dri-devel August 2012

dri-devel@lists.freedesktop.org

145 participants
811 discussions

[PATCH 0/2] dma-parms and helpers for dma-buf
by Rob Clark 06 Aug '12

06 Aug '12

From: Rob Clark <rob(a)ti.com> Re-sending first patch, with a wider audience. Apparently I didn't spam enough inboxes the first time. And, at Daniel Vetter's suggestion, adding some helper functions in dma-buf to get the most restrictive parameters of all the attached devices. Rob Clark (2): device: add dma_params->max_segment_count dma-buf: add helpers for attacher dma-parms drivers/base/dma-buf.c | 63 +++++++++++++++++++++++++++++++++++++++++++ include/linux/device.h | 1 + include/linux/dma-buf.h | 19 +++++++++++++ include/linux/dma-mapping.h | 16 +++++++++++ 4 files changed, 99 insertions(+) -- 1.7.9.5

5 8

[PATCH] vmwgfx: add missing mutex_unlocks
by Devendra Naga 06 Aug '12

06 Aug '12

we have done a proper mutex_unlock in the error cases of the vmw_fifo_reserve, but there are missing mutex unlocks where we return a valid pointer in vmw_fifo_reserve. Signed-off-by: Devendra Naga <develkernel412222(a)gmail.com> --- drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c index a0c2f12..e3abd7a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c @@ -355,6 +355,7 @@ void *vmw_fifo_reserve(struct vmw_private *dev_priv, uint32_t bytes) if (reserveable) iowrite32(bytes, fifo_mem + SVGA_FIFO_RESERVED); + mutex_unlock(&fifo_state->fifo_mutex); return fifo_mem + (next_cmd >> 2); } else { need_bounce = true; @@ -363,10 +364,12 @@ void *vmw_fifo_reserve(struct vmw_private *dev_priv, uint32_t bytes) if (need_bounce) { fifo_state->using_bounce_buffer = true; - if (bytes < fifo_state->static_buffer_size) + if (bytes < fifo_state->static_buffer_size) { + mutex_unlock(&fifo_state->fifo_mutex); return fifo_state->static_buffer; - else { + } else { fifo_state->dynamic_buffer = vmalloc(bytes); + mutex_unlock(&fifo_state->fifo_mutex); return fifo_state->dynamic_buffer; } } -- 1.7.9.5

2 2

[RFC PATCH 1/3] dma-fence: dma-buf synchronization (v5)
by Maarten Lankhorst 06 Aug '12

06 Aug '12

A dma-fence can be attached to a buffer which is being filled or consumed by hw, to allow userspace to pass the buffer without waiting to another device. For example, userspace can call page_flip ioctl to display the next frame of graphics after kicking the GPU but while the GPU is still rendering. The display device sharing the buffer with the GPU would attach a callback to get notified when the GPU's rendering-complete IRQ fires, to update the scan-out address of the display, without having to wake up userspace. A dma-fence is transient, one-shot deal. It is allocated and attached to one or more dma-buf's. When the one that attached it is done, with the pending operation, it can signal the fence. + dma_fence_signal() The dma-buf-mgr handles tracking, and waiting on, the fences associated with a dma-buf. TODO maybe need some helper fxn for simple devices, like a display- only drm/kms device which simply wants to wait for exclusive fence to be signaled, and then attach a non-exclusive fence while scanout is in progress. The one pending on the fence can add an async callback (and optionally cancel it.. for example, to recover from GPU hangs): + dma_fence_add_callback() + dma_fence_cancel_callback() Or wait synchronously (optionally with timeout or from atomic context): + dma_fence_wait() A default software-only implementation is provided, which can be used by drivers attaching a fence to a buffer when they have no other means for hw sync. But a memory backed fence is also envisioned, because it is common that GPU's can write to, or poll on some memory location for synchronization. For example: fence = dma_buf_get_fence(dmabuf); if (fence->ops == &bikeshed_fence_ops) { dma_buf *fence_buf; dma_bikeshed_fence_get_buf(fence, &fence_buf, &offset); ... tell the hw the memory location to wait on ... } else { /* fall-back to sw sync * / dma_fence_add_callback(fence, my_cb); } On SoC platforms, if some other hw mechanism is provided for synchronizing between IP blocks, it could be supported as an alternate implementation with it's own fence ops in a similar way. To facilitate other non-sw implementations, the enable_signaling callback can be used to keep track if a device not supporting hw sync is waiting on the fence, and in this case should arrange to call dma_fence_signal() at some point after the condition has changed, to notify other devices waiting on the fence. If there are no sw waiters, this can be skipped to avoid waking the CPU unnecessarily. The intention is to provide a userspace interface (presumably via eventfd) later, to be used in conjunction with dma-buf's mmap support for sw access to buffers (or for userspace apps that would prefer to do their own synchronization). v1: Original v2: After discussion w/ danvet and mlankhorst on #dri-devel, we decided that dma-fence didn't need to care about the sw->hw signaling path (it can be handled same as sw->sw case), and therefore the fence->ops can be simplified and more handled in the core. So remove the signal, add_callback, cancel_callback, and wait ops, and replace with a simple enable_signaling() op which can be used to inform a fence supporting hw->hw signaling that one or more devices which do not support hw signaling are waiting (and therefore it should enable an irq or do whatever is necessary in order that the CPU is notified when the fence is passed). v3: Fix locking fail in attach_fence() and get_fence() v4: Remove tie-in w/ dma-buf.. after discussion w/ danvet and mlankorst we decided that we need to be able to attach one fence to N dma-buf's, so using the list_head in dma-fence struct would be problematic. v5: [ Maarten Lankhorst ] Updated for dma-bikeshed-fence and dma-buf-manager. --- drivers/base/Makefile | 2 drivers/base/dma-fence.c | 317 +++++++++++++++++++++++++++++++++++++++++++++ include/linux/dma-fence.h | 123 +++++++++++++++++ 3 files changed, 441 insertions(+), 1 deletion(-) create mode 100644 drivers/base/dma-fence.c create mode 100644 include/linux/dma-fence.h diff --git a/drivers/base/Makefile b/drivers/base/Makefile index 5aa2d70..6e9f217 100644 --- a/drivers/base/Makefile +++ b/drivers/base/Makefile @@ -10,7 +10,7 @@ obj-$(CONFIG_CMA) += dma-contiguous.o obj-y += power/ obj-$(CONFIG_HAS_DMA) += dma-mapping.o obj-$(CONFIG_HAVE_GENERIC_DMA_COHERENT) += dma-coherent.o -obj-$(CONFIG_DMA_SHARED_BUFFER) += dma-buf.o +obj-$(CONFIG_DMA_SHARED_BUFFER) += dma-buf.o dma-fence.o obj-$(CONFIG_ISA) += isa.o obj-$(CONFIG_FW_LOADER) += firmware_class.o obj-$(CONFIG_NUMA) += node.o diff --git a/drivers/base/dma-fence.c b/drivers/base/dma-fence.c new file mode 100644 index 0000000..6798dc4 --- /dev/null +++ b/drivers/base/dma-fence.c @@ -0,0 +1,317 @@ +/* + * Fence mechanism for dma-buf to allow for asynchronous dma access + * + * Copyright (C) 2012 Texas Instruments + * Author: Rob Clark <rob.clark(a)linaro.org> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include <linux/slab.h> +#include <linux/sched.h> +#include <linux/export.h> +#include <linux/dma-fence.h> + +/** + * dma_fence_signal - Signal a fence. + * + * @fence: The fence to signal + * + * All registered callbacks will be called directly (synchronously) and + * all blocked waters will be awoken. + * + * TODO: any value in adding a dma_fence_cancel(), for example to recov + * from hung gpu? It would behave like dma_fence_signal() but return + * an error to waiters and cb's to let them know that the condition they + * are waiting for will never happen. + */ +int dma_fence_signal(struct dma_fence *fence) +{ + unsigned long flags; + int ret = -EINVAL; + + if (WARN_ON(!fence)) + return -EINVAL; + + spin_lock_irqsave(&fence->event_queue.lock, flags); + if (!fence->signaled) { + fence->signaled = true; + wake_up_all_locked(&fence->event_queue); + ret = 0; + } else WARN(1, "Already signaled"); + spin_unlock_irqrestore(&fence->event_queue.lock, flags); + + return ret; +} +EXPORT_SYMBOL_GPL(dma_fence_signal); + +static void release_fence(struct kref *kref) +{ + struct dma_fence *fence = + container_of(kref, struct dma_fence, refcount); + + WARN_ON(waitqueue_active(&fence->event_queue)); + if (fence->ops->release) + fence->ops->release(fence); + + kfree(fence); +} + +/** + * dma_fence_put - Release a reference to the fence. + */ +void dma_fence_put(struct dma_fence *fence) +{ + WARN_ON(!fence); + kref_put(&fence->refcount, release_fence); +} +EXPORT_SYMBOL_GPL(dma_fence_put); + +/** + * dma_fence_get - Take a reference to the fence. + * + * In most cases this is used only internally by dma-fence. + */ +void dma_fence_get(struct dma_fence *fence) +{ + WARN_ON(!fence); + kref_get(&fence->refcount); +} +EXPORT_SYMBOL_GPL(dma_fence_get); + +static int check_signaling(struct dma_fence *fence) +{ + bool enable_signaling = false, signaled; + unsigned long flags; + + spin_lock_irqsave(&fence->event_queue.lock, flags); + if (!fence->needs_sw_signal) + enable_signaling = fence->needs_sw_signal = true; + signaled = fence->signaled; + spin_unlock_irqrestore(&fence->event_queue.lock, flags); + + if (enable_signaling) { + int ret; + + /* At this point, if enable_signaling returns any error + * a wakeup has to be performanced regardless. + * -ENOENT signals fence was already signaled. Any other error + * inidicates a catastrophic hardware error. + * + * If any hardware error occurs, nothing can be done against + * it, so it's treated like the fence was already signaled. + * No synchronization can be performed, so we have to assume + * the fence was already signaled. + */ + ret = fence->ops->enable_signaling(fence); + if (ret) { + signaled = true; + dma_fence_signal(fence); + } + } + + if (!signaled) + return 0; + else + return -ENOENT; +} + +/** + * dma_fence_add_callback - Add a callback to be called when the fence + * is signaled. + * + * @fence: The fence to wait on + * @cb: The callback to register + * + * Any number of callbacks can be registered to a fence, but a callback + * can only be registered to once fence at a time. + * + * Note that the callback can be called from an atomic context. If + * fence is already signaled, this function will return -ENOENT (and + * *not* call the callback) + */ +int dma_fence_add_callback(struct dma_fence *fence, + struct dma_fence_cb *cb) +{ + unsigned long flags; + int ret; + + if (WARN_ON(!fence || !cb)) + return -EINVAL; + + ret = check_signaling(fence); + + spin_lock_irqsave(&fence->event_queue.lock, flags); + if (!ret && !fence->signaled) { + cb->fence = fence; + __add_wait_queue(&fence->event_queue, &cb->base); + ret = 0; + } else if (!ret) + ret = -ENOENT; + spin_unlock_irqrestore(&fence->event_queue.lock, flags); + + return ret; +} +EXPORT_SYMBOL_GPL(dma_fence_add_callback); + +/** + * dma_fence_cancel_callback - Remove a previously registered callback. + * + * @cb: The callback to unregister + * + * The callback will not be called after this function returns, but could + * be called before this function returns. + */ +int dma_fence_cancel_callback(struct dma_fence_cb *cb) +{ + struct dma_fence *fence; + unsigned long flags; + int ret = -EINVAL; + + if (WARN_ON(!cb)) + return -EINVAL; + + fence = cb->fence; + + spin_lock_irqsave(&fence->event_queue.lock, flags); + if (fence) { + __remove_wait_queue(&fence->event_queue, &cb->base); + cb->fence = NULL; + ret = 0; + } + spin_unlock_irqrestore(&fence->event_queue.lock, flags); + + return ret; +} +EXPORT_SYMBOL_GPL(dma_fence_cancel_callback); + +/** + * dma_fence_wait - Wait for a fence to be signaled. + * + * @fence: The fence to wait on + * @interruptible: if true, do an interruptible wait + * @timeout: absolute time for timeout, in jiffies. + * + * Returns 0 on success, -EBUSY if a timeout occured, + * -ERESTARTSYS if the wait was interrupted by a signal. + */ +int dma_fence_wait(struct dma_fence *fence, bool interruptible, unsigned long timeout) +{ + unsigned long cur; + int ret; + + if (WARN_ON(!fence)) + return -EINVAL; + + cur = jiffies; + if (time_after_eq(cur, timeout)) + return -EBUSY; + + timeout -= cur; + + ret = check_signaling(fence); + if (ret == -ENOENT) + return 0; + else if (ret) + return ret; + + if (interruptible) + ret = wait_event_interruptible_timeout(fence->event_queue, + fence->signaled, + timeout); + else + ret = wait_event_timeout(fence->event_queue, + fence->signaled, timeout); + + WARN(1, "wait_event_timeout(%u) returns %i", interruptible, ret); + if (ret > 0) + return 0; + else if (!ret) + return -EBUSY; + else + return ret; +} +EXPORT_SYMBOL_GPL(dma_fence_wait); + +int __dma_fence_wake_func(wait_queue_t *wait, unsigned mode, + int flags, void *key) +{ + struct dma_fence_cb *cb = + container_of(wait, struct dma_fence_cb, base); + struct dma_fence *fence = cb->fence; + int ret; + + ret = cb->func(cb, fence); + cb->fence = NULL; + + return ret; +} +EXPORT_SYMBOL_GPL(__dma_fence_wake_func); + +/* + * Helpers intended to be used by the ops of the dma_fence implementation: + * + * NOTE: helpers and fxns intended to be used by other dma-fence + * implementations are not exported.. I'm not really sure if it makes + * sense to have a dma-fence implementation that is itself a module. + */ + +void __dma_fence_init(struct dma_fence *fence, struct dma_fence_ops *ops, void *priv) +{ + WARN_ON(!ops || !ops->enable_signaling); + + kref_init(&fence->refcount); + fence->ops = ops; + fence->priv = priv; + init_waitqueue_head(&fence->event_queue); +} +EXPORT_SYMBOL_GPL(__dma_fence_init); + +/* + * Pure sw implementation for dma-fence. The CPU always gets involved. + */ + +static int sw_enable_signaling(struct dma_fence *fence) +{ + /* + * pure sw, no irq's to enable, because the fence creator will + * always call dma_fence_signal() + */ + return 0; +} + +static struct dma_fence_ops sw_fence_ops = { + .enable_signaling = sw_enable_signaling, +}; + +/** + * dma_fence_create - Create a simple sw-only fence. + * + * This fence only supports signaling from/to CPU. Other implementations + * of dma-fence can be used to support hardware to hardware signaling, if + * supported by the hardware, and use the dma_fence_helper_* functions for + * compatibility with other devices that only support sw signaling. + */ +struct dma_fence *dma_fence_create(void) +{ + struct dma_fence *fence; + + fence = kzalloc(sizeof(struct dma_fence), GFP_KERNEL); + if (!fence) + return ERR_PTR(-ENOMEM); + + __dma_fence_init(fence, &sw_fence_ops, 0); + + return fence; +} +EXPORT_SYMBOL_GPL(dma_fence_create); diff --git a/include/linux/dma-fence.h b/include/linux/dma-fence.h new file mode 100644 index 0000000..648f136 --- /dev/null +++ b/include/linux/dma-fence.h @@ -0,0 +1,123 @@ +/* + * Fence mechanism for dma-buf to allow for asynchronous dma access + * + * Copyright (C) 2012 Texas Instruments + * Author: Rob Clark <rob.clark(a)linaro.org> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef __DMA_FENCE_H__ +#define __DMA_FENCE_H__ + +#include <linux/err.h> +#include <linux/list.h> +#include <linux/wait.h> +#include <linux/list.h> +#include <linux/dma-buf.h> + +struct dma_fence; +struct dma_fence_ops; +struct dma_fence_cb; + +struct dma_fence { + struct kref refcount; + struct dma_fence_ops *ops; + wait_queue_head_t event_queue; + void *priv; + + /* has this fence been signaled yet? */ + bool signaled : 1; + + /* do we have one or more waiters or callbacks? */ + bool needs_sw_signal : 1; +}; + +typedef int (*dma_fence_func_t)(struct dma_fence_cb *cb, + struct dma_fence *fence); + +struct dma_fence_cb { + wait_queue_t base; + dma_fence_func_t func; + + /* + * This is initialized when the cb is added, and NULL'd when it + * is canceled or expired, so can be used to for error checking + * if the cb is already pending. A dma_fence_cb can be pending + * on at most one fence at a time. + */ + struct dma_fence *fence; +}; + +struct dma_fence_ops { + /** + * For fence implementations that have the capability for hw->hw + * signaling, they can implement this op to enable the necessary + * irqs, or insert commands into cmdstream, etc. This is called + * in the first wait() or add_callback() path to let the fence + * implementation know that there is another driver waiting on + * the signal (ie. hw->sw case). + * + * A return value of -ENOENT will indicate that the fence has + * already passed. + */ + int (*enable_signaling)(struct dma_fence *fence); + void (*release)(struct dma_fence *fence); +}; + +int __dma_fence_wake_func(wait_queue_t *wait, unsigned mode, + int flags, void *key); + +#define DMA_FENCE_CB_INITIALIZER(cb_func) { \ + .base = { .func = __dma_fence_wake_func }, \ + .func = (cb_func), \ + } + +#define DECLARE_DMA_FENCE_CB(name, cb_func) \ + struct dma_fence_cb name = DMA_FENCE_CB_INITIALIZER(cb_func) + + +/* + * TODO does it make sense to be able to enable dma-fence without dma-buf, + * or visa versa? + */ +#ifdef CONFIG_DMA_SHARED_BUFFER + +/* create a basic (pure sw) fence: */ +struct dma_fence *dma_fence_create(void); + +/* intended to be used by other dma_fence implementations: */ +void __dma_fence_init(struct dma_fence *fence, struct dma_fence_ops *ops, void *priv); + +void dma_fence_get(struct dma_fence *fence); +void dma_fence_put(struct dma_fence *fence); +int dma_fence_signal(struct dma_fence *fence); + +int dma_fence_add_callback(struct dma_fence *fence, + struct dma_fence_cb *cb); +int dma_fence_cancel_callback(struct dma_fence_cb *cb); +int dma_fence_wait(struct dma_fence *fence, bool interruptible, unsigned long timeout); + +/* helpers intended to be used by the ops of the dma_fence implementation: */ +int dma_fence_helper_signal(struct dma_fence *fence); +int dma_fence_helper_add_callback(struct dma_fence *fence, + struct dma_fence_cb *cb); +int dma_fence_helper_cancel_callback(struct dma_fence_cb *cb); +int dma_fence_helper_wait(struct dma_fence *fence, bool interruptible, + long timeout); + +#else +// TODO +#endif /* CONFIG_DMA_SHARED_BUFFER */ + +#endif /* __DMA_FENCE_H__ */

2 4

[PATCH] gpu/mfd/usb: Fix USB randconfig problems
by Guenter Roeck 06 Aug '12

06 Aug '12

Fix config warning: warning: ( ... && DRM_USB) selects USB which has unmet direct dependencies (USB_SUPPORT && USB_ARCH_HAS_HCD) by adding the missing dependency on USB_ARCH_HAS_HCD to DRM_UDL and DRM_USB. This exposes: drivers/video/Kconfig:36:error: recursive dependency detected! drivers/video/Kconfig:36: symbol FB is selected by DRM_KMS_HELPER drivers/gpu/drm/Kconfig:28: symbol DRM_KMS_HELPER is selected by DRM_UDL drivers/gpu/drm/udl/Kconfig:1: symbol DRM_UDL depends on USB_ARCH_HAS_HCD drivers/usb/Kconfig:78: symbol USB_ARCH_HAS_HCD depends on USB_ARCH_HAS_OHCI drivers/usb/Kconfig:16: symbol USB_ARCH_HAS_OHCI depends on I2C drivers/i2c/Kconfig:5: symbol I2C is selected by FB_DDC drivers/video/Kconfig:86: symbol FB_DDC is selected by FB_CYBER2000_DDC drivers/video/Kconfig:385: symbol FB_CYBER2000_DDC depends on FB_CYBER2000 drivers/video/Kconfig:373: symbol FB_CYBER2000 depends on FB which is due to drivers/usb/Kconfig: config USB_ARCH_HAS_OHCI ... default y if ARCH_PNX4008 && I2C Fix by dropping I2C from the above dependency; logic is that this is not a platform dependency but a configuration dependency: the _architecture_ still supports USB even is I2C is not selected. This exposes: drivers/video/Kconfig:36:error: recursive dependency detected! drivers/video/Kconfig:36: symbol FB is selected by DRM_KMS_HELPER drivers/gpu/drm/Kconfig:28: symbol DRM_KMS_HELPER is selected by DRM_UDL drivers/gpu/drm/udl/Kconfig:1: symbol DRM_UDL depends on USB_ARCH_HAS_HCD drivers/usb/Kconfig:78: symbol USB_ARCH_HAS_HCD depends on USB_ARCH_HAS_OHCI drivers/usb/Kconfig:17: symbol USB_ARCH_HAS_OHCI depends on MFD_TC6393XB drivers/mfd/Kconfig:396: symbol MFD_TC6393XB depends on GPIOLIB drivers/gpio/Kconfig:35: symbol GPIOLIB is selected by FB_VIA drivers/video/Kconfig:1560: symbol FB_VIA depends on FB which can be fixed by having MFD_TC6393XB select GPIOLIB instead of depending on it. Signed-off-by: Guenter Roeck <linux(a)roeck-us.net> --- If anyone has a better idea how to fix this set of problems, please let me know. Also, I'll be happy to split the patch into three parts if that helps to get it integrated. drivers/gpu/drm/Kconfig | 1 + drivers/gpu/drm/udl/Kconfig | 1 + drivers/mfd/Kconfig | 3 ++- drivers/usb/Kconfig | 2 +- 4 files changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index 23120c0..90e2808 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -22,6 +22,7 @@ menuconfig DRM config DRM_USB tristate depends on DRM + depends on USB_ARCH_HAS_HCD select USB config DRM_KMS_HELPER diff --git a/drivers/gpu/drm/udl/Kconfig b/drivers/gpu/drm/udl/Kconfig index 0b5e096..56e0bf3 100644 --- a/drivers/gpu/drm/udl/Kconfig +++ b/drivers/gpu/drm/udl/Kconfig @@ -1,6 +1,7 @@ config DRM_UDL tristate "DisplayLink" depends on DRM && EXPERIMENTAL + depends on USB_ARCH_HAS_HCD select DRM_USB select FB_SYS_FILLRECT select FB_SYS_COPYAREA diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig index d1facef..b1a1462 100644 --- a/drivers/mfd/Kconfig +++ b/drivers/mfd/Kconfig @@ -395,7 +395,8 @@ config MFD_TC6387XB config MFD_TC6393XB bool "Support Toshiba TC6393XB" - depends on GPIOLIB && ARM && HAVE_CLK + depends on ARM && HAVE_CLK + select GPIOLIB select MFD_CORE select MFD_TMIO help diff --git a/drivers/usb/Kconfig b/drivers/usb/Kconfig index a7773a3..7065df6 100644 --- a/drivers/usb/Kconfig +++ b/drivers/usb/Kconfig @@ -13,7 +13,7 @@ config USB_ARCH_HAS_OHCI default y if PXA3xx default y if ARCH_EP93XX default y if ARCH_AT91 - default y if ARCH_PNX4008 && I2C + default y if ARCH_PNX4008 default y if MFD_TC6393XB default y if ARCH_W90X900 default y if ARCH_DAVINCI_DA8XX -- 1.7.9.7

1 0

[Bug 36522] Caught 16-bit read from uninitialized memory in drm_fb_helper_setcmap
by bugzilla-daemon＠bugzilla.kernel.org 05 Aug '12

05 Aug '12

https://bugzilla.kernel.org/show_bug.cgi?id=36522 --- Comment #9 from Christian Casteyde <casteyde.christian(a)free.fr> 2012-08-05 21:28:16 --- Update: Still present in 3.6-rc1 -- Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching the assignee of the bug.

1 0

[PATCH] drm/i915: Fix mem leak in intel_sdvo_write_cmd()
by Jesper Juhl 05 Aug '12

05 Aug '12

If the allocation of 'buf' succeeds but the allocation of 'msgs' fails we'll return false and leak 'buf' when it goes out of scope. Signed-off-by: Jesper Juhl <jj(a)chaosbits.net> --- drivers/gpu/drm/i915/intel_sdvo.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) note: compile tested only due to lack of hardware. diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 26a6a4d..1f73e24 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -444,13 +444,12 @@ static bool intel_sdvo_write_cmd(struct intel_sdvo *intel_sdvo, u8 cmd, struct i2c_msg *msgs; int i, ret = true; - buf = (u8 *)kzalloc(args_len * 2 + 2, GFP_KERNEL); - if (!buf) - return false; - + buf = kzalloc(args_len * 2 + 2, GFP_KERNEL); msgs = kcalloc(args_len + 3, sizeof(*msgs), GFP_KERNEL); - if (!msgs) - return false; + if (!msgs || !buf) { + ret = false; + goto out; + } intel_sdvo_debug_write(intel_sdvo, cmd, args, args_len); -- 1.7.11.3 -- Jesper Juhl <jj(a)chaosbits.net> http://www.chaosbits.net/ Don't top-post http://www.catb.org/jargon/html/T/top-post.html Plain text mails only, please.

2 2

[PATCH] drm: ignore disconnected <-> unknown status changes
by Knut Petersen 05 Aug '12

05 Aug '12

On an AOpen i915GMm-hfs the hotplug events generated by transitions between connector_status_unknown and connector_status_disconnected cause screen distortions. The attached patch cures the problem by disabling the generation of hotplug events in those cases. That should be safe for everybody as the only relevant changes are those from / to connector_status_connected. cu, Knut

3 2

[PATCH] i915: don't map imported dma-bufs for dmar.
by Dave Airlie 05 Aug '12

05 Aug '12

From: Dave Airlie <airlied(a)redhat.com> The exporter should have given us pages in the correct place, avoid the prepare object mapping phase on dmar systems. This fixes an oops on a GM45/R600 machine, when running the intel/radeon tests. Signed-off-by: Dave Airlie <airlied(a)redhat.com> --- drivers/gpu/drm/i915/i915_gem_gtt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 9fd25a4..ee9b68f 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -361,7 +361,8 @@ int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj) struct drm_device *dev = obj->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - if (dev_priv->mm.gtt->needs_dmar) + /* don't map imported dma buf objects */ + if (dev_priv->mm.gtt->needs_dmar && !obj->sg_table) return intel_gtt_map_memory(obj->pages, obj->base.size >> PAGE_SHIFT, &obj->sg_list, -- 1.7.11.2

2 1

[PATCH] drm/i915: remove unused variable
by Devendra Naga 05 Aug '12

05 Aug '12

the following warning was produced, drivers/gpu/drm/i915/i915_gem_context.c: In function ‘i915_switch_context’: drivers/gpu/drm/i915/i915_gem_context.c:454:6: warning: unused variable ‘ret’ [-Wunused-variable] fix up by removing it Signed-off-by: Devendra Naga <devendra.aaru(a)gmail.com> --- This warning occurs in current v3.5 with HEAD at 173f8654746c. drivers/gpu/drm/i915/i915_gem_context.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index da8b01f..a9d58d7 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -451,7 +451,6 @@ int i915_switch_context(struct intel_ring_buffer *ring, struct drm_i915_file_private *file_priv = NULL; struct i915_hw_context *to; struct drm_i915_gem_object *from_obj = ring->last_context_obj; - int ret; if (dev_priv->hw_contexts_disabled) return 0; -- 1.7.9.5

2 1

[XDC 2012] Conference Update #1
by Egbert Eich 05 Aug '12

05 Aug '12

We are still 1 1/2 month into XDC 2012 so it's time to give a status update and beat the drums some more: - Registration Reminder: So far we have 24 registered participants: there's plenty of room for more - so if you haven't done so: register! If you are in need for travel sponsorship please consider contacting the X.Org Foundation Board of Directors: send email board(a)foundation.x.org. - Presentation Reminder: So far we don't have all too many proposals for presentations. Matt Dew has just announced a more formal CFP if you would like to submit a conference paper: http://lists.x.org/archives/xorg/2012-August/054967.html However you can still register as described on the conference page if you prefer a more informal process. I expect that Matt will send out a clarification on this issue. If at doubt feel free to contact me. - Hotel Reservation Reminder: We've got a number of rooms allocated for the conference which you will get for EUR 68/night for a single room (breakfast and free wifi included) at the Azimut Hotel (http://www.azimuthotels.com) Kaulbachstr 1 90408 Nuernberg tel.: +49 911 3657 0 FAX.: +49 911 3657 448 email: reservierung.nuernberg(a)azimuthotels.com This deal will expire on April 18th, so please make your reservation! There are still a number of rooms available - first come first serve. For the reservation you need to state the reservation number #142733. PLEASE NOTE: ============ The hotel has informed me that you should either place your reservation by phone or by email if you want to take advantage of this special deal. Please *don't* use the online reservation tool. - Beer Hiking thru the Franconian Countryside: We will do a beer hiking trip thru the Frankonian Countryside on Saturday after the conference: on this trip we will walk from one beer garden of a micro-brewery to the next. Please let me know if you would like to participate so that I get a rough estimate on the number of people who would like to join. - Contact Information Update: If you have any further inquiries please feel free to email me at egbert.eich _at_ gmail.com (Please note I've updated the email address at which you can reach me and replaced it by one that is not hampered by broken spam filters.) So if you have tried to contact me in the past and weren't able to, please resend! You may also contact me at freenode: my nick is 'egbert'. I'm always present - though not always physical. I will respond sometimes with a delay. So see you all in Nuernberg! Cheers, Egbert.

1 0

← Newer
1
...
75
76
77
78
79
80
81
82
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel August 2012