Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Could you explain the device driver related questioin?
6 views
Skip to first unread message
fl
May 11, 2013, 4:15:39 PM5/11/13
to
Hi,
I read IOCtl description on a device driver book. I do not understand the last two sentences. What meaning is "break no end of binary programs"? Why the kernel developers are not willing to do? In fact, it looks like a question on English comprehension. Please help me if you know that. Thanks a lot.
.......................................
To help programmers create unique ioctl command codes, these codes have been
split up into several bitfields. The first versions of Linux used 16-bit numbers: the top eight were the “magic” numbers associated with the device, and the bottom eight were a sequential number, unique within the device. This happened because Linus was “clueless” (his own word); a better division of bitfields was conceived only later. Unfortunately, quite a few drivers still use the old convention. They have to: changing the command codes would break no end of binary programs, and that is not something the kernel developers are willing to do.
I read IOCtl description on a device driver book. I do not understand the last two sentences. What meaning is "break no end of binary programs"? Why the kernel developers are not willing to do? In fact, it looks like a question on English comprehension. Please help me if you know that. Thanks a lot.
.......................................
To help programmers create unique ioctl command codes, these codes have been
split up into several bitfields. The first versions of Linux used 16-bit numbers: the top eight were the “magic” numbers associated with the device, and the bottom eight were a sequential number, unique within the device. This happened because Linus was “clueless” (his own word); a better division of bitfields was conceived only later. Unfortunately, quite a few drivers still use the old convention. They have to: changing the command codes would break no end of binary programs, and that is not something the kernel developers are willing to do.
Richard Kettlewell
May 11, 2013, 4:22:52 PM5/11/13
to
fl <rxj...@gmail.com> writes:
> I read IOCtl description on a device driver book. I do not understand
> the last two sentences. What meaning is "break no end of binary
> programs"? Why the kernel developers are not willing to do? In fact,
> it looks like a question on English comprehension. Please help me if
> you know that. Thanks a lot.
“no end of” = “lots of” or “many”.
> I read IOCtl description on a device driver book. I do not understand
> the last two sentences. What meaning is "break no end of binary
> programs"? Why the kernel developers are not willing to do? In fact,
> it looks like a question on English comprehension. Please help me if
> you know that. Thanks a lot.
--
http://www.greenend.org.uk/rjk/
Chick Tower
May 12, 2013, 4:41:19 PM5/12/13
to
On 2013-05-11, fl wrote:
> I read IOCtl description on a device driver book. I do not understand
> the last two sentences. What meaning is "break no end of binary
> programs"? Why the kernel developers are not willing to do? In fact,
> it looks like a question on English comprehension. Please help me if
> you know that. Thanks a lot.
You're right, it is an English comprehension problem. Or, perhaps more
> I read IOCtl description on a device driver book. I do not understand
> the last two sentences. What meaning is "break no end of binary
> programs"? Why the kernel developers are not willing to do? In fact,
> it looks like a question on English comprehension. Please help me if
> you know that. Thanks a lot.
specifically, an unknown (to you) English idiom.
"Break no end of binary programs" means the list of binary programs it
would break would have no end. Thus, many programs would break. And
that is why kernel developers don't want to do it.
--
Chick Tower
For e-mail: colm DOT sent DOT towerboy AT xoxy DOT net
J G Miller
May 12, 2013, 4:54:49 PM5/12/13
to
On Sunday, May 12th, 2013, at 20:41:19h +0000, Chick Tower wrote:
> And that is why kernel developers don't want to do it.
And even more importantly, at 09:36:15h -0800
> And that is why kernel developers don't want to do it.
on Sunday, December 23rd, Linus Benedict Torvalds declared,
"WE DO NOT BREAK USERSPACE!"
at <https://lkml.ORG/lkml/2012/12/23/75>
0 new messages