From ecf61c78bd787b9dfb4a6b1dfb7561c2a95c5c17 Mon Sep 17 00:00:00 2001 From: Michal Marek Date: Mon, 11 Nov 2013 14:40:35 +0100 Subject: PCI: Omit PCI ID macro strings to shorten quirk names Pasting the verbatim PCI_(VENDOR|DEVICE)_* macros in the __pci_fixup_* symbol names results in insanely long names such as __pci_fixup_resumePCI_VENDOR_ID_SERVERWORKSPCI_DEVICE_ID_SERVERWORKS_HT1000SBquirk_disable_broadcom_boot_interrupt When Link-Time Optimization adds its numeric suffix to such symbol, it overflows the namebuf[KSYM_NAME_LEN] array in kernel/kallsyms.c. Use the line number instead to create (nearly) unique symbol names. Reported-by: Joe Mario Signed-off-by: Michal Marek Signed-off-by: Bjorn Helgaas Cc: Andi Kleen --- include/linux/pci.h | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/linux/pci.h b/include/linux/pci.h index 1084a15175e..eb8078aeadc 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1567,65 +1567,65 @@ enum pci_fixup_pass { /* Anonymous variables would be nice... */ #define DECLARE_PCI_FIXUP_SECTION(section, name, vendor, device, class, \ class_shift, hook) \ - static const struct pci_fixup __pci_fixup_##name __used \ + static const struct pci_fixup __PASTE(__pci_fixup_##name,__LINE__) __used \ __attribute__((__section__(#section), aligned((sizeof(void *))))) \ = { vendor, device, class, class_shift, hook }; #define DECLARE_PCI_FIXUP_CLASS_EARLY(vendor, device, class, \ class_shift, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_early, \ - vendor##device##hook, vendor, device, class, class_shift, hook) + hook, vendor, device, class, class_shift, hook) #define DECLARE_PCI_FIXUP_CLASS_HEADER(vendor, device, class, \ class_shift, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_header, \ - vendor##device##hook, vendor, device, class, class_shift, hook) + hook, vendor, device, class, class_shift, hook) #define DECLARE_PCI_FIXUP_CLASS_FINAL(vendor, device, class, \ class_shift, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_final, \ - vendor##device##hook, vendor, device, class, class_shift, hook) + hook, vendor, device, class, class_shift, hook) #define DECLARE_PCI_FIXUP_CLASS_ENABLE(vendor, device, class, \ class_shift, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_enable, \ - vendor##device##hook, vendor, device, class, class_shift, hook) + hook, vendor, device, class, class_shift, hook) #define DECLARE_PCI_FIXUP_CLASS_RESUME(vendor, device, class, \ class_shift, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_resume, \ - resume##vendor##device##hook, vendor, device, class, \ + resume##hook, vendor, device, class, \ class_shift, hook) #define DECLARE_PCI_FIXUP_CLASS_RESUME_EARLY(vendor, device, class, \ class_shift, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_resume_early, \ - resume_early##vendor##device##hook, vendor, device, \ + resume_early##hook, vendor, device, \ class, class_shift, hook) #define DECLARE_PCI_FIXUP_CLASS_SUSPEND(vendor, device, class, \ class_shift, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_suspend, \ - suspend##vendor##device##hook, vendor, device, class, \ + suspend##hook, vendor, device, class, \ class_shift, hook) #define DECLARE_PCI_FIXUP_EARLY(vendor, device, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_early, \ - vendor##device##hook, vendor, device, PCI_ANY_ID, 0, hook) + hook, vendor, device, PCI_ANY_ID, 0, hook) #define DECLARE_PCI_FIXUP_HEADER(vendor, device, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_header, \ - vendor##device##hook, vendor, device, PCI_ANY_ID, 0, hook) + hook, vendor, device, PCI_ANY_ID, 0, hook) #define DECLARE_PCI_FIXUP_FINAL(vendor, device, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_final, \ - vendor##device##hook, vendor, device, PCI_ANY_ID, 0, hook) + hook, vendor, device, PCI_ANY_ID, 0, hook) #define DECLARE_PCI_FIXUP_ENABLE(vendor, device, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_enable, \ - vendor##device##hook, vendor, device, PCI_ANY_ID, 0, hook) + hook, vendor, device, PCI_ANY_ID, 0, hook) #define DECLARE_PCI_FIXUP_RESUME(vendor, device, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_resume, \ - resume##vendor##device##hook, vendor, device, \ + resume##hook, vendor, device, \ PCI_ANY_ID, 0, hook) #define DECLARE_PCI_FIXUP_RESUME_EARLY(vendor, device, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_resume_early, \ - resume_early##vendor##device##hook, vendor, device, \ + resume_early##hook, vendor, device, \ PCI_ANY_ID, 0, hook) #define DECLARE_PCI_FIXUP_SUSPEND(vendor, device, hook) \ DECLARE_PCI_FIXUP_SECTION(.pci_fixup_suspend, \ - suspend##vendor##device##hook, vendor, device, \ + suspend##hook, vendor, device, \ PCI_ANY_ID, 0, hook) #ifdef CONFIG_PCI_QUIRKS -- cgit v1.2.3-70-g09d2 From 358f24704f2f016af7d504b357cdf32606091d07 Mon Sep 17 00:00:00 2001 From: Pali Rohár Date: Tue, 26 Nov 2013 11:17:02 -0800 Subject: Input: add key code for ambient light sensor button MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Many notebooks have a special button for enabling/disabling ambient light sensor. Signed-off-by: Pali Rohár Signed-off-by: Dmitry Torokhov --- include/uapi/linux/input.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/uapi/linux/input.h b/include/uapi/linux/input.h index a3726275876..7bacdb51437 100644 --- a/include/uapi/linux/input.h +++ b/include/uapi/linux/input.h @@ -719,6 +719,8 @@ struct input_keymap_entry { #define BTN_DPAD_LEFT 0x222 #define BTN_DPAD_RIGHT 0x223 +#define KEY_ALS_TOGGLE 0x230 /* Ambient light sensor */ + #define BTN_TRIGGER_HAPPY 0x2c0 #define BTN_TRIGGER_HAPPY1 0x2c0 #define BTN_TRIGGER_HAPPY2 0x2c1 -- cgit v1.2.3-70-g09d2 From 1e31aa9270daab40c7aef9d5488982e3475b87ef Mon Sep 17 00:00:00 2001 From: Ashutosh Dixit Date: Wed, 27 Nov 2013 08:58:41 -0800 Subject: misc: mic: Fix user space namespace pollution from mic_common.h. Avoid declaring ALIGN() and __aligned() in include/uapi/linux/mic_common.h since they pollute user space namespace. Also, mic_aligned_size() can be simply replaced simply by sizeof() since all structures where mic_aligned_size() is used are declared using __attribute__ ((aligned(8))); -- >From mail from H Peter Anvin about this: On Fri, Nov 08, 2013 H Peter Anvin wrote: Subject: Namespace pollution in mic_common.h This puts two macros, ALIGN() and __aligned(), into arbitrary user space namespace. This really isn't safe or acceptable, especially since those symbols are highly generic. ... When these structures are forced-aligned, they will in fact have padding automatically added by the compiler to an 8-byte boundary anyway, so mic_aligned_size() does nothing. ... Reported-by: H Peter Anvin Reviewed-by: Sudeep Dutt Signed-off-by: Nikhil Rao Signed-off-by: Ashutosh Dixit Signed-off-by: Greg Kroah-Hartman --- Documentation/mic/mpssd/mpssd.c | 4 ++-- drivers/misc/mic/card/mic_virtio.c | 4 ++-- drivers/misc/mic/card/mic_virtio.h | 7 +++---- drivers/misc/mic/host/mic_virtio.c | 2 +- include/uapi/linux/mic_common.h | 26 +++++++++----------------- 5 files changed, 17 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/Documentation/mic/mpssd/mpssd.c b/Documentation/mic/mpssd/mpssd.c index 5c7fdda56f0..91f9e8e4ef7 100644 --- a/Documentation/mic/mpssd/mpssd.c +++ b/Documentation/mic/mpssd/mpssd.c @@ -313,7 +313,7 @@ static struct mic_device_desc *get_device_desc(struct mic_info *mic, int type) int i; void *dp = get_dp(mic, type); - for (i = mic_aligned_size(struct mic_bootparam); i < PAGE_SIZE; + for (i = sizeof(struct mic_bootparam); i < PAGE_SIZE; i += mic_total_desc_size(d)) { d = dp + i; @@ -520,7 +520,7 @@ static void * virtio_net(void *arg) { static __u8 vnet_hdr[2][sizeof(struct virtio_net_hdr)]; - static __u8 vnet_buf[2][MAX_NET_PKT_SIZE] __aligned(64); + static __u8 vnet_buf[2][MAX_NET_PKT_SIZE] __attribute__ ((aligned(64))); struct iovec vnet_iov[2][2] = { { { .iov_base = vnet_hdr[0], .iov_len = sizeof(vnet_hdr[0]) }, { .iov_base = vnet_buf[0], .iov_len = sizeof(vnet_buf[0]) } }, diff --git a/drivers/misc/mic/card/mic_virtio.c b/drivers/misc/mic/card/mic_virtio.c index 4dce912f9ea..20c567b1760 100644 --- a/drivers/misc/mic/card/mic_virtio.c +++ b/drivers/misc/mic/card/mic_virtio.c @@ -520,8 +520,8 @@ static void mic_scan_devices(struct mic_driver *mdrv, bool remove) struct device *dev; int ret; - for (i = mic_aligned_size(struct mic_bootparam); - i < MIC_DP_SIZE; i += mic_total_desc_size(d)) { + for (i = sizeof(struct mic_bootparam); i < MIC_DP_SIZE; + i += mic_total_desc_size(d)) { d = mdrv->dp + i; dc = (void __iomem *)d + mic_aligned_desc_size(d); /* diff --git a/drivers/misc/mic/card/mic_virtio.h b/drivers/misc/mic/card/mic_virtio.h index 2c5c22c93ba..d0407ba53bb 100644 --- a/drivers/misc/mic/card/mic_virtio.h +++ b/drivers/misc/mic/card/mic_virtio.h @@ -42,8 +42,8 @@ static inline unsigned mic_desc_size(struct mic_device_desc __iomem *desc) { - return mic_aligned_size(*desc) - + ioread8(&desc->num_vq) * mic_aligned_size(struct mic_vqconfig) + return sizeof(*desc) + + ioread8(&desc->num_vq) * sizeof(struct mic_vqconfig) + ioread8(&desc->feature_len) * 2 + ioread8(&desc->config_len); } @@ -67,8 +67,7 @@ mic_vq_configspace(struct mic_device_desc __iomem *desc) } static inline unsigned mic_total_desc_size(struct mic_device_desc __iomem *desc) { - return mic_aligned_desc_size(desc) + - mic_aligned_size(struct mic_device_ctrl); + return mic_aligned_desc_size(desc) + sizeof(struct mic_device_ctrl); } int mic_devices_init(struct mic_driver *mdrv); diff --git a/drivers/misc/mic/host/mic_virtio.c b/drivers/misc/mic/host/mic_virtio.c index 945a3d0fe22..efe8da4401e 100644 --- a/drivers/misc/mic/host/mic_virtio.c +++ b/drivers/misc/mic/host/mic_virtio.c @@ -467,7 +467,7 @@ static int mic_copy_dp_entry(struct mic_vdev *mvdev, } /* Find the first free device page entry */ - for (i = mic_aligned_size(struct mic_bootparam); + for (i = sizeof(struct mic_bootparam); i < MIC_DP_SIZE - mic_total_desc_size(dd_config); i += mic_total_desc_size(devp)) { devp = mdev->dp + i; diff --git a/include/uapi/linux/mic_common.h b/include/uapi/linux/mic_common.h index 17e7d95e4f5..d0ec46d88d2 100644 --- a/include/uapi/linux/mic_common.h +++ b/include/uapi/linux/mic_common.h @@ -23,12 +23,7 @@ #include -#ifndef __KERNEL__ -#define ALIGN(a, x) (((a) + (x) - 1) & ~((x) - 1)) -#define __aligned(x) __attribute__ ((aligned(x))) -#endif - -#define mic_aligned_size(x) ALIGN(sizeof(x), 8) +#define __mic_align(a, x) (((a) + (x) - 1) & ~((x) - 1)) /** * struct mic_device_desc: Virtio device information shared between the @@ -49,7 +44,7 @@ struct mic_device_desc { __u8 config_len; __u8 status; __u64 config[0]; -} __aligned(8); +} __attribute__ ((aligned(8))); /** * struct mic_device_ctrl: Per virtio device information in the device page @@ -74,7 +69,7 @@ struct mic_device_ctrl { __u8 used_address_updated; __s8 c2h_vdev_db; __s8 h2c_vdev_db; -} __aligned(8); +} __attribute__ ((aligned(8))); /** * struct mic_bootparam: Virtio device independent information in device page @@ -93,7 +88,7 @@ struct mic_bootparam { __s8 h2c_config_db; __u8 shutdown_status; __u8 shutdown_card; -} __aligned(8); +} __attribute__ ((aligned(8))); /** * struct mic_device_page: High level representation of the device page @@ -119,7 +114,7 @@ struct mic_vqconfig { __u64 address; __u64 used_address; __u16 num; -} __aligned(8); +} __attribute__ ((aligned(8))); /* * The alignment to use between consumer and producer parts of vring. @@ -173,15 +168,13 @@ struct mic_vring { int len; }; -#define mic_aligned_desc_size(d) ALIGN(mic_desc_size(d), 8) +#define mic_aligned_desc_size(d) __mic_align(mic_desc_size(d), 8) #ifndef INTEL_MIC_CARD static inline unsigned mic_desc_size(const struct mic_device_desc *desc) { - return mic_aligned_size(*desc) - + desc->num_vq * mic_aligned_size(struct mic_vqconfig) - + desc->feature_len * 2 - + desc->config_len; + return sizeof(*desc) + desc->num_vq * sizeof(struct mic_vqconfig) + + desc->feature_len * 2 + desc->config_len; } static inline struct mic_vqconfig * @@ -201,8 +194,7 @@ static inline __u8 *mic_vq_configspace(const struct mic_device_desc *desc) } static inline unsigned mic_total_desc_size(struct mic_device_desc *desc) { - return mic_aligned_desc_size(desc) + - mic_aligned_size(struct mic_device_ctrl); + return mic_aligned_desc_size(desc) + sizeof(struct mic_device_ctrl); } #endif -- cgit v1.2.3-70-g09d2 From 173c07278763850bfee57eec442dce38855d6f13 Mon Sep 17 00:00:00 2001 From: Ashutosh Dixit Date: Wed, 27 Nov 2013 08:58:42 -0800 Subject: misc: mic: Fix endianness issues. Endianness issues are now consistent as per the documentation in host/mic_virtio.h. Sparse warnings related to endianness are also fixed. Note that the MIC driver implementation assumes that the host can be both BE or LE whereas the card is always LE. Reported-by: Fengguang Wu Reviewed-by: Sudeep Dutt Reviewed-by: Nikhil Rao Signed-off-by: Ashutosh Dixit Signed-off-by: Greg Kroah-Hartman --- Documentation/mic/mpssd/mpssd.c | 8 ++++---- drivers/misc/mic/card/mic_virtio.c | 14 +++++++------- drivers/misc/mic/host/mic_boot.c | 2 +- drivers/misc/mic/host/mic_virtio.c | 14 +++++++------- drivers/misc/mic/host/mic_x100.c | 4 ++-- include/uapi/linux/mic_common.h | 14 +++++++------- 6 files changed, 28 insertions(+), 28 deletions(-) (limited to 'include') diff --git a/Documentation/mic/mpssd/mpssd.c b/Documentation/mic/mpssd/mpssd.c index 91f9e8e4ef7..4d17487d5ad 100644 --- a/Documentation/mic/mpssd/mpssd.c +++ b/Documentation/mic/mpssd/mpssd.c @@ -445,8 +445,8 @@ init_vr(struct mic_info *mic, int fd, int type, __func__, mic->name, vr0->va, vr0->info, vr_size, vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN)); mpsslog("magic 0x%x expected 0x%x\n", - vr0->info->magic, MIC_MAGIC + type); - assert(vr0->info->magic == MIC_MAGIC + type); + le32toh(vr0->info->magic), MIC_MAGIC + type); + assert(le32toh(vr0->info->magic) == MIC_MAGIC + type); if (vr1) { vr1->va = (struct mic_vring *) &va[MIC_DEVICE_PAGE_END + vr_size]; @@ -458,8 +458,8 @@ init_vr(struct mic_info *mic, int fd, int type, __func__, mic->name, vr1->va, vr1->info, vr_size, vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN)); mpsslog("magic 0x%x expected 0x%x\n", - vr1->info->magic, MIC_MAGIC + type + 1); - assert(vr1->info->magic == MIC_MAGIC + type + 1); + le32toh(vr1->info->magic), MIC_MAGIC + type + 1); + assert(le32toh(vr1->info->magic) == MIC_MAGIC + type + 1); } done: return va; diff --git a/drivers/misc/mic/card/mic_virtio.c b/drivers/misc/mic/card/mic_virtio.c index 20c567b1760..ca0445fa3d8 100644 --- a/drivers/misc/mic/card/mic_virtio.c +++ b/drivers/misc/mic/card/mic_virtio.c @@ -248,17 +248,16 @@ static struct virtqueue *mic_find_vq(struct virtio_device *vdev, /* First assign the vring's allocated in host memory */ vqconfig = mic_vq_config(mvdev->desc) + index; memcpy_fromio(&config, vqconfig, sizeof(config)); - _vr_size = vring_size(config.num, MIC_VIRTIO_RING_ALIGN); + _vr_size = vring_size(le16_to_cpu(config.num), MIC_VIRTIO_RING_ALIGN); vr_size = PAGE_ALIGN(_vr_size + sizeof(struct _mic_vring_info)); - va = mic_card_map(mvdev->mdev, config.address, vr_size); + va = mic_card_map(mvdev->mdev, le64_to_cpu(config.address), vr_size); if (!va) return ERR_PTR(-ENOMEM); mvdev->vr[index] = va; memset_io(va, 0x0, _vr_size); - vq = vring_new_virtqueue(index, - config.num, MIC_VIRTIO_RING_ALIGN, vdev, - false, - va, mic_notify, callback, name); + vq = vring_new_virtqueue(index, le16_to_cpu(config.num), + MIC_VIRTIO_RING_ALIGN, vdev, false, va, + mic_notify, callback, name); if (!vq) { err = -ENOMEM; goto unmap; @@ -273,7 +272,8 @@ static struct virtqueue *mic_find_vq(struct virtio_device *vdev, /* Allocate and reassign used ring now */ mvdev->used_size[index] = PAGE_ALIGN(sizeof(__u16) * 3 + - sizeof(struct vring_used_elem) * config.num); + sizeof(struct vring_used_elem) * + le16_to_cpu(config.num)); used = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, get_order(mvdev->used_size[index])); if (!used) { diff --git a/drivers/misc/mic/host/mic_boot.c b/drivers/misc/mic/host/mic_boot.c index 7558d918643..b75c6b5cc20 100644 --- a/drivers/misc/mic/host/mic_boot.c +++ b/drivers/misc/mic/host/mic_boot.c @@ -62,7 +62,7 @@ void mic_bootparam_init(struct mic_device *mdev) { struct mic_bootparam *bootparam = mdev->dp; - bootparam->magic = MIC_MAGIC; + bootparam->magic = cpu_to_le32(MIC_MAGIC); bootparam->c2h_shutdown_db = mdev->shutdown_db; bootparam->h2c_shutdown_db = -1; bootparam->h2c_config_db = -1; diff --git a/drivers/misc/mic/host/mic_virtio.c b/drivers/misc/mic/host/mic_virtio.c index efe8da4401e..453d7409da9 100644 --- a/drivers/misc/mic/host/mic_virtio.c +++ b/drivers/misc/mic/host/mic_virtio.c @@ -293,8 +293,8 @@ static void mic_virtio_init_post(struct mic_vdev *mvdev) continue; } mvdev->mvr[i].vrh.vring.used = - mvdev->mdev->aper.va + - le64_to_cpu(vqconfig[i].used_address); + mvdev->mdev->aper.va + + le64_to_cpu(vqconfig[i].used_address); } mvdev->dc->used_address_updated = 0; @@ -525,6 +525,7 @@ int mic_virtio_add_device(struct mic_vdev *mvdev, char irqname[10]; struct mic_bootparam *bootparam = mdev->dp; u16 num; + dma_addr_t vr_addr; mutex_lock(&mdev->mic_mutex); @@ -559,17 +560,16 @@ int mic_virtio_add_device(struct mic_vdev *mvdev, } vr->len = vr_size; vr->info = vr->va + vring_size(num, MIC_VIRTIO_RING_ALIGN); - vr->info->magic = MIC_MAGIC + mvdev->virtio_id + i; - vqconfig[i].address = mic_map_single(mdev, - vr->va, vr_size); - if (mic_map_error(vqconfig[i].address)) { + vr->info->magic = cpu_to_le32(MIC_MAGIC + mvdev->virtio_id + i); + vr_addr = mic_map_single(mdev, vr->va, vr_size); + if (mic_map_error(vr_addr)) { free_pages((unsigned long)vr->va, get_order(vr_size)); ret = -ENOMEM; dev_err(mic_dev(mvdev), "%s %d err %d\n", __func__, __LINE__, ret); goto err; } - vqconfig[i].address = cpu_to_le64(vqconfig[i].address); + vqconfig[i].address = cpu_to_le64(vr_addr); vring_init(&vr->vr, num, vr->va, MIC_VIRTIO_RING_ALIGN); ret = vringh_init_kern(&mvr->vrh, diff --git a/drivers/misc/mic/host/mic_x100.c b/drivers/misc/mic/host/mic_x100.c index 81e9541b784..0dfa8a81436 100644 --- a/drivers/misc/mic/host/mic_x100.c +++ b/drivers/misc/mic/host/mic_x100.c @@ -397,8 +397,8 @@ mic_x100_load_ramdisk(struct mic_device *mdev) * so copy over the ramdisk @ 128M. */ memcpy_toio(mdev->aper.va + (mdev->bootaddr << 1), fw->data, fw->size); - iowrite32(cpu_to_le32(mdev->bootaddr << 1), &bp->hdr.ramdisk_image); - iowrite32(cpu_to_le32(fw->size), &bp->hdr.ramdisk_size); + iowrite32(mdev->bootaddr << 1, &bp->hdr.ramdisk_image); + iowrite32(fw->size, &bp->hdr.ramdisk_size); release_firmware(fw); error: return rc; diff --git a/include/uapi/linux/mic_common.h b/include/uapi/linux/mic_common.h index d0ec46d88d2..6eb40244e01 100644 --- a/include/uapi/linux/mic_common.h +++ b/include/uapi/linux/mic_common.h @@ -43,7 +43,7 @@ struct mic_device_desc { __u8 feature_len; __u8 config_len; __u8 status; - __u64 config[0]; + __le64 config[0]; } __attribute__ ((aligned(8))); /** @@ -61,7 +61,7 @@ struct mic_device_desc { * @h2c_vdev_db: The doorbell number to be used by host. Set by guest. */ struct mic_device_ctrl { - __u64 vdev; + __le64 vdev; __u8 config_change; __u8 vdev_reset; __u8 guest_ack; @@ -82,7 +82,7 @@ struct mic_device_ctrl { * @shutdown_card: Set to 1 by the host when a card shutdown is initiated */ struct mic_bootparam { - __u32 magic; + __le32 magic; __s8 c2h_shutdown_db; __s8 h2c_shutdown_db; __s8 h2c_config_db; @@ -111,9 +111,9 @@ struct mic_device_page { * @num: The number of entries in the virtio_ring */ struct mic_vqconfig { - __u64 address; - __u64 used_address; - __u16 num; + __le64 address; + __le64 used_address; + __le16 num; } __attribute__ ((aligned(8))); /* @@ -149,7 +149,7 @@ struct mic_vqconfig { */ struct _mic_vring_info { __u16 avail_idx; - int magic; + __le32 magic; }; /** -- cgit v1.2.3-70-g09d2 From 23fd78d76415729b338ff1802a0066b4a62f7fb8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 2 Dec 2013 11:24:18 +0000 Subject: KEYS: Fix multiple key add into associative array If sufficient keys (or keyrings) are added into a keyring such that a node in the associative array's tree overflows (each node has a capacity N, currently 16) and such that all N+1 keys have the same index key segment for that level of the tree (the level'th nibble of the index key), then assoc_array_insert() calls ops->diff_objects() to indicate at which bit position the two index keys vary. However, __key_link_begin() passes a NULL object to assoc_array_insert() with the intention of supplying the correct pointer later before we commit the change. This means that keyring_diff_objects() is given a NULL pointer as one of its arguments which it does not expect. This results in an oops like the attached. With the previous patch to fix the keyring hash function, this can be forced much more easily by creating a keyring and only adding keyrings to it. Add any other sort of key and a different insertion path is taken - all 16+1 objects must want to cluster in the same node slot. This can be tested by: r=`keyctl newring sandbox @s` for ((i=0; i<=16; i++)); do keyctl newring ring$i $r; done This should work fine, but oopses when the 17th keyring is added. Since ops->diff_objects() is always called with the first pointer pointing to the object to be inserted (ie. the NULL pointer), we can fix the problem by changing the to-be-inserted object pointer to point to the index key passed into assoc_array_insert() instead. Whilst we're at it, we also switch the arguments so that they are the same as for ->compare_object(). BUG: unable to handle kernel NULL pointer dereference at 0000000000000088 IP: [] hash_key_type_and_desc+0x18/0xb0 ... RIP: 0010:[] hash_key_type_and_desc+0x18/0xb0 ... Call Trace: [] keyring_diff_objects+0x21/0xd2 [] assoc_array_insert+0x3b6/0x908 [] __key_link_begin+0x78/0xe5 [] key_create_or_update+0x17d/0x36a [] SyS_add_key+0x123/0x183 [] tracesys+0xdd/0xe2 Signed-off-by: David Howells Tested-by: Stephen Gallagher --- Documentation/assoc_array.txt | 6 +++--- include/linux/assoc_array.h | 6 +++--- lib/assoc_array.c | 4 ++-- security/keys/keyring.c | 7 +++---- 4 files changed, 11 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/Documentation/assoc_array.txt b/Documentation/assoc_array.txt index f4faec0f66e..2f2c6cdd73c 100644 --- a/Documentation/assoc_array.txt +++ b/Documentation/assoc_array.txt @@ -164,10 +164,10 @@ This points to a number of methods, all of which need to be provided: (4) Diff the index keys of two objects. - int (*diff_objects)(const void *a, const void *b); + int (*diff_objects)(const void *object, const void *index_key); - Return the bit position at which the index keys of two objects differ or - -1 if they are the same. + Return the bit position at which the index key of the specified object + differs from the given index key or -1 if they are the same. (5) Free an object. diff --git a/include/linux/assoc_array.h b/include/linux/assoc_array.h index 9a193b84238..a89df3be168 100644 --- a/include/linux/assoc_array.h +++ b/include/linux/assoc_array.h @@ -41,10 +41,10 @@ struct assoc_array_ops { /* Is this the object we're looking for? */ bool (*compare_object)(const void *object, const void *index_key); - /* How different are two objects, to a bit position in their keys? (or - * -1 if they're the same) + /* How different is an object from an index key, to a bit position in + * their keys? (or -1 if they're the same) */ - int (*diff_objects)(const void *a, const void *b); + int (*diff_objects)(const void *object, const void *index_key); /* Method to free an object. */ void (*free_object)(void *object); diff --git a/lib/assoc_array.c b/lib/assoc_array.c index 17edeaf1918..1b6a44f1ec3 100644 --- a/lib/assoc_array.c +++ b/lib/assoc_array.c @@ -759,8 +759,8 @@ all_leaves_cluster_together: pr_devel("all leaves cluster together\n"); diff = INT_MAX; for (i = 0; i < ASSOC_ARRAY_FAN_OUT; i++) { - int x = ops->diff_objects(assoc_array_ptr_to_leaf(edit->leaf), - assoc_array_ptr_to_leaf(node->slots[i])); + int x = ops->diff_objects(assoc_array_ptr_to_leaf(node->slots[i]), + index_key); if (x < diff) { BUG_ON(x < 0); diff = x; diff --git a/security/keys/keyring.c b/security/keys/keyring.c index 0adbc77a59b..3dd8445cd48 100644 --- a/security/keys/keyring.c +++ b/security/keys/keyring.c @@ -279,12 +279,11 @@ static bool keyring_compare_object(const void *object, const void *data) * Compare the index keys of a pair of objects and determine the bit position * at which they differ - if they differ. */ -static int keyring_diff_objects(const void *_a, const void *_b) +static int keyring_diff_objects(const void *object, const void *data) { - const struct key *key_a = keyring_ptr_to_key(_a); - const struct key *key_b = keyring_ptr_to_key(_b); + const struct key *key_a = keyring_ptr_to_key(object); const struct keyring_index_key *a = &key_a->index_key; - const struct keyring_index_key *b = &key_b->index_key; + const struct keyring_index_key *b = data; unsigned long seg_a, seg_b; int level, i; -- cgit v1.2.3-70-g09d2 From c7277090927a5e71871e799a355ed2940f6c8fc6 Mon Sep 17 00:00:00 2001 From: Eric Paris Date: Mon, 2 Dec 2013 11:24:19 +0000 Subject: security: shmem: implement kernel private shmem inodes We have a problem where the big_key key storage implementation uses a shmem backed inode to hold the key contents. Because of this detail of implementation LSM checks are being done between processes trying to read the keys and the tmpfs backed inode. The LSM checks are already being handled on the key interface level and should not be enforced at the inode level (since the inode is an implementation detail, not a part of the security model) This patch implements a new function shmem_kernel_file_setup() which returns the equivalent to shmem_file_setup() only the underlying inode has S_PRIVATE set. This means that all LSM checks for the inode in question are skipped. It should only be used for kernel internal operations where the inode is not exposed to userspace without proper LSM checking. It is possible that some other users of shmem_file_setup() should use the new interface, but this has not been explored. Reproducing this bug is a little bit difficult. The steps I used on Fedora are: (1) Turn off selinux enforcing: setenforce 0 (2) Create a huge key k=`dd if=/dev/zero bs=8192 count=1 | keyctl padd big_key test-key @s` (3) Access the key in another context: runcon system_u:system_r:httpd_t:s0-s0:c0.c1023 keyctl print $k >/dev/null (4) Examine the audit logs: ausearch -m AVC -i --subject httpd_t | audit2allow If the last command's output includes a line that looks like: allow httpd_t user_tmpfs_t:file { open read }; There was an inode check between httpd and the tmpfs filesystem. With this patch no such denial will be seen. (NOTE! you should clear your audit log if you have tested for this previously) (Please return you box to enforcing) Signed-off-by: Eric Paris Signed-off-by: David Howells cc: Hugh Dickins cc: linux-mm@kvack.org --- include/linux/shmem_fs.h | 2 ++ mm/shmem.c | 36 +++++++++++++++++++++++++++++------- security/keys/big_key.c | 2 +- 3 files changed, 32 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index 30aa0dc60d7..9d55438bc4a 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -47,6 +47,8 @@ extern int shmem_init(void); extern int shmem_fill_super(struct super_block *sb, void *data, int silent); extern struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags); +extern struct file *shmem_kernel_file_setup(const char *name, loff_t size, + unsigned long flags); extern int shmem_zero_setup(struct vm_area_struct *); extern int shmem_lock(struct file *file, int lock, struct user_struct *user); extern void shmem_unlock_mapping(struct address_space *mapping); diff --git a/mm/shmem.c b/mm/shmem.c index 8297623fcae..902a14842b7 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2918,13 +2918,8 @@ static struct dentry_operations anon_ops = { .d_dname = simple_dname }; -/** - * shmem_file_setup - get an unlinked file living in tmpfs - * @name: name for dentry (to be seen in /proc//maps - * @size: size to be set for the file - * @flags: VM_NORESERVE suppresses pre-accounting of the entire object size - */ -struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags) +static struct file *__shmem_file_setup(const char *name, loff_t size, + unsigned long flags, unsigned int i_flags) { struct file *res; struct inode *inode; @@ -2957,6 +2952,7 @@ struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags if (!inode) goto put_dentry; + inode->i_flags |= i_flags; d_instantiate(path.dentry, inode); inode->i_size = size; clear_nlink(inode); /* It is unlinked */ @@ -2977,6 +2973,32 @@ put_memory: shmem_unacct_size(flags, size); return res; } + +/** + * shmem_kernel_file_setup - get an unlinked file living in tmpfs which must be + * kernel internal. There will be NO LSM permission checks against the + * underlying inode. So users of this interface must do LSM checks at a + * higher layer. The one user is the big_key implementation. LSM checks + * are provided at the key level rather than the inode level. + * @name: name for dentry (to be seen in /proc//maps + * @size: size to be set for the file + * @flags: VM_NORESERVE suppresses pre-accounting of the entire object size + */ +struct file *shmem_kernel_file_setup(const char *name, loff_t size, unsigned long flags) +{ + return __shmem_file_setup(name, size, flags, S_PRIVATE); +} + +/** + * shmem_file_setup - get an unlinked file living in tmpfs + * @name: name for dentry (to be seen in /proc//maps + * @size: size to be set for the file + * @flags: VM_NORESERVE suppresses pre-accounting of the entire object size + */ +struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags) +{ + return __shmem_file_setup(name, size, flags, 0); +} EXPORT_SYMBOL_GPL(shmem_file_setup); /** diff --git a/security/keys/big_key.c b/security/keys/big_key.c index 7f44c3207a9..8137b27d641 100644 --- a/security/keys/big_key.c +++ b/security/keys/big_key.c @@ -70,7 +70,7 @@ int big_key_instantiate(struct key *key, struct key_preparsed_payload *prep) * * TODO: Encrypt the stored data with a temporary key. */ - file = shmem_file_setup("", datalen, 0); + file = shmem_kernel_file_setup("", datalen, 0); if (IS_ERR(file)) { ret = PTR_ERR(file); goto err_quota; -- cgit v1.2.3-70-g09d2 From 35773dac5f862cb1c82ea151eba3e2f6de51ec3e Mon Sep 17 00:00:00 2001 From: David Laight Date: Mon, 11 Nov 2013 12:26:54 +0000 Subject: usb: xhci: Link TRB must not occur within a USB payload burst Section 4.11.7.1 of rev 1.0 of the xhci specification states that a link TRB can only occur at a boundary between underlying USB frames (512 bytes for high speed devices). If this isn't done the USB frames aren't formatted correctly and, for example, the USB3 ethernet ax88179_178a card will stop sending (while still receiving) when running a netperf tcp transmit test with (say) and 8k buffer. This should be a candidate for stable, the ax88179_178a driver defaults to gso and tso enabled so it passes a lot of fragmented skb to the USB stack. Notes from Sarah: Discussion: http://marc.info/?l=linux-usb&m=138384509604981&w=2 This patch fixes a long-standing xHCI driver bug that was revealed by a change in 3.12 in the usb-net driver. Commit 638c5115a794981441246fa8fa5d95c1875af5ba "USBNET: support DMA SG" added support to use bulk endpoint scatter-gather (urb->sg). Only the USB ethernet drivers trigger this bug, because the mass storage driver sends sg list entries in page-sized chunks. This patch only fixes the issue for bulk endpoint scatter-gather. The problem will still occur for periodic endpoints, because hosts will interpret no-op transfers as a request to skip a service interval, which is not what we want. Luckily, the USB core isn't set up for scatter-gather on isochronous endpoints, and no USB drivers use scatter-gather for interrupt endpoints. Document this known limitation so that developers won't try to use urb->sg for interrupt endpoints until this issue is fixed. The more comprehensive fix would be to allow link TRBs in the middle of the endpoint ring and revert this patch, but that fix would touch too much code to be allowed in for stable. This patch should be backported to kernels as old as 3.12, that contain the commit 638c5115a794981441246fa8fa5d95c1875af5ba "USBNET: support DMA SG". Without this patch, the USB network device gets wedged, and stops sending packets. Mark Lord confirms this patch fixes the regression: http://marc.info/?l=linux-netdev&m=138487107625966&w=2 Signed-off-by: David Laight Signed-off-by: Sarah Sharp Tested-by: Mark Lord Cc: stable@vger.kernel.org --- drivers/usb/host/xhci-ring.c | 54 ++++++++++++++++++++++++++++++++++++++++++-- include/linux/usb.h | 2 ++ 2 files changed, 54 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 1e2f3f49584..53c2e296467 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -2973,8 +2973,58 @@ static int prepare_ring(struct xhci_hcd *xhci, struct xhci_ring *ep_ring, } while (1) { - if (room_on_ring(xhci, ep_ring, num_trbs)) - break; + if (room_on_ring(xhci, ep_ring, num_trbs)) { + union xhci_trb *trb = ep_ring->enqueue; + unsigned int usable = ep_ring->enq_seg->trbs + + TRBS_PER_SEGMENT - 1 - trb; + u32 nop_cmd; + + /* + * Section 4.11.7.1 TD Fragments states that a link + * TRB must only occur at the boundary between + * data bursts (eg 512 bytes for 480M). + * While it is possible to split a large fragment + * we don't know the size yet. + * Simplest solution is to fill the trb before the + * LINK with nop commands. + */ + if (num_trbs == 1 || num_trbs <= usable || usable == 0) + break; + + if (ep_ring->type != TYPE_BULK) + /* + * While isoc transfers might have a buffer that + * crosses a 64k boundary it is unlikely. + * Since we can't add NOPs without generating + * gaps in the traffic just hope it never + * happens at the end of the ring. + * This could be fixed by writing a LINK TRB + * instead of the first NOP - however the + * TRB_TYPE_LINK_LE32() calls would all need + * changing to check the ring length. + */ + break; + + if (num_trbs >= TRBS_PER_SEGMENT) { + xhci_err(xhci, "Too many fragments %d, max %d\n", + num_trbs, TRBS_PER_SEGMENT - 1); + return -ENOMEM; + } + + nop_cmd = cpu_to_le32(TRB_TYPE(TRB_TR_NOOP) | + ep_ring->cycle_state); + ep_ring->num_trbs_free -= usable; + do { + trb->generic.field[0] = 0; + trb->generic.field[1] = 0; + trb->generic.field[2] = 0; + trb->generic.field[3] = nop_cmd; + trb++; + } while (--usable); + ep_ring->enqueue = trb; + if (room_on_ring(xhci, ep_ring, num_trbs)) + break; + } if (ep_ring == xhci->cmd_ring) { xhci_err(xhci, "Do not support expand command ring\n"); diff --git a/include/linux/usb.h b/include/linux/usb.h index 7454865ad14..512ab162832 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -1264,6 +1264,8 @@ typedef void (*usb_complete_t)(struct urb *); * @sg: scatter gather buffer list, the buffer size of each element in * the list (except the last) must be divisible by the endpoint's * max packet size if no_sg_constraint isn't set in 'struct usb_bus' + * (FIXME: scatter-gather under xHCI is broken for periodic transfers. + * Do not use urb->sg for interrupt endpoints for now, only bulk.) * @num_mapped_sgs: (internal) number of mapped sg entries * @num_sgs: number of entries in the sg list * @transfer_buffer_length: How big is transfer_buffer. The transfer may -- cgit v1.2.3-70-g09d2 From 8496e85c20e7836b3dec97780e40f420a3ae2801 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 1 Dec 2013 02:34:37 +0100 Subject: PCI / tg3: Give up chip reset and carrier loss handling if PCI device is not present Modify tg3_chip_reset() and tg3_close() to check if the PCI network adapter device is accessible at all in order to skip poking it or trying to handle a carrier loss in vain when that's not the case. Introduce a special PCI helper function pci_device_is_present() for this purpose. Of course, this uncovers the lack of the appropriate RTNL locking in tg3_suspend() and tg3_resume(), so add that locking in there too. These changes prevent tg3 from burning a CPU at 100% load level for solid several seconds after the Thunderbolt link is disconnected from a Matrox DS1 docking station. Signed-off-by: Rafael J. Wysocki Acked-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/tg3.c | 26 +++++++++++++++++++------- drivers/pci/pci.c | 8 ++++++++ include/linux/pci.h | 1 + 3 files changed, 28 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 369b736dde0..472305cdff4 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -8932,6 +8932,9 @@ static int tg3_chip_reset(struct tg3 *tp) void (*write_op)(struct tg3 *, u32, u32); int i, err; + if (!pci_device_is_present(tp->pdev)) + return -ENODEV; + tg3_nvram_lock(tp); tg3_ape_lock(tp, TG3_APE_LOCK_GRC); @@ -11581,10 +11584,11 @@ static int tg3_close(struct net_device *dev) memset(&tp->net_stats_prev, 0, sizeof(tp->net_stats_prev)); memset(&tp->estats_prev, 0, sizeof(tp->estats_prev)); - tg3_power_down_prepare(tp); - - tg3_carrier_off(tp); + if (pci_device_is_present(tp->pdev)) { + tg3_power_down_prepare(tp); + tg3_carrier_off(tp); + } return 0; } @@ -17726,10 +17730,12 @@ static int tg3_suspend(struct device *device) struct pci_dev *pdev = to_pci_dev(device); struct net_device *dev = pci_get_drvdata(pdev); struct tg3 *tp = netdev_priv(dev); - int err; + int err = 0; + + rtnl_lock(); if (!netif_running(dev)) - return 0; + goto unlock; tg3_reset_task_cancel(tp); tg3_phy_stop(tp); @@ -17771,6 +17777,8 @@ out: tg3_phy_start(tp); } +unlock: + rtnl_unlock(); return err; } @@ -17779,10 +17787,12 @@ static int tg3_resume(struct device *device) struct pci_dev *pdev = to_pci_dev(device); struct net_device *dev = pci_get_drvdata(pdev); struct tg3 *tp = netdev_priv(dev); - int err; + int err = 0; + + rtnl_lock(); if (!netif_running(dev)) - return 0; + goto unlock; netif_device_attach(dev); @@ -17806,6 +17816,8 @@ out: if (!err) tg3_phy_start(tp); +unlock: + rtnl_unlock(); return err; } #endif /* CONFIG_PM_SLEEP */ diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 33120d15666..07369f32e8b 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -4165,6 +4165,14 @@ int pci_set_vga_state(struct pci_dev *dev, bool decode, return 0; } +bool pci_device_is_present(struct pci_dev *pdev) +{ + u32 v; + + return pci_bus_read_dev_vendor_id(pdev->bus, pdev->devfn, &v, 0); +} +EXPORT_SYMBOL_GPL(pci_device_is_present); + #define RESOURCE_ALIGNMENT_PARAM_SIZE COMMAND_LINE_SIZE static char resource_alignment_param[RESOURCE_ALIGNMENT_PARAM_SIZE] = {0}; static DEFINE_SPINLOCK(resource_alignment_lock); diff --git a/include/linux/pci.h b/include/linux/pci.h index 1084a15175e..25a389566e4 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -960,6 +960,7 @@ void pci_update_resource(struct pci_dev *dev, int resno); int __must_check pci_assign_resource(struct pci_dev *dev, int i); int __must_check pci_reassign_resource(struct pci_dev *dev, int i, resource_size_t add_size, resource_size_t align); int pci_select_bars(struct pci_dev *dev, unsigned long flags); +bool pci_device_is_present(struct pci_dev *pdev); /* ROM control related routines */ int pci_enable_rom(struct pci_dev *pdev); -- cgit v1.2.3-70-g09d2 From 9f740ffa8134aaef770f964485dac3ed6780d8b7 Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Wed, 27 Nov 2013 22:19:00 +0000 Subject: HID: hid-sensor-hub: Add logical min and max Exporting logical minimum and maximum of HID fields as part of the hid sensor attribute info. This can be used for range checking and to calculate enumeration base for NAry fields of HID sensor hub. Signed-off-by: Srinivas Pandruvada Signed-off-by: Jonathan Cameron --- drivers/hid/hid-sensor-hub.c | 20 ++++++++------------ include/linux/hid-sensor-hub.h | 2 ++ 2 files changed, 10 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/drivers/hid/hid-sensor-hub.c b/drivers/hid/hid-sensor-hub.c index a184e1921c1..d87f7cb4bee 100644 --- a/drivers/hid/hid-sensor-hub.c +++ b/drivers/hid/hid-sensor-hub.c @@ -112,13 +112,15 @@ static int sensor_hub_get_physical_device_count( static void sensor_hub_fill_attr_info( struct hid_sensor_hub_attribute_info *info, - s32 index, s32 report_id, s32 units, s32 unit_expo, s32 size) + s32 index, s32 report_id, struct hid_field *field) { info->index = index; info->report_id = report_id; - info->units = units; - info->unit_expo = unit_expo; - info->size = size/8; + info->units = field->unit; + info->unit_expo = field->unit_exponent; + info->size = (field->report_size * field->report_count)/8; + info->logical_minimum = field->logical_minimum; + info->logical_maximum = field->logical_maximum; } static struct hid_sensor_hub_callbacks *sensor_hub_get_callback( @@ -325,9 +327,7 @@ int sensor_hub_input_get_attribute_info(struct hid_sensor_hub_device *hsdev, if (field->physical == usage_id && field->logical == attr_usage_id) { sensor_hub_fill_attr_info(info, i, report->id, - field->unit, field->unit_exponent, - field->report_size * - field->report_count); + field); ret = 0; } else { for (j = 0; j < field->maxusage; ++j) { @@ -336,11 +336,7 @@ int sensor_hub_input_get_attribute_info(struct hid_sensor_hub_device *hsdev, field->usage[j].collection_index == collection_index) { sensor_hub_fill_attr_info(info, - i, report->id, - field->unit, - field->unit_exponent, - field->report_size * - field->report_count); + i, report->id, field); ret = 0; break; } diff --git a/include/linux/hid-sensor-hub.h b/include/linux/hid-sensor-hub.h index 206a2af6b62..b914ca3f57b 100644 --- a/include/linux/hid-sensor-hub.h +++ b/include/linux/hid-sensor-hub.h @@ -42,6 +42,8 @@ struct hid_sensor_hub_attribute_info { s32 units; s32 unit_expo; s32 size; + s32 logical_minimum; + s32 logical_maximum; }; /** -- cgit v1.2.3-70-g09d2 From 751d17e23a9f7c8e0bca5c0b2e8d39af655ecd2a Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Wed, 27 Nov 2013 22:19:00 +0000 Subject: iio: hid-sensors: Fix power and report state In the original HID sensor hub firmwares all Named array enums were to 0-based. But the most recent hub implemented as 1-based, because of the implementation by one of the major OS vendor. Using logical minimum for the field as the base of enum. So we add logical minimum to the selector values before setting those fields. Some sensor hub FWs already changed logical minimum from 0 to 1 to reflect this and hope every other vendor will follow. There is no easy way to add a common HID quirk for NAry elements, even if the standard specifies these field as NAry, the collection used to describe selectors is still just "logical". Signed-off-by: Srinivas Pandruvada Signed-off-by: Jonathan Cameron --- drivers/iio/common/hid-sensors/Kconfig | 9 --------- drivers/iio/common/hid-sensors/hid-sensor-trigger.c | 20 +++++++++++++++----- include/linux/hid-sensor-ids.h | 12 ++++++++++++ 3 files changed, 27 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/drivers/iio/common/hid-sensors/Kconfig b/drivers/iio/common/hid-sensors/Kconfig index 1178121b55b..39188b72cd3 100644 --- a/drivers/iio/common/hid-sensors/Kconfig +++ b/drivers/iio/common/hid-sensors/Kconfig @@ -25,13 +25,4 @@ config HID_SENSOR_IIO_TRIGGER If this driver is compiled as a module, it will be named hid-sensor-trigger. -config HID_SENSOR_ENUM_BASE_QUIRKS - bool "ENUM base quirks for HID Sensor IIO drivers" - depends on HID_SENSOR_IIO_COMMON - help - Say yes here to build support for sensor hub FW using - enumeration, which is using 1 as base instead of 0. - Since logical minimum is still set 0 instead of 1, - there is no easy way to differentiate. - endmenu diff --git a/drivers/iio/common/hid-sensors/hid-sensor-trigger.c b/drivers/iio/common/hid-sensors/hid-sensor-trigger.c index bbd6426c972..7dcf83998e6 100644 --- a/drivers/iio/common/hid-sensors/hid-sensor-trigger.c +++ b/drivers/iio/common/hid-sensors/hid-sensor-trigger.c @@ -33,24 +33,34 @@ static int hid_sensor_data_rdy_trigger_set_state(struct iio_trigger *trig, { struct hid_sensor_common *st = iio_trigger_get_drvdata(trig); int state_val; + int report_val; if (state) { if (sensor_hub_device_open(st->hsdev)) return -EIO; - } else + state_val = + HID_USAGE_SENSOR_PROP_POWER_STATE_D0_FULL_POWER_ENUM; + report_val = + HID_USAGE_SENSOR_PROP_REPORTING_STATE_ALL_EVENTS_ENUM; + + } else { sensor_hub_device_close(st->hsdev); + state_val = + HID_USAGE_SENSOR_PROP_POWER_STATE_D4_POWER_OFF_ENUM; + report_val = + HID_USAGE_SENSOR_PROP_REPORTING_STATE_NO_EVENTS_ENUM; + } - state_val = state ? 1 : 0; - if (IS_ENABLED(CONFIG_HID_SENSOR_ENUM_BASE_QUIRKS)) - ++state_val; st->data_ready = state; + state_val += st->power_state.logical_minimum; + report_val += st->report_state.logical_minimum; sensor_hub_set_feature(st->hsdev, st->power_state.report_id, st->power_state.index, (s32)state_val); sensor_hub_set_feature(st->hsdev, st->report_state.report_id, st->report_state.index, - (s32)state_val); + (s32)report_val); return 0; } diff --git a/include/linux/hid-sensor-ids.h b/include/linux/hid-sensor-ids.h index 4f945d3ed49..8323775ac21 100644 --- a/include/linux/hid-sensor-ids.h +++ b/include/linux/hid-sensor-ids.h @@ -117,4 +117,16 @@ #define HID_USAGE_SENSOR_PROP_REPORT_STATE 0x200316 #define HID_USAGE_SENSOR_PROY_POWER_STATE 0x200319 +/* Power state enumerations */ +#define HID_USAGE_SENSOR_PROP_POWER_STATE_UNDEFINED_ENUM 0x00 +#define HID_USAGE_SENSOR_PROP_POWER_STATE_D0_FULL_POWER_ENUM 0x01 +#define HID_USAGE_SENSOR_PROP_POWER_STATE_D1_LOW_POWER_ENUM 0x02 +#define HID_USAGE_SENSOR_PROP_POWER_STATE_D2_STANDBY_WITH_WAKE_ENUM 0x03 +#define HID_USAGE_SENSOR_PROP_POWER_STATE_D3_SLEEP_WITH_WAKE_ENUM 0x04 +#define HID_USAGE_SENSOR_PROP_POWER_STATE_D4_POWER_OFF_ENUM 0x05 + +/* Report State enumerations */ +#define HID_USAGE_SENSOR_PROP_REPORTING_STATE_NO_EVENTS_ENUM 0x00 +#define HID_USAGE_SENSOR_PROP_REPORTING_STATE_ALL_EVENTS_ENUM 0x01 + #endif -- cgit v1.2.3-70-g09d2 From 7ce5a27f2ef8ed4f8f892f46391f933c0e2ac0f1 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 2 Dec 2013 17:26:05 -0500 Subject: Revert "net: Handle CHECKSUM_COMPLETE more adequately in pskb_trim_rcsum()." This reverts commit 018c5bba052b3a383d83cf0c756da0e7bc748397. It causes regressions for people using chips driven by the sungem driver. Suspicion is that the skb->csum value isn't being adjusted properly. The change also has a bug in that if __pskb_trim() fails, we'll leave a corruped skb->csum value in there. We would really need to revert it to it's original value in that case. Signed-off-by: David S. Miller --- include/linux/skbuff.h | 39 ++++++++++++++++++--------------------- 1 file changed, 18 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index bec1cc7d5e3..215b5ea1cb3 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2263,6 +2263,24 @@ static inline void skb_postpull_rcsum(struct sk_buff *skb, unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len); +/** + * pskb_trim_rcsum - trim received skb and update checksum + * @skb: buffer to trim + * @len: new length + * + * This is exactly the same as pskb_trim except that it ensures the + * checksum of received packets are still valid after the operation. + */ + +static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len) +{ + if (likely(len >= skb->len)) + return 0; + if (skb->ip_summed == CHECKSUM_COMPLETE) + skb->ip_summed = CHECKSUM_NONE; + return __pskb_trim(skb, len); +} + #define skb_queue_walk(queue, skb) \ for (skb = (queue)->next; \ skb != (struct sk_buff *)(queue); \ @@ -2360,27 +2378,6 @@ __wsum __skb_checksum(const struct sk_buff *skb, int offset, int len, __wsum skb_checksum(const struct sk_buff *skb, int offset, int len, __wsum csum); -/** - * pskb_trim_rcsum - trim received skb and update checksum - * @skb: buffer to trim - * @len: new length - * - * This is exactly the same as pskb_trim except that it ensures the - * checksum of received packets are still valid after the operation. - */ - -static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len) -{ - if (likely(len >= skb->len)) - return 0; - if (skb->ip_summed == CHECKSUM_COMPLETE) { - __wsum adj = skb_checksum(skb, len, skb->len - len, 0); - - skb->csum = csum_sub(skb->csum, adj); - } - return __pskb_trim(skb, len); -} - static inline void *skb_header_pointer(const struct sk_buff *skb, int offset, int len, void *buffer) { -- cgit v1.2.3-70-g09d2 From 471e42ad148c05d091219096726d751684ebf918 Mon Sep 17 00:00:00 2001 From: Thomas Pugliese Date: Mon, 2 Dec 2013 15:39:45 -0600 Subject: usb: wusbcore: fix deadlock in wusbhc_gtk_rekey When multiple wireless USB devices are connected and one of the devices disconnects, the host will distribute a new group key to the remaining devicese using wusbhc_gtk_rekey. wusbhc_gtk_rekey takes the wusbhc->mutex and holds it while it submits a URB to set the new key. This causes a deadlock in wa_urb_enqueue when it calls a device lookup helper function that takes the same lock. This patch changes wusbhc_gtk_rekey to submit a work item to set the GTK so that the URB is submitted without holding wusbhc->mutex. Signed-off-by: Thomas Pugliese Signed-off-by: Greg Kroah-Hartman --- drivers/usb/wusbcore/devconnect.c | 24 +--------- drivers/usb/wusbcore/security.c | 98 +++++++++++++++++++++++---------------- drivers/usb/wusbcore/wusbhc.h | 6 +-- include/linux/usb/wusb.h | 2 + 4 files changed, 62 insertions(+), 68 deletions(-) (limited to 'include') diff --git a/drivers/usb/wusbcore/devconnect.c b/drivers/usb/wusbcore/devconnect.c index 5107ca999d4..f14e7929ba2 100644 --- a/drivers/usb/wusbcore/devconnect.c +++ b/drivers/usb/wusbcore/devconnect.c @@ -97,18 +97,12 @@ static void wusbhc_devconnect_acked_work(struct work_struct *work); static void wusb_dev_free(struct wusb_dev *wusb_dev) { - if (wusb_dev) { - kfree(wusb_dev->set_gtk_req); - usb_free_urb(wusb_dev->set_gtk_urb); - kfree(wusb_dev); - } + kfree(wusb_dev); } static struct wusb_dev *wusb_dev_alloc(struct wusbhc *wusbhc) { struct wusb_dev *wusb_dev; - struct urb *urb; - struct usb_ctrlrequest *req; wusb_dev = kzalloc(sizeof(*wusb_dev), GFP_KERNEL); if (wusb_dev == NULL) @@ -118,22 +112,6 @@ static struct wusb_dev *wusb_dev_alloc(struct wusbhc *wusbhc) INIT_WORK(&wusb_dev->devconnect_acked_work, wusbhc_devconnect_acked_work); - urb = usb_alloc_urb(0, GFP_KERNEL); - if (urb == NULL) - goto err; - wusb_dev->set_gtk_urb = urb; - - req = kmalloc(sizeof(*req), GFP_KERNEL); - if (req == NULL) - goto err; - wusb_dev->set_gtk_req = req; - - req->bRequestType = USB_DIR_OUT | USB_TYPE_STANDARD | USB_RECIP_DEVICE; - req->bRequest = USB_REQ_SET_DESCRIPTOR; - req->wValue = cpu_to_le16(USB_DT_KEY << 8 | wusbhc->gtk_index); - req->wIndex = 0; - req->wLength = cpu_to_le16(wusbhc->gtk.descr.bLength); - return wusb_dev; err: wusb_dev_free(wusb_dev); diff --git a/drivers/usb/wusbcore/security.c b/drivers/usb/wusbcore/security.c index dd88441c8f7..4c40d0dbf53 100644 --- a/drivers/usb/wusbcore/security.c +++ b/drivers/usb/wusbcore/security.c @@ -29,19 +29,16 @@ #include #include "wusbhc.h" -static void wusbhc_set_gtk_callback(struct urb *urb); -static void wusbhc_gtk_rekey_done_work(struct work_struct *work); +static void wusbhc_gtk_rekey_work(struct work_struct *work); int wusbhc_sec_create(struct wusbhc *wusbhc) { wusbhc->gtk.descr.bLength = sizeof(wusbhc->gtk.descr) + sizeof(wusbhc->gtk.data); wusbhc->gtk.descr.bDescriptorType = USB_DT_KEY; wusbhc->gtk.descr.bReserved = 0; + wusbhc->gtk_index = 0; - wusbhc->gtk_index = wusb_key_index(0, WUSB_KEY_INDEX_TYPE_GTK, - WUSB_KEY_INDEX_ORIGINATOR_HOST); - - INIT_WORK(&wusbhc->gtk_rekey_done_work, wusbhc_gtk_rekey_done_work); + INIT_WORK(&wusbhc->gtk_rekey_work, wusbhc_gtk_rekey_work); return 0; } @@ -113,7 +110,7 @@ int wusbhc_sec_start(struct wusbhc *wusbhc) wusbhc_generate_gtk(wusbhc); result = wusbhc->set_gtk(wusbhc, wusbhc->gtk_tkid, - &wusbhc->gtk.descr.bKeyData, key_size); + &wusbhc->gtk.descr.bKeyData, key_size); if (result < 0) dev_err(wusbhc->dev, "cannot set GTK for the host: %d\n", result); @@ -129,7 +126,7 @@ int wusbhc_sec_start(struct wusbhc *wusbhc) */ void wusbhc_sec_stop(struct wusbhc *wusbhc) { - cancel_work_sync(&wusbhc->gtk_rekey_done_work); + cancel_work_sync(&wusbhc->gtk_rekey_work); } @@ -185,12 +182,14 @@ static int wusb_dev_set_encryption(struct usb_device *usb_dev, int value) static int wusb_dev_set_gtk(struct wusbhc *wusbhc, struct wusb_dev *wusb_dev) { struct usb_device *usb_dev = wusb_dev->usb_dev; + u8 key_index = wusb_key_index(wusbhc->gtk_index, + WUSB_KEY_INDEX_TYPE_GTK, WUSB_KEY_INDEX_ORIGINATOR_HOST); return usb_control_msg( usb_dev, usb_sndctrlpipe(usb_dev, 0), USB_REQ_SET_DESCRIPTOR, USB_DIR_OUT | USB_TYPE_STANDARD | USB_RECIP_DEVICE, - USB_DT_KEY << 8 | wusbhc->gtk_index, 0, + USB_DT_KEY << 8 | key_index, 0, &wusbhc->gtk.descr, wusbhc->gtk.descr.bLength, 1000); } @@ -520,24 +519,55 @@ error_kzalloc: * Once all connected and authenticated devices have received the new * GTK, switch the host to using it. */ -static void wusbhc_gtk_rekey_done_work(struct work_struct *work) +static void wusbhc_gtk_rekey_work(struct work_struct *work) { - struct wusbhc *wusbhc = container_of(work, struct wusbhc, gtk_rekey_done_work); + struct wusbhc *wusbhc = container_of(work, + struct wusbhc, gtk_rekey_work); size_t key_size = sizeof(wusbhc->gtk.data); + int port_idx; + struct wusb_dev *wusb_dev, *wusb_dev_next; + LIST_HEAD(rekey_list); mutex_lock(&wusbhc->mutex); + /* generate the new key */ + wusbhc_generate_gtk(wusbhc); + /* roll the gtk index. */ + wusbhc->gtk_index = (wusbhc->gtk_index + 1) % (WUSB_KEY_INDEX_MAX + 1); + /* + * Save all connected devices on a list while holding wusbhc->mutex and + * take a reference to each one. Then submit the set key request to + * them after releasing the lock in order to avoid a deadlock. + */ + for (port_idx = 0; port_idx < wusbhc->ports_max; port_idx++) { + wusb_dev = wusbhc->port[port_idx].wusb_dev; + if (!wusb_dev || !wusb_dev->usb_dev + || !wusb_dev->usb_dev->authenticated) + continue; - if (--wusbhc->pending_set_gtks == 0) - wusbhc->set_gtk(wusbhc, wusbhc->gtk_tkid, &wusbhc->gtk.descr.bKeyData, key_size); - + wusb_dev_get(wusb_dev); + list_add_tail(&wusb_dev->rekey_node, &rekey_list); + } mutex_unlock(&wusbhc->mutex); -} -static void wusbhc_set_gtk_callback(struct urb *urb) -{ - struct wusbhc *wusbhc = urb->context; + /* Submit the rekey requests without holding wusbhc->mutex. */ + list_for_each_entry_safe(wusb_dev, wusb_dev_next, &rekey_list, + rekey_node) { + list_del_init(&wusb_dev->rekey_node); + dev_dbg(&wusb_dev->usb_dev->dev, "%s: rekey device at port %d\n", + __func__, wusb_dev->port_idx); + + if (wusb_dev_set_gtk(wusbhc, wusb_dev) < 0) { + dev_err(&wusb_dev->usb_dev->dev, "%s: rekey device at port %d failed\n", + __func__, wusb_dev->port_idx); + } + wusb_dev_put(wusb_dev); + } - queue_work(wusbd, &wusbhc->gtk_rekey_done_work); + /* Switch the host controller to use the new GTK. */ + mutex_lock(&wusbhc->mutex); + wusbhc->set_gtk(wusbhc, wusbhc->gtk_tkid, + &wusbhc->gtk.descr.bKeyData, key_size); + mutex_unlock(&wusbhc->mutex); } /** @@ -553,26 +583,12 @@ static void wusbhc_set_gtk_callback(struct urb *urb) */ void wusbhc_gtk_rekey(struct wusbhc *wusbhc) { - static const size_t key_size = sizeof(wusbhc->gtk.data); - int p; - - wusbhc_generate_gtk(wusbhc); - - for (p = 0; p < wusbhc->ports_max; p++) { - struct wusb_dev *wusb_dev; - - wusb_dev = wusbhc->port[p].wusb_dev; - if (!wusb_dev || !wusb_dev->usb_dev || !wusb_dev->usb_dev->authenticated) - continue; - - usb_fill_control_urb(wusb_dev->set_gtk_urb, wusb_dev->usb_dev, - usb_sndctrlpipe(wusb_dev->usb_dev, 0), - (void *)wusb_dev->set_gtk_req, - &wusbhc->gtk.descr, wusbhc->gtk.descr.bLength, - wusbhc_set_gtk_callback, wusbhc); - if (usb_submit_urb(wusb_dev->set_gtk_urb, GFP_KERNEL) == 0) - wusbhc->pending_set_gtks++; - } - if (wusbhc->pending_set_gtks == 0) - wusbhc->set_gtk(wusbhc, wusbhc->gtk_tkid, &wusbhc->gtk.descr.bKeyData, key_size); + /* + * We need to submit a URB to the downstream WUSB devices in order to + * change the group key. This can't be done while holding the + * wusbhc->mutex since that is also taken in the urb_enqueue routine + * and will cause a deadlock. Instead, queue a work item to do + * it when the lock is not held + */ + queue_work(wusbd, &wusbhc->gtk_rekey_work); } diff --git a/drivers/usb/wusbcore/wusbhc.h b/drivers/usb/wusbcore/wusbhc.h index 711b1952b11..6bd3b819a6b 100644 --- a/drivers/usb/wusbcore/wusbhc.h +++ b/drivers/usb/wusbcore/wusbhc.h @@ -97,6 +97,7 @@ struct wusb_dev { struct kref refcnt; struct wusbhc *wusbhc; struct list_head cack_node; /* Connect-Ack list */ + struct list_head rekey_node; /* GTK rekey list */ u8 port_idx; u8 addr; u8 beacon_type:4; @@ -107,8 +108,6 @@ struct wusb_dev { struct usb_wireless_cap_descriptor *wusb_cap_descr; struct uwb_mas_bm availability; struct work_struct devconnect_acked_work; - struct urb *set_gtk_urb; - struct usb_ctrlrequest *set_gtk_req; struct usb_device *usb_dev; }; @@ -296,8 +295,7 @@ struct wusbhc { } __attribute__((packed)) gtk; u8 gtk_index; u32 gtk_tkid; - struct work_struct gtk_rekey_done_work; - int pending_set_gtks; + struct work_struct gtk_rekey_work; struct usb_encryption_descriptor *ccm1_etd; }; diff --git a/include/linux/usb/wusb.h b/include/linux/usb/wusb.h index 0c4d4ca370e..eeb28329fa3 100644 --- a/include/linux/usb/wusb.h +++ b/include/linux/usb/wusb.h @@ -271,6 +271,8 @@ static inline u8 wusb_key_index(int index, int type, int originator) #define WUSB_KEY_INDEX_TYPE_GTK 2 #define WUSB_KEY_INDEX_ORIGINATOR_HOST 0 #define WUSB_KEY_INDEX_ORIGINATOR_DEVICE 1 +/* bits 0-3 used for the key index. */ +#define WUSB_KEY_INDEX_MAX 15 /* A CCM Nonce, defined in WUSB1.0[6.4.1] */ struct aes_ccm_nonce { -- cgit v1.2.3-70-g09d2 From 3b5a7ab40c2a1bfd89dd5b4bbfe98f8692d5fd62 Mon Sep 17 00:00:00 2001 From: Ping Cheng Date: Thu, 5 Dec 2013 12:42:09 -0800 Subject: Input: add SW_MUTE_DEVICE switch definition Some devices, such as new Intuos series tablets, have a hardware switch to turn touch data on/off. To report the state, SW_MUTE_DEVICE is added in include/uapi/linux/input.h. Reviewed_by: Chris Bagwell Acked-by: Peter Hutterer Tested-by: Jason Gerecke Signed-off-by: Ping Cheng Signed-off-by: Dmitry Torokhov --- include/uapi/linux/input.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/uapi/linux/input.h b/include/uapi/linux/input.h index 7bacdb51437..ecc88592ecb 100644 --- a/include/uapi/linux/input.h +++ b/include/uapi/linux/input.h @@ -858,6 +858,7 @@ struct input_keymap_entry { #define SW_FRONT_PROXIMITY 0x0b /* set = front proximity sensor active */ #define SW_ROTATE_LOCK 0x0c /* set = rotate locked/disabled */ #define SW_LINEIN_INSERT 0x0d /* set = inserted */ +#define SW_MUTE_DEVICE 0x0e /* set = device disabled */ #define SW_MAX 0x0f #define SW_CNT (SW_MAX+1) -- cgit v1.2.3-70-g09d2 From 1431fb31ecbaba1b5718006128f0f2ed0b94e1c3 Mon Sep 17 00:00:00 2001 From: Paul Durrant Date: Tue, 3 Dec 2013 17:39:29 +0000 Subject: xen-netback: fix fragment detection in checksum setup The code to detect fragments in checksum_setup() was missing for IPv4 and too eager for IPv6. (It transpires that Windows seems to send IPv6 packets with a fragment header even if they are not a fragment - i.e. offset is zero, and M bit is not set). This patch also incorporates a fix to callers of maybe_pull_tail() where skb->network_header was being erroneously added to the length argument. Signed-off-by: Paul Durrant Signed-off-by: Zoltan Kiss Cc: Wei Liu Cc: Ian Campbell Cc: David Vrabel cc: David Miller Acked-by: Wei Liu Signed-off-by: David S. Miller --- drivers/net/xen-netback/netback.c | 236 ++++++++++++++++++++++---------------- include/linux/ipv6.h | 1 + include/net/ipv6.h | 3 +- 3 files changed, 140 insertions(+), 100 deletions(-) (limited to 'include') diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c index 64f0e0d18b8..acf13920e6d 100644 --- a/drivers/net/xen-netback/netback.c +++ b/drivers/net/xen-netback/netback.c @@ -1149,49 +1149,72 @@ static int xenvif_set_skb_gso(struct xenvif *vif, return 0; } -static inline void maybe_pull_tail(struct sk_buff *skb, unsigned int len) +static inline int maybe_pull_tail(struct sk_buff *skb, unsigned int len, + unsigned int max) { - if (skb_is_nonlinear(skb) && skb_headlen(skb) < len) { - /* If we need to pullup then pullup to the max, so we - * won't need to do it again. - */ - int target = min_t(int, skb->len, MAX_TCP_HEADER); - __pskb_pull_tail(skb, target - skb_headlen(skb)); - } + if (skb_headlen(skb) >= len) + return 0; + + /* If we need to pullup then pullup to the max, so we + * won't need to do it again. + */ + if (max > skb->len) + max = skb->len; + + if (__pskb_pull_tail(skb, max - skb_headlen(skb)) == NULL) + return -ENOMEM; + + if (skb_headlen(skb) < len) + return -EPROTO; + + return 0; } +/* This value should be large enough to cover a tagged ethernet header plus + * maximally sized IP and TCP or UDP headers. + */ +#define MAX_IP_HDR_LEN 128 + static int checksum_setup_ip(struct xenvif *vif, struct sk_buff *skb, int recalculate_partial_csum) { - struct iphdr *iph = (void *)skb->data; - unsigned int header_size; unsigned int off; - int err = -EPROTO; + bool fragment; + int err; + + fragment = false; + + err = maybe_pull_tail(skb, + sizeof(struct iphdr), + MAX_IP_HDR_LEN); + if (err < 0) + goto out; - off = sizeof(struct iphdr); + if (ip_hdr(skb)->frag_off & htons(IP_OFFSET | IP_MF)) + fragment = true; - header_size = skb->network_header + off + MAX_IPOPTLEN; - maybe_pull_tail(skb, header_size); + off = ip_hdrlen(skb); - off = iph->ihl * 4; + err = -EPROTO; - switch (iph->protocol) { + switch (ip_hdr(skb)->protocol) { case IPPROTO_TCP: if (!skb_partial_csum_set(skb, off, offsetof(struct tcphdr, check))) goto out; if (recalculate_partial_csum) { - struct tcphdr *tcph = tcp_hdr(skb); - - header_size = skb->network_header + - off + - sizeof(struct tcphdr); - maybe_pull_tail(skb, header_size); - - tcph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, - skb->len - off, - IPPROTO_TCP, 0); + err = maybe_pull_tail(skb, + off + sizeof(struct tcphdr), + MAX_IP_HDR_LEN); + if (err < 0) + goto out; + + tcp_hdr(skb)->check = + ~csum_tcpudp_magic(ip_hdr(skb)->saddr, + ip_hdr(skb)->daddr, + skb->len - off, + IPPROTO_TCP, 0); } break; case IPPROTO_UDP: @@ -1200,24 +1223,20 @@ static int checksum_setup_ip(struct xenvif *vif, struct sk_buff *skb, goto out; if (recalculate_partial_csum) { - struct udphdr *udph = udp_hdr(skb); - - header_size = skb->network_header + - off + - sizeof(struct udphdr); - maybe_pull_tail(skb, header_size); - - udph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, - skb->len - off, - IPPROTO_UDP, 0); + err = maybe_pull_tail(skb, + off + sizeof(struct udphdr), + MAX_IP_HDR_LEN); + if (err < 0) + goto out; + + udp_hdr(skb)->check = + ~csum_tcpudp_magic(ip_hdr(skb)->saddr, + ip_hdr(skb)->daddr, + skb->len - off, + IPPROTO_UDP, 0); } break; default: - if (net_ratelimit()) - netdev_err(vif->dev, - "Attempting to checksum a non-TCP/UDP packet, " - "dropping a protocol %d packet\n", - iph->protocol); goto out; } @@ -1227,75 +1246,99 @@ out: return err; } +/* This value should be large enough to cover a tagged ethernet header plus + * an IPv6 header, all options, and a maximal TCP or UDP header. + */ +#define MAX_IPV6_HDR_LEN 256 + +#define OPT_HDR(type, skb, off) \ + (type *)(skb_network_header(skb) + (off)) + static int checksum_setup_ipv6(struct xenvif *vif, struct sk_buff *skb, int recalculate_partial_csum) { - int err = -EPROTO; - struct ipv6hdr *ipv6h = (void *)skb->data; + int err; u8 nexthdr; - unsigned int header_size; unsigned int off; + unsigned int len; bool fragment; bool done; + fragment = false; done = false; off = sizeof(struct ipv6hdr); - header_size = skb->network_header + off; - maybe_pull_tail(skb, header_size); + err = maybe_pull_tail(skb, off, MAX_IPV6_HDR_LEN); + if (err < 0) + goto out; - nexthdr = ipv6h->nexthdr; + nexthdr = ipv6_hdr(skb)->nexthdr; - while ((off <= sizeof(struct ipv6hdr) + ntohs(ipv6h->payload_len)) && - !done) { + len = sizeof(struct ipv6hdr) + ntohs(ipv6_hdr(skb)->payload_len); + while (off <= len && !done) { switch (nexthdr) { case IPPROTO_DSTOPTS: case IPPROTO_HOPOPTS: case IPPROTO_ROUTING: { - struct ipv6_opt_hdr *hp = (void *)(skb->data + off); + struct ipv6_opt_hdr *hp; - header_size = skb->network_header + - off + - sizeof(struct ipv6_opt_hdr); - maybe_pull_tail(skb, header_size); + err = maybe_pull_tail(skb, + off + + sizeof(struct ipv6_opt_hdr), + MAX_IPV6_HDR_LEN); + if (err < 0) + goto out; + hp = OPT_HDR(struct ipv6_opt_hdr, skb, off); nexthdr = hp->nexthdr; off += ipv6_optlen(hp); break; } case IPPROTO_AH: { - struct ip_auth_hdr *hp = (void *)(skb->data + off); + struct ip_auth_hdr *hp; + + err = maybe_pull_tail(skb, + off + + sizeof(struct ip_auth_hdr), + MAX_IPV6_HDR_LEN); + if (err < 0) + goto out; + + hp = OPT_HDR(struct ip_auth_hdr, skb, off); + nexthdr = hp->nexthdr; + off += ipv6_authlen(hp); + break; + } + case IPPROTO_FRAGMENT: { + struct frag_hdr *hp; - header_size = skb->network_header + - off + - sizeof(struct ip_auth_hdr); - maybe_pull_tail(skb, header_size); + err = maybe_pull_tail(skb, + off + + sizeof(struct frag_hdr), + MAX_IPV6_HDR_LEN); + if (err < 0) + goto out; + + hp = OPT_HDR(struct frag_hdr, skb, off); + + if (hp->frag_off & htons(IP6_OFFSET | IP6_MF)) + fragment = true; nexthdr = hp->nexthdr; - off += (hp->hdrlen+2)<<2; + off += sizeof(struct frag_hdr); break; } - case IPPROTO_FRAGMENT: - fragment = true; - /* fall through */ default: done = true; break; } } - if (!done) { - if (net_ratelimit()) - netdev_err(vif->dev, "Failed to parse packet header\n"); - goto out; - } + err = -EPROTO; - if (fragment) { - if (net_ratelimit()) - netdev_err(vif->dev, "Packet is a fragment!\n"); + if (!done || fragment) goto out; - } switch (nexthdr) { case IPPROTO_TCP: @@ -1304,17 +1347,17 @@ static int checksum_setup_ipv6(struct xenvif *vif, struct sk_buff *skb, goto out; if (recalculate_partial_csum) { - struct tcphdr *tcph = tcp_hdr(skb); - - header_size = skb->network_header + - off + - sizeof(struct tcphdr); - maybe_pull_tail(skb, header_size); - - tcph->check = ~csum_ipv6_magic(&ipv6h->saddr, - &ipv6h->daddr, - skb->len - off, - IPPROTO_TCP, 0); + err = maybe_pull_tail(skb, + off + sizeof(struct tcphdr), + MAX_IPV6_HDR_LEN); + if (err < 0) + goto out; + + tcp_hdr(skb)->check = + ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, + &ipv6_hdr(skb)->daddr, + skb->len - off, + IPPROTO_TCP, 0); } break; case IPPROTO_UDP: @@ -1323,25 +1366,20 @@ static int checksum_setup_ipv6(struct xenvif *vif, struct sk_buff *skb, goto out; if (recalculate_partial_csum) { - struct udphdr *udph = udp_hdr(skb); - - header_size = skb->network_header + - off + - sizeof(struct udphdr); - maybe_pull_tail(skb, header_size); - - udph->check = ~csum_ipv6_magic(&ipv6h->saddr, - &ipv6h->daddr, - skb->len - off, - IPPROTO_UDP, 0); + err = maybe_pull_tail(skb, + off + sizeof(struct udphdr), + MAX_IPV6_HDR_LEN); + if (err < 0) + goto out; + + udp_hdr(skb)->check = + ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, + &ipv6_hdr(skb)->daddr, + skb->len - off, + IPPROTO_UDP, 0); } break; default: - if (net_ratelimit()) - netdev_err(vif->dev, - "Attempting to checksum a non-TCP/UDP packet, " - "dropping a protocol %d packet\n", - nexthdr); goto out; } diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 5d89d1b808a..c56c350324e 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -4,6 +4,7 @@ #include #define ipv6_optlen(p) (((p)->hdrlen+1) << 3) +#define ipv6_authlen(p) (((p)->hdrlen+2) << 2) /* * This structure contains configuration options per IPv6 link. */ diff --git a/include/net/ipv6.h b/include/net/ipv6.h index eb198acaac1..488316e339a 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -110,7 +110,8 @@ struct frag_hdr { __be32 identification; }; -#define IP6_MF 0x0001 +#define IP6_MF 0x0001 +#define IP6_OFFSET 0xFFF8 #include -- cgit v1.2.3-70-g09d2 From 7f2cbdc28c034ef2c3be729681f631d5744e3cd5 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 4 Dec 2013 20:12:04 -0800 Subject: tcp_memcontrol: Cleanup/fix cg_proto->memory_pressure handling. kill memcg_tcp_enter_memory_pressure. The only function of memcg_tcp_enter_memory_pressure was to reduce deal with the unnecessary abstraction that was tcp_memcontrol. Now that struct tcp_memcontrol is gone remove this unnecessary function, the unnecessary function pointer, and modify sk_enter_memory_pressure to set this field directly, just as sk_leave_memory_pressure cleas this field directly. This fixes a small bug I intruduced when killing struct tcp_memcontrol that caused memcg_tcp_enter_memory_pressure to never be called and thus failed to ever set cg_proto->memory_pressure. Remove the cg_proto enter_memory_pressure function as it now serves no useful purpose. Don't test cg_proto->memory_presser in sk_leave_memory_pressure before clearing it. The test was originally there to ensure that the pointer was non-NULL. Now that cg_proto is not a pointer the pointer does not matter. Signed-off-by: "Eric W. Biederman" Signed-off-by: David S. Miller --- include/net/sock.h | 6 ++---- net/ipv4/tcp_memcontrol.c | 7 ------- 2 files changed, 2 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index e3a18ff0c38..2ef3c3eca47 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1035,7 +1035,6 @@ enum cg_proto_flags { }; struct cg_proto { - void (*enter_memory_pressure)(struct sock *sk); struct res_counter memory_allocated; /* Current allocated memory. */ struct percpu_counter sockets_allocated; /* Current number of sockets. */ int memory_pressure; @@ -1155,8 +1154,7 @@ static inline void sk_leave_memory_pressure(struct sock *sk) struct proto *prot = sk->sk_prot; for (; cg_proto; cg_proto = parent_cg_proto(prot, cg_proto)) - if (cg_proto->memory_pressure) - cg_proto->memory_pressure = 0; + cg_proto->memory_pressure = 0; } } @@ -1171,7 +1169,7 @@ static inline void sk_enter_memory_pressure(struct sock *sk) struct proto *prot = sk->sk_prot; for (; cg_proto; cg_proto = parent_cg_proto(prot, cg_proto)) - cg_proto->enter_memory_pressure(sk); + cg_proto->memory_pressure = 1; } sk->sk_prot->enter_memory_pressure(sk); diff --git a/net/ipv4/tcp_memcontrol.c b/net/ipv4/tcp_memcontrol.c index 269a89ecd2f..f7e522c558b 100644 --- a/net/ipv4/tcp_memcontrol.c +++ b/net/ipv4/tcp_memcontrol.c @@ -6,13 +6,6 @@ #include #include -static void memcg_tcp_enter_memory_pressure(struct sock *sk) -{ - if (sk->sk_cgrp->memory_pressure) - sk->sk_cgrp->memory_pressure = 1; -} -EXPORT_SYMBOL(memcg_tcp_enter_memory_pressure); - int tcp_init_cgroup(struct mem_cgroup *memcg, struct cgroup_subsys *ss) { /* -- cgit v1.2.3-70-g09d2 From 4fc9bbf98fd66f879e628d8537ba7c240be2b58e Mon Sep 17 00:00:00 2001 From: Khalid Aziz Date: Wed, 27 Nov 2013 15:19:25 -0700 Subject: PCI: Disable Bus Master only on kexec reboot Add a flag to tell the PCI subsystem that kernel is shutting down in preparation to kexec a kernel. Add code in PCI subsystem to use this flag to clear Bus Master bit on PCI devices only in case of kexec reboot. This fixes a power-off problem on Acer Aspire V5-573G and likely other machines and avoids any other issues caused by clearing Bus Master bit on PCI devices in normal shutdown path. The problem was introduced by b566a22c2332 ("PCI: disable Bus Master on PCI device shutdown"). This patch is based on discussion at http://marc.info/?l=linux-pci&m=138425645204355&w=2 Link: https://bugzilla.kernel.org/show_bug.cgi?id=63861 Reported-by: Chang Liu Signed-off-by: Khalid Aziz Signed-off-by: Bjorn Helgaas Acked-by: Konstantin Khlebnikov Cc: stable@vger.kernel.org # v3.5+ --- drivers/pci/pci-driver.c | 12 +++++++++--- include/linux/kexec.h | 3 +++ kernel/kexec.c | 4 ++++ 3 files changed, 16 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 7edd5c30744..25f0bc65916 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -19,6 +19,7 @@ #include #include #include +#include #include "pci.h" struct pci_dynid { @@ -415,12 +416,17 @@ static void pci_device_shutdown(struct device *dev) pci_msi_shutdown(pci_dev); pci_msix_shutdown(pci_dev); +#ifdef CONFIG_KEXEC /* - * Turn off Bus Master bit on the device to tell it to not - * continue to do DMA. Don't touch devices in D3cold or unknown states. + * If this is a kexec reboot, turn off Bus Master bit on the + * device to tell it to not continue to do DMA. Don't touch + * devices in D3cold or unknown states. + * If it is not a kexec reboot, firmware will hit the PCI + * devices with big hammer and stop their DMA any way. */ - if (pci_dev->current_state <= PCI_D3hot) + if (kexec_in_progress && (pci_dev->current_state <= PCI_D3hot)) pci_clear_master(pci_dev); +#endif } #ifdef CONFIG_PM diff --git a/include/linux/kexec.h b/include/linux/kexec.h index d78d28a733b..5fd33dc1fe3 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -198,6 +198,9 @@ extern u32 vmcoreinfo_note[VMCOREINFO_NOTE_SIZE/4]; extern size_t vmcoreinfo_size; extern size_t vmcoreinfo_max_size; +/* flag to track if kexec reboot is in progress */ +extern bool kexec_in_progress; + int __init parse_crashkernel(char *cmdline, unsigned long long system_ram, unsigned long long *crash_size, unsigned long long *crash_base); int parse_crashkernel_high(char *cmdline, unsigned long long system_ram, diff --git a/kernel/kexec.c b/kernel/kexec.c index 490afc03627..d0d8fca5406 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -47,6 +47,9 @@ u32 vmcoreinfo_note[VMCOREINFO_NOTE_SIZE/4]; size_t vmcoreinfo_size; size_t vmcoreinfo_max_size = sizeof(vmcoreinfo_data); +/* Flag to indicate we are going to kexec a new kernel */ +bool kexec_in_progress = false; + /* Location of the reserved area for the crash kernel */ struct resource crashk_res = { .name = "Crash kernel", @@ -1675,6 +1678,7 @@ int kernel_kexec(void) } else #endif { + kexec_in_progress = true; kernel_restart_prepare(NULL); printk(KERN_EMERG "Starting new kernel\n"); machine_shutdown(); -- cgit v1.2.3-70-g09d2 From 12205a4b79bef56ef618a4b7caa840c5c971cff2 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 8 Dec 2013 01:04:17 +0100 Subject: Revert "cpufreq: suspend governors on system suspend/hibernate" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 5a87182aa21d (cpufreq: suspend governors on system suspend/hibernate) causes hibernation problems to happen on Bjørn Mork's and Paul Bolle's systems, so revert it. Fixes: 5a87182aa21d (cpufreq: suspend governors on system suspend/hibernate) Reported-by: Bjørn Mork Reported-by: Paul Bolle Signed-off-by: Rafael J. Wysocki --- drivers/base/power/main.c | 3 --- drivers/cpufreq/cpufreq.c | 43 ------------------------------------------- include/linux/cpufreq.h | 8 -------- 3 files changed, 54 deletions(-) (limited to 'include') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index e3219dfd736..1b41fca3d65 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -29,7 +29,6 @@ #include #include #include -#include #include #include @@ -541,7 +540,6 @@ static void dpm_resume_noirq(pm_message_t state) dpm_show_time(starttime, state, "noirq"); resume_device_irqs(); cpuidle_resume(); - cpufreq_resume(); } /** @@ -957,7 +955,6 @@ static int dpm_suspend_noirq(pm_message_t state) ktime_t starttime = ktime_get(); int error = 0; - cpufreq_suspend(); cpuidle_pause(); suspend_device_irqs(); mutex_lock(&dpm_list_mtx); diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 81e9d4412db..b7c3b877da4 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -26,7 +26,6 @@ #include #include #include -#include #include #include #include @@ -48,9 +47,6 @@ static LIST_HEAD(cpufreq_policy_list); static DEFINE_PER_CPU(char[CPUFREQ_NAME_LEN], cpufreq_cpu_governor); #endif -/* Flag to suspend/resume CPUFreq governors */ -static bool cpufreq_suspended; - static inline bool has_target(void) { return cpufreq_driver->target_index || cpufreq_driver->target; @@ -1466,41 +1462,6 @@ static struct subsys_interface cpufreq_interface = { .remove_dev = cpufreq_remove_dev, }; -void cpufreq_suspend(void) -{ - struct cpufreq_policy *policy; - - if (!has_target()) - return; - - pr_debug("%s: Suspending Governors\n", __func__); - - list_for_each_entry(policy, &cpufreq_policy_list, policy_list) - if (__cpufreq_governor(policy, CPUFREQ_GOV_STOP)) - pr_err("%s: Failed to stop governor for policy: %p\n", - __func__, policy); - - cpufreq_suspended = true; -} - -void cpufreq_resume(void) -{ - struct cpufreq_policy *policy; - - if (!has_target()) - return; - - pr_debug("%s: Resuming Governors\n", __func__); - - cpufreq_suspended = false; - - list_for_each_entry(policy, &cpufreq_policy_list, policy_list) - if (__cpufreq_governor(policy, CPUFREQ_GOV_START) - || __cpufreq_governor(policy, CPUFREQ_GOV_LIMITS)) - pr_err("%s: Failed to start governor for policy: %p\n", - __func__, policy); -} - /** * cpufreq_bp_suspend - Prepare the boot CPU for system suspend. * @@ -1803,10 +1764,6 @@ static int __cpufreq_governor(struct cpufreq_policy *policy, struct cpufreq_governor *gov = NULL; #endif - /* Don't start any governor operations if we are entering suspend */ - if (cpufreq_suspended) - return 0; - if (policy->governor->max_transition_latency && policy->cpuinfo.transition_latency > policy->governor->max_transition_latency) { diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index ee5fe9d77ae..dc196bbcf22 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -280,14 +280,6 @@ cpufreq_verify_within_cpu_limits(struct cpufreq_policy *policy) policy->cpuinfo.max_freq); } -#ifdef CONFIG_CPU_FREQ -void cpufreq_suspend(void); -void cpufreq_resume(void); -#else -static inline void cpufreq_suspend(void) {} -static inline void cpufreq_resume(void) {} -#endif - /********************************************************************* * CPUFREQ NOTIFIER INTERFACE * *********************************************************************/ -- cgit v1.2.3-70-g09d2 From 11e6a09fba446ce799a9de51d2d99586024bbdaa Mon Sep 17 00:00:00 2001 From: Jakob Bornecrantz Date: Sun, 8 Dec 2013 23:39:04 -0800 Subject: drm/vmwgfx: Add max surface memory param Userspace uses this to workaround overcommit issues by flushing the command stream early. Signed-off-by: Jakob Bornecrantz Reviewed-by: Thomas Hellstrom --- drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c | 3 +++ include/uapi/drm/vmwgfx_drm.h | 1 + 2 files changed, 4 insertions(+) (limited to 'include') diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c index a51f48e3e91..45d5b5ab6ca 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c @@ -68,6 +68,9 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, SVGA_FIFO_3D_HWVERSION)); break; } + case DRM_VMW_PARAM_MAX_SURF_MEMORY: + param->value = dev_priv->memory_size; + break; default: DRM_ERROR("Illegal vmwgfx get param request: %d\n", param->param); diff --git a/include/uapi/drm/vmwgfx_drm.h b/include/uapi/drm/vmwgfx_drm.h index bcb0912afe7..f854ca4a137 100644 --- a/include/uapi/drm/vmwgfx_drm.h +++ b/include/uapi/drm/vmwgfx_drm.h @@ -75,6 +75,7 @@ #define DRM_VMW_PARAM_FIFO_CAPS 4 #define DRM_VMW_PARAM_MAX_FB_SIZE 5 #define DRM_VMW_PARAM_FIFO_HW_VERSION 6 +#define DRM_VMW_PARAM_MAX_SURF_MEMORY 7 /** * struct drm_vmw_getparam_arg -- cgit v1.2.3-70-g09d2 From 389a5390583a18e45bc4abd4439291abec5e7a63 Mon Sep 17 00:00:00 2001 From: Tom Lendacky Date: Thu, 5 Dec 2013 13:09:53 -0600 Subject: crypto: scatterwalk - Use sg_chain_ptr on chain entries Now that scatterwalk_sg_chain sets the chain pointer bit the sg_page call in scatterwalk_sg_next hits a BUG_ON when CONFIG_DEBUG_SG is enabled. Use sg_chain_ptr instead of sg_page on a chain entry. Cc: stable@vger.kernel.org Signed-off-by: Tom Lendacky Signed-off-by: Herbert Xu --- include/crypto/scatterwalk.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/crypto/scatterwalk.h b/include/crypto/scatterwalk.h index 64ebede184f..6a626a507b8 100644 --- a/include/crypto/scatterwalk.h +++ b/include/crypto/scatterwalk.h @@ -44,7 +44,7 @@ static inline struct scatterlist *scatterwalk_sg_next(struct scatterlist *sg) if (sg_is_last(sg)) return NULL; - return (++sg)->length ? sg : (void *)sg_page(sg); + return (++sg)->length ? sg : sg_chain_ptr(sg); } static inline void scatterwalk_crypto_chain(struct scatterlist *head, -- cgit v1.2.3-70-g09d2 From c1b96a236e94d49d9396d0bbceb5524519c5c66e Mon Sep 17 00:00:00 2001 From: Philipp Zabel Date: Tue, 21 May 2013 05:11:35 -0300 Subject: [media] videobuf2: Add support for file access mode flags for DMABUF exporting Currently it is not possible for userspace to map a DMABUF exported buffer with write permissions. This patch allows to also pass O_RDONLY/O_RDWR when exporting the buffer, so that userspace may map it with write permissions. Signed-off-by: Philipp Zabel Signed-off-by: Sylwester Nawrocki Signed-off-by: Mauro Carvalho Chehab --- Documentation/DocBook/media/v4l/vidioc-expbuf.xml | 8 +++++--- drivers/media/v4l2-core/videobuf2-core.c | 8 ++++---- drivers/media/v4l2-core/videobuf2-dma-contig.c | 4 ++-- include/media/videobuf2-core.h | 2 +- 4 files changed, 12 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/Documentation/DocBook/media/v4l/vidioc-expbuf.xml b/Documentation/DocBook/media/v4l/vidioc-expbuf.xml index e287c8fc803..4165e7bfa4f 100644 --- a/Documentation/DocBook/media/v4l/vidioc-expbuf.xml +++ b/Documentation/DocBook/media/v4l/vidioc-expbuf.xml @@ -73,7 +73,8 @@ range from zero to the maximal number of valid planes for the currently active format. For the single-planar API, applications must set plane to zero. Additional flags may be posted in the flags field. Refer to a manual for open() for details. -Currently only O_CLOEXEC is supported. All other fields must be set to zero. +Currently only O_CLOEXEC, O_RDONLY, O_WRONLY, and O_RDWR are supported. All +other fields must be set to zero. In the case of multi-planar API, every plane is exported separately using multiple VIDIOC_EXPBUF calls. @@ -170,8 +171,9 @@ multi-planar API. Otherwise this value must be set to zero. __u32 flags Flags for the newly created file, currently only -O_CLOEXEC is supported, refer to the manual of open() for more -details. +O_CLOEXEC , O_RDONLY, O_WRONLY +, and O_RDWR are supported, refer to the manual +of open() for more details. __s32 diff --git a/drivers/media/v4l2-core/videobuf2-core.c b/drivers/media/v4l2-core/videobuf2-core.c index b19b306c8f7..57ba131f08a 100644 --- a/drivers/media/v4l2-core/videobuf2-core.c +++ b/drivers/media/v4l2-core/videobuf2-core.c @@ -1824,8 +1824,8 @@ int vb2_expbuf(struct vb2_queue *q, struct v4l2_exportbuffer *eb) return -EINVAL; } - if (eb->flags & ~O_CLOEXEC) { - dprintk(1, "Queue does support only O_CLOEXEC flag\n"); + if (eb->flags & ~(O_CLOEXEC | O_ACCMODE)) { + dprintk(1, "Queue does support only O_CLOEXEC and access mode flags\n"); return -EINVAL; } @@ -1848,14 +1848,14 @@ int vb2_expbuf(struct vb2_queue *q, struct v4l2_exportbuffer *eb) vb_plane = &vb->planes[eb->plane]; - dbuf = call_memop(q, get_dmabuf, vb_plane->mem_priv); + dbuf = call_memop(q, get_dmabuf, vb_plane->mem_priv, eb->flags & O_ACCMODE); if (IS_ERR_OR_NULL(dbuf)) { dprintk(1, "Failed to export buffer %d, plane %d\n", eb->index, eb->plane); return -EINVAL; } - ret = dma_buf_fd(dbuf, eb->flags); + ret = dma_buf_fd(dbuf, eb->flags & ~O_ACCMODE); if (ret < 0) { dprintk(3, "buffer %d, plane %d failed to export (%d)\n", eb->index, eb->plane, ret); diff --git a/drivers/media/v4l2-core/videobuf2-dma-contig.c b/drivers/media/v4l2-core/videobuf2-dma-contig.c index 646f08f4f50..33d3871d1e1 100644 --- a/drivers/media/v4l2-core/videobuf2-dma-contig.c +++ b/drivers/media/v4l2-core/videobuf2-dma-contig.c @@ -393,7 +393,7 @@ static struct sg_table *vb2_dc_get_base_sgt(struct vb2_dc_buf *buf) return sgt; } -static struct dma_buf *vb2_dc_get_dmabuf(void *buf_priv) +static struct dma_buf *vb2_dc_get_dmabuf(void *buf_priv, unsigned long flags) { struct vb2_dc_buf *buf = buf_priv; struct dma_buf *dbuf; @@ -404,7 +404,7 @@ static struct dma_buf *vb2_dc_get_dmabuf(void *buf_priv) if (WARN_ON(!buf->sgt_base)) return NULL; - dbuf = dma_buf_export(buf, &vb2_dc_dmabuf_ops, buf->size, 0); + dbuf = dma_buf_export(buf, &vb2_dc_dmabuf_ops, buf->size, flags); if (IS_ERR(dbuf)) return NULL; diff --git a/include/media/videobuf2-core.h b/include/media/videobuf2-core.h index bd8218b1500..941055e9d12 100644 --- a/include/media/videobuf2-core.h +++ b/include/media/videobuf2-core.h @@ -83,7 +83,7 @@ struct vb2_fileio_data; struct vb2_mem_ops { void *(*alloc)(void *alloc_ctx, unsigned long size, gfp_t gfp_flags); void (*put)(void *buf_priv); - struct dma_buf *(*get_dmabuf)(void *buf_priv); + struct dma_buf *(*get_dmabuf)(void *buf_priv, unsigned long flags); void *(*get_userptr)(void *alloc_ctx, unsigned long vaddr, unsigned long size, int write); -- cgit v1.2.3-70-g09d2 From 932e9dec380c67ec15ac3eb073bb55797d8b4801 Mon Sep 17 00:00:00 2001 From: Stefano Panella Date: Tue, 10 Dec 2013 14:20:28 +0000 Subject: ALSA: memalloc.h - fix wrong truncation of dma_addr_t When running a 32bit kernel the hda_intel driver is still reporting a 64bit dma_mask if the HW supports it. From sound/pci/hda/hda_intel.c: /* allow 64bit DMA address if supported by H/W */ if ((gcap & ICH6_GCAP_64OK) && !pci_set_dma_mask(pci, DMA_BIT_MASK(64))) pci_set_consistent_dma_mask(pci, DMA_BIT_MASK(64)); else { pci_set_dma_mask(pci, DMA_BIT_MASK(32)); pci_set_consistent_dma_mask(pci, DMA_BIT_MASK(32)); } which means when there is a call to dma_alloc_coherent from snd_malloc_dev_pages a machine address bigger than 32bit can be returned. This can be true in particular if running the 32bit kernel as a pv dom0 under the Xen Hypervisor or PAE on bare metal. The problem is that when calling setup_bdle to program the BLE the dma_addr_t returned from the dma_alloc_coherent is wrongly truncated from snd_sgbuf_get_addr if running a 32bit kernel: static inline dma_addr_t snd_sgbuf_get_addr(struct snd_dma_buffer *dmab, size_t offset) { struct snd_sg_buf *sgbuf = dmab->private_data; dma_addr_t addr = sgbuf->table[offset >> PAGE_SHIFT].addr; addr &= PAGE_MASK; return addr + offset % PAGE_SIZE; } where PAGE_MASK in a 32bit kernel is zeroing the upper 32bit af addr. Without this patch the HW will fetch the 32bit truncated address, which is not the one obtained from dma_alloc_coherent and will result to a non working audio but can corrupt host memory at a random location. The current patch apply to v3.13-rc3-74-g6c843f5 Signed-off-by: Stefano Panella Reviewed-by: Frediano Ziglio Cc: Signed-off-by: Takashi Iwai --- include/sound/memalloc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/sound/memalloc.h b/include/sound/memalloc.h index af998397041..5f73785f597 100644 --- a/include/sound/memalloc.h +++ b/include/sound/memalloc.h @@ -108,7 +108,7 @@ static inline dma_addr_t snd_sgbuf_get_addr(struct snd_dma_buffer *dmab, { struct snd_sg_buf *sgbuf = dmab->private_data; dma_addr_t addr = sgbuf->table[offset >> PAGE_SHIFT].addr; - addr &= PAGE_MASK; + addr &= ~((dma_addr_t)PAGE_SIZE - 1); return addr + offset % PAGE_SIZE; } -- cgit v1.2.3-70-g09d2 From 6733cf572a9e20db2b7580a5dd39d5782d571eec Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 10 Dec 2013 12:15:52 +0100 Subject: ALSA: compress: Fix 64bit ABI incompatibility snd_pcm_uframes_t is defined as unsigned long so it would take different sizes depending on 32 or 64bit architectures. As we don't want this ABI incompatibility, and there is no real 64bit user yet, let's make it the fixed size with __u32. Also bump the protocol version number to 0.1.2. Acked-by: Vinod Koul Cc: Signed-off-by: Takashi Iwai --- include/uapi/sound/compress_offload.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/uapi/sound/compress_offload.h b/include/uapi/sound/compress_offload.h index d630163b9a2..5759810e1c1 100644 --- a/include/uapi/sound/compress_offload.h +++ b/include/uapi/sound/compress_offload.h @@ -30,7 +30,7 @@ #include -#define SNDRV_COMPRESS_VERSION SNDRV_PROTOCOL_VERSION(0, 1, 1) +#define SNDRV_COMPRESS_VERSION SNDRV_PROTOCOL_VERSION(0, 1, 2) /** * struct snd_compressed_buffer: compressed buffer * @fragment_size: size of buffer fragment in bytes @@ -67,8 +67,8 @@ struct snd_compr_params { struct snd_compr_tstamp { __u32 byte_offset; __u32 copied_total; - snd_pcm_uframes_t pcm_frames; - snd_pcm_uframes_t pcm_io_frames; + __u32 pcm_frames; + __u32 pcm_io_frames; __u32 sampling_rate; }; -- cgit v1.2.3-70-g09d2 From 503cf95c061a0551eb684da364509297efbe55d9 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Tue, 10 Dec 2013 14:56:06 -0800 Subject: x86, build, icc: Remove uninitialized_var() from compiler-intel.h When compiling with icc, ends up included because the icc environment defines __GNUC__. Thus, we neither need nor want to have this macro defined in both compiler-gcc.h and compiler-intel.h, and the fact that they are inconsistent just makes the compiler spew warnings. Reported-by: Sunil K. Pandey Cc: Kevin B. Smith Signed-off-by: H. Peter Anvin Link: http://lkml.kernel.org/n/tip-0mbwou1zt7pafij09b897lg3@git.kernel.org Cc: --- include/linux/compiler-intel.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/linux/compiler-intel.h b/include/linux/compiler-intel.h index 973ce10c40b..dc1bd3dcf11 100644 --- a/include/linux/compiler-intel.h +++ b/include/linux/compiler-intel.h @@ -28,8 +28,6 @@ #endif -#define uninitialized_var(x) x - #ifndef __HAVE_BUILTIN_BSWAP16__ /* icc has this, but it's called _bswap16 */ #define __HAVE_BUILTIN_BSWAP16__ -- cgit v1.2.3-70-g09d2 From 12663bfc97c8b3fdb292428105dd92d563164050 Mon Sep 17 00:00:00 2001 From: Sasha Levin Date: Sat, 7 Dec 2013 17:26:27 -0500 Subject: net: unix: allow set_peek_off to fail unix_dgram_recvmsg() will hold the readlock of the socket until recv is complete. In the same time, we may try to setsockopt(SO_PEEK_OFF) which will hang until unix_dgram_recvmsg() will complete (which can take a while) without allowing us to break out of it, triggering a hung task spew. Instead, allow set_peek_off to fail, this way userspace will not hang. Signed-off-by: Sasha Levin Acked-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/linux/net.h | 2 +- net/core/sock.c | 2 +- net/unix/af_unix.c | 8 ++++++-- 3 files changed, 8 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/net.h b/include/linux/net.h index 4bcee94cef9..69be3e6079c 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -181,7 +181,7 @@ struct proto_ops { int offset, size_t size, int flags); ssize_t (*splice_read)(struct socket *sock, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags); - void (*set_peek_off)(struct sock *sk, int val); + int (*set_peek_off)(struct sock *sk, int val); }; #define DECLARE_SOCKADDR(type, dst, src) \ diff --git a/net/core/sock.c b/net/core/sock.c index ab20ed9b0f3..5393b4b719d 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -882,7 +882,7 @@ set_rcvbuf: case SO_PEEK_OFF: if (sock->ops->set_peek_off) - sock->ops->set_peek_off(sk, val); + ret = sock->ops->set_peek_off(sk, val); else ret = -EOPNOTSUPP; break; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 01625ccc3ae..a0ca162e5bd 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -530,13 +530,17 @@ static int unix_seqpacket_sendmsg(struct kiocb *, struct socket *, static int unix_seqpacket_recvmsg(struct kiocb *, struct socket *, struct msghdr *, size_t, int); -static void unix_set_peek_off(struct sock *sk, int val) +static int unix_set_peek_off(struct sock *sk, int val) { struct unix_sock *u = unix_sk(sk); - mutex_lock(&u->readlock); + if (mutex_lock_interruptible(&u->readlock)) + return -EINTR; + sk->sk_peek_off = val; mutex_unlock(&u->readlock); + + return 0; } -- cgit v1.2.3-70-g09d2 From 9f70f46bd4c7267d48ef461a1d613ec9ec0d520c Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Tue, 10 Dec 2013 06:48:15 -0500 Subject: sctp: properly latch and use autoclose value from sock to association Currently, sctp associations latch a sockets autoclose value to an association at association init time, subject to capping constraints from the max_autoclose sysctl value. This leads to an odd situation where an application may set a socket level autoclose timeout, but sliently sctp will limit the autoclose timeout to something less than that. Fix this by modifying the autoclose setsockopt function to check the limit, cap it and warn the user via syslog that the timeout is capped. This will allow getsockopt to return valid autoclose timeout values that reflect what subsequent associations actually use. While were at it, also elimintate the assoc->autoclose variable, it duplicates whats in the timeout array, which leads to multiple sources for the same information, that may differ (as the former isn't subject to any capping). This gives us the timeout information in a canonical place and saves some space in the association structure as well. Signed-off-by: Neil Horman Acked-by: Vlad Yasevich CC: Wang Weidong CC: David Miller CC: Vlad Yasevich CC: netdev@vger.kernel.org Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 6 ------ net/sctp/associola.c | 5 +---- net/sctp/output.c | 3 ++- net/sctp/sm_statefuns.c | 12 ++++++------ net/sctp/socket.c | 4 ++++ 5 files changed, 13 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index ea0ca5f6e62..67b5d006827 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1726,12 +1726,6 @@ struct sctp_association { /* How many duplicated TSNs have we seen? */ int numduptsns; - /* Number of seconds of idle time before an association is closed. - * In the association context, this is really used as a boolean - * since the real timeout is stored in the timeouts array - */ - __u32 autoclose; - /* These are to support * "SCTP Extensions for Dynamic Reconfiguration of IP Addresses * and Enforcement of Flow and Message Limits" diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 68a27f9796d..31ed008c8e1 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -154,8 +154,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a asoc->timeouts[SCTP_EVENT_TIMEOUT_HEARTBEAT] = 0; asoc->timeouts[SCTP_EVENT_TIMEOUT_SACK] = asoc->sackdelay; - asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE] = - min_t(unsigned long, sp->autoclose, net->sctp.max_autoclose) * HZ; + asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE] = sp->autoclose * HZ; /* Initializes the timers */ for (i = SCTP_EVENT_TIMEOUT_NONE; i < SCTP_NUM_TIMEOUT_TYPES; ++i) @@ -291,8 +290,6 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a asoc->peer.ipv6_address = 1; INIT_LIST_HEAD(&asoc->asocs); - asoc->autoclose = sp->autoclose; - asoc->default_stream = sp->default_stream; asoc->default_ppid = sp->default_ppid; asoc->default_flags = sp->default_flags; diff --git a/net/sctp/output.c b/net/sctp/output.c index 0e2644d0a77..0fb140f8f08 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -581,7 +581,8 @@ int sctp_packet_transmit(struct sctp_packet *packet) unsigned long timeout; /* Restart the AUTOCLOSE timer when sending data. */ - if (sctp_state(asoc, ESTABLISHED) && asoc->autoclose) { + if (sctp_state(asoc, ESTABLISHED) && + asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) { timer = &asoc->timers[SCTP_EVENT_TIMEOUT_AUTOCLOSE]; timeout = asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index dfe3f36ff2a..a26065be728 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -820,7 +820,7 @@ sctp_disposition_t sctp_sf_do_5_1D_ce(struct net *net, SCTP_INC_STATS(net, SCTP_MIB_PASSIVEESTABS); sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMERS_START, SCTP_NULL()); - if (new_asoc->autoclose) + if (new_asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); @@ -908,7 +908,7 @@ sctp_disposition_t sctp_sf_do_5_1E_ca(struct net *net, SCTP_INC_STATS(net, SCTP_MIB_CURRESTAB); SCTP_INC_STATS(net, SCTP_MIB_ACTIVEESTABS); sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMERS_START, SCTP_NULL()); - if (asoc->autoclose) + if (asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); @@ -2970,7 +2970,7 @@ sctp_disposition_t sctp_sf_eat_data_6_2(struct net *net, if (chunk->chunk_hdr->flags & SCTP_DATA_SACK_IMM) force = SCTP_FORCE(); - if (asoc->autoclose) { + if (asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) { sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); } @@ -3878,7 +3878,7 @@ sctp_disposition_t sctp_sf_eat_fwd_tsn(struct net *net, SCTP_CHUNK(chunk)); /* Count this as receiving DATA. */ - if (asoc->autoclose) { + if (asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) { sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); } @@ -5267,7 +5267,7 @@ sctp_disposition_t sctp_sf_do_9_2_start_shutdown( sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); - if (asoc->autoclose) + if (asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); @@ -5346,7 +5346,7 @@ sctp_disposition_t sctp_sf_do_9_2_shutdown_ack( sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); - if (asoc->autoclose) + if (asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 72046b9729a..5455043f449 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -2196,6 +2196,7 @@ static int sctp_setsockopt_autoclose(struct sock *sk, char __user *optval, unsigned int optlen) { struct sctp_sock *sp = sctp_sk(sk); + struct net *net = sock_net(sk); /* Applicable to UDP-style socket only */ if (sctp_style(sk, TCP)) @@ -2205,6 +2206,9 @@ static int sctp_setsockopt_autoclose(struct sock *sk, char __user *optval, if (copy_from_user(&sp->autoclose, optval, optlen)) return -EFAULT; + if (sp->autoclose > net->sctp.max_autoclose) + sp->autoclose = net->sctp.max_autoclose; + return 0; } -- cgit v1.2.3-70-g09d2 From ba1f14fbe70965ae0fb1655a5275a62723f65b77 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 28 Nov 2013 14:26:41 +0100 Subject: sched: Remove PREEMPT_NEED_RESCHED from generic code While hunting a preemption issue with Alexander, Ben noticed that the currently generic PREEMPT_NEED_RESCHED stuff is horribly broken for load-store architectures. We currently rely on the IPI to fold TIF_NEED_RESCHED into PREEMPT_NEED_RESCHED, but when this IPI lands while we already have a load for the preempt-count but before the store, the store will erase the PREEMPT_NEED_RESCHED change. The current preempt-count only works on load-store archs because interrupts are assumed to be completely balanced wrt their preempt_count fiddling; the previous preempt_count load will match the preempt_count state after the interrupt and therefore nothing gets lost. This patch removes the PREEMPT_NEED_RESCHED usage from generic code and pushes it into x86 arch code; the generic code goes back to relying on TIF_NEED_RESCHED. Boot tested on x86_64 and compile tested on ppc64. Reported-by: Benjamin Herrenschmidt Reported-and-Tested-by: Alexander Graf Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Link: http://lkml.kernel.org/r/20131128132641.GP10022@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- arch/x86/include/asm/preempt.h | 11 +++++++++++ include/asm-generic/preempt.h | 35 +++++++++++------------------------ include/linux/sched.h | 2 -- 3 files changed, 22 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/arch/x86/include/asm/preempt.h b/arch/x86/include/asm/preempt.h index 8729723636f..c8b051933b1 100644 --- a/arch/x86/include/asm/preempt.h +++ b/arch/x86/include/asm/preempt.h @@ -7,6 +7,12 @@ DECLARE_PER_CPU(int, __preempt_count); +/* + * We use the PREEMPT_NEED_RESCHED bit as an inverted NEED_RESCHED such + * that a decrement hitting 0 means we can and should reschedule. + */ +#define PREEMPT_ENABLED (0 + PREEMPT_NEED_RESCHED) + /* * We mask the PREEMPT_NEED_RESCHED bit so as not to confuse all current users * that think a non-zero value indicates we cannot preempt. @@ -74,6 +80,11 @@ static __always_inline void __preempt_count_sub(int val) __this_cpu_add_4(__preempt_count, -val); } +/* + * Because we keep PREEMPT_NEED_RESCHED set when we do _not_ need to reschedule + * a decrement which hits zero means we have no preempt_count and should + * reschedule. + */ static __always_inline bool __preempt_count_dec_and_test(void) { GEN_UNARY_RMWcc("decl", __preempt_count, __percpu_arg(0), "e"); diff --git a/include/asm-generic/preempt.h b/include/asm-generic/preempt.h index ddf2b420ac8..1cd3f5d767a 100644 --- a/include/asm-generic/preempt.h +++ b/include/asm-generic/preempt.h @@ -3,13 +3,11 @@ #include -/* - * We mask the PREEMPT_NEED_RESCHED bit so as not to confuse all current users - * that think a non-zero value indicates we cannot preempt. - */ +#define PREEMPT_ENABLED (0) + static __always_inline int preempt_count(void) { - return current_thread_info()->preempt_count & ~PREEMPT_NEED_RESCHED; + return current_thread_info()->preempt_count; } static __always_inline int *preempt_count_ptr(void) @@ -17,11 +15,6 @@ static __always_inline int *preempt_count_ptr(void) return ¤t_thread_info()->preempt_count; } -/* - * We now loose PREEMPT_NEED_RESCHED and cause an extra reschedule; however the - * alternative is loosing a reschedule. Better schedule too often -- also this - * should be a very rare operation. - */ static __always_inline void preempt_count_set(int pc) { *preempt_count_ptr() = pc; @@ -41,28 +34,17 @@ static __always_inline void preempt_count_set(int pc) task_thread_info(p)->preempt_count = PREEMPT_ENABLED; \ } while (0) -/* - * We fold the NEED_RESCHED bit into the preempt count such that - * preempt_enable() can decrement and test for needing to reschedule with a - * single instruction. - * - * We invert the actual bit, so that when the decrement hits 0 we know we both - * need to resched (the bit is cleared) and can resched (no preempt count). - */ - static __always_inline void set_preempt_need_resched(void) { - *preempt_count_ptr() &= ~PREEMPT_NEED_RESCHED; } static __always_inline void clear_preempt_need_resched(void) { - *preempt_count_ptr() |= PREEMPT_NEED_RESCHED; } static __always_inline bool test_preempt_need_resched(void) { - return !(*preempt_count_ptr() & PREEMPT_NEED_RESCHED); + return false; } /* @@ -81,7 +63,12 @@ static __always_inline void __preempt_count_sub(int val) static __always_inline bool __preempt_count_dec_and_test(void) { - return !--*preempt_count_ptr(); + /* + * Because of load-store architectures cannot do per-cpu atomic + * operations; we cannot use PREEMPT_NEED_RESCHED because it might get + * lost. + */ + return !--*preempt_count_ptr() && tif_need_resched(); } /* @@ -89,7 +76,7 @@ static __always_inline bool __preempt_count_dec_and_test(void) */ static __always_inline bool should_resched(void) { - return unlikely(!*preempt_count_ptr()); + return unlikely(!preempt_count() && tif_need_resched()); } #ifdef CONFIG_PREEMPT diff --git a/include/linux/sched.h b/include/linux/sched.h index 768b037dfac..96d674ba387 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -440,8 +440,6 @@ struct task_cputime { .sum_exec_runtime = 0, \ } -#define PREEMPT_ENABLED (PREEMPT_NEED_RESCHED) - #ifdef CONFIG_PREEMPT_COUNT #define PREEMPT_DISABLED (1 + PREEMPT_ENABLED) #else -- cgit v1.2.3-70-g09d2 From be5e610c0fd6ef772cafb9e0bd4128134804aef3 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 18 Nov 2013 18:27:06 +0100 Subject: math64: Add mul_u64_u32_shr() Introduce mul_u64_u32_shr() as proposed by Andy a while back; it allows using 64x64->128 muls on 64bit archs and recent GCC which defines __SIZEOF_INT128__ and __int128. (This new method will be used by the scheduler.) Signed-off-by: Peter Zijlstra Cc: fweisbec@gmail.com Cc: Andy Lutomirski Cc: Linus Torvalds Link: http://lkml.kernel.org/n/tip-hxjoeuzmrcaumR0uZwjpe2pv@git.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 1 + include/linux/math64.h | 30 ++++++++++++++++++++++++++++++ init/Kconfig | 6 ++++++ 3 files changed, 37 insertions(+) (limited to 'include') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index e903c71f7e6..0952ecd60ec 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -26,6 +26,7 @@ config X86 select HAVE_AOUT if X86_32 select HAVE_UNSTABLE_SCHED_CLOCK select ARCH_SUPPORTS_NUMA_BALANCING + select ARCH_SUPPORTS_INT128 if X86_64 select ARCH_WANTS_PROT_NUMA_PROT_NONE select HAVE_IDE select HAVE_OPROFILE diff --git a/include/linux/math64.h b/include/linux/math64.h index 69ed5f5e9f6..c45c089bfda 100644 --- a/include/linux/math64.h +++ b/include/linux/math64.h @@ -133,4 +133,34 @@ __iter_div_u64_rem(u64 dividend, u32 divisor, u64 *remainder) return ret; } +#if defined(CONFIG_ARCH_SUPPORTS_INT128) && defined(__SIZEOF_INT128__) + +#ifndef mul_u64_u32_shr +static inline u64 mul_u64_u32_shr(u64 a, u32 mul, unsigned int shift) +{ + return (u64)(((unsigned __int128)a * mul) >> shift); +} +#endif /* mul_u64_u32_shr */ + +#else + +#ifndef mul_u64_u32_shr +static inline u64 mul_u64_u32_shr(u64 a, u32 mul, unsigned int shift) +{ + u32 ah, al; + u64 ret; + + al = a; + ah = a >> 32; + + ret = ((u64)al * mul) >> shift; + if (ah) + ret += ((u64)ah * mul) << (32 - shift); + + return ret; +} +#endif /* mul_u64_u32_shr */ + +#endif + #endif /* _LINUX_MATH64_H */ diff --git a/init/Kconfig b/init/Kconfig index 79383d3aa5d..4e5d96ab203 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -809,6 +809,12 @@ config GENERIC_SCHED_CLOCK config ARCH_SUPPORTS_NUMA_BALANCING bool +# +# For architectures that know their GCC __int128 support is sound +# +config ARCH_SUPPORTS_INT128 + bool + # For architectures that (ab)use NUMA to represent different memory regions # all cpu-local but of different latencies, such as SuperH. # -- cgit v1.2.3-70-g09d2 From 9dbdb155532395ba000c5d5d187658b0e17e529f Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 18 Nov 2013 18:27:06 +0100 Subject: sched/fair: Rework sched_fair time accounting Christian suffers from a bad BIOS that wrecks his i5's TSC sync. This results in him occasionally seeing time going backwards - which crashes the scheduler ... Most of our time accounting can actually handle that except the most common one; the tick time update of sched_fair. There is a further problem with that code; previously we assumed that because we get a tick every TICK_NSEC our time delta could never exceed 32bits and math was simpler. However, ever since Frederic managed to get NO_HZ_FULL merged; this is no longer the case since now a task can run for a long time indeed without getting a tick. It only takes about ~4.2 seconds to overflow our u32 in nanoseconds. This means we not only need to better deal with time going backwards; but also means we need to be able to deal with large deltas. This patch reworks the entire code and uses mul_u64_u32_shr() as proposed by Andy a long while ago. We express our virtual time scale factor in a u32 multiplier and shift right and the 32bit mul_u64_u32_shr() implementation reduces to a single 32x32->64 multiply if the time delta is still short (common case). For 64bit a 64x64->128 multiply can be used if ARCH_SUPPORTS_INT128. Reported-and-Tested-by: Christian Engelmayer Signed-off-by: Peter Zijlstra Cc: fweisbec@gmail.com Cc: Paul Turner Cc: Stanislaw Gruszka Cc: Andy Lutomirski Cc: Linus Torvalds Cc: Andrew Morton Link: http://lkml.kernel.org/r/20131118172706.GI3866@twins.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- include/linux/sched.h | 3 +- kernel/sched/fair.c | 144 ++++++++++++++++++++++---------------------------- 2 files changed, 66 insertions(+), 81 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 96d674ba387..53f97eb8dbc 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -930,7 +930,8 @@ struct pipe_inode_info; struct uts_namespace; struct load_weight { - unsigned long weight, inv_weight; + unsigned long weight; + u32 inv_weight; }; struct sched_avg { diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index fd773ade1a3..9030da7bcb1 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -178,59 +178,61 @@ void sched_init_granularity(void) update_sysctl(); } -#if BITS_PER_LONG == 32 -# define WMULT_CONST (~0UL) -#else -# define WMULT_CONST (1UL << 32) -#endif - +#define WMULT_CONST (~0U) #define WMULT_SHIFT 32 -/* - * Shift right and round: - */ -#define SRR(x, y) (((x) + (1UL << ((y) - 1))) >> (y)) +static void __update_inv_weight(struct load_weight *lw) +{ + unsigned long w; + + if (likely(lw->inv_weight)) + return; + + w = scale_load_down(lw->weight); + + if (BITS_PER_LONG > 32 && unlikely(w >= WMULT_CONST)) + lw->inv_weight = 1; + else if (unlikely(!w)) + lw->inv_weight = WMULT_CONST; + else + lw->inv_weight = WMULT_CONST / w; +} /* - * delta *= weight / lw + * delta_exec * weight / lw.weight + * OR + * (delta_exec * (weight * lw->inv_weight)) >> WMULT_SHIFT + * + * Either weight := NICE_0_LOAD and lw \e prio_to_wmult[], in which case + * we're guaranteed shift stays positive because inv_weight is guaranteed to + * fit 32 bits, and NICE_0_LOAD gives another 10 bits; therefore shift >= 22. + * + * Or, weight =< lw.weight (because lw.weight is the runqueue weight), thus + * weight/lw.weight <= 1, and therefore our shift will also be positive. */ -static unsigned long -calc_delta_mine(unsigned long delta_exec, unsigned long weight, - struct load_weight *lw) +static u64 __calc_delta(u64 delta_exec, unsigned long weight, struct load_weight *lw) { - u64 tmp; + u64 fact = scale_load_down(weight); + int shift = WMULT_SHIFT; - /* - * weight can be less than 2^SCHED_LOAD_RESOLUTION for task group sched - * entities since MIN_SHARES = 2. Treat weight as 1 if less than - * 2^SCHED_LOAD_RESOLUTION. - */ - if (likely(weight > (1UL << SCHED_LOAD_RESOLUTION))) - tmp = (u64)delta_exec * scale_load_down(weight); - else - tmp = (u64)delta_exec; + __update_inv_weight(lw); - if (!lw->inv_weight) { - unsigned long w = scale_load_down(lw->weight); - - if (BITS_PER_LONG > 32 && unlikely(w >= WMULT_CONST)) - lw->inv_weight = 1; - else if (unlikely(!w)) - lw->inv_weight = WMULT_CONST; - else - lw->inv_weight = WMULT_CONST / w; + if (unlikely(fact >> 32)) { + while (fact >> 32) { + fact >>= 1; + shift--; + } } - /* - * Check whether we'd overflow the 64-bit multiplication: - */ - if (unlikely(tmp > WMULT_CONST)) - tmp = SRR(SRR(tmp, WMULT_SHIFT/2) * lw->inv_weight, - WMULT_SHIFT/2); - else - tmp = SRR(tmp * lw->inv_weight, WMULT_SHIFT); + /* hint to use a 32x32->64 mul */ + fact = (u64)(u32)fact * lw->inv_weight; + + while (fact >> 32) { + fact >>= 1; + shift--; + } - return (unsigned long)min(tmp, (u64)(unsigned long)LONG_MAX); + return mul_u64_u32_shr(delta_exec, fact, shift); } @@ -443,7 +445,7 @@ find_matching_se(struct sched_entity **se, struct sched_entity **pse) #endif /* CONFIG_FAIR_GROUP_SCHED */ static __always_inline -void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, unsigned long delta_exec); +void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec); /************************************************************** * Scheduling class tree data structure manipulation methods: @@ -612,11 +614,10 @@ int sched_proc_update_handler(struct ctl_table *table, int write, /* * delta /= w */ -static inline unsigned long -calc_delta_fair(unsigned long delta, struct sched_entity *se) +static inline u64 calc_delta_fair(u64 delta, struct sched_entity *se) { if (unlikely(se->load.weight != NICE_0_LOAD)) - delta = calc_delta_mine(delta, NICE_0_LOAD, &se->load); + delta = __calc_delta(delta, NICE_0_LOAD, &se->load); return delta; } @@ -665,7 +666,7 @@ static u64 sched_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) update_load_add(&lw, se->load.weight); load = &lw; } - slice = calc_delta_mine(slice, se->load.weight, load); + slice = __calc_delta(slice, se->load.weight, load); } return slice; } @@ -703,47 +704,32 @@ void init_task_runnable_average(struct task_struct *p) #endif /* - * Update the current task's runtime statistics. Skip current tasks that - * are not in our scheduling class. + * Update the current task's runtime statistics. */ -static inline void -__update_curr(struct cfs_rq *cfs_rq, struct sched_entity *curr, - unsigned long delta_exec) -{ - unsigned long delta_exec_weighted; - - schedstat_set(curr->statistics.exec_max, - max((u64)delta_exec, curr->statistics.exec_max)); - - curr->sum_exec_runtime += delta_exec; - schedstat_add(cfs_rq, exec_clock, delta_exec); - delta_exec_weighted = calc_delta_fair(delta_exec, curr); - - curr->vruntime += delta_exec_weighted; - update_min_vruntime(cfs_rq); -} - static void update_curr(struct cfs_rq *cfs_rq) { struct sched_entity *curr = cfs_rq->curr; u64 now = rq_clock_task(rq_of(cfs_rq)); - unsigned long delta_exec; + u64 delta_exec; if (unlikely(!curr)) return; - /* - * Get the amount of time the current task was running - * since the last time we changed load (this cannot - * overflow on 32 bits): - */ - delta_exec = (unsigned long)(now - curr->exec_start); - if (!delta_exec) + delta_exec = now - curr->exec_start; + if (unlikely((s64)delta_exec <= 0)) return; - __update_curr(cfs_rq, curr, delta_exec); curr->exec_start = now; + schedstat_set(curr->statistics.exec_max, + max(delta_exec, curr->statistics.exec_max)); + + curr->sum_exec_runtime += delta_exec; + schedstat_add(cfs_rq, exec_clock, delta_exec); + + curr->vruntime += calc_delta_fair(delta_exec, curr); + update_min_vruntime(cfs_rq); + if (entity_is_task(curr)) { struct task_struct *curtask = task_of(curr); @@ -3015,8 +3001,7 @@ static void expire_cfs_rq_runtime(struct cfs_rq *cfs_rq) } } -static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, - unsigned long delta_exec) +static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec) { /* dock delta_exec before expiring quota (as it could span periods) */ cfs_rq->runtime_remaining -= delta_exec; @@ -3034,7 +3019,7 @@ static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, } static __always_inline -void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, unsigned long delta_exec) +void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec) { if (!cfs_bandwidth_used() || !cfs_rq->runtime_enabled) return; @@ -3574,8 +3559,7 @@ static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq) return rq_clock_task(rq_of(cfs_rq)); } -static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, - unsigned long delta_exec) {} +static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec) {} static void check_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} static void check_enqueue_throttle(struct cfs_rq *cfs_rq) {} static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} -- cgit v1.2.3-70-g09d2 From 4bd7b5127bd02c12c1cc837a7a0b6ce295eb2505 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Tue, 10 Dec 2013 02:20:41 +0300 Subject: micrel: add support for KSZ8041RNLI Renesas R-Car development boards use KSZ8041RNLI PHY which for some reason has ID of 0x00221537 that is not documented for KSZ8041-family PHYs and does not match the documented ID of 0x0022151x (where 'x' is the revision). We have to add the new #define PHY_ID_* and new ksphy_driver[] entry, almost the same as KSZ8041 one, differing only in the 'phy_id' and 'name' fields. Signed-off-by: Sergei Shtylyov Tested-by: Geert Uytterhoeven Signed-off-by: David S. Miller --- drivers/net/phy/micrel.c | 15 +++++++++++++++ include/linux/micrel_phy.h | 2 ++ 2 files changed, 17 insertions(+) (limited to 'include') diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 3ae28f42086..26fa05a472b 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -335,6 +335,21 @@ static struct phy_driver ksphy_driver[] = { .suspend = genphy_suspend, .resume = genphy_resume, .driver = { .owner = THIS_MODULE,}, +}, { + .phy_id = PHY_ID_KSZ8041RNLI, + .phy_id_mask = 0x00fffff0, + .name = "Micrel KSZ8041RNLI", + .features = PHY_BASIC_FEATURES | + SUPPORTED_Pause | SUPPORTED_Asym_Pause, + .flags = PHY_HAS_MAGICANEG | PHY_HAS_INTERRUPT, + .config_init = kszphy_config_init, + .config_aneg = genphy_config_aneg, + .read_status = genphy_read_status, + .ack_interrupt = kszphy_ack_interrupt, + .config_intr = kszphy_config_intr, + .suspend = genphy_suspend, + .resume = genphy_resume, + .driver = { .owner = THIS_MODULE,}, }, { .phy_id = PHY_ID_KSZ8051, .phy_id_mask = 0x00fffff0, diff --git a/include/linux/micrel_phy.h b/include/linux/micrel_phy.h index ad05ce60c1c..2e5b194b9b1 100644 --- a/include/linux/micrel_phy.h +++ b/include/linux/micrel_phy.h @@ -22,6 +22,8 @@ #define PHY_ID_KSZ8021 0x00221555 #define PHY_ID_KSZ8031 0x00221556 #define PHY_ID_KSZ8041 0x00221510 +/* undocumented */ +#define PHY_ID_KSZ8041RNLI 0x00221537 #define PHY_ID_KSZ8051 0x00221550 /* same id: ks8001 Rev. A/B, and ks8721 Rev 3. */ #define PHY_ID_KSZ8001 0x0022161A -- cgit v1.2.3-70-g09d2 From a51d5229d10dd3a337b674ce8603437d2996c5c3 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Sat, 23 Nov 2013 10:35:58 -0800 Subject: target: Remove write-only stats fields and lock from struct se_node_acl Commit 04f3b31bff72 ("iscsi-target: Convert iscsi_session statistics to atomic_long_t") removed the updating of these fields in iscsi (the only fabric driver that ever touched these counters), and the core has no way to report or otherwise use the values. Remove the last remnants of these counters. Signed-off-by: Roland Dreier Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_tpg.c | 2 -- include/target/target_core_base.h | 4 ---- 2 files changed, 6 deletions(-) (limited to 'include') diff --git a/drivers/target/target_core_tpg.c b/drivers/target/target_core_tpg.c index f697f8baec5..f755712a9a0 100644 --- a/drivers/target/target_core_tpg.c +++ b/drivers/target/target_core_tpg.c @@ -278,7 +278,6 @@ struct se_node_acl *core_tpg_check_initiator_node_acl( snprintf(acl->initiatorname, TRANSPORT_IQN_LEN, "%s", initiatorname); acl->se_tpg = tpg; acl->acl_index = scsi_get_new_index(SCSI_AUTH_INTR_INDEX); - spin_lock_init(&acl->stats_lock); acl->dynamic_node_acl = 1; tpg->se_tpg_tfo->set_default_node_attributes(acl); @@ -406,7 +405,6 @@ struct se_node_acl *core_tpg_add_initiator_node_acl( snprintf(acl->initiatorname, TRANSPORT_IQN_LEN, "%s", initiatorname); acl->se_tpg = tpg; acl->acl_index = scsi_get_new_index(SCSI_AUTH_INTR_INDEX); - spin_lock_init(&acl->stats_lock); tpg->se_tpg_tfo->set_default_node_attributes(acl); diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h index 45412a6afa6..9f1dda659c5 100644 --- a/include/target/target_core_base.h +++ b/include/target/target_core_base.h @@ -517,10 +517,6 @@ struct se_node_acl { u32 acl_index; #define MAX_ACL_TAG_SIZE 64 char acl_tag[MAX_ACL_TAG_SIZE]; - u64 num_cmds; - u64 read_bytes; - u64 write_bytes; - spinlock_t stats_lock; /* Used for PR SPEC_I_PT=1 and REGISTER_AND_MOVE */ atomic_t acl_pr_ref_count; struct se_dev_entry **device_list; -- cgit v1.2.3-70-g09d2 From a5c21dcefa1c3d759457a604b3cfc4af29c8713f Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 12 Dec 2013 17:40:21 +0000 Subject: dcache: allow word-at-a-time name hashing with big-endian CPUs When explicitly hashing the end of a string with the word-at-a-time interface, we have to be careful which end of the word we pick up. On big-endian CPUs, the upper-bits will contain the data we're after, so ensure we generate our masks accordingly (and avoid hashing whatever random junk may have been sitting after the string). This patch adds a new dcache helper, bytemask_from_count, which creates a mask appropriate for the CPU endianness. Cc: Al Viro Signed-off-by: Will Deacon Signed-off-by: Linus Torvalds --- fs/dcache.c | 2 +- fs/namei.c | 7 +------ include/linux/dcache.h | 2 ++ 3 files changed, 4 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/fs/dcache.c b/fs/dcache.c index 4bdb300b16e..6055d61811d 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -192,7 +192,7 @@ static inline int dentry_string_cmp(const unsigned char *cs, const unsigned char if (!tcount) return 0; } - mask = ~(~0ul << tcount*8); + mask = bytemask_from_count(tcount); return unlikely(!!((a ^ b) & mask)); } diff --git a/fs/namei.c b/fs/namei.c index c53d3a9547f..3531deebad3 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1598,11 +1598,6 @@ static inline int nested_symlink(struct path *path, struct nameidata *nd) * do a "get_unaligned()" if this helps and is sufficiently * fast. * - * - Little-endian machines (so that we can generate the mask - * of low bytes efficiently). Again, we *could* do a byte - * swapping load on big-endian architectures if that is not - * expensive enough to make the optimization worthless. - * * - non-CONFIG_DEBUG_PAGEALLOC configurations (so that we * do not trap on the (extremely unlikely) case of a page * crossing operation. @@ -1646,7 +1641,7 @@ unsigned int full_name_hash(const unsigned char *name, unsigned int len) if (!len) goto done; } - mask = ~(~0ul << len*8); + mask = bytemask_from_count(len); hash += mask & a; done: return fold_hash(hash); diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 57e87e749a4..bf72e9ac6de 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -29,8 +29,10 @@ struct vfsmount; /* The hash is always the low bits of hash_len */ #ifdef __LITTLE_ENDIAN #define HASH_LEN_DECLARE u32 hash; u32 len; + #define bytemask_from_count(cnt) (~(~0ul << (cnt)*8)) #else #define HASH_LEN_DECLARE u32 len; u32 hash; + #define bytemask_from_count(cnt) (~(~0ul >> (cnt)*8)) #endif /* -- cgit v1.2.3-70-g09d2 From 11ec50caedb56e3a87715edeff6a1852e6ae5416 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 12 Dec 2013 17:40:22 +0000 Subject: word-at-a-time: provide generic big-endian zero_bytemask implementation Whilst architectures may be able to do better than this (which they can, by simply defining their own macro), this is a generic stab at a zero_bytemask implementation for the asm-generic, big-endian word-at-a-time implementation. On arm64, a clz instruction is used to implement the fls efficiently. Signed-off-by: Will Deacon Signed-off-by: Linus Torvalds --- include/asm-generic/word-at-a-time.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include') diff --git a/include/asm-generic/word-at-a-time.h b/include/asm-generic/word-at-a-time.h index 3f21f1b72e4..d3909effd72 100644 --- a/include/asm-generic/word-at-a-time.h +++ b/include/asm-generic/word-at-a-time.h @@ -49,4 +49,12 @@ static inline bool has_zero(unsigned long val, unsigned long *data, const struct return (val + c->high_bits) & ~rhs; } +#ifndef zero_bytemask +#ifdef CONFIG_64BIT +#define zero_bytemask(mask) (~0ul << fls64(mask)) +#else +#define zero_bytemask(mask) (~0ul << fls(mask)) +#endif /* CONFIG_64BIT */ +#endif /* zero_bytemask */ + #endif /* _ASM_WORD_AT_A_TIME_H */ -- cgit v1.2.3-70-g09d2 From a0a9663dd2146d54339237764e1bfc60e8a39328 Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Thu, 12 Dec 2013 10:15:59 +0100 Subject: net: make neigh_priv_len in struct net_device 16bit instead of 8bit neigh_priv_len is defined as u8. With all debug enabled struct ipoib_neigh has 200 bytes. The largest part is sk_buff_head with 96 bytes and here the spinlock with 72 bytes. The size value still fits in this u8 leaving some room for more. On -RT struct ipoib_neigh put on weight and has 392 bytes. The main reason is sk_buff_head with 288 and the fatty here is spinlock with 192 bytes. This does no longer fit into into neigh_priv_len and gcc complains. This patch changes neigh_priv_len from being 8bit to 16bit. Since the following element (dev_id) is 16bit followed by a spinlock which is aligned, the struct remains with a total size of 3200 (allmodconfig) / 2048 (with as much debug off as possible) bytes on x86-64. On x86-32 the struct is 1856 (allmodconfig) / 1216 (with as much debug off as possible) bytes long. The numbers were gained with and without the patch to prove that this change does not increase the size of the struct. Cc: Steven Rostedt Cc: Thomas Gleixner Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 7f0ed423a36..d9a550bf3e8 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1255,7 +1255,7 @@ struct net_device { unsigned char perm_addr[MAX_ADDR_LEN]; /* permanent hw address */ unsigned char addr_assign_type; /* hw address assignment type */ unsigned char addr_len; /* hardware address length */ - unsigned char neigh_priv_len; + unsigned short neigh_priv_len; unsigned short dev_id; /* Used to differentiate devices * that share the same link * layer address -- cgit v1.2.3-70-g09d2 From f40386a4e976acb2bd3e0f9ead11e8e769acbe98 Mon Sep 17 00:00:00 2001 From: Naoya Horiguchi Date: Thu, 12 Dec 2013 17:12:19 -0800 Subject: include/linux/hugetlb.h: make isolate_huge_page() an inline With CONFIG_HUGETLBFS=n: mm/migrate.c: In function `do_move_page_to_node_array': include/linux/hugetlb.h:140:33: warning: statement with no effect [-Wunused-value] #define isolate_huge_page(p, l) false ^ mm/migrate.c:1170:4: note: in expansion of macro `isolate_huge_page' isolate_huge_page(page, &pagelist); Reported-by: Borislav Petkov Tested-by: Borislav Petkov Signed-off-by: Naoya Horiguchi Acked-by: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/hugetlb.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 9649ff0c63f..bd7e9875222 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -142,7 +142,10 @@ static inline int dequeue_hwpoisoned_huge_page(struct page *page) return 0; } -#define isolate_huge_page(p, l) false +static inline bool isolate_huge_page(struct page *page, struct list_head *list) +{ + return false; +} #define putback_active_hugepage(p) do {} while (0) #define is_hugepage_active(x) false -- cgit v1.2.3-70-g09d2 From 386e79066f04850352f144e67edfd5b636c0f95c Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Thu, 12 Dec 2013 17:12:24 -0800 Subject: include/linux/kernel.h: make might_fault() a nop for !MMU The machine cannot fault if !MUU, so make might_fault() a nop for !MMU. This fixes below build error if !CONFIG_MMU && (CONFIG_PROVE_LOCKING=y || CONFIG_DEBUG_ATOMIC_SLEEP=y): arch/arm/kernel/built-in.o: In function `arch_ptrace': arch/arm/kernel/ptrace.c:852: undefined reference to `might_fault' arch/arm/kernel/built-in.o: In function `restore_sigframe': arch/arm/kernel/signal.c:173: undefined reference to `might_fault' ... arch/arm/kernel/built-in.o:arch/arm/kernel/signal.c:177: more undefined references to `might_fault' follow make: *** [vmlinux] Error 1 Signed-off-by: Axel Lin Acked-by: Michael S. Tsirkin Cc: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index d4e98d13eff..ecb87544cc5 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -193,7 +193,8 @@ extern int _cond_resched(void); (__x < 0) ? -__x : __x; \ }) -#if defined(CONFIG_PROVE_LOCKING) || defined(CONFIG_DEBUG_ATOMIC_SLEEP) +#if defined(CONFIG_MMU) && \ + (defined(CONFIG_PROVE_LOCKING) || defined(CONFIG_DEBUG_ATOMIC_SLEEP)) void might_fault(void); #else static inline void might_fault(void) { } -- cgit v1.2.3-70-g09d2 From 3e1e4a5f3a324502c27c4e8808e06ac2ea842360 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Thu, 12 Dec 2013 17:12:31 -0800 Subject: mfd/rtc: s5m: fix register updating by adding regmap for RTC Rename old regmap field of "struct sec_pmic_dev" to "regmap_pmic" and add new regmap for RTC. On S5M8767A registers were not properly updated and read due to usage of the same regmap as the PMIC. This could be observed in various hangs, e.g. in infinite loop during waiting for UDR field change. On this chip family the RTC has different I2C address than PMIC so additional regmap is needed. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Kyungmin Park Reviewed-by: Mark Brown Acked-by: Sangbeom Kim Cc: Samuel Ortiz Cc: Lee Jones Cc: Liam Girdwood Cc: Alessandro Zummo Cc: Marek Szyprowski Cc: Geert Uytterhoeven Cc: Kyungmin Park Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/mfd/sec-core.c | 30 ++++++++++++++++++++++-------- drivers/mfd/sec-irq.c | 6 +++--- drivers/regulator/s5m8767.c | 2 +- drivers/rtc/rtc-s5m.c | 2 +- include/linux/mfd/samsung/core.h | 3 ++- 5 files changed, 29 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/drivers/mfd/sec-core.c b/drivers/mfd/sec-core.c index 34c18fb8c08..54cc2554659 100644 --- a/drivers/mfd/sec-core.c +++ b/drivers/mfd/sec-core.c @@ -81,31 +81,31 @@ static struct of_device_id sec_dt_match[] = { int sec_reg_read(struct sec_pmic_dev *sec_pmic, u8 reg, void *dest) { - return regmap_read(sec_pmic->regmap, reg, dest); + return regmap_read(sec_pmic->regmap_pmic, reg, dest); } EXPORT_SYMBOL_GPL(sec_reg_read); int sec_bulk_read(struct sec_pmic_dev *sec_pmic, u8 reg, int count, u8 *buf) { - return regmap_bulk_read(sec_pmic->regmap, reg, buf, count); + return regmap_bulk_read(sec_pmic->regmap_pmic, reg, buf, count); } EXPORT_SYMBOL_GPL(sec_bulk_read); int sec_reg_write(struct sec_pmic_dev *sec_pmic, u8 reg, u8 value) { - return regmap_write(sec_pmic->regmap, reg, value); + return regmap_write(sec_pmic->regmap_pmic, reg, value); } EXPORT_SYMBOL_GPL(sec_reg_write); int sec_bulk_write(struct sec_pmic_dev *sec_pmic, u8 reg, int count, u8 *buf) { - return regmap_raw_write(sec_pmic->regmap, reg, buf, count); + return regmap_raw_write(sec_pmic->regmap_pmic, reg, buf, count); } EXPORT_SYMBOL_GPL(sec_bulk_write); int sec_reg_update(struct sec_pmic_dev *sec_pmic, u8 reg, u8 val, u8 mask) { - return regmap_update_bits(sec_pmic->regmap, reg, mask, val); + return regmap_update_bits(sec_pmic->regmap_pmic, reg, mask, val); } EXPORT_SYMBOL_GPL(sec_reg_update); @@ -166,6 +166,11 @@ static struct regmap_config s5m8767_regmap_config = { .cache_type = REGCACHE_FLAT, }; +static const struct regmap_config sec_rtc_regmap_config = { + .reg_bits = 8, + .val_bits = 8, +}; + #ifdef CONFIG_OF /* * Only the common platform data elements for s5m8767 are parsed here from the @@ -266,9 +271,9 @@ static int sec_pmic_probe(struct i2c_client *i2c, break; } - sec_pmic->regmap = devm_regmap_init_i2c(i2c, regmap); - if (IS_ERR(sec_pmic->regmap)) { - ret = PTR_ERR(sec_pmic->regmap); + sec_pmic->regmap_pmic = devm_regmap_init_i2c(i2c, regmap); + if (IS_ERR(sec_pmic->regmap_pmic)) { + ret = PTR_ERR(sec_pmic->regmap_pmic); dev_err(&i2c->dev, "Failed to allocate register map: %d\n", ret); return ret; @@ -277,6 +282,15 @@ static int sec_pmic_probe(struct i2c_client *i2c, sec_pmic->rtc = i2c_new_dummy(i2c->adapter, RTC_I2C_ADDR); i2c_set_clientdata(sec_pmic->rtc, sec_pmic); + sec_pmic->regmap_rtc = devm_regmap_init_i2c(sec_pmic->rtc, + &sec_rtc_regmap_config); + if (IS_ERR(sec_pmic->regmap_rtc)) { + ret = PTR_ERR(sec_pmic->regmap_rtc); + dev_err(&i2c->dev, "Failed to allocate RTC register map: %d\n", + ret); + return ret; + } + if (pdata && pdata->cfg_pmic_irq) pdata->cfg_pmic_irq(); diff --git a/drivers/mfd/sec-irq.c b/drivers/mfd/sec-irq.c index 0dd84e99081..b441b1be27c 100644 --- a/drivers/mfd/sec-irq.c +++ b/drivers/mfd/sec-irq.c @@ -280,19 +280,19 @@ int sec_irq_init(struct sec_pmic_dev *sec_pmic) switch (type) { case S5M8763X: - ret = regmap_add_irq_chip(sec_pmic->regmap, sec_pmic->irq, + ret = regmap_add_irq_chip(sec_pmic->regmap_pmic, sec_pmic->irq, IRQF_TRIGGER_FALLING | IRQF_ONESHOT, sec_pmic->irq_base, &s5m8763_irq_chip, &sec_pmic->irq_data); break; case S5M8767X: - ret = regmap_add_irq_chip(sec_pmic->regmap, sec_pmic->irq, + ret = regmap_add_irq_chip(sec_pmic->regmap_pmic, sec_pmic->irq, IRQF_TRIGGER_FALLING | IRQF_ONESHOT, sec_pmic->irq_base, &s5m8767_irq_chip, &sec_pmic->irq_data); break; case S2MPS11X: - ret = regmap_add_irq_chip(sec_pmic->regmap, sec_pmic->irq, + ret = regmap_add_irq_chip(sec_pmic->regmap_pmic, sec_pmic->irq, IRQF_TRIGGER_FALLING | IRQF_ONESHOT, sec_pmic->irq_base, &s2mps11_irq_chip, &sec_pmic->irq_data); diff --git a/drivers/regulator/s5m8767.c b/drivers/regulator/s5m8767.c index cbf91e25cf7..aeb40aad0ae 100644 --- a/drivers/regulator/s5m8767.c +++ b/drivers/regulator/s5m8767.c @@ -925,7 +925,7 @@ static int s5m8767_pmic_probe(struct platform_device *pdev) config.dev = s5m8767->dev; config.init_data = pdata->regulators[i].initdata; config.driver_data = s5m8767; - config.regmap = iodev->regmap; + config.regmap = iodev->regmap_pmic; config.of_node = pdata->regulators[i].reg_node; rdev[i] = devm_regulator_register(&pdev->dev, ®ulators[id], diff --git a/drivers/rtc/rtc-s5m.c b/drivers/rtc/rtc-s5m.c index 0ba56b7cb38..ae8119dc284 100644 --- a/drivers/rtc/rtc-s5m.c +++ b/drivers/rtc/rtc-s5m.c @@ -567,7 +567,7 @@ static int s5m_rtc_probe(struct platform_device *pdev) info->dev = &pdev->dev; info->s5m87xx = s5m87xx; - info->regmap = s5m87xx->regmap; + info->regmap = s5m87xx->regmap_rtc; info->device_type = s5m87xx->device_type; info->wtsr_smpl = s5m87xx->wtsr_smpl; diff --git a/include/linux/mfd/samsung/core.h b/include/linux/mfd/samsung/core.h index 2d0c9071bcf..cab2dd27907 100644 --- a/include/linux/mfd/samsung/core.h +++ b/include/linux/mfd/samsung/core.h @@ -39,7 +39,8 @@ enum sec_device_type { struct sec_pmic_dev { struct device *dev; struct sec_platform_data *pdata; - struct regmap *regmap; + struct regmap *regmap_pmic; + struct regmap *regmap_rtc; struct i2c_client *i2c; struct i2c_client *rtc; -- cgit v1.2.3-70-g09d2 From 380108d891acf8db5cf0d477176c7ed2b62b7928 Mon Sep 17 00:00:00 2001 From: Julien Grall Date: Tue, 3 Dec 2013 15:40:37 +0000 Subject: xen/block: Correctly define structures in public headers on ARM32 and ARM64 On ARM (32 bits and 64 bits), the double-word is 8-bytes aligned. This will result on different structure from Xen and Linux repositories. As Linux is using __packed__ attribute, it must have a 4-bytes padding before each "id" field. This change breaks guest block support with older kernel. IMHO, it's acceptable because Xen on ARM is still on Tech Preview and the hypercall ABI is not yet freezed. Only one architecture (x86_32) doesn't have 64-bit ABI for the block interface. Don't add padding if Linux is compiled for this architecture. Signed-off-by: Julien Grall Acked-by: Roger Pau Monne Acked-by: David Vrabel Cc: Boris Ostrovsky Cc: Ian Campbell Acked-by: Stefano Stabellini [I had asked for confirmation that it did not break x86 and Ian went beyound the call of duty to confirm it. Also a internal regression bucket with 32/64 dom0 with 32/64 domU (PV and HVM) confirmed no regressions. ABI changes are a drag..] Signed-off-by: Konrad Rzeszutek Wilk --- include/xen/interface/io/blkif.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/xen/interface/io/blkif.h b/include/xen/interface/io/blkif.h index 65e12099ef8..ae665ac59c3 100644 --- a/include/xen/interface/io/blkif.h +++ b/include/xen/interface/io/blkif.h @@ -146,7 +146,7 @@ struct blkif_request_segment_aligned { struct blkif_request_rw { uint8_t nr_segments; /* number of segments */ blkif_vdev_t handle; /* only for read/write requests */ -#ifdef CONFIG_X86_64 +#ifndef CONFIG_X86_32 uint32_t _pad1; /* offsetof(blkif_request,u.rw.id) == 8 */ #endif uint64_t id; /* private guest value, echoed in resp */ @@ -163,7 +163,7 @@ struct blkif_request_discard { uint8_t flag; /* BLKIF_DISCARD_SECURE or zero. */ #define BLKIF_DISCARD_SECURE (1<<0) /* ignored if discard-secure=0 */ blkif_vdev_t _pad1; /* only for read/write requests */ -#ifdef CONFIG_X86_64 +#ifndef CONFIG_X86_32 uint32_t _pad2; /* offsetof(blkif_req..,u.discard.id)==8*/ #endif uint64_t id; /* private guest value, echoed in resp */ @@ -175,7 +175,7 @@ struct blkif_request_discard { struct blkif_request_other { uint8_t _pad1; blkif_vdev_t _pad2; /* only for read/write requests */ -#ifdef CONFIG_X86_64 +#ifndef CONFIG_X86_32 uint32_t _pad3; /* offsetof(blkif_req..,u.other.id)==8*/ #endif uint64_t id; /* private guest value, echoed in resp */ @@ -184,7 +184,7 @@ struct blkif_request_other { struct blkif_request_indirect { uint8_t indirect_op; uint16_t nr_segments; -#ifdef CONFIG_X86_64 +#ifndef CONFIG_X86_32 uint32_t _pad1; /* offsetof(blkif_...,u.indirect.id) == 8 */ #endif uint64_t id; @@ -192,7 +192,7 @@ struct blkif_request_indirect { blkif_vdev_t handle; uint16_t _pad2; grant_ref_t indirect_grefs[BLKIF_MAX_INDIRECT_PAGES_PER_REQUEST]; -#ifdef CONFIG_X86_64 +#ifndef CONFIG_X86_32 uint32_t _pad3; /* make it 64 byte aligned */ #else uint64_t _pad3; /* make it 64 byte aligned */ -- cgit v1.2.3-70-g09d2 From 189b84fb54490ae24111124346a8e63f8e019385 Mon Sep 17 00:00:00 2001 From: Vince Weaver Date: Fri, 13 Dec 2013 15:52:25 -0500 Subject: perf: Document the new transaction sample type Commit fdfbbd07e91f8fe3871 ("perf: Add generic transaction flags") added support for PERF_SAMPLE_TRANSACTION but forgot to add documentation for the sample type to include/uapi/linux/perf_event.h Signed-off-by: Vince Weaver Signed-off-by: Peter Zijlstra Cc: Andi Kleen Link: http://lkml.kernel.org/r/alpine.DEB.2.02.1312131548450.10372@pianoman.cluster.toy Signed-off-by: Ingo Molnar --- include/uapi/linux/perf_event.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h index e1802d6153a..959d454f76a 100644 --- a/include/uapi/linux/perf_event.h +++ b/include/uapi/linux/perf_event.h @@ -679,6 +679,7 @@ enum perf_event_type { * * { u64 weight; } && PERF_SAMPLE_WEIGHT * { u64 data_src; } && PERF_SAMPLE_DATA_SRC + * { u64 transaction; } && PERF_SAMPLE_TRANSACTION * }; */ PERF_RECORD_SAMPLE = 9, -- cgit v1.2.3-70-g09d2 From c97102ba96324da330078ad8619ba4dfe840dbe3 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Wed, 18 Dec 2013 17:08:31 -0800 Subject: kexec: migrate to reboot cpu Commit 1b3a5d02ee07 ("reboot: move arch/x86 reboot= handling to generic kernel") moved reboot= handling to generic code. In the process it also removed the code in native_machine_shutdown() which are moving reboot process to reboot_cpu/cpu0. I guess that thought must have been that all reboot paths are calling migrate_to_reboot_cpu(), so we don't need this special handling. But kexec reboot path (kernel_kexec()) is not calling migrate_to_reboot_cpu() so above change broke kexec. Now reboot can happen on non-boot cpu and when INIT is sent in second kerneo to bring up BP, it brings down the machine. So start calling migrate_to_reboot_cpu() in kexec reboot path to avoid this problem. Bisected by WANG Chao. Reported-by: Matthew Whitehead Reported-by: Dave Young Signed-off-by: Vivek Goyal Tested-by: Baoquan He Tested-by: WANG Chao Acked-by: H. Peter Anvin Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/reboot.h | 1 + kernel/kexec.c | 1 + kernel/reboot.c | 2 +- 3 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/reboot.h b/include/linux/reboot.h index 8e00f9f6f96..9e7db9e73cc 100644 --- a/include/linux/reboot.h +++ b/include/linux/reboot.h @@ -43,6 +43,7 @@ extern int unregister_reboot_notifier(struct notifier_block *); * Architecture-specific implementations of sys_reboot commands. */ +extern void migrate_to_reboot_cpu(void); extern void machine_restart(char *cmd); extern void machine_halt(void); extern void machine_power_off(void); diff --git a/kernel/kexec.c b/kernel/kexec.c index d0d8fca5406..9c970167e40 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1680,6 +1680,7 @@ int kernel_kexec(void) { kexec_in_progress = true; kernel_restart_prepare(NULL); + migrate_to_reboot_cpu(); printk(KERN_EMERG "Starting new kernel\n"); machine_shutdown(); } diff --git a/kernel/reboot.c b/kernel/reboot.c index f813b347464..662c83fc16b 100644 --- a/kernel/reboot.c +++ b/kernel/reboot.c @@ -104,7 +104,7 @@ int unregister_reboot_notifier(struct notifier_block *nb) } EXPORT_SYMBOL(unregister_reboot_notifier); -static void migrate_to_reboot_cpu(void) +void migrate_to_reboot_cpu(void) { /* The boot cpu is always logical cpu 0 */ int cpu = reboot_cpu; -- cgit v1.2.3-70-g09d2 From de466bd628e8d663fdf3f791bc8db318ee85c714 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 18 Dec 2013 17:08:42 -0800 Subject: mm: numa: avoid unnecessary disruption of NUMA hinting during migration do_huge_pmd_numa_page() handles the case where there is parallel THP migration. However, by the time it is checked the NUMA hinting information has already been disrupted. This patch adds an earlier check with some helpers. Signed-off-by: Mel Gorman Reviewed-by: Rik van Riel Cc: Alex Thorlton Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/migrate.h | 9 +++++++++ mm/huge_memory.c | 22 ++++++++++++++++------ mm/migrate.c | 12 ++++++++++++ 3 files changed, 37 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/migrate.h b/include/linux/migrate.h index f5096b58b20..b7717d74da7 100644 --- a/include/linux/migrate.h +++ b/include/linux/migrate.h @@ -90,10 +90,19 @@ static inline int migrate_huge_page_move_mapping(struct address_space *mapping, #endif /* CONFIG_MIGRATION */ #ifdef CONFIG_NUMA_BALANCING +extern bool pmd_trans_migrating(pmd_t pmd); +extern void wait_migrate_huge_page(struct anon_vma *anon_vma, pmd_t *pmd); extern int migrate_misplaced_page(struct page *page, struct vm_area_struct *vma, int node); extern bool migrate_ratelimited(int node); #else +static inline bool pmd_trans_migrating(pmd_t pmd) +{ + return false; +} +static inline void wait_migrate_huge_page(struct anon_vma *anon_vma, pmd_t *pmd) +{ +} static inline int migrate_misplaced_page(struct page *page, struct vm_area_struct *vma, int node) { diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 70e7429fd8e..7de1bf85f68 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -882,6 +882,10 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm, ret = 0; goto out_unlock; } + + /* mmap_sem prevents this happening but warn if that changes */ + WARN_ON(pmd_trans_migrating(pmd)); + if (unlikely(pmd_trans_splitting(pmd))) { /* split huge page running from under us */ spin_unlock(src_ptl); @@ -1299,6 +1303,17 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, if (unlikely(!pmd_same(pmd, *pmdp))) goto out_unlock; + /* + * If there are potential migrations, wait for completion and retry + * without disrupting NUMA hinting information. Do not relock and + * check_same as the page may no longer be mapped. + */ + if (unlikely(pmd_trans_migrating(*pmdp))) { + spin_unlock(ptl); + wait_migrate_huge_page(vma->anon_vma, pmdp); + goto out; + } + page = pmd_page(pmd); BUG_ON(is_huge_zero_page(page)); page_nid = page_to_nid(page); @@ -1329,12 +1344,7 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, goto clear_pmdnuma; } - /* - * If there are potential migrations, wait for completion and retry. We - * do not relock and check_same as the page may no longer be mapped. - * Furtermore, even if the page is currently misplaced, there is no - * guarantee it is still misplaced after the migration completes. - */ + /* Migration could have started since the pmd_trans_migrating check */ if (!page_locked) { spin_unlock(ptl); wait_on_page_locked(page); diff --git a/mm/migrate.c b/mm/migrate.c index a987525810a..cfb41908526 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1655,6 +1655,18 @@ int numamigrate_isolate_page(pg_data_t *pgdat, struct page *page) return 1; } +bool pmd_trans_migrating(pmd_t pmd) +{ + struct page *page = pmd_page(pmd); + return PageLocked(page); +} + +void wait_migrate_huge_page(struct anon_vma *anon_vma, pmd_t *pmd) +{ + struct page *page = pmd_page(*pmd); + wait_on_page_locked(page); +} + /* * Attempt to migrate a misplaced page to the specified destination * node. Caller is expected to have an elevated reference count on -- cgit v1.2.3-70-g09d2 From 20841405940e7be0617612d521e206e4b6b325db Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Wed, 18 Dec 2013 17:08:44 -0800 Subject: mm: fix TLB flush race between migration, and change_protection_range There are a few subtle races, between change_protection_range (used by mprotect and change_prot_numa) on one side, and NUMA page migration and compaction on the other side. The basic race is that there is a time window between when the PTE gets made non-present (PROT_NONE or NUMA), and the TLB is flushed. During that time, a CPU may continue writing to the page. This is fine most of the time, however compaction or the NUMA migration code may come in, and migrate the page away. When that happens, the CPU may continue writing, through the cached translation, to what is no longer the current memory location of the process. This only affects x86, which has a somewhat optimistic pte_accessible. All other architectures appear to be safe, and will either always flush, or flush whenever there is a valid mapping, even with no permissions (SPARC). The basic race looks like this: CPU A CPU B CPU C load TLB entry make entry PTE/PMD_NUMA fault on entry read/write old page start migrating page change PTE/PMD to new page read/write old page [*] flush TLB reload TLB from new entry read/write new page lose data [*] the old page may belong to a new user at this point! The obvious fix is to flush remote TLB entries, by making sure that pte_accessible aware of the fact that PROT_NONE and PROT_NUMA memory may still be accessible if there is a TLB flush pending for the mm. This should fix both NUMA migration and compaction. [mgorman@suse.de: fix build] Signed-off-by: Rik van Riel Signed-off-by: Mel Gorman Cc: Alex Thorlton Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/sparc/include/asm/pgtable_64.h | 4 ++-- arch/x86/include/asm/pgtable.h | 11 ++++++++-- include/asm-generic/pgtable.h | 2 +- include/linux/mm_types.h | 44 +++++++++++++++++++++++++++++++++++++ kernel/fork.c | 1 + mm/huge_memory.c | 7 ++++++ mm/mprotect.c | 2 ++ mm/pgtable-generic.c | 5 +++-- 8 files changed, 69 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index 8358dc14495..0f9e94537ee 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -619,7 +619,7 @@ static inline unsigned long pte_present(pte_t pte) } #define pte_accessible pte_accessible -static inline unsigned long pte_accessible(pte_t a) +static inline unsigned long pte_accessible(struct mm_struct *mm, pte_t a) { return pte_val(a) & _PAGE_VALID; } @@ -847,7 +847,7 @@ static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr, * SUN4V NOTE: _PAGE_VALID is the same value in both the SUN4U * and SUN4V pte layout, so this inline test is fine. */ - if (likely(mm != &init_mm) && pte_accessible(orig)) + if (likely(mm != &init_mm) && pte_accessible(mm, orig)) tlb_batch_add(mm, addr, ptep, orig, fullmm); } diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 3d199945870..bbc8b12fa44 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -452,9 +452,16 @@ static inline int pte_present(pte_t a) } #define pte_accessible pte_accessible -static inline int pte_accessible(pte_t a) +static inline bool pte_accessible(struct mm_struct *mm, pte_t a) { - return pte_flags(a) & _PAGE_PRESENT; + if (pte_flags(a) & _PAGE_PRESENT) + return true; + + if ((pte_flags(a) & (_PAGE_PROTNONE | _PAGE_NUMA)) && + mm_tlb_flush_pending(mm)) + return true; + + return false; } static inline int pte_hidden(pte_t pte) diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h index f330d28e4d0..b12079afbd5 100644 --- a/include/asm-generic/pgtable.h +++ b/include/asm-generic/pgtable.h @@ -217,7 +217,7 @@ static inline int pmd_same(pmd_t pmd_a, pmd_t pmd_b) #endif #ifndef pte_accessible -# define pte_accessible(pte) ((void)(pte),1) +# define pte_accessible(mm, pte) ((void)(pte), 1) #endif #ifndef flush_tlb_fix_spurious_fault diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index bd299418a93..e5c49c30460 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -442,6 +442,14 @@ struct mm_struct { /* numa_scan_seq prevents two threads setting pte_numa */ int numa_scan_seq; +#endif +#if defined(CONFIG_NUMA_BALANCING) || defined(CONFIG_COMPACTION) + /* + * An operation with batched TLB flushing is going on. Anything that + * can move process memory needs to flush the TLB when moving a + * PROT_NONE or PROT_NUMA mapped page. + */ + bool tlb_flush_pending; #endif struct uprobes_state uprobes_state; }; @@ -459,4 +467,40 @@ static inline cpumask_t *mm_cpumask(struct mm_struct *mm) return mm->cpu_vm_mask_var; } +#if defined(CONFIG_NUMA_BALANCING) || defined(CONFIG_COMPACTION) +/* + * Memory barriers to keep this state in sync are graciously provided by + * the page table locks, outside of which no page table modifications happen. + * The barriers below prevent the compiler from re-ordering the instructions + * around the memory barriers that are already present in the code. + */ +static inline bool mm_tlb_flush_pending(struct mm_struct *mm) +{ + barrier(); + return mm->tlb_flush_pending; +} +static inline void set_tlb_flush_pending(struct mm_struct *mm) +{ + mm->tlb_flush_pending = true; + barrier(); +} +/* Clearing is done after a TLB flush, which also provides a barrier. */ +static inline void clear_tlb_flush_pending(struct mm_struct *mm) +{ + barrier(); + mm->tlb_flush_pending = false; +} +#else +static inline bool mm_tlb_flush_pending(struct mm_struct *mm) +{ + return false; +} +static inline void set_tlb_flush_pending(struct mm_struct *mm) +{ +} +static inline void clear_tlb_flush_pending(struct mm_struct *mm) +{ +} +#endif + #endif /* _LINUX_MM_TYPES_H */ diff --git a/kernel/fork.c b/kernel/fork.c index 728d5be9548..5721f0e3f2d 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -537,6 +537,7 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p) spin_lock_init(&mm->page_table_lock); mm_init_aio(mm); mm_init_owner(mm, p); + clear_tlb_flush_pending(mm); if (likely(!mm_alloc_pgd(mm))) { mm->def_flags = 0; diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 7de1bf85f68..3d2783e1059 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1376,6 +1376,13 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, goto clear_pmdnuma; } + /* + * The page_table_lock above provides a memory barrier + * with change_protection_range. + */ + if (mm_tlb_flush_pending(mm)) + flush_tlb_range(vma, haddr, haddr + HPAGE_PMD_SIZE); + /* * Migrate the THP to the requested node, returns with page unlocked * and pmd_numa cleared. diff --git a/mm/mprotect.c b/mm/mprotect.c index f8421722acb..bb53a6591ae 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -188,6 +188,7 @@ static unsigned long change_protection_range(struct vm_area_struct *vma, BUG_ON(addr >= end); pgd = pgd_offset(mm, addr); flush_cache_range(vma, addr, end); + set_tlb_flush_pending(mm); do { next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) @@ -199,6 +200,7 @@ static unsigned long change_protection_range(struct vm_area_struct *vma, /* Only flush the TLB if we actually modified any entries: */ if (pages) flush_tlb_range(vma, start, end); + clear_tlb_flush_pending(mm); return pages; } diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c index e84cad27a80..a8b91992593 100644 --- a/mm/pgtable-generic.c +++ b/mm/pgtable-generic.c @@ -110,9 +110,10 @@ int pmdp_clear_flush_young(struct vm_area_struct *vma, pte_t ptep_clear_flush(struct vm_area_struct *vma, unsigned long address, pte_t *ptep) { + struct mm_struct *mm = (vma)->vm_mm; pte_t pte; - pte = ptep_get_and_clear((vma)->vm_mm, address, ptep); - if (pte_accessible(pte)) + pte = ptep_get_and_clear(mm, address, ptep); + if (pte_accessible(mm, pte)) flush_tlb_page(vma, address); return pte; } -- cgit v1.2.3-70-g09d2 From af2c1401e6f9177483be4fad876d0073669df9df Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 18 Dec 2013 17:08:45 -0800 Subject: mm: numa: guarantee that tlb_flush_pending updates are visible before page table updates According to documentation on barriers, stores issued before a LOCK can complete after the lock implying that it's possible tlb_flush_pending can be visible after a page table update. As per revised documentation, this patch adds a smp_mb__before_spinlock to guarantee the correct ordering. Signed-off-by: Mel Gorman Acked-by: Paul E. McKenney Reviewed-by: Rik van Riel Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm_types.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index e5c49c30460..ad0616f2fe2 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -482,7 +482,12 @@ static inline bool mm_tlb_flush_pending(struct mm_struct *mm) static inline void set_tlb_flush_pending(struct mm_struct *mm) { mm->tlb_flush_pending = true; - barrier(); + + /* + * Guarantee that the tlb_flush_pending store does not leak into the + * critical section updating the page tables + */ + smp_mb__before_spinlock(); } /* Clearing is done after a TLB flush, which also provides a barrier. */ static inline void clear_tlb_flush_pending(struct mm_struct *mm) -- cgit v1.2.3-70-g09d2 From 95cadace8f3959282e76ebf8b382bd0930807d2c Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 12 Dec 2013 12:24:11 -0800 Subject: target/file: Update hw_max_sectors based on current block_size This patch allows FILEIO to update hw_max_sectors based on the current max_bytes_per_io. This is required because vfs_[writev,readv]() can accept a maximum of 2048 iovecs per call, so the enforced hw_max_sectors really needs to be calculated based on block_size. This addresses a >= v3.5 bug where block_size=512 was rejecting > 1M sized I/O requests, because FD_MAX_SECTORS was hardcoded to 2048 for the block_size=4096 case. (v2: Use max_bytes_per_io instead of ->update_hw_max_sectors) Reported-by: Henrik Goldman Cc: #3.5+ Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_device.c | 5 +++++ drivers/target/target_core_file.c | 8 ++++---- drivers/target/target_core_file.h | 5 ++++- include/target/target_core_base.h | 1 + 4 files changed, 14 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c index 207b340498a..d06de84b069 100644 --- a/drivers/target/target_core_device.c +++ b/drivers/target/target_core_device.c @@ -1106,6 +1106,11 @@ int se_dev_set_block_size(struct se_device *dev, u32 block_size) dev->dev_attrib.block_size = block_size; pr_debug("dev[%p]: SE Device block_size changed to %u\n", dev, block_size); + + if (dev->dev_attrib.max_bytes_per_io) + dev->dev_attrib.hw_max_sectors = + dev->dev_attrib.max_bytes_per_io / block_size; + return 0; } diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c index 0e34cda3271..78241a53b55 100644 --- a/drivers/target/target_core_file.c +++ b/drivers/target/target_core_file.c @@ -66,9 +66,8 @@ static int fd_attach_hba(struct se_hba *hba, u32 host_id) pr_debug("CORE_HBA[%d] - TCM FILEIO HBA Driver %s on Generic" " Target Core Stack %s\n", hba->hba_id, FD_VERSION, TARGET_CORE_MOD_VERSION); - pr_debug("CORE_HBA[%d] - Attached FILEIO HBA: %u to Generic" - " MaxSectors: %u\n", - hba->hba_id, fd_host->fd_host_id, FD_MAX_SECTORS); + pr_debug("CORE_HBA[%d] - Attached FILEIO HBA: %u to Generic\n", + hba->hba_id, fd_host->fd_host_id); return 0; } @@ -220,7 +219,8 @@ static int fd_configure_device(struct se_device *dev) } dev->dev_attrib.hw_block_size = fd_dev->fd_block_size; - dev->dev_attrib.hw_max_sectors = FD_MAX_SECTORS; + dev->dev_attrib.max_bytes_per_io = FD_MAX_BYTES; + dev->dev_attrib.hw_max_sectors = FD_MAX_BYTES / fd_dev->fd_block_size; dev->dev_attrib.hw_queue_depth = FD_MAX_DEVICE_QUEUE_DEPTH; if (fd_dev->fbd_flags & FDBD_HAS_BUFFERED_IO_WCE) { diff --git a/drivers/target/target_core_file.h b/drivers/target/target_core_file.h index 37ffc5bd239..d7772c16768 100644 --- a/drivers/target/target_core_file.h +++ b/drivers/target/target_core_file.h @@ -7,7 +7,10 @@ #define FD_DEVICE_QUEUE_DEPTH 32 #define FD_MAX_DEVICE_QUEUE_DEPTH 128 #define FD_BLOCKSIZE 512 -#define FD_MAX_SECTORS 2048 +/* + * Limited by the number of iovecs (2048) per vfs_[writev,readv] call + */ +#define FD_MAX_BYTES 8388608 #define RRF_EMULATE_CDB 0x01 #define RRF_GOT_LBA 0x02 diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h index 9f1dda659c5..321301c0a64 100644 --- a/include/target/target_core_base.h +++ b/include/target/target_core_base.h @@ -620,6 +620,7 @@ struct se_dev_attrib { u32 unmap_granularity; u32 unmap_granularity_alignment; u32 max_write_same_len; + u32 max_bytes_per_io; struct se_device *da_dev; struct config_group da_group; }; -- cgit v1.2.3-70-g09d2 From ee53664bda169f519ce3c6a22d378f0b946c8178 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Fri, 20 Dec 2013 15:10:03 +0200 Subject: mm: Fix NULL pointer dereference in madvise(MADV_WILLNEED) support Sasha Levin found a NULL pointer dereference that is due to a missing page table lock, which in turn is due to the pmd entry in question being a transparent huge-table entry. The code - introduced in commit 1998cc048901 ("mm: make madvise(MADV_WILLNEED) support swap file prefetch") - correctly checks for this situation using pmd_none_or_trans_huge_or_clear_bad(), but it turns out that that function doesn't work correctly. pmd_none_or_trans_huge_or_clear_bad() expected that pmd_bad() would trigger if the transparent hugepage bit was set, but it doesn't do that if pmd_numa() is also set. Note that the NUMA bit only gets set on real NUMA machines, so people trying to reproduce this on most normal development systems would never actually trigger this. Fix it by removing the very subtle (and subtly incorrect) expectation, and instead just checking pmd_trans_huge() explicitly. Reported-by: Sasha Levin Acked-by: Andrea Arcangeli [ Additionally remove the now stale test for pmd_trans_huge() inside the pmd_bad() case - Linus ] Signed-off-by: Linus Torvalds --- include/asm-generic/pgtable.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h index b12079afbd5..db092345894 100644 --- a/include/asm-generic/pgtable.h +++ b/include/asm-generic/pgtable.h @@ -599,11 +599,10 @@ static inline int pmd_none_or_trans_huge_or_clear_bad(pmd_t *pmd) #ifdef CONFIG_TRANSPARENT_HUGEPAGE barrier(); #endif - if (pmd_none(pmdval)) + if (pmd_none(pmdval) || pmd_trans_huge(pmdval)) return 1; if (unlikely(pmd_bad(pmdval))) { - if (!pmd_trans_huge(pmdval)) - pmd_clear_bad(pmd); + pmd_clear_bad(pmd); return 1; } return 0; -- cgit v1.2.3-70-g09d2 From 597d795a2a786d22dd872332428e2b9439ede639 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Fri, 20 Dec 2013 13:35:58 +0200 Subject: mm: do not allocate page->ptl dynamically, if spinlock_t fits to long In struct page we have enough space to fit long-size page->ptl there, but we use dynamically-allocated page->ptl if size(spinlock_t) is larger than sizeof(int). It hurts 64-bit architectures with CONFIG_GENERIC_LOCKBREAK, where sizeof(spinlock_t) == 8, but it easily fits into struct page. Signed-off-by: Kirill A. Shutemov Acked-by: Hugh Dickins Signed-off-by: Linus Torvalds --- include/linux/lockref.h | 2 +- include/linux/mm.h | 6 +++--- include/linux/mm_types.h | 3 ++- kernel/bounds.c | 2 +- mm/memory.c | 2 +- 5 files changed, 8 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/linux/lockref.h b/include/linux/lockref.h index c8929c3832d..4bfde0e99ed 100644 --- a/include/linux/lockref.h +++ b/include/linux/lockref.h @@ -19,7 +19,7 @@ #define USE_CMPXCHG_LOCKREF \ (IS_ENABLED(CONFIG_ARCH_USE_CMPXCHG_LOCKREF) && \ - IS_ENABLED(CONFIG_SMP) && !BLOATED_SPINLOCKS) + IS_ENABLED(CONFIG_SMP) && SPINLOCK_SIZE <= 4) struct lockref { union { diff --git a/include/linux/mm.h b/include/linux/mm.h index 1cedd000cf2..35527173cf5 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1317,7 +1317,7 @@ static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long a #endif /* CONFIG_MMU && !__ARCH_HAS_4LEVEL_HACK */ #if USE_SPLIT_PTE_PTLOCKS -#if BLOATED_SPINLOCKS +#if ALLOC_SPLIT_PTLOCKS extern bool ptlock_alloc(struct page *page); extern void ptlock_free(struct page *page); @@ -1325,7 +1325,7 @@ static inline spinlock_t *ptlock_ptr(struct page *page) { return page->ptl; } -#else /* BLOATED_SPINLOCKS */ +#else /* ALLOC_SPLIT_PTLOCKS */ static inline bool ptlock_alloc(struct page *page) { return true; @@ -1339,7 +1339,7 @@ static inline spinlock_t *ptlock_ptr(struct page *page) { return &page->ptl; } -#endif /* BLOATED_SPINLOCKS */ +#endif /* ALLOC_SPLIT_PTLOCKS */ static inline spinlock_t *pte_lockptr(struct mm_struct *mm, pmd_t *pmd) { diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index ad0616f2fe2..290901a8c1d 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -26,6 +26,7 @@ struct address_space; #define USE_SPLIT_PTE_PTLOCKS (NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS) #define USE_SPLIT_PMD_PTLOCKS (USE_SPLIT_PTE_PTLOCKS && \ IS_ENABLED(CONFIG_ARCH_ENABLE_SPLIT_PMD_PTLOCK)) +#define ALLOC_SPLIT_PTLOCKS (SPINLOCK_SIZE > BITS_PER_LONG/8) /* * Each physical page in the system has a struct page associated with @@ -155,7 +156,7 @@ struct page { * system if PG_buddy is set. */ #if USE_SPLIT_PTE_PTLOCKS -#if BLOATED_SPINLOCKS +#if ALLOC_SPLIT_PTLOCKS spinlock_t *ptl; #else spinlock_t ptl; diff --git a/kernel/bounds.c b/kernel/bounds.c index 5253204afdc..9fd4246b04b 100644 --- a/kernel/bounds.c +++ b/kernel/bounds.c @@ -22,6 +22,6 @@ void foo(void) #ifdef CONFIG_SMP DEFINE(NR_CPUS_BITS, ilog2(CONFIG_NR_CPUS)); #endif - DEFINE(BLOATED_SPINLOCKS, sizeof(spinlock_t) > sizeof(int)); + DEFINE(SPINLOCK_SIZE, sizeof(spinlock_t)); /* End of constants */ } diff --git a/mm/memory.c b/mm/memory.c index 5d9025f3b3e..b6e211b779d 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4271,7 +4271,7 @@ void copy_user_huge_page(struct page *dst, struct page *src, } #endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */ -#if USE_SPLIT_PTE_PTLOCKS && BLOATED_SPINLOCKS +#if ALLOC_SPLIT_PTLOCKS bool ptlock_alloc(struct page *page) { spinlock_t *ptl; -- cgit v1.2.3-70-g09d2 From df36ac1bc2a166eef90785d584e4cfed6f52bd32 Mon Sep 17 00:00:00 2001 From: "Luck, Tony" Date: Wed, 18 Dec 2013 15:17:10 -0800 Subject: pstore: Don't allow high traffic options on fragile devices Some pstore backing devices use on board flash as persistent storage. These have limited numbers of write cycles so it is a poor idea to use them from high frequency operations. Signed-off-by: Tony Luck Signed-off-by: Linus Torvalds --- drivers/acpi/apei/erst.c | 1 + drivers/firmware/efi/efi-pstore.c | 1 + fs/pstore/platform.c | 7 +++++-- include/linux/pstore.h | 3 +++ 4 files changed, 10 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/acpi/apei/erst.c b/drivers/acpi/apei/erst.c index 26311f23c82..cb1d557fc22 100644 --- a/drivers/acpi/apei/erst.c +++ b/drivers/acpi/apei/erst.c @@ -942,6 +942,7 @@ static int erst_clearer(enum pstore_type_id type, u64 id, int count, static struct pstore_info erst_info = { .owner = THIS_MODULE, .name = "erst", + .flags = PSTORE_FLAGS_FRAGILE, .open = erst_open_pstore, .close = erst_close_pstore, .read = erst_reader, diff --git a/drivers/firmware/efi/efi-pstore.c b/drivers/firmware/efi/efi-pstore.c index 743fd426f21..4b9dc836dcf 100644 --- a/drivers/firmware/efi/efi-pstore.c +++ b/drivers/firmware/efi/efi-pstore.c @@ -356,6 +356,7 @@ static int efi_pstore_erase(enum pstore_type_id type, u64 id, int count, static struct pstore_info efi_pstore_info = { .owner = THIS_MODULE, .name = "efi", + .flags = PSTORE_FLAGS_FRAGILE, .open = efi_pstore_open, .close = efi_pstore_close, .read = efi_pstore_read, diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c index b8e93a40a5d..78c3c209778 100644 --- a/fs/pstore/platform.c +++ b/fs/pstore/platform.c @@ -443,8 +443,11 @@ int pstore_register(struct pstore_info *psi) pstore_get_records(0); kmsg_dump_register(&pstore_dumper); - pstore_register_console(); - pstore_register_ftrace(); + + if ((psi->flags & PSTORE_FLAGS_FRAGILE) == 0) { + pstore_register_console(); + pstore_register_ftrace(); + } if (pstore_update_ms >= 0) { pstore_timer.expires = jiffies + diff --git a/include/linux/pstore.h b/include/linux/pstore.h index abd437d0a8a..ece0c6bbfcc 100644 --- a/include/linux/pstore.h +++ b/include/linux/pstore.h @@ -51,6 +51,7 @@ struct pstore_info { char *buf; size_t bufsize; struct mutex read_mutex; /* serialize open/read/close */ + int flags; int (*open)(struct pstore_info *psi); int (*close)(struct pstore_info *psi); ssize_t (*read)(u64 *id, enum pstore_type_id *type, @@ -70,6 +71,8 @@ struct pstore_info { void *data; }; +#define PSTORE_FLAGS_FRAGILE 1 + #ifdef CONFIG_PSTORE extern int pstore_register(struct pstore_info *); extern bool pstore_cannot_block_path(enum kmsg_dump_reason reason); -- cgit v1.2.3-70-g09d2 From 8e321fefb0e60bae4e2a28d20fc4fa30758d27c6 Mon Sep 17 00:00:00 2001 From: Benjamin LaHaise Date: Sat, 21 Dec 2013 17:56:08 -0500 Subject: aio/migratepages: make aio migrate pages sane The arbitrary restriction on page counts offered by the core migrate_page_move_mapping() code results in rather suspicious looking fiddling with page reference counts in the aio_migratepage() operation. To fix this, make migrate_page_move_mapping() take an extra_count parameter that allows aio to tell the code about its own reference count on the page being migrated. While cleaning up aio_migratepage(), make it validate that the old page being passed in is actually what aio_migratepage() expects to prevent misbehaviour in the case of races. Signed-off-by: Benjamin LaHaise --- fs/aio.c | 52 +++++++++++++++++++++++++++++++++++++++++-------- include/linux/migrate.h | 3 ++- mm/migrate.c | 13 +++++++------ 3 files changed, 53 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/fs/aio.c b/fs/aio.c index fd1c0baf15b..efa708b2905 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -244,9 +244,14 @@ static void aio_free_ring(struct kioctx *ctx) int i; for (i = 0; i < ctx->nr_pages; i++) { + struct page *page; pr_debug("pid(%d) [%d] page->count=%d\n", current->pid, i, page_count(ctx->ring_pages[i])); - put_page(ctx->ring_pages[i]); + page = ctx->ring_pages[i]; + if (!page) + continue; + ctx->ring_pages[i] = NULL; + put_page(page); } put_aio_ring_file(ctx); @@ -280,18 +285,38 @@ static int aio_migratepage(struct address_space *mapping, struct page *new, unsigned long flags; int rc; + rc = 0; + + /* Make sure the old page hasn't already been changed */ + spin_lock(&mapping->private_lock); + ctx = mapping->private_data; + if (ctx) { + pgoff_t idx; + spin_lock_irqsave(&ctx->completion_lock, flags); + idx = old->index; + if (idx < (pgoff_t)ctx->nr_pages) { + if (ctx->ring_pages[idx] != old) + rc = -EAGAIN; + } else + rc = -EINVAL; + spin_unlock_irqrestore(&ctx->completion_lock, flags); + } else + rc = -EINVAL; + spin_unlock(&mapping->private_lock); + + if (rc != 0) + return rc; + /* Writeback must be complete */ BUG_ON(PageWriteback(old)); - put_page(old); + get_page(new); - rc = migrate_page_move_mapping(mapping, new, old, NULL, mode); + rc = migrate_page_move_mapping(mapping, new, old, NULL, mode, 1); if (rc != MIGRATEPAGE_SUCCESS) { - get_page(old); + put_page(new); return rc; } - get_page(new); - /* We can potentially race against kioctx teardown here. Use the * address_space's private data lock to protect the mapping's * private_data. @@ -303,13 +328,24 @@ static int aio_migratepage(struct address_space *mapping, struct page *new, spin_lock_irqsave(&ctx->completion_lock, flags); migrate_page_copy(new, old); idx = old->index; - if (idx < (pgoff_t)ctx->nr_pages) - ctx->ring_pages[idx] = new; + if (idx < (pgoff_t)ctx->nr_pages) { + /* And only do the move if things haven't changed */ + if (ctx->ring_pages[idx] == old) + ctx->ring_pages[idx] = new; + else + rc = -EAGAIN; + } else + rc = -EINVAL; spin_unlock_irqrestore(&ctx->completion_lock, flags); } else rc = -EBUSY; spin_unlock(&mapping->private_lock); + if (rc == MIGRATEPAGE_SUCCESS) + put_page(old); + else + put_page(new); + return rc; } #endif diff --git a/include/linux/migrate.h b/include/linux/migrate.h index b7717d74da7..f015c059e15 100644 --- a/include/linux/migrate.h +++ b/include/linux/migrate.h @@ -55,7 +55,8 @@ extern int migrate_huge_page_move_mapping(struct address_space *mapping, struct page *newpage, struct page *page); extern int migrate_page_move_mapping(struct address_space *mapping, struct page *newpage, struct page *page, - struct buffer_head *head, enum migrate_mode mode); + struct buffer_head *head, enum migrate_mode mode, + int extra_count); #else static inline void putback_lru_pages(struct list_head *l) {} diff --git a/mm/migrate.c b/mm/migrate.c index e9b71020133..9194375b230 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -317,14 +317,15 @@ static inline bool buffer_migrate_lock_buffers(struct buffer_head *head, */ int migrate_page_move_mapping(struct address_space *mapping, struct page *newpage, struct page *page, - struct buffer_head *head, enum migrate_mode mode) + struct buffer_head *head, enum migrate_mode mode, + int extra_count) { - int expected_count = 0; + int expected_count = 1 + extra_count; void **pslot; if (!mapping) { /* Anonymous page without mapping */ - if (page_count(page) != 1) + if (page_count(page) != expected_count) return -EAGAIN; return MIGRATEPAGE_SUCCESS; } @@ -334,7 +335,7 @@ int migrate_page_move_mapping(struct address_space *mapping, pslot = radix_tree_lookup_slot(&mapping->page_tree, page_index(page)); - expected_count = 2 + page_has_private(page); + expected_count += 1 + page_has_private(page); if (page_count(page) != expected_count || radix_tree_deref_slot_protected(pslot, &mapping->tree_lock) != page) { spin_unlock_irq(&mapping->tree_lock); @@ -584,7 +585,7 @@ int migrate_page(struct address_space *mapping, BUG_ON(PageWriteback(page)); /* Writeback must be complete */ - rc = migrate_page_move_mapping(mapping, newpage, page, NULL, mode); + rc = migrate_page_move_mapping(mapping, newpage, page, NULL, mode, 0); if (rc != MIGRATEPAGE_SUCCESS) return rc; @@ -611,7 +612,7 @@ int buffer_migrate_page(struct address_space *mapping, head = page_buffers(page); - rc = migrate_page_move_mapping(mapping, newpage, page, head, mode); + rc = migrate_page_move_mapping(mapping, newpage, page, head, mode, 0); if (rc != MIGRATEPAGE_SUCCESS) return rc; -- cgit v1.2.3-70-g09d2