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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Mon, 4 Apr 2022 16:59:23 +0000
From:   David Laight <David.Laight@...LAB.COM>
To:     'Pavel Skripkin' <paskripkin@...il.com>,
        Michael Straube <straube.linux@...il.com>,
        "Fabio M. De Francesco" <fmdefrancesco@...il.com>
CC:     Greg KH <gregkh@...uxfoundation.org>,
        Larry Finger <Larry.Finger@...inger.net>,
        Phillip Potter <phil@...lpotter.co.uk>,
        "open list:STAGING SUBSYSTEM" <linux-staging@...ts.linux.dev>,
        "Linux Kernel Mailing List" <linux-kernel@...r.kernel.org>
Subject: RE: staging: r8188eu: how to handle nested mutex under spinlock

From: Pavel Skripkin
> Sent: 04 April 2022 17:39
> 
> Hi David,
> 
> On 4/4/22 11:50, David Laight wrote:
> >>
> >> > 		while (pwrpriv->bInSuspend &&
> >>
> >> I've looked into what gcc11 produced from this function and looks like
> >> my compiler is smart enough to not cache that value, but I am afraid not
> >> all compilers are that smart.
> >
> > The compiler can't cache the value because of the function call.
> >
> 
> Hm, I am a newbie in compilers, so can you, please, explain (or give a
> link to any resource where I can read about it) how function call here
> prevent caching.
> 
> IIUC compiler generates code that works well in scope of single-threaded
> application, so why can't compiler cache that value instead of accessing
> memory on each iteration... Isn't register access a way faster than even
> cache hit?

Because calls to external functions are allowed to change
any data via 'other' references.
For instance the structure pointer the function has could
also be in global data somewhere.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ