linux-next: manual merge of the nfsd tree with the nfsd-fixes tree

linux-next: manual merge of the nfsd tree with the nfsd-fixes tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 2432 bytes --]

Hi all,

Today's linux-next merge of the nfsd tree got a conflict in:

  fs/nfsd/nfsctl.c

between commit:

  76d296a82657 ("nfsd: drop the nfsd_put helper")

from the nfsd-fixes tree and commits:

  3a0b966ab40f ("SUNRPC: discard sv_refcnt, and svc_get/svc_put")
  9bf4b41b79a3 ("nfsd: rename nfsd_last_thread() to nfsd_destroy_serv()")

from the nfsd tree.

I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging.  You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc fs/nfsd/nfsctl.c
index 87fed75808ff,cca1dd7b8c55..000000000000
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@@ -704,15 -707,12 +708,12 @@@ static ssize_t __write_ports_addfd(cha
  	if (err != 0)
  		return err;
  
 -	err = svc_addsock(nn->nfsd_serv, net, fd, buf, SIMPLE_TRANSACTION_LIMIT, cred);
 +	serv = nn->nfsd_serv;
 +	err = svc_addsock(serv, net, fd, buf, SIMPLE_TRANSACTION_LIMIT, cred);
  
- 	if (err < 0 && !serv->sv_nrthreads && !nn->keep_active)
- 		nfsd_last_thread(net);
- 	else if (err >= 0 && !serv->sv_nrthreads && !xchg(&nn->keep_active, 1))
- 		svc_get(serv);
 -	if (!nn->nfsd_serv->sv_nrthreads &&
 -	    list_empty(&nn->nfsd_serv->sv_permsocks))
++	if (!serv->sv_nrthreads && list_empty(&serv->sv_permsocks))
+ 		nfsd_destroy_serv(net);
  
- 	svc_put(serv);
  	return err;
  }
  
@@@ -750,22 -748,18 +751,17 @@@ static ssize_t __write_ports_addxprt(ch
  	if (err < 0 && err != -EAFNOSUPPORT)
  		goto out_close;
  
- 	if (!serv->sv_nrthreads && !xchg(&nn->keep_active, 1))
- 		svc_get(serv);
- 
- 	svc_put(serv);
  	return 0;
  out_close:
 -	xprt = svc_find_xprt(nn->nfsd_serv, transport, net, PF_INET, port);
 +	xprt = svc_find_xprt(serv, transport, net, PF_INET, port);
  	if (xprt != NULL) {
  		svc_xprt_close(xprt);
  		svc_xprt_put(xprt);
  	}
  out_err:
- 	if (!serv->sv_nrthreads && !nn->keep_active)
- 		nfsd_last_thread(net);
 -	if (!nn->nfsd_serv->sv_nrthreads &&
 -	    list_empty(&nn->nfsd_serv->sv_permsocks))
++	if (!serv->sv_nrthreads && list_empty(&serv->sv_permsocks))
+ 		nfsd_destroy_serv(net);
  
- 	svc_put(serv);
  	return err;
  }
  

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the nfsd tree with the nfsd-fixes tree

[Chuck Lever]

On Fri, Jan 05, 2024 at 11:00:07AM +1100, Stephen Rothwell wrote:
> Hi all,
> 
> Today's linux-next merge of the nfsd tree got a conflict in:
> 
>   fs/nfsd/nfsctl.c
> 
> between commit:
> 
>   76d296a82657 ("nfsd: drop the nfsd_put helper")
> 
> from the nfsd-fixes tree and commits:
> 
>   3a0b966ab40f ("SUNRPC: discard sv_refcnt, and svc_get/svc_put")
>   9bf4b41b79a3 ("nfsd: rename nfsd_last_thread() to nfsd_destroy_serv()")
> 
> from the nfsd tree.

Ugh.

I plan to rebase nfsd-next as soon as 76d296a82657 ("nfsd: drop the
nfsd_put helper") is merged. I'll have a closer look at this
conflict tomorrow (my time).


> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
> 
> -- 
> Cheers,
> Stephen Rothwell
> 
> diff --cc fs/nfsd/nfsctl.c
> index 87fed75808ff,cca1dd7b8c55..000000000000
> --- a/fs/nfsd/nfsctl.c
> +++ b/fs/nfsd/nfsctl.c
> @@@ -704,15 -707,12 +708,12 @@@ static ssize_t __write_ports_addfd(cha
>   	if (err != 0)
>   		return err;
>   
>  -	err = svc_addsock(nn->nfsd_serv, net, fd, buf, SIMPLE_TRANSACTION_LIMIT, cred);
>  +	serv = nn->nfsd_serv;
>  +	err = svc_addsock(serv, net, fd, buf, SIMPLE_TRANSACTION_LIMIT, cred);
>   
> - 	if (err < 0 && !serv->sv_nrthreads && !nn->keep_active)
> - 		nfsd_last_thread(net);
> - 	else if (err >= 0 && !serv->sv_nrthreads && !xchg(&nn->keep_active, 1))
> - 		svc_get(serv);
>  -	if (!nn->nfsd_serv->sv_nrthreads &&
>  -	    list_empty(&nn->nfsd_serv->sv_permsocks))
> ++	if (!serv->sv_nrthreads && list_empty(&serv->sv_permsocks))
> + 		nfsd_destroy_serv(net);
>   
> - 	svc_put(serv);
>   	return err;
>   }
>   
> @@@ -750,22 -748,18 +751,17 @@@ static ssize_t __write_ports_addxprt(ch
>   	if (err < 0 && err != -EAFNOSUPPORT)
>   		goto out_close;
>   
> - 	if (!serv->sv_nrthreads && !xchg(&nn->keep_active, 1))
> - 		svc_get(serv);
> - 
> - 	svc_put(serv);
>   	return 0;
>   out_close:
>  -	xprt = svc_find_xprt(nn->nfsd_serv, transport, net, PF_INET, port);
>  +	xprt = svc_find_xprt(serv, transport, net, PF_INET, port);
>   	if (xprt != NULL) {
>   		svc_xprt_close(xprt);
>   		svc_xprt_put(xprt);
>   	}
>   out_err:
> - 	if (!serv->sv_nrthreads && !nn->keep_active)
> - 		nfsd_last_thread(net);
>  -	if (!nn->nfsd_serv->sv_nrthreads &&
>  -	    list_empty(&nn->nfsd_serv->sv_permsocks))
> ++	if (!serv->sv_nrthreads && list_empty(&serv->sv_permsocks))
> + 		nfsd_destroy_serv(net);
>   
> - 	svc_put(serv);
>   	return err;
>   }
>   



-- 
Chuck Lever

Re: linux-next: manual merge of the nfsd tree with the nfsd-fixes tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 410 bytes --]

Hi Chuck,

On Thu, 4 Jan 2024 22:51:07 -0500 Chuck Lever <chuck.lever@oracle.com> wrote:
>
> I plan to rebase nfsd-next as soon as 76d296a82657 ("nfsd: drop the
> nfsd_put helper") is merged. I'll have a closer look at this
> conflict tomorrow (my time).

Instead of rebasing your tree just before the merge window, why not
just merge nfsd-fixes into the nfsd tree?

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the nfsd tree with the nfsd-fixes tree

[Chuck Lever III]

> On Jan 5, 2024, at 5:33 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> 
> Hi Chuck,
> 
> On Thu, 4 Jan 2024 22:51:07 -0500 Chuck Lever <chuck.lever@oracle.com> wrote:
>> 
>> I plan to rebase nfsd-next as soon as 76d296a82657 ("nfsd: drop the
>> nfsd_put helper") is merged. I'll have a closer look at this
>> conflict tomorrow (my time).
> 
> Instead of rebasing your tree just before the merge window, why not
> just merge nfsd-fixes into the nfsd tree?

My development toolchain is stgit, where rebasing is butt-
simple.

I'm not smart enough to make all that work out, and merges
make my head hurt. That's why I let you and Linus handle it,
and then never rebase once my PRs are merged into
torvalds/linux.git.

I'm open to learning more, but I might resist making my dev
workflow more complicated.


--
Chuck Lever

Re: linux-next: manual merge of the nfsd tree with the nfsd-fixes tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 606 bytes --]

Hi Chuck,

On Sat, 6 Jan 2024 00:26:53 +0000 Chuck Lever III <chuck.lever@oracle.com> wrote:
>
> My development toolchain is stgit, where rebasing is butt-
> simple.
> 
> I'm not smart enough to make all that work out, and merges
> make my head hurt. That's why I let you and Linus handle it,
> and then never rebase once my PRs are merged into
> torvalds/linux.git.
> 
> I'm open to learning more, but I might resist making my dev
> workflow more complicated.

Understood, but if you are interested, see
Documentation/maintainer/rebasing-and-merging.rst

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

linux-next: manual merge of the nfsd tree with the nfsd-fixes tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1906 bytes --]

Hi all,

Today's linux-next merge of the nfsd tree got a conflict in:

  fs/nfsd/nfs4callback.c

between commits:

  173936c54b8f ("Revert "NFSD: Reschedule CB operations when backchannel rpc_clnt is shut down"")
  7d32b207859f ("Revert "NFSD: Convert the callback workqueue to use delayed_work"")

from the nfsd-fixes tree and commit:

  7c33236d7e12 ("NFSD: Move callback_wq into struct nfs4_client")

from the nfsd tree.

I fixed it up (I think - see below) and can carry the fix as
necessary. This is now fixed as far as linux-next is concerned, but any
non trivial conflicts should be mentioned to your upstream maintainer
when your tree is submitted for merging.  You may also want to consider
cooperating with the maintainer of the conflicting tree to minimise any
particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

diff --cc fs/nfsd/nfs4callback.c
index e88aca0c6e8e,cf87ace7a1b0..000000000000
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@@ -978,12 -978,22 +978,12 @@@ static int max_cb_time(struct net *net
  	return max(((u32)nn->nfsd4_lease)/10, 1u) * HZ;
  }
  
- static struct workqueue_struct *callback_wq;
- 
  static bool nfsd4_queue_cb(struct nfsd4_callback *cb)
  {
- 	trace_nfsd_cb_queue(cb->cb_clp, cb);
- 	return queue_work(callback_wq, &cb->cb_work);
+ 	struct nfs4_client *clp = cb->cb_clp;
+ 
+ 	trace_nfsd_cb_queue(clp, cb);
 -	return queue_delayed_work(clp->cl_callback_wq, &cb->cb_work, 0);
 -}
 -
 -static void nfsd4_queue_cb_delayed(struct nfsd4_callback *cb,
 -				   unsigned long msecs)
 -{
 -	struct nfs4_client *clp = cb->cb_clp;
 -
 -	trace_nfsd_cb_queue(clp, cb);
 -	queue_delayed_work(clp->cl_callback_wq, &cb->cb_work,
 -			   msecs_to_jiffies(msecs));
++	return queue_work(clp->cl_callback_wq, &cb->cb_work);
  }
  
  static void nfsd41_cb_inflight_begin(struct nfs4_client *clp)

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: manual merge of the nfsd tree with the nfsd-fixes tree

[Chuck Lever]

On Tue, Aug 15, 2023 at 11:06:27AM +1000, Stephen Rothwell wrote:
> Hi all,
> 
> Today's linux-next merge of the nfsd tree got a conflict in:
> 
>   net/sunrpc/svcsock.c
> 
> between commit:
> 
>   c96e2a695e00 ("sunrpc: set the bv_offset of first bvec in svc_tcp_sendmsg")
> 
> from the nfsd-fixes tree and commit:
> 
>   62c25ceb29a6 ("SUNRPC: Convert svc_tcp_sendmsg to use bio_vecs directly")
> 
> from the nfsd tree.
> 
> I fixed it up (I just used the latter) and can carry the fix as
> necessary.

Thanks, using the latter is the correct fix. I will address this in
my repo, since c96e2a695e00 is destined for v6.5-rc soon.


> This is now fixed as far as linux-next is concerned, but any
> non trivial conflicts should be mentioned to your upstream maintainer
> when your tree is submitted for merging.  You may also want to consider
> cooperating with the maintainer of the conflicting tree to minimise any
> particularly complex conflicts.
> 
> -- 
> Cheers,
> Stephen Rothwell



-- 
Chuck Lever

linux-next: manual merge of the nfsd tree with the nfsd-fixes tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 765 bytes --]

Hi all,

Today's linux-next merge of the nfsd tree got a conflict in:

  net/sunrpc/svcsock.c

between commit:

  c96e2a695e00 ("sunrpc: set the bv_offset of first bvec in svc_tcp_sendmsg")

from the nfsd-fixes tree and commit:

  62c25ceb29a6 ("SUNRPC: Convert svc_tcp_sendmsg to use bio_vecs directly")

from the nfsd tree.

I fixed it up (I just used the latter) and can carry the fix as
necessary. This is now fixed as far as linux-next is concerned, but any
non trivial conflicts should be mentioned to your upstream maintainer
when your tree is submitted for merging.  You may also want to consider
cooperating with the maintainer of the conflicting tree to minimise any
particularly complex conflicts.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]