[PATCH] tcp_splice: Fix side in OUT_WAIT flag setting

[PATCH] tcp_splice: Fix side in OUT_WAIT flag setting

[Stefano Brivio]

If the "from" (input) side for a given transfer is 0, and we can't
complete the write right away, what we need to be waiting for is for
output readiness on side 1, not 0, and the other way around as well.

This causes random transfer failures for local TCP connections,
depending if we ever need to wait for output readiness.

Reported-by: Paul Holzinger <pholzing@redhat.com>
Link: https://github.com/containers/podman/issues/23517
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
---
 tcp_splice.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tcp_splice.c b/tcp_splice.c
index 473562b..483e45d 100644
--- a/tcp_splice.c
+++ b/tcp_splice.c
@@ -577,7 +577,7 @@ eintr:
 			if (conn->read[fromsidei] == conn->written[fromsidei])
 				break;
 
-			conn_event(c, conn, OUT_WAIT(fromsidei));
+			conn_event(c, conn, OUT_WAIT(!fromsidei));
 			break;
 		}
 
-- 
@@ -577,7 +577,7 @@ eintr:
 			if (conn->read[fromsidei] == conn->written[fromsidei])
 				break;
 
-			conn_event(c, conn, OUT_WAIT(fromsidei));
+			conn_event(c, conn, OUT_WAIT(!fromsidei));
 			break;
 		}
 
-- 
2.43.0

Re: [PATCH] tcp_splice: Fix side in OUT_WAIT flag setting

[David Gibson]

[-- Attachment #1: Type: text/plain, Size: 1300 bytes --]

On Tue, Aug 06, 2024 at 02:12:49PM +0200, Stefano Brivio wrote:
> If the "from" (input) side for a given transfer is 0, and we can't
> complete the write right away, what we need to be waiting for is for
> output readiness on side 1, not 0, and the other way around as well.
> 
> This causes random transfer failures for local TCP connections,
> depending if we ever need to wait for output readiness.
> 
> Reported-by: Paul Holzinger <pholzing@redhat.com>
> Link: https://github.com/containers/podman/issues/23517
> Signed-off-by: Stefano Brivio <sbrivio@redhat.com>

Reviewed-by: David Gibson <david@gibson.dropbear.id.au>

And mea culpa.

> ---
>  tcp_splice.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tcp_splice.c b/tcp_splice.c
> index 473562b..483e45d 100644
> --- a/tcp_splice.c
> +++ b/tcp_splice.c
> @@ -577,7 +577,7 @@ eintr:
>  			if (conn->read[fromsidei] == conn->written[fromsidei])
>  				break;
>  
> -			conn_event(c, conn, OUT_WAIT(fromsidei));
> +			conn_event(c, conn, OUT_WAIT(!fromsidei));
>  			break;
>  		}
>  

-- 
David Gibson (he or they)	| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you, not the other way
				| around.
http://www.ozlabs.org/~dgibson

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH] tcp_splice: Fix side in OUT_WAIT flag setting

[Paul Holzinger]

On 06/08/2024 14:12, Stefano Brivio wrote:
> If the "from" (input) side for a given transfer is 0, and we can't
> complete the write right away, what we need to be waiting for is for
> output readiness on side 1, not 0, and the other way around as well.
>
> This causes random transfer failures for local TCP connections,
> depending if we ever need to wait for output readiness.
>
> Reported-by: Paul Holzinger <pholzing@redhat.com>
> Link: https://github.com/containers/podman/issues/23517
> Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
Tested-by: Paul Holzinger <pholzing@redhat.com>
> ---
>   tcp_splice.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tcp_splice.c b/tcp_splice.c
> index 473562b..483e45d 100644
> --- a/tcp_splice.c
> +++ b/tcp_splice.c
> @@ -577,7 +577,7 @@ eintr:
>   			if (conn->read[fromsidei] == conn->written[fromsidei])
>   				break;
>   
> -			conn_event(c, conn, OUT_WAIT(fromsidei));
> +			conn_event(c, conn, OUT_WAIT(!fromsidei));
>   			break;
>   		}
>   

-- 
Paul Holzinger