This is the mail archive of the mailing list for the Cygwin project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Help on posix file lock semantics

On Mon, Aug 20, 2001 at 10:55:54PM +0200, Gunnar Andre Dalsnes wrote:
> At 20.08.01 13:18 , you wrote:
> >On Sun, Aug 19, 2001 at 11:06:13PM +0200, Gunnar Andre Dalsnes wrote:
> >
> >      There will be at most one type of lock set for each byte in the
> >      file. Before a successful return from an F_SETLK or an F_SETLKW
> >      request when the calling process has previously existing locks on
> >      bytes in the region specified by the request, the previous lock
> >      type for each byte in the specified region will be replaced by the
> And locks outside of upgraded regions are resized to fit and kept as standalone locks?
> Example:
> A file has write lock from off. 10 to 20 and read lock from off. 30 to 40.
> A new read lock from off. 15 to 35 upgrades both existing overlapped regions.
> Now we have three locks?
> -write lock off. 10 to 15
> -read lock off. 15 to 35
> -read lock off. 35 to 40
> Or maybe they merged? 
> -write lock off. 10 to 15
> -read lock off. 15 to 40
> The reason i ask is that i want F_GETLK to behave correctly if called afterwards.

You got me there. :)

I don't see anything in the standard that covers it, so I guess
it's implementation defined.

Otoh, is it really that important?

> >If someone already holds an exclusive/write lock, and you try to
> >get a read/shared lock, and used F_SETLKW, you will have to wait
> >until that lock is gone, and you can get it.
> >
> >Same goes for getting any lock when someone already has an
> >exclusive/write lock.
> >
> >Read locks can't block each other, but write locks block
> >everything else.
> Stupid me:-) I get it!
> But one thing's for sure:
> A lock of any type can't block any lock attempt from same process (because of upgrading), right?

Right, It only says the request will fail if the lock is hold by
an other process.


Unsubscribe info:
Bug reporting:

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]