grub-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GRUB2: *BSD and more patch


From: Yoshinori K. Okuji
Subject: Re: GRUB2: *BSD and more patch
Date: Mon, 22 Mar 2004 14:49:41 +0100
User-agent: KMail/1.5.3

On Monday 22 March 2004 09:20, Sergey Matveychuk wrote:
> Yoshinori K. Okuji wrote:
>   >>-  if (! S_ISBLK (st.st_mode))
> >>
> >>-    return make_device_name (drive, -1, -1);
> >>-
> >
> > This part is not good. The problem here is that we want to support
> > installing GRUB into a normal file as well as a device file.
>
> OK. I've taken it back.

I got an advise from Mr.Soda personally about the block device issue. 
And, I now think it is better to use S_ISREG rather than S_ISBLK when 
determining if a file is a disk device or a normal file. That is, use 
S_ISREG for a normal file, and use ! S_ISREG for a disk device.

So, regardless of your OS, let's use:

if (S_ISREG (st.st_mode))
  return make_device_name (drive, -1, -1);

The same applies to other places where S_ISBLK is used.

> > BTW, I confirmed that you haven't assigned your copyright on GRUB
> > to the FSF yet. For a GNU project, it is a custom to assign your
> > copyright to the FSF, so that the FSF can fight instead of you in a
> > court when someone claims that our software is illegal. Also, this
> > step of a copyright assignment makes sure that your contribution
> > will be free in freedom forever. So, would you like to sign a
> > copyright assignment for GRUB? If you need more information, don't
> > hesitate to ask me. We can talk privately if you want.
>
> Do you mean GRUB1?

There is no difference between GRUB 1 and GRUB 2, as far as we talk 
about the legal issue, since they are the same software from legal 
point of view. It's GRUB.

> @@ -194,7 +197,11 @@ pupa_memalign (pupa_size_t align, pupa_s
>  {
>    void *p;
>    
> +#ifdef HAVE_MEMALIGN
>    p = memalign (align, size);
> +#else
> +  pupa_util_error ("memalign not available");
> +#endif
>    if (! p)
>      pupa_util_error ("out of memory");

This segfaults if pupa_memalign is used actually. It must return a NULL 
pointer if fails.

Okuji





reply via email to

[Prev in Thread] Current Thread [Next in Thread]