linux-kernel - Re: [PATCH] tpm_tis: fix stall after iowrite*()s

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]

Message-ID: <1619753.FtnHesjV4E@ws-stein>
Date:   Tue, 15 Aug 2017 08:11:07 +0200
From:   Alexander Stein <alexander.stein@...tec-electronic.com>
To:     Haris Okanovic <haris.okanovic@...com>
Cc:     linux-rt-users@...r.kernel.org, linux-kernel@...r.kernel.org,
        tpmdd-devel@...ts.sourceforge.net, harisokn@...il.com,
        julia.cartwright@...com, gratian.crisan@...com,
        scott.hartman@...com, chris.graf@...com, brad.mouring@...com,
        jonathan.david@...com, peterhuewe@....de, tpmdd@...horst.net,
        jarkko.sakkinen@...ux.intel.com, jgunthorpe@...idianresearch.com,
        eric.gardiner@...com
Subject: Re: [PATCH] tpm_tis: fix stall after iowrite*()s

On Monday 14 August 2017 17:53:47, Haris Okanovic wrote:
> --- a/drivers/char/tpm/tpm_tis.c
> +++ b/drivers/char/tpm/tpm_tis.c
> @@ -52,6 +52,22 @@ static inline struct tpm_tis_tcg_phy
> *to_tpm_tis_tcg_phy(struct tpm_tis_data *da return container_of(data,
> struct tpm_tis_tcg_phy, priv);
>  }
> 
> +static inline void tpm_tis_iowrite8(u8 b, void __iomem *iobase, u32 addr)
> +{
> +	iowrite8(b, iobase + addr);
> +#ifdef CONFIG_PREEMPT_RT_FULL
> +	ioread8(iobase + TPM_ACCESS(0));
> +#endif
> +}

Maybe add some comment why an iorad8 is actually requried after each write on 
RT. Currently it is rather obvious why this additional read is necessary. But 
is this still the case in a year?

> +static inline void tpm_tis_iowrite32(u32 b, void __iomem *iobase, u32 addr)
> +{
> +	iowrite32(b, iobase + addr);
> +#ifdef CONFIG_PREEMPT_RT_FULL
> +	ioread8(iobase + TPM_ACCESS(0));
> +#endif
> +}

Same applies here. Or add a comment above both functions describing their 
purpose.

Just my 2 cents

Best regards,
Alexander