qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v3 24/25] nbd: Fix error_append_hint usage


From: Vladimir Sementsov-Ogievskiy
Subject: Re: [PATCH v3 24/25] nbd: Fix error_append_hint usage
Date: Wed, 25 Sep 2019 07:39:47 +0000

25.09.2019 0:12, Eric Blake wrote:
> On 9/24/19 3:09 PM, Vladimir Sementsov-Ogievskiy wrote:
>> If we want append hint to errp, we must use ERRP_FUNCTION_BEGIN macro.
>> Otherwise hint will not be appended in case of errp == &fatal_err
>> (program will exit before error_append_hint() call). Fix such cases.
>>
> 
> Copy-and-pasted, but if you want to tweak the grammar to all of the
> patches with identical bodies:
> 
> If we want to append a hint to errp, we must use the ERRP_FUNCTION_BEGIN
> macro.  Otherwise, the hint will not be appended when errp == &fatal_err
> (the program will exit prior to the error_append_hint() call).  Fix such
> cases.

Will do, thanks.

> 
>> This commit (together with its neighbors) was generated by
>>
>> git grep -l 'error_append_hint(errp' | while read f; do \
>> spatch --sp-file scripts/coccinelle/fix-error_append_hint-usage.cocci \
>> --in-place $f; done
>>
>> and then
>>
>> ./python/commit-per-subsystem.py MAINTAINERS "$(< auto-msg)"
>>
>> (auto-msg was a file with this commit message)
>>
>> Still, for backporting it may be more comfortable to use only the first
>> command and then do one huge commit.
>>
>> Reported-by: Greg Kurz <address@hidden>
>> Signed-off-by: Vladimir Sementsov-Ogievskiy <address@hidden>
>> ---
> 
> Should the commit-per-subsystem.py script append a distinct CC: line as
> long as it is already grouping files by maintainer?


Hmm, actually, it was in a first version. But then I decided that:
1. I have to use --cc-cmd to handle first three patches anyway
2. I don't like commit messages with "CC:" inside (any reason to keep this
    information in history forever, increasing false-recipients for those
    who forget to add --suppress-cc=all ?)

So, I decided to use --cc-cmd instead.


> 
>>   nbd/client.c | 1 +
>>   nbd/server.c | 1 +
>>   2 files changed, 2 insertions(+)
> 
> Reviewed-by: Eric Blake <address@hidden>
> 
> It's probably easier to take this entire series through one maintainer
> (Markus, since it is error-related), than to have me pick up this patch
> by itself through the NBD tree.

Yes, it simplifies backporting (even backporting something other)

> 
>>
>> diff --git a/nbd/client.c b/nbd/client.c
>> index b9dc829175..4d90a26c00 100644
>> --- a/nbd/client.c
>> +++ b/nbd/client.c
>> @@ -153,6 +153,7 @@ static int nbd_receive_option_reply(QIOChannel *ioc, 
>> uint32_t opt,
>>   static int nbd_handle_reply_err(QIOChannel *ioc, NBDOptionReply *reply,
>>                                   bool strict, Error **errp)
>>   {
>> +    ERRP_FUNCTION_BEGIN();
>>       g_autofree char *msg = NULL;
>>   
>>       if (!(reply->type & (1 << 31))) {
>> diff --git a/nbd/server.c b/nbd/server.c
>> index 28c3c8be85..09565ad8dc 100644
>> --- a/nbd/server.c
>> +++ b/nbd/server.c
>> @@ -1616,6 +1616,7 @@ void nbd_export_close(NBDExport *exp)
>>   
>>   void nbd_export_remove(NBDExport *exp, NbdServerRemoveMode mode, Error 
>> **errp)
>>   {
>> +    ERRP_FUNCTION_BEGIN();
>>       if (mode == NBD_SERVER_REMOVE_MODE_HARD || 
>> QTAILQ_EMPTY(&exp->clients)) {
>>           nbd_export_close(exp);
>>           return;
>>
> 


-- 
Best regards,
Vladimir

reply via email to

[Prev in Thread] Current Thread [Next in Thread]