linux-kernel - Re: [RFC V3 3/3] phy: freescale: fsl-samsung-hdmi: Support dynamic integer

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ZtZcit7HVzzedgdk@atmark-techno.com>
Date: Tue, 3 Sep 2024 09:47:06 +0900
From: Dominique Martinet <dominique.martinet@...ark-techno.com>
To: Adam Ford <aford173@...il.com>
Cc: linux-phy@...ts.infradead.org, linux-imx@....com, festevam@...il.com,
	frieder.schrempf@...tron.de, aford@...conembedded.com,
	Vinod Koul <vkoul@...nel.org>,
	Kishon Vijay Abraham I <kishon@...nel.org>,
	Marco Felsch <m.felsch@...gutronix.de>,
	Uwe Kleine-König <u.kleine-koenig@...gutronix.de>,
	Lucas Stach <l.stach@...gutronix.de>, linux-kernel@...r.kernel.org,
	Makoto Sato <makoto.sato@...ark-techno.com>
Subject: Re: [RFC V3 3/3] phy: freescale: fsl-samsung-hdmi: Support dynamic
 integer

Adam Ford wrote on Mon, Sep 02, 2024 at 04:20:11PM -0500:
> > -       const struct phy_config *cur_cfg;
> > +       struct phy_config cur_cfg;
> 
> Wouldn't converting this from a pointer require me to do a memcpy
> later?  It seems like that's more work than just pointing it to an
> address.
> 
> > -       phy->cur_cfg = &phy_pll_cfg[i];
> > +       phy->cur_cfg = phy_pll_cfg[i];
> 
> I think this is would have to be a memcpy instead of just an equal
> statement since phy->cur_cfg  would no longer be a pointer.

C allows copying structs like this, it's fine to write it as just an
equal.
It's not 100% equivalent, iiuc simple assignment is undefined behaviour
if the elements aren't aligned but memcpy will work even in that case,
but for us this is not a proble mand the generated code should be
identical... Also note I'm only suggesting that because the struct is
tiny (1*u32+7*u8 is less than two u64), but this code isn't meant to run
very often anyway so we should prioritize readability -- if you think
it's harder to understand than an extra pointer somewhere I have no
strong opinion; as said in the previous mail if parallel uses are
possible it'd be better kept on the stack anyway...

-- 
Dominique