mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Sascha Hauer <sha@pengutronix.de>
To: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Cc: barebox@lists.infradead.org
Subject: Re: [PATCH v2 12/12] firmware: add support for compressed images
Date: Thu, 28 Jan 2021 16:27:33 +0100	[thread overview]
Message-ID: <20210128152733.GD19583@pengutronix.de> (raw)
In-Reply-To: <20210127134423.31587-12-s.trumtrar@pengutronix.de>

On Wed, Jan 27, 2021 at 02:44:23PM +0100, Steffen Trumtrar wrote:
> At least bitstreams for FPGAs can consist of a lot of zeros depending on
> device utilization. These bitstreams can be compressed very effectively.
> 
> Let the firmware code accept these images and decompress them before
> handing it to the firmware-manager in question.
> 
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
> ---
>  common/firmware.c | 43 ++++++++++++++++++++++++++++++++++++++-----
>  1 file changed, 38 insertions(+), 5 deletions(-)
> 
> diff --git a/common/firmware.c b/common/firmware.c
> index 58509d5da615..c617ecc0f129 100644
> --- a/common/firmware.c
> +++ b/common/firmware.c
> @@ -14,6 +14,8 @@
>  #include <linux/list.h>
>  #include <linux/stat.h>
>  #include <linux/err.h>
> +#include <uncompress.h>
> +#include <filetype.h>
>  
>  #define BUFSIZ 4096
>  
> @@ -211,12 +213,43 @@ int firmwaremgr_register(struct firmware_handler *fh)
>   */
>  int firmwaremgr_load_file(struct firmware_mgr *mgr, const char *firmware)
>  {
> -	int ret;
> -	char *name = basprintf("/dev/%s", mgr->handler->id);
> -
> -	ret = copy_file(firmware, name, 0);
> +	char *dst;
> +	enum filetype type;
> +	int ret = -ENOENT;
> +	int srcfd = 0;
> +	int dstfd = 0;
> +
> +	if (mgr->handler->id) {
> +		dst = basprintf("/dev/%s", mgr->handler->id);
> +	} else {
> +		pr_err("id not defined for handler\n");
> +		return -ENODEV;
> +	}

I'd do the error handling for !mgr->handler->id first, then the dst =
basprintf(...) doesn't have to be done conditionally.

>  
> -	free(name);
> +	if (firmware) {

This function seems to do nothing when firmware is not given. Is this
the correct behaviour? If yes, then you could bail out early instead of
putting the rest of the function inside an if() {}

> +		type = file_name_detect_type(firmware);
> +		if (type == filetype_unknown) {
> +			ret = copy_file(firmware, dst, 0);

When going this way srcfd is closed below, but never opened.

> +		} else {
> +			srcfd = open(firmware, O_RDONLY);
> +			if (srcfd < 0)
> +				return srcfd;

'dst' is not freed here.

> +			dstfd = open(dst, O_WRONLY | O_CREAT);

You'll probably want to add a O_TRUNC here because otherwise if the file
existed before and was bigger than the data you gonna write to it you'll
have garbage at the end of the file.
OTOH I don't know what 'dst' is. When this could be a device file then
truncating it will fail.

> +
> +			if (dstfd < 0) {
> +				printf("could not open %s: %s\n", dst, errno_str());
> +				ret = dstfd;
> +				goto out;
> +			}
> +
> +			ret = uncompress_fd_to_fd(srcfd, dstfd, uncompress_err_stdout);
> +
> +			close(dstfd);
> +		}
> +	}
> +out:
> +	close(srcfd);
> +	free(dst);
>  
>  	return ret;
>  }

Sascha


-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

      reply	other threads:[~2021-01-28 15:27 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-27 13:44 [PATCH v2 01/12] ARM: mmu: inherit pte flags from pmd Steffen Trumtrar
2021-01-27 13:44 ` [PATCH v2 02/12] firmware: socfpga: set APPLYCFG after loading bitstream Steffen Trumtrar
2021-01-27 13:44 ` [PATCH v2 03/12] reset: add of_reset_control_get to header Steffen Trumtrar
2021-01-27 13:44 ` [PATCH v2 04/12] drivers: add fpga bridge framework Steffen Trumtrar
2021-01-27 13:44 ` [PATCH v2 05/12] drivers: fpga: add socfpga bridges Steffen Trumtrar
2021-01-27 13:44 ` [PATCH v2 06/12] firmware: socfpga: change function prefixes Steffen Trumtrar
2021-01-27 13:44 ` [PATCH v2 07/12] firmware: import fpga-mgr.h from linux Steffen Trumtrar
2021-01-27 13:44 ` [PATCH v2 08/12] of: kconfig: of_overlay uses firmwaremgr_load_file Steffen Trumtrar
2021-01-27 13:44 ` [PATCH v2 09/12] of: of_firmware: add support for fpga bridges Steffen Trumtrar
2021-01-28 15:10   ` Sascha Hauer
2021-01-27 13:44 ` [PATCH v2 10/12] commands: firmwareload: allow loading firmware from dt Steffen Trumtrar
2021-01-28 15:16   ` Sascha Hauer
2021-01-27 13:44 ` [PATCH v2 11/12] drivers: firmware: socfpga: remove bridges shutdown Steffen Trumtrar
2021-01-27 13:44 ` [PATCH v2 12/12] firmware: add support for compressed images Steffen Trumtrar
2021-01-28 15:27   ` Sascha Hauer [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210128152733.GD19583@pengutronix.de \
    --to=sha@pengutronix.de \
    --cc=barebox@lists.infradead.org \
    --cc=s.trumtrar@pengutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox