Re: [PATCH] tty: fix data race in tty_buffer_flush

2015-09-17 Thread Greg Kroah-Hartman
On Thu, Sep 17, 2015 at 12:28:11PM +0200, Dmitry Vyukov wrote: > One stupid question. There is a number of branches: > remotes/origin/master > remotes/origin/tty-linus > remotes/origin/tty-next > remotes/origin/tty-testing > > What branch should I base my patches on? I used master. Right

Re: [PATCH] tty: fix data race in tty_buffer_flush

2015-09-17 Thread Dmitry Vyukov
I've resent the 3 patches with version information. On Thu, Sep 17, 2015 at 12:28 PM, Dmitry Vyukov wrote: > One stupid question. There is a number of branches: > remotes/origin/master > remotes/origin/tty-linus > remotes/origin/tty-next > remotes/origin/tty-testing > > What branch should

Re: [PATCH] tty: fix data race in tty_buffer_flush

2015-09-17 Thread Dmitry Vyukov
One stupid question. There is a number of branches: remotes/origin/master remotes/origin/tty-linus remotes/origin/tty-next remotes/origin/tty-testing What branch should I base my patches on? I used master. On Wed, Sep 16, 2015 at 8:20 PM, Dmitry Vyukov wrote: > Sorry for the delay. It

Re: [PATCH] tty: fix data race in tty_buffer_flush

2015-09-16 Thread Dmitry Vyukov
Sorry for the delay. It is starred in my inbox and I plan to resend the patches. Just did not have time yet. On Wed, Sep 16, 2015 at 3:38 AM, Peter Hurley wrote: > On Tue, Sep 8, 2015 at 8:48 AM, Dmitry Vyukov wrote: >> tty_buffer_flush frees not acquired buffers. >> As the result, for example,

Re: [PATCH] tty: fix data race in tty_buffer_flush

2015-09-15 Thread Peter Hurley
On Tue, Sep 8, 2015 at 8:48 AM, Dmitry Vyukov wrote: > tty_buffer_flush frees not acquired buffers. > As the result, for example, read of b->size in tty_buffer_free > can return garbage value which will lead to a huge buffer > hanging in the freelist. This is just the benignest > manifestation of

Re: [PATCH] tty: fix data race in tty_buffer_flush

2015-09-08 Thread Greg KH
On Tue, Sep 08, 2015 at 02:48:26PM +0200, Dmitry Vyukov wrote: > tty_buffer_flush frees not acquired buffers. > As the result, for example, read of b->size in tty_buffer_free > can return garbage value which will lead to a huge buffer > hanging in the freelist. This is just the benignest > manifest