public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
From: AKASHI Takahiro <takahiro.akashi@linaro.org>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH v4 8/9] cmd: efitool: export uefi variable helper functions
Date: Thu, 17 Jan 2019 16:30:56 +0900	[thread overview]
Message-ID: <20190117073054.GW20286@linaro.org> (raw)
In-Reply-To: <f6d3b97b-c251-eb50-4130-a9b3949d7890@gmx.de>

On Tue, Jan 15, 2019 at 06:28:38AM +0100, Heinrich Schuchardt wrote:
> On 1/15/19 3:55 AM, AKASHI Takahiro wrote:
> > Those function will be used for integration with 'env' command
> > so as to handle uefi variables.
> > 
> > Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
> > ---
> >  cmd/efitool.c     | 38 ++++++++++++++++++++++++++++++++++----
> >  include/command.h |  4 ++++
> >  2 files changed, 38 insertions(+), 4 deletions(-)
> > 
> > diff --git a/cmd/efitool.c b/cmd/efitool.c
> > index f06718ea580d..b8fe28c53aaf 100644
> > --- a/cmd/efitool.c
> > +++ b/cmd/efitool.c
> > @@ -67,7 +67,7 @@ static void dump_var_data(char *data, unsigned long len)
> >   *
> >   *   efi_$guid_$varname = {attributes}(type)value
> >   */
> > -static int do_efi_dump_var(int argc, char * const argv[])
> > +static int _do_efi_dump_var(int argc, char * const argv[])
> 
> Please, do not use two function names only distinguished by and
> underscore. A a reader I will always wonder which does what.

While I believe that "_" and "__" are a very common practice to
show an internal version of function,

> Please, use names that describe what the difference is.

I will revert _do_efi_dump_var() to do_efi_dump_var() and
change do_efi_dump_var() to do_efi_dump_var_ext().

> >  {
> >  	char regex[256];
> >  	char * const regexlist[] = {regex};
> > @@ -130,6 +130,21 @@ static int do_efi_dump_var(int argc, char * const argv[])
> >  	return CMD_RET_SUCCESS;
> >  }
> >  
> > +int do_efi_dump_var(int argc, char * const argv[])
> > +{
> > +	efi_status_t r;
> > +
> > +	/* Initialize EFI drivers */
> > +	r = efi_init_obj_list();
> > +	if (r != EFI_SUCCESS) {
> > +		printf("Error: Cannot set up EFI drivers, r = %lu\n",
> > +		       r & ~EFI_ERROR_MASK);
> 
> You duplicate this code below.

Do you want another function for just calling one function?
I don't think it's worth doing so.

> So, please, put this into a separate function.
> 
> That could also help to avoid having separate do_efi_set_var and
> _do_efi_set_var.

I don't get your point.
Do you want to call efi_init_obj_list() at every do_efi_xxx()?
It just doesn't make sense.

> > +		return CMD_RET_FAILURE;
> > +	}
> > +
> > +	return _do_efi_dump_var(argc, argv);
> > +}
> > +
> >  static int append_value(char **bufp, unsigned long *sizep, char *data)
> >  {
> >  	char *tmp_buf = NULL, *new_buf = NULL, *value;
> > @@ -227,7 +242,7 @@ out:
> >  	return 0;
> >  }
> >  
> > -static int do_efi_set_var(int argc, char * const argv[])
> > +static int _do_efi_set_var(int argc, char * const argv[])
> >  {
> >  	char *var_name, *value = NULL;
> >  	efi_uintn_t size = 0;
> > @@ -270,6 +285,21 @@ out:
> >  	return ret;
> >  }
> >  
> > +int do_efi_set_var(int argc, char * const argv[])
> > +{
> > +	efi_status_t r;
> > +
> > +	/* Initialize EFI drivers */
> > +	r = efi_init_obj_list();
> > +	if (r != EFI_SUCCESS) {
> > +		printf("Error: Cannot set up EFI drivers, r = %lu\n",
> 
> It is more then drivers that we setup. So perhaps
> 
> "Cannot initialize UEFI sub-system."

I don't mind, but that string directly comes from the original code
in bootefi.c.

> > +		       r & ~EFI_ERROR_MASK);
> > +		return CMD_RET_FAILURE;
> > +	}
> > +
> > +	return _do_efi_set_var(argc, argv);
> > +}
> > +
> >  static int efi_get_handles_by_proto(efi_guid_t *guid, efi_handle_t **handlesp,
> >  				    int *num)
> >  {
> > @@ -1016,9 +1046,9 @@ static int do_efitool(cmd_tbl_t *cmdtp, int flag,
> >  	if (!strcmp(command, "boot"))
> >  		return do_efi_boot_opt(argc, argv);
> >  	else if (!strcmp(command, "dumpvar") || !strcmp(command, "dmpstore"))
> > -		return do_efi_dump_var(argc, argv);
> > +		return _do_efi_dump_var(argc, argv);
> >  	else if (!strcmp(command, "setvar"))
> > -		return do_efi_set_var(argc, argv);
> > +		return _do_efi_set_var(argc, argv);
> >  	else if (!strcmp(command, "devices"))
> >  		return do_efi_show_devices(argc, argv);
> >  	else if (!strcmp(command, "drivers"))
> > diff --git a/include/command.h b/include/command.h
> > index feddef300ccc..315e4b9aabfb 100644
> > --- a/include/command.h
> > +++ b/include/command.h
> > @@ -51,6 +51,10 @@ extern int do_run(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]);
> >  #if defined(CONFIG_CMD_BOOTEFI)
> >  extern int do_bootefi_run(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]);
> >  #endif
> > +#if defined(CONFIG_CMD_EFITOOL)
> 
> The definition has not dependencies. No need for an #if here.

Currently has, but as you and Alex suggested in a separate thread,
we may want to enable it (and do_efi_dump/set_var() as well)
whether or not CMD_BOOTEFI and CMD_EFIDEBUG, respectively, are defined.

Thanks,
-Takahiro Akashi

> > +int do_efi_dump_var(int argc, char * const argv[]);
> > +int do_efi_set_var(int argc, char * const argv[]);
> > +#endi
> 
> Shouldn't we put this into some include/efi*?
> 
> Best regards
> 
> Heinrich
> 
> >  
> >  /* common/command.c */
> >  int _do_help (cmd_tbl_t *cmd_start, int cmd_items, cmd_tbl_t * cmdtp, int
> > 
> 

  reply	other threads:[~2019-01-17  7:30 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-15  2:55 [U-Boot] [PATCH v4 0/9] cmd: add efitool for efi environment AKASHI Takahiro
2019-01-15  2:55 ` [U-Boot] [PATCH v4 1/9] cmd: add efitool command AKASHI Takahiro
2019-01-15  3:31   ` Heinrich Schuchardt
2019-01-17  4:27     ` AKASHI Takahiro
2019-01-15  2:55 ` [U-Boot] [PATCH v4 2/9] cmd: efitool: add devices command AKASHI Takahiro
2019-01-15  5:09   ` Heinrich Schuchardt
2019-01-17  4:48     ` AKASHI Takahiro
2019-01-15  2:55 ` [U-Boot] [PATCH v4 3/9] efi_driver: add name to driver binding protocol AKASHI Takahiro
2019-01-15  3:41   ` Heinrich Schuchardt
2019-01-17  5:33     ` AKASHI Takahiro
2019-01-17  6:58       ` Heinrich Schuchardt
2019-01-21  7:47         ` AKASHI Takahiro
2019-01-15  2:55 ` [U-Boot] [PATCH v4 4/9] cmd: efitool: add drivers command AKASHI Takahiro
2019-01-15  3:39   ` Heinrich Schuchardt
2019-01-15  2:55 ` [U-Boot] [PATCH v4 5/9] cmd: efitool: add dh command AKASHI Takahiro
2019-01-15  4:55   ` Heinrich Schuchardt
2019-01-17  6:01     ` AKASHI Takahiro
2019-01-15  2:55 ` [U-Boot] [PATCH v4 6/9] cmd: efitool: add images command AKASHI Takahiro
2019-01-15  4:58   ` Heinrich Schuchardt
2019-01-17  6:02     ` AKASHI Takahiro
2019-01-15  2:55 ` [U-Boot] [PATCH v4 7/9] cmd: efitool: add memmap command AKASHI Takahiro
2019-01-15  4:26   ` Heinrich Schuchardt
2019-01-17  7:03     ` AKASHI Takahiro
2019-01-15  2:55 ` [U-Boot] [PATCH v4 8/9] cmd: efitool: export uefi variable helper functions AKASHI Takahiro
2019-01-15  5:28   ` Heinrich Schuchardt
2019-01-17  7:30     ` AKASHI Takahiro [this message]
2019-01-15  2:55 ` [U-Boot] [PATCH v4 9/9] cmd: env: add "-e" option for handling UEFI variables AKASHI Takahiro
2019-01-15  3:47   ` Heinrich Schuchardt
2019-01-15 13:23     ` Alexander Graf
2019-01-15 13:26   ` Alexander Graf
2019-01-15  9:16 ` [U-Boot] [PATCH v4 0/9] cmd: add efitool for efi environment Alexander Graf

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=20190117073054.GW20286@linaro.org \
    --to=takahiro.akashi@linaro.org \
    --cc=u-boot@lists.denx.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