Re: [PATCH 1/2] sched/wait: Break up long wake list walk
Linus Torvalds writes: (Summary) On Tue, Aug 22, 2017 at 1:53 PM, Peter Zijlstra <peterz@infradead.org> wrote: [...] lock to make progress at all. I thought we had explicitly defined yield() to...
View ArticleRe: [PATCH 1/2] sched/wait: Break up long wake list walk
Linus Torvalds writes: (Summary) But if it's some kind of "try and try again" pattern, the second time you try and there are people waiting for the page, the page count (not the map count) would be...
View ArticleRe: [PATCH 1/2] sched/wait: Break up long wake list walk
Linus Torvalds writes: (Summary) On Tue, Aug 22, 2017 at 3:52 PM, Linus Torvalds<torvalds@linux-foundation.org> wrote:[...] MPOL_PREFERRED/MPOL_F_LOCAL really looks completely broken. Of course,...
View ArticleRe: ext4 bug fixes for 4.13
Linus Torvalds writes: (Summary) On Tue, Aug 22, 2017 at 9:16 PM, Theodore Ts'o <tytso@mit.edu> wrote: [...] ext4: fix clang build regression Tssk. The authorship attribution here looks wrong,...
View ArticleRe: Kernels v4.9+ cause short reads of block devices
Linus Torvalds writes: (Summary) wrote: [...] type=bdev Oh, right you are - I'm much too used to 64-bit, where MAX_LFS_FILESIZE is basically infinite, and was jusr assuming that it was something like...
View ArticleRe: [PATCH] x86: enable RCU based table free when PARAVIRT
Linus Torvalds writes: (Summary) You just want to populate *one* page, in order to build up the page directory structure, not allocate all the final points. structure, not allocate all the final...
View ArticleRe: [PATCH] x86: enable RCU based table free when PARAVIRT
Linus Torvalds writes: (Summary) Doing a profile (without the RCU freeing, obviously) gives = me me 0.77% a.out [kernel.vmlinux] [k] free_pgd_range 0.77% a.out [kernel.vmlinux] [k] free_pgd_range 0.77%...
View ArticleRe: [PATCH 1/2] sched/wait: Break up long wake list walk
Linus Torvalds writes: (Summary) */ -static void __wake_up_common(struct wait_queue_head *wq_head, unsigned int mode, - int nr_exclusive, int wake_flags, void *key) +static int __wake_up_common(struct...
View ArticleRe: [PATCH 0/1] devpts: use dynamic_dname() to generate proc name
Linus Torvalds writes: (Summary) wrote: [...] W: Could not unmount dev/ptmx: umount: /var/cache/pbuilder/build/1340/dev/ptmx: target is busy Yes, that patch definitely keeps a reference to the pts...
View ArticleRe: [PATCH 0/1] devpts: use dynamic_dname() to generate proc name
Linus Torvalds writes: (Summary) There's no way around it. no way around it. So I think I will have to just revert that fix. So I think I will have to just revert that fix. Damn. Damn. Now, I think...
View ArticleRe: [PATCH 0/1] devpts: use dynamic_dname() to generate proc name
Linus Torvalds writes: (Summary) It's not strictly "new", though - we've done that for a while, and if you used /dev/pts/ptmx you'd even have had the *right* path for a while ;) for a while ;) And this...
View ArticleRe: [PATCH 0/1] devpts: use dynamic_dname() to generate proc name
Linus Torvalds writes: (Summary) And it's *not* fairly straightforward, because the tty_operations "ioctl()" function pointer only gets 'struct tty *'. So in the TIOCGPTPEER path, we don't actually...
View ArticleRe: [PATCH 0/1] devpts: use dynamic_dname() to generate proc name
Linus Torvalds writes: (Summary) So we're back to the old behavior that gives the wrong pathname in /proc. that gives the wrong pathname in /proc. And I think I can handle the lack of a 'struct file *'...
View ArticleRe: [PATCH 0/1] devpts: use dynamic_dname() to generate proc name
Linus Torvalds writes: (Summary) > */ > - pts_path = kmalloc(sizeof(struct path), GFP_KERNEL); > - if (!pts_path) > > > - pts_path->dentry = dentry; > > > +...
View ArticleRe: [PATCH 0/1] devpts: use dynamic_dname() to generate proc name
Linus Torvalds writes: (Summary) I just committed Eric's final patch, which is pretty much exactly the same as what I sent out except with an added verification that the mount has not changed. mount...
View ArticleRe: [PATCH 0/1] devpts: use dynamic_dname() to generate proc name
Linus Torvalds writes: (Summary) /dev/pts/ptmx simply is *not*. /dev/pts/ptmx simply is *not*. So pretty much every single user that ever uses pty's will use /dev/ptmx, it's just how it has always...
View ArticleRe: [PATCH 0/1] devpts: use dynamic_dname() to generate proc name
Linus Torvalds writes: (Summary) It's the SysV STREAMS standard location, and it is what Sysv pty users _will_ use directly. pty users _will_ use directly. Linux didn't make up that name. Linux didn't...
View ArticleRe: [PATCH 2/2 v2] sched/wait: Introduce lock breaker in wake_up_p ...
Linus Torvalds writes: (Summary) - an exclusive wait (ie somebody who waits to get the bit for locking) adds itself to the 'exclusive' llist - non-locking waiters add themselves to the 'all' list - we...
View ArticleRe: [PATCH 2/2 v2] sched/wait: Introduce lock breaker in wake_up_p ...
Linus Torvalds writes: (Summary) By using a hash table of + * waitqueues where the bucket discipline is to maintain all + * waiters on the same queue and wake all when any of the pages + * become...
View ArticleRe: [PATCH 2/2 v2] sched/wait: Introduce lock breaker in wake_up_p ...
Linus Torvalds writes: (Summary) + * + * Note that for any particular page, only one of the + * structs will actually be visible at the head of the + * wait queue hash table at any particular time, but...
View Article