Some frameworks (e.g. Ion) may need to do explicit cache management
to meet performance/correctness requirements. Rather than piggy-back
on another API and hope the semantics don't change, introduce a
set of APIs to force a page to be cleaned/invalidated in the cache.
Signed-off-by: Laura Abbott
arm may need the kernel_force_cache APIs to guarantee data consistency.
Implement versions of these APIs based on the DMA APIs.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
arch/arm/include/asm/cacheflush.h | 4 ++
arch/arm/mm/dma-mapping.c
driver could potentially be converted
with the addition of an x86 implementation.
Feedback appreciated as always.
Thanks,
Laura
[1]
https://lkml.kernel.org/g/<1464205684-5587-1-git-send-email-labb...@redhat.com>
Laura Abbott (5):
Documentation: Introduce kernel_force_cache_* APIs
ion_carveout_allocate and ion_carveout_free aren't used outside of the
carveout heap. Get rid of the definitions.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion_carveout_heap.c | 2 ++
drivers/staging/android/ion/ion_priv.h | 14 ---
ion_reserve was supposed to be used to reserve memory in board files.
These days, board files are no more and there are other more controlled
mechanisms for reserving memory. Get rid of this function.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion.
ion_phys was an interface used for older legacy behavior. sg_tables
are the standard now. Get rid of it.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion.c | 26 -
drivers/staging/android/ion/ion.h
The map_dma API interface was designed to generate an sg_table.
Currently, every client just creates the table at allocation time and
then returns the one table. Nothing happens on unmap_dma either.
Just get rid of the API and assign the sg_table directly.
Signed-off-by: Laura Abbott <l
The ion_sg_table interface is mostly a reimplementation of
what dma_buf is doing. Clients should be using dma_buf APIs instead.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion.c | 20
drivers/staging/android/ion/ion.h | 11
On 07/30/2016 09:11 AM, Nadim almas wrote:
This patch compresses two lines in to a single line in file ion.c
if immediate return statement is found. It also removes variable
bytes_written as it is no longer needed.
This still says bytes_written.
None of the patches after your first one apply
ion_reserve was supposed to be used to reserve memory in board files.
These days, board files are no more and there are other more controlled
mechanisms for reserving memory. Get rid of this function.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion.
The map_dma API interface was designed to generate an sg_table.
Currently, every client just creates the table at allocation time and
then returns the one table. Nothing happens on unmap_dma either.
Just get rid of the API and assign the sg_table directly.
Signed-off-by: Laura Abbott <l
ion_phys was an interface used for older legacy behavior. sg_tables
are the standard now. Get rid of it.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion.c | 26 -
drivers/staging/android/ion/ion.h
ion_carveout_allocate and ion_carveout_free aren't used outside of the
carveout heap. Get rid of the definitions.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion_carveout_heap.c | 2 ++
drivers/staging/android/ion/ion_priv.h | 14 ---
The ion_sg_table interface is mostly a reimplementation of
what dma_buf is doing. Clients should be using dma_buf APIs instead.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion.c | 20
drivers/staging/android/ion/ion.h | 11
to be extended.
Thanks,
Laura
Laura Abbott (5):
staging: android: ion: Get rid of ion_sg_table
staging: android: ion: Drop ion_phys interface
staging: android: ion: Get rid of map_dma/unmap_dma
staging: android: ion: Drop ion_carveout_allocate definitions
staging: android: ion: Get rid
n comments not about the heap ID mapping
because I'm leaning towards dropping the heap ID mapping.
I've a few inline comments below.
-Brian
On Mon, Jun 06, 2016 at 11:23:33AM -0700, Laura Abbott wrote:
From: Laura Abbott <labb...@fedoraproject.org>
The Ion ABI for heaps is limiting to work with
On 06/08/2016 08:15 AM, Brian Starkey wrote:
Hi Laura,
On Mon, Jun 06, 2016 at 11:23:27AM -0700, Laura Abbott wrote:
The ABI for Ion's ioctl interface are a pain to work with. The heap IDs
are a 32-bit non-discoverable namespace that form part of the ABI. There's
no way to determine what ABI
On 06/08/2016 06:50 AM, Liviu Dudau wrote:
On Mon, Jun 06, 2016 at 11:23:33AM -0700, Laura Abbott wrote:
From: Laura Abbott <labb...@fedoraproject.org>
The Ion ABI for heaps is limiting to work with for more complex systems.
Heaps have to be registered at boot time with known ids ava
?
Thanks,
Laura
But beyond this, The new alloc2 with not-handle flag is good.
And the pull out of ioctl interface is also a good cleanup.
On 2016/6/7 2:23, Laura Abbott wrote:
The ABI for Ion's ioctl interface are a pain to work with. The heap IDs
are a 32-bit non-discoverable namespace that form
The current Ion ioctls lack a good way to tell what ioctls are
available. Introduce an ioctl to give an ABI version. This way when the
ABI inevitably gets screwed up userspace will have a way to tell what
version of the screw up is available.
Signed-off-by: Laura Abbott <labb...@redhat.
From: Laura Abbott <labb...@fedoraproject.org>
The Ion ABI for heaps is limiting to work with for more complex systems.
Heaps have to be registered at boot time with known ids available to
userspace. This becomes a tight ABI which is prone to breakage.
Introduce a new mechanism for regis
The number of Ion ioctls may continue to grow along with necessary
validation. Pull it out into a separate file for easier management
and review.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/Makefile| 3 +-
drivers/staging/android/ion/ion-ioctl.c
From: Laura Abbott <labb...@fedoraproject.org>
There is no advantage to having heap types be a mask. The ion client has
long since dropped the mask. Drop the notion of heap type masks as well.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/uapi/ion.h |
From: Laura Abbott <labb...@fedoraproject.org>
In anticipation of dynamic registration of heaps, switch to using
an idr for heaps. The idr makes it easier to control the assignment
and management + lookup of heap numbers.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
dri
From: Laura Abbott <labb...@fedoraproject.org>
ion_device_add_heap doesn't return an error value. Change it to return
information to callers.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion.c | 7 +--
drivers/staging/android/ion/io
is established.
Thanks,
Laura
P.S. Not to turn this into a bike shedding session but if you have suggestions
for a name for this framework other than Ion I would be interested to hear
them. Too many other things are already named Ion.
Laura Abbott (6):
staging: android: ion: return error value
to call cache operations outside of the
DMA API.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/Kconfig | 14 -
drivers/staging/android/ion/Makefile| 3 +
drivers/staging/android/ion/ion-arm.c
to the underlying
DMA APIs so the final patch in the series adds the appropriate calls.
Feedback is appreciated.
Thanks,
Laura
Laura Abbott (3):
staging: ion: Move away from the DMA APIs for cache flushing
staging: ion: Add support for syncing with DMA_BUF_IOCTL_SYNC
staging: ion: Add dma_map
call backs.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion.c | 12
1 file changed, 12 insertions(+)
diff --git a/drivers/staging/android/ion/ion.c
b/drivers/staging/android/ion/ion.c
index 2096592..52dbea4 100644
--- a/drivers/staging/andro
dma_buf added support for a userspace syncing ioctl. It is implemented
by calling dma_buf_begin_cpu_access and dma_buf_end_cpu_access. Ion
currently lacks cache operations on this code path. Add them for
compatibility with the dma_buf ioctl.
Signed-off-by: Laura Abbott <labb...@redhat.
On 05/22/2016 02:05 AM, Geert Uytterhoeven wrote:
Submitters of device tree binding documentation may forget to CC
the subsystem maintainer if this is missing.
Signed-off-by: Geert Uytterhoeven <ge...@linux-m68k.org>
Cc: Laura Abbott <labb...@redhat.com>
Cc: Sumit Semwa
licon.com>
Reviewed-by: Fu Jun <oliver...@hisilicon.com>
Acked-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion_page_pool.c | 10 +-
drivers/staging/android/ion/ion_priv.h| 5 +-
drivers/staging/android/ion/ion_system_heap.c | 183 +++
On 05/09/2016 01:37 AM, Chen Feng wrote:
Add ion cached pool in system heap.
Signed-off-by: Chen Feng
Signed-off-by: Xia Qing
Reviewed-by: Fu Jun
---
drivers/staging/android/ion/ion_system_heap.c | 145
On 05/05/2016 07:48 PM, Chen Feng wrote:
On 2016/5/6 1:09, Laura Abbott wrote:
On 05/04/2016 08:27 PM, Chen Feng wrote:
Makes the ion buffer always alloced from page pool, no matter
it's cached or not. In this way, it can improve the efficiency
of it.
Currently, there is no difference from
On 05/04/2016 08:27 PM, Chen Feng wrote:
Makes the ion buffer always alloced from page pool, no matter
it's cached or not. In this way, it can improve the efficiency
of it.
Currently, there is no difference from cached or non-cached buffer
for the page pool.
The advantage of the uncached
On 04/17/2016 05:35 PM, Tobin C Harding wrote:
drivers/staging/android/ion/ion.c calls BUG_ON in places where WARN_ON will
suffice.
This patch replaces two such occurences. Two other occurences remain.
You got the logic flow correct but I'd prefer to keep the BUG_ON.
It's catching actual
The android drivers have a few other people reviewing patches.
Add a separate entry to ensure patches go to the right people.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
v2: Drop RFC tag
---
MAINTAINERS | 9 +
1 file changed, 9 insertions(+)
diff --git a/MAINTAI
The android drivers have a few other people reviewing patches.
Add a separate entry to ensure patches go to the right people.
Signed-off-by: Laura Abbott <labb...@redhat.com>
---
Sumit and I have been doing review anyway so I think it makes sense for
us to be cc-ed on patches in ad
On 04/07/2016 11:56 PM, John Einar Reitan wrote:
On Thu, Apr 07, 2016 at 12:37:50PM -0700, Laura Abbott wrote:
On 04/07/2016 04:29 AM, John Einar Reitan wrote:
ion's default user/kernel page mapping code don't honor the offset
option for scatterlists. It uses sg_page and expect the whole page
rkus Böhme <markus.boe...@mailbox.org>
Reviewed-by: Laura Abbott <labb...@redhat.com>
---
drivers/staging/android/ion/ion.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/staging/android/ion/ion.c
b/drivers/staging/android/ion/ion.c
index d
On 04/07/2016 09:32 AM, Sudip Mukherjee wrote:
ion_device_create() can fail and if it fails then it returns the error
value in ERR_PTR.
Reviewed-by: Laura Abbott <labb...@redhat.com>
Signed-off-by: Sudip Mukherjee <sudip.mukher...@codethink.co.uk>
---
drivers/staging
On 04/07/2016 04:29 AM, John Einar Reitan wrote:
ion's default user/kernel page mapping code don't honor the offset
option for scatterlists. It uses sg_page and expect the whole page to be
mapped, while the offset could dictate an offset within a large page.
sg_phys correctly accounts for the
On 03/22/2016 04:08 PM, Moritz Fischer wrote:
Hi Laura,
On Tue, Mar 22, 2016 at 3:51 PM, Laura Abbott <labb...@redhat.com> wrote:
In the past what drivers have done is a foo_ion_client_create which has the
reference
to the ion_device created from ion_device_create. Drivers the
On 03/22/2016 03:33 PM, Moritz Fischer wrote:
Hi all,
probably I'm doing this all wrong. I'm playing around with a bunch of
accelerators
and I need to share buffers between them. I have my heaps implemented as a
platform device and from what I understand I'll need a reference to the
struct
probably
best if you send out a v3 version with the commit message fixed up.
For the actual patch itself, you can add
Reviewed-by: Laura Abbott <labb...@redhat.com>
when you resend.
Thanks,
Laura
Signed-off-by: Derek Yerger <d...@drexel.edu>
---
diff --git a/drivers/staging/android/
d the
> ion driver decrements the reference after it has been
> freed.
>
> To fix this problem client->lock mutex is extended
> to protect all the codes that uses the handle.
>
> Signed-off-by: Eun Taik Lee <eun.taik@samsung.com>
Reviewed-by: Laura Abbott <labb...@redha
On 02/20/2016 07:15 PM, Derek Yerger wrote:
From drivers/staging/android TODO file - sparse errors.
The current code attempts assignment of -1 to unsigned type gfp_t.
Assignment should be an enumerated type of GFP_KERNEL, GFP_ATOMIC,
GFP_HIGHMEM, or __GFP_HIGH.
The original 2014 patch by Gioh
On 02/19/2016 04:03 AM, EunTaik Lee wrote:
> There is a use-after-free problem in the ion driver.
> This is caused by a race condition in the ion_ioctl()
> function.
>
> A handle has ref count of 1 and two tasks on different
> cpus calls ION_IOC_FREE simultaneously.
>
> cpu 0
On 02/16/2016 10:32 PM, EunTaik Lee wrote:
> There was a use-after-free problem in the ion driver.
>
> The problem is detected as an unaligned access in the
> spin lock functions since it uses load exclusive
> instruction. In some cases it corrupts the slub's
> free pointer which causes a
On 02/03/2016 04:03 AM, Robin Murphy wrote:
On 03/02/16 06:49, Gujulan Elango, Hari Prasath (H.) wrote:
From: Hari Prasath Gujulan Elango
Use the managed version of the dma_alloc_coherent() i.e. the
dmam_alloc_coherent() & accordingly cleanup the error handling
On 01/21/2016 03:57 AM, Jon Medhurst (Tixy) wrote:
From: Liviu Dudau
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
Signed-off-by: Jon
On 01/21/2016 12:19 PM, Jon Medhurst (Tixy) wrote:
On Thu, 2016-01-21 at 09:39 -0800, Laura Abbott wrote:
On 01/21/2016 03:57 AM, Jon Medhurst (Tixy) wrote:
From: Liviu Dudau <liviu.du...@arm.com>
ion_buffer_create() will allocate a buffer and then create a DMA
mapping for it, but it
(adding linux-arm and a few people)
On 01/14/2016 06:42 PM, Chen Feng wrote:
The page is already alloc at ion_alloc function,
ion_mmap map the alloced pages to user-space.
The default prot can be PTE_RDONLY. Take a look at
here:
set_pte_at()
arch/arm64/include/asm:
if
he data presented
here.
Reviewed-by: Laura Abbott <labb...@redhat.com>
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
Small sample driver to show what setup would look like with the dt
bindings
Signed-off-by: Laura Abbott <la...@labbott.name>
---
drivers/staging/android/ion/Kconfig | 6 ++
drivers/staging/android/ion/Makefile| 1 +
drivers/staging/android/ion/ion_of_sample.
Devicetree is the preferred mechanism for platform definition
these days. Add a set of files for supporting Ion with devicetree.
This is a set of functions for drivers to call to parse the Ion
devicetree along with definitons for defining heaps.
Signed-off-by: Laura Abbott <la...@labbott.n
This adds a base set of devicetree bindings for the Ion memory
manager. This supports setting up the generic set of heaps and
their properties.
Signed-off-by: Laura Abbott <la...@labbott.name>
---
drivers/staging/android/ion/devicetree.txt | 50 ++
1 file chang
On 1/11/16 3:28 PM, Rob Herring wrote:
On Mon, Jan 11, 2016 at 3:39 PM, Laura Abbott <la...@labbott.name> wrote:
This adds a base set of devicetree bindings for the Ion memory
manager. This supports setting up the generic set of heaps and
their properties.
Signed-off-by: Laura Abbo
Hi,
There was some recent discussion
(http://thread.gmane.org/gmane.linux.ports.arm.kernel/468051/)
about Ion DT bindings and a request that I resend my proposal for common DT
bindings. This is a straight resend, no bux fixed that were pointed out.
Previous thread:
Laura Abbott <labb...@redhat.com>
---
v2: Updated commit message with use case explanation, as suggested by
Laura Abbott<labb...@redhat.com>
drivers/staging/android/ion/ion.c | 21 +++--
drivers/staging/android/ion/ion.h | 20
2
On 01/05/2016 05:03 AM, Rohit kumar wrote:
Currently we can only import dma buf fd's to get ion_handle.
Adding support to import dma buf handles to support kernel
specific use cases.
Signed-off-by: Rohit kumar
---
Currently, ION is the memory manager for graphics in
On 11/17/15 7:15 AM, Arnd Bergmann wrote:
On Monday 16 November 2015 16:57:32 Laura Abbott wrote:
Hi,
This is another attempt at devicetree bindings for Ion. The big complaint from
v1 was that too much unnecessary data was being pushed into devicetree.
v2 takes a different approach of using
Devicetree is the preferred mechanism for platform definition
these days. Add a set of files for supporting Ion with devicetree.
This is a set of functions for drivers to call to parse the Ion
devicetree along with definitons for defining heaps.
Signed-off-by: Laura Abbott <la...@labbott.n
Small sample driver to show what setup would look like with the dt
bindings
Signed-off-by: Laura Abbott <la...@labbott.name>
---
drivers/staging/android/ion/Kconfig | 6 ++
drivers/staging/android/ion/Makefile| 1 +
drivers/staging/android/ion/ion_of_sample.
might look like. As always, your comments and reviews are appreciated.
Thanks,
Laura
Laura Abbott (3):
ion: Devicetree bindings for Ion
staging: ion: Add files for parsing the devicetree (WIP)
NOMERGE: Sample driver
drivers/staging/android/ion/Kconfig | 16 +++
drivers/staging
This adds a base set of devicetree bindings for the Ion memory
manager. This supports setting up the generic set of heaps and
their properties.
Signed-off-by: Laura Abbott <la...@labbott.name>
---
drivers/staging/android/ion/devicetree.txt | 50 ++
1 file chang
On 10/25/15 3:16 PM, Dzonatans Melgalvis wrote:
Fixed checkpatch.pl warning about comment block coding style.
The right aligned comments aren't great style.
How about moving the comments (with the correct style) right
above the macro they are describing.
Signed-off-by: Dzonatans Melgalvis
:35:41 PM, Rob Herring <robherri...@gmail.com>
wrote:
On Tue, Oct 6, 2015 at 3:47 PM, Laura Abbott <labb...@fedoraproject.org>
wrote:
[...]
+Example:
+
+ ion {
+ compatbile = "linux,ion";
+ #address-cells = <1>;
+ #size-cells = <0>;
+
+ ion-system-heap {
+ li
On 10/7/15 11:36 AM, Rob Herring wrote:
On Wed, Oct 7, 2015 at 5:36 AM, Andrew <and...@ncrmnt.org> wrote:
On 2015-10-07 02:01, Laura Abbott wrote:
On 10/6/15 3:35 PM, Rob Herring wrote:
On Tue, Oct 6, 2015 at 3:47 PM, Laura Abbott <labb...@fedoraproject.org>
wrote:
From: Laur
From: Laura Abbott <la...@labbott.name>
This adds a base set of devicetree bindings for the Ion memory
manager. This supports setting up the generic set of heaps and
their properties.
Signed-off-by: Laura Abbott <la...@labbott.name>
Signed-off-by: Andrew Andrianov <an
From: Laura Abbott <la...@labbott.name>
Hi,
At the last Plumbers and Linaro Connect, there was some discussion
related to Ion. One of the items that came up towards eventually
getting Ion out of staging was some concept of stable devicetree
bindings. This is a proof of concept for bi
From: Laura Abbott <la...@labbott.name>
Devicetree is the preferred mechanism for platform definition
these days. Add a set of files for supporting Ion with devicetree.
This includes a set of bindings for heaps common across all
devices and parsing methods. Clients may use the standard
bi
On 10/6/15 3:35 PM, Rob Herring wrote:
On Tue, Oct 6, 2015 at 3:47 PM, Laura Abbott <labb...@fedoraproject.org> wrote:
From: Laura Abbott <la...@labbott.name>
This adds a base set of devicetree bindings for the Ion memory
manager. This supports setting up the generic
On 09/29/2015 10:37 PM, Rohit kumar wrote:
This patch fixes error handling case when buffer->pages allocation
fails. Also, it removes unreachable code of checking ret variable
although it is not updated.
Reviewed-by: Laura Abbott <labb...@redhat.com>
Signed-off-by: Rohit kum
On 09/09/2015 10:41 PM, Colin Cross wrote:
On Wed, Sep 9, 2015 at 10:19 AM, Laura Abbott <labb...@redhat.com> wrote:
(adding Colin and John)
On 09/09/2015 12:41 AM, Shawn Lin wrote:
we found this issue but still exit in lastest kernel. Simply
keep ion_handle_create under mutex_lock to
(adding Colin and John)
On 09/09/2015 12:41 AM, Shawn Lin wrote:
we found this issue but still exit in lastest kernel. Simply
keep ion_handle_create under mutex_lock to avoid this race.
WARNING: CPU: 2 PID: 2648 at drivers/staging/android/ion/ion.c:512
ion_handle_add+0xb4/0xc0()
On 08/03/2015 05:21 PM, Radhakrishna, Pradeep wrote:
From b4069e7fe1b2aa1660b3f944e246c13b4947db21 Mon Sep 17 00:00:00 2001
From: Zhebin Jin zhebin@intel.com
Date: Wed, 8 Jul 2015 10:35:06 +0800
Subject: [PATCH] ion: Handle the memory mapping correctly on x86
This patch modifies the ion
On 07/17/2015 08:21 AM, Robin Murphy wrote:
Hi Tixy,
On 17/07/15 12:01, Jon Medhurst (Tixy) wrote:
Use dma_get_sgtable rather than dma_common_get_sgtable so a device's
dma_ops aren't bypassed. This is essential in situations where a device
uses an IOMMU and the physical memory is not
ion_page_pool_shrink() to count only pages,
not chunks. And then ion_system_heap_shrink() to work on pages.
Signed-off-by: Gioh Kim gioh@lge.com
Reviewed-by: Laura Abbott labb...@redhat.com
---
drivers/staging/android/ion/ion_page_pool.c |5 +++--
drivers/staging/android/ion/ion_system_heap.c
pool size. It is already implemented
but not complete. This patch completes and enables it.
Reading the file returns pool size
in page unit and writing the number of pages shrinks pool.
It flushes all pages to write zero at the file.
Signed-off-by: Gioh Kim gioh@lge.com
Reviewed-by: Laura
(adding devicetree mailing list since I didn't see it cc-ed)
Please also remember to cc the staging list since Ion is
still a staging framework.
On 06/30/2015 08:34 AM, Andrew Andrianov wrote:
Signed-off-by: Andrew Andrianov and...@ncrmnt.org
---
On 06/30/2015 08:34 AM, Andrew Andrianov wrote:
This patch adds a generic ion driver that allows
ion heaps to be added via devicetree. It provides
a simple and generic way to feed physical memory regions
to ion without writing a custom driver, e.g.
ion_sram: ion@0x0010 {
On 11/28/2014 5:31 PM, Zeng Tao wrote:
when cma is located in highmem, virt_to_page will not
work the right way, use pfn_to_page instead.
Signed-off-by: Zeng Tao prime.z...@huawei.com
---
drivers/staging/android/ion/ion_cma_heap.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
On 11/12/2014 11:18 PM, Gioh Kim wrote:
Hi,
I'm wondering why dma_alloc_coherent needs GFP_HIGHUSER | __GFP_ZERO flags in
ion_cma_allocate().
I think dma_alloc_coherent can go to __dma_alloc if the system has CMA area.
If __GFP_WAIT flag is set, it allocates memory from atomic pool.
So
On 10/28/2014 9:57 PM, Gioh Kim wrote:
Hello,
I found that ion insert freed-page at the tail of list
and extract at the head.
I think it is good for cache to use the most recently used page
but these pages in pool are not cached memory.
So I'm not sure using hot page is better or not
Hi,
On 10/23/2014 11:47 PM, Gioh Kim wrote:
This patch shrink page-pool by page unit.
Can you explain a bit more about what this patch is
fixing? The description in the cover letter would
be helpful here.
Thanks,
Laura
--
Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc. is
Hi,
On 10/23/2014 11:47 PM, Gioh Kim wrote:
This patch creates debugfs files, /sys/kernel/debug/ion/heaps/system_shrink,
to shrink pool or get pool size.
Reading the file returns pool size and writing occurs to shrink pool.
Can you clarify here that you are updating the existing debugfs
Hi,
On 10/23/2014 11:47 PM, Gioh Kim wrote:
This patch limits pool size by page unit.
This looks useful. Might be nice to add a debugfs option
to change this at runtime as well.
Signed-off-by: Gioh Kim gioh@lge.com
---
drivers/staging/android/ion/Kconfig |4
On 10/6/2014 3:27 AM, Heesub Shin wrote:
Hello Kumar,
On 10/06/2014 05:31 PM, Pintu Kumar wrote:
The Android ion_system_heap uses allocation fallback mechanism
based on 8,4,0 order pages available in the system.
It changes gfp flags based on higher order allocation request.
This higher order
On 8/10/2014 7:06 PM, Zengtao (B) wrote:
From: Zeng Tao prime.z...@hisilicon.com
Date: Tue, 5 Aug 2014 17:58:10 +0800
Subject: [PATCH] drivers: android: fix watermark when cma enabled
when cma is enabled, the watermark caculation will include the cma pages,
while it can't use cma pages.
On 5/7/2014 4:17 AM, Dan Carpenter wrote:
These days most people don't use git to send patches so I have added a
section about that.
Do you mean most people *do* use git to send patches? Or most people don't
use e-mail clients?
Laura
--
Qualcomm Innovation Center, Inc. is a member of Code
Instead of manually checking the bounds of VMALLOC_START and
VMALLOC_END, just use is_vmalloc_addr. That's what the function
was designed for.
Signed-off-by: Laura Abbott lau...@codeaurora.org
---
.../staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c |3 +--
1 files changed, 1 insertions(+), 2
301 - 392 of 392 matches
Mail list logo