Re: [Nfs-ganesha-devel] async dispatch not good
Hi, - Original Message - > From: "William Allen Simpson"> To: "Matt Benjamin" > Cc: "NFS Ganesha Developers" > Sent: Monday, June 19, 2017 10:03:53 PM > Subject: Re: [Nfs-ganesha-devel] async dispatch not good > > On 6/19/17 3:41 PM, Matt Benjamin wrote: > > it's not about memory, this is the problem we're trying to avoid > > > > but, referring for context to our verbal discussion earlier today, your > > suggestion to hybridize the existing output side (which depends on > > blocking sockets) and an async input side using recv() seems work at least > > exploring; I assume you are proposing to use recv() with MSG_DONTWAIT? > > Yes. Linux does support MSG_DONTWAIT, and it should be possible to try > the recv() writev() hybrid approach. At least there's one Oracle > article that says it works > > The underlying problem is EPOLL reaallly isn't a good design. What we > need for speed is callbacks that tell us that the read/write is done, > not signals that there might be more data pending -- which cause us to > do more system calls to find out. System calls are the problem. They have latency, sure. > > kqueue is a much better design. We should try to get kqueue support in > the Linux kernel. That would aid portability, too. You're welcome to try, seems political. > > But what I'm doing right now is backing out my previous attempt. Even > after dumping the mass code, awful lot of hooks to undo Sorry. > > My thought now is it's better to get the big changes in, then work on > TCP I-O re-write separately (as I was doing for UDP and RDMA). Quick > and dirty shims, but only temporarily. One of the key goals I have is read-frags-ahead/non-blocking decode. Has been at the top of the queue since our initial meetings. Seems like your recv() technique should work. > > While I'm thinking about it, why does Ganesha call svc_reg()? AFAICT, > that's just filling in a tree that is never used anymore. > > Can I remove that code in Ganesha? It's a pain to maintain in ntirpc. If it's no longer effective, then eventually, sure. Is it a substantial help to your work? Matt > -- Matt Benjamin Red Hat, Inc. 315 West Huron Street, Suite 140A Ann Arbor, Michigan 48103 http://www.redhat.com/en/technologies/storage tel. 734-821-5101 fax. 734-769-8938 cel. 734-216-5309 -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
Re: [Nfs-ganesha-devel] async dispatch not good
On 6/19/17 3:41 PM, Matt Benjamin wrote: it's not about memory, this is the problem we're trying to avoid but, referring for context to our verbal discussion earlier today, your suggestion to hybridize the existing output side (which depends on blocking sockets) and an async input side using recv() seems work at least exploring; I assume you are proposing to use recv() with MSG_DONTWAIT? Yes. Linux does support MSG_DONTWAIT, and it should be possible to try the recv() writev() hybrid approach. At least there's one Oracle article that says it works The underlying problem is EPOLL reaallly isn't a good design. What we need for speed is callbacks that tell us that the read/write is done, not signals that there might be more data pending -- which cause us to do more system calls to find out. System calls are the problem. kqueue is a much better design. We should try to get kqueue support in the Linux kernel. That would aid portability, too. But what I'm doing right now is backing out my previous attempt. Even after dumping the mass code, awful lot of hooks to undo My thought now is it's better to get the big changes in, then work on TCP I-O re-write separately (as I was doing for UDP and RDMA). Quick and dirty shims, but only temporarily. While I'm thinking about it, why does Ganesha call svc_reg()? AFAICT, that's just filling in a tree that is never used anymore. Can I remove that code in Ganesha? It's a pain to maintain in ntirpc. -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
Re: [Nfs-ganesha-devel] UDP VSOCK?
On 6/19/17 3:44 PM, Matt Benjamin wrote: there is no UDP vsock, it's always a stream socket, this could be done differently, as desired Good, 'cause I've ready submitted the patch. Also, VSOCK only needs to support NFS v3 and v4, not the other programs? But I could be wrong? This question still needs to be answered. My unpublished big patch removes VSOCK support for anything other than v3 and v4. I've got no way to test. -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
Re: [Nfs-ganesha-devel] Mount connection timeout
That’s just LRU running and not finding any work to do. I’m not sure those messages should be LogDebug, maybe they should be LogFullDebug. Frank From: Supriti Singh [mailto:supriti.si...@suse.com] Sent: Monday, June 19, 2017 4:47 AM To: nfs-ganesha-devel@lists.sourceforge.net Subject: [Nfs-ganesha-devel] Mount connection timeout I am using nfs-ganesha v2.5-final + CephFS FSAL. I have noticed that when I mount, sometimes first mount attempt fails with connection time out. And it succeed in later attempts. In event of timeout, the log contains the following lines many times: ganesha.nfsd-4274[cache_lru] lru_run :INODE LRU :DEBUG :After work, open_fd_count:0 count:5 fdrate:1 threadwait=90 ganesha.nfsd-4274[cache_lru] lru_run :INODE LRU :DEBUG :FD count is 0 and low water mark is 2048: not reaping. Can someone please explain what could be possible reason? Thanks, Supriti -- Supriti Singh SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
[Nfs-ganesha-devel] IRC Channel
I'm not sure how well it's known, we do have a #ganesha channel on FreeNode where many of the developers and a few users hang out. We are pretty friendly and willing to answer questions. Frank --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
[Nfs-ganesha-devel] Change in ffilz/nfs-ganesha[next]: Determine size of ACLs to be encoded when getting ACLs
>From: madhu.punj...@in.ibm.com has uploaded this change for review. ( https://review.gerrithub.io/366139 Change subject: Determine size of ACLs to be encoded when getting ACLs .. Determine size of ACLs to be encoded when getting ACLs If nfs4_getfacl is run by the client when there are large number of ACLs and if the server is not able to fit in all ACLs in the buffer of size NFS4_ATTRVALS_BUFFLEN then the client gets Input/Output error. To avoid this we now calculate at run time the size required by ACLs, to determine the size of buffer where the ACLs would be encoded. Change-Id: I4ace8223abe2c2957f6e40cb619a1b838ca78677 Signed-off-by: Madhu Thorat --- M src/Protocols/NFS/nfs_proto_tools.c 1 file changed, 16 insertions(+), 2 deletions(-) git pull ssh://review.gerrithub.io:29418/ffilz/nfs-ganesha refs/changes/39/366139/1 -- To view, visit https://review.gerrithub.io/366139 To unsubscribe, visit https://review.gerrithub.io/settings Gerrit-Project: ffilz/nfs-ganesha Gerrit-Branch: next Gerrit-MessageType: newchange Gerrit-Change-Id: I4ace8223abe2c2957f6e40cb619a1b838ca78677 Gerrit-Change-Number: 366139 Gerrit-PatchSet: 1 Gerrit-Owner: madhu.punj...@in.ibm.com -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
Re: [Nfs-ganesha-devel] Mount connection timeout
yeah - Original Message - > From: "Frank Filz"> To: "Supriti Singh" , > nfs-ganesha-devel@lists.sourceforge.net > Sent: Monday, June 19, 2017 9:46:32 AM > Subject: Re: [Nfs-ganesha-devel] Mount connection timeout > > > > That’s just LRU running and not finding any work to do. I’m not sure those > messages should be LogDebug, maybe they should be LogFullDebug. > > > > Frank > > > > > From: Supriti Singh [mailto:supriti.si...@suse.com] > Sent: Monday, June 19, 2017 4:47 AM > To: nfs-ganesha-devel@lists.sourceforge.net > Subject: [Nfs-ganesha-devel] Mount connection timeout > > > > > I am using nfs-ganesha v2.5-final + CephFS FSAL. I have noticed that when I > mount, sometimes first mount attempt fails with connection time out. And it > succeed in later attempts. > > In event of timeout, the log contains the following lines many times: > > ganesha.nfsd-4274[cache_lru] lru_run :INODE LRU :DEBUG :After work, > open_fd_count:0 count:5 fdrate:1 threadwait=90 > ganesha.nfsd-4274[cache_lru] lru_run :INODE LRU :DEBUG :FD count is 0 and low > water mark is 2048: not reaping. > > > Can someone please explain what could be possible reason? > > Thanks, > Supriti > > > -- > > > Supriti Singh > > > SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, > > > HRB 21284 (AG Nürnberg) > > > > > Virus-free. www.avast.com > > -- > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > ___ > Nfs-ganesha-devel mailing list > Nfs-ganesha-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel > -- Matt Benjamin Red Hat, Inc. 315 West Huron Street, Suite 140A Ann Arbor, Michigan 48103 http://www.redhat.com/en/technologies/storage tel. 734-821-5101 fax. 734-769-8938 cel. 734-216-5309 -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
Re: [Nfs-ganesha-devel] async dispatch not good
Hi Bill, inline - Original Message - > From: "William Allen Simpson"> To: "NFS Ganesha Developers" > Sent: Monday, June 19, 2017 3:04:52 AM > Subject: [Nfs-ganesha-devel] async dispatch not good > > As folks may have noticed, I've been re-working my old 2015 dispatch > patches that eliminate the network input-side queues in Ganesha. > > Matt had wanted fully async non-blocking I-O. I've been poking at it > for a week, and now am sure that's the wrong way to go. I don't think so, but, below > > It might still be good for FSALs. Remains to be seen. DanG and > Soumya are looking at that now. > > The devil in userland network I-O is system calls. Each epoll_wait > is a system call. Each read or write is a system call. Each thread > switch is a system call. > > My code in Ganesha v2.5 (NTIRPC v1.5) gets the network output down to > one system call per request on a very hot thread. Cannot do better, > as trying harder would just push the data into kernel buffers, > possibly slowing our own output (for various reasons). > > Trying to re-work that for async non-blocking calls instead means > many more system calls. Instead of one clean writev with the TCP > fragment header and all ready buffers in one single call, we'd at > minimum have a call, an epoll_wait, spawn another work thread, then > another call and/or release the buffer, rinse and repeat. the expensive part of this (spawn) is necessary only due to aspects of the old design, but, considering effort, ok, below > > For a long buffer chain (the times we want more performance), we'd > have much less performance -- roughly 2 + (3 * number of buffers) > additional system calls. For common short response chains, still > have the extra overhead of the epoll system call, doubling calls. > > Also, using writev minimizes buffer copies. Eliminating data > copying will usually give far better performance. > > The only thing async output is saving is waiting threads. But I've > already got the output threads down to the minimum (per interface). > No gain here! > > On the input side, the truly optimum reduction in system calls would > be one read to get the TCP fragment header and up to 1500 bytes of > data, followed (only when needed) by another read to get the entire > rest of long fragments in one fell swoop. well, maybe, not considering blocking? I think we really do want avoid blocking in the paths that now can/do, but, below > > With async input I've tried level triggered, and am getting spurious > epoll read data signals. Googling shows that's been a problem since > at least 2014, but possible to program around. ok > > Still, this could be better, had it not been terrible for output-side. > > Changing to edge triggered means that every good read would be > followed by another read to make sure that we've gotten all the data. > That is, common small reads turn into two (2) reads. Doubling our > system calls in the common case is not the way to go > > In conclusion, with epoll we know when input data is available, so > input threads aren't sitting around waiting anyway, and trying to > minimize threads results in more system calls and poorer performance. > > NTIRPC already defaults to 200 worker threads. If we need more, we > should allocate more. Memory should not be an issue. it's not about memory, this is the problem we're trying to avoid but, referring for context to our verbal discussion earlier today, your suggestion to hybridize the existing output side (which depends on blocking sockets) and an async input side using recv() seems work at least exploring; I assume you are proposing to use recv() with MSG_DONTWAIT? Matt > > -- > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > ___ > Nfs-ganesha-devel mailing list > Nfs-ganesha-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel > -- Matt Benjamin Red Hat, Inc. 315 West Huron Street, Suite 140A Ann Arbor, Michigan 48103 http://www.redhat.com/en/technologies/storage tel. 734-821-5101 fax. 734-769-8938 cel. 734-216-5309 -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
Re: [Nfs-ganesha-devel] UDP VSOCK?
there is no UDP vsock, it's always a stream socket, this could be done differently, as desired Matt - Original Message - > From: "William Allen Simpson"> To: "NFS Ganesha Developers" > Sent: Friday, June 16, 2017 4:38:09 PM > Subject: [Nfs-ganesha-devel] UDP VSOCK? > > Tried to talk to DanG today, but he went home earlier than usual. So > maybe somebody else knows: > > void Create_SVCXPRTs(void) > { > protos p; > > LogFullDebug(COMPONENT_DISPATCH, "Allocation of the SVCXPRT"); > for (p = P_NFS; p < P_COUNT; p++) > if (nfs_protocol_enabled(p)) { > Create_udp(p); > Create_tcp(p); > } > #ifdef RPC_VSOCK > if (vsock) > create_vsock(); > #endif /* RPC_VSOCK */ > } > > This creates a UDP VSOCK fd, a TCP VSOCK fd, and then another TCP VSOCK > fd. I'm fairly sure the the current code won't work properly for the > UDP VSOCK, and I'm fairly sure that two TCP VSOCKs won't be used. > > Also, VSOCK only needs to support NFS v3 and v4, not the other programs? > > But I could be wrong? > > -- > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > ___ > Nfs-ganesha-devel mailing list > Nfs-ganesha-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel > -- Matt Benjamin Red Hat, Inc. 315 West Huron Street, Suite 140A Ann Arbor, Michigan 48103 http://www.redhat.com/en/technologies/storage tel. 734-821-5101 fax. 734-769-8938 cel. 734-216-5309 -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
[Nfs-ganesha-devel] async dispatch not good
As folks may have noticed, I've been re-working my old 2015 dispatch patches that eliminate the network input-side queues in Ganesha. Matt had wanted fully async non-blocking I-O. I've been poking at it for a week, and now am sure that's the wrong way to go. It might still be good for FSALs. Remains to be seen. DanG and Soumya are looking at that now. The devil in userland network I-O is system calls. Each epoll_wait is a system call. Each read or write is a system call. Each thread switch is a system call. My code in Ganesha v2.5 (NTIRPC v1.5) gets the network output down to one system call per request on a very hot thread. Cannot do better, as trying harder would just push the data into kernel buffers, possibly slowing our own output (for various reasons). Trying to re-work that for async non-blocking calls instead means many more system calls. Instead of one clean writev with the TCP fragment header and all ready buffers in one single call, we'd at minimum have a call, an epoll_wait, spawn another work thread, then another call and/or release the buffer, rinse and repeat. For a long buffer chain (the times we want more performance), we'd have much less performance -- roughly 2 + (3 * number of buffers) additional system calls. For common short response chains, still have the extra overhead of the epoll system call, doubling calls. Also, using writev minimizes buffer copies. Eliminating data copying will usually give far better performance. The only thing async output is saving is waiting threads. But I've already got the output threads down to the minimum (per interface). No gain here! On the input side, the truly optimum reduction in system calls would be one read to get the TCP fragment header and up to 1500 bytes of data, followed (only when needed) by another read to get the entire rest of long fragments in one fell swoop. With async input I've tried level triggered, and am getting spurious epoll read data signals. Googling shows that's been a problem since at least 2014, but possible to program around. Still, this could be better, had it not been terrible for output-side. Changing to edge triggered means that every good read would be followed by another read to make sure that we've gotten all the data. That is, common small reads turn into two (2) reads. Doubling our system calls in the common case is not the way to go In conclusion, with epoll we know when input data is available, so input threads aren't sitting around waiting anyway, and trying to minimize threads results in more system calls and poorer performance. NTIRPC already defaults to 200 worker threads. If we need more, we should allocate more. Memory should not be an issue. -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot ___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
[Nfs-ganesha-devel] Mount connection timeout
I am using nfs-ganesha v2.5-final + CephFS FSAL. I have noticed that when I mount, sometimes first mount attempt fails with connection time out. And it succeed in later attempts. In event of timeout, the log contains the following lines many times: ganesha.nfsd-4274[cache_lru] lru_run :INODE LRU :DEBUG :After work, open_fd_count:0 count:5 fdrate:1 threadwait=90 ganesha.nfsd-4274[cache_lru] lru_run :INODE LRU :DEBUG :FD count is 0 and low water mark is 2048: not reaping. Can someone please explain what could be possible reason? Thanks, Supriti -- Supriti Singh��SUSE Linux GmbH, GF: Felix Imend��rffer, Jane Smithard, Graham Norton, HRB 21284 (AG N��rnberg) -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot___ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel