netdev - Re: [PATCH net-next v11 4/5] net: rnpgbe: Add basic mbx

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] [day] [month] [year] [list]

Message-ID: <D0FA3B21AAFDF094+20250911014612.GA1855388@nic-Precision-5820-Tower>
Date: Thu, 11 Sep 2025 09:46:12 +0800
From: Yibo Dong <dong100@...se.com>
To: Jörg Sommer <joerg@...so.de>
Cc: Jakub Kicinski <kuba@...nel.org>, "Anwar, Md Danish" <a0501179@...com>,
	andrew+netdev@...n.ch, davem@...emloft.net, edumazet@...gle.com,
	pabeni@...hat.com, horms@...nel.org, corbet@....net,
	gur.stavi@...wei.com, maddy@...ux.ibm.com, mpe@...erman.id.au,
	danishanwar@...com, lee@...ger.us, gongfan1@...wei.com,
	lorenzo@...nel.org, geert+renesas@...der.be,
	Parthiban.Veerasooran@...rochip.com, lukas.bulwahn@...hat.com,
	alexanderduyck@...com, richardcochran@...il.com, kees@...nel.org,
	gustavoars@...nel.org, rdunlap@...radead.org,
	vadim.fedorenko@...ux.dev, netdev@...r.kernel.org,
	linux-doc@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-hardening@...r.kernel.org
Subject: Re: [PATCH net-next v11 4/5] net: rnpgbe: Add basic mbx_fw support

On Wed, Sep 10, 2025 at 10:21:58AM +0200, Jörg Sommer wrote:
> Yibo Dong schrieb am Mi 10. Sep, 14:08 (+0800):
> > On Tue, Sep 09, 2025 at 01:58:22PM -0700, Jakub Kicinski wrote:
> > > On Tue, 9 Sep 2025 19:59:11 +0530 Anwar, Md Danish wrote:
> > > > > +int mucse_mbx_sync_fw(struct mucse_hw *hw)
> > > > > +{
> > > > > +	int try_cnt = 3;
> > > > > +	int err;
> > > > > +
> > > > > +	do {
> > > > > +		err = mucse_mbx_get_info(hw);
> > > > > +		if (err == -ETIMEDOUT)
> > > > > +			continue;
> > > > > +		break;
> > > > > +	} while (try_cnt--);
> > > > > +
> > > > > +	return err;
> > > > > +}  
> > > > 
> > > > There's a logical issue in the code. The loop structure attempts to
> > > > retry on ETIMEDOUT errors, but the unconditional break statement after
> > > > the if-check will always exit the loop after the first attempt,
> > > > regardless of the error. The do-while loop will never actually retry
> > > > because the break statement is placed outside of the if condition that
> > > > checks for timeout errors.
> > > 
> > 
> > What is expected is 'retry on ETIMEDOUT' and 'no retry others'. 
> > https://lore.kernel.org/netdev/a066746c-2f12-4e70-b63a-7996392a9132@lunn.ch/
> > 
> > > The other way around. continue; in a do {} while () look does *not*
> > > evaluate the condition. So this can loop forever.
> > > 
> > 
> > Maybe I can update like this ?
> > 
> > int mucse_mbx_sync_fw(struct mucse_hw *hw)
> > {
> > 	int try_cnt = 3;
> > 	int err;
> > 
> > 	do {
> > 		err = mucse_mbx_get_info(hw);
> > 		if (err != -ETIMEDOUT)
> > 			break;
> > 		/* only retry with ETIMEDOUT, others just return */
> > 	} while (try_cnt--);
> > 
> > 	return err;
> > }  
> 
> How about something like this?
> 
> int mucse_mbx_sync_fw(struct mucse_hw *hw)
> {
> 	for (int try = 3; try; --try) {
> 		int err = mucse_mbx_get_info(hw);
> 		if (err != -ETIMEDOUT)
> 			return err;
> 	}
> 
> 	return ETIMEDOUT;
> }
> 

I think Jakub Kicinski's suggestion is clearer, right?
	do {
		err = mucse_mbx_get_info(hw);
	} while (err == -ETIMEDOUT && try_cnt--);

> 
> My 2cent.
> 
> Regards Jörg
> 
> -- 
> „Es wurden und werden zu viele sprachlose Bücher gedruckt, nach deren
> schon flüchtiger Lektüre man all die Bäume um Vergebung bitten möchte,
> die für den Schund ihr Leben lassen mussten.“ (Michael Jürgs,
>                       Seichtgebiete – Warum wir hemmungslos verblöden)

Thanks for your feedback.