[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v31 20/22] Add rx-softmmu
From: |
Eric Blake |
Subject: |
Re: [PATCH v31 20/22] Add rx-softmmu |
Date: |
Mon, 24 Feb 2020 12:31:21 -0600 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.4.1 |
On 2/24/20 11:53 AM, Aleksandar Markovic wrote:
Missing mention of the new enum member with a 'since 5.0' designation
in the documentation.
Just want to bring to your attention a point that is not clear to me here.
(forgive me for being an outsider in this area)
No problem - asking questions is how we learn.
Could you please take a look at commit:
bb5ccf225e81d2801c03e63d16c371f0617270e8
<https://github.com/qemu/qemu/commit/bb5ccf225e81d2801c03e63d16c371f0617270e8#diff-d7db5c644ce52a6080e77ef1e7a3811b>
I am not familiar with QAPI doc generator, but shouldn't "Since: 3.0"
cause the same problem as "ppcemb: dropped in 3.1"? If not, why?
That was the opposite direction - in that commit, we were removing stale
comments about a member that was no longer present (if the member is not
present, the doc engine doesn't know what to attach the tag to, and we
now have the better qemu-deprecated.texi file for tracking removals).
And even if the removal is not documented, users that depended on it
(which should be few, or we would not have removed it), it becomes
fairly obvious at the attempted point of use that it no longer works.
But when the member is present, but added later than the original
struct, knowing what release the additional members were added is useful
documentation when deciding whether support for the new member is
present on all versions of qemu you care about, or whether your
management code has to consider the case when the member was absent.
Yes, you can get the same information from introspection
programmatically (in fact, when libvirt has to deal with new additions,
that's how it probes whether the addition is there), but having the
consistent documentation reminds someone to think about whether checking
the introspection is necessary.
--
Eric Blake, Principal Software Engineer
Red Hat, Inc. +1-919-301-3226
Virtualization: qemu.org | libvirt.org
- [PATCH v31 10/22] target/rx: Use prt_ldmi for XCHG_mr disassembly, (continued)
- [PATCH v31 10/22] target/rx: Use prt_ldmi for XCHG_mr disassembly, Yoshinori Sato, 2020/02/23
- [PATCH v31 08/22] target/rx: Disassemble rx_index_addr into a string, Yoshinori Sato, 2020/02/23
- [PATCH v31 19/22] hw/rx: Restrict the RX62N microcontroller to the RX62N CPU core, Yoshinori Sato, 2020/02/23
- [PATCH v31 05/22] target/rx: TCG helper, Yoshinori Sato, 2020/02/23
- [PATCH v31 21/22] BootLinuxConsoleTest: Test the RX-Virt machine, Yoshinori Sato, 2020/02/23
- [PATCH v31 20/22] Add rx-softmmu, Yoshinori Sato, 2020/02/23
[PATCH v31 17/22] hw/rx: RX Target hardware definition, Yoshinori Sato, 2020/02/23
[PATCH v31 15/22] hw/timer: RX62N internal timer modules, Yoshinori Sato, 2020/02/23
[PATCH v31 07/22] target/rx: RX disassembler, Yoshinori Sato, 2020/02/23
[PATCH v31 14/22] hw/intc: RX62N interrupt controller (ICUa), Yoshinori Sato, 2020/02/23
[PATCH v31 22/22] qemu-doc.texi: Add RX section., Yoshinori Sato, 2020/02/23
[PATCH v31 04/22] target/rx: TCG translation, Yoshinori Sato, 2020/02/23
Re: [PATCH v31 00/22] Add RX archtecture support, no-reply, 2020/02/23
Re: [PATCH v31 00/22] Add RX archtecture support, no-reply, 2020/02/23
Re: [PATCH v31 00/22] Add RX archtecture support, no-reply, 2020/02/23
Re: [PATCH v31 00/22] Add RX archtecture support, no-reply, 2020/02/23