- 08 Feb, 2016 40 commits
-
-
Gustavo Padovan authored
sync_file has a more close meaning to what a sync_fence really, a struct that represent a file that can be used by userspace to get information on a fence, or wait for it to be signaled. Signed-off-by:
Gustavo Padovan <gustavo.padovan@collabora.co.uk> Reviewed-by:
Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Gustavo Padovan authored
This remove CONFIG_SW_SYNC_USER and instead compile the sw_sync file into debugpfs under <debugfs>/sync/sw_sync. Signed-off-by:
-
Gustavo Padovan authored
Creates the 'sync' dir on debugfs root dir and move the 'sync' file to sync/info. This is the preparation to add more debug info and control. Signed-off-by:
-
Gustavo Padovan authored
.dup and .compare are not used by the sync framework, so remove them from sw_sync. Signed-off-by:
-
Gustavo Padovan authored
These interfaces are not used nor have plans to be used in the near future so remove them for a cleaner solution before de-staging the sync framework. Signed-off-by:
-
Gustavo Padovan authored
Updates comments about functions and structures. Signed-off-by:
-
Chen Feng authored
Sanity check at binder ioctl function, Only allow the shared mm_struct to use the same binder-object to do binder operate. And add proc->vma_vm_mm = current->mm at the open function. The libbinder do ioctl before mmap called. V2: Fix compile error for error commit V3: Change the condition to proc->vma_vm_mm Signed-off-by:
Chen Feng <puck.chen@hisilicon.com> Signed-off-by:
Wei Dong <weidong2@hisilicon.com> Signed-off-by:
Junmin Zhao <zhaojunmin@huawei.com> Reviewed-by:
Zhuangluan Su <suzhuangluan@hisilicon.com> Signed-off-by:
-
Liviu Dudau authored
ion_buffer_create() will allocate a buffer and then create a DMA mapping for it, but it forgot to set the length of the page entries. Signed-off-by:
Liviu Dudau <Liviu.Dudau@arm.com> Signed-off-by:
Jon Medhurst <tixy@linaro.org> Acked-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
-
Chen Feng authored
Since ion alloc can be called by userspace,eg gralloc. When it is called frequently, the efficiency of kswapd is to low. And the reclaimed memory is too lower. In this way, the kswapd can use to much cpu resources. With 3.5GB DMA Zone and 0.5 Normal Zone. pgsteal_kswapd_dma 9364140 pgsteal_kswapd_normal 7071043 pgscan_kswapd_dma 10428250 pgscan_kswapd_normal 37840094 With this change the reclaim ratio has greatly improved 18.9% -> 72.5% Signed-off-by:
Lu bing <albert.lubing@hisilicon.com> Reviewed-by:
-
Bopamo Osaisai authored
This is a patch to the sync_debug.c file that rectifies a brace warning that was found with the checkpatch.pl tool Signed-off-by:
Bopamo Osaisai <bopamo@gmail.com> Signed-off-by:
-
Rajmal Menariya authored
In carveout heap, change minimum allocation order from 12 to PAGE_SHIFT. After this change each bit in bitmap (genalloc - General purpose special memory pool) represents one page size memory. Cc: sprd-ind-kernel-group@googlegroups.com Cc: sanjeev.yadav@spreadtrum.com Cc: Colin Cross <ccross@android.com> Cc: Android Kernel Team <kernel-team@android.com> Cc: Greg KH <gregkh@linuxfoundation.org> Cc: Sumit Semwal <sumit.semwal@linaro.org> Signed-off-by:
Rajmal Menariya <rajmal.menariya@spreadtrum.com> [jstultz: Reworked commit message] Signed-off-by:
John Stultz <john.stultz@linaro.org> Acked-by:
-
Colin Cross authored
lowmemorykiller debug messages are inscrutable and mostly useful for debugging the lowmemorykiller, not explaining why a process was killed. Make the messages more useful by prefixing them with "lowmemorykiller: " and explaining in more readable terms what was killed, who it was killed for, and why it was killed. The messages now look like: [ 76.997631] lowmemorykiller: Killing 'droid.gallery3d' (2172), adj 1000, [ 76.997635] to free 27436kB on behalf of 'kswapd0' (29) because [ 76.997638] cache 122624kB is below limit 122880kB for oom_score_adj 1000 [ 76.997641] Free memory is -53356kB above reserved A negative number for free memory above reserved means some of the reserved memory has been used and is being regenerated by kswapd, which is likely what called the shrinkers. Cc: Android Kernel Team <kernel-team@android.com> Cc: Greg KH <gregkh@linuxfoundation.org> Signed-off-by:
Colin Cross <ccross@android.com> [jstultz: Minor checkpatch tweaks] Signed-off-by:
-
Rom Lemarchand authored
Include <linux/types.h> into ashmem.h to ensure referenced types are defined Cc: Android Kernel Team <kernel-team@android.com> Cc: Greg KH <gregkh@linuxfoundation.org> Signed-off-by:
Rom Lemarchand <romlem@android.com> [jstultz: Minor commit message tweaks] Signed-off-by:
-
Laura Abbott authored
Both ashmem_mmap and ashmem_shrink take the ashmem_lock. It may be possible for ashmem_mmap to invoke ashmem_shrink: -000|mutex_lock(lock = 0x0) -001|ashmem_shrink(?, sc = 0x0) <--- try to take ashmem_mutex again -002|shrink_slab(shrink = 0xDA5F1CC0, nr_pages_scanned = 0, lru_pages -002|= -002|124) -003|try_to_free_pages(zonelist = 0x0, ?, ?, ?) -004|__alloc_pages_nodemask(gfp_mask = 21200, order = 1, zonelist = -004|0xC11D0940, -005|new_slab(s = 0xE4841E80, ?, node = -1) -006|__slab_alloc.isra.43.constprop.50(s = 0xE4841E80, gfpflags = -006|2148925462, ad -007|kmem_cache_alloc(s = 0xE4841E80, gfpflags = 208) -008|shmem_alloc_inode(?) -009|alloc_inode(sb = 0xE480E800) -010|new_inode_pseudo(?) -011|new_inode(?) -012|shmem_get_inode(sb = 0xE480E800, dir = 0x0, ?, dev = 0, flags = -012|187) -013|shmem_file_setup(?, ?, flags = 187) -014|ashmem_mmap(?, vma = 0xC5D64210) <---- Acquire ashmem_mutex -015|mmap_region(file = 0xDF8E2C00, addr = 1772974080, len = 233472, -015|flags = 57, -016|sys_mmap_pgoff(addr = 0, len = 230400, prot = 3, flags = 1, fd = -016|157, pgoff -017|ret_fast_syscall(asm) -->|exception -018|NUR:0x40097508(asm) ---|end of frame Avoid this deadlock by using mutex_trylock in ashmem_shrink; if the mutex is already held, do not attempt to shrink. Cc: Greg KH <gregkh@linuxfoundation.org> Cc: Android Kernel Team <kernel-team@android.com> Reported-by:
Matt Wagantall <mattw@codeaurora.org> Reported-by:
Syed Rameez Mustafa <rameezmustafa@codeaurora.org> Reported-by:
Osvaldo Banuelos <osvaldob@codeaurora.org> Reported-by:
Subbaraman Narayanamurthy <subbaram@codeaurora.org> Signed-off-by:
Laura Abbott <lauraa@codeaurora.org> [jstultz: Minor commit message tweaks] Signed-off-by:
-
Hugo Camboulive authored
This removes a few Sparse warnings. Signed-off-by:
Hugo Camboulive <hugo.camboulive@gmail.com> Signed-off-by:
-
Hari Prasath Gujulan Elango authored
Replace multiple if..else if..statements with simple table lookup in two functions. Signed-off-by:
Hari Prasath Gujulan Elango <hgujulan@visteon.com> Signed-off-by:
-
Sudip Mukherjee authored
device_create() returns ERR_PTR on error, it does not return NULL. Signed-off-by:
Sudip Mukherjee <sudip@vectorindia.org> Signed-off-by:
-
Christian Gromm authored
This patch removes the second forwared declaration of struct most_aim. Signed-off-by:
Christian Gromm <christian.gromm@microchip.com> Signed-off-by:
-
Christian Gromm authored
This patch adds a counter for dropped packets. It needed for statistical analysis. Signed-off-by:
-
Christian Gromm authored
This patch straightens and rearranges the code of function aim_write() of module aim-cdev. Signed-off-by:
-
Christian Gromm authored
This patch makes use of kfifo_peek and kfifo_skip, which renders the variable stacked_mbo useless. It is therefore removed. Signed-off-by:
-
Christian Gromm authored
This patch changes the type of the access reference from atomit_t to int. It is needed, because the reference variable is secured by synchronization locks and does not need to be atomic anymore. Signed-off-by:
-
Christian Gromm authored
This patch fixes race conditions that might emerge from functions aim_open, aim_close, aim_read, aim_write and aim_disconnect_channel within module cdev. Signed-off-by:
-
Christian Gromm authored
This patch renames the variable 'channel' to 'c'. This is needed to have the code look more homogeneous and to prevent format violations. Signed-off-by:
-
Christian Gromm authored
This patch fixes the function channel_has_mbo that delivers the false information in case two AIMs are using the same tx channel. Signed-off-by:
-
Christian Gromm authored
This patch encapsulates shared code. It therefore creates the new functions stop_channel and destroy_cdev. Signed-off-by:
-
Christian Gromm authored
This patch prevents the cdev module from rousing the channel wait queue in case the channel is about to be closed. It is safe to do so, because the application can not be waiting within read or write and at the same time be calling close. Signed-off-by:
-
Christian Gromm authored
This patch removes the mutexes stop_task_mutex and deregister mutex, since they can safely be left out. Signed-off-by:
-
Christian Gromm authored
This patch removes unnecessary code to destroy channel objects. It is needed, because function most_stop_channel, which is indirectly triggered by function most_deregister_interface, already destroys the channels. Signed-off-by:
-
Christian Gromm authored
This patch removes the unnecessary reference conter mod_ref. Signed-off-by:
-
Christian Gromm authored
This patch removes the atomic tainted flag. It is needed to get rid of logical overhead. Signed-off-by:
-
Christian Gromm authored
This patch moves the code that disconnects linked channels. It is needed to have cleaning things up done right. Signed-off-by:
-
Christian Gromm authored
This patch removes mutex from code that doesn't need any locking. Signed-off-by:
-
Christian Gromm authored
This patch makes function store_add_link initialize the pointer to an AIM right before the channel is probed. It is needed, the AIM may already call most_start_channel while probe_channel is still running. At this point the pointer to the AIM must not be NULL. Signed-off-by:
-
Christian Gromm authored
This patch invokes AIM's disconnect_channel callback before the corresponding pointers are re-initialized to NULL. Signed-off-by:
-
Christian Gromm authored
This patch adds two missing calls to function ida_simpel_remove. Signed-off-by:
-
Christian Gromm authored
This patch removes the function destroy_most_c_obj and executes its code within function destroy_most_inst_obj. Signed-off-by:
-
Christian Gromm authored
This patch makes use of functions readl and writel instead of the __raw_* variants. Signed-off-by:
-
Christian Gromm authored
This patch increments mbo_offs by the number of bytes that have been copied and resets it in case a complete mbo has been transferred to user buffer. Signed-off-by:
-
Christian Gromm authored
This patch fixes a potential MBO leak in case function aim_read() exits right after the MBO has been fetched from kfifo and before it has been saved to the variable stacked_mbo. Signed-off-by:
-
