Re: [V9fs-developer] 9pfs hangs since 4.7

On Sun, 8 Jan 2017 05:46:39 + Al Viro wrote: > On Sat, Jan 07, 2017 at 06:15:23PM +, Al Viro wrote: > > On Sat, Jan 07, 2017 at 05:19:10PM +, Al Viro wrote: > > > > > released) simply trigger virtio_queue_notify_vq() again? It *is* a bug > > > (if we get

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sun, 8 Jan 2017 05:46:39 + Al Viro wrote: > On Sat, Jan 07, 2017 at 06:15:23PM +, Al Viro wrote: > > On Sat, Jan 07, 2017 at 05:19:10PM +, Al Viro wrote: > > > > > released) simply trigger virtio_queue_notify_vq() again? It *is* a bug > > > (if we get a burst filling a

Re: [V9fs-developer] 9pfs hangs since 4.7

On Mon, Jan 09, 2017 at 08:39:31PM +0200, Tuomas Tynkkynen wrote: > Yes, this does seem to be related to this or otherwise MAX_REQ related! > - Bumping MAX_REQ up to 1024 makes the hang go away (on 4.7). > - Dropping it to 64 makes the same hang happen on kernels where it worked > before (I

Re: [V9fs-developer] 9pfs hangs since 4.7

On Mon, Jan 09, 2017 at 08:39:31PM +0200, Tuomas Tynkkynen wrote: > Yes, this does seem to be related to this or otherwise MAX_REQ related! > - Bumping MAX_REQ up to 1024 makes the hang go away (on 4.7). > - Dropping it to 64 makes the same hang happen on kernels where it worked > before (I

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, 7 Jan 2017 17:19:10 + Al Viro wrote: > On Sat, Jan 07, 2017 at 04:10:45PM +0100, Greg Kurz wrote: > > > virtqueue_push(), but pdu freeing is delayed until v9fs_flush() gets woken > > > up. In the meanwhile, another request arrives into the slot of freed by >

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, 7 Jan 2017 17:19:10 + Al Viro wrote: > On Sat, Jan 07, 2017 at 04:10:45PM +0100, Greg Kurz wrote: > > > virtqueue_push(), but pdu freeing is delayed until v9fs_flush() gets woken > > > up. In the meanwhile, another request arrives into the slot of freed by > > > that

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, 7 Jan 2017 17:19:10 + Al Viro wrote: > On Sat, Jan 07, 2017 at 04:10:45PM +0100, Greg Kurz wrote: > > > virtqueue_push(), but pdu freeing is delayed until v9fs_flush() gets woken > > > up. In the meanwhile, another request arrives into the slot of freed by >

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, 7 Jan 2017 17:19:10 + Al Viro wrote: > On Sat, Jan 07, 2017 at 04:10:45PM +0100, Greg Kurz wrote: > > > virtqueue_push(), but pdu freeing is delayed until v9fs_flush() gets woken > > > up. In the meanwhile, another request arrives into the slot of freed by > > > that

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, Jan 07, 2017 at 06:15:23PM +, Al Viro wrote: > On Sat, Jan 07, 2017 at 05:19:10PM +, Al Viro wrote: > > > released) simply trigger virtio_queue_notify_vq() again? It *is* a bug > > (if we get a burst filling a previously empty queue all at once, there won't > > be any slots

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, Jan 07, 2017 at 06:15:23PM +, Al Viro wrote: > On Sat, Jan 07, 2017 at 05:19:10PM +, Al Viro wrote: > > > released) simply trigger virtio_queue_notify_vq() again? It *is* a bug > > (if we get a burst filling a previously empty queue all at once, there won't > > be any slots

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, Jan 07, 2017 at 05:19:10PM +, Al Viro wrote: > released) simply trigger virtio_queue_notify_vq() again? It *is* a bug > (if we get a burst filling a previously empty queue all at once, there won't > be any slots becoming freed Umm... Nope, even in that scenario we'll have all

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, Jan 07, 2017 at 05:19:10PM +, Al Viro wrote: > released) simply trigger virtio_queue_notify_vq() again? It *is* a bug > (if we get a burst filling a previously empty queue all at once, there won't > be any slots becoming freed Umm... Nope, even in that scenario we'll have all

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, Jan 07, 2017 at 04:10:45PM +0100, Greg Kurz wrote: > > virtqueue_push(), but pdu freeing is delayed until v9fs_flush() gets woken > > up. In the meanwhile, another request arrives into the slot of freed by > > that virtqueue_push() and we are out of pdus. > > > > Indeed. Even if this

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, Jan 07, 2017 at 04:10:45PM +0100, Greg Kurz wrote: > > virtqueue_push(), but pdu freeing is delayed until v9fs_flush() gets woken > > up. In the meanwhile, another request arrives into the slot of freed by > > that virtqueue_push() and we are out of pdus. > > > > Indeed. Even if this

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, 7 Jan 2017 06:26:47 + Al Viro wrote: > On Fri, Jan 06, 2017 at 02:52:35PM +0100, Greg Kurz wrote: > > > Looking at the tag numbers, I think we're hitting the hardcoded limit of 128 > > simultaneous requests in QEMU (which doesn't produce any error, new

Re: [V9fs-developer] 9pfs hangs since 4.7

On Sat, 7 Jan 2017 06:26:47 + Al Viro wrote: > On Fri, Jan 06, 2017 at 02:52:35PM +0100, Greg Kurz wrote: > > > Looking at the tag numbers, I think we're hitting the hardcoded limit of 128 > > simultaneous requests in QEMU (which doesn't produce any error, new requests > > are silently

Re: [V9fs-developer] 9pfs hangs since 4.7

On Fri, Jan 06, 2017 at 02:52:35PM +0100, Greg Kurz wrote: > Looking at the tag numbers, I think we're hitting the hardcoded limit of 128 > simultaneous requests in QEMU (which doesn't produce any error, new requests > are silently dropped). > > Tuomas, can you change MAX_REQ to some higher

Re: [V9fs-developer] 9pfs hangs since 4.7

On Fri, Jan 06, 2017 at 02:52:35PM +0100, Greg Kurz wrote: > Looking at the tag numbers, I think we're hitting the hardcoded limit of 128 > simultaneous requests in QEMU (which doesn't produce any error, new requests > are silently dropped). > > Tuomas, can you change MAX_REQ to some higher

Re: [V9fs-developer] 9pfs hangs since 4.7

On Wed, 4 Jan 2017 23:01:01 + Al Viro wrote: > > Here's logs that should be complete this time: > > > > https://gist.githubusercontent.com/dezgeg/08629d4c8ca79da794bc087e5951e518/raw/a1a82b9bc24e5282c82beb43a9dc91974ffcf75a/9p.qemu.log > >

Re: [V9fs-developer] 9pfs hangs since 4.7

On Wed, 4 Jan 2017 23:01:01 + Al Viro wrote: > > Here's logs that should be complete this time: > > > > https://gist.githubusercontent.com/dezgeg/08629d4c8ca79da794bc087e5951e518/raw/a1a82b9bc24e5282c82beb43a9dc91974ffcf75a/9p.qemu.log > >