June 2020 - Kernel - mailweb.openeuler.org

perf spe-c2c record失败的问题
by haoxin 15 Dec '21

15 Dec '21

您好，我采用最细的openEuler-1.0测试perf spe-c2c的功能，但是发现/sys/devices下没有arm_spe_0这个目录执行perf spe-c2c record时显示： event syntax error: 'arm_spe_0/ts_enable=1,pct_enable=1,pa_enable=1,load_filter=1,jitter=1,store_filter=1,min_latency=0/' \___ Cannot find PMU `arm_spe_0'. Missing kernel support? 想问下您这是为什么？需要哪里开启spe功能吗？

3 2

[PATCH] usb: usbtest: fix missing kfree(dev->buf) in usbtest_disconnect
by Yang Yingliang 01 Jul '20

01 Jul '20

From: Zqiang <qiang.zhang(a)windriver.com> mainline inclusion from mainline-v5.8-rc3 commit 28ebeb8db77035e058a510ce9bd17c2b9a009dba category: bugfix bugzilla: NA CVE: CVE-2020-15393 --------------------------- BUG: memory leak unreferenced object 0xffff888055046e00 (size 256): comm "kworker/2:9", pid 2570, jiffies 4294942129 (age 1095.500s) hex dump (first 32 bytes): 00 70 04 55 80 88 ff ff 18 bb 5a 81 ff ff ff ff .p.U......Z..... f5 96 78 81 ff ff ff ff 37 de 8e 81 ff ff ff ff ..x.....7....... backtrace: [<00000000d121dccf>] kmemleak_alloc_recursive include/linux/kmemleak.h:43 [inline] [<00000000d121dccf>] slab_post_alloc_hook mm/slab.h:586 [inline] [<00000000d121dccf>] slab_alloc_node mm/slub.c:2786 [inline] [<00000000d121dccf>] slab_alloc mm/slub.c:2794 [inline] [<00000000d121dccf>] kmem_cache_alloc_trace+0x15e/0x2d0 mm/slub.c:2811 [<000000005c3c3381>] kmalloc include/linux/slab.h:555 [inline] [<000000005c3c3381>] usbtest_probe+0x286/0x19d0 drivers/usb/misc/usbtest.c:2790 [<000000001cec6910>] usb_probe_interface+0x2bd/0x870 drivers/usb/core/driver.c:361 [<000000007806c118>] really_probe+0x48d/0x8f0 drivers/base/dd.c:551 [<00000000a3308c3e>] driver_probe_device+0xfc/0x2a0 drivers/base/dd.c:724 [<000000003ef66004>] __device_attach_driver+0x1b6/0x240 drivers/base/dd.c:831 [<00000000eee53e97>] bus_for_each_drv+0x14e/0x1e0 drivers/base/bus.c:431 [<00000000bb0648d0>] __device_attach+0x1f9/0x350 drivers/base/dd.c:897 [<00000000838b324a>] device_initial_probe+0x1a/0x20 drivers/base/dd.c:944 [<0000000030d501c1>] bus_probe_device+0x1e1/0x280 drivers/base/bus.c:491 [<000000005bd7adef>] device_add+0x131d/0x1c40 drivers/base/core.c:2504 [<00000000a0937814>] usb_set_configuration+0xe84/0x1ab0 drivers/usb/core/message.c:2030 [<00000000e3934741>] generic_probe+0x6a/0xe0 drivers/usb/core/generic.c:210 [<0000000098ade0f1>] usb_probe_device+0x90/0xd0 drivers/usb/core/driver.c:266 [<000000007806c118>] really_probe+0x48d/0x8f0 drivers/base/dd.c:551 [<00000000a3308c3e>] driver_probe_device+0xfc/0x2a0 drivers/base/dd.c:724 Acked-by: Alan Stern <stern(a)rowland.harvard.edu> Reported-by: Kyungtae Kim <kt0755(a)gmail.com> Signed-off-by: Zqiang <qiang.zhang(a)windriver.com> Link: https://lore.kernel.org/r/20200612035210.20494-1-qiang.zhang@windriver.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/usb/misc/usbtest.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/misc/usbtest.c b/drivers/usb/misc/usbtest.c index c7f82310e73e..fc3fc9d48a55 100644 --- a/drivers/usb/misc/usbtest.c +++ b/drivers/usb/misc/usbtest.c @@ -2853,6 +2853,7 @@ static void usbtest_disconnect(struct usb_interface *intf) usb_set_intfdata(intf, NULL); dev_dbg(&intf->dev, "disconnect\n"); + kfree(dev->buf); kfree(dev); } -- 2.25.1

1 0

[PATCH] genirq/debugfs: Add missing sanity checks to interrupt injection
by Yang Yingliang 01 Jul '20

01 Jul '20

From: Thomas Gleixner <tglx(a)linutronix.de> mainline inclusion from mainline-v5.6 commit a740a423c36932695b01a3e920f697bc55b05fec category: bugfix bugzilla: 32425 CVE: NA ------------------- Interrupts cannot be injected when the interrupt is not activated and when a replay is already in progress. Fixes: 536e2e34bd00 ("genirq/debugfs: Triggering of interrupts from userspace") Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Acked-by: Marc Zyngier <maz(a)kernel.org> Link: https://lkml.kernel.org/r/20200306130623.500019114@linutronix.de Signed-off-by: Yuan Can <yuancan(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- kernel/irq/debugfs.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/kernel/irq/debugfs.c b/kernel/irq/debugfs.c index 0a7365456e72..4eb5c62991fe 100644 --- a/kernel/irq/debugfs.c +++ b/kernel/irq/debugfs.c @@ -206,8 +206,15 @@ static ssize_t irq_debug_write(struct file *file, const char __user *user_buf, chip_bus_lock(desc); raw_spin_lock_irqsave(&desc->lock, flags); - if (irq_settings_is_level(desc) || desc->istate & IRQS_NMI) { - /* Can't do level nor NMIs, sorry */ + /* + * Don't allow injection when the interrupt is: + * - Level or NMI type + * - not activated + * - replaying already + */ + if (irq_settings_is_level(desc) || + !irqd_is_activated(&desc->irq_data) || + (desc->istate & (IRQS_NMI | IRQS_REPLAY))) { err = -EINVAL; } else { desc->istate |= IRQS_PENDING; -- 2.25.1

1 0

[PATCH] perf top: Fix stdio interface input handling with glibc 2.28+
by Yang Yingliang 01 Jul '20

01 Jul '20

From: Tommi Rantala <tommi.t.rantala(a)nokia.com> mainline inclusion from mainline-5.6-rc6 commit 29b4f5f18857 category: bugfix bugzilla: 31774 CVE: NA ------------------------------------------------- Since glibc 2.28 when running 'perf top --stdio', input handling no longer works, but hitting any key always just prints the "Mapped keys" help text. To fix it, call clearerr() in the display_thread() loop to clear any EOF sticky errors, as instructed in the glibc NEWS file (https://sourceware.org/git/?p=glibc.git;a=blob;f=NEWS) * All stdio functions now treat end-of-file as a sticky condition. If you read from a file until EOF, and then the file is enlarged by another process, you must call clearerr or another function with the same effect (e.g. fseek, rewind) before you can read the additional data. This corrects a longstanding C99 conformance bug. It is most likely to affect programs that use stdio to read interactive input from a terminal. (Bug #1190.) Signed-off-by: Tommi Rantala <tommi.t.rantala(a)nokia.com> Tested-by: Arnaldo Carvalho de Melo <acme(a)redhat.com> Cc: Alexander Shishkin <alexander.shishkin(a)linux.intel.com> Cc: Jiri Olsa <jolsa(a)redhat.com> Cc: Mark Rutland <mark.rutland(a)arm.com> Cc: Namhyung Kim <namhyung(a)kernel.org> Cc: Peter Zijlstra <peterz(a)infradead.org> Link: http://lore.kernel.org/lkml/20200305083714.9381-2-tommi.t.rantala@nokia.com Signed-off-by: Arnaldo Carvalho de Melo <acme(a)redhat.com> Signed-off-by: Wei Li <liwei391(a)huawei.com> Reviewed-by: Jian Cheng <cj.chengjian(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- tools/perf/builtin-top.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index 781f6cf1cda6..265c074cbe3b 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -661,7 +661,9 @@ static void *display_thread(void *arg) delay_msecs = top->delay_secs * MSEC_PER_SEC; set_term_quiet_input(&save); /* trash return*/ - getc(stdin); + clearerr(stdin); + if (poll(&stdin_poll, 1, 0) > 0) + getc(stdin); while (!done) { perf_top__print_sym_table(top); -- 2.25.1

1 0

[PATCH] vfio/pci: Fix SR-IOV VF handling with MMIO blocking
by Yang Yingliang 30 Jun '20

30 Jun '20

From: Alex Williamson <alex.williamson(a)redhat.com> mainline inclusion from mainline-v5.8-rc3 commit ebfa440ce38b7e2e04c3124aa89c8a9f4094cf21 category: bugfix bugzilla: NA CVE: CVE-2020-12888 --------------------------- SR-IOV VFs do not implement the memory enable bit of the command register, therefore this bit is not set in config space after pci_enable_device(). This leads to an unintended difference between PF and VF in hand-off state to the user. We can correct this by setting the initial value of the memory enable bit in our virtualized config space. There's really no need however to ever fault a user on a VF though as this would only indicate an error in the user's management of the enable bit, versus a PF where the same access could trigger hardware faults. Fixes: abafbc551fdd ("vfio-pci: Invalidate mmaps and block MMIO access on disabled memory") Signed-off-by: Alex Williamson <alex.williamson(a)redhat.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/vfio/pci/vfio_pci_config.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/drivers/vfio/pci/vfio_pci_config.c b/drivers/vfio/pci/vfio_pci_config.c index 4fe71fbce194..a1a26465d224 100644 --- a/drivers/vfio/pci/vfio_pci_config.c +++ b/drivers/vfio/pci/vfio_pci_config.c @@ -401,9 +401,15 @@ static inline void p_setd(struct perm_bits *p, int off, u32 virt, u32 write) /* Caller should hold memory_lock semaphore */ bool __vfio_pci_memory_enabled(struct vfio_pci_device *vdev) { + struct pci_dev *pdev = vdev->pdev; u16 cmd = le16_to_cpu(*(__le16 *)&vdev->vconfig[PCI_COMMAND]); - return cmd & PCI_COMMAND_MEMORY; + /* + * SR-IOV VF memory enable is handled by the MSE bit in the + * PF SR-IOV capability, there's therefore no need to trigger + * faults based on the virtual value. + */ + return pdev->is_virtfn || (cmd & PCI_COMMAND_MEMORY); } /* @@ -1732,6 +1738,15 @@ int vfio_config_init(struct vfio_pci_device *vdev) vconfig[PCI_INTERRUPT_PIN]); vconfig[PCI_INTERRUPT_PIN] = 0; /* Gratuitous for good VFs */ + + /* + * VFs do no implement the memory enable bit of the COMMAND + * register therefore we'll not have it set in our initial + * copy of config space after pci_enable_device(). For + * consistency with PFs, set the virtual enable bit here. + */ + *(__le16 *)&vconfig[PCI_COMMAND] |= + cpu_to_le16(PCI_COMMAND_MEMORY); } if (!IS_ENABLED(CONFIG_VFIO_PCI_INTX) || vdev->nointx) -- 2.25.1

1 0

[PATCH 1/5] sched/core: Create task_has_idle_policy() helper
by Yang Yingliang 30 Jun '20

30 Jun '20

From: Viresh Kumar <viresh.kumar(a)linaro.org> mainline inclusion from mainline-v5.0-rc1 commit 1da1843f9f0334e2428308945d396ffecc2acfe1 category: feature bugzilla: 38260 CVE: NA --------------------------- We already have task_has_rt_policy() and task_has_dl_policy() helpers, create task_has_idle_policy() as well and update sched core to start using it. While at it, use task_has_dl_policy() at one more place. Signed-off-by: Viresh Kumar <viresh.kumar(a)linaro.org> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Acked-by: Daniel Lezcano <daniel.lezcano(a)linaro.org> Cc: Linus Torvalds <torvalds(a)linux-foundation.org> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Thomas Gleixner <tglx(a)linutronix.de> Cc: Vincent Guittot <vincent.guittot(a)linaro.org> Link: http://lkml.kernel.org/r/ce3915d5b490fc81af926a3b6bfb775e7188e005.154141689… Signed-off-by: Ingo Molnar <mingo(a)kernel.org> Signed-off-by: Cheng Jian <cj.chengjian(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- kernel/sched/core.c | 4 ++-- kernel/sched/debug.c | 2 +- kernel/sched/fair.c | 10 +++++----- kernel/sched/sched.h | 5 +++++ 4 files changed, 13 insertions(+), 8 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 0fbdc620697b..c1513905c0de 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -724,7 +724,7 @@ static void set_load_weight(struct task_struct *p, bool update_load) /* * SCHED_IDLE tasks get minimal weight: */ - if (idle_policy(p->policy)) { + if (task_has_idle_policy(p)) { load->weight = scale_load(WEIGHT_IDLEPRIO); load->inv_weight = WMULT_IDLEPRIO; return; @@ -4270,7 +4270,7 @@ static int __sched_setscheduler(struct task_struct *p, * Treat SCHED_IDLE as nice 20. Only allow a switch to * SCHED_NORMAL if the RLIMIT_NICE would normally permit it. */ - if (idle_policy(p->policy) && !idle_policy(policy)) { + if (task_has_idle_policy(p) && !idle_policy(policy)) { if (!can_nice(p, task_nice(p))) return -EPERM; } diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index c542d84dafce..f12382877390 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -1012,7 +1012,7 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, #endif P(policy); P(prio); - if (p->policy == SCHED_DEADLINE) { + if (task_has_dl_policy(p)) { P(dl.runtime); P(dl.deadline); } diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 09b0d83d5bbc..807c7fb78b6f 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -6609,7 +6609,7 @@ wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se) static void set_last_buddy(struct sched_entity *se) { - if (entity_is_task(se) && unlikely(task_of(se)->policy == SCHED_IDLE)) + if (entity_is_task(se) && unlikely(task_has_idle_policy(task_of(se)))) return; for_each_sched_entity(se) { @@ -6621,7 +6621,7 @@ static void set_last_buddy(struct sched_entity *se) static void set_next_buddy(struct sched_entity *se) { - if (entity_is_task(se) && unlikely(task_of(se)->policy == SCHED_IDLE)) + if (entity_is_task(se) && unlikely(task_has_idle_policy(task_of(se)))) return; for_each_sched_entity(se) { @@ -6679,8 +6679,8 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ return; /* Idle tasks are by definition preempted by non-idle tasks. */ - if (unlikely(curr->policy == SCHED_IDLE) && - likely(p->policy != SCHED_IDLE)) + if (unlikely(task_has_idle_policy(curr)) && + likely(!task_has_idle_policy(p))) goto preempt; /* @@ -7090,7 +7090,7 @@ static int task_hot(struct task_struct *p, struct lb_env *env) if (p->sched_class != &fair_sched_class) return 0; - if (unlikely(p->policy == SCHED_IDLE)) + if (unlikely(task_has_idle_policy(p))) return 0; /* diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 955abd645ff9..fbe3b3b8a19f 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -184,6 +184,11 @@ static inline bool valid_policy(int policy) rt_policy(policy) || dl_policy(policy); } +static inline int task_has_idle_policy(struct task_struct *p) +{ + return idle_policy(p->policy); +} + static inline int task_has_rt_policy(struct task_struct *p) { return rt_policy(p->policy); -- 2.25.1

1 4

[PATCH 001/204] power: supply: bq24257_charger: Replace depends on REGMAP_I2C with select
by Yang Yingliang 29 Jun '20

29 Jun '20

From: Enric Balletbo i Serra <enric.balletbo(a)collabora.com> [ Upstream commit 87c3d579c8ed0eaea6b1567d529a8daa85a2bc6c ] regmap is a library function that gets selected by drivers that need it. No driver modules should depend on it. Depending on REGMAP_I2C makes this driver only build if another driver already selected REGMAP_I2C, as the symbol can't be selected through the menu kernel configuration. Fixes: 2219a935963e ("power_supply: Add TI BQ24257 charger driver") Signed-off-by: Enric Balletbo i Serra <enric.balletbo(a)collabora.com> Signed-off-by: Sebastian Reichel <sebastian.reichel(a)collabora.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/power/supply/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/power/supply/Kconfig b/drivers/power/supply/Kconfig index ff6dab0bf0dd..76c699b5abda 100644 --- a/drivers/power/supply/Kconfig +++ b/drivers/power/supply/Kconfig @@ -555,7 +555,7 @@ config CHARGER_BQ24257 tristate "TI BQ24250/24251/24257 battery charger driver" depends on I2C depends on GPIOLIB || COMPILE_TEST - depends on REGMAP_I2C + select REGMAP_I2C help Say Y to enable support for the TI BQ24250, BQ24251, and BQ24257 battery chargers. -- 2.25.1

1 203

[PATCH 1/2] Revert "ipmi: fix sleep-in-atomic in free_user at cleanup SRCU user->release_barrier"
by Yang Yingliang 29 Jun '20

29 Jun '20

hulk inclusion category: bugfix bugzilla: 5499 CVE: NA --------------------------- cleanup_srcu_struct_quiesced won't free memory if work_pending() is true, and it will cause a memory leak, so revert it and using mainline patch instead. Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/char/ipmi/ipmi_msghandler.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 0085d32cef46..7510013f28f8 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -1190,12 +1190,7 @@ EXPORT_SYMBOL(ipmi_get_smi_info); static void free_user(struct kref *ref) { struct ipmi_user *user = container_of(ref, struct ipmi_user, refcount); - - /* - * Cleanup without waiting. This could be called in atomic context. - * Refcount is zero: all read-sections should have been ended. - */ - cleanup_srcu_struct_quiesced(&user->release_barrier); + cleanup_srcu_struct(&user->release_barrier); kfree(user); } -- 2.25.1

1 1

[PATCH 001/257] ipv6: fix IPV6_ADDRFORM operation logic
by Yang Yingliang 29 Jun '20

29 Jun '20

From: Hangbin Liu <liuhangbin(a)gmail.com> [ Upstream commit 79a1f0ccdbb4ad700590f61b00525b390cb53905 ] Socket option IPV6_ADDRFORM supports UDP/UDPLITE and TCP at present. Previously the checking logic looks like: if (sk->sk_protocol == IPPROTO_UDP || sk->sk_protocol == IPPROTO_UDPLITE) do_some_check; else if (sk->sk_protocol != IPPROTO_TCP) break; After commit b6f6118901d1 ("ipv6: restrict IPV6_ADDRFORM operation"), TCP was blocked as the logic changed to: if (sk->sk_protocol == IPPROTO_UDP || sk->sk_protocol == IPPROTO_UDPLITE) do_some_check; else if (sk->sk_protocol == IPPROTO_TCP) do_some_check; break; else break; Then after commit 82c9ae440857 ("ipv6: fix restrict IPV6_ADDRFORM operation") UDP/UDPLITE were blocked as the logic changed to: if (sk->sk_protocol == IPPROTO_UDP || sk->sk_protocol == IPPROTO_UDPLITE) do_some_check; if (sk->sk_protocol == IPPROTO_TCP) do_some_check; if (sk->sk_protocol != IPPROTO_TCP) break; Fix it by using Eric's code and simply remove the break in TCP check, which looks like: if (sk->sk_protocol == IPPROTO_UDP || sk->sk_protocol == IPPROTO_UDPLITE) do_some_check; else if (sk->sk_protocol == IPPROTO_TCP) do_some_check; else break; Fixes: 82c9ae440857 ("ipv6: fix restrict IPV6_ADDRFORM operation") Signed-off-by: Hangbin Liu <liuhangbin(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- net/ipv6/ipv6_sockglue.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index f255eaaf0136..9ca4e8657534 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -187,14 +187,15 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, retv = -EBUSY; break; } - } - if (sk->sk_protocol == IPPROTO_TCP && - sk->sk_prot != &tcpv6_prot) { - retv = -EBUSY; + } else if (sk->sk_protocol == IPPROTO_TCP) { + if (sk->sk_prot != &tcpv6_prot) { + retv = -EBUSY; + break; + } + } else { break; } - if (sk->sk_protocol != IPPROTO_TCP) - break; + if (sk->sk_state != TCP_ESTABLISHED) { retv = -ENOTCONN; break; -- 2.25.1

1 256

[PATCH 1/3] vfio/type1: Support faulting PFNMAP vmas
by Yang Yingliang 28 Jun '20

28 Jun '20

From: Alex Williamson <alex.williamson(a)redhat.com> mainline inclusion from mainline-v5.8-rc1 commit 41311242221e3482b20bfed10fa4d9db98d87016 category: bugfix bugzilla: NA CVE: CVE-2020-12888 --------------------------- With conversion to follow_pfn(), DMA mapping a PFNMAP range depends on the range being faulted into the vma. Add support to manually provide that, in the same way as done on KVM with hva_to_pfn_remapped(). Reviewed-by: Peter Xu <peterx(a)redhat.com> Signed-off-by: Alex Williamson <alex.williamson(a)redhat.com> Conflicts: drivers/vfio/vfio_iommu_type1.c [yyl: adjust context] Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/vfio/vfio_iommu_type1.c | 36 ++++++++++++++++++++++++++++++--- 1 file changed, 33 insertions(+), 3 deletions(-) diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c index 4bfc79635c6a..01be1496307b 100644 --- a/drivers/vfio/vfio_iommu_type1.c +++ b/drivers/vfio/vfio_iommu_type1.c @@ -346,6 +346,32 @@ static int put_pfn(unsigned long pfn, int prot) return 0; } +static int follow_fault_pfn(struct vm_area_struct *vma, struct mm_struct *mm, + unsigned long vaddr, unsigned long *pfn, + bool write_fault) +{ + int ret; + + ret = follow_pfn(vma, vaddr, pfn); + if (ret) { + bool unlocked = false; + + ret = fixup_user_fault(NULL, mm, vaddr, + FAULT_FLAG_REMOTE | + (write_fault ? FAULT_FLAG_WRITE : 0), + &unlocked); + if (unlocked) + return -EAGAIN; + + if (ret) + return ret; + + ret = follow_pfn(vma, vaddr, pfn); + } + + return ret; +} + static int vaddr_get_pfn(struct mm_struct *mm, unsigned long vaddr, int prot, unsigned long *pfn) { @@ -385,12 +411,16 @@ static int vaddr_get_pfn(struct mm_struct *mm, unsigned long vaddr, down_read(&mm->mmap_sem); +retry: vma = find_vma_intersection(mm, vaddr, vaddr + 1); if (vma && vma->vm_flags & VM_PFNMAP) { - if (!follow_pfn(vma, vaddr, pfn) && - is_invalid_reserved_pfn(*pfn)) - ret = 0; + ret = follow_fault_pfn(vma, mm, vaddr, pfn, prot & IOMMU_WRITE); + if (ret == -EAGAIN) + goto retry; + + if (!ret && !is_invalid_reserved_pfn(*pfn)) + ret = -EFAULT; } up_read(&mm->mmap_sem); -- 2.25.1

1 2