* [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full
@ 2019-06-11 17:11 Nicholas Tsirakis
2019-06-11 17:11 ` [Xen-devel] [PATCH 2/2] argo: correctly report pending message length Nicholas Tsirakis
2019-06-11 18:43 ` [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full Christopher Clark
0 siblings, 2 replies; 9+ messages in thread
From: Nicholas Tsirakis @ 2019-06-11 17:11 UTC (permalink / raw)
To: xen-devel; +Cc: Nicholas Tsirakis, Christopher Clark
In its current state, if the destination ring is full, sendv()
will requeue the message and return the rc of pending_requeue(),
which will return 0 on success. This prevents the caller from
distinguishing the difference between a successful write and a
message that needs to be resent at a later time.
Instead, capture the -EAGAIN value returned from ringbuf_insert()
and *only* overwrite it if the rc of pending_requeue() is non-zero.
This allows the caller to make intelligent decisions on -EAGAIN and
still be alerted if the pending message fails to requeue.
Signed-off-by: Nicholas Tsirakis <tsirakisn@ainfosec.com>
---
xen/common/argo.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/xen/common/argo.c b/xen/common/argo.c
index 13052b9239..2f874a570d 100644
--- a/xen/common/argo.c
+++ b/xen/common/argo.c
@@ -2048,9 +2048,13 @@ sendv(struct domain *src_d, xen_argo_addr_t *src_addr,
message_type, &len);
if ( ret == -EAGAIN )
{
+ int rc;
+
argo_dprintk("argo_ringbuf_sendv failed, EAGAIN\n");
/* requeue to issue a notification when space is there */
- ret = pending_requeue(dst_d, ring_info, src_id.domain_id, len);
+ rc = pending_requeue(dst_d, ring_info, src_id.domain_id, len);
+ if ( rc )
+ ret = rc;
}
spin_unlock(&ring_info->L3_lock);
--
2.17.1
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [Xen-devel] [PATCH 2/2] argo: correctly report pending message length
2019-06-11 17:11 [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full Nicholas Tsirakis
@ 2019-06-11 17:11 ` Nicholas Tsirakis
2019-06-11 18:49 ` Christopher Clark
2019-06-11 18:43 ` [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full Christopher Clark
1 sibling, 1 reply; 9+ messages in thread
From: Nicholas Tsirakis @ 2019-06-11 17:11 UTC (permalink / raw)
To: xen-devel; +Cc: Nicholas Tsirakis, Christopher Clark
When a message is requeue'd in Xen's internal queue, the queue
entry contains the length of the message so that Xen knows to
send a VIRQ to the respective domain when enough space frees up
in the ring. Due to a small bug, however, Xen doesn't populate
the length of the msg if a given write fails, so this length is
always reported as zero. This causes Xen to spurriously wake up
a domain even when the ring doesn't have enough space.
This patch makes sure that the msg len is properly reported by
populating it in the event of a write failure.
Signed-off-by: Nicholas Tsirakis <tsirakisn@ainfosec.com>
---
xen/common/argo.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/xen/common/argo.c b/xen/common/argo.c
index 2f874a570d..eb541829d6 100644
--- a/xen/common/argo.c
+++ b/xen/common/argo.c
@@ -2050,6 +2050,12 @@ sendv(struct domain *src_d, xen_argo_addr_t *src_addr,
{
int rc;
+ /*
+ * if ringbuf_insert fails, then len will never be populated.
+ * make sure to populate it here.
+ */
+ iov_count(iovs, niov, &len);
+
argo_dprintk("argo_ringbuf_sendv failed, EAGAIN\n");
/* requeue to issue a notification when space is there */
rc = pending_requeue(dst_d, ring_info, src_id.domain_id, len);
--
2.17.1
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
^ permalink raw reply related [flat|nested] 9+ messages in thread
* Re: [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full
2019-06-11 17:11 [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full Nicholas Tsirakis
2019-06-11 17:11 ` [Xen-devel] [PATCH 2/2] argo: correctly report pending message length Nicholas Tsirakis
@ 2019-06-11 18:43 ` Christopher Clark
2019-06-11 19:16 ` Andrew Cooper
1 sibling, 1 reply; 9+ messages in thread
From: Christopher Clark @ 2019-06-11 18:43 UTC (permalink / raw)
To: Nicholas Tsirakis; +Cc: xen-devel, Nicholas Tsirakis
On Tue, Jun 11, 2019 at 10:11 AM Nicholas Tsirakis
<niko.tsirakis@gmail.com> wrote:
>
> In its current state, if the destination ring is full, sendv()
> will requeue the message and return the rc of pending_requeue(),
> which will return 0 on success. This prevents the caller from
> distinguishing the difference between a successful write and a
> message that needs to be resent at a later time.
>
> Instead, capture the -EAGAIN value returned from ringbuf_insert()
> and *only* overwrite it if the rc of pending_requeue() is non-zero.
> This allows the caller to make intelligent decisions on -EAGAIN and
> still be alerted if the pending message fails to requeue.
>
> Signed-off-by: Nicholas Tsirakis <tsirakisn@ainfosec.com>
Thanks for the correct identification of the problem and the patch.
Reviewed-by: Christopher Clark <christopher.clark6@baesystems.com>
> ---
> xen/common/argo.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/xen/common/argo.c b/xen/common/argo.c
> index 13052b9239..2f874a570d 100644
> --- a/xen/common/argo.c
> +++ b/xen/common/argo.c
> @@ -2048,9 +2048,13 @@ sendv(struct domain *src_d, xen_argo_addr_t *src_addr,
> message_type, &len);
> if ( ret == -EAGAIN )
> {
> + int rc;
> +
> argo_dprintk("argo_ringbuf_sendv failed, EAGAIN\n");
> /* requeue to issue a notification when space is there */
> - ret = pending_requeue(dst_d, ring_info, src_id.domain_id, len);
> + rc = pending_requeue(dst_d, ring_info, src_id.domain_id, len);
> + if ( rc )
> + ret = rc;
> }
>
> spin_unlock(&ring_info->L3_lock);
> --
> 2.17.1
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Xen-devel] [PATCH 2/2] argo: correctly report pending message length
2019-06-11 17:11 ` [Xen-devel] [PATCH 2/2] argo: correctly report pending message length Nicholas Tsirakis
@ 2019-06-11 18:49 ` Christopher Clark
2019-06-11 19:55 ` Nicholas Tsirakis
0 siblings, 1 reply; 9+ messages in thread
From: Christopher Clark @ 2019-06-11 18:49 UTC (permalink / raw)
To: Nicholas Tsirakis; +Cc: xen-devel, Nicholas Tsirakis
On Tue, Jun 11, 2019 at 10:11 AM Nicholas Tsirakis
<niko.tsirakis@gmail.com> wrote:
>
> When a message is requeue'd in Xen's internal queue, the queue
> entry contains the length of the message so that Xen knows to
> send a VIRQ to the respective domain when enough space frees up
> in the ring. Due to a small bug, however, Xen doesn't populate
> the length of the msg if a given write fails, so this length is
> always reported as zero. This causes Xen to spurriously wake up
> a domain even when the ring doesn't have enough space.
>
> This patch makes sure that the msg len is properly reported by
> populating it in the event of a write failure.
You're correct that this is an issue to be fixed, but unfortunately
this patch doesn't compile, at least with gcc 8.2 with warnings as
errors, reporting:
argo.c: In function 'sendv':
argo.c:2057:35: error: passing argument 3 of 'iov_count' from
incompatible pointer type [-Werror=incompatible-pointer-types]
iov_count(iovs, niov, &len);
^~~~
argo.c:723:25: note: expected 'unsigned int *' but argument is of type
'long unsigned int *'
unsigned int *count)
~~~~~~~~~~~~~~^~~~~
Even without this error, the logic it implements can unnecessarily
invoke iov_count twice upon the same guest-supplied buffers; it would
be better to avoid that, so: looking at the original section of code:
* sendv's "len" variable can be int, rather than long.
* iov_count can be called from sendv, just before ringbuf_insert,
instead of within ringbuf_insert. It can populate sendv's "len"
variable.
* the len obtained from iov_count (if successful) can be passed into
ringbuf_insert as a parameter, and replace ringbuf_insert's existing
"len" variable.
* ringbuf_insert's "out_len" pointer argument can then be dropped as
unnecessary.
* pending_requeue will be fine to use sendv's populated "len" variable.
Christopher
> Signed-off-by: Nicholas Tsirakis <tsirakisn@ainfosec.com>
> ---
> xen/common/argo.c | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/xen/common/argo.c b/xen/common/argo.c
> index 2f874a570d..eb541829d6 100644
> --- a/xen/common/argo.c
> +++ b/xen/common/argo.c
> @@ -2050,6 +2050,12 @@ sendv(struct domain *src_d, xen_argo_addr_t *src_addr,
> {
> int rc;
>
> + /*
> + * if ringbuf_insert fails, then len will never be populated.
> + * make sure to populate it here.
> + */
> + iov_count(iovs, niov, &len);
> +
> argo_dprintk("argo_ringbuf_sendv failed, EAGAIN\n");
> /* requeue to issue a notification when space is there */
> rc = pending_requeue(dst_d, ring_info, src_id.domain_id, len);
> --
> 2.17.1
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full
2019-06-11 18:43 ` [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full Christopher Clark
@ 2019-06-11 19:16 ` Andrew Cooper
2019-06-11 19:22 ` Christopher Clark
0 siblings, 1 reply; 9+ messages in thread
From: Andrew Cooper @ 2019-06-11 19:16 UTC (permalink / raw)
To: Christopher Clark, Nicholas Tsirakis; +Cc: xen-devel, Nicholas Tsirakis
On 11/06/2019 19:43, Christopher Clark wrote:
> On Tue, Jun 11, 2019 at 10:11 AM Nicholas Tsirakis
> <niko.tsirakis@gmail.com> wrote:
>> In its current state, if the destination ring is full, sendv()
>> will requeue the message and return the rc of pending_requeue(),
>> which will return 0 on success. This prevents the caller from
>> distinguishing the difference between a successful write and a
>> message that needs to be resent at a later time.
>>
>> Instead, capture the -EAGAIN value returned from ringbuf_insert()
>> and *only* overwrite it if the rc of pending_requeue() is non-zero.
>> This allows the caller to make intelligent decisions on -EAGAIN and
>> still be alerted if the pending message fails to requeue.
>>
>> Signed-off-by: Nicholas Tsirakis <tsirakisn@ainfosec.com>
> Thanks for the correct identification of the problem and the patch.
>
> Reviewed-by: Christopher Clark <christopher.clark6@baesystems.com>
So I was coming to commit this, but technically according to the
maintainers file, ARGO is maintained by <christopher.w.clark@gmail.com>
Looking at the ARGO series as committed, the patches where all From:
gmail, SoB: baesystems.
Which is the correct alias to use?
~Andrew
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full
2019-06-11 19:16 ` Andrew Cooper
@ 2019-06-11 19:22 ` Christopher Clark
2019-06-11 19:29 ` Andrew Cooper
0 siblings, 1 reply; 9+ messages in thread
From: Christopher Clark @ 2019-06-11 19:22 UTC (permalink / raw)
To: Andrew Cooper; +Cc: xen-devel, Nicholas Tsirakis, Nicholas Tsirakis
On Tue, Jun 11, 2019 at 12:16 PM Andrew Cooper
<andrew.cooper3@citrix.com> wrote:
>
> On 11/06/2019 19:43, Christopher Clark wrote:
> > On Tue, Jun 11, 2019 at 10:11 AM Nicholas Tsirakis
> > <niko.tsirakis@gmail.com> wrote:
> >> In its current state, if the destination ring is full, sendv()
> >> will requeue the message and return the rc of pending_requeue(),
> >> which will return 0 on success. This prevents the caller from
> >> distinguishing the difference between a successful write and a
> >> message that needs to be resent at a later time.
> >>
> >> Instead, capture the -EAGAIN value returned from ringbuf_insert()
> >> and *only* overwrite it if the rc of pending_requeue() is non-zero.
> >> This allows the caller to make intelligent decisions on -EAGAIN and
> >> still be alerted if the pending message fails to requeue.
> >>
> >> Signed-off-by: Nicholas Tsirakis <tsirakisn@ainfosec.com>
> > Thanks for the correct identification of the problem and the patch.
> >
> > Reviewed-by: Christopher Clark <christopher.clark6@baesystems.com>
>
> So I was coming to commit this, but technically according to the
> maintainers file, ARGO is maintained by <christopher.w.clark@gmail.com>
>
> Looking at the ARGO series as committed, the patches where all From:
> gmail, SoB: baesystems.
>
> Which is the correct alias to use?
For this purpose:
Reviewed-by: Christopher Clark <christopher.w.clark@gmail.com>
thanks
Christopher
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full
2019-06-11 19:22 ` Christopher Clark
@ 2019-06-11 19:29 ` Andrew Cooper
0 siblings, 0 replies; 9+ messages in thread
From: Andrew Cooper @ 2019-06-11 19:29 UTC (permalink / raw)
To: Christopher Clark; +Cc: xen-devel, Nicholas Tsirakis, Nicholas Tsirakis
On 11/06/2019 20:22, Christopher Clark wrote:
> On Tue, Jun 11, 2019 at 12:16 PM Andrew Cooper
> <andrew.cooper3@citrix.com> wrote:
>> On 11/06/2019 19:43, Christopher Clark wrote:
>>> On Tue, Jun 11, 2019 at 10:11 AM Nicholas Tsirakis
>>> <niko.tsirakis@gmail.com> wrote:
>>>> In its current state, if the destination ring is full, sendv()
>>>> will requeue the message and return the rc of pending_requeue(),
>>>> which will return 0 on success. This prevents the caller from
>>>> distinguishing the difference between a successful write and a
>>>> message that needs to be resent at a later time.
>>>>
>>>> Instead, capture the -EAGAIN value returned from ringbuf_insert()
>>>> and *only* overwrite it if the rc of pending_requeue() is non-zero.
>>>> This allows the caller to make intelligent decisions on -EAGAIN and
>>>> still be alerted if the pending message fails to requeue.
>>>>
>>>> Signed-off-by: Nicholas Tsirakis <tsirakisn@ainfosec.com>
>>> Thanks for the correct identification of the problem and the patch.
>>>
>>> Reviewed-by: Christopher Clark <christopher.clark6@baesystems.com>
>> So I was coming to commit this, but technically according to the
>> maintainers file, ARGO is maintained by <christopher.w.clark@gmail.com>
>>
>> Looking at the ARGO series as committed, the patches where all From:
>> gmail, SoB: baesystems.
>>
>> Which is the correct alias to use?
> For this purpose:
> Reviewed-by: Christopher Clark <christopher.w.clark@gmail.com>
Fixed up and pushed. Thanks.
~Andrew
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Xen-devel] [PATCH 2/2] argo: correctly report pending message length
2019-06-11 18:49 ` Christopher Clark
@ 2019-06-11 19:55 ` Nicholas Tsirakis
2019-06-11 19:59 ` Christopher Clark
0 siblings, 1 reply; 9+ messages in thread
From: Nicholas Tsirakis @ 2019-06-11 19:55 UTC (permalink / raw)
To: Christopher Clark; +Cc: xen-devel, Nicholas Tsirakis
On Tue, Jun 11, 2019 at 2:49 PM Christopher Clark
<christopher.w.clark@gmail.com> wrote:
>
> On Tue, Jun 11, 2019 at 10:11 AM Nicholas Tsirakis
> <niko.tsirakis@gmail.com> wrote:
> >
> > When a message is requeue'd in Xen's internal queue, the queue
> > entry contains the length of the message so that Xen knows to
> > send a VIRQ to the respective domain when enough space frees up
> > in the ring. Due to a small bug, however, Xen doesn't populate
> > the length of the msg if a given write fails, so this length is
> > always reported as zero. This causes Xen to spurriously wake up
> > a domain even when the ring doesn't have enough space.
> >
> > This patch makes sure that the msg len is properly reported by
> > populating it in the event of a write failure.
>
> You're correct that this is an issue to be fixed, but unfortunately
> this patch doesn't compile, at least with gcc 8.2 with warnings as
> errors, reporting:
>
> argo.c: In function 'sendv':
> argo.c:2057:35: error: passing argument 3 of 'iov_count' from
> incompatible pointer type [-Werror=incompatible-pointer-types]
> iov_count(iovs, niov, &len);
> ^~~~
> argo.c:723:25: note: expected 'unsigned int *' but argument is of type
> 'long unsigned int *'
> unsigned int *count)
> ~~~~~~~~~~~~~~^~~~~
Shoot, sorry about that, it compiles on my end just fine.
> Even without this error, the logic it implements can unnecessarily
> invoke iov_count twice upon the same guest-supplied buffers; it would
> be better to avoid that, so: looking at the original section of code:
>
> * sendv's "len" variable can be int, rather than long.
> * iov_count can be called from sendv, just before ringbuf_insert,
> instead of within ringbuf_insert. It can populate sendv's "len"
> variable.
> * the len obtained from iov_count (if successful) can be passed into
> ringbuf_insert as a parameter, and replace ringbuf_insert's existing
> "len" variable.
> * ringbuf_insert's "out_len" pointer argument can then be dropped as
> unnecessary.
> * pending_requeue will be fine to use sendv's populated "len" variable.
>
> Christopher
This was an alternative that I had considered. Ultimately I went with my current
implementation as it had less of a SLOC change, though I see now that that was
a poor choice. Shall I submit as a v2 or reply to this thread
directly? Being that
the first patch was already pushed.
> > Signed-off-by: Nicholas Tsirakis <tsirakisn@ainfosec.com>
> > ---
> > xen/common/argo.c | 6 ++++++
> > 1 file changed, 6 insertions(+)
> >
> > diff --git a/xen/common/argo.c b/xen/common/argo.c
> > index 2f874a570d..eb541829d6 100644
> > --- a/xen/common/argo.c
> > +++ b/xen/common/argo.c
> > @@ -2050,6 +2050,12 @@ sendv(struct domain *src_d, xen_argo_addr_t *src_addr,
> > {
> > int rc;
> >
> > + /*
> > + * if ringbuf_insert fails, then len will never be populated.
> > + * make sure to populate it here.
> > + */
> > + iov_count(iovs, niov, &len);
> > +
> > argo_dprintk("argo_ringbuf_sendv failed, EAGAIN\n");
> > /* requeue to issue a notification when space is there */
> > rc = pending_requeue(dst_d, ring_info, src_id.domain_id, len);
> > --
> > 2.17.1
> >
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Xen-devel] [PATCH 2/2] argo: correctly report pending message length
2019-06-11 19:55 ` Nicholas Tsirakis
@ 2019-06-11 19:59 ` Christopher Clark
0 siblings, 0 replies; 9+ messages in thread
From: Christopher Clark @ 2019-06-11 19:59 UTC (permalink / raw)
To: Nicholas Tsirakis; +Cc: xen-devel, Nicholas Tsirakis
On Tue, Jun 11, 2019 at 12:55 PM Nicholas Tsirakis
<niko.tsirakis@gmail.com> wrote:
>
> On Tue, Jun 11, 2019 at 2:49 PM Christopher Clark
> <christopher.w.clark@gmail.com> wrote:
> >
> > On Tue, Jun 11, 2019 at 10:11 AM Nicholas Tsirakis
> > <niko.tsirakis@gmail.com> wrote:
> > >
> > > When a message is requeue'd in Xen's internal queue, the queue
> > > entry contains the length of the message so that Xen knows to
> > > send a VIRQ to the respective domain when enough space frees up
> > > in the ring. Due to a small bug, however, Xen doesn't populate
> > > the length of the msg if a given write fails, so this length is
> > > always reported as zero. This causes Xen to spurriously wake up
> > > a domain even when the ring doesn't have enough space.
> > >
> > > This patch makes sure that the msg len is properly reported by
> > > populating it in the event of a write failure.
> >
> > You're correct that this is an issue to be fixed, but unfortunately
> > this patch doesn't compile, at least with gcc 8.2 with warnings as
> > errors, reporting:
> >
> > argo.c: In function 'sendv':
> > argo.c:2057:35: error: passing argument 3 of 'iov_count' from
> > incompatible pointer type [-Werror=incompatible-pointer-types]
> > iov_count(iovs, niov, &len);
> > ^~~~
> > argo.c:723:25: note: expected 'unsigned int *' but argument is of type
> > 'long unsigned int *'
> > unsigned int *count)
> > ~~~~~~~~~~~~~~^~~~~
>
> Shoot, sorry about that, it compiles on my end just fine.
>
> > Even without this error, the logic it implements can unnecessarily
> > invoke iov_count twice upon the same guest-supplied buffers; it would
> > be better to avoid that, so: looking at the original section of code:
> >
> > * sendv's "len" variable can be int, rather than long.
> > * iov_count can be called from sendv, just before ringbuf_insert,
> > instead of within ringbuf_insert. It can populate sendv's "len"
> > variable.
> > * the len obtained from iov_count (if successful) can be passed into
> > ringbuf_insert as a parameter, and replace ringbuf_insert's existing
> > "len" variable.
> > * ringbuf_insert's "out_len" pointer argument can then be dropped as
> > unnecessary.
> > * pending_requeue will be fine to use sendv's populated "len" variable.
>
> This was an alternative that I had considered. Ultimately I went with my current
> implementation as it had less of a SLOC change, though I see now that that was
> a poor choice. Shall I submit as a v2 or reply to this thread
> directly? Being that
> the first patch was already pushed.
v2 please, ensuring that it applies to staging on top of the prior
patch already applied.
thanks
Christopher
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2019-06-11 19:59 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-06-11 17:11 [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full Nicholas Tsirakis
2019-06-11 17:11 ` [Xen-devel] [PATCH 2/2] argo: correctly report pending message length Nicholas Tsirakis
2019-06-11 18:49 ` Christopher Clark
2019-06-11 19:55 ` Nicholas Tsirakis
2019-06-11 19:59 ` Christopher Clark
2019-06-11 18:43 ` [Xen-devel] [PATCH 1/2] argo: warn sendv() caller when ring is full Christopher Clark
2019-06-11 19:16 ` Andrew Cooper
2019-06-11 19:22 ` Christopher Clark
2019-06-11 19:29 ` Andrew Cooper
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).