* [PATCH] Clarify the cases where BLKIF_RSP_EOPNOTSUPP can be returned.
@ 2025-08-28 9:38 Mark Syms
2025-08-29 7:41 ` Roger Pau Monné
2025-08-29 8:56 ` [PATCH v2] " Mark Syms
0 siblings, 2 replies; 6+ messages in thread
From: Mark Syms @ 2025-08-28 9:38 UTC (permalink / raw)
To: jgross, roger.pau, andrew.cooper3, xen-devel; +Cc: Mark Syms
Previously this said it would only happen on barrier writes. Except
the documentation blocks for
* feature-flush-cache
* feature-discard
Also say that they can return this error.
Signed-off-by: Mark Syms <mark.syms@cloud.com>
---
xen/include/public/io/blkif.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/xen/include/public/io/blkif.h b/xen/include/public/io/blkif.h
index 8407453324..5e617b0c67 100644
--- a/xen/include/public/io/blkif.h
+++ b/xen/include/public/io/blkif.h
@@ -710,7 +710,7 @@ typedef struct blkif_response blkif_response_t;
/*
* STATUS RETURN CODES.
*/
- /* Operation not supported (only happens on barrier writes). */
+ /* Operation not supported, see feature-barrier, feature-flush-cache, feature-discard. */
#define BLKIF_RSP_EOPNOTSUPP -2
/* Operation failed for some unspecified reason (-EIO). */
#define BLKIF_RSP_ERROR -1
--
2.50.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] Clarify the cases where BLKIF_RSP_EOPNOTSUPP can be returned.
2025-08-28 9:38 [PATCH] Clarify the cases where BLKIF_RSP_EOPNOTSUPP can be returned Mark Syms
@ 2025-08-29 7:41 ` Roger Pau Monné
2025-08-29 8:47 ` Mark Syms
2025-08-29 8:56 ` [PATCH v2] " Mark Syms
1 sibling, 1 reply; 6+ messages in thread
From: Roger Pau Monné @ 2025-08-29 7:41 UTC (permalink / raw)
To: Mark Syms; +Cc: jgross, andrew.cooper3, xen-devel
On Thu, Aug 28, 2025 at 10:38:21AM +0100, Mark Syms wrote:
> Previously this said it would only happen on barrier writes. Except
> the documentation blocks for
> * feature-flush-cache
> * feature-discard
>
> Also say that they can return this error.
>
> Signed-off-by: Mark Syms <mark.syms@cloud.com>
> ---
> xen/include/public/io/blkif.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/xen/include/public/io/blkif.h b/xen/include/public/io/blkif.h
> index 8407453324..5e617b0c67 100644
> --- a/xen/include/public/io/blkif.h
> +++ b/xen/include/public/io/blkif.h
> @@ -710,7 +710,7 @@ typedef struct blkif_response blkif_response_t;
> /*
> * STATUS RETURN CODES.
> */
> - /* Operation not supported (only happens on barrier writes). */
> + /* Operation not supported, see feature-barrier, feature-flush-cache, feature-discard. */
> #define BLKIF_RSP_EOPNOTSUPP -2
I think there's no need to mention the specific operations. FWIW,
blkback will return BLKIF_RSP_EOPNOTSUPP for any request type it
doesn't understand (see dispatch_other_io()), which covers any
possible request type that's not yet defined:
/* Operation not supported. */
#define BLKIF_RSP_EOPNOTSUPP -2
Should be fine.
Thanks, Roger.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] Clarify the cases where BLKIF_RSP_EOPNOTSUPP can be returned.
2025-08-29 7:41 ` Roger Pau Monné
@ 2025-08-29 8:47 ` Mark Syms
0 siblings, 0 replies; 6+ messages in thread
From: Mark Syms @ 2025-08-29 8:47 UTC (permalink / raw)
To: Roger Pau Monné; +Cc: jgross, andrew.cooper3, xen-devel
On Fri, 29 Aug 2025 at 08:41, Roger Pau Monné <roger.pau@citrix.com> wrote:
> I think there's no need to mention the specific operations. FWIW,
> blkback will return BLKIF_RSP_EOPNOTSUPP for any request type it
> doesn't understand (see dispatch_other_io()), which covers any
> possible request type that's not yet defined:
>
> /* Operation not supported. */
> #define BLKIF_RSP_EOPNOTSUPP -2
>
> Should be fine.
OK, I'll reduce that comment to the minimum then, I was just trying to
be helpful to future us :)
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH v2] Clarify the cases where BLKIF_RSP_EOPNOTSUPP can be returned.
2025-08-28 9:38 [PATCH] Clarify the cases where BLKIF_RSP_EOPNOTSUPP can be returned Mark Syms
2025-08-29 7:41 ` Roger Pau Monné
@ 2025-08-29 8:56 ` Mark Syms
2025-09-01 11:09 ` Roger Pau Monné
2025-09-01 12:09 ` Jürgen Groß
1 sibling, 2 replies; 6+ messages in thread
From: Mark Syms @ 2025-08-29 8:56 UTC (permalink / raw)
To: jgross, roger.pau, andrew.cooper3, xen-devel; +Cc: Mark Syms
Previously this said it would only happen on barrier writes. Except
the documentation blocks for
* feature-flush-cache
* feature-discard
Also say that they can return this error.
Signed-off-by: Mark Syms <mark.syms@cloud.com>
---
v2: simplify to remove references to features as requested.
---
xen/include/public/io/blkif.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/xen/include/public/io/blkif.h b/xen/include/public/io/blkif.h
index 8407453324..e96c9af7c6 100644
--- a/xen/include/public/io/blkif.h
+++ b/xen/include/public/io/blkif.h
@@ -710,7 +710,7 @@ typedef struct blkif_response blkif_response_t;
/*
* STATUS RETURN CODES.
*/
- /* Operation not supported (only happens on barrier writes). */
+ /* Operation not supported. */
#define BLKIF_RSP_EOPNOTSUPP -2
/* Operation failed for some unspecified reason (-EIO). */
#define BLKIF_RSP_ERROR -1
--
2.50.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH v2] Clarify the cases where BLKIF_RSP_EOPNOTSUPP can be returned.
2025-08-29 8:56 ` [PATCH v2] " Mark Syms
@ 2025-09-01 11:09 ` Roger Pau Monné
2025-09-01 12:09 ` Jürgen Groß
1 sibling, 0 replies; 6+ messages in thread
From: Roger Pau Monné @ 2025-09-01 11:09 UTC (permalink / raw)
To: Mark Syms; +Cc: jgross, andrew.cooper3, xen-devel
On Fri, Aug 29, 2025 at 09:56:27AM +0100, Mark Syms wrote:
> Previously this said it would only happen on barrier writes. Except
> the documentation blocks for
> * feature-flush-cache
> * feature-discard
I cannot parse what "documentation blocks" means in this context. I
would maybe write this as:
"Previously this said it would only happen on barrier writes. However
Linux blkback and possibly other backends already return it when the
type of the requests is unknown.
Fix the comment to clarify the return code can be used as a reply to
any request types not understood by the backend, not just barrier
writes."
>
> Also say that they can return this error.
>
> Signed-off-by: Mark Syms <mark.syms@cloud.com>
Possibly with the commit message adjusted:
Acked-by: Roger Pau Monné <roger.pau@citrix.com>
Thanks, Roger.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2] Clarify the cases where BLKIF_RSP_EOPNOTSUPP can be returned.
2025-08-29 8:56 ` [PATCH v2] " Mark Syms
2025-09-01 11:09 ` Roger Pau Monné
@ 2025-09-01 12:09 ` Jürgen Groß
1 sibling, 0 replies; 6+ messages in thread
From: Jürgen Groß @ 2025-09-01 12:09 UTC (permalink / raw)
To: Mark Syms, roger.pau, andrew.cooper3, xen-devel
[-- Attachment #1.1.1: Type: text/plain, Size: 354 bytes --]
On 29.08.25 10:56, Mark Syms wrote:
> Previously this said it would only happen on barrier writes. Except
> the documentation blocks for
> * feature-flush-cache
> * feature-discard
>
> Also say that they can return this error.
>
> Signed-off-by: Mark Syms <mark.syms@cloud.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Juergen
[-- Attachment #1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 3743 bytes --]
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 495 bytes --]
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2025-09-01 12:10 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-08-28 9:38 [PATCH] Clarify the cases where BLKIF_RSP_EOPNOTSUPP can be returned Mark Syms
2025-08-29 7:41 ` Roger Pau Monné
2025-08-29 8:47 ` Mark Syms
2025-08-29 8:56 ` [PATCH v2] " Mark Syms
2025-09-01 11:09 ` Roger Pau Monné
2025-09-01 12:09 ` Jürgen Groß
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.