android: binder: use VM_ALLOC to get vm area
PATCH v2 1/1 android: binder: use VM_ALLOC to get vm area
VM_IOREMAP is used to access hardware through a mechanism called I/O mapped memory. Android binder is a IPC machanism which will not access I/O memory.
Also VM_IOREMAP has alignment requiement which may not needed in binder.
__get_vm_area_node()
{
...
if (flags & VM_IOREMAP)
align = 1ul << clamp_t(int, fls_long(size),
PAGE_SHIFT, IOREMAP_MAX_ORDER);
...
}This patch use VM_ALLOC to get vm area.
Below is the throughput test result:
# ./binderThroughputTest -w 100
I run this command 10 times:
before after
average iterations per sec: 11199.9 11886.9
No performance regression found throgh binder test.
--- drivers/android/binder.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/android/binder.c +1 -1
@@ -2901,7 +2901,7 @@ static int binder_mmap(struct file *filp, struct vm_area_struct *vma) goto err_already_mapped; } - area =get_vm_area (vma->vm_end - vma->vm_start,VM_IOREMAP ); + area =get_vm_area (vma->vm_end - vma->vm_start,VM_ALLOC );if (area ==NULL ) { ret = -ENOMEM ; failure_string ="get_vm_area" ;
Signature · 1 line
1.9.1
Greg KH
maintainerQuoted 24 lines · Ganesh Mahendran
VM_IOREMAPis used to access hardware through a mechanism called I/O mapped memory. Android binder is a IPC machanism which will not access I/O memory.Also
VM_IOREMAPhas alignment requiement which may not needed in binder.__get_vm_area_node() { ... if (flags & VM_IOREMAP) align = 1ul << clamp_t(int, fls_long(size), PAGE_SHIFT, IOREMAP_MAX_ORDER); ... }This patch use
VM_ALLOCto get vm area.Below is the throughput test result:
# ./binderThroughputTest -w 100
I run this command 10 times:
before after
average iterations per sec: 11199.9 11886.9
Quoted 6 lines
No performance regression found throgh binder test.
Signed-off-by Ganesh Mahendran <[email protected]>--- drivers/android/binder.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
What changed from v1?
Always list that below the --- line.
thanks,
greg k-h
Hi, Greg
2016-11-15 18:18 GMT+08:00 Greg KH <[email protected]>:
Quoted 25 lines · Ganesh Mahendran
Quoted 24 lines
VM_IOREMAPis used to access hardware through a mechanism called I/O mapped memory. Android binder is a IPC machanism which will not access I/O memory.Also
VM_IOREMAPhas alignment requiement which may not needed in binder.__get_vm_area_node() { ... if (flags & VM_IOREMAP) align = 1ul << clamp_t(int, fls_long(size), PAGE_SHIFT, IOREMAP_MAX_ORDER); ... }This patch use
VM_ALLOCto get vm area.Below is the throughput test result:
# ./binderThroughputTest -w 100
I run this command 10 times:
before after
average iterations per sec: 11199.9 11886.9
Quoted 8 lines
Quoted 6 lines
No performance regression found throgh binder test.
Signed-off-by Ganesh Mahendran <[email protected]>--- drivers/android/binder.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)What changed from v1?
Sorry for missing the change information.
In V2, I run the binder test. And there is no side effect with this patch.
Quoted 2 lines
Always list that below the --- line.
Thanks for reminder.
Quoted 4 lines
thanks,
greg k-h
A gentle ping.
Thanks.
2016-11-15 21:18 GMT+08:00 Ganesh Mahendran <[email protected]>:
Quoted 51 lines
Hi, Greg
2016-11-15 18:18 GMT+08:00 Greg KH <[email protected]>:
Quoted 25 lines · Ganesh Mahendran
Quoted 24 lines
VM_IOREMAPis used to access hardware through a mechanism called I/O mapped memory. Android binder is a IPC machanism which will not access I/O memory.Also
VM_IOREMAPhas alignment requiement which may not needed in binder.__get_vm_area_node() { ... if (flags & VM_IOREMAP) align = 1ul << clamp_t(int, fls_long(size), PAGE_SHIFT, IOREMAP_MAX_ORDER); ... }This patch use
VM_ALLOCto get vm area.Below is the throughput test result:
# ./binderThroughputTest -w 100
I run this command 10 times:
before after
average iterations per sec: 11199.9 11886.9Quoted 8 lines
Quoted 6 lines
No performance regression found throgh binder test.
Signed-off-by Ganesh Mahendran <[email protected]>--- drivers/android/binder.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)What changed from v1?
Sorry for missing the change information.
In V2, I run the binder test. And there is no side effect with this patch.
Quoted 2 lines
Always list that below the --- line.
Thanks for reminder.
Quoted 4 lines
thanks,
greg k-h
Greg KH
maintainerQuoted 1 line · Ganesh Mahendran
A gentle ping.
I don't see a patch here that can be accepted, what are you asking for a response from?
confused,
greg k-h
Hi, Greg:
2017-02-09 18:17 GMT+08:00 Greg KH <[email protected]>:
Quoted 5 lines · Ganesh Mahendran
Quoted 1 line
A gentle ping.
I don't see a patch here that can be accepted, what are you asking for a response from?
I sent a patch before:
https://patchwork.kernel.org/patch/9429257/
Please help to review.
Thanks.
Quoted 4 lines
confused,
greg k-h
Greg KH
maintainerQuoted 11 lines · Ganesh Mahendran
Hi, Greg:
2017-02-09 18:17 GMT+08:00 Greg KH <[email protected]>:
Quoted 5 lines · Ganesh Mahendran
Quoted 1 line
A gentle ping.
I don't see a patch here that can be accepted, what are you asking for a response from?
I sent a patch before:
https://patchwork.kernel.org/patch/9429257/
I don't use patchwork, please respond to the patch you are referring to, or resend it.
thanks,
greg k-h