linux-kernel - Re: [PATCH] fbtft: limit transfer length by spi device limit

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: <99987e07-1540-b1c7-d6c8-989e9a3464a7@tronnes.org>
Date:	Sun, 29 May 2016 19:29:44 +0200
From:	Noralf Trønnes <noralf@...nnes.org>
To:	Michal Suchanek <hramrach@...il.com>,
	Thomas Petazzoni <thomas.petazzoni@...e-electrons.com>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	devel@...verdev.osuosl.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] fbtft: limit transfer length by spi device limit


Den 26.05.2016 21:25, skrev Michal Suchanek:
> Some SPI controllers can transfer only small piece of data at a time.
> Since SPI core gained a function to get the maximum transfer length use
> it.
>
> Signed-off-by: Michal Suchanek <hramrach@...il.com>
> ---
>
>   Tested on sunxi spi with DMA enabled and disabled. Makes a visible speed
> difference and display works in either case.
>
>   drivers/staging/fbtft/fbtft-core.c | 9 +++++++++
>   1 file changed, 9 insertions(+)
>
> diff --git a/drivers/staging/fbtft/fbtft-core.c b/drivers/staging/fbtft/fbtft-core.c
> index 0c1a77c..f3bdc8f 100644
> --- a/drivers/staging/fbtft/fbtft-core.c
> +++ b/drivers/staging/fbtft/fbtft-core.c
> @@ -1346,6 +1346,15 @@ int fbtft_probe_common(struct fbtft_display *display,
>   			return PTR_ERR(pdata);
>   	}
>   
> +	if (sdev && (spi_max_transfer_size(sdev) < SIZE_MAX))
> +		if ((pdata->txbuflen <= 0) || (pdata->txbuflen > spi_max_transfer_size(sdev))) {
> +			dev_warn(dev,
> +				 "Limiting used buffer size %i -> %i due to device %s transfer size limitation",
> +				 pdata->txbuflen, spi_max_transfer_size(sdev),
> +				 dev_name(&sdev->dev));
> +			pdata->txbuflen = spi_max_transfer_size(sdev);
> +	}
> +
>   	info = fbtft_framebuffer_alloc(display, dev, pdata);
>   	if (!info)
>   		return -ENOMEM;

(Ugh, this code of mine looks worse each time I'm confronted with it.)

You have even taken care of the special txbuflen == -1 value I see,
so I guess this is as good as it gets without any major refactoring, so:

Acked-by: Noralf Trønnes <noralf@...nnes.org>

And there's no point in doing any refactoring since I'm working on a DRM
successor for fbtft. I have been working on it since the fbdev maintainer
issued the "No more new fbdev drivers, please" call in September and now
it has reached a tipping point where I can say that I will keep working
on it until it's done.
Some info: https://github.com/notro/tinydrm/wiki

Thanks,
Noralf.