Re: [PATCH] concrete /proc/mounts

2019-05-26 Thread J. R. Okajima
Al Viro: > Translation: let's generate the entire contents on the first read() and keep > it until the sucker's closed; that way userland wont' see anything changing > under it. Oh, wait... > > NAK. Do you mean that the change can hide other mountpoints which are kept unchanged before/after read(

Re: [PATCH] concrete /proc/mounts

2019-05-26 Thread Al Viro
On Sun, May 26, 2019 at 06:11:34PM +0900, J. R. Okajima wrote: > commit 1e83f8634c6efe7dd4e6036ee202ca10bdbca0b3 > Author: J. R. Okajima > Date: Sat May 25 18:35:13 2019 +0900 > > concrete /proc/mounts > > When the size of /proc/mounts exceeds PAGE_SIZE, seq_read() has to > rel

[PATCH] concrete /proc/mounts

2019-05-26 Thread J. R. Okajima
commit 1e83f8634c6efe7dd4e6036ee202ca10bdbca0b3 Author: J. R. Okajima Date: Sat May 25 18:35:13 2019 +0900 concrete /proc/mounts When the size of /proc/mounts exceeds PAGE_SIZE, seq_read() has to release namespace_sem via mounts_op.m_stop(). It means if someone else issues