diff mbox series

[v3,1/7] util/error: Introduce warn_report_err_once()

Message ID 20250206131438.1505542-2-clg@redhat.com (mailing list archive)
State New
Headers show
Series vfio: Improve error reporting when MMIO region mapping fails | expand

Commit Message

Cédric Le Goater Feb. 6, 2025, 1:14 p.m. UTC
Depending on the configuration of the host and VM, a passthrough
device may generate recurring DMA mapping errors at runtime. In such
cases, reporting the issue once is sufficient.

We have already the warn/error_report_once() routines taking a format
and arguments. Using the same design pattern, add a new warning
variant taking an 'Error *' parameter.

Cc: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
 include/qapi/error.h | 12 ++++++++++++
 util/error.c         | 11 +++++++++++
 2 files changed, 23 insertions(+)

Comments

Cédric Le Goater Feb. 10, 2025, 4:26 p.m. UTC | #1
On 2/6/25 14:14, Cédric Le Goater wrote:
> Depending on the configuration of the host and VM, a passthrough
> device may generate recurring DMA mapping errors at runtime. In such
> cases, reporting the issue once is sufficient.
> 
> We have already the warn/error_report_once() routines taking a format
> and arguments. Using the same design pattern, add a new warning
> variant taking an 'Error *' parameter.
> 
> Cc: Markus Armbruster <armbru@redhat.com>
> Signed-off-by: Cédric Le Goater <clg@redhat.com>
> ---
>   include/qapi/error.h | 12 ++++++++++++
>   util/error.c         | 11 +++++++++++
>   2 files changed, 23 insertions(+)

Hello Markus,

Are you ok with this change ? Should we take it through the vfio queue ?

Thanks,

C.





> diff --git a/include/qapi/error.h b/include/qapi/error.h
> index 71f8fb2c50eee9a544992d0c05263c9793956fe1..f5fe2162623e5770d652f7415ebc25172d97616e 100644
> --- a/include/qapi/error.h
> +++ b/include/qapi/error.h
> @@ -466,6 +466,18 @@ void warn_reportf_err(Error *err, const char *fmt, ...)
>   void error_reportf_err(Error *err, const char *fmt, ...)
>       G_GNUC_PRINTF(2, 3);
>   
> +/*
> + * Similar to warn_report_err(), except it prints the message just once.
> + * Return true when it prints, false otherwise.
> + */
> +bool warn_report_err_once_cond(bool *printed, Error *err);
> +
> +#define warn_report_err_once(err)                           \
> +    ({                                                      \
> +        static bool print_once_;                            \
> +        warn_report_err_once_cond(&print_once_, err);       \
> +    })
> +
>   /*
>    * Just like error_setg(), except you get to specify the error class.
>    * Note: use of error classes other than ERROR_CLASS_GENERIC_ERROR is
> diff --git a/util/error.c b/util/error.c
> index e5e247209a9e0796074a9794f5598325f22f8d35..673011b89e95f488817b86c31cd389386b2558bb 100644
> --- a/util/error.c
> +++ b/util/error.c
> @@ -247,6 +247,17 @@ void warn_report_err(Error *err)
>       error_free(err);
>   }
>   
> +bool warn_report_err_once_cond(bool *printed, Error *err)
> +{
> +    if (*printed) {
> +        error_free(err);
> +        return false;
> +    }
> +    *printed = true;
> +    warn_report_err(err);
> +    return true;
> +}
> +
>   void error_reportf_err(Error *err, const char *fmt, ...)
>   {
>       va_list ap;
Markus Armbruster Feb. 11, 2025, 6:53 a.m. UTC | #2
Cédric Le Goater <clg@redhat.com> writes:

> On 2/6/25 14:14, Cédric Le Goater wrote:
>> Depending on the configuration of the host and VM, a passthrough
>> device may generate recurring DMA mapping errors at runtime. In such
>> cases, reporting the issue once is sufficient.
>>
>> We have already the warn/error_report_once() routines taking a format
>> and arguments. Using the same design pattern, add a new warning
>> variant taking an 'Error *' parameter.
>>
>> Cc: Markus Armbruster <armbru@redhat.com>
>> Signed-off-by: Cédric Le Goater <clg@redhat.com>
>> ---
>>   include/qapi/error.h | 12 ++++++++++++
>>   util/error.c         | 11 +++++++++++
>>   2 files changed, 23 insertions(+)
>
> Hello Markus,
>
> Are you ok with this change ? Should we take it through the vfio queue ?

Yes, please.

Reviewed-by: Markus Armbruster <armbru@redhat.com>
diff mbox series

Patch

diff --git a/include/qapi/error.h b/include/qapi/error.h
index 71f8fb2c50eee9a544992d0c05263c9793956fe1..f5fe2162623e5770d652f7415ebc25172d97616e 100644
--- a/include/qapi/error.h
+++ b/include/qapi/error.h
@@ -466,6 +466,18 @@  void warn_reportf_err(Error *err, const char *fmt, ...)
 void error_reportf_err(Error *err, const char *fmt, ...)
     G_GNUC_PRINTF(2, 3);
 
+/*
+ * Similar to warn_report_err(), except it prints the message just once.
+ * Return true when it prints, false otherwise.
+ */
+bool warn_report_err_once_cond(bool *printed, Error *err);
+
+#define warn_report_err_once(err)                           \
+    ({                                                      \
+        static bool print_once_;                            \
+        warn_report_err_once_cond(&print_once_, err);       \
+    })
+
 /*
  * Just like error_setg(), except you get to specify the error class.
  * Note: use of error classes other than ERROR_CLASS_GENERIC_ERROR is
diff --git a/util/error.c b/util/error.c
index e5e247209a9e0796074a9794f5598325f22f8d35..673011b89e95f488817b86c31cd389386b2558bb 100644
--- a/util/error.c
+++ b/util/error.c
@@ -247,6 +247,17 @@  void warn_report_err(Error *err)
     error_free(err);
 }
 
+bool warn_report_err_once_cond(bool *printed, Error *err)
+{
+    if (*printed) {
+        error_free(err);
+        return false;
+    }
+    *printed = true;
+    warn_report_err(err);
+    return true;
+}
+
 void error_reportf_err(Error *err, const char *fmt, ...)
 {
     va_list ap;