROR; + return; + } + page = hmm_device_entry_to_page(range, entry); + if (!page) { + *perm = HMM_DMIRROR_PROT_NONE; + return; + } + if (entry & range->flags[HMM_PFN_DEVICE_PRIVATE]) { + /* Is the page migrated to this device or some other? */ + if (dmirror->mdevice == dmirror_page_to_device(page)) + *perm = HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL; + else + *perm = HMM_DMIRROR_PROT_DEV_PRIVATE_REMOTE; + } else if (is_zero_pfn(page_to_pfn(page))) + *perm = HMM_DMIRROR_PROT_ZERO; + else + *perm = HMM_DMIRROR_PROT_NONE; + if (entry & range->flags[HMM_PFN_WRITE]) + *perm |= HMM_DMIRROR_PROT_WRITE; + else + *perm |= HMM_DMIRROR_PROT_READ; +} + +static int dmirror_snapshot(struct dmirror *dmirror, + struct hmm_dmirror_cmd *cmd) +{ + struct mm_struct *mm = dmirror->mirror.hmm->mmu_notifier.mm; + unsigned long start, end; + unsigned long size = cmd->npages << PAGE_SHIFT; + unsigned long addr; + unsigned long next; + uint64_t pfns[64]; + unsigned char perm[64]; + char __user *uptr; + struct hmm_range range = { + .pfns = pfns, + .flags = dmirror_hmm_flags, + .values = dmirror_hmm_values, + .pfn_shift = DPT_SHIFT, + .pfn_flags_mask = ~0ULL, + }; + int ret = 0; + + start = cmd->addr; + end = start + size; + uptr = (void __user *)cmd->ptr; + + for (addr = start; addr < end; ) { + long count; + unsigned long i; + unsigned long n; + + next = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end); + range.start = addr; + range.end = next; + + down_read(&mm->mmap_sem); + + ret = hmm_range_register(&range, &dmirror->mirror); + if (ret) { + up_read(&mm->mmap_sem); + break; + } + + if (!hmm_range_wait_until_valid(&range, + DMIRROR_RANGE_FAULT_TIMEOUT)) { + hmm_range_unregister(&range); + up_read(&mm->mmap_sem); + continue; + } + + count = hmm_range_fault(&range, HMM_FAULT_SNAPSHOT); + if (count < 0) { + ret = count; + hmm_range_unregister(&range); + up_read(&mm->mmap_sem); + if (ret == -EBUSY) + continue; + break; + } + + if (!hmm_range_valid(&range)) { + hmm_range_unregister(&range); + up_read(&mm->mmap_sem); + continue; + } + + n = (next - addr) >> PAGE_SHIFT; + for (i = 0; i < n; i++) + dmirror_mkentry(dmirror, &range, perm + i, pfns[i]); + hmm_range_unregister(&range); + up_read(&mm->mmap_sem); + + ret = copy_to_user(uptr, perm, n); + if (ret) + break; + + cmd->cpages += n; + uptr += n; + addr = next; + } + + return ret; +} + +static long dmirror_fops_unlocked_ioctl(struct file *filp, + unsigned int command, + unsigned long arg) +{ + void __user *uarg = (void __user *)arg; + struct hmm_dmirror_cmd cmd; + struct dmirror *dmirror; + int ret; + + dmirror = filp->private_data; + if (!dmirror) + return -EINVAL; + + ret = copy_from_user(&cmd, uarg, sizeof(cmd)); + if (ret) + return ret; + + if (cmd.addr & ~PAGE_MASK) + return -EINVAL; + if (cmd.addr >= (cmd.addr + (cmd.npages << PAGE_SHIFT))) + return -EINVAL; + + cmd.cpages = 0; + cmd.faults = 0; + + switch (command) { + case HMM_DMIRROR_READ: + ret = dmirror_read(dmirror, &cmd); + break; + + case HMM_DMIRROR_WRITE: + ret = dmirror_write(dmirror, &cmd); + break; + + case HMM_DMIRROR_MIGRATE: + ret = dmirror_migrate(dmirror, &cmd); + break; + + case HMM_DMIRROR_SNAPSHOT: + ret = dmirror_snapshot(dmirror, &cmd); + break; + + default: + return -EINVAL; + } + if (ret) + return ret; + + return copy_to_user(uarg, &cmd, sizeof(cmd)); +} + +static const struct file_operations dmirror_fops = { + .read = dmirror_fops_read, + .write = dmirror_fops_write, + .mmap = dmirror_fops_mmap, + .open = dmirror_fops_open, + .release = dmirror_fops_release, + .unlocked_ioctl = dmirror_fops_unlocked_ioctl, + .llseek = default_llseek, + .owner = THIS_MODULE, +}; + +static void dmirror_devmem_free(struct page *page) +{ + struct page *rpage = page->zone_device_data; + struct dmirror_device *mdevice; + + if (rpage) + __free_page(rpage); + + mdevice = dmirror_page_to_device(page); + + spin_lock(&mdevice->lock); + mdevice->cfree++; + page->zone_device_data = mdevice->free_pages; + mdevice->free_pages = page; + spin_unlock(&mdevice->lock); +} + +static vm_fault_t dmirror_devmem_fault_alloc_and_copy(struct migrate_vma *args, + struct dmirror_device *mdevice) +{ + struct vm_area_struct *vma = args->vma; + const unsigned long *src = args->src; + unsigned long *dst = args->dst; + unsigned long start = args->start; + unsigned long end = args->end; + unsigned long addr; + + for (addr = start; addr < end; addr += PAGE_SIZE, + src++, dst++) { + struct page *dpage, *spage; + + spage = migrate_pfn_to_page(*src); + if (!spage || !(*src & MIGRATE_PFN_MIGRATE)) + continue; + if (!dmirror_device_is_mine(mdevice, spage)) + continue; + spage = spage->zone_device_data; + + dpage = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, addr); + if (!dpage) + continue; + + lock_page(dpage); + copy_highpage(dpage, spage); + *dst = migrate_pfn(page_to_pfn(dpage)) | + MIGRATE_PFN_LOCKED; + if (*src & MIGRATE_PFN_WRITE) + *dst |= MIGRATE_PFN_WRITE; + } + return 0; +} + +static void dmirror_devmem_fault_finalize_and_map(struct migrate_vma *args, + struct dmirror *dmirror) +{ + /* Invalidate the device's page table mapping. */ + mutex_lock(&dmirror->mutex); + dmirror_pt_walk(dmirror, dmirror_do_update, args->start, args->end, + NULL, false); + mutex_unlock(&dmirror->mutex); +} + +static vm_fault_t dmirror_devmem_fault(struct vm_fault *vmf) +{ + struct migrate_vma args; + unsigned long src_pfns; + unsigned long dst_pfns; + struct page *rpage; + struct dmirror *dmirror; + vm_fault_t ret; + + /* FIXME demonstrate how we can adjust migrate range */ + args.vma = vmf->vma; + args.start = vmf->address; + args.end = args.start + PAGE_SIZE; + args.src = &src_pfns; + args.dst = &dst_pfns; + + if (migrate_vma_setup(&args)) + return VM_FAULT_SIGBUS; + + /* + * Normally, a device would use the page->zone_device_data to point to + * the mirror but here we use it to hold the page for the simulated + * device memory and that page holds the pointer to the mirror. + */ + rpage = vmf->page->zone_device_data; + dmirror = rpage->zone_device_data; + + ret = dmirror_devmem_fault_alloc_and_copy(&args, dmirror->mdevice); + if (ret) + return ret; + migrate_vma_pages(&args); + dmirror_devmem_fault_finalize_and_map(&args, dmirror); + migrate_vma_finalize(&args); + return 0; +} + +static const struct dev_pagemap_ops dmirror_devmem_ops = { + .page_free = dmirror_devmem_free, + .migrate_to_ram = dmirror_devmem_fault, +}; + +static void dmirror_pdev_del(void *arg) +{ + struct dmirror_device *mdevice = arg; + unsigned int i; + + if (mdevice->devmem_chunks) { + for (i = 0; i < mdevice->devmem_count; i++) + kfree(mdevice->devmem_chunks[i]); + kfree(mdevice->devmem_chunks); + } + + cdev_del(&mdevice->cdevice); + kfree(mdevice); +} + +static int dmirror_probe(struct platform_device *pdev) +{ + struct dmirror_device *mdevice; + int ret; + + mdevice = kzalloc(sizeof(*mdevice), GFP_KERNEL); + if (!mdevice) + return -ENOMEM; + + mdevice->pdevice = pdev; + mutex_init(&mdevice->devmem_lock); + spin_lock_init(&mdevice->lock); + + cdev_init(&mdevice->cdevice, &dmirror_fops); + ret = cdev_add(&mdevice->cdevice, pdev->dev.devt, 1); + if (ret) { + kfree(mdevice); + return ret; + } + + platform_set_drvdata(pdev, mdevice); + ret = devm_add_action_or_reset(&pdev->dev, dmirror_pdev_del, mdevice); + if (ret) + return ret; + + /* Build list of free struct page */ + dmirror_allocate_chunk(mdevice, NULL); + + return 0; +} + +static int dmirror_remove(struct platform_device *pdev) +{ + /* all probe actions are unwound by devm */ + return 0; +} + +static struct platform_driver dmirror_device_driver = { + .probe = dmirror_probe, + .remove = dmirror_remove, + .driver = { + .name = "HMM_DMIRROR", + }, +}; + +static int __init hmm_dmirror_init(void) +{ + int ret; + int id; + + ret = platform_driver_register(&dmirror_device_driver); + if (ret) + return ret; + + ret = alloc_chrdev_region(&dmirror_dev, 0, DMIRROR_NDEVICES, + "HMM_DMIRROR"); + if (ret) + goto err_unreg; + + for (id = 0; id < DMIRROR_NDEVICES; id++) { + struct platform_device *pd; + + pd = platform_device_alloc("HMM_DMIRROR", id); + if (!pd) { + ret = -ENOMEM; + goto err_chrdev; + } + pd->dev.devt = MKDEV(MAJOR(dmirror_dev), id); + ret = platform_device_add(pd); + if (ret) { + platform_device_put(pd); + goto err_chrdev; + } + dmirror_platform_devices[id] = pd; + } + + /* + * Allocate a zero page to simulate a reserved page of device private + * memory which is always zero. The zero_pfn page isn't used just to + * make the code here simpler (i.e., we need a struct page for it). + */ + dmirror_zero_page = alloc_page(GFP_HIGHUSER | __GFP_ZERO); + if (!dmirror_zero_page) + goto err_chrdev; + + pr_info("hmm_dmirror loaded. This is only for testing HMM.\n"); + return 0; + +err_chrdev: + while (--id >= 0) { + platform_device_unregister(dmirror_platform_devices[id]); + dmirror_platform_devices[id] = NULL; + } + unregister_chrdev_region(dmirror_dev, 1); +err_unreg: + platform_driver_unregister(&dmirror_device_driver); + return ret; +} + +static void __exit hmm_dmirror_exit(void) +{ + int id; + + if (dmirror_zero_page) + __free_page(dmirror_zero_page); + for (id = 0; id < DMIRROR_NDEVICES; id++) + platform_device_unregister(dmirror_platform_devices[id]); + unregister_chrdev_region(dmirror_dev, DMIRROR_NDEVICES); + platform_driver_unregister(&dmirror_device_driver); + mmu_notifier_synchronize(); +} + +module_init(hmm_dmirror_init); +module_exit(hmm_dmirror_exit); +MODULE_LICENSE("GPL"); diff --git a/include/Kbuild b/include/Kbuild index ffba79483cc5..6ffb44a45957 100644 --- a/include/Kbuild +++ b/include/Kbuild @@ -1063,6 +1063,7 @@ header-test- += uapi/linux/coda_psdev.h header-test- += uapi/linux/errqueue.h header-test- += uapi/linux/eventpoll.h header-test- += uapi/linux/hdlc/ioctl.h +header-test- += uapi/linux/hmm_dmirror.h header-test- += uapi/linux/input.h header-test- += uapi/linux/kvm.h header-test- += uapi/linux/kvm_para.h diff --git a/include/uapi/linux/hmm_dmirror.h b/include/uapi/linux/hmm_dmirror.h new file mode 100644 index 000000000000..61d3643aff95 --- /dev/null +++ b/include/uapi/linux/hmm_dmirror.h @@ -0,0 +1,74 @@ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +/* + * Copyright 2013 Red Hat Inc. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2 of + * the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Authors: Jérôme Glisse <jglisse@redhat.com> + */ +/* + * This is a dummy driver to exercise the HMM (heterogeneous memory management) + * API of the kernel. It allows a userspace program to expose its entire address + * space through the HMM dummy driver file. + */ +#ifndef _UAPI_LINUX_HMM_DMIRROR_H +#define _UAPI_LINUX_HMM_DMIRROR_H + +#include <linux/types.h> +#include <linux/ioctl.h> + +/* + * Structure to pass to the HMM test driver to mimic a device accessing + * system memory and ZONE_DEVICE private memory through device page tables. + * + * @addr: (in) user address the device will read/write + * @ptr: (in) user address where device data is copied to/from + * @npages: (in) number of pages to read/write + * @cpages: (out) number of pages copied + * @faults: (out) number of device page faults seen + */ +struct hmm_dmirror_cmd { + __u64 addr; + __u64 ptr; + __u64 npages; + __u64 cpages; + __u64 faults; +}; + +/* Expose the address space of the calling process through hmm dummy dev file */ +#define HMM_DMIRROR_READ _IOWR('H', 0x00, struct hmm_dmirror_cmd) +#define HMM_DMIRROR_WRITE _IOWR('H', 0x01, struct hmm_dmirror_cmd) +#define HMM_DMIRROR_MIGRATE _IOWR('H', 0x02, struct hmm_dmirror_cmd) +#define HMM_DMIRROR_SNAPSHOT _IOWR('H', 0x03, struct hmm_dmirror_cmd) + +/* + * Values returned in hmm_dmirror_cmd.ptr for HMM_DMIRROR_SNAPSHOT. + * HMM_DMIRROR_PROT_ERROR: no valid mirror PTE for this page + * HMM_DMIRROR_PROT_NONE: unpopulated PTE or PTE with no access + * HMM_DMIRROR_PROT_READ: read-only PTE + * HMM_DMIRROR_PROT_WRITE: read/write PTE + * HMM_DMIRROR_PROT_ZERO: special read-only zero page + * HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL: Migrated device private page on the + * device the ioctl() is made + * HMM_DMIRROR_PROT_DEV_PRIVATE_REMOTE: Migrated device private page on some + * other device + */ +enum { + HMM_DMIRROR_PROT_ERROR = 0xFF, + HMM_DMIRROR_PROT_NONE = 0x00, + HMM_DMIRROR_PROT_READ = 0x01, + HMM_DMIRROR_PROT_WRITE = 0x02, + HMM_DMIRROR_PROT_ZERO = 0x10, + HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL = 0x20, + HMM_DMIRROR_PROT_DEV_PRIVATE_REMOTE = 0x30, +}; + +#endif /* _UAPI_LINUX_HMM_DMIRROR_H */ diff --git a/tools/testing/selftests/vm/.gitignore b/tools/testing/selftests/vm/.gitignore index 31b3c98b6d34..3054565b3f07 100644 --- a/tools/testing/selftests/vm/.gitignore +++ b/tools/testing/selftests/vm/.gitignore @@ -14,3 +14,4 @@ virtual_address_range gup_benchmark va_128TBswitch map_fixed_noreplace +hmm-tests diff --git a/tools/testing/selftests/vm/Makefile b/tools/testing/selftests/vm/Makefile index 9534dc2bc929..5643cfb5e3d6 100644 --- a/tools/testing/selftests/vm/Makefile +++ b/tools/testing/selftests/vm/Makefile @@ -5,6 +5,7 @@ CFLAGS = -Wall -I ../../../../usr/include $(EXTRA_CFLAGS) LDLIBS = -lrt TEST_GEN_FILES = compaction_test TEST_GEN_FILES += gup_benchmark +TEST_GEN_FILES += hmm-tests TEST_GEN_FILES += hugepage-mmap TEST_GEN_FILES += hugepage-shm TEST_GEN_FILES += map_hugetlb @@ -26,6 +27,8 @@ TEST_FILES := test_vmalloc.sh KSFT_KHDR_INSTALL := 1 include ../lib.mk +$(OUTPUT)/hmm-tests: LDLIBS += -lhugetlbfs -lpthread + $(OUTPUT)/userfaultfd: LDLIBS += -lpthread $(OUTPUT)/mlock-random-test: LDLIBS += -lcap diff --git a/tools/testing/selftests/vm/config b/tools/testing/selftests/vm/config index 1c0d76cb5adf..34cfab18e737 100644 --- a/tools/testing/selftests/vm/config +++ b/tools/testing/selftests/vm/config @@ -1,2 +1,5 @@ CONFIG_SYSVIPC=y CONFIG_USERFAULTFD=y +CONFIG_HMM_MIRROR=y +CONFIG_DEVICE_PRIVATE=y +CONFIG_HMM_DMIRROR=m diff --git a/tools/testing/selftests/vm/hmm-tests.c b/tools/testing/selftests/vm/hmm-tests.c new file mode 100644 index 000000000000..f4ae6188fd0e --- /dev/null +++ b/tools/testing/selftests/vm/hmm-tests.c @@ -0,0 +1,1311 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright 2013 Red Hat Inc. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2 of + * the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Authors: Jérôme Glisse <jglisse@redhat.com> + */ + +/* + * HMM stands for Heterogeneous Memory Management, it is a helper layer inside + * the linux kernel to help device drivers mirror a process address space in + * the device. This allows the device to use the same address space which + * makes communication and data exchange a lot easier. + * + * This framework's sole purpose is to exercise various code paths inside + * the kernel to make sure that HMM performs as expected and to flush out any + * bugs. + */ + +#include "../kselftest_harness.h" + +#include <errno.h> +#include <fcntl.h> +#include <stdio.h> +#include <stdlib.h> +#include <stdint.h> +#include <unistd.h> +#include <strings.h> +#include <time.h> +#include <pthread.h> +#include <hugetlbfs.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/mman.h> +#include <sys/ioctl.h> +#include <linux/hmm_dmirror.h> + +struct hmm_buffer { + void *ptr; + void *mirror; + unsigned long size; + int fd; + uint64_t cpages; + uint64_t faults; +}; + +#define TWOMEG (1 << 21) +#define HMM_BUFFER_SIZE (1024 << 12) +#define HMM_PATH_MAX 64 +#define NTIMES 256 + +#define ALIGN(x, a) (((x) + (a - 1)) & (~((a) - 1))) + +FIXTURE(hmm) +{ + int fd; + unsigned int page_size; + unsigned int page_shift; +}; + +FIXTURE(hmm2) +{ + int fd0; + int fd1; + unsigned int page_size; + unsigned int page_shift; +}; + +static int hmm_open(int unit) +{ + char pathname[HMM_PATH_MAX]; + int fd; + + snprintf(pathname, sizeof(pathname), "/dev/hmm_dmirror%d", unit); + fd = open(pathname, O_RDWR, 0); + if (fd < 0) + fprintf(stderr, "could not open hmm dmirror driver (%s)\n", + pathname); + return fd; +} + +FIXTURE_SETUP(hmm) +{ + self->page_size = sysconf(_SC_PAGE_SIZE); + self->page_shift = ffs(self->page_size) - 1; + + self->fd = hmm_open(0); + ASSERT_GE(self->fd, 0); +} + +FIXTURE_SETUP(hmm2) +{ + self->page_size = sysconf(_SC_PAGE_SIZE); + self->page_shift = ffs(self->page_size) - 1; + + self->fd0 = hmm_open(0); + ASSERT_GE(self->fd0, 0); + self->fd1 = hmm_open(1); + ASSERT_GE(self->fd1, 0); +} + +FIXTURE_TEARDOWN(hmm) +{ + int ret = close(self->fd); + + ASSERT_EQ(ret, 0); + self->fd = -1; +} + +FIXTURE_TEARDOWN(hmm2) +{ + int ret = close(self->fd0); + + ASSERT_EQ(ret, 0); + self->fd0 = -1; + + ret = close(self->fd1); + ASSERT_EQ(ret, 0); + self->fd1 = -1; +} + +static int hmm_dmirror_cmd(int fd, + unsigned long request, + struct hmm_buffer *buffer, + unsigned long npages) +{ + struct hmm_dmirror_cmd cmd; + int ret; + + /* Simulate a device reading system memory. */ + cmd.addr = (__u64)buffer->ptr; + cmd.ptr = (__u64)buffer->mirror; + cmd.npages = npages; + + for (;;) { + ret = ioctl(fd, request, &cmd); + if (ret == 0) + break; + if (errno == EINTR) + continue; + return -errno; + } + buffer->cpages = cmd.cpages; + buffer->faults = cmd.faults; + + return 0; +} + +static void hmm_buffer_free(struct hmm_buffer *buffer) +{ + if (buffer == NULL) + return; + + if (buffer->ptr) + munmap(buffer->ptr, buffer->size); + free(buffer->mirror); + free(buffer); +} + +/* + * Create a temporary file that will be deleted on close. + */ +static int hmm_create_file(unsigned long size) +{ + char path[HMM_PATH_MAX]; + int fd; + + strcpy(path, "/tmp"); + fd = open(path, O_TMPFILE | O_EXCL | O_RDWR, 0600); + if (fd >= 0) { + int r; + + do { + r = ftruncate(fd, size); + } while (r == -1 && errno == EINTR); + if (!r) + return fd; + close(fd); + } + return -1; +} + +/* + * Return a random unsigned number. + */ +static unsigned int hmm_random(void) +{ + static int fd = -1; + unsigned int r; + + if (fd < 0) { + fd = open("/dev/urandom", O_RDONLY); + if (fd < 0) { + fprintf(stderr, "%s:%d failed to open /dev/urandom\n", + __FILE__, __LINE__); + return ~0U; + } + } + read(fd, &r, sizeof(r)); + return r; +} + +static void hmm_nanosleep(unsigned int n) +{ + struct timespec t; + + t.tv_sec = 0; + t.tv_nsec = n; + nanosleep(&t, NULL); +} + +/* + * Simple NULL test of device open/close. + */ +TEST_F(hmm, open_close) +{ +} + +/* + * Read private anonymous memory. + */ +TEST_F(hmm, anon_read) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + int val; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* + * Initialize buffer in system memory but leave the first two pages + * zero (pte_none and pfn_zero). + */ + i = 2 * self->page_size / sizeof(*ptr); + for (ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Set buffer permission to read-only. */ + ret = mprotect(buffer->ptr, size, PROT_READ); + ASSERT_EQ(ret, 0); + + /* Populate the CPU page table with a special zero page. */ + val = *(int *)(buffer->ptr + self->page_size); + ASSERT_EQ(val, 0); + + /* Simulate a device reading system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + ptr = buffer->mirror; + for (i = 0; i < 2 * self->page_size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], 0); + for (; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Read private anonymous memory which has been protected with + * mprotect() PROT_NONE. + */ +TEST_F(hmm, anon_read_prot) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Initialize mirror buffer so we can verify it isn't written. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = -i; + + /* Protect buffer from reading. */ + ret = mprotect(buffer->ptr, size, PROT_NONE); + ASSERT_EQ(ret, 0); + + /* Simulate a device reading system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, npages); + ASSERT_EQ(ret, -EFAULT); + + /* Allow CPU to read the buffer so we can check it. */ + ret = mprotect(buffer->ptr, size, PROT_READ); + ASSERT_EQ(ret, 0); + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + hmm_buffer_free(buffer); +} + +/* + * Write private anonymous memory. + */ +TEST_F(hmm, anon_write) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Write private anonymous memory which has been protected with + * mprotect() PROT_READ. + */ +TEST_F(hmm, anon_write_prot) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Simulate a device reading a zero page of memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, 1); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, 1); + ASSERT_EQ(buffer->faults, 1); + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, -EPERM); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], 0); + + /* Now allow writing and see that the zero page is replaced. */ + ret = mprotect(buffer->ptr, size, PROT_WRITE | PROT_READ); + ASSERT_EQ(ret, 0); + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Check that a device writing an anonymous private mapping + * will copy-on-write if a child process inherits the mapping. + */ +TEST_F(hmm, anon_write_child) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + pid_t pid; + int child_fd; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer->ptr so we can tell if it is written. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = -i; + + pid = fork(); + if (pid == -1) + ASSERT_EQ(pid, 0); + if (pid != 0) { + waitpid(pid, &ret, 0); + ASSERT_EQ(WIFEXITED(ret), 1); + + /* Check that the parent's buffer did not change. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + return; + } + + /* Check that we see the parent's values. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + /* The child process needs its own mirror to its own mm. */ + child_fd = hmm_open(0); + ASSERT_GE(child_fd, 0); + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(child_fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + close(child_fd); + exit(0); +} + +/* + * Check that a device writing an anonymous shared mapping + * will not copy-on-write if a child process inherits the mapping. + */ +TEST_F(hmm, anon_write_child_shared) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + pid_t pid; + int child_fd; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_SHARED | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer->ptr so we can tell if it is written. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = -i; + + pid = fork(); + if (pid == -1) + ASSERT_EQ(pid, 0); + if (pid != 0) { + waitpid(pid, &ret, 0); + ASSERT_EQ(WIFEXITED(ret), 1); + + /* Check that the parent's buffer did change. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + return; + } + + /* Check that we see the parent's values. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + /* The child process needs its own mirror to its own mm. */ + child_fd = hmm_open(0); + ASSERT_GE(child_fd, 0); + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(child_fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + close(child_fd); + exit(0); +} + +/* + * Write private anonymous huge page. + */ +TEST_F(hmm, anon_write_huge) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + void *old_ptr; + void *map; + int *ptr; + int ret; + + size = 2 * TWOMEG; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + size = TWOMEG; + npages = size >> self->page_shift; + map = (void *)ALIGN((uintptr_t)buffer->ptr, size); + ret = madvise(map, size, MADV_HUGEPAGE); + ASSERT_EQ(ret, 0); + old_ptr = buffer->ptr; + buffer->ptr = map; + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + buffer->ptr = old_ptr; + hmm_buffer_free(buffer); +} + +/* + * Write huge TLBFS page. + */ +TEST_F(hmm, anon_write_hugetlbfs) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + long pagesizes[4]; + int n, idx; + + /* Skip test if we can't allocate a hugetlbfs page. */ + + n = gethugepagesizes(pagesizes, 4); + if (n <= 0) + return; + for (idx = 0; --n > 0; ) { + if (pagesizes[n] < pagesizes[idx]) + idx = n; + } + size = ALIGN(TWOMEG, pagesizes[idx]); + npages = size >> self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->ptr = get_hugepage_region(size, GHR_STRICT); + if (buffer->ptr == NULL) { + free(buffer); + return; + } + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + free_hugepage_region(buffer->ptr); + buffer->ptr = NULL; + hmm_buffer_free(buffer); +} + +/* + * Read mmap'ed file memory. + */ +TEST_F(hmm, file_read) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + int fd; + off_t off; + ssize_t len; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + fd = hmm_create_file(size); + ASSERT_GE(fd, 0); + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = fd; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + /* Write initial contents of the file. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + off = lseek(fd, 0, SEEK_SET); + ASSERT_EQ(off, 0); + len = write(fd, buffer->mirror, size); + ASSERT_EQ(len, size); + memset(buffer->mirror, 0, size); + + buffer->ptr = mmap(NULL, size, + PROT_READ, + MAP_SHARED, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Simulate a device reading system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Write mmap'ed file memory. + */ +TEST_F(hmm, file_write) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + int fd; + off_t off; + ssize_t len; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + fd = hmm_create_file(size); + ASSERT_GE(fd, 0); + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = fd; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_SHARED, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Check that the device also wrote the file. */ + off = lseek(fd, 0, SEEK_SET); + ASSERT_EQ(off, 0); + len = read(fd, buffer->mirror, size); + ASSERT_EQ(len, size); + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Migrate anonymous memory to device private memory. + */ +TEST_F(hmm, migrate) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Migrate memory to device. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_MIGRATE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Migrate anonymous memory to device private memory and fault it back to system + * memory. + */ +TEST_F(hmm, migrate_fault) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Migrate memory to device. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_MIGRATE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Fault pages back to system memory and check them. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Try to migrate various memory types to device private memory. + */ +TEST_F(hmm2, migrate_mixed) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + int *ptr; + unsigned char *p; + int ret; + int val; + + npages = 6; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + /* Reserve a range of addresses. */ + buffer->ptr = mmap(NULL, size, + PROT_NONE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + p = buffer->ptr; + + /* Now try to migrate everything to device 1. */ + ret = hmm_dmirror_cmd(self->fd1, HMM_DMIRROR_MIGRATE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, 6); + + /* Punch a hole after the first page address. */ + ret = munmap(buffer->ptr + self->page_size, self->page_size); + ASSERT_EQ(ret, 0); + + /* We expect an error if the vma doesn't cover the range. */ + ret = hmm_dmirror_cmd(self->fd1, HMM_DMIRROR_MIGRATE, buffer, 3); + ASSERT_EQ(ret, -EINVAL); + + /* Page 2 will be a read-only zero page. */ + ret = mprotect(buffer->ptr + 2 * self->page_size, self->page_size, + PROT_READ); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 2 * self->page_size); + val = *ptr + 3; + ASSERT_EQ(val, 3); + + /* Page 3 will be read-only. */ + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, + PROT_READ | PROT_WRITE); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 3 * self->page_size); + *ptr = val; + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, + PROT_READ); + ASSERT_EQ(ret, 0); + + /* Page 4 will be read-write. */ + ret = mprotect(buffer->ptr + 4 * self->page_size, self->page_size, + PROT_READ | PROT_WRITE); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 4 * self->page_size); + *ptr = val; + + /* Page 5 won't be migrated to device 0 because it's on device 1. */ + buffer->ptr = p + 5 * self->page_size; + ret = hmm_dmirror_cmd(self->fd0, HMM_DMIRROR_MIGRATE, buffer, 1); + ASSERT_EQ(ret, -ENOENT); + buffer->ptr = p; + + /* Now try to migrate pages 2-3 to device 1. */ + buffer->ptr = p + 2 * self->page_size; + ret = hmm_dmirror_cmd(self->fd1, HMM_DMIRROR_MIGRATE, buffer, 2); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, 2); + buffer->ptr = p; + + hmm_buffer_free(buffer); +} + +/* + * Migrate anonymous memory to device private memory and fault it back to system + * memory multiple times. + */ +TEST_F(hmm, migrate_multiple) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + unsigned long c; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + for (c = 0; c < NTIMES; c++) { + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Migrate memory to device. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_MIGRATE, buffer, + npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Fault pages back to system memory and check them. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); + } +} + +/* + * Read anonymous memory multiple times. + */ +TEST_F(hmm, anon_read_multiple) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + unsigned long c; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + for (c = 0; c < NTIMES; c++) { + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i + c; + + /* Simulate a device reading system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, + npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i + c); + + hmm_buffer_free(buffer); + } +} + +void *unmap_buffer(void *p) +{ + struct hmm_buffer *buffer = p; + + /* Delay for a bit and then unmap buffer while it is being read. */ + hmm_nanosleep(hmm_random() % 32000); + munmap(buffer->ptr + buffer->size / 2, buffer->size / 2); + buffer->ptr = NULL; + + return NULL; +} + +/* + * Try reading anonymous memory while it is being unmapped. + */ +TEST_F(hmm, anon_teardown) +{ + unsigned long npages; + unsigned long size; + unsigned long c; + void *ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + for (c = 0; c < NTIMES; ++c) { + pthread_t thread; + struct hmm_buffer *buffer; + unsigned long i; + int *ptr; + int rc; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i + c; + + rc = pthread_create(&thread, NULL, unmap_buffer, buffer); + ASSERT_EQ(rc, 0); + + /* Simulate a device reading system memory. */ + rc = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, + npages); + if (rc == 0) { + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; + i < size / sizeof(*ptr); + ++i) + ASSERT_EQ(ptr[i], i + c); + } + + pthread_join(thread, &ret); + hmm_buffer_free(buffer); + } +} + +/* + * Test memory snapshot without faulting in pages accessed by the device. + */ +TEST_F(hmm2, snapshot) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + int *ptr; + unsigned char *p; + unsigned char *m; + int ret; + int val; + + npages = 7; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(npages); + ASSERT_NE(buffer->mirror, NULL); + + /* Reserve a range of addresses. */ + buffer->ptr = mmap(NULL, size, + PROT_NONE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + p = buffer->ptr; + + /* Punch a hole after the first page address. */ + ret = munmap(buffer->ptr + self->page_size, self->page_size); + ASSERT_EQ(ret, 0); + + /* Page 2 will be read-only zero page. */ + ret = mprotect(buffer->ptr + 2 * self->page_size, self->page_size, + PROT_READ); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 2 * self->page_size); + val = *ptr + 3; + ASSERT_EQ(val, 3); + + /* Page 3 will be read-only. */ + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, + PROT_READ | PROT_WRITE); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 3 * self->page_size); + *ptr = val; + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, + PROT_READ); + ASSERT_EQ(ret, 0); + + /* Page 4-6 will be read-write. */ + ret = mprotect(buffer->ptr + 4 * self->page_size, 3 * self->page_size, + PROT_READ | PROT_WRITE); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 4 * self->page_size); + *ptr = val; + + /* Page 5 will be migrated to device 0. */ + buffer->ptr = p + 5 * self->page_size; + ret = hmm_dmirror_cmd(self->fd0, HMM_DMIRROR_MIGRATE, buffer, 1); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, 1); + + /* Page 6 will be migrated to device 1. */ + buffer->ptr = p + 6 * self->page_size; + ret = hmm_dmirror_cmd(self->fd1, HMM_DMIRROR_MIGRATE, buffer, 1); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, 1); + + /* Simulate a device snapshotting CPU pagetables. */ + buffer->ptr = p; + ret = hmm_dmirror_cmd(self->fd0, HMM_DMIRROR_SNAPSHOT, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device saw. */ + m = buffer->mirror; + ASSERT_EQ(m[0], HMM_DMIRROR_PROT_NONE); + ASSERT_EQ(m[1], HMM_DMIRROR_PROT_NONE); + ASSERT_EQ(m[2], HMM_DMIRROR_PROT_ZERO | HMM_DMIRROR_PROT_READ); + ASSERT_EQ(m[3], HMM_DMIRROR_PROT_READ); + ASSERT_EQ(m[4], HMM_DMIRROR_PROT_WRITE); + ASSERT_EQ(m[5], HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL | + HMM_DMIRROR_PROT_WRITE); + ASSERT_EQ(m[6], HMM_DMIRROR_PROT_DEV_PRIVATE_REMOTE | + HMM_DMIRROR_PROT_WRITE); + + hmm_buffer_free(buffer); +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/vm/run_vmtests b/tools/testing/selftests/vm/run_vmtests index 951c507a27f7..634cfefdaffd 100755 --- a/tools/testing/selftests/vm/run_vmtests +++ b/tools/testing/selftests/vm/run_vmtests @@ -227,4 +227,20 @@ else exitcode=1 fi +echo "------------------------------------" +echo "running HMM smoke test" +echo "------------------------------------" +./test_hmm.sh smoke +ret_val=$? + +if [ $ret_val -eq 0 ]; then + echo "[PASS]" +elif [ $ret_val -eq $ksft_skip ]; then + echo "[SKIP]" + exitcode=$ksft_skip +else + echo "[FAIL]" + exitcode=1 +fi + exit $exitcode diff --git a/tools/testing/selftests/vm/test_hmm.sh b/tools/testing/selftests/vm/test_hmm.sh new file mode 100755 index 000000000000..268d32752045 --- /dev/null +++ b/tools/testing/selftests/vm/test_hmm.sh @@ -0,0 +1,97 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# +# Copyright (C) 2018 Uladzislau Rezki (Sony) <urezki@gmail.com> +# +# This is a test script for the kernel test driver to analyse vmalloc +# allocator. Therefore it is just a kernel module loader. You can specify +# and pass different parameters in order to: +# a) analyse performance of vmalloc allocations; +# b) stressing and stability check of vmalloc subsystem. + +TEST_NAME="test_hmm" +DRIVER="hmm_dmirror" + +# 1 if fails +exitcode=1 + +# Kselftest framework requirement - SKIP code is 4. +ksft_skip=4 + +check_test_requirements() +{ + uid=$(id -u) + if [ $uid -ne 0 ]; then + echo "$0: Must be run as root" + exit $ksft_skip + fi + + if ! which modprobe > /dev/null 2>&1; then + echo "$0: You need modprobe installed" + exit $ksft_skip + fi + + if ! modinfo $DRIVER > /dev/null 2>&1; then + echo "$0: You must have the following enabled in your kernel:" + echo "CONFIG_HMM_DMIRROR=m" + exit $ksft_skip + fi +} + +load_driver() +{ + modprobe $DRIVER > /dev/null 2>&1 + if [ $? == 0 ]; then + major=$(awk "\$2==\"HMM_DMIRROR\" {print \$1}" /proc/devices) + mknod /dev/hmm_dmirror0 c $major 0 + mknod /dev/hmm_dmirror1 c $major 1 + fi +} + +unload_driver() +{ + modprobe -r $DRIVER > /dev/null 2>&1 + rm -f /dev/hmm_dmirror? +} + +run_smoke() +{ + echo "Running smoke test. Note, this test provides basic coverage." + + load_driver + ./hmm-tests + unload_driver +} + +usage() +{ + echo -n "Usage: $0" + echo + echo "Example usage:" + echo + echo "# Shows help message" + echo "./${TEST_NAME}.sh" + echo + echo "# Smoke testing" + echo "./${TEST_NAME}.sh smoke" + echo + exit 0 +} + +function run_test() +{ + if [ $# -eq 0 ]; then + usage + else + if [ "$1" = "smoke" ]; then + run_smoke + else + usage + fi + fi +} + +check_test_requirements +run_test $@ + +exit 0
Add self tests for HMM. Signed-off-by: Ralph Campbell <rcampbell@nvidia.com> --- MAINTAINERS | 3 + drivers/char/Kconfig | 11 + drivers/char/Makefile | 1 + drivers/char/hmm_dmirror.c | 1566 ++++++++++++++++++++++++ include/Kbuild | 1 + include/uapi/linux/hmm_dmirror.h | 74 ++ tools/testing/selftests/vm/.gitignore | 1 + tools/testing/selftests/vm/Makefile | 3 + tools/testing/selftests/vm/config | 3 + tools/testing/selftests/vm/hmm-tests.c | 1311 ++++++++++++++++++++ tools/testing/selftests/vm/run_vmtests | 16 + tools/testing/selftests/vm/test_hmm.sh | 97 ++ 12 files changed, 3087 insertions(+) create mode 100644 drivers/char/hmm_dmirror.c create mode 100644 include/uapi/linux/hmm_dmirror.h create mode 100644 tools/testing/selftests/vm/hmm-tests.c create mode 100755 tools/testing/selftests/vm/test_hmm.sh