From: "Alexandre Courbot" <acourbot@nvidia.com>
To: "Joel Fernandes" <joelagnelf@nvidia.com>,
"Alice Ryhl" <aliceryhl@google.com>
Cc: "Rhys Lloyd" <krakow20@gmail.com>, <dakr@kernel.org>,
<rust-for-linux@vger.kernel.org>, <airlied@gmail.com>,
<simona@ffwll.ch>, <nouveau@lists.freedesktop.org>,
<dri-devel@lists.freedesktop.org>, <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2] gpu: nova-core: vbios: change PmuLookupTableEntry to relax alignment
Date: Wed, 16 Jul 2025 10:28:22 +0900 [thread overview]
Message-ID: <DBD38ILN8ANI.K75X0M10PLNL@nvidia.com> (raw)
In-Reply-To: <b56a6923-7795-4b5f-a57a-5775a3c08f19@nvidia.com>
On Tue Jul 15, 2025 at 12:22 AM JST, Joel Fernandes wrote:
>
>
> On 7/14/2025 10:53 AM, Alice Ryhl wrote:
>> On Mon, Jul 14, 2025 at 4:49 PM Joel Fernandes <joelagnelf@nvidia.com> wrote:
>>>
>>> Hello, Rhys,
>>>
>>> On Mon, Jul 14, 2025 at 04:02:23AM -0700, Rhys Lloyd wrote:
>>>> Instead of the data field containing a u32 and changing the alignment,
>>>> change data to [u8; 4] and convert to u32 with a helper function.
>>>> Removes another magic number by making the struct the same size as
>>>> the data it needs to read, allowing the use of
>>>> `size_of::<PmuLookupTableEntry>()`
>>>>
>>>> Signed-off-by: Rhys Lloyd <krakow20@gmail.com>
>>>> ---
>>>> Changes in v2:
>>>> - get_data helper function renamed to data
>>>>
>>>> ---
>>>> drivers/gpu/nova-core/vbios.rs | 13 +++++++++----
>>>> 1 file changed, 9 insertions(+), 4 deletions(-)
>>>>
>>>> diff --git a/drivers/gpu/nova-core/vbios.rs b/drivers/gpu/nova-core/vbios.rs
>>>> index 5b5d9f38cbb3..339c66e63c7e 100644
>>>> --- a/drivers/gpu/nova-core/vbios.rs
>>>> +++ b/drivers/gpu/nova-core/vbios.rs
>>>> @@ -896,21 +896,26 @@ fn try_from(base: BiosImageBase) -> Result<Self> {
>>>> struct PmuLookupTableEntry {
>>>> application_id: u8,
>>>> target_id: u8,
>>>> - data: u32,
>>>> + data: [u8; 4],
>>>
>>> Instead of this, could we make the struct as #repr[(C, packed)] or does that
>>> not work for some reason?
>>
>> It would probably, but packed structs aren't very nice to work with
>> because Rust has to be really careful to never generate a reference to
>> unaligned fields.
> Oh, interesting. I am Ok with the [u8; 4] then. Btw, we do have several
> #[repr(C, packed)] in vbios.rs already.
Yeah, in this particular case this is a module-local struct for which
(AFAICT) we don't need to generate references to, so unless there are
other issues I think making it packed and storing the properly-ordered
u32 at construction time is both simpler and safer.
prev parent reply other threads:[~2025-07-16 1:28 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-14 11:02 [PATCH v2] gpu: nova-core: vbios: change PmuLookupTableEntry to relax alignment Rhys Lloyd
2025-07-14 11:28 ` Danilo Krummrich
2025-07-14 14:48 ` Joel Fernandes
2025-07-14 14:53 ` Alice Ryhl
2025-07-14 15:22 ` Joel Fernandes
2025-07-16 1:28 ` Alexandre Courbot [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=DBD38ILN8ANI.K75X0M10PLNL@nvidia.com \
--to=acourbot@nvidia.com \
--cc=airlied@gmail.com \
--cc=aliceryhl@google.com \
--cc=dakr@kernel.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=joelagnelf@nvidia.com \
--cc=krakow20@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=nouveau@lists.freedesktop.org \
--cc=rust-for-linux@vger.kernel.org \
--cc=simona@ffwll.ch \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).