June 2018 - Linux-stable-mirror

[PATCH v2 3/4] drm/amdgpu: Add amdgpu_atpx_get_dhandle()

by Lyude Paul

The handles for ACPI methods like ATPX and ATIF will live under the integrated GPU's namespace on systems with more then one GPU. ATPX is already detected regardless of the namespace it lives in, and it will always be in the same namespace as ATIF. So we can make things easier on ourselves by just adding some functions to retrieve said namespace so it can be searched later for ATIF. Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: stable(a)vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 6 ++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c | 6 ++++++ 2 files changed, 12 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 7df5d3d11aff..7dcbac8af9a7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1858,6 +1858,12 @@ static inline bool amdgpu_atpx_dgpu_req_power_for_displays(void) { return false; static inline bool amdgpu_has_atpx(void) { return false; } #endif +#if defined(CONFIG_VGA_SWITCHEROO) && defined(CONFIG_ACPI) +void *amdgpu_atpx_get_dhandle(void); +#else +static inline void *amdgpu_atpx_get_dhandle(void) { return NULL; } +#endif + /* * KMS */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c index a3896412a019..b33f1680c9a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c @@ -90,6 +90,12 @@ bool amdgpu_atpx_dgpu_req_power_for_displays(void) { return amdgpu_atpx_priv.atpx.dgpu_req_power_for_displays; } +#if defined(CONFIG_ACPI) +void *amdgpu_atpx_get_dhandle(void) { + return amdgpu_atpx_priv.dhandle; +} +#endif + /** * amdgpu_atpx_call - call an ATPX method * -- 2.17.1

6 years, 7 months

1
0
0 0

[PATCH v2 1/4] drm/amdgpu: Make struct amdgpu_atif private to amdgpu_acpi.c

by Lyude Paul

Currently, there is nothing in amdgpu that actually uses these structs other than amdgpu_acpi.c. Additionally, since we're about to start saving the correct ACPI handle to use for calling ATIF in this struct this saves us from having to handle making sure that the acpi_handle (and by proxy, the type definition for acpi_handle and all of the other acpi headers) doesn't need to be included within the amdgpu_drv struct itself. This follows the example set by amdgpu_atpx_handler.c. Signed-off-by: Lyude Paul <lyude(a)redhat.com> Cc: stable(a)vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 40 +----------------- drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c | 54 ++++++++++++++++++++++-- 2 files changed, 53 insertions(+), 41 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index a59c07590cee..7df5d3d11aff 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -190,6 +190,7 @@ struct amdgpu_job; struct amdgpu_irq_src; struct amdgpu_fpriv; struct amdgpu_bo_va_mapping; +struct amdgpu_atif; enum amdgpu_cp_irq { AMDGPU_CP_IRQ_GFX_EOP = 0, @@ -1269,43 +1270,6 @@ struct amdgpu_vram_scratch { /* * ACPI */ -struct amdgpu_atif_notification_cfg { - bool enabled; - int command_code; -}; - -struct amdgpu_atif_notifications { - bool display_switch; - bool expansion_mode_change; - bool thermal_state; - bool forced_power_state; - bool system_power_state; - bool display_conf_change; - bool px_gfx_switch; - bool brightness_change; - bool dgpu_display_event; -}; - -struct amdgpu_atif_functions { - bool system_params; - bool sbios_requests; - bool select_active_disp; - bool lid_state; - bool get_tv_standard; - bool set_tv_standard; - bool get_panel_expansion_mode; - bool set_panel_expansion_mode; - bool temperature_change; - bool graphics_device_types; -}; - -struct amdgpu_atif { - struct amdgpu_atif_notifications notifications; - struct amdgpu_atif_functions functions; - struct amdgpu_atif_notification_cfg notification_cfg; - struct amdgpu_encoder *encoder_for_bl; -}; - struct amdgpu_atcs_functions { bool get_ext_state; bool pcie_perf_req; @@ -1466,7 +1430,7 @@ struct amdgpu_device { #if defined(CONFIG_DEBUG_FS) struct dentry *debugfs_regs[AMDGPU_DEBUGFS_MAX_COMPONENTS]; #endif - struct amdgpu_atif atif; + struct amdgpu_atif *atif; struct amdgpu_atcs atcs; struct mutex srbm_mutex; /* GRBM index mutex. Protects concurrent access to GRBM index */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index 8fa850a070e0..22c7e8ec0b9a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -34,6 +34,43 @@ #include "amd_acpi.h" #include "atom.h" +struct amdgpu_atif_notification_cfg { + bool enabled; + int command_code; +}; + +struct amdgpu_atif_notifications { + bool display_switch; + bool expansion_mode_change; + bool thermal_state; + bool forced_power_state; + bool system_power_state; + bool display_conf_change; + bool px_gfx_switch; + bool brightness_change; + bool dgpu_display_event; +}; + +struct amdgpu_atif_functions { + bool system_params; + bool sbios_requests; + bool select_active_disp; + bool lid_state; + bool get_tv_standard; + bool set_tv_standard; + bool get_panel_expansion_mode; + bool set_panel_expansion_mode; + bool temperature_change; + bool graphics_device_types; +}; + +struct amdgpu_atif { + struct amdgpu_atif_notifications notifications; + struct amdgpu_atif_functions functions; + struct amdgpu_atif_notification_cfg notification_cfg; + struct amdgpu_encoder *encoder_for_bl; +}; + /* Call the ATIF method */ /** @@ -292,7 +329,7 @@ static int amdgpu_atif_get_sbios_requests(acpi_handle handle, static int amdgpu_atif_handler(struct amdgpu_device *adev, struct acpi_bus_event *event) { - struct amdgpu_atif *atif = &adev->atif; + struct amdgpu_atif *atif = adev->atif; struct atif_sbios_requests req; acpi_handle handle; int count; @@ -303,7 +340,8 @@ static int amdgpu_atif_handler(struct amdgpu_device *adev, if (strcmp(event->device_class, ACPI_VIDEO_CLASS) != 0) return NOTIFY_DONE; - if (!atif->notification_cfg.enabled || + if (!atif || + !atif->notification_cfg.enabled || event->type != atif->notification_cfg.command_code) /* Not our event */ return NOTIFY_DONE; @@ -642,7 +680,7 @@ static int amdgpu_acpi_event(struct notifier_block *nb, int amdgpu_acpi_init(struct amdgpu_device *adev) { acpi_handle handle; - struct amdgpu_atif *atif = &adev->atif; + struct amdgpu_atif *atif; struct amdgpu_atcs *atcs = &adev->atcs; int ret; @@ -659,11 +697,19 @@ int amdgpu_acpi_init(struct amdgpu_device *adev) } /* Call the ATIF method */ + atif = kzalloc(sizeof(*atif), GFP_KERNEL); + if (!atif) { + DRM_WARN("Not enough memory to initialize ATIF\n"); + goto out; + } + ret = amdgpu_atif_verify_interface(handle, atif); if (ret) { DRM_DEBUG_DRIVER("Call to ATIF verify_interface failed: %d\n", ret); + kfree(atif); goto out; } + adev->atif = atif; if (atif->notifications.brightness_change) { struct drm_encoder *tmp; @@ -720,4 +766,6 @@ int amdgpu_acpi_init(struct amdgpu_device *adev) void amdgpu_acpi_fini(struct amdgpu_device *adev) { unregister_acpi_notifier(&adev->acpi_nb); + if (adev->atif) + kfree(adev->atif); } -- 2.17.1

6 years, 7 months

1
0
0 0

[PATCH] block: Fix cloning of requests with a special payload

by Bart Van Assche

This patch avoids that removing a path controlled by the dm-mpath driver while mkfs is running triggers the following kernel bug: kernel BUG at block/blk-core.c:3347! invalid opcode: 0000 [#1] PREEMPT SMP KASAN CPU: 20 PID: 24369 Comm: mkfs.ext4 Not tainted 4.18.0-rc1-dbg+ #2 RIP: 0010:blk_end_request_all+0x68/0x70 Call Trace: <IRQ> dm_softirq_done+0x326/0x3d0 [dm_mod] blk_done_softirq+0x19b/0x1e0 __do_softirq+0x128/0x60d irq_exit+0x100/0x110 smp_call_function_single_interrupt+0x90/0x330 call_function_single_interrupt+0xf/0x20 </IRQ> Fixes: f9d03f96b988 ("block: improve handling of the magic discard payload") Signed-off-by: Bart Van Assche <bart.vanassche(a)wdc.com> Cc: Christoph Hellwig <hch(a)lst.de> Cc: Mike Snitzer <snitzer(a)redhat.com> Cc: Ming Lei <ming.lei(a)redhat.com> Cc: Hannes Reinecke <hare(a)suse.com> Cc: Johannes Thumshirn <jthumshirn(a)suse.de> Cc: <stable(a)vger.kernel.org> --- block/blk-core.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/block/blk-core.c b/block/blk-core.c index 118dd17eb71f..f1e07ed1513c 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -3529,6 +3529,10 @@ static void __blk_rq_prep_clone(struct request *dst, struct request *src) dst->cpu = src->cpu; dst->__sector = blk_rq_pos(src); dst->__data_len = blk_rq_bytes(src); + if (src->rq_flags & RQF_SPECIAL_PAYLOAD) { + dst->rq_flags |= RQF_SPECIAL_PAYLOAD; + dst->special_vec = src->special_vec; + } dst->nr_phys_segments = src->nr_phys_segments; dst->ioprio = src->ioprio; dst->extra_len = src->extra_len; -- 2.17.1

6 years, 7 months

6
6
0 0

[PATCH v7] tpm: separate cmd_ready/go_idle from runtime_pm

by Tomas Winkler

Fix tpm ptt initialization error: tpm tpm0: A TPM error (378) occurred get tpm pcr allocation. We cannot use go_idle cmd_ready commands via runtime_pm handles as with the introduction of localities this is no longer an optional feature, while runtime pm can be not enabled. Though cmd_ready/go_idle provides a power saving, it's also a part of TPM2 protocol and should be called explicitly. This patch exposes cmd_read/go_idle via tpm class ops and removes runtime pm support as it is not used by any driver. A new tpm transmit flag is added TPM_TRANSMIT_NESTED, which implies TPM_TRANSMIT_UNLOCKED and TPM_TRANSMIT_RAW. Both are needed to resolve tpm spaces and locality request recursive calls to tpm_transmit(). New wrappers are added tpm_cmd_ready() and tpm_go_idle() to streamline tpm_try_transmit code. tpm_crb no longer needs own power saving functions and can drop using tpm_pm_suspend/resume. This patch cannot be really separated from the locality fix. Fixes: 888d867df441 (tpm: cmd_ready command can be issued only after granting locality) Cc: stable(a)vger.kernel.org Fixes: 888d867df441 (tpm: cmd_ready command can be issued only after granting locality) Signed-off-by: Tomas Winkler <tomas.winkler(a)intel.com> --- V2-3:resent V4: 1. Use tpm_pm_suspend/resume in tpm_crb 2. Drop cmd_ready/go_idle around tpm_chip_register, as there is no usage counter like in runtime_pm. V5: 1. add tpm_cmd_ready and tpm_go_idle wrappers. 2. Abuse TPM_TRANSMIT_UNLOCKED flag to resolve tpm space recursion. V6: 1. Add new flags TPM_TRANSMIT_NESTED which implies TPM_TRANSMIT_UNLOCKED and TPM_TRANSMIT_RAW, as not all unlocked flows are recursive i.e. tpm2_unseal_trusted. V7: 1. Add dmesg dump. drivers/char/tpm/tpm-interface.c | 51 +++++++++++++++---- drivers/char/tpm/tpm.h | 13 ++++- drivers/char/tpm/tpm2-space.c | 12 ++--- drivers/char/tpm/tpm_crb.c | 101 ++++++++++---------------------------- drivers/char/tpm/tpm_vtpm_proxy.c | 2 +- include/linux/tpm.h | 2 + 6 files changed, 88 insertions(+), 93 deletions(-) diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c index e32f6e85dc6d..1abd8261651b 100644 --- a/drivers/char/tpm/tpm-interface.c +++ b/drivers/char/tpm/tpm-interface.c @@ -29,7 +29,6 @@ #include <linux/mutex.h> #include <linux/spinlock.h> #include <linux/freezer.h> -#include <linux/pm_runtime.h> #include <linux/tpm_eventlog.h> #include "tpm.h" @@ -369,10 +368,13 @@ static int tpm_validate_command(struct tpm_chip *chip, return -EINVAL; } -static int tpm_request_locality(struct tpm_chip *chip) +static int tpm_request_locality(struct tpm_chip *chip, unsigned int flags) { int rc; + if (flags & __TPM_TRANSMIT_RAW) + return 0; + if (!chip->ops->request_locality) return 0; @@ -385,10 +387,13 @@ static int tpm_request_locality(struct tpm_chip *chip) return 0; } -static void tpm_relinquish_locality(struct tpm_chip *chip) +static void tpm_relinquish_locality(struct tpm_chip *chip, unsigned int flags) { int rc; + if (flags & __TPM_TRANSMIT_RAW) + return; + if (!chip->ops->relinquish_locality) return; @@ -399,6 +404,28 @@ static void tpm_relinquish_locality(struct tpm_chip *chip) chip->locality = -1; } +static int tpm_cmd_ready(struct tpm_chip *chip, unsigned int flags) +{ + if (flags & __TPM_TRANSMIT_RAW) + return 0; + + if (!chip->ops->cmd_ready) + return 0; + + return chip->ops->cmd_ready(chip); +} + +static int tpm_go_idle(struct tpm_chip *chip, unsigned int flags) +{ + if (flags & __TPM_TRANSMIT_RAW) + return 0; + + if (!chip->ops->go_idle) + return 0; + + return chip->ops->go_idle(chip); +} + static ssize_t tpm_try_transmit(struct tpm_chip *chip, struct tpm_space *space, u8 *buf, size_t bufsiz, @@ -449,14 +476,15 @@ static ssize_t tpm_try_transmit(struct tpm_chip *chip, /* Store the decision as chip->locality will be changed. */ need_locality = chip->locality == -1; - if (!(flags & TPM_TRANSMIT_RAW) && need_locality) { - rc = tpm_request_locality(chip); + if (need_locality) { + rc = tpm_request_locality(chip, flags); if (rc < 0) goto out_no_locality; } - if (chip->dev.parent) - pm_runtime_get_sync(chip->dev.parent); + rc = tpm_cmd_ready(chip, flags); + if (rc) + goto out; rc = tpm2_prepare_space(chip, space, ordinal, buf); if (rc) @@ -516,13 +544,16 @@ static ssize_t tpm_try_transmit(struct tpm_chip *chip, } rc = tpm2_commit_space(chip, space, ordinal, buf, &len); + if (rc) + dev_err(&chip->dev, "tpm2_commit_space: error %d\n", rc); out: - if (chip->dev.parent) - pm_runtime_put_sync(chip->dev.parent); + rc = tpm_go_idle(chip, flags); + if (rc) + goto out; if (need_locality) - tpm_relinquish_locality(chip); + tpm_relinquish_locality(chip, flags); out_no_locality: if (chip->ops->clk_enable != NULL) diff --git a/drivers/char/tpm/tpm.h b/drivers/char/tpm/tpm.h index 9824cccb2c76..a711901a9cf5 100644 --- a/drivers/char/tpm/tpm.h +++ b/drivers/char/tpm/tpm.h @@ -512,9 +512,18 @@ extern const struct file_operations tpm_fops; extern const struct file_operations tpmrm_fops; extern struct idr dev_nums_idr; +/** + * enum tpm_transmit_flags + * + * @TPM_TRANSMIT_UNLOCKED: used to lock sequence of tpm_transmit calls. + * @__TPM_TRANSMIT_RAW: prevent recursive calls into setup steps + * (go idle, locality,..). Don't use directly. + * @TPM_TRANSMIT_NESTED: Use from nested tpm_transmit calls + */ enum tpm_transmit_flags { - TPM_TRANSMIT_UNLOCKED = BIT(0), - TPM_TRANSMIT_RAW = BIT(1), + TPM_TRANSMIT_UNLOCKED = BIT(0), + __TPM_TRANSMIT_RAW = BIT(1), + TPM_TRANSMIT_NESTED = TPM_TRANSMIT_UNLOCKED | __TPM_TRANSMIT_RAW, }; ssize_t tpm_transmit(struct tpm_chip *chip, struct tpm_space *space, diff --git a/drivers/char/tpm/tpm2-space.c b/drivers/char/tpm/tpm2-space.c index 6122d3276f72..d2e101b32482 100644 --- a/drivers/char/tpm/tpm2-space.c +++ b/drivers/char/tpm/tpm2-space.c @@ -39,7 +39,7 @@ static void tpm2_flush_sessions(struct tpm_chip *chip, struct tpm_space *space) for (i = 0; i < ARRAY_SIZE(space->session_tbl); i++) { if (space->session_tbl[i]) tpm2_flush_context_cmd(chip, space->session_tbl[i], - TPM_TRANSMIT_UNLOCKED); + TPM_TRANSMIT_NESTED); } } @@ -84,7 +84,7 @@ static int tpm2_load_context(struct tpm_chip *chip, u8 *buf, tpm_buf_append(&tbuf, &buf[*offset], body_size); rc = tpm_transmit_cmd(chip, NULL, tbuf.data, PAGE_SIZE, 4, - TPM_TRANSMIT_UNLOCKED, NULL); + TPM_TRANSMIT_NESTED, NULL); if (rc < 0) { dev_warn(&chip->dev, "%s: failed with a system error %d\n", __func__, rc); @@ -133,7 +133,7 @@ static int tpm2_save_context(struct tpm_chip *chip, u32 handle, u8 *buf, tpm_buf_append_u32(&tbuf, handle); rc = tpm_transmit_cmd(chip, NULL, tbuf.data, PAGE_SIZE, 0, - TPM_TRANSMIT_UNLOCKED, NULL); + TPM_TRANSMIT_NESTED, NULL); if (rc < 0) { dev_warn(&chip->dev, "%s: failed with a system error %d\n", __func__, rc); @@ -170,7 +170,7 @@ static void tpm2_flush_space(struct tpm_chip *chip) for (i = 0; i < ARRAY_SIZE(space->context_tbl); i++) if (space->context_tbl[i] && ~space->context_tbl[i]) tpm2_flush_context_cmd(chip, space->context_tbl[i], - TPM_TRANSMIT_UNLOCKED); + TPM_TRANSMIT_NESTED); tpm2_flush_sessions(chip, space); } @@ -377,7 +377,7 @@ static int tpm2_map_response_header(struct tpm_chip *chip, u32 cc, u8 *rsp, return 0; out_no_slots: - tpm2_flush_context_cmd(chip, phandle, TPM_TRANSMIT_UNLOCKED); + tpm2_flush_context_cmd(chip, phandle, TPM_TRANSMIT_NESTED); dev_warn(&chip->dev, "%s: out of slots for 0x%08X\n", __func__, phandle); return -ENOMEM; @@ -465,7 +465,7 @@ static int tpm2_save_space(struct tpm_chip *chip) return rc; tpm2_flush_context_cmd(chip, space->context_tbl[i], - TPM_TRANSMIT_UNLOCKED); + TPM_TRANSMIT_NESTED); space->context_tbl[i] = ~0; } diff --git a/drivers/char/tpm/tpm_crb.c b/drivers/char/tpm/tpm_crb.c index 34fbc6cb097b..36952ef98f90 100644 --- a/drivers/char/tpm/tpm_crb.c +++ b/drivers/char/tpm/tpm_crb.c @@ -132,7 +132,7 @@ static bool crb_wait_for_reg_32(u32 __iomem *reg, u32 mask, u32 value, } /** - * crb_go_idle - request tpm crb device to go the idle state + * __crb_go_idle - request tpm crb device to go the idle state * * @dev: crb device * @priv: crb private data @@ -147,7 +147,7 @@ static bool crb_wait_for_reg_32(u32 __iomem *reg, u32 mask, u32 value, * * Return: 0 always */ -static int crb_go_idle(struct device *dev, struct crb_priv *priv) +static int __crb_go_idle(struct device *dev, struct crb_priv *priv) { if ((priv->sm == ACPI_TPM2_START_METHOD) || (priv->sm == ACPI_TPM2_COMMAND_BUFFER_WITH_START_METHOD) || @@ -163,11 +163,20 @@ static int crb_go_idle(struct device *dev, struct crb_priv *priv) dev_warn(dev, "goIdle timed out\n"); return -ETIME; } + return 0; } +static int crb_go_idle(struct tpm_chip *chip) +{ + struct device *dev = &chip->dev; + struct crb_priv *priv = dev_get_drvdata(dev); + + return __crb_go_idle(dev, priv); +} + /** - * crb_cmd_ready - request tpm crb device to enter ready state + * __crb_cmd_ready - request tpm crb device to enter ready state * * @dev: crb device * @priv: crb private data @@ -181,7 +190,7 @@ static int crb_go_idle(struct device *dev, struct crb_priv *priv) * * Return: 0 on success -ETIME on timeout; */ -static int crb_cmd_ready(struct device *dev, struct crb_priv *priv) +static int __crb_cmd_ready(struct device *dev, struct crb_priv *priv) { if ((priv->sm == ACPI_TPM2_START_METHOD) || (priv->sm == ACPI_TPM2_COMMAND_BUFFER_WITH_START_METHOD) || @@ -200,6 +209,14 @@ static int crb_cmd_ready(struct device *dev, struct crb_priv *priv) return 0; } +static int crb_cmd_ready(struct tpm_chip *chip) +{ + struct device *dev = &chip->dev; + struct crb_priv *priv = dev_get_drvdata(dev); + + return __crb_cmd_ready(dev, priv); +} + static int __crb_request_locality(struct device *dev, struct crb_priv *priv, int loc) { @@ -401,6 +418,8 @@ static const struct tpm_class_ops tpm_crb = { .send = crb_send, .cancel = crb_cancel, .req_canceled = crb_req_canceled, + .go_idle = crb_go_idle, + .cmd_ready = crb_cmd_ready, .request_locality = crb_request_locality, .relinquish_locality = crb_relinquish_locality, .req_complete_mask = CRB_DRV_STS_COMPLETE, @@ -520,7 +539,7 @@ static int crb_map_io(struct acpi_device *device, struct crb_priv *priv, * PTT HW bug w/a: wake up the device to access * possibly not retained registers. */ - ret = crb_cmd_ready(dev, priv); + ret = __crb_cmd_ready(dev, priv); if (ret) goto out_relinquish_locality; @@ -565,7 +584,7 @@ static int crb_map_io(struct acpi_device *device, struct crb_priv *priv, if (!ret) priv->cmd_size = cmd_size; - crb_go_idle(dev, priv); + __crb_go_idle(dev, priv); out_relinquish_locality: @@ -628,32 +647,7 @@ static int crb_acpi_add(struct acpi_device *device) chip->acpi_dev_handle = device->handle; chip->flags = TPM_CHIP_FLAG_TPM2; - rc = __crb_request_locality(dev, priv, 0); - if (rc) - return rc; - - rc = crb_cmd_ready(dev, priv); - if (rc) - goto out; - - pm_runtime_get_noresume(dev); - pm_runtime_set_active(dev); - pm_runtime_enable(dev); - - rc = tpm_chip_register(chip); - if (rc) { - crb_go_idle(dev, priv); - pm_runtime_put_noidle(dev); - pm_runtime_disable(dev); - goto out; - } - - pm_runtime_put_sync(dev); - -out: - __crb_relinquish_locality(dev, priv, 0); - - return rc; + return tpm_chip_register(chip); } static int crb_acpi_remove(struct acpi_device *device) @@ -663,52 +657,11 @@ static int crb_acpi_remove(struct acpi_device *device) tpm_chip_unregister(chip); - pm_runtime_disable(dev); - return 0; } -static int __maybe_unused crb_pm_runtime_suspend(struct device *dev) -{ - struct tpm_chip *chip = dev_get_drvdata(dev); - struct crb_priv *priv = dev_get_drvdata(&chip->dev); - - return crb_go_idle(dev, priv); -} - -static int __maybe_unused crb_pm_runtime_resume(struct device *dev) -{ - struct tpm_chip *chip = dev_get_drvdata(dev); - struct crb_priv *priv = dev_get_drvdata(&chip->dev); - - return crb_cmd_ready(dev, priv); -} - -static int __maybe_unused crb_pm_suspend(struct device *dev) -{ - int ret; - - ret = tpm_pm_suspend(dev); - if (ret) - return ret; - - return crb_pm_runtime_suspend(dev); -} - -static int __maybe_unused crb_pm_resume(struct device *dev) -{ - int ret; - - ret = crb_pm_runtime_resume(dev); - if (ret) - return ret; - - return tpm_pm_resume(dev); -} - static const struct dev_pm_ops crb_pm = { - SET_SYSTEM_SLEEP_PM_OPS(crb_pm_suspend, crb_pm_resume) - SET_RUNTIME_PM_OPS(crb_pm_runtime_suspend, crb_pm_runtime_resume, NULL) + SET_SYSTEM_SLEEP_PM_OPS(tpm_pm_suspend, tpm_pm_resume) }; static const struct acpi_device_id crb_device_ids[] = { diff --git a/drivers/char/tpm/tpm_vtpm_proxy.c b/drivers/char/tpm/tpm_vtpm_proxy.c index e4f79f920450..87a0ce47f201 100644 --- a/drivers/char/tpm/tpm_vtpm_proxy.c +++ b/drivers/char/tpm/tpm_vtpm_proxy.c @@ -418,7 +418,7 @@ static int vtpm_proxy_request_locality(struct tpm_chip *chip, int locality) proxy_dev->state |= STATE_DRIVER_COMMAND; rc = tpm_transmit_cmd(chip, NULL, buf.data, tpm_buf_length(&buf), 0, - TPM_TRANSMIT_UNLOCKED | TPM_TRANSMIT_RAW, + TPM_TRANSMIT_NESTED, "attempting to set locality"); proxy_dev->state &= ~STATE_DRIVER_COMMAND; diff --git a/include/linux/tpm.h b/include/linux/tpm.h index 06639fb6ab85..8eb5e5ebe136 100644 --- a/include/linux/tpm.h +++ b/include/linux/tpm.h @@ -43,6 +43,8 @@ struct tpm_class_ops { u8 (*status) (struct tpm_chip *chip); bool (*update_timeouts)(struct tpm_chip *chip, unsigned long *timeout_cap); + int (*go_idle)(struct tpm_chip *chip); + int (*cmd_ready)(struct tpm_chip *chip); int (*request_locality)(struct tpm_chip *chip, int loc); int (*relinquish_locality)(struct tpm_chip *chip, int loc); void (*clk_enable)(struct tpm_chip *chip, bool value); -- 2.14.4

6 years, 7 months

3
9
0 0

Re: Patch "x86/platform/UV: Use new set memory block size function" has been added to the 4.14-stable tree

by Mike Travis

Until the original patch that adds the set_memory_block_size_order() function is added, then the following two patches should also not be applied. In fact I'm surprised it built with that function undefined. Thanks, Mike On 6/27/2018 7:08 PM, gregkh(a)linuxfoundation.org wrote: > > This is a note to let you know that I've just added the patch titled > > x86/platform/UV: Use new set memory block size function > > to the 4.14-stable tree which can be found at: > http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… > > The filename of the patch is: > x86-platform-uv-use-new-set-memory-block-size-function.patch > and it can be found in the queue-4.14 subdirectory. > > If you, or anyone else, feels it should not be added to the stable tree, > please let <stable(a)vger.kernel.org> know about it. > > > From bbbd2b51a2aa0d76b3676271e216cf3647773397 Mon Sep 17 00:00:00 2001 > From: "mike.travis(a)hpe.com" <mike.travis(a)hpe.com> > Date: Thu, 24 May 2018 15:17:13 -0500 > Subject: x86/platform/UV: Use new set memory block size function > > From: mike.travis(a)hpe.com <mike.travis(a)hpe.com> > > commit bbbd2b51a2aa0d76b3676271e216cf3647773397 upstream. > > Add a call to the new function to "adjust" the current fixed UV memory > block size of 2GB so it can be changed to a different physical boundary. > This accommodates changes in the Intel BIOS, and therefore UV BIOS, > which now can align boundaries different than the previous UV standard > of 2GB. It also flags any UV Global Address boundaries from BIOS that > cause a change in the mem block size (boundary). > > The current boundary of 2GB has been used on UV since the first system > release in 2009 with Linux 2.6 and has worked fine. But the new NVDIMM > persistent memory modules (PMEM), along with the Intel BIOS changes to > support these modules caused the memory block size boundary to be set > to a lower limit. Intel only guarantees that this minimum boundary at > 64MB though the current Linux limit is 128MB. > > Note that the default remains 2GB if no changes occur. > > Signed-off-by: Mike Travis <mike.travis(a)hpe.com> > Reviewed-by: Andrew Banman <andrew.banman(a)hpe.com> > Cc: Andrew Morton <akpm(a)linux-foundation.org> > Cc: Dimitri Sivanich <dimitri.sivanich(a)hpe.com> > Cc: Linus Torvalds <torvalds(a)linux-foundation.org> > Cc: Peter Zijlstra <peterz(a)infradead.org> > Cc: Russ Anderson <russ.anderson(a)hpe.com> > Cc: Thomas Gleixner <tglx(a)linutronix.de> > Cc: dan.j.williams(a)intel.com > Cc: jgross(a)suse.com > Cc: kirill.shutemov(a)linux.intel.com > Cc: mhocko(a)suse.com > Cc: stable(a)vger.kernel.org > Link: https://lkml.kernel.org/lkml/20180524201711.732785782@stormcage.americas.sg… > Signed-off-by: Ingo Molnar <mingo(a)kernel.org> > Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> > > --- > arch/x86/kernel/apic/x2apic_uv_x.c | 49 ++++++++++++++++++++++++++++++++++--- > 1 file changed, 46 insertions(+), 3 deletions(-) > > --- a/arch/x86/kernel/apic/x2apic_uv_x.c > +++ b/arch/x86/kernel/apic/x2apic_uv_x.c > @@ -26,6 +26,7 @@ > #include <linux/delay.h> > #include <linux/crash_dump.h> > #include <linux/reboot.h> > +#include <linux/memory.h> > > #include <asm/uv/uv_mmrs.h> > #include <asm/uv/uv_hub.h> > @@ -346,6 +347,40 @@ extern int uv_hub_info_version(void) > } > EXPORT_SYMBOL(uv_hub_info_version); > > +/* Default UV memory block size is 2GB */ > +static unsigned long mem_block_size = (2UL << 30); > + > +static __init int adj_blksize(u32 lgre) > +{ > + unsigned long base = (unsigned long)lgre << UV_GAM_RANGE_SHFT; > + unsigned long size; > + > + for (size = mem_block_size; size > MIN_MEMORY_BLOCK_SIZE; size >>= 1) > + if (IS_ALIGNED(base, size)) > + break; > + > + if (size >= mem_block_size) > + return 0; > + > + mem_block_size = size; > + return 1; > +} > + > +static __init void set_block_size(void) > +{ > + unsigned int order = ffs(mem_block_size); > + > + if (order) { > + /* adjust for ffs return of 1..64 */ > + set_memory_block_size_order(order - 1); > + pr_info("UV: mem_block_size set to 0x%lx\n", mem_block_size); > + } else { > + /* bad or zero value, default to 1UL << 31 (2GB) */ > + pr_err("UV: mem_block_size error with 0x%lx\n", mem_block_size); > + set_memory_block_size_order(31); > + } > +} > + > /* Build GAM range lookup table: */ > static __init void build_uv_gr_table(void) > { > @@ -1144,23 +1179,30 @@ static void __init decode_gam_rng_tbl(un > << UV_GAM_RANGE_SHFT); > int order = 0; > char suffix[] = " KMGTPE"; > + int flag = ' '; > > while (size > 9999 && order < sizeof(suffix)) { > size /= 1024; > order++; > } > > + /* adjust max block size to current range start */ > + if (gre->type == 1 || gre->type == 2) > + if (adj_blksize(lgre)) > + flag = '*'; > + > if (!index) { > pr_info("UV: GAM Range Table...\n"); > - pr_info("UV: # %20s %14s %5s %4s %5s %3s %2s\n", "Range", "", "Size", "Type", "NASID", "SID", "PN"); > + pr_info("UV: # %20s %14s %6s %4s %5s %3s %2s\n", "Range", "", "Size", "Type", "NASID", "SID", "PN"); > } > - pr_info("UV: %2d: 0x%014lx-0x%014lx %5lu%c %3d %04x %02x %02x\n", > + pr_info("UV: %2d: 0x%014lx-0x%014lx%c %5lu%c %3d %04x %02x %02x\n", > index++, > (unsigned long)lgre << UV_GAM_RANGE_SHFT, > (unsigned long)gre->limit << UV_GAM_RANGE_SHFT, > - size, suffix[order], > + flag, size, suffix[order], > gre->type, gre->nasid, gre->sockid, gre->pnode); > > + /* update to next range start */ > lgre = gre->limit; > if (sock_min > gre->sockid) > sock_min = gre->sockid; > @@ -1391,6 +1433,7 @@ static void __init uv_system_init_hub(vo > > build_socket_tables(); > build_uv_gr_table(); > + set_block_size(); > uv_init_hub_info(&hub_info); > uv_possible_blades = num_possible_nodes(); > if (!_node_to_pnode) > > > Patches currently in stable-queue which might be from mike.travis(a)hpe.com are > > queue-4.14/x86-platform-uv-add-kernel-parameter-to-set-memory-block-size.patch > queue-4.14/x86-platform-uv-use-new-set-memory-block-size-function.patch >

6 years, 7 months

2
1
0 0

patch "serdev: fix memleak on module unload" added to tty-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled serdev: fix memleak on module unload to my tty git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty.git in the tty-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. >From bc6cf3669d22371f573ab0305b3abf13887c0786 Mon Sep 17 00:00:00 2001 From: Johan Hovold <johan(a)kernel.org> Date: Wed, 13 Jun 2018 17:08:59 +0200 Subject: serdev: fix memleak on module unload Make sure to free all resources associated with the ida on module exit. Fixes: cd6484e1830b ("serdev: Introduce new bus for serial attached devices") Cc: stable <stable(a)vger.kernel.org> # 4.11 Signed-off-by: Johan Hovold <johan(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/tty/serdev/core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/tty/serdev/core.c b/drivers/tty/serdev/core.c index df93b727e984..9e59f4788589 100644 --- a/drivers/tty/serdev/core.c +++ b/drivers/tty/serdev/core.c @@ -617,6 +617,7 @@ EXPORT_SYMBOL_GPL(__serdev_device_driver_register); static void __exit serdev_exit(void) { bus_unregister(&serdev_bus_type); + ida_destroy(&ctrl_ida); } module_exit(serdev_exit); -- 2.18.0

6 years, 7 months

1
0
0 0

patch "vt: prevent leaking uninitialized data to userspace via /dev/vcs*" added to tty-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled vt: prevent leaking uninitialized data to userspace via /dev/vcs* to my tty git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty.git in the tty-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. >From 21eff69aaaa0e766ca0ce445b477698dc6a9f55a Mon Sep 17 00:00:00 2001 From: Alexander Potapenko <glider(a)google.com> Date: Thu, 14 Jun 2018 12:23:09 +0200 Subject: vt: prevent leaking uninitialized data to userspace via /dev/vcs* KMSAN reported an infoleak when reading from /dev/vcs*: BUG: KMSAN: kernel-infoleak in vcs_read+0x18ba/0x1cc0 Call Trace: ... kmsan_copy_to_user+0x7a/0x160 mm/kmsan/kmsan.c:1253 copy_to_user ./include/linux/uaccess.h:184 vcs_read+0x18ba/0x1cc0 drivers/tty/vt/vc_screen.c:352 __vfs_read+0x1b2/0x9d0 fs/read_write.c:416 vfs_read+0x36c/0x6b0 fs/read_write.c:452 ... Uninit was created at: kmsan_save_stack_with_flags mm/kmsan/kmsan.c:279 kmsan_internal_poison_shadow+0xb8/0x1b0 mm/kmsan/kmsan.c:189 kmsan_kmalloc+0x94/0x100 mm/kmsan/kmsan.c:315 __kmalloc+0x13a/0x350 mm/slub.c:3818 kmalloc ./include/linux/slab.h:517 vc_allocate+0x438/0x800 drivers/tty/vt/vt.c:787 con_install+0x8c/0x640 drivers/tty/vt/vt.c:2880 tty_driver_install_tty drivers/tty/tty_io.c:1224 tty_init_dev+0x1b5/0x1020 drivers/tty/tty_io.c:1324 tty_open_by_driver drivers/tty/tty_io.c:1959 tty_open+0x17b4/0x2ed0 drivers/tty/tty_io.c:2007 chrdev_open+0xc25/0xd90 fs/char_dev.c:417 do_dentry_open+0xccc/0x1440 fs/open.c:794 vfs_open+0x1b6/0x2f0 fs/open.c:908 ... Bytes 0-79 of 240 are uninitialized Consistently allocating |vc_screenbuf| with kzalloc() fixes the problem Reported-by: syzbot+17a8efdf800000(a)syzkaller.appspotmail.com Signed-off-by: Alexander Potapenko <glider(a)google.com> Cc: stable <stable(a)vger.kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/tty/vt/vt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 1eb1a376a041..15eb6c829d39 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -784,7 +784,7 @@ int vc_allocate(unsigned int currcons) /* return 0 on success */ if (!*vc->vc_uni_pagedir_loc) con_set_default_unimap(vc); - vc->vc_screenbuf = kmalloc(vc->vc_screenbuf_size, GFP_KERNEL); + vc->vc_screenbuf = kzalloc(vc->vc_screenbuf_size, GFP_KERNEL); if (!vc->vc_screenbuf) goto err_free; @@ -871,7 +871,7 @@ static int vc_do_resize(struct tty_struct *tty, struct vc_data *vc, if (new_screen_size > (4 << 20)) return -EINVAL; - newscreen = kmalloc(new_screen_size, GFP_USER); + newscreen = kzalloc(new_screen_size, GFP_USER); if (!newscreen) return -ENOMEM; -- 2.18.0

6 years, 7 months

1
0
0 0

patch "serial: 8250_pci: Remove stalled entries in blacklist" added to tty-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled serial: 8250_pci: Remove stalled entries in blacklist to my tty git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty.git in the tty-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. >From 20dcff436e9fcd2e106b0ccc48a52206bc176d70 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Date: Wed, 6 Jun 2018 21:00:41 +0300 Subject: serial: 8250_pci: Remove stalled entries in blacklist After the commit 7d8905d06405 ("serial: 8250_pci: Enable device after we check black list") pure serial multi-port cards, such as CH355, got blacklisted and thus not being enumerated anymore. Previously, it seems, blacklisting them was on purpose to shut up pciserial_init_one() about record duplication. So, remove the entries from blacklist in order to get cards enumerated. Fixes: 7d8905d06405 ("serial: 8250_pci: Enable device after we check black list") Reported-by: Matt Turner <mattst88(a)gmail.com> Cc: Sergej Pupykin <ml(a)sergej.pp.ru> Cc: Alexandr Petrenko <petrenkoas83(a)gmail.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Reviewed-and-Tested-by: Matt Turner <mattst88(a)gmail.com> Cc: stable <stable(a)vger.kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/tty/serial/8250/8250_pci.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/tty/serial/8250/8250_pci.c b/drivers/tty/serial/8250/8250_pci.c index 3296a05cda2d..f80a300b5d68 100644 --- a/drivers/tty/serial/8250/8250_pci.c +++ b/drivers/tty/serial/8250/8250_pci.c @@ -3339,9 +3339,7 @@ static const struct pci_device_id blacklist[] = { /* multi-io cards handled by parport_serial */ { PCI_DEVICE(0x4348, 0x7053), }, /* WCH CH353 2S1P */ { PCI_DEVICE(0x4348, 0x5053), }, /* WCH CH353 1S1P */ - { PCI_DEVICE(0x4348, 0x7173), }, /* WCH CH355 4S */ { PCI_DEVICE(0x1c00, 0x3250), }, /* WCH CH382 2S1P */ - { PCI_DEVICE(0x1c00, 0x3470), }, /* WCH CH384 4S */ /* Moxa Smartio MUE boards handled by 8250_moxa */ { PCI_VDEVICE(MOXA, 0x1024), }, -- 2.18.0

6 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] x86/platform/UV: Add adjustable set memory block size" failed to apply to 4.14-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.14-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From f642fb5864a6e3645edce6f85ffe7b44d5e9b990 Mon Sep 17 00:00:00 2001 From: "mike.travis(a)hpe.com" <mike.travis(a)hpe.com> Date: Thu, 24 May 2018 15:17:12 -0500 Subject: [PATCH] x86/platform/UV: Add adjustable set memory block size function Add a new function to "adjust" the current fixed UV memory block size of 2GB so it can be changed to a different physical boundary. This is out of necessity so arch dependent code can accommodate specific BIOS requirements which can align these new PMEM modules at less than the default boundaries. A "set order" type of function was used to insure that the memory block size will be a power of two value without requiring a validity check. 64GB was chosen as the upper limit for memory block size values to accommodate upcoming 4PB systems which have 6 more bits of physical address space (46 becoming 52). Signed-off-by: Mike Travis <mike.travis(a)hpe.com> Reviewed-by: Andrew Banman <andrew.banman(a)hpe.com> Cc: Andrew Morton <akpm(a)linux-foundation.org> Cc: Dimitri Sivanich <dimitri.sivanich(a)hpe.com> Cc: Linus Torvalds <torvalds(a)linux-foundation.org> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Russ Anderson <russ.anderson(a)hpe.com> Cc: Thomas Gleixner <tglx(a)linutronix.de> Cc: dan.j.williams(a)intel.com Cc: jgross(a)suse.com Cc: kirill.shutemov(a)linux.intel.com Cc: mhocko(a)suse.com Cc: stable(a)vger.kernel.org Link: https://lkml.kernel.org/lkml/20180524201711.609546602@stormcage.americas.sg… Signed-off-by: Ingo Molnar <mingo(a)kernel.org> diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 0a400606dea0..20d8bf5fbceb 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -1350,16 +1350,28 @@ int kern_addr_valid(unsigned long addr) /* Amount of ram needed to start using large blocks */ #define MEM_SIZE_FOR_LARGE_BLOCK (64UL << 30) +/* Adjustable memory block size */ +static unsigned long set_memory_block_size; +int __init set_memory_block_size_order(unsigned int order) +{ + unsigned long size = 1UL << order; + + if (size > MEM_SIZE_FOR_LARGE_BLOCK || size < MIN_MEMORY_BLOCK_SIZE) + return -EINVAL; + + set_memory_block_size = size; + return 0; +} + static unsigned long probe_memory_block_size(void) { unsigned long boot_mem_end = max_pfn << PAGE_SHIFT; unsigned long bz; - /* If this is UV system, always set 2G block size */ - if (is_uv_system()) { - bz = MAX_BLOCK_SIZE; + /* If memory block size has been set, then use it */ + bz = set_memory_block_size; + if (bz) goto done; - } /* Use regular block if RAM is smaller than MEM_SIZE_FOR_LARGE_BLOCK */ if (boot_mem_end < MEM_SIZE_FOR_LARGE_BLOCK) { diff --git a/include/linux/memory.h b/include/linux/memory.h index 31ca3e28b0eb..a6ddefc60517 100644 --- a/include/linux/memory.h +++ b/include/linux/memory.h @@ -38,6 +38,7 @@ struct memory_block { int arch_get_memory_phys_device(unsigned long start_pfn); unsigned long memory_block_size_bytes(void); +int set_memory_block_size_order(unsigned int order); /* These states are exposed to userspace as text strings in sysfs */ #define MEM_ONLINE (1<<0) /* exposed to userspace */

6 years, 7 months

2
1
0 0

FAILED: patch "[PATCH] mm, page_alloc: do not break __GFP_THISNODE by zonelist reset" failed to apply to 4.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 7810e6781e0fcbca78b91cf65053f895bf59e85f Mon Sep 17 00:00:00 2001 From: Vlastimil Babka <vbabka(a)suse.cz> Date: Thu, 7 Jun 2018 17:09:29 -0700 Subject: [PATCH] mm, page_alloc: do not break __GFP_THISNODE by zonelist reset In __alloc_pages_slowpath() we reset zonelist and preferred_zoneref for allocations that can ignore memory policies. The zonelist is obtained from current CPU's node. This is a problem for __GFP_THISNODE allocations that want to allocate on a different node, e.g. because the allocating thread has been migrated to a different CPU. This has been observed to break SLAB in our 4.4-based kernel, because there it relies on __GFP_THISNODE working as intended. If a slab page is put on wrong node's list, then further list manipulations may corrupt the list because page_to_nid() is used to determine which node's list_lock should be locked and thus we may take a wrong lock and race. Current SLAB implementation seems to be immune by luck thanks to commit 511e3a058812 ("mm/slab: make cache_grow() handle the page allocated on arbitrary node") but there may be others assuming that __GFP_THISNODE works as promised. We can fix it by simply removing the zonelist reset completely. There is actually no reason to reset it, because memory policies and cpusets don't affect the zonelist choice in the first place. This was different when commit 183f6371aac2 ("mm: ignore mempolicies when using ALLOC_NO_WATERMARK") introduced the code, as mempolicies provided their own restricted zonelists. We might consider this for 4.17 although I don't know if there's anything currently broken. SLAB is currently not affected, but in kernels older than 4.7 that don't yet have 511e3a058812 ("mm/slab: make cache_grow() handle the page allocated on arbitrary node") it is. That's at least 4.4 LTS. Older ones I'll have to check. So stable backports should be more important, but will have to be reviewed carefully, as the code went through many changes. BTW I think that also the ac->preferred_zoneref reset is currently useless if we don't also reset ac->nodemask from a mempolicy to NULL first (which we probably should for the OOM victims etc?), but I would leave that for a separate patch. Link: http://lkml.kernel.org/r/20180525130853.13915-1-vbabka@suse.cz Signed-off-by: Vlastimil Babka <vbabka(a)suse.cz> Fixes: 183f6371aac2 ("mm: ignore mempolicies when using ALLOC_NO_WATERMARK") Acked-by: Mel Gorman <mgorman(a)techsingularity.net> Cc: Michal Hocko <mhocko(a)kernel.org> Cc: David Rientjes <rientjes(a)google.com> Cc: Joonsoo Kim <iamjoonsoo.kim(a)lge.com> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> diff --git a/mm/page_alloc.c b/mm/page_alloc.c index ef1811531999..07b3c23762ad 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4169,7 +4169,6 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, * orientated. */ if (!(alloc_flags & ALLOC_CPUSET) || reserve_flags) { - ac->zonelist = node_zonelist(numa_node_id(), gfp_mask); ac->preferred_zoneref = first_zones_zonelist(ac->zonelist, ac->high_zoneidx, ac->nodemask); }

6 years, 7 months

2
1
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror June 2018