How to disable fault-around by debugfs?
Hi, I'm just the guy who asked a question about the page fault handler days ago. Thanks for telling me about the fault-around feature. You also told me that this feature could be disabled in debugfs, but I still don't know how to achieve it by debugfs. I only know that debugfs is something that developers could put information about kernel there. So could you tell me how to disable fault-around feature by debugfs? Or is it OK to directly modify kernel? I find where it is located in source: 3518 if (vma->vm_ops->map_pages) { 3519 pte = pte_offset_map_lock(mm, pmd, address, &ptl); 3520 do_fault_around(vma, address, pte, pgoff, flags); 3521 if (!pte_same(*pte, orig_pte)) 3522 goto unlock_out; 3523 pte_unmap_unlock(pte, ptl); 3524 } Thank you very much! Best regards
Hi, I have a question regarding open systemcall "open(dev_name, O_RDWR, 0)" -->userspace suppose devname is /dev/my_device Now when we look into open file operation, we have prototype "int (*open) (struct inode *, struct file *);" could anyone tell me how inode and file structure pointer are filled. I think logic is written in fs/open.c but i got confused in callbacks used in it.could anyone please tell me which topics should i cover to understand open system call. Best regards, Kernel newbie :)
Hi, Sorry i forgot to change subject line :( Please ignore previous mail On Wed, Nov 12, 2014 at 6:38 PM, lokesh kumar <lokeshkumar.222@gmail.com> wrote:
Hi,
I have a question regarding open systemcall
"open(dev_name, O_RDWR, 0)" -->userspace
suppose devname is /dev/my_device
Now when we look into open file operation, we have prototype "int (*open) (struct inode *, struct file *);"
could anyone tell me how inode and file structure pointer are filled.
I think logic is written in fs/open.c
but i got confused in callbacks used in it.could anyone please tell me which topics should i cover to understand open system call.
Best regards, Kernel newbie :)
On Wed, 12 Nov 2014 18:39:08 +0530, lokesh kumar said:
Hi, Sorry i forgot to change subject line :(
Note that for people that use mail software that supports threading, even changing the subject line may not be sufficient - there's another E_mail header. For this note, it says: References: <267163c2.1f570.149a3108a46.Coremail.michaelbest002@126.com> <CAAZ733vy8Q952mYLrDVFHPd_fULd0=9Ja=XczgFbtCg7APONKg@mail.gmail.com> <CAAZ733sk1+SweRe-D8kwx4tQnStUbBBSavZh3PZ0Gu2Bc5tHHg@mail.gmail.com> and can be used by the software to group all posts in the thread together, and even ignore an entire thread. It's also what most web archives use for "display by thread". So for instance, since I changed the Subject: line, but did *not* clean up References:, this still shows up under "How to disable fault-around by debugfs" in the archives: http://lists.kernelnewbies.org/pipermail/kernelnewbies/2014-November/thread.... Moral: Just start with a fresh new post to the list. :)
participants (3)
-
lokesh kumar -
Valdis.Kletnieks＠vt.edu -
秦弋戈